doc.anet.be faculty of medicine and health sciences characterization and modulation of the enteric...

Front cover Image: Bully by Hanne Van Spaendonk Back cover Image: Baseline tracing of murine lumbar colonic nerve activity; what I would have finished if time was on my side. Cover design: Natacha Hoevenaegel Printed by: D Provo nv, Brulens 23B, B-2275 Gierle The research described in this dissertation was performed at the Laboratory of Experimental Medicine and Pediatrics, Division of Gastroenterology and Hepatology at the University of Anwerp. Sara Nullens is an aspirant of the Fund for Scientific Research Flanders (FWO) (11G7415N). This work was supported financially by the FWO (grant G028615N) © Sara Nullens, Antwerp, 2016. All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the holder of copyright.

Faculty of Medicine and Health Sciences

Characterization and modulation of the

enteric neuroimmune environment during sepsis-induced ileus

Karakterisering en modulatie van de enterische

neuro-immune omgeving gedurende sepsis-geïnduceerde ileus

Proefschrift voorgelegd tot het behalen van de graad van Doctor in de Medische Wetenschappen aan de Universiteit Antwerpen te verdedigen door:

Sara NULLENS Promotoren: Prof. dr. Benedicte De Winter Prof. dr. Sven Francque Begeleider: Ing. Joris De Man

Antwerpen, 2016

II

Members of the jury

Doctoral Committee

Prof. dr. Guy Hubens – Chairman

Laboratory of Human Anatomy and Embryology, Faculty of Medicine and

Health Sciences, University of Antwerp, Belgium

Prof. dr. Nathalie Cools – Member

Laboratory of Experimental Hematology, Vaxinfectio,

University of Antwerp, Belgium

External members of the jury

Prof. dr. Jörg Kalff - Member

Department for General, Visceral, Thoracic and Vascular Surgery,

University Hospital Bonn, Germany

Prof. dr. Romain Lefebvre – Member

Heymans Institute of Pharmacology, University of Ghent, Belgium

III

TABLE OF CONTENTS

TABLE OF CONTENTS III ABBREVIATIONS AND UNITS OF MEASURE VIII Chapter 1: INTRODUCTION 1 1.1 Sepsis……………………………………………………………………………………………………………………… 2

1.1.1 Definition and pathophysiology of sepsis………………………………………………. 2 1.1.2 Current guidelines on the treatment of sepsis………………………………………. 9

1.2 The gut and inflammation………………………………………………………………………………………. 13 1.2.1 Permeability………………………………………………………………………………………….. 13 1.2.1.1 Organization of the gastrointestinal epithelial barrier………….. 13 1.2.1.2 The epithelium during inflammation…………………………………….. 14 1.2.2 Inflammation – the immune cells in the gut………………………………………..… 16 1.2.3 Innervation……………………………………………………………………………………………. 21 1.2.3.1 Intrinsic innervation of the gut……………………………………………. 21 1.2.3.2 Extrinsic innervation of the gut…………………………………………… 22 1.2.3.3 Ascending spinal pathways and central processing……………… 26 1.2.3.4 Central descending inhibitory pathways………………………………. 26 1.2.4 Impaired gastrointestinal motility – ileus……………………..……………………….. 27 1.2.4.1 Post-operative ileus……………………………………………………………… 28 1.2.4.2 Septic ileus……………………………………………………………………………. 31 1.2.4.3 The contribution of ileus to sepsis………………………………………… 31

1.3 The gut during sepsis……………………………………………………………………………………………… 32 1.3.1 Bacterial translocation………………………………………………………………………….. 32 1.3.2 The nervous system of the gut during sepsis………………………………………… 34 1.3.3 The cholinergic anti-inflammatory pathway………………………………………….. 36 1.3.4 The contribution of the ‘failing’ gastrointestinal tract to sepsis……………. 39 1.4 Possible therapeutic treatment strategies…………………………………………………………….. 40 1.5 Studying new treatment strategies – animal models of sepsis………………………………..40 1.5.1 The lipopolysaccharide model………………………..……………………………………… 40 1.5.2 The cecal ligation and puncture-model…………………………………………………. 42 1.5.3 The colon ascendens stent peritonitis model…………………………………………. 44 1.6 Summary……………………………………………………………………………………………………………….. 44 Chapter 2: AIMS 45 Chapter 3: MATERIAL AND METHODS 51 3.1 Animals………………………………………………………………………………………………………………….. 52 3.2 The cecal ligation and puncture animal model of sepsis………………………………………… 52 3.3 Clinical signs of disease and sickness behavior………………………………………………………. 53 3.4 Assessment of gastrointestinal motility…………………………………………………………………. 54 3.4.1 The Evans blue method: gastrointestinal propulsion of semi-liquids.……. 56 3.4.2 The solid beads method: gastrointestinal propulsion of solids.……………… 56

IV

3.4.3 The solid bead expulsion test………………………………………………………………… 57 3.4.4 In vitro study of the colonic peristaltic activity………………………………………. 57 3.5 Assessment of inflammation…………………………………………………………………………………..57 3.5.1 Complete blood count………………………………………………………………………….. 57 3.5.2 Serum and colonic cytokines – protein level…………………………………………. 58 3.5.3 Colonic cytokines – mRNA level…………………………………………………………….. 58 3.5.4 Myeloperoxidase activity assay……………………………………………………..……… 59 3.6 Assessment of colonic permeability………………………………………………………………………. 59 3.6.1 The Evans blue method (colonic permeability)……………………………………… 59 3.6.2 Tight junction proteins………………………………………………………………………….. 60 3.7 Quantification of bacterial load……………………………………………………………………………… 60 3.8 Histology and immunohistochemistry…………………………………………………………………… 60 3.9 Cell culture experiments……………………………………………………………………………….………. 61 3.10 Data presentation and statistical analysis……………………………………………………………… 62 Chapter 4: A CHARACTERIZATION STUDY OF THE CECAL LIGATION AND PUNCTURE MODEL 63 4.1 Abstract……………………………………………………………………………………………………………….… 64 4.2 Introduction…………………………………………………………………………………………………………… 66 4.3 Material and methods…………………………………………………………………………………………….69 4.3.1 Animals…………………………………………………………………………………………………. 69 4.3.2 The cecal ligation and puncture animal model of sepsis………………………… 69 4.3.3 Survival analysis and clinical disease score……………………………………………. 70 4.3.4 Final experimental design……………………………………………………………………… 70 4.3.5 Measurement of gastrointestinal transit………………………………………………. 70 4.3.5.1 The Evans blue method………………………………………………………… 70 4.3.5.2 The solid beads method……………………………………………………….. 71 4.3.5.3 The bead expulsion test………………………………………………………… 71 4.3.5.4 In vitro study of colonic peristaltic activity………………………….… 71 4.3.6 Peripheral blood count…………………………………………………………………………. 72

4.3.7 Cytokine measurements……………………………………………………………………….. 72 4.3.8 Flow cytometric characterization of gastrointestinal tissues………….……… 72 4.3.8.1 Instrument characteristics……………………………………………………. 72 4.3.8.2 Tissue preparation – spleen and MLN single cell suspension… 73 4.3.8.3 Tissue preparation – isolation of LPMC from colon and ileum. 74 4.3.8.4 Staining protocols…………………………………………………………………. 75 4.3.8.5 Flow cytometric characterization of immune cells……………..… 78 4.3.9 Statistical analysis…………………………………………………………………….…………… 78

4.4 Results………………………………………………………………………………………………………………..…. 81 4.4.1 Survival analysis of the implemented CLP-model of sepsis………………….… 81 4.4.2 Implementation of the CLP-model to study septic ileus and gastrointestinal motility disturbances…………………………………………….….. 83 4.4.3 Peripheral blood count in the CLP-model……………………………………………… 85 4.4.4 Cytokine measurements in the CLP-model………………………………………….… 86 4.4.5 Flow cytometric characterization of gastrointestinal tissues………………… 88 4.4.5.1 T and B cell staining………………………………………………………………. 88 4.4.5.2 Dendritic cell staining…………………………………………………………… 91 4.4.5.3 Other leukocyte populations………………………………………………… 94

V

4.4.5.4 General summary of the flow cytometry data………………………. 95 4.4.6 Myeloperoxidase-activity……………………………………………………………………….97 4.4.7 Histology…………………………………………………………………..…………………………… 98 4.5 Discussion………………………………………………………..……………………………………………..…….. 101 Chapter 5: STUDY OF THE AFFERENT NERVE ACTIVITY AND CENTRAL NERVOUS SYSTEM ACTIVATION IN THE CECAL LIGATION AND PUNCTURE MODEL 107 5.1 Abstract……………………………………………………………………………………………………………….… 108 5.2 Introduction…………………………………………………………………………………………………………...109 5.3 Material and methods…………………………………………………………………………………………….113 5.3.1 Animals…………………………………………………………………………………………………. 113 5.3.2 Induction of sepsis………………………………………………………………………………… 113 5.3.3 Splanchnic afferent nerve recordings……………………………………………………. 113 5.3.3.1 Tissue preparation of jejunal afferent nerves……………………….. 113 5.3.3.2 Data acquisition and analysis………………………………………………… 115 5.3.3.3 Experimental design…………………………………………….………………. 117 5.3.4 Central nervous system imaging……………………………………………………………. 118 5.3.4.1 Experimental protocol……………………………………………….…………. 118 5.3.4.2 Data acquisition and analysis………………………………………………… 118 5.3.5 Dorsal root ganglia. …………………………………………………………….………………… 119 5.3.6 Statistics. ……………………………………………………………………………………………… 120 5.4 Results…………………………………………………………………………………………..………………………. 121 5.4.1 Multi-unit jejunal afferent nerve activity………………………………………………. 121 5.4.2 Single unit jejunal afferent nerve activity…………………………………………….… 122 5.4.3 Incubation of colonic segments with septic colonic samples…………………. 124 5.4.4 Central nervous system imaging………………………………………………………….… 126 5.4.5 Dorsal root ganglia……………………………………………………………………..…………. 130 5.5 Discussion……………………………………………………………………………………………………………… 131 Chapter 6: NEURONAL MODULATION OF SEPTIC ILEUS: EFFECT OF GTS-21, AN α7nAChR AGONIST, ON CLP-INDUCED ILEUS 135 6.1 Abstract…………………………………………………………………………………………………………………. 136 6.2 Introduction…………………………………………………………………………………………..……………….137 6.3 Material and methods…………………………………………………………………………………………….139 6.3.1 Animals…………………………………………………………………………………………………. 139 6.3.2 Induction of sepsis – the CLP-model……………………………………………………… 139 6.3.3 Dose optimization of GTS-21………………………………………………….…………….. 139 6.3.4 Experimental design…………………………………………………………….……………….. 140 6.3.5 In vivo measurement of gastrointestinal transit…………….……………………… 141 6.3.6 Cytokine measurements……………………………………………..………………………… 142 6.3.7 Colonic cell adhesion proteins………………………………………………………………. 142 6.3.8 Colonic permeability and bacterial translocation…………………………………… 142 6.3.9 In vitro culture of bone-marrow derived macrophages and dendritic cells………….………………………………………………………………………………….……. 143 6.3.10 Histology………………………………………………………………..…………………………..… 143 6.3.11 Presentation of results and statistical analysis……………………………………… 143 6.4 Results……………………………………………………………………………………………………..……………. 144

VI

6.4.1 Dose-dependent effects of GTS-21 on disturbed GI motility post-CLP….. 144 6.4.2 Effect of GTS-21 on CLP-induced septic ileus…………………………………………. 144 6.4.3 CLP and GTS-21 in Chrna7-/- mice…………………………………………………………… 149 6.4.4 Effects of splenectomy on CLP-induced septic ileus………………………………. 150 6.4.5 In vitro culture of bone-marrow derived macrophages and dendritic cells………….………………………………………………………………………………………… 152 6.5 Discussion……………………………………………………………………………..………………………………. 153 Chapter 7: IMMUNOLOGICAL MODULATION OF SEPTIC ILEUS: EFFECT OF ANTIBODIES TO INTERLEUKIN-6 ON CLP-INDUCED ILEUS 159 7.1 Abstract…………………………………………………………………………………………………………………. 160 7.2 Introduction………………………………………………………………………………………..………………….161 7.3 Material and methods…………………………………………………………………………………………….166 7.3.1 Animals……………………………………………………………………………….………………... 166 7.3.2 Induction of sepsis…………………………………………………………….………………….. 166 7.3.3 Experimental design……………………………………………………….…………………….. 166 7.3.4 In vivo measurement of gastrointestinal transit……………………………………. 167 7.3.5 Cytokine measurements…………………………………………..…………………………… 167 7.3.6 Colonic permeability, tight junction molecules and bacterial translocation……………………………………………………………………………………… 169 7.3.7 Histology………………………………………………………………..……………………………… 169 7.3.8 Statistical analysis………………………………………………………..……………………….. 169 7.4 Results………………………………………………………………………………………..…………………….….. 170

7.4.1 Effect of preventive administration of anti-IL-6 antibodies on septic ileus…………..………………………………………………………………………………………. 170

7.4.2 Effect of preventive administration of anti-IL-6 antibodies with repeat injection 24h later on CLP-induced septic ileus…………………………………… 174

7.4.3 Effect of curative administration of anti-IL-6 antibodies on septic ileus… 175 7.5 Discussion…………………………………………………………………………………………..…………………. 182 Chapter 8: GENERAL DISCUSSION 187 8.1 Characterization of the different neuroimmune players in the digestive tract involved in the pathogenesis of sepsis-induced ileus in order to identify new therapeutic targets……………………………………………………………………………………………………………… 190 Implementation of an animal model of sepsis in our lab………………………….………. 190 Flow cytometric characterization of the cecal ligation and puncture model of sepsis…….…………………………………………………………………………………………… 191 Afferent mesenteric nerve recordings and activation status of the dorsal root

ganglia and central nervous system in the CLP-model……………............. 194 8.2 Neuronal modulation of the neuroimmune environment…………………..…………………. 198 8.3 Immunological modulation of the neuroimmune environment……….………………….….200 8.4 Clinical implications of the research presented in this thesis…………………………………. 201 8.5 Future challenges and perspectives…………………………………………….…………………………. 202 Chapter 9: SUMMARY 205 Chapter 10: SAMENVATTING 209

VII

Chapter 11: REFERENCES 213 SCIENTIFIC CURRICULUM VITAE 246 DANKWOORD 257

VIII

ABBREVIATIONS AND UNITS OF MEASURE

Abbreviations %GE percentage of gastric emptying [18F]-FDG fluorodeoxyglucose α7nAChR alpha7 nicotinic acetylcholine receptor ACC anterior cingulate cortex AF alexa fluor (fluorochrome) ANOVA analysis of variance APC antigen presenting cell or allophycocyanin (fluorochrome) BB brilliant blue (fluorochrome) CARS compensatory anti-inflammatory response syndrome CASP colon ascendens stent peritonitis CB cannabinoid receptor CBA cytometric bead assay CBC cell blood count CCL chemokines C-C motif ligand CCR C-C chemokines receptor CD cluster of differentiation cDC conventional dendritic cell cDNA complement DNA CDS clinical disease score CFU colony forming units CGRP calcitonin gene related peptide Chrna7 neuronal acetylcholine receptor subunit alpha-7 CLP cecal ligation and puncture CLPd2 cecal ligation and puncture – sacrificed at day 2 post-procedure CLPd7 cecal ligation and puncture – sacrificed at day 7 post-procedure CNS central nervous system CTLA-4 cytotoxic T lymphocyte associated protein 4 CXCL chemokines C-X-C motif ligands DAB diaminobenzidine DAMP danger associated molecular pattern DC dendritic cell DNA deoxyribonucleic acid DRG dorsal root ganglion DSS dextran sulphate sodium EB Evans blue EDTA ethylenediaminetetracetic acide eEF2 eukaryotic elongation factor 2 ENS enteric nervous system FAAH fatty acid amide hydrolase enzyme FALGPA Furylacryloyl-Leucine-Glycyl-Propyl-Alanine FITC fluorescein isothiocyanate FMO fluorescence minus one Foxp3 forkhead box P3 FSC forward scatter

IX

GABA gamma-aminobutyric acid GALT gut-associated lymphoid tissue GAPDH glyceraldehyde-3-phosphate dehydrogenase GC geometric center of transit GI gastrointestinal GIT gastrointestinal tract GLM generalized linear models HLA-DR Human Leukocyte Antigen – antigen D Related HMGB high-mobility group protein B1 HT high threshold fiber IBD inflammatory bowel disease IBS irritable bowel syndrome ICC interstitial cells of Cajal ICU intensive care unit IEL intraepithelial lymphocytes IFN interferon Ig immunoglobulin IL interleukin ILC innate lymphoid cells iNOS inducible nitric oxide synthase INR international normalized ratio i.p. intraperitoneally IRAK interleukin-1 receptor-associated kinase 4 KO knock-out LGM lymphocyte growth medium LPMC lamina propria mononuclear cells LPS lipopolysaccharide LSD least significant difference LT low threshold fiber M cell microfold cell Mφ macrophage M1 Mφ M1 macrophage MAP mean arterial pressure MAPK mitogen associated protein kinase MC mast cell MCP-1 monocyte chemoattractant protein-1 MEM minimum essential medium MHCII major histocompatibility complex II MIA mechano-insensitive afferent fiber MLN mesenteric lymph nodes Mo monocyte MODS multiple organ dysfunction syndrome MOF multiple organ failure MPO myeloperoxidase mRNA messenger ribonucleic acid MyD88 myeloid differentiation primary response gene 88 NA noradrenalin nAChR nicotinic acetylcholine receptor

X

NaCl saline, 0.9% sodium chloride solution NF-κB nuclear factor kappa B NK natural killer cell NKT natural killer T cell NMDA N-Methyl-D-aspartate NO nitric oxide NPh neutrophil PAF platelet activating factor PAG periaqueductal gray PAMP pathogen associated molecular pattern PCR polymerase chain reaction PD-1 programmed cell death protein 1 PD-L1 programmed death ligand 1 PE phycoerythrin (fluorochromes) PET/CT positron emission tomography – computed tomography POI postoperative ileus PRR pattern recognition receptor RNA ribonucleic acid ROS reactive oxygen species rpm rounds per minute RPMI Roswell Park Memorial Institute medium, cell culture medium rRNA ribosomal RNA RT-PCR reverse transcription polymerase chain reaction SBP systolic blood pressure s.c. subcutaneously SDD selective digestive tract decontamination SEM standard error of the mean SIRS systemic inflammatory response syndrome SMC smooth muscle cell SPLX splenectomy SPM statistical parametric mapping analysis SSC side scatter STAT signal transducer and activator of transcription SUVglc standard uptake value corrected for endogenous glucose levels TCR T cell receptor TGF tumor growth factor Th T helper TLR toll-like receptor TNF tumor necrosis factor TRPA transient receptor potential ion channel ankyrin Treg regulatory T cell TRPV transient receptor potential cation channel vanilloids TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling VIP vasoactive intestinal peptide VNS vagal nerve stimulation VOI volume-of-interest WBC white blood cells WDR wide dynamic range fiber

XI

Units of measure % percentage <L/D below the limit of detection µL microliter µm micrometer µmol micromoles µV microvolt °C degrees Celsius cm centimeter dL deciliter g gravity g gram G gauge h hour Hz hertz imp impulses kD kilodalton kg kilogram L liter LT% afferent firing at 20 mmHg/maximal afferent firing M molar mg milligram min minute mL milliliter mm3 cubic millimeter mm millimeter mM millimolar mmHg millimeters of mercury mV millivolt nm nanometer Pa arterial pressure pg pictogram pH potential of hydrogen s seconds U units w/v weight/volume

Chapter 1

Introduction

Based upon :

Nullens S, Staessens M, Peleman C, Jorens PG, Francque SM, De Man JG, De Winter BY.

Interplay between the gut and the immune system during sepsis: from characterizing

the gastrointestinal neuroimmune environment to the development of new

therapeutic agents. In preparation.

Chapter 1 Introduction

2

Sepsis, defined by experts in the field as a systemic inflammatory response secondary to

infection, is an important cause of death in critically ill patients (Bone et al., 1992). The clinical

importance of sepsis is emphasized in epidemiological studies, reporting an estimated 19

million cases a year worldwide and representing somewhere between 10% of all intensive care

unit (ICU) admissions in the United States (Angus et al., 2013) and 24.7% in European Intensive

Care Units (Vincent et al., 2006), depending on the definition criteria that were used. A recent

worldwide study even reported in-hospital mortality rates from sepsis to be as high as 35.3%

(Vincent et al., 2014).

Growing evidence states that the gastrointestinal (GI) tract is not just an innocent bystander in

the pathogenesis of sepsis, as it is nowadays generally accepted that the gut can initiate,

perpetuate and/or exacerbate sepsis and its complications, thus contributing to many fatalities

(Deitch et al., 2006; Deitch, 2010). Several gastrointestinal processes are believed to play an

important pathogenic role, amongst which there are ileus, mucosal barrier dysfunction and the

occurrence of bacterial translocation. Ileus, defined as the loss of normal bowel peristalsis and

a delay in gastric emptying, can be induced by sepsis. Together with an increased mucosal

permeability, this can lead to bacterial translocation and further deterioration of the patient’s

clinical picture (Hassoun et al., 2001).

In this first Chapter, we will provide an overview on the exact definition of sepsis, on what is

known concerning the pathogenesis and treatment guidelines, while focusing on the GI tract

and immune system.

1.1 Sepsis

1.1.1 Definition and pathophysiology of sepsis

In 1991, the American College of Chest Physicians and the Society of Critical Care Medicine

formulated the uniform definitions for the concepts of ‘infection’, ‘bacteremia’, ‘systemic

inflammatory response syndrome (SIRS)’, ‘sepsis’, ‘severe sepsis’, ‘septic shock’ and ‘multiple

organ dysfunction syndrome (MODS)’ (table 1.1), designed to be implemented in daily clinical

practice as well as in research facilities (Bone, 1992). Sepsis and its consequences represent a

continuum of clinical and pathophysiological severity, and can be complicated by organ

dysfunction resulting in severe sepsis and septic shock with multiple organ failure (MOF), and

eventually death (Bone et al., 1992). These definitions are now being updated regularly on a

biannual basis during the ‘International sepsis definitions conference’, during which treatment


3

guidelines are formulated as well in the ‘Surviving Sepsis Campaign’ (Levy et al., 2010; Dellinger

et al., 2013).

Recently, the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3)

were published (Singer et al., 2016; Shankar-Hari et al., 2016). In this consensus report, a task

force with many renowned leaders in the sepsis field stated that the previous definitions

focused excessively on inflammation, that the current SIRS criteria lacked specificity and

sensitivity, and that a major limitation of previous guidelines concerns the misleading model

that sepsis follows a continuum through severe sepsis and septic shock. They therefore suggest

to replace the previous definitions, and to define sepsis as a ‘life-threatening organ dysfunction

caused by a dysregulated host response to infection’, and organ dysfunction is suggested to be

represented by an increase in the Sepsis-related Organ Failure Assessment or SOFA-score of 2

points or more. Septic shock on the other hand should be defined as ‘a subset of sepsis in

which particularly profound circulatory, cellular, and metabolic abnormalities are associated

with a greater risk of mortality than with sepsis alone’. Similar to the previous guidelines,

patients with septic shock can be clinically identified by a vasopressor requirement or serum

lactate level greater than 2 mmol/L (>18mg/dL) in the absence of hypovolemia.

The causes of multiple organ dysfunction in sepsis and its analogous syndromes aren’t fully

elucidated, but it has been shown that impaired tissue oxygenation due to hypotension, a

reduced red blood cell deformability and microvascular thrombosis play a key role in septic

shock. Endothelial dysfunction and loss of cell membrane integrity lead to edema in tissues and

swelling of cells. In addition, mitochondrial damage caused by reactive oxygen species (ROS)

impairs cellular respiration and danger-associated molecular patterns (DAMPs) released from

damaged cells will worsen tissue injury. The progressing MOF ultimately can result in the

demise of the septic patient (Angus et al., 2013). It has been acknowledged over three decades

ago that the host, and not the germ, is the main player that drives the pathogenesis of sepsis

(Cerra et al., 1985). The immune response generated by the patient in response to the

pathogens or injury is often exaggerated, thus aggravating tissue damage, giving rise to the

cytokine theory of disease (Bone et al., 1997; Pinsky et al., 1993). This cytokine hypothesis is

only a small part of the complicated story underlying the pathogenesis of sepsis, as not only

cytokines, but also the cells secreting them as well as other mediators need to be taken into


4

account, more specifically the complete local immune response interfering with other cell

types in the environment (Tracey, 2007; Ulloa & Tracey, 2005).

In the vast majority of patients, an initial local reaction occurs at the focus of infection,

triggering both pro- and anti-inflammatory responses at a local level resulting in a mixture of

mediators secreted in the micro-environment. The immune cells of the innate immune system,

such as dendritic cells, natural killer cells and innate lymphoid cells, carry pattern recognition

receptors (PRRs) which interact with pathogen associated molecular patterns (PAMPs),

structures that are conserved among microbial species, such as lipopolysaccharide (LPS) or

endotoxin. The same receptors also interact with endogenous molecules released from injured

cells, the danger associated molecular patterns (DAMPS) or alarmins, of which the high-

mobility group protein B1 (HMGB), extracellular DNA and histones are the most infamous ones

(Chan et al., 2012). The PRRs comprise 4 main classes: toll-like receptors (TLR), C-type lectine

receptors, retinoic acid inducible gene1-like receptors and nucleotide-binding oligomerization

domain receptors (Takeuchi & Akira, 2010). Binding of LPS onto the TLR4-receptor, which is

described in detail below, will ultimately result in the activation of the nuclear factor kappa B

(NF-κB) and the release of reactive oxygen and nitrogen species, contributing to the sepsis-

related organ dysfunctions. In case of a severe initial insult which the micro-environment fails

to handle, inflammatory mediators will subsequently appear in the systemic circulation. This

will result in the recruitment of neutrophils, T cells, B cells, monocytes, macrophages, platelets

and coagulation factors to the location of the initial insult. This initial systemic response

comprises both pro- and anti-inflammatory responses during which the proinflammatory

response will eliminate pathogens and promote systemic healing, while the anti-inflammatory

response is necessary to subdue its counterpart, limiting local and systemic tissue damage and

promoting tissue repair (Platzer et al. 1995)(figure 1.1).


5

Table 1.1 The definition of sepsis and organ failure, based upon the original definitions as formulated by the American College of Chest Physicians and the Society of Critical Care Medicine in 1991 (Bone et al., 1992).

Disorder Definition Requirements

Infection inflammatory response to presence of micro-organisms or the invasion of normally sterile host tissue by micro-organisms

Bacteremia presence of viable bacteria in the blood

SIRS (systemic anti-inflammatory response syndrome)

clinical syndrome that results from a dysregulated, systemic inflammatory response to an infectious insult (such as pancreatitis, ischemia, polytrauma, hemorrhagic shock, auto-immune disorder, burns, trombo-embolism, surgery, …)

Two or more of the following must be present:

Temperature >38.3°C or <36°C Heart rate >90 beats/min Respiratory rate >20 breaths/min

or PaCO2 <32 mmHg WBC >12,000 cells/mm3, <4000

cells/mm3, or >10 percent immature (band) forms neutrophils

Sepsis clinical syndrome that results from a dysregulated systemic inflammatory response to an infection (SIRS + documented infection)

Two or more of the following must be present:

Temperature >38.3°C or <36°C Heart rate >90 beats/min Respiratory rate >20 breaths/min

or PaCO2 <32 mmHg WBC >12,000 cells/mm3, <4000

cells/mm3, or >10 percent immature (band) forms neutrophils

Severe sepsis Sepsis and organ dysfunction or hypoperfusion, including but not limited to: lactic acidosis, oliguria, or an acute alteration in mental status.

Septic shock Circulatory failure due to inappropriate vasodilation and/or vasoplegia (= severe sepsis + hypotension despite adequate fluid resuscitation)

This hypotension is defined as one or both of the following:

Systemic mean blood pressure is <60 mmHg (or <80 mmHg if the patient has baseline hypertension) despite adequate fluid resuscitation

Maintaining the systemic mean blood pressure >60 mmHg (or >80 mmHg if the patient has baseline hypertension) requires dopamine >5 µg/kg per min, norepinephrine <0.25 µg/kg per min, or epinephrine <0.25 µg/kg per min despite adequate fluid resuscitation


6

Table 1.2 Current diagnostic criteria for ‘sepsis’ and ‘severe sepsis’.

Sepsis Infection, documented or suspected, and some of the following: General variables

Fever (> 38.3 °C) Hypothermia (core temperature< 36 °C) Heart rate (90/min or more than two standard deviations above the normal

value for age) Tachypnea Altered mental status Significant edema or positive fluid balance (>20 mL/kg over 24 h) Hyperglycemia (plasma glucose>140 mg/dL or 7.7 mmol/L) in the absence of

diabetes Inflammatory variables

Leukocytosis (WBC count >12000/µL) Leukopenia (WBC count <4000/µL) Normal WBC count with greater than 10 % immature forms Plasma C-reactive protein more than two standard deviations above the

normal value Plasma procalcitonin more than two standard deviations above the normal

value Hemodynamic variables

Arterial hypotension (SBP <90 mmHg, MAP <70 mmHg, or an SBP decrease >40 mmHg in adults or less than two standard deviations below normal for age)

Organ dysfunction variables Arterial hypoxemia (PaO2/FiO2<300) Acute oliguria (urine output <0.5 mL/kg/h for at least 2 h despite adequate

fluid resuscitation) Creatinine increase >0.5 mg/dL or 44.2 µmol/L Coagulation abnormalities (INR >1.5 or aPTT >60 s) Ileus (absent bowel sounds) Thrombocytopenia (platelet count <100000/µL) Hyperbilirubinemia (plasma total bilirubin >4 mg/dL or 70 µmol/L)

Tissue perfusion variables Hyperlactatemia >1 mmol/L Decreased capillary filling or mottling

Severe sepsis

Sepsis-induced tissue hypoperfusion or organ dysfunction (any of the following thought to be due to the infection):

Sepsis-induced hypotension Lactate above upper limits laboratory normal Urine output <0.5 mL/kg/h for more than 2 h despite adequate fluid resuscitation Acute lung injury with PaO2/FiO2<250 in the absence of pneumonia as infection

source Acute lung injury with PaO2/FiO2<200 in the presence of pneumonia as infection

source Creatinine >2.0 mg/dL (176.8 µmol/L) Bilirubin >2 mg/dL (34.2 µmol/L) Platelet count <100000/µL Coagulopathy (international normalized ratio >1.5)

Current diagnostic criteria for ‘sepsis’ and ‘severe sepsis’, as proposed by the 2012 Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock (Based upon: Dellinger et al., 2013). aPTT: activated partial thromboplastin time; h: hours; INR: international normalized ratio; MAP: mean arterial pressure; SBP: systolic blood pressure; WBC: white blood cells


7

Figure 1.1 Examples of possible different immune profiles during sepsis. Based upon and modified from Hotchkiss et al., 2013.

If this subtle balance however sweeps over, a massive proinflammatory response known as the

systemic inflammatory response syndrome (SIRS) can be the result. On the other hand, this

inflammatory equilibrium might also lose its balance towards the opposite direction when a

massive anti-inflammatory response, termed the compensatory anti-inflammatory response

syndrome (CARS), characterized by immunosuppression and anergy of the immune system,

takes over (Bone RC, 1996). SIRS results in the destruction of healthy tissue, possibly

culminating into the succumbing of the patient to the cytokine storm (Dofferhoff et al., 1992),

whereas CARS makes the patient prone to secondary infections, the possible reactivation of

latent viral infections, or renders him or her unable to tackle the primary focus of infection,

leading to a state of severe immune suppression (Bone et al., 1997; Hotchkiss et al., 2013;

Hotchkiss et al., 2013).

During sepsis a decrease in the absolute number of several leukocyte subpopulations can be

observed in the blood and spleen. The apoptosis of CD4+ helper T cells, CD8+ cytotoxic T cells, B

cells and dendritic cells (DCs) is accelerated in septic patients, hence the term ‘sepsis-induced

immune cell apoptosis’ (Hotchkiss et al., 2013). This sepsis-induced immune cell apoptosis also


8

affects immune cells in the gut-associated lymphoid tissue (GALT) and lamina propria of the

human GI tract (Hotchkiss et al., 2000). Besides the decrease in absolute number, macrophages

will undergo endotoxin tolerance, meaning that the binding of LPS to its TLR4 receptor will not

induce the release of proinflammatory cytokines but on the contrary shall result in the

secretion of anti-inflammatory cytokines (Biswas et al., 2009). Similarly, the remaining DC

population produces more interleukine (IL)-10 (Poehlmann et al., 2009), a cytokine known for

its anti-inflammatory capacities, inducing T cell anergy. As such, human type 1 helper T cell

(Th1), Th2 and Th17 lineages are suppressed, and their effector functions declined (Pachot et

al., 2005) resulting in ‘T cell exhaustion’, characterized by a decreased production of interferon

γ (IFN-γ) and tumor-necrosis factor α (TNF-α) by T cells, and an increased expression of

programmed death 1 (PD1) and its ligand PD-L1 (PD-Ligand1) correlating with an increased risk

of mortality (Guignant et al., 2011), and of IL-7 receptor α chain on their cell surface

membranes (Hutchins et al., 2014). The regulatory T cell (Tregs) population, a cell subset that

suppresses the effector activity of CD4+ and CD8+ lymphocytes, undergoes the opposite

changes with an increase in their relative numbers that actively inhibits monocyte and

neutrophil function (Monneret et al., 2003). As such, an increased proportion of Tregs has been

described in septic patients (Monneret et al., 2003; Monneret & Venet, 2012; Leng et al., Crit

Care 2012), contributing even more to the already existing lymphocyte anergy (Venet et al.,

2009). There are currently several theories concerning the host immune response during

sepsis, as was clearly summarized in a recent extensive review by Hotchkiss and colleagues

(Hotchkiss et al., 2013). Numerous researchers demonstrated the simultaneous occurrence of

proinflammatory and anti-inflammatory events, where the net effect of both processes will

determine the actual immunological state the patient resides in (Hotchkiss & Karl, 2003;

Munford et al., 2001; Stearns-Kurosawa et al., 2011). Persistent activation of the innate

immune response will result in uncontrollable inflammation with subsequent organ failure

(Xiao et al., 2011).

Several cellular and biochemical determinants guide the immune reaction towards a

predominantly pro- or anti-inflammatory direction; some of these determinants are host-

dependent, such as comorbidities, age, drugs administered or genetical background (Angus et

al., 2001), whereas others are pathogen-dependent, more specifically the features of the

pathogen such as pathogen load, virulence and PAMPs, may determine the evolution of sepsis

as well (Angus et al,. 2013).


9

1.1.2 Current guidelines on the treatment of sepsis

In 2004 the Surviving sepsis campaign guidelines for management of severe sepsis and septic

shock were published with the original goal to increase awareness and improve outcome in

severe sepsis (Dellinger et al., 2004), with the most recent revised edition being published in

2013 (Dellinger et al., 2013). Recommendations assessed by means of the GRADE® system

were grouped in different categories, such as guidelines concerning initial resuscitation and

infection issues, hemodynamic support, adjunctive therapies, supportive therapies, prophylaxis

of deep vein thrombosis and stress ulcers, and special considerations in the pediatric

population. For the application by the bedside clinician these recommendations have been

reformulated into two ‘bundles of care’: an initial management bundle to accomplish within

the first 6 hours, and a management bundle for the ICU (Levy et al., 2010).

Due to the controversy surrounding the use of corticosteroids (Bollaert et al., 1998; et al.,

1999; Sprung et al., 2008) and the failure of activated protein C (Abraham E et al., 2005; Nadel

et al., 2007; Vincent JL et al., 2005; Eliézer et al., 2010; Ranieri et al., 2012), the current

management of sepsis consists mainly of supportive therapy again. The present-day

enhancement in the outcome of sepsis is due to general improvements in care, rather than the

usage of etiological therapies for sepsis, since there are hardly any interventions targeting the

modulation of the disrupted host response.

Moreover, during the past 30 years numerous clinical trials for new drugs that proved to be

successful in animal experiments, have failed in human clinical trials (Bone RC, 1996). Angus

and others attribute this discrepancy to the large between-species differences in host

response, and the fact that young, healthy mice subjected to a septic insult are not similar to

the generally elderly human patient that has been treated with antibiotics, resuscitation, renal

dialysis and so on (Rittirsch et al., 2007; Dyson & Singer, 2009; Angus et al., 2013; Pene et al.,

2015). Studies focusing on anti-TNF-antibodies and anti-LPS-antibodies in septic shock patients

did not result in the expected amelioration or increased survival in human studies, as targeting

only one mediator of the activated cytokine cascade in an ICU patient will not consequently

lead to the subduing of the entire inflammatory cascade (Bone et al., 1997; Venet et al., 2013).

Hope however should not be abandoned as new immunotherapies are being developed at the

moment (Hotchkiss et al., 2013), of which a concise overview can be found in table 1.4. The

prerequisite for successful immunotherapy is by many deemed to be the correct identification


10

of the phase of the immune response the patient is in, as stimulation of the immune system

would worsen the over-exuberant proinflammatory response if administered to a patient in

SIRS, and, conversely, inhibiting the immune response could be deleterious in the CARS phase.

Therefore, new clinical and laboratory biomarkers are studied to identify a patient in CARS

(table 1.3). The goal is to define a biomarker-guided therapy, which is based on an individual

approach to every sepsis patient (Hotchkiss et al., 2013, Venet et al., 2013).

Table 1.3 Clinical and laboratory biomarkers than could be utilized to identify patients in the CARS phase.

Innate immunity

HLA‑DR expression on monocytes ↓

Tumour necrosis factor (TNF) production by lipopolysaccharide-stimulated whole-blood cells ↓

Programmed cell death ligand 1 expression on monocytes ↑

Adaptive immunity

Persistent severe lymphopenia Reactivation of cytomegalovirus or herpes simplex virus infections Programmed cell death 1 expression on CD4+ or CD8+ cells ↑ Number of circulating regulatory T cells ↑ Interferon-γ production by T cells ↓ T cell proliferation ↓

Innate and adaptive immunity

Infections with relatively avirulent or opportunistic pathogens, such as Enteroroccus spp., Acinetobacter spp. or Candida spp.

Interleukin‑10/TNF ratio ↑

Delayed-type hypersensitivity response


11

Table 1.4 Possible new immunotherapies for sepsis in experimental studies

Molecule of interest

Physiological action

Type of study

Study result Reference

Recombinant human IL-7 (rhIL-7)

Endogenous IL-7 promotes T cell development and function and stimulates the differentiation of stem cells into lymphoid progenitor cells, resulting in an increased number of T cells

Lymphocytes from septic patients

rhIL-7 reverses crucial sepsis-induced defects in T cell function in vitro, which proves that the IL-7 signaling pathway remains operative during sepsis.

Venet et al., 2012

Mouse model of septic peritonitis (CLP)

rhIL-7 prevents the depletion in absolute cell counts of CD4+ and CD8+ T cells in spleen , increased cell proliferation in CD8+ T cells at 72 h after ‘cecal ligation and puncture’, improved survival in sepsis, increases expression of leukocyte adhesion markers in sepsis.

Unsinger et al., 2010

Antibodies to PD-1 / PD-L1

Programmed cell death 1 (PD-1) and programmed cell death ligand 1 (PD-L1) are co-inhibitory molecules. When PD-1 on circulating T cells interacts with PD-L1 (on macrophages, capillary endothelial cells and/or bronchial epithelial cells), T cell proliferation and function are impaired. PD-1 and PD-L1 can also be used as biomarkers of immune suppression.

Mouse model of septic peritonitis (CLP)

Anti-PD-1 antibody administered 24h after sepsis prevented sepsis-induced depletion of lymphocytes and DCs, blocked apoptosis and improved survival.

Brahmamdam et al., 2010

Blood from septic patients

In vitro blockade of the PD-1:PD-L1 pathway decreases apoptosis and increased IFN-γ and IL-2 production in septic patients.

Chang et al., 2014

Mouse model of fungal sepsis

Anti-PD-1 and anti-PD-L1 antibodies improved survival in primary and secondary fungal sepsis; both antibodies reversed sepsis-induced suppression of IFN-γ and increased expression of MHC II on antigen presenting cells.

Chang et al., 2013

SIVmac239-infected rhesus macaques

In vivo administration of rPD-Fc induced SIV-specific CD4 and CD8 T cell proliferation in the blood and gut, but did not alter plasma viremia.

Amancha et al., 2013


12

Antibodies to CTLA-4

Cytotoxic T-lymphocyte antigen-4 (CTLA-4) is a negative costimulatory molecule, expressed on the cell surface of T cells or intracellular in Tregs, that is up-regulated in sepsis and acts like PD-1 to suppress T cell function.

Mouse model of fungal sepsis

Blockade of CTLA-4 improved survival in fungal sepsis.

Chang et al., 2013

Mouse model of septic peritonitis (CLP) / Two-hit model of CLP (Candida albicans)

Anti-CTLA-4 therapy decreased sepsis-induced apoptosis but did not alter cytokine production in sepsis. Anti-CTLA-4 improved survival following CLP, even when animals were challenged with intravenous Candida albicans (two-hit model).

Inoue et al., 2011

Recombinant IFN-γ (rIFN-γ)

IFN-γ is an immunostimulatory and immunomodulatory cytokine critical for immunity against mainly viral and fungal infections. It activates macrophages and induces MHCII molecule expression. IFN-γ is mainly secreted by NK cells, and by CD8+ cytotoxic T cells.

In vitro monocyte function

In vitro restoration of monocytic function was demonstrated by enhanced LPS-induced TNF-α levels following treatment with IFN-γ.

Döcke et al., 1997

Case series: eight patients with invasive fungal infection

rIFN-γ treatment in patients with invasive fungal infection increased the HLA-DR expression in those patients with a baseline and enhanced the capacity of leukocytes from treated patients to produce proinflammatory cytokines involved in antifungal defence.

Delsing et al., 2014


13

1.2 The gut and inflammation

1.2.1 Permeability

The entire gastrointestinal tract is lined on the luminal side with epithelium, stratified

squamous non-keratinized in nature in the esophagus and anal canal, and columnar from the

stomach onwards covered with microvilli present in the small bowel. The epithelium is part of

the physical, chemical as well as immunological barrier that exists between the human body

and the possible harmful effects of the intraluminal contents. In the small intestine, its main

principal cell type is the enterocyte, next to the goblet cells, enteroendocrine cells, M-cells and

Paneth cells. The epithelium is of major importance in the pathogenesis of gut-derived sepsis,

as impairment of its permeability might predispose to translocation of whole bacteria as well as

bacterial compounds. In order to adequately fulfill its selective barrier function, the presence of

several specialized physical cell-cell junctions is mandatory. These however constitute only one

element in the intricate intestinal barrier.

1.2.1.1 Organization of the gastrointestinal epithelial barrier

The mucosal barrier consists out of extracellular and cellular components. The intestinal lumen

is colonized by commensal bacteria, that will impair the growth of pathogenic bacteria via

alterations in microenvironment (nutrients, pH,…) and competition for nutrients. Below, the

glycocalyx layer in direct contact with the GI epithelium contains antimicrobial products and

IgA secreted by enterocytes (Turner, 2009; Brandtzaeg et al., 2011). Finally, the intestinal

epithelium constitutes a true physical barrier that separates the intestinal lumen from the host.

Uptake of all compounds from the gastrointestinal lumen, including antigens and

microorganisms, can occur via either the paracellular way (via the tight junction complex), or

via transcellular pathways (endocytosis). The paracellular route is impermeable under normal

circumstances to protein-sized molecules, thus forming an efficient barrier against antigens

and macromolecules (Keita & Soderholm, 2010). Besides this passive transport, active

transcellular transport over the epithelium is also possible by means of energy-dependent

carrier mechanisms via transcytosis, believed to be of utmost importance in the antigen

surveillance in the GI tract and the pathophysiology of intestinal diseases (Ponda & Mayer,

2005), as alterations of the paracellular pathway and transcytosis are of particular importance

during intestinal disorders (Keita & Soderholm, 2010). The enterocytes that line the GI tract are

in close contact one another by means of the junctional complexes, consisting out of tight


14

junctions, adherens junctions, desmosomes and gap junctions (Farquhar & Pallade, 1963). The

tight junction complex can be found on the apical side of the lateral membrane of enterocytes,

and is of particular importance in regulating the selective paracellular transport of ions. Three

transmembrane proteins, namely occludins, claudins and junctional adhesion molecules or

JAMs, span the intercellular spaces. The intracellular domain of these occludins and claudins is

anchored to intracellular scaffolds, such as zonula occludens proteins, that will connect the

occludins and claudins to the cytoskeleton of the cell (Turner et al., 2009; Vanheel & Farré,

2013). Claudins represent a rate-limiting, size- and charge-selective route for the

transportation of ions (Van Itallie & Anderson, 2006). In the claudin-superfamily, at least 24

claudins have been described, whereas occludin is encoded by only one gene. Some claudins

are ubiquitously expressed in all cell types (e.g. claudin-1), whereas others are restricted to a

limited number of cells. In the rat colon, there is a strong expression of claudin-1, -3, -4, -5 and

-8, which all mediate restrictive passage of molecules, whereas claudin-2, -7 and -12 mediate

permeability (Camilleri et al., 2010; Veshnyakova et al., 2010). Adherens junctions can be found

basal from the tight junction complex, and consist out of a transmembrane protein (E-

cadherin) that once again connects to the intracellular cytoskeleton via β-catenin. They are of

importance in the cell-cell junction, but also in the regulation of cell proliferation and

polarization. Finally, desmosomes (or ‘maculae adherentes’) can be found scattered alongside

the lateral border of the cell, but usually basal from the adherens junction. Junctions exist out

of connections between transcellular desmoglein, desmocollin and desmoplakin with

intracellular keratin cytoskeletal filaments (Green & Jones, 1996). Finally, the family of gap

junctions (or macula communicans) consisting out of connexins allow for intercellular

connections between the cytoplasm of two cells, allowing for direct communication. CLP-

induced sepsis reduces the number of connexins and intercalated disk remodeling in the heart

(Morris et al., 2007).

1.2.1.2 The epithelium during inflammation

The exposure of an epithelium to LPS will result in loosening of the intestinal mucosal villi and

necrosis of the epithelial cells, as was shown by Liu et al. (2009) in a rat model of endotoxemia.

Endotoxemia was furthermore associated with decreased expression of occludin and zonulin-1

in the epithelium and subsequent increased intestinal permeability (Zhou et al., 2013). The


15

effects of LPS have recently been demonstrated to be mediated via the TLR4 - MyD88 – IRAK4

signaling pathway (vide infra)(Guo et al., 2015; Johnston & Corr, 2016).

Figure 1.2 Schematic drawing of the gastrointestinal epithelial barrier, with focus on the cell-cell adhesion molecules. Based upon and modified from Turner et al., 2009 and Vanheel & Farré, 2013.

Next, several immune cells recruited to a site of inflammation will secrete mediators that can

affect epithelial permeability (vide infra). Mucosal mast cells can release a myriad of

compounds that can alter transcellular and/or paracellular mucosal permeability, such as

proteases, tryptase, TNF-α and several interleukins, which has been extensively summarized by

Keita et al. (Keita & Soderholm, 2010). Tryptase and other proteases will exert their effects via

binding onto the PAR2-receptor on epithelial cells, increasing paracellular permeability and

facilitating bacterial translocation (Cenac et al., 2004; Jacob et al., 2005; Bueno & Fioramonti,

2008).

Finally, several proinflammatory cytokines have been shown to directly deteriorate the

epithelial barrier function. TNF-α and IL-13 can alter size- and charge-selectivity in gut

epithelium (Odenwald & Turner, 2013). IFN, TNF-α and IL-1β all decrease the transepithelial

resistance, which in its turn resulted in an increased expression of claudin-1 mRNA in a Caco-2

cell line, presumably an attempted repair response (Han et al., 2003; Van Itallie & Anderson,

2006). IFN-γ directly reduces the transepithelial resistance by affecting zonulin-1 levels, and

Desmosome

Claudin

Occludin

Zonulin

CytoskeletonTight junction

Adherens junction E-cadherin

β-catenin α-catenin-1

Desmoglein

Desmocollin

Desmoplakin

Keratin filament


16

influences paracellular as well as transcellular transport (Wang et al., 2005). TNF-α will

upregulate claudin-2 and induce apoptosis of enterocytes, resulting in increased paracellular

permeability (Gitter et al., 2000).

Gut permeability can be assessed either in vitro or in vivo. Concerning the in vitro

measurements, different techniques can be distinguished as was reviewed by Shen et al.

(2011): transepithelial resistance of the flux of macromolecules can be assessed in isolated

tissue segments using Ussing chambers, tight junction molecules can be studied at the mRNA

level, the protein level, or morphologically measured, and pore pathways (having a high

capacity, but being charge selective) versus leak pathways (low capacity) can be distinguished

using polyethylene glycol (Shen et al., 2011; Camilleri et al., 2012). In vivo murine experiments

typically comprise the gavage of a transit marker, such as FITC-dextran or Evans blue, with

subsequent determination of the concentration of these markers in the serum as an estimation

of gastrointestinal permeability (Lange et al., 1994).

1.2.2 Inflammation – the immune cells in the gut

The gastrointestinal tract hosts the biggest number of immune cells in the entire body (Liu et

al., 2009) where many interactions occur between the colonizing bacteria, antigens and

immune cells. Complex regulatory pathways maintain the homeostasis in a healthy

gastrointestinal tract, being aware of possible pathogens (Liu et al., 2007). The GI innate

immune system is the first line of defence against these luminal pathogens, and does so by

recognizing several macromolecules that can be divided into the two major categories

mentioned before, namely the pathogen-associated molecular patterns (PAMPs) and damage-

associated molecular patterns (DAMPs) (Pacheco et al., 2010; Izcue et al., 2006). Both PAMPs

and DAMPs are recognized by Toll-like receptors (TLR) (Boeckxstaens & De Jonge, 2009; De

Winter & De Man, 2010). Peyer’s patches, aggregated lymph nodes found in the ileum and

colon, and isolated follicles with M cells, consisting out of many microfolds, form the gut-

associated lymphoid tissue (GALT) and translocate these antigens from the lumen to

underneath the epithelium where leukocytes, mostly dendritic cells, reside awaiting an

antigenic stimulus (Liu et al., 2009; Izcue et al., 2006) (figure 1.3). Several other cell populations

in the epihtelium possess the capacity to react to antigens as well, though somewhat less

efficient (MacDonald et al., 2011). As such, the innate lymphoid cells (ILC) represent a cell

population pertaining to the innate immune system, characterized by classic lymphocyte


17

morphology, exhibiting a similar cytokine expression and effector function as T helper cells;

ILCs however are not antigen specific. They contribute to immunity, inflammation and tissue

homeostasis. Their characteristics and functions are extensively described in a recent review

(Artis & Spits, 2015). Intraepithelial lymphocytes (IEL) on the other hand represent a subset of

immune cells belonging to the GALT. Upon encountering an antigen, IELs will release cytokines

and induce cytotoxicity upon the encountering of antigens without previous priming.

Dendritic cells, which can be found throughout the intestine, can detect pathogens through a

number of receptors that recognize several motifs, of which the TLR are of utmost importance

(Flores-Langarica et al, 2005; Bekiaris et al., 2014). When activated by an antigen, DCs residing

in the lamina propria of the gut go through a process of maturation and change their surface

markers according to the antigen presented, upregulating membrane markers such as the

major histocompatibility complex II (MHCII) and co-stimulatory molecules such as CD80 and

CD86, ultimately leading to the secretion of different cytokines that will polarize naïve

(unstimulated) T-helper lymphocytes into different subgroups such as Th1, Th2, Th17 or Tregs

(Pacheco et al., 2010; Nijhuis et al., 2010; Engel et al., 2010; Bogunovic et al., 2009; Mann et

al., 2013). DCs that lack the CX3C chemokine receptor 1 (CX3CR1) react to antigens and migrate

to the local draining lymph nodes in the peritoneum. This allows presentation of the antigen to

T lymphocytes and differentiation of these cells into regulatory T cells (Tregs), resulting in a

negative regulation of the immunological response and maintenance of gut homeostasis

(Schulz et al., 2009). On the other hand, CX3CR1 positive DCs express a larger amount of co-

stimulatory molecules, resulting in higher amounts of cytokines due to the activation of

effector T cells, ultimately leading to a proinflammatory response (Nijhuis et al., 2010; García-

López et al., 2001; Merad et al., 2013; Scott et al., 2011; del Rio et al., 2010). Intestinal

epithelial cells can modify this process by releasing several factors such as transforming growth

factor-β (TGF-β), an important anti-inflammatory molecule secreted by connective tissues. This

results in upregulation of the cell surface marker CD103 on the DC, which promotes Treg

differentiation and thus bowel homeostasis (Pacheco et al., 2010; MacDonald et al., 2011). The

DC modulation via these receptors has been studied in rodents (Johansson-Lindbom et al.,

2005); whether this is also applicable in humans, remains to be studied (Schulz et al, 2009).

Mast cells are found near the blood vessels entering the bowel wall, often in pairs, and in close

association with afferent nerves facilitating bidirectional communication between both cell

types (Boeckxstaens & de Jonge, 2009; De Winter & De Jonge, 2012). Inside their granules they


18

store preformed mediators, such as histamine, tryptases, chymase, heparin and chondroitin

(MacGlashan, 2008). Other mediators are synthesized upon activation, such as leukotrienes,

cytokines, growth factors, TGF-β and platelet activating factor (PAF). Its activation often occurs

in interaction with the specific antibody IgE bound to the cell membrane, but also cytokines,

immunoglobulins, hormones, neuropeptides and several other substances can activate mast

cells, secreting powerful mediators such as histamine, serotonin, TNF-α and PAF, that are

furthermore able to influence the enteric nervous system (Van Nassauw et al., 2007). The

importance of mast cells in the GI immune regulation has been demonstrated almost two

decades ago, for better or for worse. Mast cell deficient mice demonstrate an impaired

immunity and increased mortality in a mouse model of septic peritonitis (Echtenacher et al.,

1996) as the initial release of proinflammatory cytokines will result in the recruitment of

bactericidal neutrophils into the gut wall (Thakurdas et al., 2006; Sutherland et al., 2008;

Maurer et al., 2004; Piliponsky et al., 2008); however, more recent research suggests that

systemic mast cell degranulation increases mortality during the same polymicrobial mouse

model (Seeley et al., 2011), and mast cell deficient mice are protected from intestinal

muscularis inflammation in a mouse model of intestinal manipulation (de Jonge et al., 2004).

Finally, macrophages are immunocytes that can be roughly subdivided in two categories,

namely the resident muscularis macrophages and the mucosal macrophages. The latter

category resides permanently in an anti-inflammatory state that is skewed towards a pro-

inflammatory one upon encountering an antigen or during inflammation (Platt & Mowat, 2008;

Smith et al., 2005; Zhang & Mosser, 2008). The former population is by far the most studied

one, and is of particular importance in the pathogenesis of (post-operative) ileus (vide infra).

During sepsis, the binding of LPS on the TLR4-receptor on the macrophage’s cell surface will

induce the release of inflammatory cytokines, nitric oxide by the upregulation of inducible

nitric oxide synthase (iNOS), prostaglandins and chemokines among others, resulting in the

recruitment of circulating leukocytes towards the inflammatory focus and subsequent further

deterioration or exacerbation of inflammation (Overhaus et al., 2004) (Murray & Wynn, 2011).

A graphical representation of the different neuroimmune players in the gut during

inflammation is given in figure 1.3.

T cellzone

DC

IL-17

Th17

venule

ILCILC

venule

B

Th0B

M1

Th0

Th0

venule

IFN-γIL-2

IFN-γ,IL-2

IL-12, IL-18, IL-27

IL-23

mesenteric lymph node

Th0

Th0

IFN-γIL-2

TGFβ,IL-10IL-18, IL-27

IL-6, IL-1β, IL-23

splanchnic afferent nerves

gut lumen

villus

Peyer’spatch

villus

afferent lymphaticvessel

commensal gut flora

efferent lymphaticvessel

mucus

IEL

IEL

IEL

IEL

IEL

M cell

M cell

panethcell

mucus

epithelium

IL-18, IL-27

IFN-γIL-2

IL-6, IL-1β, IL-23

IFN-γIL-2

Increased mucosal

permeability

LPS, DAMPS, PAMPS

TLR4

pathogens

MΦ

M1

M1

MΦ

MΦ

DC

DC

DC

DC

MCP-1

ICAM-1

Th1Th17

Th1

Th1

Th1

Th1

DC

M1

Th0

DC

DC

DC

CX3XR1

CGRP, TRPV-1,Substance P

Monocytes, neutrophils

Th1

DCNO,

prostaglandins

Mast cells

Histamine, serotonine, proteases

Tissue-resident muscularis MΦ

Treg

CD103

DC


20

Figure 1.3 (previous page) The gastrointestinal tract during sepsis. A complex interplay takes place between host and microbial antigen, during which binding of lipopolysaccharide upon its TLR4-receptor, expressed on the cell surface of antigen presenting cells such as macrophages, will result in the release of nitric oxide, prostaglandins, interleukins and several chemokines. Nitric oxide and prostaglandins will directly target the smooth muscle cell, resulting in impaired gastrointestinal motility. The release of chemokines, such as MCP-1, will attract several other leukocyte populations such as monocytes and neutrophils to the site of insult, exacerbating inflammation. Dendritic cells present in the lamina propria of the gut wall will sample microbial antigens, translocate towards Peyer’s patches and mesenteric lymph nodes, and induce a proinflammatory response by releasing IL-23, IL-18 and IL-27 amongst others, thereby skewing the population of naïve T cells towards the IFN-γ and IL-2 secreting Th1 subtype. Concomitantly, mast cells lying adjacent to neuronal endings will release several mediators, such as histamine, serotonin and proteases that can directly attenuate epithelial integrity and increase mucosal permeability facilitating the translocation of noxious bacteria and microbial antigens. Finally, simultaneous activation of afferent neurons by microbial peptides and proinflammatory cytokines will result in the release of several neurotransmitters that will contribute to inflammation and impaired gastrointestinal motility. This schematic drawing is probably still an oversimplification of what happens at the GI tract level during sepsis, as our knowledge is ever increasing. B: B cell; CGRP: calcitonin gene related peptide; DAMP: danger associated molecular pattern; DC: dendritic cell; GI: gastrointestinal; IFN-γ: interferon-gamma; IEL: intraepithelial lymphocyte; IL: interleukin; ILC: innate lymphoid cell; LPS: lipopolysaccharide; M cell: microfold cell; MCP-1: monocytes chemoattractant protein-1; Mφ: macrophage; M1: M1 macrophage, proinflammatory macrophage subset; NO: nitric oxide; PAMP: pathogen associated molecular pattern; Th1: T helper 1 lymphocyte; Th17: T helper 17 lymphocyte; TRPV1: transient receptor potential cation channel subfamily V member 1 Based upon and modified from Heylen et al., 2014, depicting the different immune players during inflammatory bowel disease in the colon wall.


21

1.2.3 Innervation

The afferent nervous system conveys sensory information concerning, but not limited to,

gastrointestinal homeostasis and disturbances of this balance towards the higher brain areas.

Below, we will first describe the neuroanatomy of the sensory pathways of the GI tract. Many

excellent reviews were published on the intricate neuroanatomy of the GI tract, as can be read

in the following paragraphs.

1.2.3.1 Intrinsic innervation of the gut

One might consider the gastrointestinal tract to be a unique organ system as it is entirely

innervated by the enteric nervous system (ENS), which theoretically permits the GI tract to

function in an independent way without modulation from the central nervous system and

generate autonomous GI motility and digestive functions. The ENS comprises the neuronal

bodies that are arranged in two main plexuses, namely the myenteric or Auerbach’s plexus

located between the circular and longitudinal muscle layer lining the GI wall, and the

submucosal or Meissner’s plexus situated in the connective tissue beneath the mucosa (Lomax

et al., 2005). The former one generally modulates GI motility, whereas the latter one is

predominantly in control of the secretion and absorption of nutrients from the GI tract.

Extensive synapses exist between the neurons in both plexuses. The neurons present in both

plexuses are however by far outnumbered by the amount of glial cells, which are

morphologically and functionally similar to the astrocytes in the central nervous system (Lomax

et al., 2005), displaying an intimate proximity to and communication with the enteric neurons,

and participating in the metabolic maintenance and gut homeostasis of the GI ENS. Glial cells

are of particular importance in maintaining the intestinal mucosal barrier integrity, and can

participate in inflammation as they are able to secrete cytokines and chemokines in response

to an inflammatory focus (Neunlist et al., 2014; Sharkey, 2015; Coelho-Aguiar et al., 2015).

Finally, the ENS would not be able to perform its autonomous function were it not for the

presence of Interstitial Cells of Cajal (ICC). Often termed the ‘pacemaker of the gut’, this cell

population drives the generation of slow waves in the entire GI tract, serving as a connecting

link between myenteric neurons and smooth muscle cells and ultimately resulting in rhythmic

contractions of the smooth muscle cells (Ward et al., 2000).


22

1.2.3.2 Extrinsic innervation of the gut

The intrinsic nervous system of the gut and its functions can be modulated by the central

nervous system via the extrinsic nervous system; this extrinsic nervous system can be

subdivided into two separate entities, namely the (ortho)sympathetic and parasympathetic

nervous system, that exert complimentary functions at the GI tract level (Vermeulen et al.,

2014). In both categories, afferent as well as efferent fibers are present, coursing together

(Brookes et al., 2013).

The preganglionic efferent nerves of the orthosympathetic nervous system, which are

cholinergic in nature, are located in the greater thoracic and the lesser lumbar splanchnic

nerves, connecting to the prevertebral ganglia that are located in proximity to the major

arterial vessels in the abdomen (celiac ganglion and superior mesenteric ganglion in proximity

to the superior mesenteric artery and the inferior mesenteric ganglion adjacent to the inferior

mesenteric artery). The superior and inferior mesenteric ganglia are connected by means of

intermesenteric nerves, whereas the inferior mesenteric ganglion has projections, the so-called

hypogastric nerves, coursing aborally towards the pelvic-hypogastric plexus. From the

prevertebral ganglia onwards, postganglionic adrenergic nerves project towards the entire GI

tract through the greater splanchnic nerve (foregut), the lesser splanchnic nerve (midgut) and

the lumbar splanchnic colonic and hypogastric nerves (colon)(figure 1.4).

The afferent fibers of the sympathetic nervous system course through the splanchnic afferent

nerves together with their efferent counterparts. The cell bodies of these fibers are located in

the thoracolumbar dorsal root ganglia (DRG) adjacent to the spinal cord (Blackshaw et al.,

2007). From there on they course towards the prevertebral ganglia and further on, synapsing

with efferent fibers on the way at multiple levels. The sympathetic afferent nerves are

presumed to transfer the main bulk of the nociceptive information towards the central nervous

system, in contrast to the parasympathetic afferent fibers that mainly transmit modalities

concerning normal physiology and homeostasis (Brierley et al., 2004).

The parasympathetic nervous system comprises the vagal nerve, which innervates nearly the

entire GI tract with the exception of the distal colon and colorectum, and the pelvic nerve,

which supplies the latter two organs. Concerning the parasympathetic nervous system, we

know that the amount of afferent fibers far outnumbers the efferent ones, as over 80% of the

vagal neurons are afferent in nature (Berthoud & Neuhuber, 2000). The preganglionic efferent

parasympathetic nerves have their cell bodies situated in the dorsal motor nucleus of the vagus


23

core, whereas the pelvic nerve has its cell bodies in the sacral spinal cord. In contrast to the

sympathetic efferent postganglionic fibers are the parasympathetic ones cholinergic in nature.

The preganglionic parasympathetic fibers course all the way into the bowel wall where they

synapse onto the intramural postganglionic efferent fibers. The afferent fibers of the vagal

parasympathetic nerve have their cell bodies located in the nodose ganglion, from where they

project centrally to the nucleus tractus solitarius in the brainstem, which in itself is in close

proximity to the dorsal motor nucleus of the vagal nerve, thus facilitating rapid vagovagal

reflexes. The pelvic afferents, in contrast to the vagal ones, have their cell bodies located in the

lumbosacral DRG much alike the sympathetic afferent nerves. Vagal and pelvic afferents

predominantly mediate physiological sensations and form the afferent branch of several GI

reflexes, such as the cologastric inhibition, satiety, feeling of hunger and fullness.

Between rodents and men, it is important to know that men have 12 spinal segments, and thus

also 12 DRGs, whereas in mice 13 thoracic segments are present. Humans have five lumbar

segments, whereas rodents have 6. The pelvic afferent cell bodies are found in DRG L6-S1 in

rodents, whereas they are primarily situated in DRG S2-S4 in men.

Figure 1.4 The autonomous nervous system that innervates the gastrointestinal tract. Efferent information from the (ortho)sympathetic nervous system (left side of the image) is wired via the celiac ganglion (CG), superior mesenteric ganglion (SMG) and inferior mesenteric ganglion (IMG). The afferent fibers course alongside their efferent counterparts, however the cell bodies from the afferent sympathetic nervous system are located in the dorsal root ganglia. The afferent parasympathetic vagal nerve has its cell bodies in the nodose ganglion (GN), whereas the pelvic nerve courses via the pelvic ganglion (PG).

OS


24

The goal of the extrinsic afferent neurons is to provide sensory information on chemical and

physical stimuli from the local GI tract level (e.g. stimuli and information with regard to

homeostasis, osmolarity and acidity of the GI luminal contents, proprioceptive input from the

viscera, et cetera) towards the central nervous system. The extrinsic afferent neurons that

originate in the gut can be classified in multiple ways, either based upon their morphologic

characteristics, their responses to mechanical or chemical stimuli such as capsaicin, bradykinin

or serotonin, the expression of cell surface receptors, the types of neurotransmitters they

secrete, or whether they process information on normal physiology or noxious stimuli (for

extensive review see Brookes et al., 2013; Knowles & Aziz, 2009). Afferent neurons can be

subdivided in five different classes, based upon the localization of their terminal endings in the

GI tract wall lining (and to a lesser extent based upon their function) (Brookes et al., 2013;

Knowles & Aziz, 2009) (figure 1.5). The first class is termed the ‘intraganglionic laminar’

afferents (type I), with their endings predominantly located in the myenteric ganglia between

the circular and longitudinal layer. It is assumed that they predominantly detect harmless

mechanical distortion. Type II afferents, also termed the ‘mucosal afferents’, have their endings

below the epithelium and are sensitive to enteroendocrine cell mediators. Type III or

‘muscular-mucosal’ afferents have mechanosensitive endings located between the muscularis

mucosae and the lamina propria, and they detect both muscular as well as mucosal distortion.

Type IV afferents have intramuscular endings (circular and/or longitudinal muscle layer) and

primarily detect mechanical distortion. Finally, the type V ‘vascular afferents’ have their

endings on blood vessels, and predominantly detect a variety of chemical mediators. A sixth

class can be defined as well, namely the ‘viscerofugal’ afferents, with their cell bodies located

in the myenteric ganglia. They directly project out of the gut, towards the prevertebral ganglia

with postganglionic efferent fibers, thus forming prevertebral reflex pathways (Szurszewski et

al., 2002). All five categories project to the sympathetic as well as the parasympathetic nervous

system, however the contribution of each type may differ significantly. The first four types are

predominantly distributed in the vagal and pelvic (parasympathetic) pathway, and to a lesser

extent in the thoracolumbar splanchnic (sympathetic) nerves, whereas vascular (and enteric

viscerofugal fibers) predominantly project towards the splanchnic thoracolumbar and the

lumbosacral afferent nerves. As such, the first four types primarily wire information concerning

normal physiology and homeostasis towards the brain, whereas the latter two are mainly

involved in nociception and detection of harmful stimuli. Another commonly used classification


25

of afferent neurons is based on their mechanosensitive response to three different mechanical

stimuli, namely probing, mucosal stroking and circular stretch (Brierley et al., 2004). In short, all

visceral afferents are responsive to probing, however only the mucosal and muscular-mucosal

ones are sensitive to mucosal stroking by means of calibrated von Frey hairs, and mucosal

afferents were not responsive to circumferential stretch (Brierley et al, 2004).

Figure 1.5. Schematic drawing of the localization of the five subsets of afferent neurons in the gastrointestinal tract. Based upon and modified from Brookes et al., 2013; Knowles & Aziz, 2009.

Finally, the splanchnic afferent neurons are for the remainder of this thesis predominantly

classified based upon their mechanosensitivity, i.e. their firing rate in response to distension

(Blackshaw et al., 2002; Grundy & Schemann, 2007; Booth et al., 2008; Deiteren et al., 2015).

As such, afferent fibers can be subdivided in four categories. Low threshold (LT) fibers typically

exert increased discharge at the lowest distension pressures, with a rapidly increasing firing

rate in response to small increases in the intraluminal pressure, with saturation in firing rate at

relatively low distension levels. Occasionally, one can see a small drop in this final stable firing

rate during noxious distensions. High threshold (HT) fibers on the other hand display an

increased firing rate at distension pressures rising above 20 mmHg, thus primarily detecting

mucosa

serosa

longitudinal muscle layer

circular muscle layer

mesentery

mucosal afferent

muscular –mucosal afferent

intraganglionic afferent muscular

afferent

vascularafferent


26

noxious pressures in the GI lumen. Wide dynamic range fibers (WDR) demonstrate an

increased somewhat linear firing rate during the entire ramp distension; they are presumed to

sense the intraluminal filling status, as well as pain and discomfort. Finally, mechanoinsensitive

afferent (MIA) nerve fibers typically do not respond to ramp distensions, but they can become

mechanosensitive during a state of inflammation (Blackshaw et al., 2002; Feng & Gebhart,

2012) (figure 5.3).

1.2.3.3 Ascending spinal pathways and central processing

Information recorded by the afferent neurons is wired towards the central nervous system, but

not before being modulated at several relay stations. As such, not all information that is wired

centrally is perceived consciously by the mind. Extrinsic splanchnic afferents will synapse onto

their second order neuron in lamina I of the dorsal horn of the thoracolumbar and lumbosacral

part of the spinal cord (Vermeulen et al., 2014; Mayer, 2011). From there on, sensory

information of visceral origin is transmitted via the ascending spinoreticular,

spinohypothalamic, spinomesencephalic and spinothalamic tracts (Anand et al., 2007) towards

the brain. The former three mainly contribute to regulate physiological functions and

information conveyed via these three pathways is not consciously perceived, whereas the

latter tract is the main mediator of pain sensation, cold, heat, and contributes to the sensation

of touch. The spinothalamic tract projects to the thalamus, which is a major relay station where

somatic as well as visceral inputs converge. The thalamus will then process this nociceptive

information, and relays it via a third order neuron to several somatosensory (prefrontal)

cortical regions, and to the anterior cingulated cortex (ACC), the mid cingulate cortex, insula

and amygdala. In the cortical regions, stimuli can be perceived consciously as being ‘painful’.

1.2.3.4 Central descending inhibitory pathways

Several supraspinal centers can modulate the afferent sensory input from the periphery. The

most renowned region is probably the periaquaductal gray (PAG) one, located in the midbrain,

as it receives input from the amygdala, which in its turn receives projections from the ACC. The

endogenous opioids are key mediators via which the descending inhibitory modulation is

wired, as the ACC contains a high amount of opioids (Gebhart et al., 2004; Mayer et al., 2005).

The effects of opioids are mediated via their three main receptor types: the µ-, κ- and δ-opioid

receptor, G-protein coupled receptors that, once activated, will mediate the presynaptic


27

release of GABA. Opioid receptors are scattered throughout the central nervous system on the

one hand, and through the entire GI tract on the other. In the GI tract, activation of opioid

receptors results in the inhibition of gastric emptying, induction of stationary motor patterns

and impairment of the peristaltic activity of the GI tract (Holzer et al., 2009). Their endogenous

ligands (enkephalins) interact with the enteric nervous system regulating GI motility and

secretion. In the brain, µ-receptors are predominantly present in the PAG area, the dorsal horn

of the spinal cord, and to a lesser extent in the amygdala and the nucleus of the solitary tract.

In the GI tract, µ-opioid receptors are most notorious for their established role in the

development of post-operative ileus and opioid-induced bowel dysfunction (Mosinska et al.,

2016). They are primarily present in the myenteric and submucosal plexus, where their

neurons project to the circular muscle layer, and in neurons of descending pathways. Besides,

they are present on the cell surface of immune cells in the lamina propria (Sternini et al., 2004).

The δ-opioid receptors are highly distributed in the forebrain, striatum and cortex, whereas the

majority of δ-receptors in the GI tract is located in the enteric ganglia and submucosal neurons,

and in the latter population they are often coexpressed with neuropeptide Y. Kappa-opioid

receptors can be found in the spinal cord (Cahil et al., 2014); data on their GI distribution are

for the major part obtained from animal research, showing a widespread distribution in rat ICC,

and in submucosal and myenteric neurons (Sternini et al., 2004). Mu-opioid receptors have

been shown to undergo endocytosis in neurons in animals that underwent abdominal surgery,

inducing a delay in GI transit (Sternini et al., 2004; Patierno et al., 2004).

1.2.4 Impaired gastrointestinal motility - ileus

Ileus can be defined as a transient decrease in gastrointestinal motility (Carroll et al., 2009).

The clinical presentation often comprises absence of bowel sounds, lack of flatus and/or

defecation, inability to tolerate oral feeding, abdominal discomfort and/or distension

(Boeckxstaens & De Jonge, 2009). Typically the entire gastrointestinal tract is involved,

however not all segments are equally affected: the small bowel motility is least impaired,

whereas colonic motility is the most affected (De Winter et al, 2012). Therefore, clinical

parameters defining colonic motility, such as flatus and defecation, are often used as outcome

parameters in clinical practice (Boeckxstaens & De Jonge, 2009; Alfonsi et al., 2014). As already

mentioned before, two clinical entities often result in ileus, namely surgical manipulation of the

gut (i.e. post-operative ileus) or sepsis.


28

1.2.4.1 Post-operative ileus

The occurrence and pathophysiology of impaired gastrointestinal motility, i.e. ‘ileus’, has been

extensively studied in the post-operative setting by many renowned research groups. Much of

the aforementioned players in the first paragraphs of this chapter contribute to the

pathophysiological changes that are observed during post-operative ileus (POI).

The immediate manipulation of the gut by the surgeon will result in neuroimmunological

changes at the GI tract level, that typically manifest themselves in a biphasic manner, as was

originally described first by Bueno et al. (1978). Several causative factors were suggested to

play a role in the development of post-operative ileus, such as the involvement of sympathetic

reflexes, inflammation, use of anesthetics and humoral factors (Livingston & Passaro, 1990;

Bueno et al., 1978; Kalff et al., 1998; Bauer & Boeckxstaens, 2009). In a rat model of post-

operative ileus, Kalff et al. (1998) were able to demonstrate that an increasing degree of

mucosal damage was present in conjunction with increasing myeloperoxidase-activity

corresponding with a more aggressive surgical intervention. Myeloperoxidase-positive cells

were predominantly present throughout the muscularis layer. One day following the surgical

protocol, the infiltrate consisted mainly out of polymorphonuclear cells, monocytes and mast

cells, whereas at the later time points more matured macrophages were detected.

Bauer postulated that at least three mechanisms are involved in the manipulation-induced POI,

namely the neurogenic, inflammatory and pharmacological mechanisms (Bauer & Boeckstaens,

2004). The manipulation in itself will initially lead to a neutrally mediated interruption in GI

motility (Boeckxstaens & de Jonge, 2009). This initial neural inhibition in motility mediated via

efferents is adrenergic in nature. Capsaicin-sensitive afferents are also involved in this

mechanism, and the application of capsaicin on the celiac ganglion partially prevented the

occurrence of postoperative gastric ileus (Plourde et al., 1993; Zittel et al., 1994; Zittel et al.,

1994). The involvement of these afferent fibers results in the central release of corticotrophin-

releasing factor with subsequent inhibition of the entire GI motility via the splanchnic nerves

(the adrenergic inhibitory mechanism). In addition, Boeckxstaens et al. (1999) described that a

second non-adrenergic, vagally mediated pathway contributed to this initial phase. Finally,

nitrergic and VIP-ergic neurotransmission also plays a pivotal role in this neurogenic phase, as

blockade of nitrergic neurotransmission as well as a prostaglandin inhibitor combined with L-

nitroarginine reversed the inhibition of the transit induced by POI in rats (De Winter et al.,

1997). A VIP-antagonist significantly enhanced transit in the same animal, whereas a VIP-


29

agonist further aggravated transit (De Winter et al., 1998). Three to four hours following

intestinal manipulation, an influx of leukocytes can be seen into the manipulated segments,

leading to a sustained and clinically relevant ileus (Kalff et al., 1999). The innate immune

response is particularly involved in this phenomenon. The manipulation of GI tissue will result

in the degranulation of peritoneal mast cells and the subsequent release of histamine and

other mediators in the bowel wall, leading to an increase in mucosal permeability that might

predispose to the translocation of (whole) bacteria and bacterial compounds. Consequently,

binding of LPS to the TLR4-receptor expressed on the cell surface of immune cells will result in

the release of proinflammatory cytokines, chemokines and nitric oxide, all together leading to a

recruitment of polymorphonuclear cells into the bowel wall and subsequent impaired GI

motility. Several experimental studies, in rodents as well as men, thus targeted the mast cell in

an effort to prevent manipulation-induced ileus. Using mast cell stabilizing agents prevented

the delay in GI emptying following POI in mice, whereas mutant mice that lack mast cells do

not develop an inflammatory infiltrate (de Jonge et al., 2004). The degranulation of mast cells is

thought to be provoked by the release of neuropeptides, such as substance P or CGRP from

afferent neurons. The pivotal role of the mast cell has been confirmed in human subjects,

where an immediate rise in tryptase levels could be detected in the peritoneal lavage fluid of

women undergoing open gynecological surgery, followed by an increase in IL-6 and IL-8 levels

(The et al., 2008), whereas administration of the mast cell stabilizer ketotifen ameliorated POI-

induced impaired gastric emptying (The et al., 2009). However also conflicting data have been

published on the role of mast cells in the pathogenesis of POI, as they trigger epithelial

dysfunction and barrier disruption, however they might not be mandatory for the development

of POI as such (Snoek et al., 2012; Gomez-Pinilla et al., 2014). Intestinal handling can activate

residential macrophages by the release of DAMPS, the facilitation of bacterial translocation and

microbial products, and the infiltration of LPS and cytokines (Boeckxstaens & de Jonge, 2009).

The subsequent release of proinflammatory cytokines, prostaglandins, nitric oxide via iNOS

upregulation, chemokines and cell adhesion molecules all represent interesting therapeutic

targets in the approach to POI, and several research papers already verified the validity of this

therapeutic approach in animal models of sepsis (Bauer & Boeckxstaens, 2004; Kreiss et al.,

2004; De Winter et al., 2002). More recently, the role of other immunotypes has been

acknowledged in the pathophysiology of POI. Not only cells pertaining to the innate immune

system, but also members of the adaptive immune system contribute to the pathogenesis of


30

POI, and more specifically of the ‘dissemination’ of inflammation throughout the entire GI

tract. Engel et al. showed that, upon intestinal manipulation, Th1 cells are primed to secrete

IFN-γ upon encountering IL-12 secreted by CD103+CD11b+ DCs. IFN-γ will subsequently induce

macrophages to upregulate iNOS. A subpopulation of these Th1 cells will also migrate to

unmanipulated intestinal tissue via the bloodstream, subsequently leading to impaired GI

motility in distant GI tissues (Engel et al. 2010; Koscielny & Kalff, 2011) This gastrointestinal

‘field-effect’, as it was called by Koscielny and colleagues, was furthermore shown to be

dependent upon CCR7, a chemokine that is predominantly responsible for the migration of

activated dendritic cells and T cells; prior to this proceeding they proved that intestinal

manipulation, as well as challenge with LPS, significantly increased the number of DCs in the

mesenteric lymph nodes and muscularis externa with concomitant upregulation of

costimulatory molecules and migratory markers on their cell surface, thus allocating a pivotal

role to the dendritic cell in the initiation of POI (Koscielny et al., 2006; Koscielny et al., 2011).

This field effect in POI could be abrogated in mice that lacked mesenteric lymph node and GALT

tissue, resulting in a significant decrease in inflammatory parameters and improvement in GI

transit times (Koscielny et al., 2013). T cells have been shown to presumably play a role in the

vagal anti-inflammatory pathway. Their comprehensive role however is yet to be fully

elucidated, as the group of Matteoli et al. showed that the anti-inflammatory response elicited

by vagus nerve stimulation (VNS) was independent of splenic T cells in a POI mouse model, and

that cholinergic efferent neurons were in close contact with macrophages residing in the

muscularis externa (Matteoli et al., 2014; Cailotto et al., 2014).

Furthermore, impaired neurotransmission has also been shown to play a role during the more

prolonged inflammatory phase, as POI-induced intestinal inflammation results in impaired

intestinal nitrergic neurotransmission due to decreased NOS-activity in nitrergic nerves

(Vanneste et al., 2008). Nitrite, a compound that has demonstrated ameliorating effects on

ischemia/reperfusion injury, displayed anti-inflammatory properties in a mouse model of POI

when administered exogenously (Cosyns et al., 2015). Finally, carbon monoxide released from

GI neurons displays antioxidant as well as anti-inflammatory effects (Miller et al., 2001;

Colpaert et al., 2002), and its administration reduces the development of POI (De Backer et al.,

2009). Successful efforts have been made in order to standardize an animal model of POI,

obtaining a reproducible degree of GI motility impairment that allows researchers to

consistently study POI. As such, the small bowel of the rodent animal can be positioned


31

between cotton buds and subjected to a standardized pressure for a predefined time, with

increasing pressures resulting in a more pronounced disturbance in GI motility (van Bree et al.,

2012). Data obtained in animal research should be extrapolated to the human setting with

great caution, as several differences already exist between mouse and rats concerning the

pathophysiology of POI: in rat, the ileus is predominantly mediated via the influx of leukocytes

and the release of nitric oxide, whereas the murine ileus is often of a more intense nature and

involves local cytokine-responses instead of leukocyte infiltration (Schmidt et al., 2012).

1.2.4.2 Septic ileus

Most of the pathophysiological mechanisms that are of importance in the development of POI,

also imply to the development of sepsis-induced ileus (Overhaus et al., 2004; De Winter & De

Man, 2010). Most knowledge has been derived from the injection of LPS in laboratory animals,

however this model certainly has its limitations despite being an easy and elegant one. In short,

an upregulation of inflammatory mediators, such as IL-6, IL-1β, MCP, iNOS and COX-2 is

observed during sepsis in isolated muscularis segments from animals challenged with a

polymicrobial abdominal sepsis, resulting in impaired motility. Furthermore was an influx of

monocytes and altered morphology of macrophages witnessed in an animal model of

polymicrobial abdominal sepsis (Overhaus et al., 2004). Much remains to be elucidated

concerning the role of other immunocytes in the development of sepsis-induced ileus.

1.2.4.3 The contribution of ileus to sepsis

The occurrence of a delayed gastrointestinal transit, either by surgical manipulation or by the

use of opioids or other pharmacological compounds amongst others, will lead to the stasis of

impaired GI transit, with subsequent facilitated translocation of luminal compounds and

antigens through the bowel wall. Normal intestinal peristalsis is pivotal in this regard, as

delayed GI transit times are associated with an increased risk of bacterial translocation, a

feature that has been predominantly studied in the clinical setting of cirrhosis or pancreatitis as

was already demonstrated two decades ago (Wang et al., 1996; Erbil et al., 1998; Samel et al.,

2002; Zhang et al., 2003).


32

1.3 The gut during sepsis

1.3.1 Bacterial translocation

The role of the gut during sepsis is quite complicated as the GI tract is both the target as well as

the source of sepsis. Given the enormous amount of intestinal commensal flora and bacteria, a

firm mucosal barrier must be present to keep them intraluminally. The occurrence of intestinal

barrier dysfunction has been described in an endotoxin model of the rat (Liu et al., 2009), and

mucosal barrier leakage also occurs after splanchnic hypoperfusion (Rahman et al., 2003). The

combination of increased mucosal permeability and sepsis-induced ileus (as discussed below)

predisposes to bacterial translocation. Bacterial adherence to the intestinal epithelium rises,

inducing the release of mucosa-derived cytokines (Mizock et al., 2009). These bacteria will

translocate, and the majority will be phagocytosed by intestinal immune cells that produce

large amounts of proinflammatory as well as anti-inflammatory mediators. Some translocating

bacteria and their toxins will be trapped in intestinal lymph nodes where they will provoke a

similar strong immune response (Tsukamoto et al., 2010). Clinically, this bacterial translocation

is confirmed in the human setting by culturing intestinal lymph nodes showing presence of

intestinal bacteria (Deitch, 2012). The importance of bacterial translocation has also been

proven by MacFie et al., who found intestinal bacteria to be the instigating agents in

pneumonia in septic patients (MacFie et al., 1999).

The occurrence of bacterial translocation alone is however insufficient to elucidate the

development of MODS in all critically ill patients. Since bacteria or endotoxins cannot be

detected in the blood of every ICU patient with MODS (Moore et al., 1991), the pathogenesis of

MODS might be more elaborately outlined in what is called the ‘gut-origin sepsis hypothesis’

(MacFie et al., 1999; Deitch et al., 2006; Deitch et al., 2006; Senthil et al., 2006). This

hypothesis recognizes the importance of bacterial translocation, but adds the concept of gut-

derived sepsis, with the gut being a source of proinflammatory and other non-microbial factors

that induce tissue damage or will contribute to the continuous development of SIRS, acute

respiratory distress syndrome (ARDS), MODS and MOF (Deitch, 2012). These non-microbial

proinflammatory mediators will drain predominantly via the intestinal lymphatics, in favor of

the portal circulation, and reach the systemic circulation. The first major vascular bed these

mediators encounter is the lung, which explains the clinical observation that the lung is

generally the first organ to fail in severely ill patients (Deitch, 2012). Via the intestinal

lymphatics and systemic circulation, these mediators can also recirculate to the gut and


33

damage other cell types at a distance from the initial infectious focus. The non-microbial

mediators also act at a distance on several centers in the brain, triggering neural reflexes

(Pavlov et al., 2003). In reality one can assume a combination of these action mechanisms is

likely to occur.

Not all gut-derived mediators have been extensively described up until now, but many authors

showed that the gut is capable of producing cytokines such as IL-6 and TNF-α (Deitch et al.,

2012; Tsukamoto et al., 2010). Other suspected mediators leading to MODS are reactive

oxygen species (Rotstein et al., 2000) and arachidonic acid liberated from epithelial

phospholipids (Mizock et al., 2009). As was already acknowledged, shock with the resulting

(splanchnic) hypoperfusion is an important inciting event in MOF. The reperfused gut is a

source of proinflammatory mediators that can amplify the early SIRS, thus contributing to early

MOF. The damage caused by reperfusion of the ischemic intestine will equally contribute to the

damage caused by the preceding ischemia, and is typically termed ischemia/reperfusion injury

(Hassoun et al., 2011; Pavlov et al., 2008).

In addition to the concept of bacterial translocation, early gut hypoperfusion is associated with

an impairment of the GI motility or ‘ileus’ of the entire GI tract and concurrent declining of the

mucosal barrier (Tsukamoto et al., 2010; Liu et al., 2009). The proximal gut becomes a reservoir

for pathogens and toxins, thus promoting bacterial overgrowth (Deitch et al., 1996).

Furthermore, during ileus nutritional support is restricted, resulting in less nutritional

components passing through the lumen and prohibiting the stimulation of the entire epithelial-

based immune system, GALT. In addition, mesenteric lymph collected during sepsis potently

inhibited gastric motility in animal experiments (Glatze et al., 2004), illustrating the potential of

mediators of gut origin in the systemic circulation influencing gut motility. The foregoing thus

supports the hypothesis that the gut is both an instigator as well as a victim of MOF.

These new insights in the importance of the gut should have implications for the treatment of

sepsis in all critically ill patients. Up until now therapeutic efforts have focused on lung,

cardiovascular and kidney failure, but therapeutical management should include targeting the

gastrointestinal tract dysfunction (Deitch, 2012). One of the proposed GI-oriented approaches

is the selective digestive tract decontamination (SDD) which has been controversial ever since

its development more than 30 years ago (Stoutenbeek et al., 1983; Stoutenbeek et al., 1984).

The general principle comprises the administration of prophylactic antibiotics to selectively

eradicate all potentially pathogenic micro-organisms so that bacterial translocation simply can


34

no longer occur (Rotstein, 2000). However, many doubt the cost-efficacy of this intervention

and are concerned about the preferential selection of resistant pathogens. In the 2008

‘International guidelines for management of severe sepsis and septic shock’ of the Surviving

Sepsis Campaign, the group stranded a 50/50 split decision, meaning that one half of the panel

members voted to include SDD in the guidelines, while the other 50% voted against, and thus

SDD was not included in the 2008 edition (Dellinger et al., 2008). In the 2012 edition the matter

of SDD was again vaguely left untouched, with the recommendation to further investigate the

efficacy of SDD in the prevention of ventilator-associated pneumonia (Dellinger et al., 2013).

1.3.2 The nervous system of the gut during sepsis

A myriad of data has been published on the effect of sepsis on the nervous system, in particular

on the effect on the afferent nerve, as well as on the modulating role of the nervous system

during a state of inflammation. The greater part of research performed focuses primarily on

the effect of pure LPS-administration, proinflammatory cytokines and/or (whole or lysated)

bacteria on afferent nerves and DRG neurons. Upon intravenous administration of LPS in the

rodent, many have confirmed the ensuing hypotensive state that coincides with an increased

firing activity of jejunal afferent nerves (Liu et al., 2005). This increase in firing activity consisted

out of an initial short rise in the afferent chemosensitivity to serotonin and mechanosensitivity

to distension five minutes post-administration, that was subsequently followed by a more

sustained rise in spontaneous afferent nerve activity approximately one to two hours later.

Others showed that the discharge of the vagal afferent and efferent nerves, as well as of the

vagal nucleus of the solitary tract, were significantly increased following intravenous

administration of LPS (Huang et al., 2010). The effects of LPS however were shown to be

dependent upon the bacterial strain that provides the LPS, as LPS from a commensal E. coli had

no effect on mesenteric afferent nerve activity whereas LPS from the pathogen S. typhimurium

did so (Liu et al., 2009).

The afferent fibers that detected a bacteriological infection in the gut have been demonstrated

to be capsaicin-sensitive (Riley et al., 2013). In this regard, the preventive administration of

hexamethonium, a blocker of neuronal transmission, prevented the occurrence of LPS-induced

delay in GI transit, pointing towards a crucial role of afferent neurons in sepsis-induced GI

motility disturbances (De Winter et al., 2009). Furthermore, animals pretreated with capsaicin,

a CGRP-antagonist or a TRPV-1 antagonist were all protected from developing this impaired


35

transit, suggesting that these afferent neurons are indeed capsaicin-sensitive and that signal

transduction was mediated via TRPV1 possessing fibers. Our group demonstrated previously

that a TRPV1-receptor antagonist could reverse the disturbed gastric emptying that occurs in a

rat-model of colitis, and that the TRPV1-receptor is present in pelvic afferent neuron cell bodies

in DRGs (De Schepper et al., 2008). This afferent nerve activity originating from the inflamed

gut can furthermore be modified by endocannabinoids, as pretreatment with a selective

endocannabinoid agonist to rats challenged with LPS significantly decreased the LPS-induced

increase in afferent firing rate in jejunal afferent nerves (Donovan & Grundy; 2009).

Concerning the effect of proinflammatory cytokines, it has been demonstrated that several of

their receptors are expressed on the afferent vagal branch (Fernandez et al., 2012), and can

either directly or indirectly cause neuropathic pain and hyperalgesia (Sommer et al., 2006).

Vagal afferents and paraganglia play a pivotal role in the so called cytokine-to-brain

communication, as subdiafragmatic vagotomy attenuated the response to i.v. administered

TNF-α and IL-1β (Fleshner et al., 2008). Injection of IL-6 could increase the activity of C-fibers

(Brenn et al., 2007), and intraperitoneal administration of IL-1β induced cFos expression in

primary afferent vagal neurons (Goehler et al., 1998); the same is true for TNF-α (Holzer et al.,

2004). Finally, mesenteric lymph collected from rats challenged with LPS increased mesenteric

afferent discharge in another animal when administered intravenously, whereas administration

of pure TNF-α could not, suggesting that other inflammatory mediators secreted from the gut

in the lymph contribute to this increased afferent activity (Wang et al., 2005).

Administration of LPS into the jejunum not only affects peripheral afferent nerves, it also

induces the expression of cFos, a marker of neuronal activation, in the vagal nucleus of the

solitary tract (Gakis et al., 2009; Wan et al., 1994); intravenous administration resulted in

increased cFos-expression in the nucleus of the solitary tract as well as in the area postrema,

since the latter is exposed to the systemic circulation. A lysate prepared from whole bacteria

was able to directly activate colonic DRG neurons (Ochoa-Cortes et al., 2010), and incubation of

DRGs with IL-1β or TNF-α increased DRG mechanosensitivity in group II units and increased the

peripheral receptive fields (Ozaktay et al., 2006).

LPS-induced sepsis significantly impaired overall neuronal activity in the rat brain as could be

detected by a decrease in [18F]-FDG uptake, presumably due to a combination of

neuroinflammation and changes in cerebral blood flow (Semmler et al., 2008).


36

1.3.3 The cholinergic anti-inflammatory pathway

In 2000, the group of Borovikova and Tracey demonstrated that an endogenous reflex pathway

is present in mammals, suppressing inflammation (Borovikova et al., Nature 2000). Upon

activation of the afferent vagal branch by inflammation or cytokines, a feedback loop is

initiated via the central nervous system. They demonstrated that, following the administration

of acetylcholine to an in vitro culture of human macrophages that was being challenged with

LPS, the release of several proinflammatory cytokines (IL-6, IL-1β, IL-18 and TNF-α) but not that

of the anti-inflammatory IL-10 could be subdued via a posttranscriptional mechanism. The

addition of α-conotoxin abolished these beneficial results and confirmed that this mechanism

was mediated by a nicotinergic acetylcholine receptor, as α-conotoxin is a selective nicotinergic

antagonist. This ‘anti-inflammatory reflex pathway’ has been shown to be mediated via the

nicotinic acetylcholine alpha7 unit receptor (α7nAChR) (Wang et al., 2003). The authors thus

suggested that acetylcholine released from vagal endings (or other sources?) would inhibit the

release of proinflammatory cytokines from macrophages via the binding of acetylcholine on

the α7nAChR expressed on the cell surface of macrophages. The exact anatomical substrate of

the cholinergic anti-inflammatory pathway has been further unraveled and critically

questioned by many over the past decade, as was nicely summarized by Martelli et al. (2014)

(figure 1.6). The role of the spleen in this elegant reflex pathway was first touched upon by

Huston et al. (2006). They showed that the vagal anti-inflammatory reflex pathway was

abolished in splenectomized mice, and that this spleen-mediated effect is wired via the

common celiac nerve, as selective ablation of this nerve abolished the suppression of TNF-α

release by vagus nerve stimulation. At the same time, they demonstrated in a mouse model of

polymicrobial sepsis that splenectomy actually protected against the development of sepsis-

induced mortality, alluding to the dual role of the spleen in inflammation on the one hand, and

neuronal modulation of inflammation on the other hand (Huston et al., 2008). The spleen

however does not receive abundant cholinergic input from the vagal nerve, only noradrenergic

fibers from the splenic sympathetic nerves, necessitating the presence of a new player in the

cholinergic anti-inflammatory pathway (Nance & Sanders, 2007). A suggested site for the

alpha7 receptor is the postganglionic splenic neuron (Rosas-Ballina et al., 2008; Anderson &

Tracey, 2012). The presence of acetylcholine-synthesizing T cells residing in the spleen was

then incorporated in this reflex pathway, with the α7nAChR located on splenic macrophages;

others demonstrated that vagal stimulation acts on the spleen by a non-neural pathway (Vida


37

et al., 2011). More recent data suggest that indirect modulation of macrophages in the GI tract

by vagal efferents is able to inhibit inflammation via VIPergic and nitrergic enteric neurons

without any contribution of the spleen in a mouse model of postoperative ileus (Cailotto et al.,

2014; Matteoli et al., 2014). Costes et al. (2014) showed that splenic denervation did not affect

intestinal inflammation or GI transit following intestinal manipulation, leading them to

conclude that endogenous activation of the vagal anti-inflammatory pathway by intestinal

inflammation directly dampens the local immune response triggered by intestinal manipulation

independently of the spleen. Martelli et al. (2014) however showed in a rat model of LPS-

induced sepsis that the TNF-α release remained unaltered following previous bilateral cervical

vagotomy in comparison to animals with intact vagal nerves, whereas levels increased

significantly following prior cutting of the greater splanchnic nerves, allocating an important

role to the splanchnic sympathetic efferent nerves and stating that the afferent arm

presumably is of humoral nature. Much thus remains to be elucidated on the role of this vagal

anti-inflammatory reflex pathway on sepsis-induced motility disturbances.

Figure 1.6 The vagal anti-inflammatory pathway (next page) – as of January 1st 2016 The anatomical substrate of the cholinergic anti-inflammatory pathway, based upon what is currently elucidated. Inflammation and increased local concentration of proinflammatory cytokines in the periphery will be detected by the afferent splanchnic nerves, and conveyed to the central nervous system. A central feedback mechanism will result in the release of acetylcholine in the celiac ganglion, via which the splenic nerves will become activated. The release of noradrenalin from the splenic nerve endings in the spleen will stimulate choline acetyltransferase positive T cells to release acetylcholine, which will subsequently bind onto the α7nAChR on the cell surface of macrophages, thereby inhibiting the release of proinflammatory cytokines from these splenic macrophages. Alternatively, vagal efferent branches will directly release acetylcholine in the gut (right), where acetylcholine can directly bind onto gut residing macrophages. Based upon Tracey, J Clin Invest 2007; Martelli et al., Auton Neurosci 2014; Matteoli et al., Gut 2014. CNS : central nervous system; GI: gastrointestinal; IL: interleukin; Mφ: macrophage; NA: noradrenalin; TNF: tumor necrosis factor


39

1.3.4 The contribution of the ‘failing’ gastrointestinal tract to sepsis

Combining all that we summarized in the previous paragraphs concerning the effects of sepsis

on the GI tract, one can appreciate the intricate interplay that occurs during sepsis between

the different neurological, cellular and immunological players at the GI tract level.

Many also believe the GI tract to be an instigator of sepsis, as the gastrointestinal tract plays a

pivotal role in other animal models of shock, without the presence of any focus of infection.

Mesenteric lymph draining the GI tract collected from rats subjected to abdominal trauma

followed by hemorrhagic shock caused several detrimental effects, such as lung injury,

neutrophil priming and red blood cell deformity upon administration to naïve mice, while vagal

nerve stimulation (VNS) was protective against the aforementioned effects (Levy et al., 2013;

Fishman et al., 2014). In a mouse model of traumatic brain injury, blunting and necrosis of villi

is commonly observed, resulting in increased intestinal permeability (Feighery et al., 2008;

Bansal et al., 2009). Intestinal segments of mice subjected to brain injury displayed impaired

mucosal barrier function and increased levels of TNF-α. VNS preserved mucosal integrity and

significantly reduced the TNF-levels (Bansal et al., 2010). In another mouse model of burn

injury, mice also displayed blunted intestinal villi, with activation of enteric glia, increased

intestinal TNF-α levels and increased mucosal permeability (Costantini et al., 2010; Costantini

et al., 2010; Krzyzaniak et al. 2010), dependent upon TLR4-signalling (Peterson et al., 2010).

The observations described here led to the development of the ‘gut-lymph hypothesis of

multiple organ dysfunction syndrome’, during which ‘biologically active’ lymph, containing

whole bacteria as well as antigens, bacterial compounds and inflammatory mediators, will

induce detrimental effects in distant organ systems, most notably the lung (Deitch, 2010;

Deitch, 2012; Gatt et al., 2015). Similarly, infusion of ‘septic lymph’ obtained from rats

intraperitoneally injected with LPS in unchallenged animals impaired their gastric and colonic

motility (Königsrainer et al., 2011). Whether these compounds comprise microbial or non-

microbial antigens, cytokines or other mediators, remains the topic of investigation. The

concept of bacterial translocation or the gut-origin-of-sepsis has been confirmed in large

clinical series, as several studies have demonstrated bacterial translocation to mesenteric

lymph nodes in up to 20% of surgical patients, associated with an increase in the rate of

infectious complications. The organisms identified in the focus of infection, such as the lungs,

were moreover often identical to the ones identified in the mesenteric lymph nodes (MacFie et

al., 1999; MacFie et al., 2006; Reddy et al., 2007; Deitch, 2012).


40

1.4 Possible therapeutic treatment strategies

From the myriad of cell types, inflammatory molecules and other players we described above,

researchers often have difficulty targeting the ‘one’ compound that might make a difference in

the treatment of the septic patients. As was already mentioned before, the failure of

translating favorable effects in the controlled laboratory research environment in animals

towards the diverse, human septic patient population has resulted in many clinical

experimental failures. This encourages us to bear the heterogeneity of the patient’s

characteristics in mind, such as the immune profile and the presence of comorbidities, when

developing new therapeutic strategies (Angus & van der Poll, 2013). Our belief in the gut as a

causative organ in the development and consolidation of sepsis, encouraged us to study the GI

tract in several animal models of sepsis, and identify possible treatment targets.

1.5 Studying new treatment strategies – animal models of sepsis

As has been previously touched upon, the development of new therapeutic strategies is

hampered by the lack of an animal model that adequately mimics the various hemodynamic

and immunological changes that occur during human sepsis. The most established and

commonly used animal models applied in sepsis research are summarized below.

1.5.1 The lipopolysaccharide-model

The lipopolysaccharide (LPS) animal model of sepsis is a commonly used and simple technique

applied to induce sepsis in animals, most notably in mice and rats. Following the

intraperitoneal (i.p.) injection of lipopolysaccharide from a specified bacterial strain, usually

Escherichia coli, the animal will display signs of disease as clinical deleterious manifestations

typically observed during sepsis set in. In short, LPS (or ‘endotoxin’) is a molecule consisting out

of a lipid and polysaccharide that is found in the outer membrane of Gram-negative bacteria

and can elicit the immune responses that can be typically observed during SIRS. LPS will bind

upon the TLR4 associated with the surface marker CD14, predominantly expressed on the cell

surface of macrophages, monocytes, dendritic cells and B cells, resulting in activation of the

innate immune system, with the subsequent release of several proinflammatory cytokines,

nitric oxide, prostaglandins and other eicosanoids (Abbas et al., 2012). TLR4 signal transduction

is mediated via either the MyD88 pathway, or the TRIF pathway. In the MyD88 pathway,

dimerization of the TLR4 will result in recruitment of several interleukin-1 receptor associated


41

kinases (IRAK), more specifically IRAK 4, IRAK 1 and IRAK 2. These will in their turn recruit other

kinases, such as TRAF-6 and IKK-γ, ultimately leading to the downstream release of cytokines,

NF-κB and mitogen-associated protein kinases that regulate several cell functions such as

proliferation and apoptosis (figure 1.7). Furthermore, the MyD88 pathway is an important

player in the transition from innate towards adaptive immunity, as activation of this pathway in

dendritic cells will result in the release of interleukin-12, leading to the skewing of naïve T cells

towards type 1 helper T cells. The TRIF-dependent pathway consists out of phosphorylation of

interferon regulatory factor 3 (IRF3), ultimately leading to the release of interferon type I.

Ultimately, both pathways will lead to the tightly controlled transcription and translation of

several pro- as well as anti-inflammatory cytokines.

Of note, several pharmacological agents daily used in the critically ill patient are actually

agonists (morphine, fentanyl, buprenorphine, methadone, oxycodone, pethidine, and ethanol

(not a drug though…)) or antagonists (naloxone, naltrexone and amitriptyllin) of the TLR4

receptor. The immune response that can be observed following bolus administration of LPS

consists out of an impressive innate immune response with predominantly secretion of TNF-α.

This immune activation is typically accompanied by an immediate decrease in blood pressure;

this hypodynamic cardiovascular state however does not typically mimic the hemodynamic

changes observed in human sepsis, that are initially characterized by a transient rise in cardiac

output provided that patients received sufficient fluid resuscitation. Bolus injection of LPS in

rodents will result in a rapid and impressive increase of serum proinflammatory cytokines,

however human serum cytokine levels are significantly lower and display a more protracted,

slow rise (Wintersteller et al., 2012; Remick et al., 2000). Within the hour following i.p.

injection, a significant rise in TNF-α levels can be witnessed in the serum, followed by an

increase in IL-10 (2h), IL-6 (3-4h) and IL-1β (6h). Furthermore, an increase in several

chemokines (CXCL1, CXCL2 and MCP-1 amongst others) could be observed already 2h following

LPS-administration. These typical manifestations, as well as the risk of mortality can be

adjusted by the researchers by means of increasing the dosage of LPS that is administered to

the rodent, and are dependent on the strain, age and sex of the rodent that is utilized by the

researcher. At the GI tract level, LPS-induced sepsis will result in a dose-dependent decrease in

GI transit times (Eskandari et al., 1997 & 1999).


42

Figure 1.7 TLR4-signalling by binding of LPS Akt: protein kinase B; CD: cluster of differentiation; IKK: IκB kinase; IRAK: Interleukin-1 receptor-associated kinase; LBP: LPS binding protein; LPS: lipopolysaccharide; MAPK: mitogen associated protein kinase; MyD88: Myeloid differentiation primary response gene 88; NF-κB: nuclear factor kappa B; TAK1: TIRAP: PI3K: phosphoinositide-3-kinase; TLR: toll like receptor. Based upon and modified from Abbas et al., 2013.

1.5.2 The cecal ligation and puncture-model

The CLP-model was first defined and established as a reproducible animal model that mimics

sepsis by Wichterman and colleagues over three decades ago (1980). The authors critically

challenged the use of endotoxins-models for the aforementioned reasons, and studied for the

first time an animal model in which the cecum was ligated and punctured at the same time in

rodents, mimicking the human state of perforated diverticultis or appendicitis resulting in a

polymicrobial abdominal sepsis with the simultaneous presence of features of SIRS as well as

CARS, and apoptosis of immune cells (Buras et al., 2005). Prior research focused merely on the

induction of sepsis by simply ligating the cecum in dogs and pigs (Clowes et al., 1968; Imamura

et al., 1975), after which devascularization of the cecum presumably led to necrosis and

subsequent spillage of fecal contents in the abdomen, resulting in sepsis. The authors tried to

reproduce these findings in rats, as rodents are more practical in use; however they were

unable to obtain the same results, leading them to complement the ligation with a puncture.

Proinflammatory cytokines, proliferation, apoptosis,…

TLR4

TLR4

MD2

CD14

LBP

LPS

TIR

AP

MyD

88 IRAK1

IRAK4

TRAF6TAK1

IKKs

IkBa

NF kB

RIP

PI3K

Akt

MAP3Ks

MEK/ERK JNK p38


43

As such, they were able to consistently demonstrate sickness behavior, obtundation and

occurrence of mortality in rats, as well as the typical hemodynamic and glycemic changes that

occur during human sepsis in both the early as well as the late phase, on the condition that

fluid resuscitation is performed by means of injecting animals with a warm (37°C) saline

solution. They furthermore showed that these results were procedure-dependent, as reducing

the number of punctures improved survival. Others showed that the distance of the ligated

cecum also influences mortality, as a ligation stretching over 30% of the cecum’s length will

result in nearly 100% mortality when followed by a double puncture with a 20G needle

(Singleton & Wischmeyer, 2003; Rittirsch et al., 2009). A strain-dependent mortality has been

demonstrated as well (Doi et al., 2009; Toscano et al., 2011). Local control of the leaked fecal

matter, the infectious focus, by abscess formation will ‘rescue’ the animal from developing a

more pronounced severity of sepsis and improve chances of survival. This feature is present in

nearly all animals that underwent CLP, as adjacent bowel loops will adhere to the punctured

cecum as well as to the peritoneum and abdominal bowel wall.

Microscopical analysis of GI small bowel tissue usually reveals signs of peritonitis and

infiltration of neutrophils in the serosa and tunica muscularis, however no notable infiltrate

could be observed in mice sacrificed 24h following CLP (15 mm ligation, one puncture with 18G

needle) (Maier et al., 2004). The hemodynamic, metabolic and immunological features of the

CLP-model are extensively described by numerous excellent reviews. The group of

Wintersteller (2012) provides us with an extensive summary of the serum and organ cytokine

profile in several animal models of sepsis. In short, CLP-induced sepsis is characterized by a

profound rise in serum IFN-γ (1-5h), TNF-α (5h), IL-6 (5-8h), IL-1β (5-6h) and IL-10 levels (5-8h

post-CLP). Lung tissue displayed a comparable cytokine profile as the serum following CLP,

however Iskander et al. showed that mice that succumbed from CLP did not display extensive

lung injury (Iskander et al., 2013).

Concerning chemokine C-C motif ligands (CCL) and chemokine C-X-C motif ligands (CXCL), a

marked rise in serum and lung tissue levels of CCL2, MIP and RANTES was observed from 1h

post-CLP and onwards, as well as an increase in CXCL1, CXCL2 and CXCL9 and an increase in the

chemokine receptor CCR1 in lungs from septic animals. These levels more accurately mimic the

human setting when compared to the LPS-model (Rivers et al., 2013).

Of note, monitoring animals with a biological parameter or biomarker (instead of behavioral

signs of disease) could allow for some sort of severity assessment; these however have not


44

been thoroughly established, with the possible exception of serum interleukin-6 levels, that are

indicative of chances of mortality but do not adequately predict it, a feature that is also present

in human sepsis (Song & Kellum, 2005). The effect of CLP on the gastrointestinal morphology

and function remains a research topic that is worth investigating, as only limited data is

available to us.

1.5.3 The colons ascendens stent peritonitis model

In the colons ascendens stent peritonitis (CASP) model, a stent with defined diameter is

surgically implanted into the rodent’s colon ascendens, resulting in the continuous leakage of

fecal matter and bacteria in the peritoneal cavity (Buras et al., 2005) and subsequent

development of widespread peritonitis. As is the case with the CLP-model, the severity can be

influenced by adjusting the diameter of the implanted stent. The model is characterized by the

presence of LPS and bacteria in the bloodstream (2h versus 12h following the procedure

respectively), and simultaneous steadily increasing secretion of pro- as well as anti-

inflammatory cytokines, albeit displaying their peak levels later on (ranging from 3 to 18h post-

CASP) in comparison to the CLP-model (Wintersteller et al., 2012). IL-10 levels are remarkably

upregulated following CASP (Maier et al., 2004). Furthermore, an upregulation of serum CCL2,

CXCL1 and CXCL10 levels could be observed as well.

1.6 Summary

In this Chapter, we provided a non-exhaustive overview on what is known concerning the

different neuro-immune players systemically as well as locally, that play a pivotal role in the

development and maintenance of sepsis-induced ileus. A full understanding of this interplay is

pivotal in the development of new therapeutic strategies for the septic patient.

Chapter 2

Aims

Chapter 2 Aims

46

The role of the gastrointestinal (GI) tract during sepsis remains a subject of much debate. The

most recent guidelines concerning the management of the septic patient only cautiously touch

upon this subject, stating that “selective oral and digestive tract decontamination by means of

antibiotics should be investigated within the scope of infection prevention (e.g. ventilator-

associated pneumonia) and that continuous research is warranted in order to ascertain

benefits, risks and cost-effectiveness” (Dellinger et al., 2012 & 2013). Research performed

during the past couple of years however substantiated that gut-derived mediators (and not

only bacteria as such) present in the mesenteric lymph during sepsis are responsible for

instigating deleterious effects in distant organ systems during sepsis, and thus can maintain or

even exacerbate the disease (Deitch, 2012; Levy et al., 2013; Dellinger et al., 2013). These

inflammatory mediators secreted by the gut are able to directly affect smooth muscle cells and

neuronal reflex pathways, leading to motility disturbances, pain or hypersensitivity (De Winter

& De Man, 2010). Further study on the different immune cells and their neuroimmune

interactions are therefore mandatory.

In Chapter 1, we elaborated on what is currently known concerning the inflammatory

environment in the GI tract, and the disturbances that occur during sepsis-induced ileus. We

started of by giving a concise review on the current definition(s) of sepsis, its pathophysiology

and the most recent guidelines on the management of the septic patient. The characterization

of the different neuroimmune players involved in the pathogenesis of sepsis can aid us in the

identification of possible therapeutic targets.

As translational research concerning sepsis is often hampered by the difficult extrapolation of

results from the murine (animal) towards the human state, we first needed to implement a

new animal model of sepsis-induced ileus in our lab in order to optimally mimic the human

septic state.

The objectives of this dissertation are as follows:

i. To implement a reliable, reproducible, and representative animal model of sepsis in

our laboratory. One of the most commonly used murine models in this regard is the

‘cecal ligation and puncture’ or ‘CLP’-model.

ii. To characterize the different neuroimmune players that play a role during the

development and maintenance of sepsis-induced ileus, inflammation, colonic

permeability and bacterial translocation.

Chapter 2 Aims

47

iii. To modulate the neuroimmune environment with the ultimate goal to ameliorate

sepsis and sepsis-induced GI inflammation, motility and permeability disturbances.

a. Neuronal modulation

b. Immunological modulation

In Chapter 3, we describe into detail the methodology applied in this dissertation. The

lipopolysaccharide (LPS) model is a well-known animal model of sepsis that consists out of a

simple intraperitoneal injection of LPS from Escherichia coli. This model was already in use in

our lab, and we therefore initiated the dissertation with this technique. But, despite being an

easy and reproducible model, this technique is impeded by several disadvantages. As the CLP-

model is by many regarded to be the gold standard animal model in animal sepsis research, we

decided to implement this animal model in our lab. The CLP-model closely resembles the

human pathological state of ‘systemic inflammatory response syndrome’ (SIRS), characterized

by a profound secretion of proinflammatory cytokines and upregulation of proinflammatory

cells following the development of an intra-abdominal inflammatory focus.

Following the induction of sepsis by performing CLP, we first assessed whether an impaired GI

motility (‘ileus’), was present in our animals at specific time points following the induction of

sepsis by CLP. Next, inflammation was assessed locally (at the GI tract level) as well as

systemically (serum). Finally, colonic permeability was objectified and the bacterial load was

quantified in our animals. The effects of several interventions or therapeutic agents we utilized

on sepsis-induced alterations were always assessed using the techniques described in this third

chapter.

The implementation, optimization and characterization of the CLP-model in our laboratory are

elaborately discussed in Chapter 4. Several modalities of the CLP-model were tested and

adjusted, with increasing severity and a mortality ranging between 0 and 100% up until 14 days

following the induction of sepsis. As we were specifically interested in the role of the GI tract

during sepsis, we opted to implement a model with a very low mortality in order to minimize

the number of animals needed in each study protocol. Based upon a clinical disease score and

individual monitoring of the animals during the implementation study, we choose two time

points (day 2 and day 7 following the procedure) that were studied later on into more detail.

Chapter 2 Aims

48

Furthermore, the neuroimmune characterization of different GI tissues is described using flow

cytometry, allowing us to identify the different cell populations.

In Chapter 5, we investigated the role of the peripheral and central nervous system during

sepsis. Both play a pivotal role in the neuroimmune interactions in the gut during sepsis, as

inflammatory mediators can activate neuronal reflexes resulting in motility disturbances and

alterations in GI sensitivity (De Winter & De Man, 2010). Vice versa neurotransmitters are able

to influence immune cells. As such, it has been demonstrated by numerous researchers that

the afferent branch of the vagal nerve can detect peripheral inflammation and signal this to the

brain, initiating an endogenously present reflex pathway called the ‘vagal cholinergic anti-

inflammatory reflex’ (Tracey, J Clin Invest 2007; Rosas-Ballina & Tracey, J Intern Med 2009).

Briefly, via a central feedback mechanism the efferent branch of the vagal nerve is activated,

leading to the release of acetylcholine. This will ultimately result in the inhibition of secretion

of proinflammatory cytokines from immune cells, most notably from the macrophages. With

this theory in mind, we sought to characterize the activity of the nervous system at different

levels: the afferent jejunal and colonic nerve, the dorsal root ganglion (DRG) and the central

nervous system.

Afferent splanchnic nerve activity was studied in an in vitro setting at baseline conditions and in

response to ramp distensions. Next, DRGs were isolated from septic and control animals, and

several markers for neuronal activation were quantified by means of RT-PCR. Finally, central

nervous system imaging was performed by means of measuring the uptake of a radioactive

tracer, [18F]-FDG, in the different brain regions by means of PET/CT imaging in collaboration

with the Molecular Imaging Center Antwerp (MICA).

The aforementioned ‘vagal anti-inflammatory reflex’ was used as a target for our first

interventional study in the CLP-model. We studied the effects of GTS-21, an agonist for the

alpha7 nicotinic acetylcholine receptor, on CLP-induced motility disturbances, GI inflammation

and impaired colonic permeability. This so called ‘neuronal modulation’ is described in Chapter

6.

In Chapter 7, the effects of immunological modulation of septic ileus by administering

antibodies to the proinflammatory cytokine interleukin-6 are studied in the CLP-model.

Chapter 2 Aims

49

Previous research by other groups demonstrated that serum levels of interleukin-6 are

predictive of mortality during sepsis; however, not all results were conclusive, and others

reported no beneficial effect at all. It was also shown that targeting IL-6 could improve survival

in several animal models of sepsis or (septic) shock. Of note, IL-6 has also been reported to be a

pivotal factor in gut permeability, as it was required to induce gut barrier dysfunction (Yang,

Am J Physiol 2003), and it impaired tight junction permeability in a human enterocyte cell

culture (Tazuke, Pediatr Surg Int 2003). IL-6 thus remains an interesting therapeutic target in

the quest for new sepsis-directed therapies, despite prior contradictory results in murine

studies.

The overall results of this thesis are discussed in Chapter 8. Based upon our integrated results

presented in the prior chapters, a general working hypothesis is put forward. Our findings are

discussed from a clinical point of view, and we speculate on whether neuronal or

immunological modulation of the GI tract can be implemented in the management of the

septic patient. A link is provided to ongoing clinical trials, and the discussion is raised on why

we preferentially ought to opt for clinical trials that target ‘patient-specific’ needs and

personalized medicine and abandon the ongoing quest for that one magic bullet.

Chapter 3

Materials and methods

Chapter 3 Materials and Methods

52

In the following chapter, we will extensively outline several laboratory techniques and methods

that were repeatedly utilized during this research project, and are hence mentioned in multiple

chapters. For the more specific techniques, we refer the reader to the proper chapters (e.g.

measurement of afferent nerve activity is described in Chapter 5 – Study of the afferent nerve

activity and central nervous system activation in the cecal ligation and puncture model). The

flow cytometric characterization of the cecal ligation and puncture model will be outlined in

Chapter 4.

3.1 Animals

Male Swiss OF-1 mice, 8 weeks old weighing 20 to25 grams, were obtained from Charles River

(France) and were allowed to acclimatize for two weeks in their cages under specific conditions

(6 to 8 animals per cage, 21 ± 1°C, humidity 40-60%, 12h light-dark cycle) with unlimited access

to regular chow and tap water. Mice were 10 weeks old at the initiation of the experiments. All

experiments listed below were approved by the Ethical Committee on Animal Experiments by

the University of Antwerp (file number 2012-42 and 2012-42-extension).

In Chapter 6, we describe the use of animals that lack the alpha7 nicotinergic acetylcholine

receptor (Chrna7-/- mice). Heterozygous breeding couples were kindly gifted to us by Prof. dr.

G. Matteoli and Prof. dr. G. Boeckxstaens (TARGID, KULeuven). Chrna7-/- and their wildtype

littermates (Chrna7+/+) were locally bred on a C57Bl/6 background, and were housed in the

aforementioned conditions.

3.2 The cecal ligation and puncture animal model of sepsis

The optimization and implementation of the cecal ligation and puncture (CLP) animal model of

sepsis in our laboratory is extensively described in Chapter 4, whereas the immunological

specifications of this model are discussed in Chapter 1. In short, the CLP-model more

adequately mimics the hemodynamic and metabolic phases of human sepsis and the immune

response mimics more accurately that of human subjects in comparison to mere LPS-

administration (Wichterman et al., 1980). The severity as well as the mortality of the animal

model can be modulated by the researcher as desired, e.g. a severe CLP-induced sepsis with

high mortality rates to study pure survival numbers, or a more mild sepsis to study

pathophysiology and mechanisms of disease (Rittirsch et al., 2007). That being said, it should

be noted that a number of animals will be able to limit the inflammatory process that unfolds


53

in the abdomen of the murine subject by containing it by means of abscess formation. The

procedure is furthermore highly operator-dependent, as the amount of stool that protrudes

from the puncture is crucial in eliciting a reproducible ‘septic state’ (Rittirsch et al., 2007).

Similar to the LPS-model, significant variability in research results and outcome can be

instigated by the strain, sex and age of mice that was chosen by the researchers.

Our goal was to implement a reproducible animal model with marked impairment of GI

functions (i.e. a decrease in transit function and increase in mucosal permeability), but without

significant mortality. The final model (50% ligation of the cecum and one puncture with a 25G

needle) was used for the remainder of this research project. As for the final procedure itself,

mice were anesthetized with a mixture of ketamine (60 mg/kg i.p.) and xylazine (6.67 mg/kg

i.p.), and placed onto a heating pad in the supine position. A midline laparotomy was

performed following abdominal shaving and disinfection with a polyvidon-iodine solution, after

which the cecum was exteriorized and positioned onto moist sterile cottons, and ligated at 50%

of its length with a 4/0 silk thread and subsequently punctured once through-and-through with

a 25G needle in order to obtain a mild sepsis without mortality up to 10 days following CLP

(figure 3.1). The ligated cecum was then repositioned into the abdominal cavity, and the

abdomen was closed in layers with 5/0 Ethilon sutures. Mice received 1 ml of a 37°C warm

isotonic glucose-saline solution subcutaneously (s.c.) for fluid resuscitation and glycemia

control, and 0.05 mg/kg of buprenorphine s.c. for pain relief. Mice were allowed to recover in a

heated cage (28°C) up until 1h until they regained consciousness, with free access to tap water.

Sham-operated mice received a midline incision without ligation or puncturing of the cecum

(Wichterman et al., 1980; Buras et al., 2005; Rittirsch et al., 2009).

3.3 Clinical signs of disease and sickness behavior

Following the initiation of the experiments, animals were assessed for clinical signs of disease

twice daily (8 am and 7 pm) by a single observer. The clinical disease score (CDS) used in our

animal model was adapted from previous researchers and has been extensively validated in

our lab in the past (Heylen et al., 2013; Ruyssers et al., 2010; De Winter et al., 2009). The

following parameters were studied (table 3.1): presence of piloerection, mobility of the animal,

grooming behavior, presence of conjunctivitis or blepharitis, signs of peritoneal irritation,

abnormal position of the ears, abnormalities of the stool, anemic appearance of the animal and


54

so-called moribund appearance of the animal. Animals were weighed daily on an individual

basis from the start of the experiment. Animals that lost over 15% of their initial bodyweight in

non-sober condition or that had a CDS >8 were prematurely sacrificed in order to minimize

animal discomfort in accordance with our ethical rules and the guidelines of our Ethical

Committee. The rectal temperature was assessed once prior to sacrifice in the sedated animal.

3.4 Assessment of gastrointestinal motility

Several techniques have been described in literature to study gastrointestinal transit in mice,

the most commonly used being the FITC-dextran (40 kD) and the Evans blue method. During

these techniques the animal is gavaged with a certain volume of a semi-liquid meal that is

labeled with a fluorescent marker that can be detected in the different parts of the GI tract by

means of spectrophotometry. As the Evans blue method was routinely applied in our lab, GI

motility was initially assessed in our experiments with this technique (De Schepper et al., 2007;

De Winter et al., 2002). However, this technique only studies transit of a semi-liquid meal, and

the kinetics of gastric emptying differ substantially when comparing liquids to solids (Camilleri

et al., 1985). Quantifying GI transit of solids was done by using small green colored glass beads

by means of gavage (Seerden et al., 2007).

Furthermore, in contrast to the Evans blue technique, the solid beads technique allows us to

assess overall GI transit as it includes stomach, small bowel and colonic motility, whereas the

Evans blue technique restricts us to the quantification of stomach and small bowel transit.


55

Figure 3.1 A schematic depiction of the cecal ligation and puncture (CLP) procedure. Adapted and modified from Buras et al., Nat Rev Drug Discov 2005. a: laparotomy, b: distal ileum, c: cecum, d: proximal colon, e: ligation (dashed lines), f: puncture, g: protrusion of stool and luminal bacteria into the abdomen; the arrow denotes the normal transit of luminal contents through the GI tract.

Table 3.1. The clinical disease score (CDS) as applied in this dissertation to score signs of disease in individual mice. Sign Score

Piloerection (not present – extensive) 0-2 Conjunctivitis (not present – bilateral) 0-2 Grooming behaviour (normal – none) 0-2 Mobility (normal – reduced – immobile) 0-3 Signs of peritoneal irritation (none – tiptoeing and broad pace) 0-2 Position of the ears (erect – flat) 0-1 Stool consistency (normal – sticky or diarrhea) 0-1 Anemic appearance (absent or present) 0-1 Moribund 0-1

Total Clinical Disease Score (no signs of disease – maximum) 0-15

100%

75%

50%

b

ce

f

da

g


56

3.4.1 The Evans blue method: gastrointestinal propulsion of semi-liquids

Following overnight fasting with unlimited access to tap water, animals were given a gavage

with a semi-liquid bolus of 100 µL of Evans blue (50 mg/mL in 0.5% methylcellulose) (Evans

blue dye content >75%, purchased from Sigma-Aldrich) during a short diethyl ether anesthesia

(30 seconds). Fifteen minutes following the gavage, animals were sacrificed by means of

cervical dislocation. Following ligation of the distal esophagus and pylorus in order to prevent

leakage of the dye, the gastrointestinal tract (stomach and small bowel) was resected and

divided into 6 segments (stomach and small bowel 1 through 5). The amount of dye in each

segment was measured spectrophotometrically (emission 565 nm), and the percentage of

gastric emptying (%GE) and the geometric center of transit (GC, score 1-6)) can be calculated

by means of the following formulas (Seerden et al., 2007):

%GE = [∑A565 (small bowel 1–5)/∑A565 (stomach + small bowel 1–5)] × 100

GC = ∑(A565 Evans blue per intestinal segment ×number of segment)/totalA565.

3.4.2 The solid beads method: gastrointestinal propulsion of solids

In order to quantify the gastrointestinal transit when challenged with a solid meal, we

performed the solid beads method. Following overnight fasting, animals received an

intragastric administration of 25 small green colored glass beads (diameter 0.4-0.5 mm).

Animals were subsequently sacrificed at different time points (30 min, 2 hours or 6 hours), the

entire GI tract resected and divided into ten segments (stomach, small bowel 1 through 5,

cecum, proximal colon, distal colon and feces) and the number of beads was counted in each

segment under a stereomicroscope, allowing us once again to calculate %GE and GC (score 1-

10):

%GE = [number of beads (small bowel 1-5 + cecum + colon 1-2 + faeces)/total number of beads]

x100

GC = Σ (beads per segment x number of segment)/total number of beads).

For the optimization and implementation of the CLP-model in our lab, we initially studied

transit at all three time points using the solid beads method; in subsequent experimental

protocols we only studied transit two hours following the gavage, in order to limit the amount

of lab animals needed and to get an estimate of overall GI transit.


57

3.4.3 The solid bead expulsion test

As colonic transit is clinically often the longest impaired during ileus, we assessed colonic

motility by means of the solid bead expulsion test. Following a short anesthesia with diethyl

ether (30 sec), a glass bead (diameter 3 mm) is inserted rectally by means of a small stainless

steel rod and pushed in the oral direction up until 2 cm from the anal verge. Animals were then

monitored individually in separate cages until expulsion of the bead.

3.4.4 In vitro study of the colonic peristaltic activity

As the clinical manifestations of ileus usually persist longer in the colon in comparison to the

more proximal parts of the GI tract, we studied colonic peristaltic activity furthermore in an in

vitro setting commonly applied in our lab (Ruyssers et al., 2010). A modified Trendelenburg set-

up was used to induce peristaltic activity in isolated colonic segments (Holzer et al., 1998;

Deiteren et al., 2011). In short, following a laparotomy in the sacrificed animal, the colon was

quickly excised and transferred to ice-cold Krebs-Ringer solution (118.3 mM NaCl, 4.7 mM KCl,

1.2 mM MgSO4, 1.2 mM KH2PO4, 2.5 mM CaCl2, 25 mM NaHCO3, 0.026 mM CaEDTA and 11.1

mM glucose) and luminal contents were flushed from the colon. The proximal and distal

segment were mounted in an organ bath filled with Krebs-Ringer solution (37°C, 95% O2, 5%

CO2), with the oral side connected to a pressure transducer for continuous measurement of

intraluminal pressure and intraluminal infusion of warm Krebs-Ringer solution, and the aboral

side to an adjustable outlet that raises in height. The outlet was gradually raised to 7.5 cm, and

repetitive colonic peristaltic contractions were recorded in control and septic animals. The

amplitude as well as frequency of the contractions was studied in septic as well as control

animals.

3.5 Assessment of inflammation

3.5.1 Complete cell blood count

Blood was obtained via cardiac puncture from terminally sedated animals in an EDTA-treated

tube (BD Vacutainer K3EDTA tube, 3 mL) and used to obtain a complete cell blood count (CBC).

A cell blood count and white blood cell differential were obtained with the Advia®120

Haematology Analyzer using the Perox method.


58

3.5.2 Serum and colonic cytokines – protein level

Serum samples were obtained via cardiac puncture in terminally sedated animals

(ketamine/xylazine). Samples were centrifuged for 5 min at 10000 rpm at room temperature

following a 30 min clotting time, and supernatants were stored at -80°C until further analysis.

Serum cytokine levels were determined by means of a Cytometric Bead Array (CBA) (BD

Biosciences, Belgium) using the BD Accuri C6 flow cytometer and FCAP Array software

according to the manufacturer’s instructions for IL-1α, IL-1β, IL-2, IL-4, IL-6, IL-10, IL-12p70, IL-

17A, IFN-γ and TNF-α. TGF-β was determined in serum samples using the Lifetechnologies

Mulitspecies TGF-β ELISA kit according to the manufacturer’s instructions.

To quantify colonic cytokine concentration at the protein level, whole colons were rinsed with

phosphate buffered saline, blotted dry, weighed and placed in an ice-cold Tris-EDTA buffer

(phosphate buffered saline with 10 mM Tris, 1 mM EDTA and 0.5% v/v Tween20) containing a

protease inhibitor cocktail (Sigma-Aldrich) at 100 mg of colonic tissue per mL of buffer. Samples

were subsequently minced, homogenized and centrifuged (11.000 rpm, 4°C, 10 min) and

supernatants were collected and stored at -80°C until further analysis.

Protein levels in the serum were expressed as pg/mL serum, whereas cytokines at the colonic

level were expressed as pg/100 mg colonic tissue.

3.5.3 Colonic cytokines – mRNA level

To determine cytokine content at the mRNA level, total RNA was isolated from a snap-frozen

piece of colonic tissue using the Qiagen RNeasy Mini Kit and purity confirmed using the

Nanodrop ND-1000 Spectrophotometer. Total RNA was treated with DNase to obtain DNA-free

RNA (Turbo DNase-free, Life Technologies) and converted to cDNA using the Transcriptor First

Strand cDNA Synthesis Kit (Roche Applied Science). Quantitative real-time PCR was performed

using the TaqMan® Universal PCR Master Mix (Life Technologies) and primers as mentioned in

table 3.2. Several housekeeping genes were included in the analysis (glyceraldehyde 3-

phosphate dehydrogenase (GAPDH), β-actin, 18S ribosomal RNA (18S rRNA) and eukaryotic

elongation factor 2 (eEF2)) to which the expression of genes were normalized against. The PCR

reaction was performed in a 25µl reaction (Bustin et al., 2009; Ledeganck et al., 2013), with the

following amplification parameters: 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of

95°C for 15 sec and 60°C for 1 min.


59

3.5.4 Myeloperoxidase activity assay

Myeloperoxidaseactivity (MPO-activity) was determined in order to measure the degree of

tissue inflammation, as MPO is an estimate of neutrophil influx in the GI tract (Ruyssers et al.,

2009). Intestinal segments were isolated, blotted dry, weighed and placed in a potassium

phosphate buffer (pH = 6.0) containing 0.5% hexadecyltrimethylammoniumbromide at 5 g of

tissue per 100 mL of buffer. Samples were then placed on ice, minced, homogenized for 30 s

and subjected to two sonication and freeze-thawing cycles. Subsequently, the samples were

centrifuged for 20 min at 4°C at 15,000 rpm. 100 µL of the supernatant was added to 2.9 mL of

o-dianisidine solution (16.7 mg of o-dianisidine dihydrochloride (Sigma-Aldrich, USA) in 1 mL of

methyl alcohol (VWR, France), 98 mL of a 50 mM potassium phosphate buffer, pH 6.0, and 1

mL of a 0.05% H2O2 solution (Merck, Germany) as a substrate for the MPO enzyme). The

change in absorbance was read 5 min following addition of the sample at 460 nm with a

Spectronic Genesys 5 spectrophotometer (Milton Roy). One unit of MPO activity equals the

amount of MPO that is necessary to degrade 1 µmol H2O2 per minute to H2O at 25°C.

3.6 Assessment of colonic permeability

3.6.1 The Evans blue method (colonic permeability)

Colonic permeability was quantified by means of the Evans blue (EB) method, a dye that

crosses epithelia via the paracellular route (Lange et al., 1994). Animals were anesthetized with

a more concentrated mixture of ketamine (90 mg/kg i.p.) and xylazine (10 mg/kg i.p.), in a way

that animals would not regain consciousness for the remainder of the final experiment. Mice

were then placed in the supine position on a heating pad to undergo an abdominal incision.

Following abdominal disinfection, the colon was visualized and ligated distally of the ileocecal

valve and proximally of the anal verge. One hundred µL of a 9% EB solution (w/v) dissolved in

PBS was injected into the colon with a Myjector U-100 insulin syringe, and the abdomen was

closed in layers. One hour following the injection mice were sacrificed by means of cardiac

exsanguination, colons resected and flushed thoroughly with PBS containing 6 mM of

acetylcysteine to flush out residual EB. Colons were blotted dry, weighed and incubated for 24h

in formamide (50°C, 95% O2, 5% CO2) to extract the EB absorbed by the GI wall as a measure

for colonic permeability. The extracted amount of EB from the colon was determined

spectrophotometrically by measuring emission at 610 nm and expressed as µg EB/100 mg

colonic tissue.


60

3.6.2 Tight junction proteins

As elaborately described in the introductory chapter, cell-cell adhesion consists out of an

impressive number of extracellular and intracellular protein complexes that anchor cells to one

another, to the surrounding tissue layers and to the intracellular cytoskeleton of a single cell.

To quantify the mRNA level of several of these tight junction proteins, the isolation of total

RNA was performed as described above. Primers are listed in table 3.2.

3.7 Quantification of bacterial load

To estimate the bacterial load into the bloodstream, one drop of EDTA-treated full blood was

obtained by cardiac puncture from the animals in which colonic permeability was studied, and

plated onto a blood agar culture plate following enrichment and incubated at 37°C for 24h in

ambient air supplied with 5% CO2. Additionally, mesenteric lymph nodes (MLN) were resected

aseptically, suspended in RPMI 1640 medium (Gibco, LifeTechnologies) and mashed manually

using a 10 mL syringe plunger through a 40 µm nylon cell strainer. Homogenized MLN were

plated onto a blood agar culture following enrichment. These experiments were performed by

the lab of Medical Microbiology, University of Antwerp, Belgium.

3.8 Histology and immunohistochemistry

As macroscopic examination revealed no obvious defects in colonic segments obtained from

septic or control animals (no ulcerations, strictures or hyperemia whatsoever), we only

assessed inflammation of GI tissues at the microscopic level. A full thickness segment (0.5x0.5

cm) was taken from the proximal colon immediately adjacent to the cecum. The segment was

fixed for 24h in 4% formaldehyde and subsequently embedded in paraffin. Transverse sections

(5 µm) were stained with haematoxylin and eosin. Inflammation was scored by assessing the

presence and degree of inflammatory infiltrates, presence of goblet cells, architecture of the

crypts and presence of mucosal erosion as previously published (Heylen et al., 2014), resulting

in a cumulative score ranging from 0 (minimum) to 13 (maximum).

Stainings for lymphocytes (CD3), macrophages (Mac-3) and neutrophils (diaminobenzidine

(DAB, detection of hydrogen peroxide) were performed as follows: paraffin-embedded sections

were deparaffinized, endogenous peroxidase was blocked (5 min) and samples were treated

with a trypsin solution at 37° (10 min) prior to antigen retrieval in a citrate buffer (microwave,


61

pH 6.0). Tissue slices were exposed to anti-CD3 antibody (1:300 dilution), anti-Mac-3 antibody

(1:2000 dilution), anti-DAB (commercial kit) or the ApopTag Plus Peroxidase in Situ Apoptosis

Kit (Millipore S7101) respectively overnight and subsequently incubated with a biotinylated

anti-rabbit secondary antibody. All antibodies were obtained from Abcam.

Table 3.2. Taqman primers used in the aforementioned protocols Protein Entrez Gene id

Interleukin-6 (IL-6) 16193 - Mm00446190_m1 Tumor Necrosis Factor-α (TNF-α) 21926 – Mm00443258_m1 Interleukin-10 (IL-10) 16153 - Mm00439614_m1 Interleukin-17A (IL-17) 16171 - Mm00439618_m1 Interleukin-4 (IL-4) 16189 - Mm00445259_m1 Interleukin-1α (IL-1α) 16175 - Mm00439620_m1 Interleukin-1β (IL-1β) 16176 – Mm00434228_m1 Tumor growth factor-β1 (TGF-β1) 21803 - Mm01178820_m1 Interferon-γ (IFN-γ) 15978 - Mm01168134_m1 Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) 14433 - Mm99999915_g1 β-actin 11461 - Mm00607939_s1 18S ribosomal RNA (18S rRNA) 19791 – Mm03928990_g1 eukaryote elongation factor 2 (eEF2) 13629 - Mm00833287_g1 Occludin 18260 - Mm00500912_1 Claudin-1 12737 - Mm00516701_m1 Claudin-2 12738 - Mm00516703_s1 Desmoglein-2 13511 - Mm00514608_m1 E-cadherin 12550 – Mm01247357_m1 Zonulin-1 21872 - Mm00493699_m1 cFos 14281 - Mm00487425_m1 N-methyl-D-aspartate (NMDA) 26012 - Mm00480341_m1 calcitonine-gene-related peptide (CGRP) 12310 - Mm00801463_g1 transient receptor potential cation channel V1 (TRPV1) 193034 - Mm01246300_m1 transient receptor potential cation channel A1 (TRPA1) 277328 - Mm01227437_m1 µ-opioid receptor (Oprm) 18390 - Mm01188089_m1 δ-opioid receptor (Oprd) 4985 - Mm01180757_m1 κ-opioid receptor (Oprk) 18387 - Mm01230885_m1

3.9 Cell culture experiments

Bone marrow-derived macrophages and conventional dendritic cells were cultured in vitro. In

short, 0.5x106 bone marrow derived stem cells were plated in full RPMI medium supplied with

5% fetal bovine serum, 1% penicillin/streptomycin , 1% L-glutamine, 1mM sodium pyruvate,

20U/mL Polymyxin B (all from Gibco - Lifetechnologies, USA) and incubated for 7 days (37°C,

95% O2, 5% CO2) with macrophage-colony stimulating factor (obtained from L-cell conditioned


62

medium) to obtain macrophages, or granulocyte and macrophage-colony stimulating factor to

obtain immature conventional dendritic cells (cDC). At day 7, cultures were stimulated with 100

ng/mL LPS and 100 U/mL interferon-gamma (IFN-γ) to polarize the macrophages towards a

proinflammatory state, or 1 µg/mL LPS and 1000 U/mL IFN-γ to obtain mature cDC. After 24h,

supernatants were collected and stored at -80°C until further analysis. The purity of bone-

marrow derived macrophages and cDCs in culture was verified using a BD Accuri C6 flow

cytometer (CD11b+F4/80+ cells and CD11c+MHCII+ cells respectively).

3.10 Data presentation and statistical analysis

Data are represented as mean ± SEM with n being the number of animals per group, unless

specifically stated otherwise. Data were analyzed using the Independent Student’s t-test, one-

way or two-way ANOVA, followed by the Student-Newman-Keuls post hoc test when

applicable. Non-parametric testing was used in case of ordinal variables or if data were not

normally distributed. Data from afferent signaling and PET/CT imaging were analyzed using

Generalized Linear Models (GLM), followed by least significant difference (LSD) post-hoc test

when appropriate. The proportions of different units of afferent fibers as well as the number of

positive cultures from blood or mesenteric lymph nodes were compared using the Chi-square

with post-hoc Fisher’s exact test. A p-value ≤ 0.05 was considered to be statistically significant.

All data analyses were performed using SPSS version 20 (IBM, Chicago, USA) and visualized

using GraphPad Prism version 5.00.

Chapter 4

A characterization study of the cecal ligation and puncture

model

Based upon:

Nullens S, De Man JG, Bridts C, Ebo D, Francque SF, De Winter BY.

Identifying new therapeutic possibilities for sepsis research: a characterization study

of the cecal ligation and puncture model.

Submitted to Journal of Leukocyte Biology (July 2016)

Chapter 4 Characterizing the cecal ligation and puncture model

64

4.1 Abstract

Background and objectives: The gastrointestinal (GI) tract remains a somewhat neglected

organ system in the battle against sepsis. However, the disturbed gut functions can maintain

and contribute to the septic state, as the impaired motility and increased mucosal permeability

can predispose to the translocation of intraluminal contents and bacterial antigens. In order to

identify new therapeutic targets, we are in need of an animal model that adequately mimics

the immunological changes in the GI tract that occur during human sepsis.

Methods: OF-1 mice were subjected to the cecal ligation and puncture procedure (CLP), the

gold standard animal model concerning sepsis research, in order to induce sepsis. Sham-

operated animals served as controls. Animals were sacrificed at day 2 and 7 post-procedure,

during which GI motility was assessed, and pro- as well as anti-inflammatory cytokines were

measured in the serum as well as at the colonic level. In the spleen, lymph nodes, ileum and

colon, different subsets of T cells, dendritic cells and other leukocyte populations were

identified by means of flow cytometry.

Results: Septic animals consistently displayed clinical signs of disease at day 2 and day 7

following the septic insult, as well as a severe impairment in the gastrointestinal motility. Two

days post-CLP, septic animals were characterized by a profound inflammatory state with

increased serum and colonic levels of proinflammatory cytokines such as IL-6 and TNF-α among

others. Flow cytometry revealed an influx of neutrophils in colon and ileum, as well as

increased numbers of macrophages in spleen and mesenteric lymph nodes, and a rise in the

number of mast cells in all analyzed tissues. At day 7 post-CLP, a profound lymphocyte

depletion was observed in all tissues studied in concurrence with profound hypothermia and

increased IL-10 and TGF-β levels, indicative of immune suppression. However, an increase in IL-

17A and IFN-γ secretion from the colon coinciding with an increase in proinflammatory CD11b+

dendritic cells could also be seen, reflecting simultaneous altered activation of several

proinflammatory players. Furthermore, the recruitment of neutrophils, monocytes and natural

killer cells into the colonic lamina propria was striking.

Conclusions: The aforementioned observations correspond well with those made previously in

human sepsis patients; however no clear distinction can be made between mere pro- and anti-

inflammatory state, as a simultaneous activation of both was witnessed at both time points in

different tissues. This makes the CLP-procedure a qualified model to study the pathological

alterations that occur at the GI tract level during sepsis, but several pivotal players need always


65

be taken into account during further research. The CLP-model could furthermore aid us in the

identification of new therapeutic targets that specifically target the impaired functions of the

gastrointestinal tract.


66

4.2 Introduction

Ileus, the inhibition of the propulsive gastrointestinal (GI) motility of the entire GI tract, is a

common complication often observed following surgical manipulation (so called post-operative

ileus or ‘POI’) of the GI tract or during sepsis. Sepsis on the other hand remains a leading cause

of mortality in Intensive Care Units; ileus not only results from sepsis, but will contribute to and

maintain it together with the occurrence of mucosal barrier dysfunction by the translocation of

bacteria and (bacterial) antigens (MacFie et al., 1999; Hassoun et al., 2001; Deitch et al., 2010).

Neuronal players as well as inflammation play a pivotal role in this initiation and maintenance

of ileus (Bauer et al., 2002; Boeckxstaens & De Jonge, Gut 2009; De Winter & De Man, 2010).

On the one hand immune cells are able to secrete mediators that directly affect smooth muscle

cells or activate neuronal reflex pathways, resulting in motility disturbances, pain and/or

hypersensitivity. This disturbance in the immune homeostasis may thus result in inflammation

(MacDonald et al., 2011). On the other hand can neurotransmitters secreted from neurons

affect immune cells; this feature has been demonstrated for mast cells, T cells and dendritic

cells (Van Nassauw et al., 2007; Pacheco et al., 2010; Nijhuis et al., 2010; Kalff et al., 1998;

Engel et al., 2010).

The pathophysiology of post-operative ileus (POI) has been extensively unraveled by many

research groups (De Jonge et al., 2004; Boeckxstaens & De Jonge, 2009; Bauer & Boeckxstaens,

2004; De Winter et al., 1997). The role of the aforementioned mast cells and macrophages has

been elaborately studied (Bauer & Boeckxstaens, 2004; Overhaus et al., 2004; De Winter et al.,

2012), however much remains to be elucidated on the role of other immunocytes, such as T

cells, in the GI tract during sepsis-induced ileus. Administration of endotoxins or

lipopolysaccharide (LPS) to rodents elicits impairment in the GI transit (De Winter et al., 2002;

Li et al., 2010). Liu et al. (2009) demonstrated that rats challenged with endotoxin displayed

mucosal injury to the small intestine, with a decrease in the number of dendritic cells (DCs),

cytotoxic T cells and B cells, together with an increase in the number of regulatory T cells

(Tregs) and apoptotic lymphocytes.

Even more interestingly, the group of Bauer showed in a rat model of POI that the

translocation of endogenous intraluminal endotoxins and gut-derived bacterial products

contribute to the development of surgical manipulation-induced ileus (Turler et al., 2007),

clearly linking both clinical syndromes. There is also a major role in the development of


67

endotoxin-induced ileus for oxidative and nitrosative stress (Wirthlin et al., 1996; De Winter et

al., 2002; De Winter et al., 2005; Eskandari et al., 1999), and both hallmarks have been

demonstrated to be of importance in POI as well (Kalff et al., 2000). Finally, inflammation and

leukocyte recruitment play a major role during endotoxemia-induced ileus (Turler et al., 2002),

a feature that is also present in POI (Kalff et al., 1999). Both disease states thus share many

common particular characteristics.

As was already discussed in Chapter 1, results obtained from animal research in sepsis are

often difficult to extrapolate towards the human clinical setting (Rittirsch et al., 2007; Dyson &

Singer, 2009; Pene et al., 2015). The cecal ligation and puncture (CLP) model, by many

considered to be the gold standard animal model with regard to sepsis research, adequately

mimics the hemodynamic and metabolic features of the ‘average’ human septic patient

(Rittirsch et al., 2009). Data on the development of CLP-induced GI inflammation and motility

disturbances are limited. Overhaus et al. (2004) elegantly showed that performing the CLP-

procedure in rats resulted in impaired GI motility, in vivo by means of the FITC-dextran method

as well as in vitro by means of organ chambers experiments, presumably caused by an influx of

neutrophils and monocytes, with subsequent increased levels of inducible nitric oxide synthase

(iNOS), interleukin (IL)-6, IL-1β and monocytes chemoattractant protein-1 (MCP-1). Most

research however is limited to the study of the muscularis externa, as it harbours a dense

network of immune cells in direct interplay with smooth muscle cells (Vilz et al., 2012). The

lamina propria however constitutes a major locus of antigen presentation in the GI tract

(Weigmann et al., 2007; Sheridan & Lefrançois, 2012), and should therefore not be overlooked

in the pathogenesis of sepsis-induced ileus.

The simultaneous occurrence of increased mucosal permeability together with ileus can

facilitate translocation of GI luminal contents, bacteria and inflammatory mediators into the

bowel wall, and subsequently into the mesenteric lymph and bloodstream (Deitch, 2012;

Königsrainer et al, 2011). Further study of the neuroimmune GI environment during sepsis

could point researchers towards new therapeutic perspectives that tackle the impaired

mucosal barrier function and/or disturbed GI motility.

We therefore aimed to fully characterize the gastrointestinal neuroimmune environment in the

cecal ligation and puncture model in time. Gastrointestinal motility was measured, as well as


68

inflammation (systemically in the blood as well as locally in the colon). Flow cytometry was

performed on spleen, mesenteric lymph nodes, ileum and colon to further characterize the

immune cell subsets that could be of importance in the development of sepsis-induced ileus.

Firstly, we specifically aimed to identify as many subsets of immune cells as possible within a

single tissue sample, focusing on the different types of T cells (CD4+ helper T cell, CD8+ cytotoxic

T cell and CD4+CD25+Foxp3+ regulatory T cells), as well as on the different subsets of dendritic

cells (DCs) in the gut. The latter cell type plays a major role in the initiation of adaptive immune

responses and the skewing of the immune response; in short, CD103+CD11b- DCs are known to

induce the development of regulatory, immune suppressing T cells (Tregs), whereas

CD11b+CD103- DCs are suggested to drive proinflammatory Th1 responses (del Rio et al., 2010;

Bekiaris et al., 2014; Schraml & Reis e Souza, 2015). Secondly, we aimed to study the

aforementioned in a specific time frame that will be later on described in this chapter.

Finally, by modulating the CLP-procedure, we hope to obtain an animal model that exhibits the

two different immune phases that have been typically described in human sepsis, namely the

systemic inflammatory response syndrome (SIRS) followed by the compensatory anti-

inflammatory response syndrome (CARS) or anergic phase (Hotchkiss et al., 2013), so that later

on we can investigate compounds specific to the immune phase.


69

4.3 Material and methods

4.3.1 Animals

Eight week old male Swiss OF-1 mice were obtained from Charles River (France) and were

allowed to acclimatize for two weeks in their cages under specific conditions with unlimited

access to regular chow and tap water. All experiments listed below were approved by the

Ethical Committee on Animal Experiments by the University of Antwerp (file number 2012-42

and 2012-42-extension).

For the flow cytometry experiments, two batches of animals were included in the experimental

set-up. In the first group, the T and B-cell stain was performed in gastrointestinal tissues, and

cytokines were determined in blood and colonic tissues by means of real-time RT-PCR and

Cytometric Bead Array (vide infra). The second group of animals was subsequently utilized for

the remainder of the staining panels, as even the most optimized isolation-protocol of

leukocytes from the lamina propria layer from small bowel and colon (the so called lamina

propria mononuclear cells or ‘LPMC’) will only yield a limited number of LPMC from each

animal, with mice aged 8 to 20 weeks yielding the highest number of LPMC (e.g. 2 to 5 million

LPMC from one ten-week old C57Bl/6 mouse; Sheridan & Lefrançois, 2012).

4.3.2 The cecal ligation and puncture animal model of sepsis

Sepsis was induced by means of CLP as described in Chapter 3 (Rittirsch et al., 2009; Buras et

al., 2005; Wintersteller et al., 2012). In short, OF-1 mice were anesthetized using a mixture of

ketamine and xylazine and placed in the supine position. A midline laparotomy was performed,

the cecum was exteriorized, and cecal contents were gently pushed towards the distal cecum.

Different approaches to the CLP-model were implemented in order to obtain a reproducible

animal model without notable mortality: the cecum was either ligated at 50 or 75% of its

length with a 4/0 silk thread, and subsequently punctured once or twice through-and-through

with a 21, 23 or 25G needle. The cecum was subsequently gently manipulated as to protrude a

consistent small amount of stool from the puncture holes. The ligated cecum was repositioned

into the abdominal cavity, and the abdomen was closed in layers. Sham-operated animals

served as controls. Mice received fluid resuscitation and pain relief. Mice were allowed to

recover in a heated cage with free access to water. Sham-operated mice served as controls.


70

4.3.3 Survival analysis and clinical disease score

Mice were monitored up until 14 days following the CLP- or sham-procedure. Animals were

weighed twice daily (8 am and 7 pm) and assessed individually for signs of illness by means of a

modified clinical disease score (CDS) (Heylen et al., 2013; De Winter et al., 2009). The presence

of piloerection, grooming behaviour, mobility, signs of peritoneal irritation, direction of the

ears, presence of conjunctivitis and consistency of stool were assessed, as well as whether

animals appeared to be anemic or moribund, resulting in a CDS ranging between 0 (no signs of

illness) and 15 (maximum). Mice were prematurely sacrificed when they lost over 15% of their

baseline body weight, when they appeared moribund or had a clinical disease score (CDS) > 8

(see also Table 3.1 in Chapter 3).

4.3.4 Final experimental design

Based upon the aforementioned survival analysis and clinical parameters (vide infra – Results),

two time points were included for the subsequent analyses. Animals were studied either 48h

(CLPd2) or 7 days (CLPd7) following the procedure. Control animals were studied 2 days

following the procedure (sham).

In a first set of experiments, GI motility was assessed by means of the solid beads method (n =

8-10/group) in order to ascertain the occurrence of ileus, whereupon animals were

anesthetized and sacrificed by means of cardiac puncture. Whole blood samples were utilized

to obtain a cell blood count and white blood cell differential, and tissue samples from colons

were harvested for histology.

In a second set of experiments, serum samples obtained by cardiac puncture were used for

cytokine analysis. Spleens and draining mesenteric lymph nodes (MLN) were harvested for flow

cytometric analysis (n = 10-12 in each group). Furthermore, lamina propria mononuclear cells

(LPMC) were isolated from ileum and colon at the same time for flow cytometry (vide infra).

4.3.5 Measurement of gastrointestinal transit

4.3.5.1 The Evans blue method

Following overnight fasting with unlimited access to tap water, animals were given a gavage

with 100 µL of Evans blue (50 mg/mL in 0.5% methylcellulose) during a short diethyl ether

anesthesia with permission from the local Ethics Committee on Animal Experiments, as other

anesthetics typically delay GI transit, whereas diethyl ether does not (De Winter et al., 1997).


71

Fifteen minutes following the gavage, animals were sacrificed, the distal esophagus and pylorus

were ligated and the gastrointestinal tract (stomach and small bowel) was resected and divided

into 6 segments (stomach and small bowel 1 through 5). The amount of dye in each segment

was measured spectrophotometrically (emission 565 nm), and the percentage of gastric

emptying (%GE) and the geometric center of transit (GC, score 1-6) was calculated by means of

the formulas mentioned in Chapter 3 (paragraph 4 – Assessment of gastrointestinal motility).

4.3.5.2 The solid beads method

Mice were overnight deprived of food with unlimited access to tap water. A gavage was given

with 0.5 ml of tapwater containing 25 glass green-colored beads (diameter 0.4 – 0.5 mm)

through a 20G flexible catheter (Terumo; outer diameter 1.10 mm, inner diameter 0.80 mm).

Mice were sacrificed 30 min, 2h or 6h following gavage and the GI tract was resected and

divided into 10 parts (stomach, 5 small bowel segments, cecum, proximal colon, distal colon

and feces). The number of beads in every segment was counted under a stereomicroscope for

calculation of percentage gastric emptying (%GE) and the geometric center of intestinal transit

(GC) as a marker for overall GI transit (Seerden et al., 2007).

4.3.5.3 The bead expulsion test

Following a short anesthesia with diethyl ether, one glass bead was inserted rectally by means

of a small rod and pushed in the oral direction up until 2 cm from the anal verge. Animals were

then monitored individually in separate cages until expulsion of the bead.

4.3.5.4 In vitro study of colonic peristaltic activity

Following a laparotomy in the sacrificed animal, the colon was quickly excised and transferred

to ice-cold Krebs-Ringer solution, luminal contents were flushed from the colon and the

proximal and distal segment were mounted in an organ bath filled with Krebs-Ringer solution

(37°C, 95% O2, 5% CO2). The outlet was gradually raised to 7.5 cm, and repetitive colonic

peristaltic contractions were recorded in control and septic animals. The amplitude as well as

frequency of the contractions was studied in septic as well as control animals.


72

4.3.6 Peripheral blood count

Following anaesthesia, animals were sacrificed 48h or 7 days following CLP or sham-procedure

with cardiac puncture while obtaining EDTA-treated blood samples. A cell blood count and

white blood cell differential were obtained with the Advia®120 Haematology Analyzer using

the Perox method.

4.3.7 Cytokine measurements

Blood samples (Multivette® 600 capillary blood collection, Sarstedt) were centrifuged (10000

g, 5 min, 20°C) and supernatants were stored at -80°C until further analysis by means of the

Cytometric Bead Array (CBA, BD) according to the manufacturer’s instructions.

Colonic cytokine levels were determined at the protein as well as the mRNA level. For the

levels of secreted protein, whole colons were rinsed with phosphate buffered saline, blotted

dry, weighed and placed in RPMI medium supplemented with 2mM glutamine, 100 U/ml

penicillin, 100 µg/ml streptomycin and 10% fetal bovine serum (“full RPMI”) and incubated for

24h (37°C, 95% O2, 5% CO2). Supernatants were collected 24h later and assessed for levels of

IL-6, TNF-α and IL-10 (pg/g colon) using the BD CBA Mouse Cytokine Kit.

To determine cytokine content at the mRNA level, total RNA was isolated from a snap-frozen

piece of proximal colonic tissue as described in Chapter 3 (paragraph 3.5.3 – Colonic cytokines-

mRNA level). In short, purified total RNA was treated with DNase and converted to cDNA.

Quantitative real-time PCR was performed using the TaqMan® Universal PCR Master Mix (Life

Technologies), and primers as mentioned previously. Out of three housekeeping genes

included in the PCR-reaction, GAPDH was determined to be the optimal housekeeping gene to

which the expression of genes of interest were normalized. The PCR reaction was performed in

a 25 µL volume reaction (Ledeganck et al., 2011; Heylen et al., 2014; Livak & Schmidt, 2001;

Bustin et al., 2009), with the following amplification parameters: 50°C for 2 min, 95°C for 10

min, followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min.

4.3.8 Flow cytometric characterization of gastrointestinal tissues

4.3.8.1 Instrument characteristics

All experiments were performed using the BD Accuri C6 flow cytometer (Becton Dickinson,

Erembodegem, Belgium). This flow cytometer is equipped with two lasers and four adjustable


73

optical filters (table 4.1), allowing the use of four different and adequately matched

fluorochromes in a multi-color flow cytometry experiment.

Table 4.1. Instrument specificities of the Accuri C6 flow cytometerTM.

BD Accuri™ C6 flow cytometer

488 nm and 640 nm laser, 3 blue 1 red configuration

Optical filter FL1 533/30 nm (FITC, BB515)

Optical filter FL2 585/40 nm (PE, PE-CF594)

Optical filter FL3 > 670 nm (PerCP-Cy5.5, PE-Cy7)

Optical filter FL4 675/25 nm (APC, AF647) Instrument specifics used for the flow cytometric characterization of the cecal ligation and puncture model. The suitable fluorochromes for each optical filter are mentioned between brackets.

4.3.8.2 Tissue preparation – spleen and MLN single cell suspension

Spleens and MLN were collected from animals sacrificed by cardiac exsanguination. Spleens

were excised aseptically from the left flank with careful removal of the surrounding vessels and

fat tissue. The mesenteric MLN, located in the mesenterium of the ascending colon and cecum

were carefully dissected free, and surrounding fat tissue was cut away (figure 4.1) (Van den

Broeck et al., 2006). All tissues were collected in ‘full’ RPMI medium (RPMI 1640 medium

(Gibco – Lifetechnologies, USA) supplemented with 10% fetal bovine serum, 1%

penicillin/streptomycin and 1% glutamine, all from Gibco – Lifetechnologies, USA), manually

dissociated using a 10 mL syringe plunger over a 40 µM cell strainer (Falcon®, BD) and rinsed

with RPMI followed by 5 min centrifugation at 4°C, 1500 rpm. Cell pellets were subsequently

incubated with 5 mL red blood cell lysis buffer (Sigma-Aldrich, USA) followed by washing and

suspension in RPMI and storage on ice until further processing of samples. Cells were counted

and viability was assessed using the Trypan blue staining method (Thermofisher Scientific,

Belgium), as Trypan blue only traverses the cell membrane when cells are no longer viable,

thus staining them blue (so-called ‘dye exclusion method’). The single cell suspensions

obtained were further on used for staining.


74

Figure 4.1 Schematic drawing (left) of the mesenteric lymph node stations that were prelevated for analysis. The MLN can be identified in the mesenterium in close proximity to the distal ileum and cecum. Left drawing obtained with permission from Elsevier (license number 3774120600400).

4.3.8.3 Tissue preparation – isolation of LPMC from

colon and ileum

Colonic and ileal lamina propria mononuclear cells (LPMC) were isolated as described

previously (Heylen et al., 2014; Setiawan et al., 2007; Lefrançois & Lycke, 2001; Weigman et al.,

2007; Sheridan & Lefrançois, 2012; Alpaerts et al., 2015). In short, ileal and colonic tissues were

opened longitudinally following meticulous removal of mesenterium and Peyer’s patches, and

rinsed of GI contents with calcium- and magnesium-free Hank’s balanced salt solution (HBSS,

Gibco – Lifetechnologies, USA). Tissues were cut into 5 mm pieces and incubated in an

Erlenmeyer flask for 20 min in HBSS supplemented with 1 mM EDTA (Ambion, USA) and 2 mM

dithiothreitol (Sigma-Aldrich, USA) at 37°C on a stir plate at 220 rpm in order to shed the


75

epithelial layer, followed by thorough washing with plain HBSS. This was repeated 4 to 6 times

depending on the turbidity of the supernatants until the supernatants were ‘clear’ (free of

epithelial cells). Tissues where then incubated in ‘lymphocyte growth medium’ (LGM): full RPMI

as described in the previous paragraph supplemented with 25 mM HEPES-buffer (Gibco –

Lifetechnologies, USA), 1 mM sodium pyruvate (Sigma-Aldrich, USA), 1% MEM non-essential

amino acids 100x, 2% MEM essential amino acids 50x (both from Gibco – Lifetechnologies,

USA) and 50 µM β-mercaptoethanol (Sigma-Aldrich, USA) containing 1 mg/ml collagenase

(collagenase from Clostridium histolyticum type I, 0.25-1.0 FALGPA units/mg solid, ≥125

CDU/mg solid; Sigma-Aldrich, USA) and 5 mM CaCl2 for enzymatic digestion. Tissues were

subsequently further dissociated mechanically using the plunger of a 1 mL syringe. To remove

debris, the suspension was sieved through wet gauze layered in a funnel. The cell suspension

was then washed with LGM (10 min, 1500 rpm) and layered onto a 30:70% Percoll (Sigma-

Aldrich, USA) gradient followed by centrifugation for 20 min at 3500 rpm at room temperature

without a brake. Cells were harvested at the Percoll interface, washed twice, resuspended in 2

mL LGM and stored on ice until further processing. Cells were counted and viability was

assessed using the Trypan blue staining method. The LPMC suspensions were then utilized for

staining.

4.3.8.4 Staining protocols

Six staining panels were designed in order to stain a substantial number of relevant immune

cell populations (table 4.2). In short, 1x106 cells were incubated for 10 min at 4°C with 50 µl rat

serum (1/50) to evade non-specific staining. Next, cells were washed and stained with 100 µl

staining mix for 40 min at 4°C in the dark. All antibodies were diluted 1/100, with the exception

of anti-mouse F4/80 APC (1/200) (tables 4.3 with cell surface markers and table 4.4 with the

different types of immune cells that were studied in these experiments). A separate propidium

iodide (PI) stain was included to assess for cell viability. Cell viability was typically > 90% in

single cell suspensions of spleen and MLN, and between 70 and 80% in the LPMC due to the

more aggressive isolation protocol. The samples were subsequently washed twice,

resuspended in 400 µl FACS buffer and kept on ice in the dark until analysis.


76

Table 4.2 Immune cell populations stained with the characterization experiments.

Antibody Manufacturer (reference) Clone

Panel 1 – dendritic cell stain Rat anti-mouse CD11b BB515 BD Horizon (564454) M1/70 Hamster anti-mouse CD11c PE-CF594 BD Horizon (562454) HL3 anti-mouse I-A/I-E (MHCII) PerCPCy5.5 Biolegend (107625) M5/114.15.2 anti-mouse CD103 APC Biolegend (121413) 2E7

Panel 2 – macrophage / mast cell stain Rat anti-mouse IgE FITC eBioscience (11-5992) 23G3 Rat anti-mouse CD117 PE eBioscience (12-1172) ACK2 Rat anti-mouse F4/80 APC eBioscience (17-4801) BM8

Panel 3 – T and B cell stain Rat anti-mouse CD3 FITC BD Pharmingen (561798) 17A2 Rat anti-mouse CD8a PE BD Pharmingen (553032) 53-6.7 Rat anti-mouse CD4 PerCPCy5.5 BD Pharmingen (550954) RM4-5 Rat anti-mouse CD19 APC BD Pharmingen (561738) 1D3

Panel 4 – regulatory T cell stain Rat anti-mouse CD25 FITC BD Pharmingen (561798) 17A2 Rat anti-mouse CD4 PerCPCy5.5 BD Pharmingen (550954) RM4-5 Rat anti-mouse Foxp3 AF647 BD Pharmingen (560401) MF23

Panel 5 – neutrophil / natural killer cell / monocyte stain Rat anti-mouse CD11b BB515 BD Horizon (564454) M1/70 Rat anti-mouse CD335 PE BD Pharmingen (560757) 29A1.4 Rat anti-mouse Ly6G PE-Cy7 BD Pharmingen (560601) 1A8 Rat anti-mouse Ly6C APC BD Pharmingen (560595) AL-21

The different antibodies used for the flow cytometry experiments.


77

Table 4.3 Cell surface markers used in the flow cytometric analysis of mouse leukocytes

Marker (CD) Signifies... Found on...

CD103 Integrin alpha E intraepithelial lymphocytes, lamina propria T cells, subset of gut mucosa and MLN dendritic cells

CD11b Integrin alpha M (macrophage-1 antigen, complement receptor 3)

innate immune cells (monocytes, macrophages, dendritic cells, natural killer cells)

CD11c Integrin alpha X dendritic cells, monocytes, macrophages

CD117 Mast/stem cell growth factor receptor / proto-oncogene c-Kit

mast cells, interstitial cells of Cajal

CD19 B-lymphocyte antigen CD19 B cells

CD25 α-chain interleukin-2 receptor activated T and B cells, regulatory T cells

CD3 complex consisting out of the γ, δand

two ε-chains of the T cell-receptor

T cells

CD335 Natural cytotoxicity triggering receptor 1 (activation receptor), synonym NKp46

NK cells

CD4 co-receptor of the T cell receptor (TCR) T helper cells, monocytes

CD8a co-receptor of the TCR, recognition of antigens presented by APCs

Cytotoxic T cells

Foxp3 regulator in the development and differentiation of immune suppressing regulatory T cells

Regulatory T cells

F4/80 cell surface glycoprotein murine macrophages

Ly6G unknown; neutrophil migration (?) neutrophils

I-A/I-E (MHCII) major histocompatibility complex II, found on APC

dendritic cells, mononuclear phagocytes, B cells

IgE Immunoglobulin E basophils, mast cells

Ly6C co-stimulatory signal for T cell activation, homing of cytotoxic T cells

monocytes

The different cell surface markers that were applied during the flow cytometry experiments in order to characterize different subsets of leukocytes. APC: antigen presenting cell; CD: cluster of differentiation; Ig: immunoglobulin; MLN: mesenteric lymph nodes; NK: natural killer; TCR:T cell receptor A simplified overview based upon current knowledge and adapted from Abbas et al., 2013.


78

4.3.8.5 Flow cytometric characterization of immune cells

Flow cytometric analysis was performed according to the MIFlowcyt guidelines (Lee et al.,

2008) using the BD Accuri™ C6 flow cytometer, of which the instrument’s specifications are

summarized in table 4.1. Compensation settings were included as appropriate, and the whole

sample was measured without a stopping gate (typically between 0.5x106 and 1x106 events per

sample). ‘Fluorescence Minus One’ samples (FMO) and unstained samples were included for

determining the correct negative populations and gate settings (Maecker & McCoy, 2010;

Hulspas et al., 2009). Results were analyzed using FCS Express 4 software (De Novo Software).

The gating strategy for the one of the panels is outlined in figure 4.2 as an example. In short,

for all samples, leukocytes were gated based upon their FSC-H and side scatter-H (SSC-H)

characteristics in order to exclude debris. Then, cell aggregates were excluded based upon the

forward scatter-area versus forward scatter-height plot (FSC-A/FSC-H). (Heylen et al., 2014;

Alpaerts et al., 2015; Van Craenenbroeck et al., 2014). Finally, cell populations were defined

based upon their expression of several cell surface markers, as described in table 4.4, and cell

subsets were expressed as a percentage of the total leukocyte count in each studied tissue.

4.3.9 Statistical analysis

Data are presented as mean ± SEM, with ‘n’ representing the number of mice. The One-way

ANOVA with post-hoc Dunnett analysis (the sham group was chosen as control) was applied to

compare the results of all the aforementioned parameters, or the Kruskal-Wallis test as non-

parametric equivalent. A p-value ≤ 0.05 was considered to be statistically significant. Data were

analyzed using SPSS version 20.0 (IBM, Chicago) and visualized using GraphPad Prism version

6.00. The flow cytometric data were visualized and analyzed using FCS Express 4 Flow, and the

CBA results were analyzed using FCAP Array (BD).


79

Figure 4.2 Example of a gating strategy for flow cytometry (panel 3 – T and B cell stain and panel 4 – Treg stain). Leukocytes were gated based upon their forward scatter (FSC) and side scatter (SSC) properties in order to exclude debris (A), followed by the exclusion of doublets based upon the forward scatter-height (FSC-H) versus forward scatter-area (FSC-A) plot (B). Next, the percentage of CD3+ cells (effector T cells) was identified in the singlet cells gate gated upon the expression of CD3 and SSC properties (C). Subsequently, the differentiation was made in this population between the number of CD4+ cells to identify helper T cells, or CD8+ cells to identify cytotoxic T cells (D). In the singlet gate, the percentage of CD4+ cells (helper T cells) was then identified (second staining panel)(E), and subsequently the CD25+Foxp3+ cells were identified in the CD4+ gate (F). In the leukocyte gate, the percentage of CD19+ cells was also determined as to identify B cells (G). Cell subsets were expressed as a percentage of total leukocyte count. FMO samples were implemented as appropriate gating controls. A: area; CD: cluster of differentiation; Foxp3: forkhead box p3; FSC: forward scatter; H: height; SSC: side scatter; Treg: regulatory T cells

A B

C

D

E

F

G


80

Table 4.4 Identification of immune cells in each panel based upon expression of cell surface markers

Antibody Cell population

Panel 1 – dendritic cell stain MHCII+CD11c+CD11b+ CD11b+ dendritic cells (“proinflammatory”) – [CD11b+DCs] MHCII+CD11c+CD103+ CD103+ dendritic cells (“anti-inflammatory”) – [CD103+DCs]

Panel 2 – macrophage/ mast cell stain F4/80+ macrophages [Mφ] IgE+CD117+ mast cells [MC]

Panel 3 – T and B cell stain CD3+CD4+ T helper cells CD3+CD8+ Cytotoxic T cells CD19+ B cells

Panel 4 – regulatory T cell stain CD4+CD25+Foxp3+ regulatory T cells [Tregs]

Panel 5 – neutrophil / natural killer cell / monocyte stain CD335+ natural killer cells [NKs] Ly6C+ monocytes Ly6G+ neutrophils

The different subsets of leukocytes that were identified during the flow cytometry experiments based upon the expression of cell surface markers, and their abbreviations as used in this chapter between square brackets.


81

4.4 Results

4.4.1 Survival analysis of the implemented CLP-model of sepsis

Several modifications of the CLP-model were studied in our lab prior to deciding which

modified procedure would be implemented in our lab. In the animals that had 75% of their

cecum length ligated, a drop in survival could be observed that was furthermore dependent on

the needle’s size that was applied, and the number of punctures (figure 4.4, upper panel). Two

punctures with a 21G needle resulted in 100% mortality by day 5, whereas 43% of the animals

that had received one puncture with a 25G needle survived up until two weeks following the

procedure. A milder version was subsequently tested, in order to minimize the number of

casualties. When 50% of the cecum’s length was ligated, all animals survived up until two

weeks past the procedure on the condition that they had received only one puncture (no

difference between the use of a 21G, 23G or 25G needle). Once again, mortality increased

when animals received two punctures, and mortality was highest in association with the large

bore needle of 21G (figure 4.4, lower panel). Based upon these results, we opted to further

characterize the model in which the cecum was ligated at 50% of its length with one 25G

puncture, as no mortality was witnessed in this group.

Interestingly, CLP50/1/25 animals displayed a biphasic course of their clinical signs of disease,

based upon their behavior and the fluctuations in their weight (figure 4.3). We therefore

choose to characterize the CLP-model mice at day 2 and at day 7 following the procedure.

Figure 4.3 Clinical disease score of the animals that underwent the CLP-procedure with the following specifics: 50% ligation, 1 puncture, 25G needle (green line), and in sham animals as control group (purple line). Note the biphasic course of disease. The dark yellow arrows denote the time points that were chosen for further investigation during the remainder of this dissertation. N = 5 animals per group.


82


83

Figure 4.4 (previous page) Cumulative survival of animals that underwent the CLP-procedure, or sham-operated control animals up until two weeks following the procedure. In the upper panel, the cecum was ligated at 75% of its length, whereas in the lower panel it was ligated at 50% of its length. Note the influence of the needle thickness on survival in the upper panel, with the large bore needles resulting in a higher mortality rate. When ligating 50% of the cecum’s length, no mortality was observed when only one puncture was performed independent of the needle size (lower panel). The needle size was then again of influence once two punctures were performed. N = 6-7 animals/group. CLP: cecal ligation and puncture; interpretation of the numeric code used following ‘CLP’: the first two numbers represent the percentage of the cecum that was ligated, the third number how many times the cecum was punctured, and the final two numbers the size of the needle, e.g. CLP50125 means CLP procedure in which the cecum is ligated at 50% of its length and punctured once (1) with a 25G needle.

4.4.2 Implementation of the CLP-model to study septic ileus and

gastrointestinal motility disturbances

Based upon results from the survival analysis and the clinical disease scores, we opted to

implement the CLP-model at our facilities during which 50% of the cecum’s length was ligated,

followed by one through-and-through puncture with a 25G needle (CLP50/1/25).

Septic animals displayed a significant increase in their clinical disease score at day 2 as well as

at day 7 following the CLP-procedure in comparison to sham-operated animals (figure 4.5,

upper left panel). The rectal temperature was only significantly decreased 7 days post-CLP

(figure 4.5, upper right panel). In order to limit the number of animals needed, GI motility was

only assessed by means of the solid beads method 2h following the gavage of the beads. A

significant decrease in %GE as well as the GC was observed both at day 2 and day 7 following

CLP (figure 4.5, lower panels). The bead expulsion test revealed a significant delay in bead

expulsion time in septic animals at day 2 (sham 738.33 ± 218.54 versus CLPd2: 1637.50 ±

202.61 seconds; p = 0.011, n = 6-8/group). In the CLPd7 group, none of the animals expelled

the bead within the first 24h (the beads were recovered in all 6 animals from the colon

following sacrifice at CLPd7).

Four days post-CLP however, in between both ‘sick’ phases, no differences were observed

concerning signs of disease or GI motility between septic and control animals applying the solid

beads method (figure 4.5).


84

Figure 4.5 Clinical disease score (upper left), rectal temperature (upper right) and gastrointestinal motility parameters (percentage of gastric emptying in lower left panel and geometric center in lower right panel, as determined by means of the solid beads method) following the CLP50/1/25-procedure at different time points post-procedure (day 2, day 4 and day 7). One-way ANOVA followed by post-hoc Dunnett testing (with sham as the control group), or its non-parametric equivalent as appropriate (Kruskall-Wallis with Dunns post hoc testing); n = 8-10/group, *p<0.05, **p<0.01, ***p<0.001.

Next, GI transit of a semiliquid meal was assessed by means of the Evans blue test.

Interestingly, no difference could be observed between the %GE (sham 26.73 ± 3.79%, CLPd2

30.03 ± 6.44%, CLPd7 25.95 ± 10.32%, p = NS) or the GC (sham 1.49 ± 0.09, CLPd2 1.59 ± 0.15,

CLPd7 1.40 ± 0.14, p = NS), as the sham-operated animals still displayed impaired GI motility to

liquids in contrast to the control animals used in the LPS-experiments.

Finally, in vitro recording of colonic motility was performed. The proximal colon peristaltic

pressure waves in control animals had a mean pressure amplitude of 3.29 ± 0.31 cm H2O and a

mean interval of 35.67 ± 4.65 sec (representative tracing displayed in figure 4.6). Colons from

septic animals did not display any spontaneous peristaltic activity.

Clinical Disease Score day 2

Sham

CLP

d2

CLP

d4

CLP

d7

0

2

4

6

8

1010

15

**

***

Cli

nic

al

Dis

ease S

co

re

Rectal temperature

Sham

CLP

d2

CLP

d4

CLP

d7

01020

25

30

35

40

***

Recta

l te

mp

era

ture

(°C

)

Percentage of gastric emptying

Sham

CLP

d2

CLP

d4

CLP

d7

0

50

100 ***

% G

E

Geometric Center

Sham

CLP

d2

CLP

d4

CLP

d7

0

1

2

3

4

5

68

10

******

Geo

metr

ic C

en

ter


85

Figure 4.6 Representative tracing of an in vitro colonic pressure recording from control animals (A) and septic animals 48h following the CLP-procedure (B), in response to increasing intraluminal pressures gradually increasing from 0 until 7.5 cm H2O. n = 6-7/group.

4.4.3 Peripheral blood count in the CLP-model

At CLPd2, animals displayed a marked drop in thrombocytes. The total number of leukocytes

remained unaltered, however an obvious lymphocytopenia and monocytosis developed. At day

7 following CLP, animals displayed a profound anemia, thrombocytosis and leukocytosis with

reduced number of neutrophils in comparison to the control animals (table 4.5). May-

Grunwald-Giemsa staining confirmed these findings (figure 4.7).


86

Figure 4.7. Representative May-Grünwald-Giemsa staining of sham animals (left), septic animals at day 2 (CLPd2 – middle) and at day 7 (CLPd7 – right). 200x magnification.

Table 4.5 Cell blood count and white blood cell differential of whole blood samples from septic and control animals

Parameter sham CLP day 2 CLP day 7

Hemoglobin (g/dl) 12.88 ± 0.74 11.97 ± 0.64 2.81 ± 0.73*** Erythrocytes (x106/µL) 8.32 ± 0.39 7.62 ± 0.42 1.78 ± 0.48** Thrombocytes (x103/µL) 1067.00 ± 64.17 517.6 ± 91.68* 1894 ± 196.6** Leukocytes (x103/µL) 2.21 ± 0.12 2.30 ± 0.30 7.23 ± 1.84* Neutrophils (%) 4.20 ± 0.58 4.87 ± 0.50 2.33 ± 0.62* Lymphocytes (%) 59.12 ± 6.98 23.30 ± 3.12* 22.19 ± 2.97* Monocytes (%) 30.24 ± 7.62 63.24 ± 3.42* 60.79 ± 4.42* Eosinophils (%) 0.40 ± 0.11 0.22 ± 0.10 0.29 ± 0.07 Basophils (%) 1.50 ± 0.83 1.87 ± 0.86 0.44 ± 0.10 “Undifferentiated/immature” 5.56 ± 0.91 7.56 ± 1.45 11.88 ± 1.39* Cell blood count of whole blood obtained by cardiac puncture in septic (CLP day 2 or CLP day 7) and control animals (sham); the different subsets of leukocytes are expressed as percentage of the leukocyte population. One-way ANOVA with post-hoc Dunnett test or Kruskall Wallis with post-hoc Dunn’s test as appropriate. N = 6/group.


87

4.4.4 Cytokine measurements in the CLP-model

Serum cytokine analysis confirmed the development of a profound proinflammatory state 48h

following CLP, as was demonstrated by a significant rise in the levels of IL-6 and TNF-α

compared to non-septic control animals. The concentrations normalized to baseline levels 7

days following the procedure, however at this time serum IL-17A levels were significantly

upregulated. These results were confirmed at the colonic level, albeit not significantly, with a

tendency towards higher concentrations of IL-17A and IFN-γ secreted in the supernatants.

Furthermore, the secreted TNF-α concentration remained significantly upregulated by day 7.

mRNA expression of the aforementioned results in the colon corresponded with levels

measured in the colonic supernatants (table 4.6).

Table 4.6 Cytokine levels in serum and colon

sham CLP day 2 CLP day 7 p-value

A. Serum cytokines (pg/mL) IL-6 2.57 ± 0.48 243.60 ± 54.89* 35.39 ± 10.97 <0.0001 IL-10 0.59 ± 0.32 2.33 ± 1.25 2.82 ± 1.82 NS TNF-α 4.95 ± 0.29 63.72 ± 21.25* 23.89 ± 3.61 0.008 IL-2 <L/D 0.38 ± 0.16 <L/D / IL-17A 0.21 ± 0.05 0.35 ± 0.71 3.57 ± 0.61* 0.006 IFN-γ <L/D 1.12 ± 0.44 0.89 ± 0.53 NS TGF-β 152.50 ± 14.33 194.10 ± 12.18* 198.10 ± 13.68* 0.04

B. Colonic cytokines (pg secreted/g colonic tissue) IL-6 103.3 ± 18.95 1904 ± 874.6 3410 ± 1429 0.1108 IL-10 0.79 ± 0.39 6.08 ± 1.47* 5.42 ± 1.71* 0.0163 TNF-α 2.53 ± 0.76 6.57 ± 1.46* 12.38 ± 3.84* <0.0001 IL-2 <L/D <L/D <L/D / IL-17A 0.87 ± 0.29 0.31 ± 0.10 2.43 ± 1.05 0.06 IFN-γ 0.76 ± 0.34 1.53 ± 0.47 8.09 ± 4.77 0.141

C. mRNA colonic cytokines IL-6 1.27 ± 0.45 28.32 ± 14.78 4.14 ± 1.78 0.096 TNF-α 1.00 ± 0.03 3.04 ± 0.78* 10.29 ± 0.49* <0.001 IL-10 1.06 ± 0.15 9.59 ± 6.30 4.83 ± 1.66 0.317 IL-17A 1.19 ± 0.30 1.59 ± 0.68 19.54 ± 6.97* 0.004 IFN-γ 1.42 ± 0.49 0.80 ± 0.15 2.73 ± 0.73* 0.007 Cytokine levels in serum and colonic supernatants, as determined by CBA (or ELISA for TGF-β) for protein content and real-time RT-PCR for mRNA content in colonic tissue. One-way ANOVA followed by Dunnett post-hoc testing or its non-parametric equivalent as appropriate; *p < 0.05 when compared to the sham group. For the PCR-results, data are expressed as relative expression (2-ΔΔCT method) and the sham group (left column) was chosen as calibrator. Data are presented as mean ± SEM. N = 10 animals per group serum cytokines, n = 5-11 animals per group for colon cytokines, n = 6 animals per group for the PCR. <L/D, below the limit of detection; IFN, interferon; IL, interleukin; TNF, tumor necrosis factor.


88

4.4.5 Flow cytometric characterization of gastrointestinal tissues

4.4.5.1 T and B cell staining

Concerning overall CD3 expression in spleen and MLN, a marker for T cells, (tables 4.7-4.10,

figure 4.8), the overall mean percentage of CD3+ cells significantly dropped at day 7 following

CLP, when compared to sham-operated animals (or septic animals at day 2, for that matter). In

cells isolated from the lamina propria from ileum and colon, the overall percentage of T

lymphocytes remained the same. When looking at the distribution in the T cell population, we

found that the absolute number of T helper cells as well as cytotoxic T cells dropped

significantly in the CLPd7 group; however the relative contribution of cytotoxic T cells to the

overall T cell subset was higher at day 7 in the MLN whereas it was significantly decreased in

the spleen. We furthermore observed a drop in the absolute number of regulatory T cells,

however the relative percentage rose in MLN and spleen, albeit not significantly in the latter

one. No obvious differences were observed in the absolute or relative number of different T

cell subsets in ileal and colonic LPMCs.

Concerning the B cell population, we observed a significant drop in the percentages of B cells in

spleen and ileal LPMC at day 7 post-CLP, whereas their numbers increased significantly in the

MLN. No differences were observed at day 2.

A more graphical approach to these data can be found in figure 4.8.

Table 4.7 Flow cytometric staining of T and B cells in the spleen.


Absolute %CD3+ (T cell) 32.11 ± 2.79 34.50 ± 2.51 9.38 ± 1.45* <0.001 %CD3+CD4+ (T helper cell) 21.86 ± 2.21 20.66 ± 1.78 4.52 ± 0.83* <0.001 %CD3+CD8+ (cytotoxic T cell) 6.44 ± 0.64 7.16 ± 0.84 1.36 ± 0.24* <0.001 %CD4+ 24.52 ± 2.06 24.76 ± 1.99 5.98 ± 0.96* <0.001 %CD4+CD25+Foxp3+ (Treg) 1.66 ± 0.30 2.43 ± 0.34a 0.59 ± 1.30* <0.001 %CD19+ (B cell) 44.31 ± 1.61 48.99 ± 3.07 12.64 ± 1.53* <0.001

Relative %CD4+ out of CD3+ 67.53 ± 1.56 59.63 ± 1.98 48.11 ± 3.84* <0.001 %CD8+ out of CD3+ 20.20 ± 1.14 20.20 ± 1.40 14.44 ± 1.38* 0.004 %CD25+Foxp3+ out of CD4+ 6.76 ± 1.05 10.08 ± 1.24 9.06 ± 1.59 NS

Absolute percentages of the different leukocyte subsets identified in the overall leukocyte population. One-way ANOVA with post-hoc Dunnett or the Kruskal-Wallis with post-hoc Dunn’s test as appropriate with the sham-group as the control group. *p<0.05 in comparison to the sham-group; ap = 0.086; n = 10 to 12/group. Treg: regulatory T cell.


89

Table 4.8 Flow cytometric staining of T and B cells in the mesenteric lymph nodes.


Absolute %CD3+ (T cell) 60.04 ± 2.30 62.75 ± 1.87 40.63 ± 2.48* <0.001 %CD3+CD4+ (T helper cell) 45.46 ± 2.33 44.39 ± 1.45 27.00 ± 1.55* <0.001 %CD3+CD8+ (cytotoxic T cell) 11.15 ± 0.74 13.17 ± 1.00 9.18 ± 0.91 0.013 %CD4+ 50.32 ± 1.94 50.37 ± 1.28 32.25 ± 1.76* <0.001 %CD4+CD25+Foxp3+ (Treg) 2.94 ± 0.47 4.22 ± 0.41 3.18 ± 0.44 NS %CD19+ (B cell) 28.67 ± 1.28 28.56 ± 0.98 36.26 ± 1.84* 0.040

Relative %CD4+ out of CD3+ 75.53 ± 1.50 70.93 ± 1.97 66.65 ± 1.64* 0.004 %CD8+ out of CD3+ 18.70 ± 1.15 20.91 ± 1.35 22.18 ± 1.36 NS %CD25+Foxp3+ out of CD4+ 5.73 ± 0.75 8.37 ± 0.82 9.78 ± 1.34* 0.034

Absolute percentages of the different leukocyte subsets identified in the overall leukocyte population. One-way ANOVA with post-hoc Dunnett or the Kruskal-Wallis with post-hoc Dunn’s test as appropriate with the sham-group as the control group. *p<0.05 in comparison to the sham-group; n = 10 to 12/group.

Table 4.9 Flow cytometric staining of T and B cells in the ileum (LPMC).


Absolute %CD3+ (T cell) 16.67 ± 1.06 16.50 ± 1.55 15.00 ± 1.08 NS %CD3+CD4+ (T helper cell) 6.45 ± 0.59 4.87 ± 0.59 5.09 ± 0.96 NS %CD3+CD8+ (cytotoxic T cell) 2.92 ± 0.46 2.67 ± 0.35 2.74 ± 0.43 NS %CD4+ 9.17 ± 1.26 8.20 ± 1.07 6.90 ± 1.27 NS %CD4+CD25+Foxp3+ (Treg) 0.20 ± 0.03 0.27 ± 0.02 0.22 ± 0.09 NS %CD19+ (B cell) 26.17 ± 2.39 24.07 ± 2.20 17.07 ± 2.23a 0.101

Relative %CD4+ out of CD3+ 39.07 ± 3.28 29.96 ± 3.25 34.53 ± 6.47 NS %CD8+ out of CD3+ 17.21 ± 2.07 17.24 ± 2.79 17.93 ± 2.16 NS %CD25+Foxp3+ out of CD4+ 2.26 ± 0.37 3.51 ± 0.38 2.89 ± 0.81 NS

Absolute percentages of the different leukocyte subsets identified in the overall leukocyte population. One-way ANOVA with post-hoc Dunnett or the Kruskal-Wallis with post-hoc Dunn’s test as appropriate with the sham-group as the control group. *p<0.05 in comparison to the sham-group; ap = 0.067; n = 6-7/group. Treg: regulatory T cell

Table 4.10 Flow cytometric staining of T and B cells in the colon (LPMC).


Absolute %CD3+ (T cell) 5.96 ± 2.68 4.50 ± 0.96 4.71 ± 1.08 NS %CD3+CD4+ (T helper cell) 1.13 ± 0.91 1.02 ± 0.26 1.32 ± 0.56 NS %CD3+CD8+ (cytotoxic T cell) 1.36 ± 0.82 0.65 ± 0.19 0.72 ± 0.22 NS %CD4+ 1.28 ± 0.23 2.54 ± 0.98 2.06 ± 0.63 NS %CD4+CD25+Foxp3+ (Treg) 0.15 ± 0.03 0.18 ± 0.04 0.19 ± 0.03 NS %CD19+ (B cell) 5.87 ± 1.01 3.28 ± 1.36 3.10 ± 2.00 NS

Relative %CD4+ out of CD3+ 25.43 ± 3.47 22.46 ± 3.45 23.03 ± 4.40 NS %CD8+ out of CD3+ 17.32 ± 4.81 12.38 ± 1.41 14.99 ± 3.37 NS %CD25+Foxp3+ out of CD4+ 11.30 ± 1.03 9.91 ± 1.29 11.21 ± 1.40 NS


Figure 4.8 Schematic representation of the percentage change in the number of different T cell subsets and B cells in spleen, mesenteric lymph nodes, and lamina propria of ileum and colon. The percentage change at day 2 or day 7 following CLP is compared to the value in sham-operated control animals.

-100,00%

0,00%

100,00%

Pe

rce

nta

ge c

han

ge

sham CLP day 2 CLP day 7

T cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

Pe

rce

nta

ge c

han

ge


T helper cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

Pe

rce

nta

ge c

han

ge


cytotoxic T cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

Pe

rce

nta

ge c

han

ge


regulatory T cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

Pe

rce

nta

ge c

han

ge


B cells

spleen

MLN

ileum LPMC

colon LPMC


91

4.4.5.2 Dendritic cell staining

The overall percentage of MHCII molecule expressing cells (predominantly antigen-presenting

cells) dramatically drops at day 7 following the CLP-procedure in the spleen, whereas its

numbers are increased in the mesenteric lymph nodes; no difference in MHCII expression was

observed at day 2 post-septic insult, or in colonic or ileal LPMCs. As for the overall expression

of the pan-DC marker CD11c, a decrease was observed in splenic tissue at day 7, in contrast to

a significant rise in the ileum. The number of cells expressing CD11b, a subunit from the

complement 3 receptor, was significantly increased during sepsis in MLN at day 2 as well as day

7, and a remarkable drop in CD103+ cells was observed in the spleen 7 days post-CLP. When

looking at the overall percentage of CD11b+ DCs (MHCII+CD11c+CD11b+) in the leukocyte gate,

their numbers remained unaffected at day 2 post-CLP in the studied tissue types. At day 7

however, a significant drop in CD11b+ DCs could be observed in the spleen, whereas their

numbers significantly increased in colon and ileum.

When looking at the overall percentage of CD103+ DCs (MHCII+CD11c+CD103+), no differences

could be observed during sepsis in none of the organs. The relative contribution of CD103+DCs

to the total number of DCs (%CD103+ out of MHCII+CD11c+) was upregulated in the spleen at

day 7, whereas they were significantly lowered in ileal and colonic LPMCs (figure 4.9, tables

4.11-4.14).

Table 4.11 Flow cytometric staining of dendritic cell markers in the spleen.


Absolute %CD11c+ 1.03 ± 0.09 0.83 ± 0.08 0.69 ± 0.12* 0.046 %CD11b+ 11.84 ± 2.13 18.92 ± 1.95 22.11 ± 6.02 0.078 %CD103+ 3.75 ± 0.32 4.39 ± 0.49 0.80 ± 0.12* <0.001 %MHCII+ 43.85 ± 2.61 48.58 ± 4.71 12.24 ± 1.33* <0.001 %CD11c+MHCII+ (DC) 1.48 ± 0.30 0.99 ± 0.13 0.48 ± 0.08* 0.012 % CD11c+MHCII+CD11b+ (proinflamm DC) 0.44 ± 0.11 0.40 ± 0.08 0.06 ± 0.01* 0.013 % CD11c+MHCII+CD103+ (anti-inflamm DC) 0.11 ± 0.02 0.09 ± 0.01 0.09 ± 0.02 NS

Relative %CD103+CD11b- out of MHCII+CD11c+ 9.19 ± 1.36 9.54 ± 0.87 17.03 ± 2.11* 0.001 %CD11b+CD103- out of MHCII+CD11c+ 25.85 ± 2.55 37.60 ± 3.02* 13.88 ± 13.88* <0.001

Absolute percentages of the different leukocyte subsets identified in the overall leukocyte population. One-way ANOVA with post-hoc Dunnett or the Kruskal-Wallis with post-hoc Dunn’s test as appropriate with the sham-group as the control group. *p<0.05 in comparison to the sham-group; n = 10 to 12/group. Anti-inflamm: anti-inflammatory; DC: dendritic cell; proinflamm: proinflammatory, antigen-presenting DC


92

Table 4.12 Flow cytometric staining of dendritic cell markers in the mesenteric lymph nodes.


Absolute %CD11c+ 1.42 ± 0.12 1.24 ± 0.23 1.48 ± 0.11 NS %CD11b+ 6.93 ± 0.77 11.43 ± 0.70* 16.06 ± 0.94* <0.001 %CD103+ 8.17 ± 0.69 8.41 ± 0.62 6.44 ± 1.15 NS %MHCII+ 31.18 ± 1.49 30.54 ± 2.62 38.35 ± 1.75* 0.040 %CD11c+MHCII+ (DC) 1.25 ± 0.11 1.01 ± 0.19 1.19 ± 0.17 NS % CD11c+MHCII+CD11b+ (proinflamm DC) 0.44 ± 0.05 0.42 ± 0.11 0.29 ± 0.05 NS % CD11c+MHCII+CD103+ (anti-inflamm DC) 0.22 ± 0.02 0.21 ± 0.05 0.23 ± 0.04 NS

Relative %CD103+ CD11b- out of MHCII+CD11c+ 19.03 ± 1.78 21.21 ± 1.93 18.48 ± 1.47 NS %CD11b+ CD103- out of MHCII+CD11c+ 35.05 ± 2.17 37.05 ± 3.44 24.01 ± 1.59* 0.005


Table 4.13 Flow cytometric staining of dendritic cell markers in the ileum (LPMC).


Absolute %CD11c+ 0.60 ± 0.09 0.99 ± 0.12 1.50 ± 0.52* 0.045 %CD11b+ 6.51 ± 0.53 9.40 ± 0.41* 12.03 ± 2.42* 0.004 %CD103+ 12.45 ± 1.30 8.35 ± 1.04 11.87 ± 2.94 NS %MHCII+ 48.80 ± 6.38 43.97 ± 4.39 46.39 ± 5.83 NS %CD11c+MHCII+ (DC) 0.63 ± 0.08 1.29 ± 0.19* 1.33 ± 0.16* 0.008 % CD11c+MHCII+CD11b+ (proinflamm DC) 0.34 ± 0.03 0.42 ± 0.05 0.81 ± 0.14* 0.007 % CD11c+MHCII+CD103+ (anti-inflamm DC) 0.14 ± 0.04 0.24 ± 0.05 0.13 ± 0.002 NS

Relative %CD103+CD11b- out of MHCII+CD11c+ 21.20 ± 3.17 20.10 ± 1.58 10.36 ± 1.29* 0.002 %CD11b+CD103- out of MHCII+CD11c+ 56.67 ± 5.46 36.23 ± 2.76 63.48 ± 8.94 0.047


Table 4.14 Flow cytometric staining of dendritic cell markers in the colon (LPMC).


Absolute %CD11c+ 0.27 ± 0.11 0.50 ± 0.08 0.37 ± 0.16 NS %CD11b+ 2.59 ± 0.24 6.00 ± 3.17 5.13 ± 1.05 NS %CD103+ 2.91 ± 0.65 2.30 ± 0.23 4.63 ± 2.72 NS %MHCII+ 8.47 ± 0.94 7.05 ± 3.70 13.32 ± 3.06 NS %CD11c+MHCII+ (DC) 0.61 ± 0.10 0.55 ± 0.03 1.43 ± 0.11* <0.001 % CD11c+MHCII+CD11b+ (proinflamm DC) 0.34 ± 0.06 0.21 ± 0.06 0.95 ± 0.17* 0.004 % CD11c+MHCII+CD103+ (anti-inflamm DC) 0.10 ± 0.03 0.11 ± 0.02 0.07 ± 0.01 NS

Relative %CD103+CD11b- out of MHCII+CD11c+ 16.30 ± 3.00 21.23 ± 5.53 5.28 ± 0.85* 0.002 %CD11b+CD103- out of MHCII+CD11c+ 58.79 ± 6.81 37.53 ± 9.15 64.64 ± 9.51 NS



93

Figure 4.9 Schematic representation of the percentage change in the number of dendritic cell subsets (overall MHCII+CD11c+ DCs in the upper panel, MHCII+CD11c+CD103+ DCs in the middle panel and MHCII+CD11c+CD11b+ DCs in the lower panel) in spleen, mesenteric lymph nodes, and lamina propria of ileum and colon. The percentage change at day 2 or day 7 following CLP is compared to the value in sham-operated control animals.

-100,00%

0,00%

100,00%

200,00%

Pe

rce

nta

ge c

han

ge


CD11b+ dendritic cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

Pe

rce

nta

ge c

han

ge


CD103+ dendritic cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

200,00%

Pe

rce

nta

ge c

han

ge


dendritic cells

spleen

MLN

ileum LPMC

colon LPMC


94

4.4.5.3 Other leukocyte populations

Septic animals displayed a significant increase in the number of F4/80+ macrophages in splenic

tissue at day 2, and in the MLN at day 2 as well as day 7 following the CLP-procedure. In the

ileum and colon however, the percentage of macrophages however dropped 7 days post-CLP.

A significant rise in CD117 (c-Kit) positive cells, presumably mast cells, was observed at day 7 in

spleen and MLN, whereas their numbers remained unaffected in ileum and colon. When

comparing the number of natural killer (NK) cells by quantifying CD335 expression, we

observed a significant drop in the percentages in the spleen at day 2, whereas at day 7 post-

CLP the numbers increased significantly in MLN, ileum and colon. Ly6C+ monocytes remained

unaltered in all tissues, with the exception of an increase in their numbers in the colon in

animals that suffered from a prolonged sepsis. Ly6G+ neutrophils increased significantly in

colon and numerically in the ileal LPMC at day 2, whereas numbers remained identical in

spleen and MLN. At day 7 however, neutrophil numbers also heightened in spleen and MLN

(figure 4.10, tables 4.15-4.18).

Table 4.15 Flow cytometric staining of other leukocyte populations in the spleen


Absolute %F4/80+ (Mφ) 10.48 ± 0.95 17.50 ± 1.96* 8.55 ± 1.12 <0.001 %CD117+ (MC) 1.26 ± 0.24 2.44 ± 0.23 4.83 ± 0.67* <0.001 %Ly6C+ (Mo) 22.16 ± 1.90 22.84 ± 1.56 22.64 ± 6.05 NS %CD335+ (NK) 4.78 ± 0.39 2.73 ± 0.20* 3.84 ± 1.26 0.048 %Ly6G+ (NPh) 9.75 ± 1.70 12.04 ± 1.17 19.25 ± 4.27* 0.025

Absolute percentages of the different leukocyte subsets identified in the overall leukocyte population. One-way ANOVA with post-hoc Dunnett or the Kruskal-Wallis with post-hoc Dunn’s test as appropriate with the sham-group as the control group. *p<0.05 in comparison to the sham-group; n = 10 to 12/group. MC: mast cell; Mφ: macrophage; Mo: monocytes; NK: natural killer cel; NPh: neutrophil.

Table 4.16 Flow cytometric staining of other leukocyte populations in mesenteric lymph nodes


Absolute %F4/80+ (Mφ) 2.40 ± 0.45 3.94 ± 0.54 a 4.73 ± 0.71* 0.018 %CD117+ (MC) 0.97 ± 0.08 1.53 ± 0.11 2.89 ± 0.52* <0.001 %Ly6C+ (Mo) 32.12 ± 2.32 30.95 ± 2.34 33.11 ± 1.97 NS %CD335+ (NK) 1.29 ± 0.10 1.61 ± 0.32 7.06 ± 2.90* 0.007 %Ly6G+ (NPh) 8.82 ± 0.66 11.58 ± 1.28 19.65 ± 3.80* 0.001

Absolute percentages of the different leukocyte subsets identified in the overall leukocyte population. One-way ANOVA with post-hoc Dunnett or the Kruskal-Wallis with post-hoc Dunn’s test as appropriate with the sham-group as the control group. *p<0.05 in comparison to the sham-group; ap = 0.075; n = 10 to 12/group. MC: mast cell; Mφ: macrophage; Mo: monocytes; NK: natural killer cel; NPh: neutrophil.


95

Table 4.17 Flow cytometric staining of other leukocyte populations in the ileum (LPMC)


Absolute %F4/80+ (Mφ) 1.00 ± 0.15 0.95 ± 0.16 0.33 ± 0.09* 0.045 %CD117+ (MC) 0.95 ± 0.05 2.00 ± 0.35 2.59 ± 0.70 NS %Ly6C+ (Mo) 6.31 ± 0.66 7.33 ± 1.11 10.14 ± 2.99 NS %CD335+ (NK) 2.31 ± 0.34 2.48 ± 0.39 4.60 ± 1.31* 0.046 %Ly6G+ (NPh) 10.56 ± 1.08 13.20 ± 1.47 13.90 ± 2.30 NS


Table 4.18 Flow cytometric staining of other leukocyte populations in the colon (LPMC)


Absolute %F4/80+ (Mφ) 0.74 ± 0.10 0.63 ± 0.13 0.20 ± 0.09* 0.013 %CD117+ (MC) 1.05 ± 0.05 2.47 ± 0.36 1.73 ± 0.67 NS %Ly6C+ (Mo) 7.74 ± 0.75 5.90 ± 0.97 14.82 ± 3.62* 0.020 %CD335+ (NK) 0.93 ± 0.11 1.48 ± 0.31 2.47 ± 0.39* 0.004 %Ly6G+ (NPh) 4.29 ± 0.46 14.58 ± 1.95* 9.24 ± 1.42* 0.004


4.4.5.4 General summary of the flow cytometry data

A more concise table describing the absolute percentage increase or decrease in each cell

population can be found at the end of this chapter (table 4.21). In summary, septic animals

displayed no obvious alterations in the number of helper T cells in all tissues studied by day 2,

whereas their numbers dropped significantly in spleen and MLN by day 7 following CLP. The

overall percentage of CD4+CD25+Foxp3+ cells (regulatory T cells) was significantly reduced as

well at day 7 in the spleen. The number of B cells dropped significantly in the spleen and a

trend towards attenuated B cell numbers could be observed in the ileum; in contrast, in the

MLN we did see a significant rise in their numbers. The lymphopenia was observed in the blood

as well at both time points, confirming the presence of lymphocyte apoptosis which is a feature

hallmark of progressive immunosuppression often observed during sepsis. The numbers of

CD8+ cytotoxic T cells were only attenuated in the spleen by day 7, and the natural killer cell

subset remained unaffected in the spleen and even increased dramatically in MLN, colon and

ileum by day 7.

Figure 4.10 Schematic representation of the percentage change in different immune cell types in spleen, mesenteric lymph nodes, and lamina propria of ileum and colon. The percentage change at day 2 or day 7 following CLP is compared to the value in sham-operated control animals.

-100,00%

0,00%

100,00%

200,00%

Pe

rce

nta

ge c

han

ge


macrophages

spleen

MLN

ileum LPMC

colon LPMC

0,00%

100,00%

200,00%

300,00%

Pe

rce

nta

ge c

han

ge


mast cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

200,00%

300,00%

400,00%

500,00%

Pe

rce

nta

ge c

han

ge


natural killer cells

spleen

MLN

ileum LPMC

colon LPMC

-100,00%

0,00%

100,00%

Pe

rce

nta

ge c

han

ge


monocytes

spleen

MLN

ileum LPMC

colon LPMC

0,00%

100,00%

200,00%

300,00%

Pe

rce

nta

ge c

han

ge


neutrophils

spleen

MLN

ileum LPMC

colon LPMC


97

The percentage of overall DCs rose significantly in the ileum (day 2 and day 7) and colon (day 7),

whereas their numbers were significantly downsized in the spleen by day 7. Percentages of CD11b+

DCs were significantly higher in colon and ileum by day 7, whereas their percentages in the spleen

dropped. The percentage of Ly6C+ cells (monocytes, but possibly macrophages as well) in spleen

and MLN remained nearly identical to values obtained in control animals, whereas the percentages

rose significantly in the colon lamina propria. In the blood, a monocytosis was observed from day 2

onwards in septic animals. The opposite was observed concerning the overall percentage of

macrophages, as numbers increased significantly at day 2 in the spleen, whereas numbers in

ileums and colons from septic animals remained identical to those in control animals. At day 7,

macrophage numbers returned to baseline in the spleen but were highly upregulated in the MLN.

In colon and ileum, the percentage of F4/80+ macrophages dropped significantly by day 7 post-CLP.

Mast cell numbers were significantly higher by day 7 in spleen and MLN, and a trend towards

increased numbers was observed in ileum and colon at both time points. At day 2 post-CLP, we

quantified a steep rise in the numbers of neutrophils that had infiltrated in the colonic LPMC; by

day 7, this increase was also observed in the spleen and MLN. This rise coincided with the

occurrence of neutropenia in the blood of these animals at day 7, which could be suggestive of the

migration of neutrophils from the blood towards the inflammatory focus.

4.4.6 Myeloperoxidase-activity

Following CLP, at day 2 as well as at day 7, a numerical albeit not significant increase could be

observed in all GI tissues except for the jejunum (table 4.19). These results confirm the data

obtained by flow cytometry from ileum and colon to a certain extent, as an increased MPO-activity

is also indicative of an increased infiltration of neutrophils in the studied tissues. Of note, a

gastrointestinal ‘field-effect’ can be observed, albeit modestly, as distant organs such as stomach,

duodenum and distal colon appear to be infiltrated as well by neutrophils to a certain extent.

Table 4.19 Myeloperoxidase activity in gastrointestinal tract tissues following CLP.


stomach 0.08 ± 0.03 0.28 ± 0.11 0.34 ± 0.07 0.077 duodenum 1.38 ± 0.41 3.52 ± 1.21 1.24 ± 0.23 0.100 jejunum 2.02 ± 0.53 2.37 ± 0.71 1.75 ± 0.38 0.733 ileum 0.49 ± 0.12 3.33 ± 2.57 4.78 ± 1.73 0.257 proximal colon 0.37 ± 0.12 0.65 ± 0.32 1.81 ± 0.94 0.209 distal colon 0.31 ± 0.07 1.90 ± 0.84 1.50 ± 0.52 0.216

The myeloperoxidase-activity (U/mg tissue) in several organs of the gastrointestinal tract in septic and control animals. One-way ANOVA with post-hoc Dunnett test; n = 6/group.


98

4.4.7 Histology

Hematoxylin and eosin staining did not reveal any obvious differences between sham-operated

animals and animals that underwent CLP (all three groups 0.0 ± 0.0, p = NS)(figure 4.11).

Figure 4.11 Representative hematoxylin and eosin staining from sham (left), CLPd2 (middle) and CLPd7 animals (right).

Next, immunohistochemistry staining was performed in order to quantify apoptosis (TUNEL) and

the infiltration of neutrophils (Dab), macrophages (Mac-3) and T cells (CD3) in colonic tissue

samples (table 4.20; figures 4.12 through 4.15).

Remarkably, TUNEL-staining did not reveal a difference in the number of apoptotic cells between

the three groups. The number of CD3-staining T cells was significantly increased at day 2 following

CLP in comparison to the control animals, and was normalized back again at day 7. No difference

was observed between the number of macrophages or neutrophils between the three groups.

Note the presence of Dab-staining neutrophils in the blood vessels present in the submucosal of

sham animals, but not in those of septic animals at day 2 or day 7 (neutropenia).

Table 4.20 Results for the immunohistochemical stains of the mucosal layer (including the lamina propria).

% area sham CLP day 2 CLP day 7 p-value

TUNEL 2.13 ± 0.29 1.75 ± 0.38 2.08 ± 0.53 NS CD3 0.64 ± 0.02 1.31 ± 0.26* 0.86 ± 0.17 0.04 Mac-3 0.04 ± 0.01 0.23 ± 0.09 0.20 ± 0.09 NS Dab 0.01 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 NS Percentage of the total area that stained positive for TUNEL (denoting apoptosis), CD3 (lymphocytes), Mac-3 (macrophages) and Dab (neutrophils) in the mucosal layer. One-way ANOVA with post-hoc Dunnett test, n = 6/group, *p<0.05 compared to the sham group

100 µm 100 µm 100 µm


99

Figure 4.12 Representative TUNEL staining from sham (left), CLPd2 (middle) and CLPd7 animals (right).

Figure 4.13 Representative CD3 staining from sham (left), CLPd2 (middle) and CLPd7 animals (right).

Figure 4.14 Representative Mac-3 staining from sham (left), CLPd2 (middle) and CLPd7 animals (right).

Figure 4.15 Representative Dab staining from sham (left), CLPd2 (middle) and CLPd7 animals (right).

100 µm100 µm100 µm

100 µm100 µm100 µm

100 µm 100 µm100 µm

100 µm100 µm100 µm

Table 4.21 Percentage increase or decrease in the different leukocyte immune populations in comparison to the sham-group.

Spleen MLN Ileum LPMC Colon LPMC

sham CLPd2

sham CLPd7

sham CLPd2

sham CLPd7

sham CLPd2

sham CLPd7

sham CLPd2

sham CLPd7

T cells +7.44% -70.79%* +4.51% -32.33%* -1.02% -10.02% -24.50% -20.97%

T helper cells -5.49% -79.32%* -2.35% -40.61%* -24.50% -21.09% -9.73% -16.81%

Cytotoxic T cells +11.18% -78.88%* +18.12% -17.67% -8.56% -6.16% -52.21% -47.06%

Regulatory T cells +46.39%a -64.46%* +43.54% -8.16% +35.00% +10.00% +20.00% +26.67%

B cells +10.56% -71.47%* -0.38% +26.47%* -8.02% -34.77%b -44.12% -47.19%

Dendritic cells -33.11% -67.57%* -19.20% -4.80% +104.76%* +111.11%* -9.84% +134.43%*

CD103+ DCs -18.18% -18.18% -4.55% +4.55% +71.43% -7.14% +10.00% -30.00%

CD11b+ DCs -9.09% -86.36%* -4.55% -34.09% +23.53% +138.24%* -38.24% +179.41%*

Macrophages +66.98%* -18.42% +64.17%c +97.08%* -5.00% -67.00%* -14.86% -72.97%*

Mast cells +93.65% +283.33%* +57.73% +197.94%* +110.53% +172.63% +135.24% +64.76%

Monocytes +3.07% +2.17% -3.64% +3.08% +16.16% +60.70% -23.77% +91.47%*

Natural killer cells -42.89%* -19.67% +24.81% +447.29%* +7.36% +99.13%* +59.14% +165.59%*

Neutrophils +23.49% +97.44%* +31.29% +122.79%* +25.00% +31.63% +239.86%* +115.38%*

Percentage increase or decrease in the absolute number of leukocyte subset in the overall leukocyte gate (based upon FSC and SSC properties) in comparison to the sham-group. One-way ANOVA with post-hoc Dunnett, with the sham-group as control. *p < 0.05; ap = 0.086; bp = 0.067; cp = 0.075 for the post-hoc Dunnett. N = 10-12/group for spleen and MLN, n = 6-7/group for ileum and colon LPMC. CLPd2: cecal ligation and puncture sacrificed at day 2; CLPd7: cecal ligation and puncture sacrificed at day 7; DC: dendritic cell; LPMC: lamina propria mononuclear cells; MLN; mesenteric lymph nodes


101

4.5 Discussion

By performing the cecal ligation and puncture (CLP) procedure, we were able to reproducibly

elicit sepsis in our animals, as was marked by the occurrence of behavioral characteristics of

disease (e.g. impaired mobility and grooming behavior of the animals), weight loss, and the rise

in proinflammatory cytokine levels in serum as well as the colon. Furthermore, the presence of

ileus was confirmed, as the gastric emptying and geometric center of transit as measured with

the solid beads method was impaired in animals that underwent CLP, both two days as well as

seven days following the induction of sepsis. Inflammation at day 2 following the CLP was

typified by an increase of IL-6 and TNF-α in the serum, and a marked rise in the amount of both

cytokines that was secreted by the colon. At day 7, serum levels of the aforementioned

cytokines normalized to that in control animals, however this time a significant rise was found

in the concentration of IL-17A; this was also the case for IL17A mRNA levels in the colon. TNF-α

levels in the colon’s supernatants remained significantly augmented seven days following the

induction of sepsis. A trend towards increased IFN-γ and IL-6 levels was noted as well.

It should be noted that a substantial amount of data failed to reach statistical significance,

despite displaying major numerical differences between septic and control animals. This is to a

certain extent due to the large variation in our data, a finding that could be caused by the fact

that we implemented a rather mild version of the CLP-model. This results on the one hand in

reproducible signs of clinical disease and impaired transit parameters, but on the other hand in

fluctuating levels of inflammatory parameters such as cytokines and different leukocyte

subsets.

We then elaborately characterized several types of immune cells in the organs that constitute

part of the immune balance between the gut and the environment.

In summary, a profound depletion in the percentages of CD4+ as well as CD8+ T cells and B cells

was predominantly observed in the spleen, but in other tissues studied as well, coinciding with

an increase in the number of natural killer cells, albeit the latter not in the spleen. These

observations have been confirmed for the greater part in human splenic tissue samples by

others, where a profound depletion of CD4+ T cells and B cells coincided with a relatively stable

number of CD8+ T cells and NK cells (Hotchkiss et al., 2001; Souza-Fonseca-Guimaraes et al.,

2012); moreover, these results confirm the previous flow cytometric observations made by

Sharma et al. in T cell subsets isolated from the spleens of mice 20h post-CLP (Sharma et al.,


102

2015). Due to technical limitations, we did not specifically look at natural killer T (NKT) cells. Of

this population, it is known that in a human population of septic patients, the number of NKT

cells remains stable in the blood, while the number of CD3+ T cells declines, and this

manifestation typified an evolving morbid state (Heffernan et al., 2014). Of note, the CD335

antigen is not expressed on these NKT cells. CD335, also termed NKp46 or natural cytotoxicity

receptor, is however also expressed on the cell surface of innate lymphoid cells type 3 located

in the intestinal lamina propria (Sonnenberg & Artis, 2015; Artis & Spits, 2015). We therefore

cannot exclude whether the number of NK cells we measured does not include a certain

percentage of type 3 innate lymphoid cells. The above findings in part endorse our hypothesis

that the CLP-model is an immunologically qualified model to study sepsis, with many

immunological features that resemble the human disease state of ‘sepsis’.

Furthermore, many research has been conducted on the role of the regulatory ‘immune

suppressing’ T cells during sepsis. Overall Treg percentages decreased tremendously in the

spleen at day 7 following CLP, however the relative amount of Tregs in the total CD4+ T helper

cell subset significantly increased in spleen and MLN which is in accordance with previously

reported data, both in rodents as well as in human subjects in the later stages of

immunoparalysis (Monneret et al., 2003; Scumpia et al., 2006; Venet et al., 2009; Jiang et al.,

2012; Leng et al., 2012).

The definition of what exactly constitutes ‘the’ dendritic cell, or ‘the’ regulatory T cell, is an

everlasting matter of debate. Renowned experts in the field are currently updating their

knowledge on an almost near-daily basis (Schraml & Reis e Souza, 2015; Bekiaris et al., 2014).

Monocytes, macrophages and dendritic cells, three categories of cells pertaining to the

phagocytic system, are classified based upon their functional and phenotypical characteristics

(Guilliams et al., 2014; Bekiaris et al., 2014). In this Chapter, we determined the MHCII+CD11c+

to be dendritic cells based upon literature, expertise and previously performed experiments in

the lab. CD11c however is also commonly expressed on the cell surface of monocytes,

macrophages and B cells, of which the latter two populations also express the major

histocompatibility complex type II, and MHCII is furthermore present on the cell surface of

group 3 innate lymphoid cells (Artis & Spits, 2015). As such, we cannot state with certainty

whether the DC we identified is actually not a macrophage. Ideally, inclusion of the cell surface

marker CD64 or IgG1 receptor is warranted, as conventional DCs do not express this marker

(Bekiaris et al, 2014; Schraml & Reis e Souza 2015). In this regard we were hampered by the


103

number of fluorochromes that can be used with the Accuri C6 flow cytometer, as it was limited

to four fluorochromes. There is currently still need for a ‘pan-dendritic cell’ expression marker,

such as the Flt3ligand as has been proposed by the group of Reis e Souza and other experts in

the field (Persson et al., 2013; Schraml & Reis e Souza, 2015; Helft et al., 2015); one however

might preferentially refer to a subset of cells based upon their expression pattern of membrane

(and intracellular) markers (e.g. MHCII+CD11c+ cell), which tells us the probable function and

plausible properties of the cell subset, instead of simply insisting on naming the population

(e.g. in this case ‘dendritic cell’).

Gut DCs, more specifically lamina propria DCs, are then typically subdivided into three major

categories based upon their expression of the cell surface markers: CD103+CD11b+ DCs, CD103-

CD11b+ DCs and CD103+CD11b-DCs. Besides its potency to induce Tregs, it is accepted that the

subset of CD103-expressing DCs is particularly capable of, upon encountering an antigen,

migration towards the draining lymph nodes MLN and induction of the priming of naïve T cells

towards CCR9 and α4β7 integrin expressing T cells, that will subsequently differentiate into

gut-homing T cells (Maldonado et al., 2010; Mann et al., 2013; Bekiaris et al., 2014).

The proportion of CD11b+CD103- DCs and CD11b-CD103+ DCs in our control animals are for the

greater part in accordance with those reported by Bekiaris et al. (2014), although we measured

a slightly larger subset of CD11b+CD103- DCs in the lamina propria of ileum as well as colon

tissues, with the final subset considered to be non-migratory and derived from monocytes

(Varol et al., 2009; Bogunovic et al., 2009; Kinnebrew et al., 2012); the number of CD103+ DCs

in the spleens of sham-animals was limited as well (Johansson-Lindbom et al., 2005). Our flow

cytometry results for the greater part confirm the observations made by the authors above in

the sham-operated controls. In septic animals, we observed a significant rise of the relative

contribution of CD103 expressing DCs to the overall number of DCs in the spleen, whereas their

percentages dropped significantly in colon and ileum; this could be indicative of the migration

of CD103+ DCs from the gut. No differences however were observed in the MLN. A significant

rise in the number of ‘non-migratory’ CD11b+CD103- DCs was observed in the gut by day 7

coinciding with a decrease in spleen and MLN, pointing towards a more inflammatory substrate

as the CD11b (or integrin alpha M) mediates leukocyte adhesion and cell-mediated cytotoxicity,

a feature that was confirmed in colon and ileum by a significant rise in the percentage of NK

cells. Moreover, as CD11b+CD103- DCs are especially capable of inducing IFN-γ secretion from T

cells, these results are in accordance with the increased concentrations of IFN-γ we measured


104

in the colonic supernatants at day 7 post-CLP (Bekiaris et al., 2014). Finally, the same subset of

CD11b+ lamina propria DCs are specialized in driving Th17 responses, a feature we confirmed

by detecting increased levels of IL-17 protein and mRNA in respectively colonic supernatants

and tissue (Schlitzer et al., 2013). Our results on the overall number of DCs in ileum and colon

confirm those made by Koscielny et al. following intestinal manipulation as well as the

administration of LPS (2006 & 2011), which demonstrated an increase in the number of DCs in

MLN and ileum, albeit for the latter tissue they only looked at the muscularis layer.

As demonstrated in this chapter, flow cytometry can be an extremely powerful tool in the

study and characterization of immune cells in the blood as well as different tissue and organs.

The use of appropriately matched FMO-controls can be difficult, as approximately no more

than 2 to 5 million immunocytes can be isolated from the colonic lamina propria prelevated

from one 10-week-old C57Bl/6 mouse (Sheridan & Lefrançois, 2012), limiting the number of

staining panels that can be applied in one single rodent as current guidelines recommend to

use sufficient cells per staining in order to correctly identify the so-called ‘rare events’, a cell

type with a frequency equal to or lower than 0.01% in a given sample (Allan & Keeney, 2010;

Hedley & Keeney, 2013; Donneneberg & Donnenberg, 2007), e.g. the regulatory T cell subset

orthe different dendritic cell subtypes (Maecker & McCoy, 2010; Hulspas et al., 2009; Lee et

al., 2008). As such, flow cytometry can be used in addition to the more broadly implemented

techniques of histology and the myeloperoxidase-activity assay.

In accordance with previous reports studying the occurrence of ileus in LPS- or CLP-induced

ileus, we did observe a significant influx of neutrophils into the colon and ileum which was

confirmed by means of an MPO-assay, albeit the latter one being only numerically, and taking

the entire GI tract wall into account including the myenteric layer, as whole snap frozen colonic

samples were processed for the MPO-assay (Buchholz et al., 2012). In our setup neutrophil

invasion was confirmed in the lamina propria of the gut, whereas previous data predominantly

studied the muscularis externa (Vilz et al., 2012; Overhaus et al., 2004), thus indicating that all

tissue layers of the GI tract are affected during sepsis-induced ileus. Of note, only Dab-staining

of the colon did not demonstrate an increase in the percentage of Dab-positive staining area,

however both the MPO-assay as well as flow cytometry data confirm the hypothesis that at day


105

7 the number of neutrophils increases in all tissues coinciding with neutropenia in the blood,

which is indicative of neutrophil recruitment to the GI tract.

In summary, the above findings confirm that the CLP-model as performed in our lab by ligating

at 50% of the cecum’s length followed by one puncture with a 25G needle, displays a biphasic

course of disease. The initial ‘peak’ was characterized by an increase in ‘prototypical’

proinflammatory cytokines, IL-6 and TNF-α, in serum as well as colon, and an influx of

neutrophils into the bowel wall. The second peak was characterized by the concurrent increase

in CD11b+ DCs and increased mRNA levels of IL-17A and IFN-γ in the colon, making it plausible

that the release of these cytokines could be mediated directly or indirectly via the stimulation

of CD4+ T cells by this DC subset (Schlitzer et al., 2013). Additional intracellular staining is

required in order to confirm this hypothesis.

Some features of a true anti-inflammatory or CARS phase were confirmed at day 7, such as the

marked drop in T cells (lymphopenia) in spleen, MLN and the blood. Besides, serum levels of

the anti-inflammatory cytokine TGF-β were increased, and a numerical rise in serum IL-10 was

noted; the latter two findings however could also be observed at day 2, reflecting a more

simultaneous activation of the pro- as well as anti-inflammatory system during sepsis. Of note,

experts in the field are currently also acknowledging the fact that features of both extreme

phenotypes can be observed in the average patient, instead of perseveringly trying to label

them as being simply pertaining to one of both categories (Hotchkiss et al., 2013). Septic

animals at day 7 also displayed a significant drop in their rectal temperature, a sign that in

human patients reflects profound lymphopenia and is associated with higher mortality (Drewry

et al., Crit Care Med 2015). The authors thus postulated that hypothermia could be a suitable

early marker of sepsis-induced immunosuppression.

Animals displayed an impaired GI motility at both time-points, making the model qualified for

studying novel compounds that tackle the GI motility changes that occur during sepsis. We

confirm the GI tract to be an interesting target in the battle against sepsis. Caution however is

warranted, as different effects could be observed in this CLP-model on different immune

subsets and immune parameters in different organs, and at different time points; a myriad of

markers should therefore always rigorously be taken into account when performing additional

studies or interventions.


106

Due to time restraints, we opted to further characterize and modulate the CLP-model at day 2

post-procedure for the remainder of this dissertation. We thus lay our focus on the more

proinflammatory, initial alteration that are caused by the CLP-procedure.

Chapter 5

Study of the afferent nerve activity and central nervous system

activation in the cecal ligation and puncture model

Parts of this chapter were published as:

Nullens S, Deiteren A, Jiang W, Keating C, Ceuleers H, Francque S, Grundy D, De Man

JG, De Winter BY. In vitro recording of mesenteric afferent nerve activity in mouse

jejunal and colonic segments

Accepted for publication in Journal of Visualized Experiments, May 2016

Nullens S, Deleye S, Deiteren A, Jiang W, Keating C, Francque SF, De Man JG, Staelens

S, De Winter BY.

Brain-gut signaling in a mouse model of polymicrobial abdominal sepsis.

Submitted for publication (Annals of Surgery, July 2016)

Chapter 5 Afferent nerve activity and CNS imaging in the CLP model

108

5.1 Abstract

Background and objectives: Afferent nerves are able to transmit inflammatory signals to the

brain and it is known that direct exposure of afferent nerves to endotoxins results in increased

afferent activity. However, little is known on the effect of prolonged sepsis on afferent nerve

activity, and on the activation status of dorsal root ganglia (DRG) and of the central nervous

system (CNS). We aimed to study the jejunal afferent nerve activity by means of in vitro

afferent nerve recording, the activation status of the CNS in vivo by means of PET/CT imaging

focusing on regions involved in visceral pain sensation, and DRG activity in vitro by means of

RT-PCR following cecal ligation and puncture (CLP), an animal model of polymicrobial

abdominal sepsis.

Methods: Jejunal afferent nerve activity from sham and septic animals was studied in vitro in

an organ bath set-up during ramp distensions. Overall activity was assessed, as well as activity

of the different types of sub-units that could be identified in one mesenteric afferent nerve.

Next, DRGs were isolated from both groups and real-time RT-PCR was performed for several

opioid receptors, as well as markers indicative of neuronal activation and/or inflammation.

Finally, PET-CT imaging of the central nervous system was performed using the radioactive

tracer [18F]-FDG, and metabolic brain activity was assessed in previously defined brain regions

involved in the processing and modulation of visceral pain in control and septic mice.

Results: CLP-induced sepsis resulted in a significant decrease in overall jejunal afferent nerve

discharge at noxious distension pressures, primarily due to a decrease in discharge from wide

dynamic range and high threshold fibers. PCR of DRGs showed a significant increase in the

neuronal activation marker cFos in septic mice, as well as increased expression of TRPA1 and

NMDA receptors and decreased expression of the delta opioid receptor. A significant global

decrease in [18F]FDG uptake (p<0.0001 in all regions) was observed in CLP mice compared to

sham, with the most significant clusters being the amygdala (left more pronounced than right),

left and right cortex, striatum and olfactory region.

Conclusion: Prolonged sepsis induced in vivo by CLP desensitizes mesenteric afferent nerves,

which is in contrast to the immediate hypersensitizing effect of pure LPS. Analysis of lumbar

DRGs however revealed increased neuronal activation markers at mRNA level as well as a

decrease in delta-opioid receptor mRNA in septic CLP-mice. CLP-induced sepsis significantly

reduced metabolic activity in all brain regions, most notably in cortex, amygdala and striatum,

all of which are involved in the processing and modulation of visceral pain.


109

5.2 Introduction

An abundant amount of research has been performed on the effect of LPS and

proinflammatory cytokines on afferent nerve activity. In short, most researchers confirm the

potentiating effect of LPS on afferent nerve activity, with the i.v. administration of LPS resulting

in a significant increase in in vivo firing activity of mesenteric afferent nerves (Liu et al., 2005),

with an initial short and transient increase in firing rate demonstrable 5 min following the

injection, and a more sustained increase approximately reaching a maximum 1.5h post-

injection. In a subsequent experiment, the same researchers found that systemic LPS-

administration increased mesenteric nerve discharge 10 min post-injection, and the firing

activity continued to rise until the end of measurements (2 h post-injection)(Liu et al., 2007).

The activation of peripheral afferents subsequently results in increased cFos-expression in the

brain stem (Gakis et al., 2009). Remarkably, the effects of LPS are dependent upon the

bacterial strain that provides the LPS, as LPS from the commensal E. coli (serotype 026:B6) had

no effect on mesenteric afferent nerve activity whereas LPS from S. typhimurium (serotype

0111:B4) did increase afferent firing, and that the latter were dependent upon release of

prostanoids (Liu et al., 2009). Furthermore, detection of a bacteriological infection in the gut

via capsaicin-sensitive vagal afferent fibers might result in activation of the vagal anti-

inflammatory pathway (Riley et al., 2013). Besides, a lysate from whole bacteria was able to

directly activate colonic DRG neurons (Ochoa-Cortes et al., 2010).

Concerning the effect of proinflammatory cytokines, it has been demonstrated that several of

their receptors are expressed on the afferent vagal branch (Fernandez et al., 2012), and can

either directly or indirectly cause neuropathic pain and hyperalgesia (Sommer et al., 2006).

Vagal afferents and paraganglia play a pivotal role in the cytokine-to-brain communication, as

subdiafragmatic vagotomy attenuated the response to i.v. administered TNF-α and IL-1β

(Fleshner et al., 2008), and the same procedure blocked induction of c-fos in the brain that is

typically witnessed following i.p. administration of LPS (Wan et al., 1994), suggesting that

neural pathways are capable of transmitting inflammatory signals to the brain. Intraluminal

administration of LPS in the jejunum did also increase the cFos expression in the vagal nucleus

of the solitary tract (Gakis et al., 2009).

Next to LPS, several cytokines are known to activate afferent firing: Injection of interleukin-6

increased the activity of C-fibers (Brenn et al., 2007), and i.p. administration of IL-1β induced

cFos expression in primary afferent vagal neurons (Goehler et al., 1998); the same holds true


110

for TNF-α (Holzer et al., 2004). Incubation of DRGs with IL-1β or TNF-α increased DRG

mechanosensitivity in group II units and increased the peripheral receptive fields (Ozaktay et

al., 2006). Finally, mesenteric lymph collected from rats challenged with LPS could increase

mesenteric afferent discharge in another animal when administered i.v., whereas i.v.

administration of pure TNF-α could not, suggesting that other inflammatory mediators

secreted from the gut in the lymph contribute to this increased afferent nerve activity (Wang et

al., 2005).

The expression of several other receptors present on afferent nerves or dorsal root ganglia is

attenuated or upregulated during a state of inflammation or following administration of LPS to

the species. Our current knowledge concerning these receptors is predominantly obtained

from animal studies with regard to abdominal pain syndromes, such as the irritable bowel

syndrome (IBS), colonic inflammation and, once again, LPS-induced sepsis. Most notably, three

main classes of afferent receptors are notorious for these effects, namely the transient

receptor potential (TRP) channels, the opioid receptors and the cannabinoid receptors (De

Winter et al., 2016). This triumvirate was recently acknowledged to play an important role, as

interactions between them can be of an inhibitory or excitatory nature, and modulate different

processes such as pain and inflammation (Zador & Wollemann, 2015). The TRP channels are

presumed to orchestrate all peripheral afferent input, and they modulate visceral

hypersensitivity by other mediators such as histamine, serotonin and protease activating

receptors (De Winter et al., 2016). As such, TRPA1 has been shown to be involved in the

hypersensitivity caused by inflammation (Holzer et al., 2011), and antagonizing TRPA1 reduced

the inflammatory response in a mouse model of ventilator-associated pneumonia (Wang et al.,

2013). TRPV1 receptors expressed on afferent neurons play a substantial role in the

pathogenesis of gastrointestinal motility disturbances, as antagonists of TRPV1 were able to

counteract the motility disorders in both colitis- as well as endotoxin-induced ileus (De

Schepper et al., 2008; De Winter et al., 2009).

Secondly, opioid receptors have been demonstrated to be present not only on DRGs, but also

on peripheral sensory neurons (Stein, 2003). These receptors, synthesized in DRGs and then

carried into central and peripheral terminals via axonal transport, are primarily responsible for

the peripheral analgesic effects of locally administered opioid agonists (Zhou et al., 1998).

During inflammation up close to the afferent nerve, a downregulation of δ-receptor in DRGs


111

can be observed, while the level of μ-opioid receptor mRNA remained unaltered (Zhang et al.,

1998; Schäfer et al., 1995). The efficacy of opioids to decrease the excitability of afferents

increases during inflammation, resulting in enhanced efficacy of opioid agonists (Antonijevic et

al., 1995). Opioidergic inhibition of viscerosensory afferent nerves has been verified in a mouse

model of DSS-induced colonic inflammation (Valdez-Morales et al., 2013) and in a translational

study during which supernatants obtained from peripheral blood mononuclear cells of patients

suffering from irritable bowel syndrome inhibited colorectal afferent mechanosensitivity in

mice (Hughes et al., 2014). Modulating opioids and their receptors thus represents an

interesting neuroimmune target during inflammation and sepsis (Al-Hashimi et al., 2013; Basso

et al., 2014; Basso et al., 2015).

Finally, several research groups have contributed to the elucidation of the involvement of the

cannabinoid receptors during sepsis. Both cannabinoid-1 as well as cannabinoid-2 receptors

are involved in the physiological regulation of gastrointestinal motility, with activation of the

CB1 receptor resulting in impairment of overall gastrointestinal motility (Izzo & Sharkey, 2010).

Antagonists of both receptors however were able to prevent the motility impairment that

occurs during LPS-induced sepsis (Li et al., 2010). Selectively inhibiting the endocannabinoid

degradation enzyme fatty acid amide hydrolase (FAAH) reduced leukocyte adhesion and

improved capillary perfusion in the gut during endotoxemia (Kianian et al., 2013). At the

afferent nerve level, Donovan and Grundy showed that the non-selective cannabinoid agonist

WIN55,212-2 caused a significant increase in afferent activity, whereas the previously

mentioned FAAH did not. Next, the LPS-induced increased afferent nerve activity was

significantly downgraded by the cannabinoid agonist or by a compound that prevented the

uptake of cannabinoids, whereas FAAH had no effect. This led the authors to conclude that

endocannabinoids are important in modulating afferent signaling, and that the uptake rather

than breakdown terminates their action in the gastrointestinal tract (Donovan & Grundy,

2012). Finally, our group showed that septic ileus in mice was associated with an upregulation

of intestinal CB1 receptors and FAAH, and that administration of a cannabinoid further

decreased the LPS-induced motility disturbances (De Winter & De Man, 2010; de Filippis et al.,

2008).

Several other receptors or channels expressed on peripheral neurons or DRGs have been

shown to play a role in sepsis, afferent nerve hypersensitization and sepsis-induced motility


112

disturbances. As such, the NMDA-receptor at the DRG level is mandatory in the development

and maintenance of human visceral pain, whilst administering an NMDA-receptor antagonist

increased the pain threshold (Willert et al., 2004). In a mouse model of LPS-induced renal

insufficiency, blocking NMDA ameliorated this renal failure (Lin et al., 2015). CGRP-receptors

are involved in the occurrence of sepsis-induced motility disturbances, as administering a

CGRP-receptor antagonist reversed the endotoxin-induced GI motility impairment in a mouse

model of LPS-induced sepsis (De Winter et al., 2009) and prevented postoperative intestinal

inflammation and ileus (Glowka et al., 2015).

Finally, in the central nervous system, LPS-induced sepsis significantly impaired overall

neuronal activity in the rat brain as could be detected by a decrease in [18F]-FDG uptake,

presumably due to a combination of neuroinflammation and changes in cerebral blood flow

(Semmler et al., 2008). Our knowledge on metabolic activity of the central nervous system

during human sepsis and/or inflammation is limited. Currently, functional brain imaging by

means of functional MRI (fMRI) of positron emission tomography/computed tomography (PET-

CT) is predominantly performed to study the higher brain areas involved in nociception or

modulation of pain in several pain syndromes, including IBS. In an interesting study, Mayer et

al. (2005) demonstrated that patients suffering from ulcerative colitis displayed decreased

brain responses to visceral pain in comparison to IBS patients. This coincided with an inhibition

of the metabolic activity of the paralimbic circuits and prefrontal cortex, regions involved in

pain sensing and processing.

The aforementioned research specifically focuses on the initial, instantaneous effects of the

administration of pure LPS and the early phases of LPS-induced sepsis on afferent nerve activity

and activation of the central nervous system. Little however is known on the effect of a

polymicrobial abdominal sepsis, as can be obtained by performing the cecal ligation and

puncture procedure (CLP) on afferent nerve activity, and even less is known on this matter

during the later stages of sepsis.

As the interaction between the neuronal players and various immune cells could play a pivotal

role in the patient with an established sepsis, we aimed to study the effect of prolonged sepsis

caused by CLP on in vitro mesenteric afferent nerve activity, activation state of dorsal root

ganglia (DRG) and in vivo activation status of the central nervous system.


113


5.3.1 Animals

Swiss-OF1 mice, all male and 8 weeks old, were obtained from Charles River (France) and

housed in pairs for the splanchnic afferent nerve recording experiments (vide infra 5.3.3) or in

groups of 6 to 8 mice for the central nervous system imaging experiments (vide infra 5.3.4), in

standardized conditions (12h light-dark cycle) with unlimited access to regular chow and

tapwater. All experiments were approved by the Committee for Medical Ethics and the use of

Experimental Animals at the University of Antwerp (file number 2012-42 and 2012-42-

extension).

5.3.2 Induction of sepsis

To induce sepsis, mice underwent the cecal ligation and puncture procedure (CLP), as has been

described in chapter 3. In short, mice were and placed in the supine position. A midline

laparotomy was performed, the cecum was exteriorized and ligated at 50% of its length

followed by one through-and-through puncture with a 25G needle; the abdomen was then

closed in layers. After receiving fluid resuscitation and pain relief, mice were allowed to recover

in a heated cage with free access to tap water. Sham-operated mice served as controls.

Animals were prematurely sacrificed when they lost over 15% of their baseline body weight or

when they appeared to be moribund or had a clinical disease score (CDS) > 8 (Heylen et al.,

2013). All experiments were performed 48h following the procedure.

5.3.3 Splanchnic afferent nerve recordings

5.3.3.1 Tissue preparation of jejunal afferent nerves

The activity of afferent gastrointestinal nerves was measured in vitro by means of a technique

described by D. Grundy and colleagues to directly measure splanchnic afferent nerve activity in

murine subjects directly adjacent to the intestinal segment of interest (Jiang et al., 2011;

Keating et al., 2011). In order to implement this technique in our lab, a colleague and I spent a

two-week research stay at the facility of Prof. D. Grundy (January 2013, Laboratory of

Biomedicine, University of Sheffield, Sheffield, United Kingdom).

For the jejunal afferent nerve set-up, mice were sacrificed by means of cervical dislocation and

several jejunal loops (segments measuring approximately 3 cm) were rapidly excised with an

intact mesentery and placed in ice-cold Krebs solution (120 mM NaCl, 6.22 mM KCl, 1.57 mM


114

NaH2PO4, 15.43 NaHCO3, 1,21 mM MgSO4, 11 mM D-glucose and 1.55 mM CaCl2). The first

jejunal segment (first segment distally of the ligament of Treitz) was placed in a purpose-built

recording chamber that was constantly perfused with oxygenated Krebs solution at a rate of 10

mL/min, after which the jejunal mesenteric nerves of each jejunal segment were dissected

free. The nerve was transected, and the myelin layer was carefully peeled off. Both oral and

aboral ends of the jejunal segment were cannulated for intraluminal perfusion (10 mL/h) with

the Krebs solution and connected to a pressure transducer for recording of the intraluminal

pressure. The tip of a suction electrode was positioned next to the transsected afferent nerve

strand, allowing suction of the nerve into a capillary electrode and recording of afferent nerve

activity (figure 5.1) (Donovan & Grundy, 2012; Rong et al., 2004; Jiang et al., J Physiol 2011).

Following the isolation of the nerve into the suction electrode, a 15 min break was introduced

in the experimental protocol in order to obtain a steady state spontaneous afferent nerve

activity. Next, the intestinal preparation was gradually distended by occlusion of the aboral

cannula, leading to gradual rise in pressure in the intestinal segment (up to 60 mmHg). Afferent

nerve activity was measured at baseline and during consecutive ramp distensions. The

experimental protocol was only performed when three consecutive ramp distensions (15 min

interval) yielded a reproducible multi-unit discharge (Deiteren et al., 2015).

Figure 5.1 Scheme depicting where jejunal afferent nerve activity was measured.


115

5.3.3.2 Data acquisition and analysis

Nerve activity was recorded via a suction electrode connected to a Neurolog Headstage

(NL100AK, Digitimer), amplified (gain 10k, NL104, Digitimer) and filtered (NL125, Digitimer,

bandpass 500-5000 Hz). The signal was digitized by means of a CED1401 analog-digital

converter (Cambridge Electronic Design) and sampled at 20 kHz for analysis using the Spike2

5.21 software. Multi-unit recordings of overall nerve activity contain action potentials of

different shape, amplitude and width, corresponding to different nerve units in each afferent

fiber (figure 5.2 and figure 5.3). These individual wavemarks were matched to predefined

templates using the computerized Spike2 software, allowing us to discriminate between the

single-units. Before allocation of a spike to a certain wavemark, a signal-to-noise ratio of >2:1

was enforced. The action potential responses were calculated by subtracting the spontaneous

activity at baseline (intraluminal pressure of 0 mm Hg) from the response during distension at

fixed time points (5 mm Hg of increment from 0 to 20 mmHg intraluminal pressure, 10 mm Hg

increment from 20 to 60 mmHg).

The single-unit analysis allowed us to identify different fibers based upon their discharge

profile during ramp distension. Low threshold fibers typically exert increased discharge at the

lower distension pressures, high threshold fibers (HT) at the highest distension pressures,

whereas wide dynamic range fibers (WDR) demonstrate increased firing during the entire ramp

distension. Mechanoinsensitive afferent (MIA) nerve fibers typically do not respond to ramp

distensions (Blackshaw et al., 2002; Booth et al., 2008; Deiteren et al., 2015). The nerve firing

response at 20 mmHg is expressed as the percentage of the maximum firing response during

distension (LT%). A value of <15% signifies a HT fiber, a value >55% a LT fiber. Values for WDR

units vary between 15 and 55%, typically ranging around 40% (Booth et al., 2008). A MIA

displays spontaneous afferent firing that is unaffected by distensions.


116

Figure 5.2 A typical recording of afferent nerve activity at baseline and in response to two ramp distensions (60 mm Hg) (upper panel), and the subsequent identification (wavemark analysis) of different subunits in the nerve signal (lower panel). The upper trace in each panel shows the nerve rate (Hertz) in response to the intraluminal pressure, which can be found in the lower trace of the upper panel. The middle trace represents the neurogram (raw signal, impulses per second). In the lower trace of the lower panel, the different action potentials based upon shape, width and amplitude are denoted using different colors (wavemark analysis).


117

Figure 5.3 Schematic representation of different units during measurement of afferent nerve activity (LT, HT, WDR and MIA). LT%: (afferent firing at 20 mmHg / maximal afferent firing)

5.3.3.3 Experimental design

Splanchnic afferent nerve recordings were performed on one pair of animals per day (housed

together in one cage). In control as well as septic animals, nerve activity was recorded in

response to ramp distension. Next, we tested the effect of incubating segments obtained from

control animals with pooled supernatants of homogenized colonic tissue from septic animals

(samples randomly collected in the past) on afferent nerve activity, following an incubation

time of exactly 5 min. In order to exclude alterations of the afferent nerve firing caused by cell

lysis and release of intracellular mediators during the homogenization process, we also

incubated jejunal segments of a limited number of control animals with pooled supernatants of

homogenized colons obtained from other control animals.

20 60 mmHg

Aff

eren

t fir

ing

Low threshold(LT% > 55)

20 60 mmHg

Aff

ere

nt f

irin

g

High threshold(LT% < 15)

20 60 mmHg

Aff

eren

t fir

ing

Wide dynamic range(15 < LT% < 55%)

20 60 mmHgA

ffe

ren

t fir

ing

Mechano-insensitive

LT% = (afferent firing at 20 mmHg )/ (max afferent firing)


118

5.3.4 Central nervous system imaging

The processing of visceral information and visceral pain can be modulated at different levels,

such as the distal afferent nerve fiber, the dorsal root ganglion, the spinal cord and at the

central nervous system level (see also Chapter 1).

Central nervous system activity was quantified using a radioactive tracer, fluorodeoxyglucose

([18F]-FDG). This study was performed at the Molecular Imaging Center Antwerp at the

University of Antwerp (Prof. S. Staelens – dr. S. Deleye). The uptake of the tracer was calculated

in the different brain regions of septic and control animals as described underneath in more

detail.

5.3.4.1 Experimental protocol

Following an overnight fast, septic and control mice were injected i.v. in the tail vein with 18.5

MBq 2-deoxy-2-[18F]-fluoro-D-glucose ([18F]-FDG). Mice were then sedated with isoflurane

(5% for induction of anesthesia, 1.5 – 2.5% for maintenance of anesthesia) and were subjected

to a static PET scan for 20 min, followed by CT imaging for 10 min on a Siemens Inveon PET-CT

scanner (Siemens Preclinical Solution, Knoxville, TN). The scanner utilizes 1.59x1.59x10 mm LSO

(lutetium oxyorthosilicate) crystals grouped in 20x20 blocks. The animals were scanned in a

feet-first-prone position. The energy and timing window were set to 350-650 keV and 3.432 ns

respectively.

5.3.4.2 Data acquisition and analysis

The PET images were reconstructed using 2 iterations with 16 subsets of the 3D ordered subset

expectation maximization (OSEM 3D) algorithm followed by 18 maximum a posteriori (MAP)

iterations with 16 subsets, including scatter and attenuation correction. PET images were

reconstructed on a 128x128x159 grid with a pixel size of 0.776 mm and a slice thickness of

0.796 mm. CT imaging was performed using a 220° rotation with 120 rotation steps. Voltage

and amperage were set to 80 keV and 500 µA, respectively. Image processing of the static

[18F]-FDG PET data was done in PMOD v3.3 (PMOD Technologies, Switzerland). Each individual

PET image was transformed into the space of a predefined mouse brain template by matching

the individual CT images to the CT of the template and applying the same transformation to the

subject’s PET images. [18F]-FDG activity concentrations (AC, in kBq/cc) were subsequently

extracted for each delineated volume of interest (VOI) using a mouse brain VOI template


119

predefined on the aforementioned CT template. The standardised uptake value corrected for

glucose (SUVglc) was then calculated as SUVglc = AC/ID x BW x glc, where ID (kBq) is the

injected dose, BW (g) is the body weight of the animal and glc (mg/dl) is the whole blood

glucose level. Voxel-based analysis was performed in SPM8 (Statistical parametric mapping,

http://www.fil.ion.ucl.ac.uk) on the SUVglc images using a one way ANOVA (threshold of 100

voxels, FWE, p < 0.01 and p < 0.001) to detect differences between both groups. To quantify

this voxel-based analysis the number of significantly changed voxels relative to the total

number of voxels in a VOI (% sign.) and the maximal T value (maxT) were presented.

5.3.5 Dorsal root ganglia

Animals were sacrificed and the lumbar dorsal root ganglia (DRGs) were isolated. In short,

animals were sacrificed via cardiac exsanguination, and the entire lumbar spine was aseptically

excised. Following the removal of muscle tissue, the spine was split longitudinally using a

surgical scalpel, the meninges removed and the DRGs from level L1 through L6 were isolated

bilaterally. RT-PCR was performed on pooled DRGs for the activation marker cFos, CGRP,

NMDA, TRPA1, TRPV1 and the mu- (μ), delta- (δ) and kappa- (κ)opioid receptors. In short, total

RNA was isolated from pooled snap-frozen DRGs using the Qiagen RNeasy Mini Kit. Total RNA

was treated with DNase and converted to cDNA using the Transcriptor First Strand cDNA

Synthesis Kit (Roche Applied Science). Quantitative real-time PCR was performed using the

TaqMan® Universal PCR Master Mix (Life Technologies) and primers as mentioned in Table 5.1.

The PCR reaction was performed in a 25 µl volume reaction (Bustin et al., 2009), with the

following amplification parameters: 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of

95°C for 15 sec and 60°C for 1 min.

Table 5.1. Taqman primers used for the analysis of the dorsal root ganglia

Protein Entrez Gene id

cFos 14281 - Mm00487425_m1 N-methyl-D-aspartate (NMDA) 26012 - Mm00480341_m1 calcitonine-gene-related peptide (CGRP) 12310 - Mm00801463_g1 transient receptor potential cation channel V1 (TRPV1) 193034 - Mm01246300_m1 transient receptor potential cation channel A1 (TRPA1) 277328 - Mm01227437_m1 µ-opioid receptor (Oprm) 18390 - Mm01188089_m1 δ-opioid receptor (Oprd) 4985 - Mm01180757_m1 κ-opioid receptor (Oprk) 18387 - Mm01230885_m1


120

5.3.6 Statistics

Data are presented as mean ± SEM, with ‘n’ representing the number of mice or the number of

units identified in a group of mice. Data obtained from the afferent nerve recording (impulses

per second or imp/s) were analyzed using the Generalized Estimating Equations (GEE) with LSD

post-hoc testing. Data describing the contribution of the different types of afferent subunits

are compared using the Chi-square with post-hoc Fisher’s exact test. The unpaired Student’s t-

test was applied to compare the PCR-results from the DRGs. A p-value ≤ 0.05 was considered to

be statistically significant. Data were analyzed using SPSS version 20.0 (IBM, Chicago) and

visualized using GraphPad Prism version 5.00.


121

5.4 Results

5.4.1 Multi-unit jejunal afferent nerve activity

Induction of sepsis by means of the CLP-procedure significantly increased the CDS measured

prior to sacrifice (sham 0.00 ± 0.00 vs CLP 4.18 ± 0.32; p < 0.05, n = 10-11 mice per group).

Septic animals with a CDS < 4 as well as control animals with a CDS > 0 two days following the

procedure were excluded from the experiment.

Both jejunal segments of control and septic animals displayed irregular spontaneous afferent

nerve discharge at baseline (intraluminal pressure 0 mmHg). Afferent nerve activity at baseline

was not statistically different between septic and control animals (sham 12.22 ± 2.85 imp/s vs

CLP 8.93 ± 1.85 imp/s). Furthermore, we could not observe a difference in the overall number

of single-units that could be observed in each analyzed afferent nerve strand between both

groups (sham 4.64 ± 0.31 vs CLP 4.40 ± 0.40 single-units/nerve). As such, the spontaneous

afferent nerve activity corrected for the number of units in each nerve strand was numerically

lower in the CLP-group, however differences were not significant (sham 2.76 ± 0.84 vs CLP 2.11

± 0.47).

Next, following three consecutive and reproducible ramp distensions, a fourth ramp distension

was performed during which the intraluminal pressure was gradually increased up until 60

mmHg. Whole nerve afferent activity in response to ramp distension was significantly lower in

septic animals when comparing overall nerve activity from 30 mmHg onward; this difference

remained when calculating overall afferent nerve activity corrected for the number of units in

each nerve strand (figure 5.4).

Figure 5.4 Mesenteric afferent nerve discharge (imp/s) in control (sham) and septic (CLP) animals during ramp distension, for the whole nerve (A) and corrected for the number of units in each nerve (B). Generalized Estimated Equation (GEE) with post-hoc LSD; *p < 0.05; n = 10-11/group.

Whole nerve

0 10 20 30 40 50 60

0

10

20

30

40

50 Sham (n = 11)

CLP (n = 10)

* * * * *

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

Multi-unit recording / #units

0 10 20 30 40 50 60

0

5

10

15 Sham (n = 11)

CLP (n = 10)

* *

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

A B


122

5.4.2 Single unit jejunal afferent nerve activity

Based upon their waveforms, single-units were discriminated in the multi-unit recordings and

classified in one of the aforementioned categories (LT, HT, WDR and MIA). In 11 sham-animals

we were able to discriminate 51 different units, whereas in 10 CLP-animals we were able to

identify 44 different units using wavemark analysis. A trend towards a decreased number of HT

units was noted in septic animals in favor of the number of LT and WDR units (figure 5.5).

Differences however were not significant (p = 0.2).

Figure 5.5 Single unit distribution of afferent fibers in control (sham) and septic (CLP) animals. Chi-square test with post-hoc Fisher’s exact, p = 0.2. HT: high threshold fiber, LT: low threshold fiber, MIA: mechanoinsensitive fiber, WDR: wide dynamic range fiber

Next, we compared the nerve activity for each specific unit subtype between septic and control

animals. Spontaneous firing rate of single-units was unaltered in septic animals for all four

single-unit subtypes (table 5.2).

Table 5.2 The spontaneous firing rate of different types of sub-units in control (sham) and septic (CLP day 2) animals.

Single unit Sham CLP day 2

Low threshold fiber (imp/s) 1.82 ± 0.36 (n = 22) 2.01 ± 0.39 (n = 23) High threshold fiber (imp/s) 0.45 ± 0.22 (n = 12) 0.01 ± 0.01 (n = 2) Wide dynamic range fiber (imp/s) 1.18 ± 0.45 (n = 14) 1.09 ± 0.34 (n = 14) Mechanoinsensitive fiber (imp/s) 9.11 ±5.61 (n = 3) 3.56 ± 1.03 (n = 5)

In response to ramp distensions, LT from control animals displayed a rapid increase in firing

activity that saturated between 10 and 20 mmHg and remained stable until the end of the

distension (60 mmHg). Of note, in some units a small drop was observed in LT firing activity at

43%

23%

28%

6%

Single unit distribution sham

52%

5%

32%

11%

Single unit distribution CLP

LT

HT

WDR

MIA

52%

5%

32%

11%

Single unit distribution CLP


123

the highest pressures (figure 5.6A). LT from animals that underwent CLP displayed a near-

identical response to increasing pressures.

HT fibers from control animals on the other hand demonstrated an enhanced firing rate when

jejunal segments were being challenged with higher (noxious) distension pressures of 20

mmHg and higher. HT from septic animals displayed significantly lower firing rates during

distensions (figure 5.6C), however these results should be interpreted with caution as only 2 HT

fibers could be identified in 44 units from septic animals. Nevertheless, these results were

significant. Afferent nerve firing from WDR isolated in jejunal nerves from control animals

displayed a linear increase in response to rising distension pressures with saturation in activity

around 40 mmHg. This is in sharp contrast with WDR activity from septic animals, which only

rose marginally during ramp distensions and was significantly lower than values in control

animals during the entire distension (figure 5.6B). As for the MIA, no increase in response to

ramp distension could be observed in control as well as septic animals (figure 5.6D).

Figure 5.6 Pressure-response curves for the different types of subunits in sham and CLP-animals. HT = high threshold, LT = low threshold, MIA = mechanoinsensitive afferent, WDR = wide dynamic range. GEE with post-hoc LSD; *p < 0.05 significantly different from sham.

LT

0 10 20 30 40 50 60

0

5

10 Sham (n = 22)

CLP (n = 23)

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

WDR

0 10 20 30 40 50 60

0

10

20CLP (n = 14)

Sham (n = 14)

** **** ** ** ** ****** ****** *********

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

HT

0 10 20 30 40 50 60

0

5

10

Sham (n = 12)

CLP (n = 2)

* * * *

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

MIA

0 10 20 30 40 50 60

-20

-10

0

10

20 Sham (n = 3)

CLP (n = 5)

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

A B

C D


124

5.4.3 Incubation of control segments with septic colonic samples

Following the baseline sham distension, jejunal segments from control animals were incubated

for 5 min with supernatants from previously collected colon homogenates from animals that

underwent CLP (sacrificed 48h post-CLP). Next, one ramp distension was performed and data

were analyzed much alike the prior experiment. Whole nerve firing activity not significantly

decreased upon exposure of control jejunal segments to the ‘septic colonic homogenate’, and

similar results were obtained when correcting for the number of units identified in each

afferent nerve fiber (figure5.7).

Figure 5.7 Mesenteric afferent nerve discharge (imp/s) in control (sham) jejunal segments incubated with supernatants from previously pooled homogenized colons from animals that had undergone CLP, for the whole nerve (A) and corrected for the number of units in each nerve (B). GEE with post-hoc LSD; p = NS; n = 11/group.

Next, activity of the different subtypes of single-units was assessed in response to ramp

distensions. Once again, we observed a significant decrease in HT activity at the noxious

distension pressures (> 30 mmHg). In contrast to the previous experiment, WDR activity

remained statistically unaltered in control animals following exposure of the segment to the

‘septic soup’. On the other hand, we now observed a sharp drop in LT activity during distension

at the lowest physiological pressures (< 20 mmHg). Activity of MIA remained unaltered (figure

5.8). In order to exclude possible deleterious effects of intracellular mediators and molecules

released by the homogenization process, we incubated a small number of segments from

control animals (n = 5) with previously randomly collected supernatants from pooled colonic

homogenates. In short, no difference was observed on overall afferent nerve activity, nor when

corrected for the number of units in each fiber (figure 5.9).

Whole nerve

0 10 20 30 40 50 60

0

10

20

30

40

50

Sham (n = 11)

Sham + supernatans CLP-colon (n = 11)

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)


0 10 20 30 40 50 60

0

5

10

15 Sham (n = 11)


Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

A B


125

Figure 5.8 Pressure-response curves for the different types of subunits in sham and sham incubated with ‘septic supernatants’. HT = high threshold, LT = low threshold, MIA = mechanoinsensitive afferent, WDR = wide dynamic range. GEE with post-hoc LSD; *p < 0.05 significantly different from sham.

Figure 5.9 Overall mesenteric nerve activity of jejunal segments from sham-operated animals incubated

with ‘control’ pooled homogenized colonic tissue for the whole nerve (A) and corrected for the number of

units in each nerve fiber (B).

LT

0 10 20 30 40 50 60

0

5

10 Sham (n = 22)


*******

** **** *

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

WDR

0 10 20 30 40 50 60

0

10

20Sham + supernatans CLP-colon (n = 14)

Sham (n = 14)

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

HT

0 10 20 30 40 50 60

0

5

10

Sham (n = 12)


* *** * *

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

MIA

0 10 20 30 40 50 60

-20

-10

0

10

20 Sham (n = 3)


Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

A B

C D

Whole nerve

0 10 20 30 40 50 60

0

5

10

15

20

Sham (n = 5)

Sham + supernatans sham-colon (n = 5)

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)


0 10 20 30 40 50 60

0

2

4

Sham (n = 5)

Sham + supernatans sham-colon (n = 5)

Pressure (mmHg)

Aff

ere

nt

dis

ch

arg

e (

imp

/s)

A B


126

5.4.4 Central nervous system imaging

Forty-eight hours following CLP or sham, the total body weight of the animals did not differ

significantly between CLP and sham animals (sham 36.3 ± 0.46 g vs CLP 35.34 ± 0.42 g),

whereas the CDS was significantly higher in septic animals (CLP 4.1±0.6 vs sham 0.5±0.2,

p<0.001). Venous glucose levels were significantly lower in the septic animals (sham 107.9 ±

4.4 vs CLP 59.8 ± 4.8 g/dL, p < 0.0001). VOI based analysis showed a significant decrease in

[18F]FDG uptake corrected for endogenous glucose levels in septic mice compared to control

animals, in all regions that were analyzed (table 5.3) (figure 5.10). The highest significance was

detected in the cortex and the amygdala.

Table 5.3 Volume-of-interest (VOI) analysis of the different brain regions using PET-CT.

Voxel based SPM analysis revealed that the most significant clusters with decreased [18F]FDG

uptake (significance level p<0.001, FWE, 100 voxels) were the amygdala (left: 37.63% sign.,

maxT = 7.13 and right: 17.20% sign., maxT = 9.71), cortex (left: 21.30% sign., maxT = 7.56 and

right: 17.88% sign., maxT = 9.05), striatum (left: 2.93% sign., maxT = 6.29 and right: 18.53%,

maxT = 6.41) and olfactory region (44.5% sign., maxT = 10.41) respectively representing the

percentage of voxels that were significantly changed and their maximal T-value. The statistical

T-map overlaid on the MR template illustrates the regions with significant decreased [18F]FDG

uptake (table 5.4) (figure 5. 11- 5.13).

Brain region Sham CLP day 2 p-value Striatum 318.20 ± 20.41 189.52 ± 12.10 <0.001

Cortex (overall) 261.55 ± 14.96 150.55 ± 11.09 <0.001

Cortex (left) 263.41 ± 14.68 151.38 ± 11.24 <0.001

Cortex (right) 259.66 ± 15.26 149.71 ± 10.97 <0.001

Hippocampus 293.85 ± 18.82 173.52 ± 12.05 <0.001

Thalamus 318.11 ± 20.88 186. 81 ± 12.99 <0.001

Cerebellum 274.90 ± 15.23 170.97 ± 12.75 <0.001

Basal forebrain septum 262.68 ± 16.71 155.63 ± 10.48 <0.001

Hypothalamus 243.72 ± 16.39 153.41 ± 10.18 <0.001

Amygdala (overall) 221.31 ± 13.93 130.67 ± 7.66 <0.001

Amygdala (left) 226.19 ± 14.32 131.17 ± 7.62 <0.001

Amygdala (right) 215.59 ± 13.60 130.10 ± 7.90 <0.001

Brain stem 288.26 ± 18.02 183.55 ± 11.77 <0.001

Central gray 314.52 ± 20.95 192.00 ± 13.49 <0.001

Superior colliculi 307.10 ± 20.25 182.33 ± 12.94 <0.001

Olfactory 261.12 ± 13.68 148.54 ± 10.86 <0.001

Midbrain 312.80 ± 21.60 188.14 ± 12.47 <0.001

Inferior colliculi 282.31 ± 17.97 171.33 ± 12.62 <0.001


127

Figure 5.10 Volume-of-interest (VOI) analysis of the different brain regions using PET-CT. *p <0.0001 significantly different to the sham group for all regions measured. Multiple t-test, n = 11 animals/group. SUVglc = standardized uptake value corrected for glucose. STR = striatum, CTX = cortex, HIP = hippocampus, THA = thalamus, CB = cerebellum, BFS = basal forebrain septum, HYP = hypothalamus, AMY = amygdala, BS = basal septum, CG = central gray, SC = superior colliculus, OLF = olfactory regions, MID = midbrain, IC = inferior colliculus

Figure 5.11 Template used for central nervous system imaging displaying the regions of interest, at the same time displaying the regions of which SPM analysis revealed the most significant clusters with reduced tracer uptake in septic animals (dark yellow regions).

STRCTX

CTX

_R HIP

THA C

BBFS

HY

P

AM

Y BS

CG SC

OLF

MID IC

0

100

200

300

400

SUVglc

sham

CLP day 2*

* ** * * * *

*

* * **

* *


128

Figure 5.12 Statistical-parametric-mapping analysis of the different brain regions using PET/CT of the central nervous system. SPM-analysis (significance level p<0.001, FWE, 100 voxels) revealed that the regions with the most significant differences in metabolic activity were the cortex (bilateral), amygdala (left more widespread than right), striatum (right more widespread than left) and olfactory region. n = 11 animals/group. LSTR/RSTR = left/right striatum, CTX = cortex, LCTX/RCTX = left/right cortex, LHIP/RHIP = left/right hippocampus, THA = thalamus, CB = cerebellum, BFS = basal forebrain septum, HYP = hypothalamus, LAMY/RAMY = left/right amygdala, BS = basal septum, CG = central gray, SC = superior colliculus, OLF = olfactory regions, LMID/RMID = left/right midbrain, RIC/LIC = right/left inferior colliculus

LSTR

RSTR

CTX

LCTX

RCTX

LHIP

RH

IPTH

A CB

BFS

HY

P

LAM

Y

RA

MY B

SCG SC

OLF

LMID

RM

ID RIC LIC

0

20

40

60

80

100

control > sepsis

significant voxels

total voxels

%

R L

CORTEX THALAMUS

STRIATUM AMYGDALA HIPPOCAMPUS MIDBRAIN

HYPOTHALAMUS

0

10

T


129

Figure 5.13 (previous page, lower figure) - Coronal slices through the MR template of the mouse brain overlaid with the T map and the volumes of interest template. The T map (significance level p<0.001, FWE, 100 voxels) demonstrates the hypometabolism of the septic mice compared to the control mice in the cortex, striatum and amygdala. No significant differences were observed between the septic and control mice in the hippocampus, hypothalamus and the thalamus. The limited spatial resolution of the PET scanner (±1.4 mm at the center of the field of view (Goertzen et al., Med 2012) in combination with the small mouse brain regions result in partial volume effects. As a consequence significant voxels defined in a VOI can be a result of spillover of activity from a nearby region or vice versa.

Table 5.4 Statistical-parametric-mapping-of-interest analysis of the different brain regions using PET-CT

Brain region Number of significant voxels (%)

Striatum (left) 46/1568 (2.93%)

Striatum (right) 290/1565 (18.53%)

Cortex (overall) 3509/18666 (18.80%)

Cortex (left) 1971/9255 (21.30%)

Cortex (right) 1655/9255 (17.88%)

Hippocampus (left) 0/1559 (0%)

Hippocampus (right) 0/1617 (0%)

Thalamus 0/3560 (0%)

Cerebellum 16/7222 (0.22%)

Basal forebrain septum 190/1574 (12.07)

Hypothalamus 0/1473 (0%)

Amygdala (left) 286/760 (37.63%)

Amygdala (right) 107/622 (17.20%)

Brain stem 0/7525 (0%)

Central gray 0/518 (0%)

Superior colliculi 0/1031 (0%)

Olfactory 1262/2835 (44.51%)

Midbrain (left) 0/660 (0%)

Midbrain (right) 0/743 (0%)

Inferior colliculi (right) 0/344 (0%)

Inferior colliculi (left) 0/343 (0%)


130

5.4.5 Dorsal root ganglia

All animals that underwent central nervous system imaging had to be quarantined for a

minimum of 24h in a contained room in order to minimize radiation exposure to researchers

and personnel following the i.v. injection of the radioactive tracer [18F]-FDG. As such, to study

the activation status of the DRGs, we isolated DRGs from animals that were sacrificed for the

afferent nerve activity measurements. Forty-eight hours following CLP, we detected a

significant increase in the neuronal activation marker cFos, as well as in mRNA levels of TRPA1,

a selective ion channel involved in somatosensory modalities such as pain, cold and itch, and in

NMDA-receptor mRNA-levels, which are involved in glutamate signaling and inducing an

excitatory postsynaptic potential resulting in increased influx of calcium in the cell.

Furthermore, opioid receptor mRNA levels remained unaltered with the exception of the δ-

opioid receptor, which was significantly downregulated in animals that had undergone CLP.

TRPV1 and CGRP levels remained unaltered 48h following CLP, as did mRNA levels of the κ and

μ-opioid receptor.

Table 5.5 mRNA expression of neuronal activation markers, TRP channels and opioid receptors in DRG.

Sham CLP day 2

cFos 1.01 ± 0.06 3.09 ± 0.52* TRPV1 1.03 ± 0.13 1.17 ± 0.12 TRPA1 1.01 ± 0.07 1.92 ± 0.11* CGRP 1.01 ± 0.06 1.15 ± 0.10 NMDA 1.02 ± 0.10 1.34 ± 0.07* µ-opioid receptor 1.02 ± 0.08 1.04 ± 0.06 κ-opioid receptor 1.03 ± 0.11 1.05 ± 0.10 δ-opioid receptor 1.03 ± 0.11 0.61 ± 0.06*

Data are expressed as relative expression (2-ΔΔCT method) and the sham group was chosen as calibrator. Data are presented as mean ± SEM. N = 6 animals per group; *p < 0.05.


131

5.5 Discussion

The enteric and autonomous nervous system play a pivotal role in the development and

maintenance of inflammation, as has been countless times demonstrated in animal models of

various pathologies. As such, the wiring of peripheral sensory information via afferent nerves is

indispensable in the pathogenesis of postoperative ileus, endotoxin-induced ileus and post-

inflammatory irritable bowel syndrome, as targeting these afferents ameliorated the motility

disturbances and/or features of pain and visceral hypersensitivity. The central nervous system

in itself can suppress the perceived peripheral inflammation by means of a negative feedback

mechanism entitled the “cholinergic (vagal) anti-inflammatory reflex pathway” or “VAIP”,

thereby impeding inflammation, as has been elaborately described in Chapter 1.

In the cecal ligation and puncture model of sepsis-induced ileus we characterize for the first

time, to the best of our knowledge, the activation status of several relay stations that are

involved in the intricate nervous system that processes all information from the peripheral

gastrointestinal tract, ranging from the peripheral afferent nerve to the central nervous system

where all information is conveyed to. This characterization can help researchers to identify

potential therapeutic targets in the quest for tackling sepsis, as previous research has

demonstrated that afferent nerve activation is mandatory in order to evoke sepsis-induced

motility disturbances in a mouse model of endotoxin-induced ileus (De Winter et al., 2009). In

this study performed at our lab, it was furthermore shown that pretreatment with the

neurotransmission blocker hexamethonium or the neurotoxin capsaicin reversed the

endotoxin-induced delay in gastrointestinal transit. Neuronal research so far exclusively

focused on the effects of administration of pure LPS on activity in mesenteric nerves, as well as

on the DRG and the brain. In short, these unanimously demonstrated the direct

hypersensitizing effect of LPS on afferent nerve activity, with an increased expression of

markers of neuronal activation at the DRG level. In sharp contrast to these observations,

following a prolonged CLP-induced sepsis, we observed a significant drop in overall jejunal

afferent nerve activity when compared to sham-operated controls following a ramp distension.

This decrease in afferent firing was predominantly due to a decrease in the activity of wide

dynamic range as well as high threshold fibers. Interestingly, baseline activity was not altered

following CLP, nor did we observe a difference in the number of units that could be identified

in each jejunal afferent fiber. More remarkably, CLP caused high threshold fibers to ‘disappear,


132

as only 2 HT fibers could be identified in 10 septic animals, in contrast to 12 HT units in 11

control animals, and these fibers displayed significantly reduced activity in response to

increasing intraluminal pressures. When incubating jejunal segments from control animals with

supernatants from homogenized colons from septic animals, a similar hyporesponsiveness was

observed during distensions, however this time due to a drop in low (and high) threshold

activity. This immediate effect on activity is therefore due to mediators in the colonic wall from

septic animals, as the hyporesponsiveness was explicitly present within 5 min post-

administration in the organ bath, and incubation of the specimen with supernatants obtained

from healthy control-animals had no effect whatsoever. More research is mandatory in order

to identify the culprit mediator (or multiple mediators which is more likely) in this ‘septic soup’.

In previous studies, we confirmed the presence of significantly increased levels of several

proinflammatory cytokines in the colon at day 2 post-CLP, and the presence of receptors for

the proinflammatory IL-6, TNF-α and IL-1β, as well as for the anti-inflammatory IL-10, on

afferents as well as DRGs has been verified by others (Hughes et al., 2013). Elaborate research

by others however has demonstrated that direct incubation of gastrointestinal specimen with

these cytokines results in overt mechanical hypersensitivity (Hughes et al., 2013; Ozaktay et al.,

2006), in contrast to the hyporesponsiveness we consistently observed. The differential effect

of prolonged sepsis compared to the immediate incubation of healthy segments with

supernatants from septic animals is undoubtedly another interesting observation made during

this study. The low threshold fibers that we were able to identify in septic animals displayed

similar afferent firing activity in response to ramp distension when compared to sham-

operated animals, however this subunit was particularly inhibited in control segments that

were challenged with the ‘septic colonic soup’ for a short period (5 min). High threshold fiber

activity on the other hand was severely impaired in both experimental settings, whereas wide

dynamic range activity was only significantly affected in the animals challenged with CLP 48h

prior to the experiment. A numeric albeit not significant decrease in wide dynamic range

activity however could also be noted in sham jejunal segments incubated with septic

supernatans.

In this matter, many have proposed that opioid-release from immunocytes could play a pivotal

role in inhibition of viscerosensory afferents (Hughes et al., 2014). Opioid receptors have been

demonstrated to be present not only on DRGs, but also on peripheral sensory neurons (Stein,

2003). These receptors are primarily responsible for the peripheral analgesic effects of locally


133

administered opioid agonists (Zhou et al., 1998). The release of opioids from T-helper cells is

well known in the context of inflammatory bowel disease (Basso et al., 2014; Basso et al.,

2015), during which innate immune cells as well as T cells have the capability to produce

endogenous opioids, which decrease the excitability of peripheral colonic sensory neurons

(Owczarek et al., 2011; Valdez-Morales et al., 2013; Verma-Gandhu et al., 2007; Hughes et al.,

2013).

More recently however, research by the group of Schemann has demonstrated evidence for a

significantly reduced response of submucous colonic neurons following overnight incubation of

these neurons with an ‘irritable bowel syndrome cocktail’. This cocktail contained TNF-α,

histamine and tryptase among others, and was obtained from the supernatans of biopsies from

IBS patients (Ostertag et al., 2015). The group concluded that desensitization of the neurons to

the ‘inflammatory mediators’ is plausible following the overnight incubation of the submucous

colonic neurons. In short, we observed desensitization at the peripheral afferent jejunal nerve

in a mouse model of prolonged, CLP-induced polymicrobial abdominal sepsis.

At the DRG level, several features we observe following the prolonged CLP-induced abdominal

sepsis, are in accordance with prior research and literature. During inflammation up close to

the afferent nerve, a downregulation of δ-receptors in DRGs could be observed by others, while

the level of the μ-opioid receptor mRNA remained unaltered (Zhang et al., 1998; Schäfer et al.,

1995). These observations are in accordance with the mRNA levels we observed in our septic

animals. No effect was noted on the mRNA-level of the κ-opioid receptor. Several markers of

neuronal inflammation were upregulated in our study, including cFos, NMDA, as well as TRPA1,

a receptor channel involved in several somatosensory modalities. In contrast to the peripheral

afferent relay station, we observed signs of excitation at the DRG level. In the central nervous

system, we observed an overall decrease in metabolic brain activity as demonstrated by a

significant drop in the [18F]-FDG uptake in all regions that were analyzed, which is in

accordance with prior research by Semmler and colleagues (Semmler et al., 2008). SPM-

analysis revealed that the area’s most notably affected were the amygdala, striatum and

cortex, regions involved in the orchestration of somatomotor, visceral and cognitive responses

to (inflammatory or painful) stimuli, but no difference in the number of significant voxels was

found in the hippocampus (at the 0.001 significance level that is), an area that is particularly

connected with the amygdala and is involved in the processing of negative stimuli (Mayer et al.,


134

2009). The same holds true for the thalamus and hypothalamus in our study. The striatum,

involved in reward responses also displayed significantly less metabolic activity in septic mice.

To the best of our knowledge, we studied for the first time ever the brain responses during a

polymicrobial sepsis using in vivo imaging, however at the same time we were faced with

technical difficulties. Spatial and temporal resolution are of poorer quality when compared

with the functional magnetic resonance imaging (fMRI), a technique that is also commonly

applied for this type of studies (Mayer et al., 2009). It would be of particular interest to focus

on the insulae, anterior cingulated cortex (ACC) and periaquaductal gray (PAG), areas all

involved in either the experience of pain (ACC) or the response to pain and inflammation, as

the PAG is known to receive a major amount of input from the amygdala and ACC, and is a key

area in the mechanism of descending inhibition of afferent activity via the presence and release

of endogenous opioids (Gebhart et al., 2004; Mayer et al., 2005).

In summary, we can state that differential effects can be observed at the different relay

stations that constitute the nervous system during a prolonged polymicrobial sepsis. Our initial

efforts to completely characterize this intricate system should certainly be confirmed by others,

as it could allow researchers to identify pivotal neuroimmune mediators in the pathogenesis of

a relevant sepsis-model and define clinical targets in the struggle against sepsis. Whilst the

information previously obtained by others on afferent and DRG neuronal activity following

exposure to pure LPS is interesting, using a polymicrobial and more clinically relevant animal

model of sepsis could certainly also be of interest to the translational and clinical scientist. Both

models are complimentary and should be studied in conjunction in order to fully elucidate the

mechanisms and pathogenesis present in several clinical syndromes.

Chapter 6

Neuronal modulation of septic ileus:

Effect of GTS-21, an α7nAChR agonist, on CLP-induced ileus


Nullens S, Staessens M, Peleman C, Schrijvers D, Malhotra-Kumar S, Francque SM,

Matteoli G, Boeckxstaens GE, De Man JG, De Winter BY. Effect of GTS-21, an alpha7

nicotinic acetylcholine receptor agonist, on CLP-induced inflammatory,

gastrointestinal motility and colonic permeability changes in mice.

Shock, 2016 Apr; 45(4): 450-9. doi: 10.1097/SHK.0000000000000519

Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21

136

6.1 Abstract

Background and objectives: During abdominal sepsis, the inhibition of gastrointestinal (GI)

motility together with mucosal barrier dysfunction will lead to increased bacterial translocation

and maintenance of sepsis. The activation of the vagal anti-inflammatory pathway remains an

appealing therapeutic strategy in sepsis. In this respect, selective alpha7 nicotinic acetylcholine

receptor (α7nAChR) agonists have shown anti-inflammatory properties in several animal

models of inflammation.

Methods: Sepsis was induced in OF-1 mice by cecal ligation and puncture (CLP). GI transit and

colonic permeability were quantified, and cytokine levels were determined in serum and colon.

We studied the effect of GTS-21, an α7nAChR agonist on the aforementioned parameters.

Splenectomized animals as well as α7nAChR-knock-out animals (Chrna7-/-) were included to

study the role of splenic macrophages and the α7nAChR during polymicrobial abdominal

sepsis.

Results: CLP-induced ileus resulted in a significant delay in overall GI transit, coinciding with the

occurrence of increased colonic permeability and systemic as well as colonic inflammation. In

septic animals, GTS-21 significantly ameliorated GI motility, lowered systemic and colonic levels

of IL-6, decreased colonic permeability and decreased the number of positive cultures obtained

from blood and mesenteric lymph nodes. The preventive excision of the spleen prevented

animals from developing sepsis-induced ileus. Chrna7-/- mice displayed a more severe septic

phenotype, whereas remarkably GTS-21 was still beneficial in these KO-animals.

Conclusion: Our results show that peripheral targeting of the vagal anti-inflammatory pathway

proves beneficial in an animal model of polymicrobial abdominal sepsis. A major role is

allocated to splenic immune cells in the development of sepsis, as preventive splenectomy was

protective for the development of sepsis. Since we could not measure an additional beneficial

effect of GTS-21 on CLP-induced sepsis in splenectomized animals, its main effects are

presumably mediated via splenic macrophages and immune cells, as opposed to those residing

in the GI tract.

Data on the Chrna7-/- mice suggest that the beneficial effects mediated by GTS-21 on

inflammation and motility might be related to activation of other receptors besides the

α7nAChR.


137

6.2 Introduction

Sepsis remains a major cause of morbidity and mortality in modern day Intensive Care units. As

was already elaborately discussed in Chapter 1, we now know that the gastrointestinal (GI)

tract is involved in these failing events with the occurrence of ileus and the disturbance of the

intestinal barrier. Compounds that are currently being studied in the context of sepsis include

agonists of the alpha7 nicotinergic acetylcholine receptor (α7nAChR). Originally demonstrated

to be present in the central nervous system, this receptor has been demonstrated on the cell

surface of immune cells, most notably macrophages (Floto & Smith, 2003), where binding of its

endogenous ligand acetylcholine results in the inhibition of the release of proinflammatory

cytokines from macrophages due to a post-transcriptional inhibition of cytokine synthesis

(Wang et al., 2003). This mechanism represents the final step in the so-called ‘cholinergic’ or

‘vagal anti-inflammatory’ pathway, an endogenously present reflex pathway that suppresses

the release of proinflammatory cytokines during inflammation (Rosas-Ballina et al., 2011;

Tracey, 2002). The exact anatomical substrate of this anti-inflammatory pathway is currently

being extensively investigated, and our knowledge concerning the organs and cells involved is

still ever increasing (Martelli et al., 2014). The group of Tracey initially showed that this

pathway is mediated via acetylcholine-synthesizing T-cells that reside in the spleen (Tracey,

2007; Rosas-Ballina et al., 2011; Borovikova et al., 2000). However, Cailotto and colleagues

(2014) demonstrated that indirect modulation of macrophages in the GI tract by vagal

efferents is also able to inhibit inflammation via VIPergic and nitrergic enteric neurons without

splenic contribution (See also Chapter 1, figure 1.6 - The vagal anti-inflammatory pathway).

GTS-21, a selective partial α7nAChR agonist, is currently under investigation for its

neuroprotective effects in Alzheimer’s disease where it enhances the cognitive impairment,

and in schizophrenia, where it ameliorates the negative symptoms that are often undertreated

with ‘classic’ antipsychotic drugs (Olincy et al., 2006; Freedman & Olincy, 2012).

GTS-21 has been demonstrated to suppress inflammation in various animal models of

inflammation, such as acute kidney injury (Chatterjee et al., 2009), pancreatitis (van Westerloo

et al., 2006) and ventilator-induced injury (Kox et al., 2011). Animal studies have already

demonstrated the beneficial effects of vagal nerve stimulation (VNS) on survival following LPS-

induced murine sepsis (Huston et al., 2007), whereas mortality was increased following

vagotomy in the colon ascendens stent peritonitis animal model (Kessler et al., 2012). These

beneficial effects were demonstrated to be dependent on the presence of intact vagus-nerve


138

to splenic neuronal circuits (Huston et al., 2006; Ji et al., 2014). Moreover, vagal nerve

stimulation has been shown to protect gut injury induced by trauma followed by hemorrhagic

shock and thereby limit further organ damage (Levy et al., 2013); this study however raised

doubts as to whether the spleen was involved. As immunohistochemistry demonstrated the

presence of α7nAChR on the cell surface of macrophages in the GI wall, this cell population

represents an interesting therapeutic target (Kawahara et al., 2011).

Given that GTS-21 improves survival in different animal models of sepsis, such as the cecal

ligation and puncture (CLP) model, burn injury (Khan et al., 2012) and endotoxemia (Pavlov et

al., 2007), we aimed to investigate the effects of GTS-21 on motility, permeability disturbances

and inflammation in the GI tract in the CLP-model, as well as the role of the α7nAChR in these

effects by studying α7nAChR-KO animals. As the role of the spleen in septic ileus currently

remains to be fully elucidated, we investigated the effect of GTS-21 in animals that underwent

splenectomy prior to the induction of sepsis.


139


6.3.1 Animals

Swiss-OF1 mice, all male and 8 weeks old, were obtained from Charles River (France) and

housed in groups in standardized conditions (12h light-dark cycle) with unlimited access to

regular chow and tapwater. Chrna7-/- (mice lacking the α7nAChR) on a C57/Bl6 background and

their wild-type littermates were a kind gift from Prof. G. Boeckxstaens and Prof. G. Matteoli

(TARGID, University of Leuven). All experiments were approved by the Committee for Medical

Ethics and the use of Experimental Animals at the University of Antwerp (file number 2012-42

and 2012-42-extension).

6.3.2 Induction of sepsis – the CLP-model

To induce sepsis, mice underwent the cecal ligation and puncture procedure (CLP) as decribed

in chapter 3. In short, mice were anesthetized and placed in the supine position. A midline

laparotomy was performed, the cecum was exteriorized and ligated at 50% of its length

followed by one through-and-through puncture with a 25G needle, followed by suturing of the

abdominal wall in layers. Sham-operated mice served as controls. Animals were prematurely

sacrificed when they lost over 15% of their baseline body weight or when they appeared to be

moribund or had a clinical disease score (CDS) > 8 (Heylen et al., 2013).

6.3.3 Dose optimization of GTS-21

In a preliminary experiment, we tested the effect of increasing dosages of GTS-21 (2, 4, 8 and

12 mg/kg i.p.) on GI motility by performing the solid beads method. Since a dose of 8 mg/kg

yielded optimal salutary effects on GI motility, this dose was utilized for the remainder of this

study (fig. 1A). A dosage of 4 mg/kg is most often used in literature, as this dosage significantly

decreased influx of neutrophils into the peritoneum in an animal model of LPS-induced sepsis

(Giebelen et al., 2007), attenuated LPS-induced acute kidney injury (Chatterjee et al., 2009),

attenuated inflammation during cerulein-induced pancreatitis (van Westerloo et al., 2006) and

ameliorated survival in an endotoxin-model (Pavlov et al., 2007). A dosage of 8 mg/kg however

was in concordance with other animal models of sepsis or organ failure, as this dosage was

necessary to achieve optimal results in a mouse model of ventilator-induced pneumonia (Kox

et al., 2011); Khan and colleagues corroborated that two daily injections with 4 mg/kg of GTS-

21 were necessary in order to improve survival in a mouse model of burn injury (Khan et al.,


140

2012). Yeboah et al. even administered a dosage of 10 mg/kg in a rat model of renal

ischemia/reperfusion injury, thereby successfully decreasing tubular necrosis (Yeboah et al.,

2008).

Figure 6.1 Experimental design of the study. In a first experimental set up (upper panel), mice (OF-1, Chrna7-/- or Chrna7+/+ mice) received vehicle (saline) or 8 mg/kg of GTS-21 on 3 consecutive days points (1h prior to CLP and then 24h and 48h after CLP). In a second set up (lower panel), mice underwent splenectomy 7 days prior to CLP (B). Mice received vehicle (saline) or 8 mg/kg of GTS-21 on 3 consecutive day points (1h prior to CLP and then 24h and 48h after CLP). 48h post-CLP, mice received a gavage and were sacrificed 2h later to measure gastrointestinal transit and for prelevation of serum and tissue samples, measurement of colonic permeability and quantification of cell adhesion molecules at the mRNA and protein level. CBA: cytometric bead array; CDS: clinical disease score; CLP: cecal ligation and puncture; MLN: mesenteric lymph nodes; mRNA: messenger ribonucleic acid; RT-PCR: reverse transcriptase real-time polymerase chain reaction; temp.: temperature.

6.3.4 Experimental design

Following the dose-optimization study, five sets of experiments were designed. The

experimental design can be consulted in figure 1. In a first set of experiments, CLP- or sham-

operated mice were administered GTS-21 (8 mg/kg) or its vehicle (sterile saline 0.9%) i.p. 1h


141

before the surgical procedure, and 24h and 48h after CLP. The same protocol was applied for

all the following experiments. Sickness behavior was quantified based upon a CDS, and GI

motility was assessed. Following terminal anaesthesia, animals were sacrificed with cardiac

puncture while obtaining blood samples (Multivette® 600 capillary blood collection, Sarstedt).

Blood samples were centrifuged (5000 rpm, 5 min, 20°C) and supernatants were stored at -

80°C until further analysis. Inflammation was furthermore assessed locally at the colonic level.

In a second set of experiments, colonic permeability was assessed in septic and control animals

under terminal anaesthesia 48h following the procedure, and blood and mesenteric lymph

nodes were obtained for culture.

In a third set of experiments, septic ileus was induced by CLP in locally bred Chrna7-/- mice that

lacked the α7nAChR and their wild type littermates (Chrna7+/+ mice) on a C57Bl/6 background

after receiving GTS-21 or vehicle solution. The absence of the Chrna7 gene was confirmed by

RT-PCR on 5 mm tail clips before the experiment using the MyTaq™ Extract-PCR Kit and

following primers: mChrna7 forward CCTGGTCCTGCTGTGTTAAACTGCTTC, mChrna7 reverse

wild-type CTGCTGGGAAATCCTAGGCACACTTGAG, mChrna7 reverse mutated

GACAAGACCGGCTTCCATCC, with following amplification parameters: 3 min at 94°C, followed

by 35 cycles of 70°C for 1 min and 72°C for 1 min.

In a fourth set of experiments we studied the role of the spleen in GTS-21-mediated effects by

including mice that underwent a splenectomy (SPLX) 7 days before CLP (fig. 1B). Briefly, mice

were anesthetized, placed on their right flank and received a midaxillary incision. The spleen

was mobilized and the splenic hilum was ligated for hemostasis, after which the spleen was

dissected free from the surrounding tissues. After controlling for bleeding, the abdomen was

closed in layers and mice were allowed to recover with fluid resuscitation and pain relief.

In a fifth experiment, the in vitro effect of GTS-21 was studied on cytokine-release from bone-

marrow derived macrophages and dendritic cells (DCs).

6.3.5 In vivo measurement of gastrointestinal transit

Following overnight deprivation of food, mice were gavaged with 0.5 mL of tapwater

containing 25 glass green-colored beads through a 20G flexible canula 1h following the final

GTS-21 or vehicle injection. Mice were sacrificed 2h later and the GI tract was resected and

divided into 10 parts. The number of beads in every segment was counted under a


142

stereomicroscope for calculation of percentage gastric emptying (%GE) and the geometric

center of intestinal transit (GC) as a marker for overall GI transit (Seerden et al., 2004).


The concentration of IL-6, TNF-α, IL-2, IL-10, IL-17 and IFN-γ in serum (pg/mL) was determined

using the BD CBA Mouse Cytokine Kit on an Accuri® flow cytometer (BD Biosciences) according

to the manufacturer’s instructions. Data were processed using FCAP Array (BD).

Colonic protein levels of cytokines were measured in supernatants from homogenized whole

colons, as previously described in Chapter 3 (3.5.2 – Colonic cytokines – protein level).

The colonic cytokine content at the mRNA level was assessed by means of RT-PCR, as

previously described in Chapter 3 (3.5.3 – Colonic cytokines – mRNA levels).

6.3.7 Colonic cell adhesion proteins

In concordance with the aforementioned protocol, real-time RT-PCR was performed for several

cell adhesion molecules.

6.3.8 Colonic permeability and bacterial translocation

Colonic permeability was quantified by means of the Evans blue (EB) method (Lange et al.,

1994), as previously described in Chapter 3 (3.6.1 – The Evans blue method). The extracted

amount of EB from the colon was determined spectrophotometrically and expressed as µg

EB/100 mg colonic tissue.

To estimate bacterial translocation into the bloodstream, one drop of EDTA-treated full blood

was plated onto a blood agar culture plate following enrichment and incubated at 37°C for 24h

in ambient air supplied with 5% CO2. Additionally, mesenteric lymph nodes (MLN) were

resected aseptically and suspended in RPMI 1640 medium. MLN were then mashed manually

using a 10 ml syringe plunger through a 40 µm nylon cell strainer. Homogenized MLN were

plated onto a blood agar culture following enrichment. The number of colonies on each plate

was counted, and bacterial strains were identified using the Maldi-Tof mass spectrometry

technique (Seng P et al., 2009). Culturing experiments were performed by the lab of Medical

Microbiology at the University of Antwerp (Prof. S. Malhotra-Kumar).


143

6.3.9 In vitro culture of bone-marrow derived macrophages and

dendritic cells

The in vitro release of proinflammatory cytokines by macrophages from Chrna7-/- and OF-1

mice was quantified in the presence of GTS-21. From three OF-1 mice or Chrna7-/-, 0.5x106

bone marrow derived stem cells were plated in triplicate in full RPMI medium supplied with 5%

FBS, 1% pen/strep, 1% L-glutamine, 1mM sodium pyruvate and 20U/mL Polymyxin B for 7 days

(37°C, 95% O2, 5% CO2) with M-CSF (obtained from L-cell conditioned medium) to obtain

macrophages, or GM-CSF to obtain immature conventional dendritic cells (cDC). At day 7,

cultures were stimulated with 100 ng/mL LPS and 100 U/mL IFN-γ to polarize the macrophages

towards a proinflammatory state, or 1 µg/mL LPS and 1000 U/mL IFN-γ to obtain mature cDC.

GTS-21 was simultaneously added (final concentration of 10 µM or 100 µM). After 24h,

supernatants were collected and stored at -80°C until analysis with CBA. The purity of bone-

marrow derived macrophages and cDCs in culture was verified using a BD Accuri C6 flow

cytometer using the following cell surface markers: CD11b+F4/80+ for macrophages and

CD11c+MHCII+ for dendritic cells respectively.

6.3.10 Histology

A full thickness segment (0.5x0.5 cm) taken from the proximal colon was fixed for 24h in 4%

formaldehyde and subsequently embedded in paraffin. Transverse sections (5 µm) were

stained with haematoxylin and eosin. Inflammation was scored as described (3.8 – Histology

and immunohistochemistry). Staining was performed for lymphocytes (CD3), macrophages

(Mac-3) and neutrophils (diaminobenzidine (DAB) detection of hydrogen peroxide), as

described in Chapter 3. Sections were analyzed in a blinded manner using a color image system

(ImageJ 1.48d). The respective positive area (200x magnification) was quantified in the mucosa.

6.3.11 Presentation of results and statistical analysis

Data are presented as mean ± SEM, with ‘n’ representing the number of mice. Two-way

ANOVA followed by one-way ANOVA with post hoc Student-Newman-Keuls (SNK) analysis was

applied to compare the results of CDS, rectal temperature, motility parameters, cytokine levels,

PCR and preliminary data. The Chi-square test with post-hoc Fisher’s exact was used to

compare between the number of cultures that came back positive. Data were analyzed using

SPSS version 20.0 (IBM, Chicago) and visualized using GraphPad Prism version 5.00.


144

6.4 Results

6.4.1 Dose-dependent effects of GTS-21 on disturbed GI motility post-CLP

Preliminary experiments in the CLP-model taught us that the commonly used dosage of 4

mg/kg body weight, administered 1h prior to CLP, did not result in a profound amelioration of

GI transit (figure 6.2). Administering 8 mg/kg however protected septic animals from

developing an impaired GI motility 48h post-procedure. A dose of 12 mg/kg did not offer an

additional increase in %GE or GC. Thus, the dose of 8 mg/kg was maintained throughout the

remainder of this study.

Figure 6.2 Dose-dependent effect of preventive administration of GTS-21 on GI transit in the CLP-model. Results from a preliminary pilot-study, One-way ANOVA with post-hoc Dunnett (sham+vehicle as the control group); *p < 0.05, ***p < 0.001; n = 9-10/group %GE: gastric emptying, CLP: cecal ligation and puncture, GC: geometric center

6.4.2 Effect of GTS-21 on CLP-induced septic ileus

The CLP-procedure resulted in a reproducible occurrence of sepsis without mortality up until

48h post-procedure. Sepsis resulted in a marked increase in CDS from 0.1±0.1 in control

animals to 4.2±0.3 48h following CLP (p<0.001, figure 6.3A). GTS-21 favourably affected the

CDS at day 2 following the induction of sepsis, with CDS declining to 2.9±0.4 (p<0.05) in the

GTS-21-treated septic group. GTS-21-treated sham animals did not display any signs of toxicity.

CLP-induced sepsis resulted in an impairment of GI transit as represented by a significant drop

Gastric emptying

Sham

+ v

ehic

le

CLP

d2 +

vehic

le

CLP

+ G

TS-21

4 m

g/kg

CLP

+ G

TS-21

8 m

g/kg

CLP

+ G

TS-21

12 m

g/kg

0

20

40

60

80

100

*

%G

E

Geometric center transit

Sham

+ v

ehic

le

CLP

d2 +

vehic

le

CLP

+ G

TS-2

1 4

mg/k

g

CLP

+ G

TS-2

1 8

mg/k

g

CLP

+ G

TS-2

1 12

mg/k

g

0

2

4

6

8

10

***

GC


145

in %GE and GC. Administration of GTS-21 however protected mice from developing an

impaired GI transit, as demonstrated by a normalization of %GE and GC (figure 6.3B-C).

Sepsis resulted in a marked rise in the serum levels of the proinflammatory IL-6 and TNF-α and

of the anti-inflammatory IL-10. The same was true for the proinflammatory IL-2 and IFN-γ, with

control values below the limit of detection. In septic mice, GTS-21 caused a significant decrease

in the concentration of IL-6 and IL-2, whereas TNF-α- and IFN-γ-levels remained unaltered

compared to vehicle treated septic mice (table 6.1A). IL-10 levels significantly increased

following administration of GTS-21 in both sham and septic mice.

Similar to the serum results, the IL-6 concentration in the colonic wall rose significantly during

sepsis. This rise in IL-6 concentration was significantly inhibited by GTS-21. In septic conditions,

a significant drop in colonic IL-10 and IL-17 was observed, without an additional effect of GTS-

21. GTS-21 treatment tended to increase colonic TNF-α levels slightly in septic and control

animals, but not significantly (p=0.072) (table 6.1B). The aforementioned results were largely

confirmed by the RT-PCR results on colonic tissue (table 6.1C). Remarkably, colonic mRNA

levels for IL-10 and IL-17 actually rose during sepsis.


Sha

m

CLP

day

2

0

1

2

3

4

5vehicle

GTS-21 8 mg/kg

* *

CD

S

% Gastric Emptying

Sha

m

CLP

day

2

0

20

40

60

80

100vehicle

GTS-21 8 mg/kg

**

% G

E

Geometric Center

Sha

m

CLP

day

2

0

2

4

6

8

10

GTS-21 8 mg/kg

**

vehicle

GC

Evans blue extracted from colon

Sha

m

CLP

day

2

0

50

100

150Vehicle

GTS-21 8 mg/kg

**

µg

/100m

g c

olo

n t

issu

e


146

Figure 6.3 (previous page) - Clinical disease score, GI motility parameters and colonic permeability measurement following CLP-induced sepsis. Sepsis significantly increased CDS (upper left panel) and impaired GI motility, as demonstrated by a lower %GE (upper right panel) and lower GC (lower left panel). GTS-21-treated septic mice displayed a lower CDS and restored impaired GI transit, as demonstrated by a higher %GE and higher GC. Sepsis resulted in a significant increase in colonic permeability, demonstrated by the rise in the amount of Evans blue that was extracted from the colon (lower right panel). GTS-21 was able to normalize this increase to baseline values. Two-way ANOVA followed by One-way ANOVA and SNK post-hoc testing when appropriate; n = 7-10/group for CDS and transit parameters; n = 9-12/group for colonic permeability testing; *p ≤ 0.05. CDS: clinical disease score; CLP: cecal ligation and puncture; GC: geometric center; %GE: percentage of gastric emptying; GI: gastrointestinal

Table 6.1 Cytokine levels in serum (A) and supernatants of homogenized colons (B) measured by CBA, and determined by RT-PCR in colon (C).

sham + vehicle sham + GTS-21 CLP d2 + vehicle CLP d2 + GTS-21

A. Serum cytokines (pg/mL) IL-6 2.57 ± 0.48 3.50 ± 0.73 243.56 ± 54.89a,b 90.26 ± 31.31c

TNF-α 4.95 ± 0.29 4.83 ± 1.03 63.72 ± 21.25# 54.23 ± 25.14# IL-2 <L/D <L/D 0.38 ± 0.16# 0.18 ± 0.06#,c IFN-γ <L/D <L/D 1.12 ± 0.44# 1.02 ± 0.58# IL-10 0.59 ± 0.33 2.25 ± 0.26§ 2.33 ± 1.25# 6.39 ± 2.69#,§ IL-17A <L/D <L/D <L/D <L/D

B. Colon supernatants (pg/g colon tissue) IL-6 97.82 ± 11.11 104.14 ± 15.21 2077.86 ± 693.75a,b 617.11 ± 184.67c

TNF-α 155.13 ± 15.22 220.17 ± 44.37 197.00 ± 32.14 250.88 ± 40.77 IL-10 13663.81 ± 1116.08 13985.63 ± 1861.44 9261.71 ± 225.22# 9611.03 ± 307.01#

IL-17A 1111.49 ± 111.75 901.62 ± 164.80 547.36 ± 34.51# 558.33 ± 32.24#

IL-1β 2726.07 ± 173.72 2937.64 ± 303.05 3098.80 ± 229.66 2874.39 ± 172.69

C. mRNA colonic tissue IL-6 1.25 ± 0.32 1.01 ± 0.27 23.23 ± 14.18# 8.49 ± 2.66# TNF-α 1.00 ± 0.03 2.57 ± 0.67§ 4.07 ± 0.87# 13.10 ± 3.01#,§ IL-10 1.69 ± 0.45 2.20 ± 0.54 15.45 ± 10.90* 8.30 ± 1.43* IL-17A 1.28 ± 0.32 3.71 ± 1.61 4.01 ± 1.78 1.91 ± 0.51 IL-1β 1.17 ± 0.39 1.88 ± 0.36§ 2.21 ± 0.22# 6.27 ± 1.89#,§

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. # significant effect of CLP, § significant effect of GTS-21; for the SNK post-hoc testing: a <0.05 compared to sham + vehicle, b <0.05 compared to sham + GTS-21, c <0.05 compared to CLP d2. *p = 0.09 for the effect of CLP. When sham concentrations were below the theoretical limit of detection, the unpaired Student’s t-test was subsequently applied to compare the CLP d2 with the CLP d2 + GTS-21 group. For the PCR-results, data are expressed as relative expression (2-ΔΔCT method) and the sham + vehicle group was chosen as calibrator. Data are presented as mean ± SEM. N = 9-10 animals per group for the serum cytokines, n = 9-11 animals per group for colonic cytokines (protein level), n = 5-11 animals per group for colon PCR. <L/D, below the limit of detection; CBA, cytometric bead array; IFN, interferon; IL, interleukin; TNF, tumor necrosis factor.


147

CLP-induced sepsis increased the permeability of the colonic wall to Evans blue and GTS-21

normalized this disturbed permeability of the GI tract (figure 6.3D) without any effect on the

sham-operated mice. Cultures from sham mice were consistently negative following 24h

incubation, whereas 67% of blood cultures and 100% of MLN cultures were positive in septic

mice (table 6.2). GTS-21 significantly reduced the number of positive cultures. The mRNA-levels

of different cell adhesion proteins were all significantly upregulated during sepsis, and

administration of GTS-21 resulted in an additional significant claudin-1 mRNA levels (table 6.3).

Table 6.2 Cultures of blood and homogenized mesenteric lymph nodes.

Sham + vehicle Sham + GTS-21 CLP d2 + vehicle CLP d2 + GTS-21

A. Blood Number of positive cultures (%)

0/4 (0%) 0/5 (0%) 8/12 (67%) 3/10* (30%)

B. Mesenteric lymph nodes Number of positive cultures (%)

N/A N/A 5/5 (100%) 2/5* (40%)

Percentage of positive cultures from blood and homogenized mesenteric lymph nodes (24h incubation at 37°C, ambient air supplied with 5% CO2). Pearson’s chi-squared test for the percentage of positive cultures; *p ≤ 0.05 versus CLP day 2 + vehicle. CLP: cecal ligation and puncture; N/A: not applicable

Table 6.3 RT-PCR of cell adhesion molecules.

Sham + vehicle Sham + GTS-21 CLP d2 + vehicle CLP d2 + GTS-21

occludin 1.07 ± 0.12 1.19 ± 0.13 1.38 ± 0.23# 1.51 ± 0.13#

zonulin-1 1.13 ± 0.18 0.75 ± 0.14 1.30 ± 0.27# 1.34 ± 0.27# claudin-1 2.59 ± 0.75 4.47 ± 1.17§ 4.07 ± 1.56# 11.50 ± 1.88#,§ E-cadherin 1.11 ± 0.60 1.00 ± 0.30 2.29 ± 0.61# 2.10 ± 0.27# desmoglein-2 1.23 ± 0.26 1.30 ± 0.25 2.25 ± 0.42# 2.71 ± 0.26#

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. # significant effect of CLP, § significant effect of GTS-21. N = 9-10 animals per group.

No obvious differences were observed on haematoxylin-eosin staining between sham- and CLP-

operated mice, nor did treatment with GTS-21 result in any histological changes (score for all

four groups: 0.0±0.0, p = NS, figure 6.4). Therefore, the beneficial GI motility changes we

observed following GTS-21 treatment in the septic animals were not associated with an

alteration of the GI histology as could be observed on haematoxylin-eosin staining. Staining for

lymphocytes, macrophages and neutrophils displayed a significantly increased infiltration of

CD3+ T cells in the mucosal layer in septic animals, while GTS-21 treated CLP-mice displayed T

cells numbers similar to those observed in control animals (table 6.4). Treatment with GTS-21


148

significantly reduced the number of apoptotic cells in the epithelium of sham-operated animals

(reduced TUNEL-staining), however this was not the case for the septic animals. Performing

CLP significantly increased the number of Mac-3 positive macrophages in the epithelium, with a

small albeit not significant drop following treatment with GTS-21. No differences were

observed concerning the numbers of neutrophils using the Dab-staining.

Table 6.4 Results for the immunohistochemical stains of the mucosal layer.

Stain (cell) Sham + vehicle Sham + GTS-21 CLP d2 + vehicle CLP d2 + GTS-21

TUNEL 2.13 ± 0.29 0.23 ± 0.07* 1.75 ± 0.38 2.32 ± 0.81 Dab-1 0.01 ± 0.00 0.01 ± 0.00 0.00 ± 0.00 0.01 ± 0.00 CD3 0.64 ± 0.02 0.59 ± 0.12 1.31 ± 0.26* 0.57 ± 0.09 Mac-3 0.04 ± 0.01 0.08 ± 0.03 0.23 ± 0.09# 0.14 ± 0.06#

Percentage of the total area that stained positive for TUNEL (denoting apoptosis), CD3 (lymphocytes), Mac-3 (macrophages) and Dab (neutrophils) in the mucosal layer including the lamina propria. Two-way ANOVA or One-way ANOVA with post-hoc SNK test as appropriate, n = 5-6/group, *p<0.05 compared to the other three groups, # significant effect of the procedure CLP.

Figure 6.4 Representative haematoxylin-eosin staining of colonic tissue in vehicle-treated sham (A) and CLP-animals (B), as well as in GTS-21 treated sham (C) and CLP-animals (D); 200x magnification. CLP: cecal ligation and puncture.


149

6.4.3 CLP and GTS-21 in Chrna7-/- mice

In Chrna7-/- mice mortality was significantly increased 48h following CLP, in contrast to their

wild-type littermates of which all survived (figure 6.5). Administration of GTS-21 was not able

to impede this increased mortality. GTS-21 still significantly decreased CDS in Chrna7-/- mice

and also prevented the development of impaired gastric emptying in septic Chrna7-/- mice

(table 6.5). Furthermore, GTS-21 treatment numerically decreased the major proinflammatory

cytokines (IL-6, TNF-α and IL-17A) as well as the anti-inflammatory IL-10, in serum as well as

colon. This effect was however not confirmed at colonic mRNA levels, with near-identical

values for animals that received GTS-21 in comparison with vehicle-treated septic Chrna7-/-

mice.

Figure 6.5 Survival curve of OF-1, Chrna7 -/- and Chrna7+/+ mice. Kaplan-Meier curve displaying survival of Chrna7-/- receiving vehicle (dash-dotted line) or GTS-21 (dashed line) and Chrna7+/+ mice (littermates, solid line) up to 48h following the CLP-procedure. Kaplan Meier estimator with log rank test, p = 0.030; n = 11-14/group. Chrna7: alpha7 nicotinic acetylcholine receptor; CLP: cecal ligation and puncture; KO + GTS: Chrna7-/- mice treated with GTS-21; KO + veh: Chrna7-/- mice treated with vehicle solution; WT: wild type-littermates


150

Table 6.5 Results for the Chrna7-/- mice

Chrna7-/- - CLP + vehicle Chrna7-/- - CLP + GTS-21 WT littermates – CLP

A. Clinical disease score CDS 6.08 ± 0.75 4.25 ± 0.35a 4.67 ± 0.84

B. Gastrointestinal motility parameters %GE 72.81 ± 10.13 95.02 ± 2.99a 81.88 ± 8.47 GC 3.52 ± 0.50 4.53 ± 0.38 3.45 ± 0.62

C. Serum cytokines (pg/mL) IL-6 4384.09 ± 2125.15 905.48 ± 373.89 962.66 ± 605.02 TNF-α 193.53 ± 74.47 58.95 ± 20.92 51.67 ± 30.67 IL-10 34.48 ± 16.87 13.11 ± 10.41 51.83 ± 50.77 IFN-γ 1.10 ± 0.29 2.68 ± 0.66 1.52 ± 0.45 IL-17A 2.70 ± 1.28 1.88 ± 0.71 0.92 ± 0.48

D. Colon cytokines (pg/g tissue) IL-6 386.53 ± 179.54 211.30 ± 78.74 N/A TNF-α 49.40 ± 39.04 20.11 ± 7.68 N/A

E. mRNA colonic tissue IL-6 2.29 ± 0.76 1.13 ± 0.37 N/A TNF-α 1.36 ± 0.34 1.02 ± 0.11 N/A IL-10 1.51 ± 0.39 1.12 ± 0.36 N/A IL-1β 1.80 ± 0.68 1.25 ± 0.50 N/A IL-17A 1.76 ± 0.60 2.17 ± 1.39 N/A

One-way ANOVA, non-parametric Kruskall-Wallis test for Independent Samples or Student’s t-test when applicable with appropriate post-hoc testing; a p<0.05 compared to CLP + vehicle. Data are presented as mean ± SEM. For the PCR-results, data are expressed as relative expression (2-ΔΔCT method) and the CLP + vehicle was chosen as calibrator. N = 5-11 animals per group for CDS, GI motility parameters, serum cytokines and PCR, n = 10-11 animals per group for colon cytokines. %GE: percentage of gastric emptying; <L/D, below the limit of detection; CLP: cecal ligation and puncture; GC: geometric center; IFN, interferon; IL, interleukin; TNF, tumor necrosis factor.

6.4.4 Effects of splenectomy on CLP-induced septic ileus

Performing SPLX one week prior to CLP protected mice from developing reduced GI transit. We

noted a significant drop in CDS as animals were protected from developing ileus, with GI transit

measurements showing a near-normal %GE and GC (table 6.6). Serum levels of the

proinflammatory IL-6 and TNF-α were significantly reduced in comparison with septic animals

with an intact spleen. The same effects could be observed at the colonic level, however results

were not significant for IL-6. SPLX had no effects on sham-operated (i.e. non-septic) animals.

The administration of GTS-21 to splenectomized animals did not result in an additional benefit

on CDS, GI motility or cytokine levels (table 6.6) suggesting that the main effect of GTS-21 in

this animal model was mediated via splenic macrophages in the preventative setting.


151

Of note, we subsequently performed splenectomy 6 or 24h following the CLP-procedure

(termed ‘curative splenectomy’) in order to assess the role of the spleen during consolidated

sepsis. It should be noted that these animals however demonstrated an impaired GI transit

identical to that of none-splenectomized septic animals; their CDS remained unaltered as well,

much in contrast to the animals that underwent splenectomy in a preventive fashion. Thus, no

beneficial effects of curative splenectomy were observed on GI transit in our experiments

(table 6.7 and table 6.8).

Table 6.6 Effects of preventive splenectomy on CLP-induced septic ileus

CLP + vehicle CLP + GTS-21 SPLX + CLP + vehicle

SPLX + CLP + GTS-21

non-septic animals

A. Clinical disease score CDS 4.38 ± 0.42 3.42 ± 0.48 2.73 ± 0.27§ 2.82 ± 0.40§ 0.1 ± 0.1

B. Gastrointestinal motility parameters %GE 53.27 ± 10.31 79.49 ± 8.98a 83.62 ± 5.91a 77.64 ± 11.80a 92.32 ± 5.27 GC 2.25 ± 0.50 4.29 ± 0.54a 3.87 ± 0.44 3.55 ± 0.52a 4.85 ± 0.30

C. Serum cytokines (pg/mL) IL-6 243.56 ± 54.89 90.26 ± 31.31 a 74.31 ± 17.45 a 64.98 ± 7.16a 2.57 ± 0.48 TNF-α 63.72 ± 21.25 53.23 ± 25.14 11.52 ± 1.98§ 12.81 ± 3.91§ 4.95 ± 0.29 IL-2 0.38 ± 0.16 0.18 ± 0.06 0.43 ± 0.04 0.33 ± 0.05 <L/D IL-10 2.33 ± 1.25 6.39 ± 2.70* 0.31 ± 0.18§ 1.18 ± 0.61§,* 0.59 ± 0.33 IFN-γ 1.12 ± 0.44 1.02 ± 0.58 <L/D 0.61 ± 0.18 <L/D IL-17A <L/D <L/D <L/D <L/D <L/D

D. Colon cytokines (pg/g tissue) IL-6 1281.50 ± 932.10 134.88 ± 36.15 214.22 ± 75.88 487.49 ± 343.87 N/A TNF-α 137.15 ± 61.36 46.46 ± 5.03a 29.97 ± 8.60a 46.04 ± 17.21a N/A

E. mRNA colonic tissue IL-6 23.23 ± 14.18 8.49 ± 2.66 19.28 ± 7.28 15.06 ± 5.52 1.25 ± 0.32 TNF-α 4.07 ± 0.82 13.10 ± 3.01 8.28 ± 1.09 9.05 ± 1.42 1.00 ± 0.03 IL-10 15.45 ± 10.90 8.31 ± 1.43 9.34 ± 1.50 9.57 ± 1.41 1.69 ± 0.45 IL-17A 4.01 ± 1.78 1.91 ± 0.51 11.15 ± 3.64 8.59 ± 4.04 1.28 ± 0.32 IL-1β 2.21 ± 0.22 6.27 ± 1.89 7.06 ± 2.13§ 6.59 ± 1.39§ 1.17 ± 0.39

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. § significant effect of splenectomy; *p = 0.07 for the effect of GTS-21, a <0.05 compared to CLP + veh. For the PCR-results, data are expressed as relative expression (2-ΔΔCT method) and the non-septic animals (right column – sham animals not subjected to splenectomy) was chosen as calibrator. Data are presented as mean ± SEM. N = 11-13 animals per group for the clinical disease score and gastrointestinal transit parameters, n = 9-11/group for serum cytokines, n = 5-11 animals per group for colon cytokines and PCR. %GE: percentage of gastric emptying; <L/D, below the limit of detection; CLP: cecal ligation and puncture; GC: geometric center; IFN, interferon; IL, interleukin; SPLX: splenectomy; TNF, tumor necrosis factor.


152

Table 6.7 Effects of curative splenectomy (6h post-CLP) on CLP-induced septic ileus

CLP + vehicle CLP + GTS-21 CLP + SPLX + vehicle

CLP + SPLX + GTS-21

non-septic animals

A. Clinical disease score CDS 4.38 ± 0.42 3.42 ± 0.48 4.00 ± 0.45 3.50 ± 0.34 0.1 ± 0.1

B. Gastrointestinal motility parameters %GE 53.27 ± 10.31 79.49 ± 8.98* 57.39 ± 17.60 43.03 ± 24.85 92.32 ± 5.27 GC 2.25 ± 0.50 4.29 ± 0.54* 2.71 ± 0.69 2.36 ± 0.79 4.85 ± 0.30

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. *p < 0.05 compared to the other three groups; n = 5-7 animals per group. %GE: percentage of gastric emptying; CLP: cecal ligation and puncture; GC: geometric center; SPLX: splenectomy.

Table 6.8 Effects of curative splenectomy (24h post-CLP) on CLP-induced septic ileus

CLP + vehicle CLP + GTS-21 CLP + SPLX + vehicle

CLP + SPLX + GTS-21

non-septic animals

A. Clinical disease score CDS 4.38 ± 0.42 3.42 ± 0.48 4.50 ± 1.56 2.75 ± 0.25 0.1 ± 0.1

B. Gastrointestinal motility parameters %GE 53.27 ± 10.31 79.49 ± 8.98* 51.51 ± 19.05 26.40 ± 18.14 92.32 ± 5.27 GC 2.25 ± 0.50 4.29 ± 0.54* 3.18 ± 0.97 2.08 ± 0.72 4.85 ± 0.30

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. *p < 0.05 compared to the other three groups; n = 5-7 animals per group. %GE: percentage of gastric emptying; CLP: cecal ligation and puncture; GC: geometric center; SPLX: splenectomy.

6.4.5 In vitro culture of bone-marrow derived macrophages and DCs

Over 90% of macrophages cultured were determined to be CD11b+F4/80+ macrophages.

Incubating proinflammatory macrophages with 10 and 100 μM GTS-21 dose-dependently

reduced the release of TNF-α with 59% and 92% respectively (p<0.001) compared to vehicle,

and reduced the release of IL-6 with respectively 26% and 59% (p<0.001). The release of IL-10

was not altered. For the cDC cultures similar results were obtained with a dose-dependent

reduction in the release of TNF-α (47% and 57%, p<0.001), IL-6 (13% and 37%, p<0.001) and IL-

12p70 (32% and 55% respectively, p<0.001). Due to surprising effects of GTS-21 in Chrna7-/-

mice, the same protocol was applied to bone marrow derived cells from Chrna7-/- mice.

Surprisingly, the release of TNF-α and IL-6 from macrophages was also reduced following

incubation with GTS-21 (TNF-α: 14% and 87% reduction respectively; IL-6: 58% decrease

following incubation with 100 µM GTS-21, all p=0.001). Similar results were obtained for cDC

cultures (IL-6: decrease with 12% and 80% and TNF-α: decrease with 75% and 95%, p<0.001).


153

6.5 Discussion

In this study, we provide evidence for the beneficial effects of GTS-21, an α7nAChR agonist, on

GI motility disturbances, inflammation and colonic permeability during polymicrobial

abdominal sepsis. Sepsis was induced by cecal ligation and puncture, a widely accepted animal

model in sepsis research (Buras et al., 2005). CLP has been demonstrated to impair in vitro

colonic peristalsis and muscle contractions (Overhaus et al., 2004). Our data show for the first

time, to the best of our knowledge, the occurrence of impaired GI motility 48h following CLP-

induced sepsis in OF-1 mice. Serum levels of the proinflammatory cytokines IL-6, TNF-α and IL-2

were upregulated following CLP whilst serum IL-10 levels, an anti-inflammatory cytokine, were

also markedly elevated during CLP-induced sepsis at this time point of investigation. This

reflects a simultaneously increased synthesis of anti-inflammatory markers counteracting the

proinflammatory cytokines and their tissue damaging effects (Rivers et al., 2013), confirming

the concept of simultaneous activation of pro- as well as anti-inflammatory agents during

sepsis as was postulated by Hotchkiss and colleagues (2013), as already discussed in Chapter 1.

GTS-21, an α7nAChR agonist, was able to protect animals from developing CLP-induced septic

ileus and caused a significant decrease in serum level of IL-6 and CDS in septic mice. Serum

TNF-α levels also increased significantly during sepsis, but levels were not reduced by GTS-21-

treatment. This is in contrast to previous literature: Giebelen et al. demonstrated that

pretreatment with GTS-21 strongly inhibited TNF-α release in the peritoneum following LPS-

injection (Giebelen et al., 2007), and Khan et al. provided us with in vitro data confirming that

GTS-21 nearly abolished the release of TNF-α in a murine macrophage cell culture (Khan et al.,

2012), much alike the data we indeed obtained from our cell culture experiments. These

results were also confirmed by the group of Tracey in a RAW264.7 mouse macrophage cell line,

and GTS-21 dose-dependently decreased TNF-α serum levels in rats challenged with

endotoxemia (Pavlov et al., 2007). In the CLP-model, the authors however only detected

HMGB-1 serum levels and unfortunately no TNF-α levels. Finally, Yeboah et al. demonstrated

that kidney TNF-α levels were reduced in rats with ischemia-reperfusion induced acute kidney

injury, if they received pretreatment with GTS-21 (Yeboah et al., 2008). However we made the

same observation at the colonic level as in the serum: treatment with GTS-21 did not decrease

TNF-α, on the contrary, levels were actually increased numerically (albeit not significantly).

These results were confirmed at the mRNA level.


154

Chatterjee et al. demonstrated that cholinergic agonists, such as GTS-21, mediate their effects

by reducing the phosphorylation of janus kinase (JAK)-2 and subsequently inhibiting the

phosphorylation of the signal transducer and activator of transcription (STAT)-3 (Chatterjee et

al., 2009). This signal transduction pathway is of particular importance for IL-6; TNF-α however

can mediate its effects via three different pathways resulting in activation of NF-κB, MAPK

and/or caspase 3, each with its own effects ranging from inflammation to apoptosis of the cell.

Others have demonstrated that TNF-α serum levels already peaked during the first 5-12h

following CLP, with a second peak at 24h and a rapid decline thereafter (Wintersteller et al.,

2012), making it plausible that we missed a possible beneficial effect of GTS-21 on TNF-α-

secretion. Furthermore, as several other cytokines peak during the initial 6 to 12h after CLP, we

could have overlooked significant effects on other cytokines at the 48h time point that play a

pivotal role in the ensuing septic cascade. Strikingly, Levy et al. also showed that TNF-α levels

remained unaltered following VNS in a mouse model of trauma and hemorrhagic shock (Levy et

al., 2013). It should however be noted that during the subsequent therapeutic interventions we

performed in our model, as described in the upcoming Chapter 7, we actually did consistently

find a significant decrease in serum TNF-α levels, thereby confirming the compound-specific

effect of GTS-21 on TNF-α in this particular animal model, and in this particular observed

timeframe.

We studied sepsis-induced changes on colonic permeability by means of the dye Evans blue,

and found a clear impairment of the GI barrier function during sepsis. We demonstrated that

GTS-21 was able to counteract the sepsis-induced deleterious effects on GI permeability as

reflected by a significant decrease of bacterial translocation to MLN and blood. As it has been

demonstrated that the draining lymphatics play an important role in the occurrence of

subsequent multi-organ failure (Deitch et al., 2010), results obtained from MLN analysis could

be of clinical relevance. These results suggest a link between inflammation and colonic

permeability (and thus bacterial translocation). In this regard, the proinflammatory cytokine IL-

6 seems an interesting player accountable for numerous deleterious effects. Others

demonstrated that GI permeability was preserved in an IL-6 knockout (IL-6-/-) mouse model,

whereas IL-6+/+ mice displayed a profound increase in GI permeability as was determined using

Ussing chambers (Wang et al., 2001). As suggested in other studies, TNF-α and HMGB1 are also

known to have a detrimental impact on gut permeability (Fink et al., 2009), with TNF-α being


155

able to regulate tight junction size- and charge-selectivity (Gitter et al., 2000; Keita &

Soderholm, 2010), and anti-TNF-α therapy has been shown to ameliorate the impaired GI

barrier function in inflammatory bowel diseases (Odenwald et al., 2014). The importance of

TNF-α was not very pronounced at this time point in the current study whereas the strong

decrease in IL-6 coincided with a normalization of colonic permeability as determined by the

Evans blue method. As bacterial translocation may occur paracellularly when tight junction

proteins are broken down (MacFie et al., 1999; Keita et al., 2010), we studied the expression of

different adhesion proteins in the colon and found that mRNA expression of different cell

adhesion proteins such as claudin-1, occludin and zonulin-1 were upregulated 48h post-CLP

coinciding with increased colonic permeability. Eadon et al. demonstrated increased mRNA

expression of various tight junctional proteins in the kidney following LPS-induced

endotoxemia, with a reduced protein expression and their impaired localization in the

intestinal epithelial barrier complex (Eadon et al., 2012). The increase in mRNA thus could

represent an attempted repair response. Our finding of increased claudin mRNA expression

following various proinflammatory insults were also communicated by others (Van Itallie et al.,

2006).

In order to study whether immune cells residing in the spleen or GI tract wall were the main

target population mediating the effects of GTS-21, we performed preventive splenectomy prior

to CLP. Splenectomized mice appear to be protected from sepsis-induced inflammation and

ileus: they displayed a normal GI transit and lower serum levels of the proinflammatory

cytokines in comparison to their non-splenectomized counterparts following CLP. Recent

papers highlight the role of the spleen as the predominant source of immune cells: preventive

splenectomy protected against the development of DSS-colitis in mice (Munyaka et al., 2014)

and against lethal sepsis (Huston et al., 2006). As GTS-21 did not lead to an additional benefit

on serum cytokine levels in splenectomized animals, the main effects of GTS-21 observed

during sepsis could be attributed to these splenic immune cells. In vitro experiments attribute

an important role to macrophages and dendritic cells.

In animals that were splenectomized in a ‘curative’ fashion, i.e. 6 or 24h following the CLP-

procedure, we did not note any amelioration of sepsis-induced GI motility changes or

inflammation, indicating that the spleen and the immune cells residing there are

predominantly involved in the initiation of the septic cascade.


156

The observation that GTS-21 still generated beneficial effects in septic Chrna7-/- mice that lack

the α7nAChR warrants further study. Mortality was increased in septic vehicle-treated Chrna7-/-

mice, allocating a protective role to the α7nAChR during sepsis. Preventive administration of

GTS-21 to Chrna7-/- mice did not improve survival following CLP. However, a clear numerical

amelioration of GI transit and serum levels of proinflammatory cytokines was observed. This

could be explained by the pharmacological properties of this compound, as GTS-21 can also

bind onto the α4β2 nAChR (Briggs et al., 1997). Van der Zanden et al. have demonstrated that

activation of α4β2nAChR also inhibits secretion of proinflammatory cytokines from intestinal

and peritoneal macrophages, with an increase in bacterial phagocytosis (Van der Zanden et al.,

2009). Others demonstrated that activation of stomach macrophages could be attenuated via

the nicotinic β2 receptor (Nemethova et al., 2013). The weak agonist activity of GTS-21 on the

α4β2nAChR could very well explain some of the beneficial effects of GTS-21, and might even

explain the lowering in the number of positive cultures from blood and MLN in wildtype OF-1

mice, due to the amplification of bacterial phagocytosis by peritoneal macrophages.

Limitations of this study include the simultaneous prophylactic and curative administration of

the compound, reducing clinical applicability, and the fixed time point studied (48h). The

results however justify the need to further study possible therapeutic agents targeting the

impaired GI mucosal barrier function and mucosal inflammation, as these drugs may prove to

be advantageous in a subpopulation of septic patients.

Furthermore, these results justify the optimization of therapies targeting the vagal anti-

inflammatory pathway during abdominal sepsis. In this respect, preliminary studies have

already demonstrated that GTS-21 administered prior to an LPS-injection in a healthy control

population of non-smoking individuals will yield a significant correlation between plasma levels

of GTS-21 and lower levels of several proinflammatory cytokines (IL-6, TNF-α and IL-1RA); there

was however no difference detected in overall plasma cytokine levels between GTS-21 treated

and vehicle-treated subjects (Kox et al., 2011). It should be noted however that the subjects

received 150 mg of GTS-21 thrice daily during three days before the onset of the experiment.

Up until then this was reported to be the safest dose that could be tolerated by a human

subject, and is also the maximum dose studied in trials targeting negative symptoms in

schizophrenic patients and cognitive impairment in Alzheimer’s disease. Rosas-Ballina et al.


157

(2009) showed in vitro that GTS-21 attenuated the TNF-α and IL-1β release from human

monocytes following LPS-stimulation. Currently, human clinical trials are ongoing studying the

effect of transcutaneous VNS prior to LPS-administration on inflammation (EudraCT number

2012-005687-97; clinicaltrials.gov identifier NCT01944228), in inflammatory bowel disease

(clinicaltrials.gov identifier NCT02311660) and in active rheumatoid arthritis (clinicaltrials.gov

identifier NCT01552941). Of note, the effect of VNS is currently also being studied in epilepsy,

esophageal hypersensitivity syndrome, heart failure and type 2 diabetes. Several phase-III trials

are also ongoing investigating the effects of GTS-21 and other α7nAChR agonists in patients

with neurocognitive deficits of various origins.

In summary, in an animal model of polymicrobial abdominal sepsis, we observed that

inflammation at the GI tract level coincides with the occurrence of GI motility disturbances and

translocation of bacteria from the GI tract into the bloodstream and lymphatics. Beneficial

effects of GTS-21 were observed, such as a reduction in inflammation at the level of the GI

tract, an amelioration of the colonic barrier function and a decrease in the number of positive

blood cultures, a relevant clinical endpoint. These results however cannot be solely attributed

to α7nAChR activation.

Chapter 7

Immunological modulation of septic ileus:

Effect of antibodies to interleukin-6 on CLP-induced ileus


Nullens S, Peleman C, Staessens M, Plaeke P, Malhotra-Kumar S, Francque SM, De

Man JG, De Winter BY. Beneficial effects of anti-interleukin-6 antibodies on impaired

gastrointestinal motility, inflammation and increased colonic permeability in a

murine model of sepsis are most pronounced when administered in a preventive

setup.

PLoS One 2016 Apr 4;11(4):e0152914. doi: 10.1371/journal.pone.0152914

Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6

160

7.1 Abstract

Background and objectives: During sepsis, gastrointestinal ileus, mucosal barrier dysfunction

and bacterial translocation are accepted to be important triggers that can maintain or

exacerbate the septic state. In the cecal ligation and puncture animal model of sepsis, we

demonstrated that systemic and colonic interleukin-6 levels are significantly increased

coinciding with an impaired colonic barrier function and ileus. We therefore aimed to study the

effect of preventive or curative administration of anti-IL6 antibodies on overall GI motility,

colonic permeability and translocation of intestinal bacteria in blood and mesenteric lymph

nodes in the mouse cecal ligation and puncture model.

Methods: OF-1 mice were randomized to either the preventive or curative protocol, in which

they received 1 mg/kg antibodies to interleukin-6, or its IgG isotype control solution. They

subsequently underwent either the cecal ligation and puncture procedure, or sham-surgery. GI

motility was assessed 48h following the procedure, as well as colonic permeability, serum and

colon cytokines as well as colonic tight junction proteins at the mRNA level. Cultures of blood

and mesenteric lymph nodes were analyzed.

Results: Preventive administration of anti-interleukin-6 antibodies successfully counteracted

the gastrointestinal motility disturbances and impaired colonic barrier function that could be

observed in vehicle-treated septic animals. Serum and colonic levels of proinflammatory

cytokines were significantly lower when animals were preventively treated with anti-

interleukin-6 antibodies. A repetitive injection 24h later resulted in the most pronounced

effects. Curative treatment significantly lowered systemic and colonic inflammation markers

while the effects on transit and permeability were no longer significant.

Conclusions: Cecal ligation and puncture resulted in septic ileus with an increased colonic

permeability. Antibodies to interleukin-6 were able to ameliorate gastrointestinal motility,

suppress inflammation and normalize the permeability of the colonic wall, with the preventive

administration combined with a repeat injection being far more efficacious than the sole

preventive or curative one.


161

7.2 Introduction

As we elaborately discussed in the first chapter, researchers nowadays consider the GI tract to

play a major role in the development and maintenance of sustained inflammation and multiple

organ failure during sepsis (Deitch, 2012). Data obtained from animal experiments showed that

gut-derived factors in lymph collected from mesenteric lymph nodes can damage distant organ

systems in an animal model of trauma followed by hemorrhagic shock (Levy et al., 2013),

supporting the ‘gut-lymph’ hypothesis and the importance of the failing GI tract during sepsis

(Deitch, 2012; Deitch et al., 2006).

A major proinflammatory cytokine that plays a pivotal role in the pathogenesis of sepsis is

interleukin-6 (IL-6). Its major detrimental proinflammatory effects are mediated via the binding

of IL-6 to its soluble receptor (sIL-6R). This IL-6 – sIL-6R complex will subsequently bind onto the

signaling receptor protein gp130 (gp130), which is ubiquitously expressed on all cells, a process

termed trans-signaling. Its effects are extensively reviewed elsewhere (Scheller et al., 2011;

Kang et al., 2015). Trans-signaling is the major pathway via which the immune system is

activated, and will play an important role in the transition from innate towards acquired

immunity, the release of acute phase reagents, the secretion of immunoglobulins from B cells

and the skewing of T cells towards a predominantly Th17 subtype in favor of the regulatory T

cell subset (Kimura et al., 2010; Mangodt et al., 2015). Classic signaling occurs when IL-6 binds

onto a membrane-bound IL-6R, and is mainly responsible for its anti-inflammatory and

regenerative effects (Scheller et al., 2011). This membrane-bound IL-6R is only expressed on

the surface of a limited number of cells, including the GI epithelial cells. At the GI tract level,

classic signaling will result in the proliferation of epithelial cells and inhibit apoptosis (Hunter &

Jones, 2015).

Besides its pro- and anti-inflammatory properties, and proliferative epithelial efects, it was

however already acknowledged decades ago that IL-6 also has a detrimental effect on several

epithelial layers (Maruo et al., 1992; Paul et al., 2003; Tazuke et al., 2003). The presence of IL-6

is mandatory for the development of gut barrier dysfunction, as Yang et al. showed that IL-6 KO

mice were protected from developing subsequent impaired GI barrier function in a mouse

model of hemorrhagic shock (Yang et al., 2003). However, IL-6 KO mice displayed more severe

inflammation in a mouse model of DSS-colitis, which can be explained by the absence of the


162

regenerative effects of IL-6 on intestinal epithelial cells (Grivennikov et al., 2009; Dann et al.,

2008).

In the search for a new therapeutic target, anti-cytokine strategies are dealing with a

somewhat bad reputation in the field of sepsis. The translation from murine studies towards

the human medical treatment was often unsuccessful (Dyson et al., 2009; Rittirsch et al., 2007),

presumably due to methodological limitations such as very extensive and comprehensive

patient inclusion criteria, design of the study or rigid clinical endpoints (Masson et al., 2014).

Besides, murine sepsis models have provided us with an extensive knowledge on the

mechanisms of sepsis, but these experimental models are often performed under tightly

controlled circumstances, making extrapolation towards the ‘average’ septic patient population

nearly impossible (Buras et al., 2005; Rittirsch et al., 2007; Dyson et al., 2009; Pène et al.,

2015). Many of these authors however postulate that stratifying patients based upon their

immunological profile prior to interventions targeting the immune system, may yield more

beneficial results, a hypothesis we firmly support. Currently, clinical trials targeting IL-6

however remain limited to pathologies such as rheumatoid arthritis, polymyalgia and various

forms of cancer. Table 7.2 provides an overview of the current clinical trials with anti-IL-6

antibodies in different disease states.

In Chapter 6, we observed a significant increase in serum and colonic levels of IL-6 in the cecal

ligation and puncture (CLP) model that coincided with ileus and a disturbance of the colonic

barrier function. So far conflicting data on the blockage of IL-6 in animal models of sepsis were

reported (table 7.1). Briefly, some authors demonstrated an improvement in survival (Gennari

et al., 1994; Gennari et al., 1995; Pallua et al., 2003; Barkhausen et al., 2011; Mostafa Anower

et al., 2011), whereas others failed to demonstrate a survival benefit (Vyas et al., 2005). This

improved survival was furthermore demonstrated to be dependent on the amount of antibody

that was infused (Pallua et al., 2003; Riedemann et al., 2003). Studies so far focused primarily

on survival and provided little data on the pivotal role of the gastrointestinal tract during

sepsis. Wang et al. showed that septic IL-6 KO mice did not display increased GI permeability,

whereas their septic wild-type counterparts displayed increased mucosal permeability (Wang

et al., 2001). We therefore aimed to further investigate the effects of directly blocking IL-6 on

gastrointestinal inflammation, motility and permeability. Antibodies were administered either


163

immediately prior to CLP (preventive setup) with or without a repeat-injection 24h following

CLP, or in a curative fashion (only one injection 24h following CLP).

Table 7.1 Overview of research papers reported on the effect of anti-interleukin-6 antibodies in mouse models of sepsis and/or (septic) shock.

Authors Animal model Intervention Observed effects

Libert et al., 1992

Injection with LPS Antimurine IL-6 antibody

Anti-IL-6 protected dose-dependently against LPS, but not against the highest dosage of LPS

Gennari et

al., 1994 Bacterial gavage and burn injury

Antimurine IL-6, IgG or placebo 1h before insult

Increased survival following anti-IL-6 therapy and a decrease in bacterial translocation in MLN.

Gennari et

al., 1995 Bacterial gavage and thermal injury

Antimurine IL-6, IgG or placebo 2/4/8h after insult

Anti-IL-6 enhanced clearance of bacteria when given 2h post-insult.

Wang et al., 2001

Cecal ligation and puncture

IL-6 WT and KO mice Intestinal permeability was increased in septic IL-6 WT mice, but not in IL-6 KO mice

Pallua et al., 2003

Contact burn followed by LPS injection

Anti-IL-6 and/or anti-IL-6R antibody before or after the LPS stimulus

Both compounds effectively prevented systemic inflammation, effect was more pronounced when administered after LPS injection.

Riedemann et al., 2003


Anti-IL-6 (different dosages) immediately or 4h after CLP

Improved survival in anti-IL-6 treated mice; highest dosage of anti-IL-6 increased mortality; delayed infusion yielded no survival benefit.

Yang et al., 2003

Hemorrhagic shock and resuscitation

WT mice and IL-6-KO mice

IL-6 was necessary to induce gut hyperpermeability and dysfunction.

Vyas et al., 2005


Early antibiotics or anti-IL-6 based upon serum IL-6 levels

Early antibiotic treatment improved outcome in mice predicted to die, no effect of anti-IL-6 on survival.

Mees et al., 2009

Hemorrhagic shock and cecal ligation and puncture

Blocking of trans-signalling with gp130Fc following shock + CLP

Treatment with gp130Fc significantly reduced IL-6 levels, but survival was unaffected.

Barkhausen et al., 2011


sgp130Fc (inhibits IL-6 trans-signalling) or anti-IL-6, 24h before or 24h after CLP

Pretreatment with sgp130Fc ameliorated survival, whereas anti-IL-6 did not.

Mostafa Anower et

al., 2011


Pretreatment with siRNA

Pretreatment with siRNA induced extended survival.

CLP: cecal ligation and puncture; IL: interleukin; KO: k,ock-out; LPS: lipopolysaccharide; MLN: mesenteric lymph nodes; WT: wild type; as of April 1st 2016.

Table 7.2 Current clinical trials studying the effect of antibodies to interleukin-6 in various pathologies.

Trial number Compound Target group

NCT00841191 CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)

Safety, efficacy and pharmacokinetic study in patients with solid tumors

NCT00718718 CNTO 136 (sirukumab, monoclonal anti-IL-6 antibody)

Patients with active rheumatoid arthritis despite MTX treatment

NCT01284569 ALX-0061 (anti-IL-6 receptor nanobody) Safety and efficacy in rheumatoid arthritis patients NCT01636557 CNTO 136 (sirukumab, monoclonal anti-IL-6

antibody) Effect on CYP450 enzyme activity in rheumatoid arthritis patients

NCT01491074 tocilizumab (monoclonal anti-IL-6 receptor antibody)

Effect in non-STEMI patients

NCT01396317 tocilizumab (monoclonal anti-IL-6 receptor antibody)

Treatment of polymyalgia rheumatica patients

2011-004763-72

BMS-945429 (monoclonal anti-IL-6 antibody) Clinical efficacy and safety of induction and maintenance therapy in Crohn patients

2011-004016-29

clazakizumab (monoclonal anti-IL-6 antibody) Efficacy and safety in patients with rheumatoid arthritis

2010-023956-99

BMS-945429 (monoclonal anti-IL-6 antibody) Efficacy and safety with or without MTX in patients with rheumatoid arthritis and other inflammatory polyarthropathies

2006-001671-38

CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)

Patients with hormone-refractory prostate cancer with mitoxantrone (versus mitoxantrone)

2009-012380-34


Castleman’s disease in combination with best supportive care (versus best supportive care)

2009-017237-22


Siltuximab or placebo combined with bortezomib and dexamethasone in patients with refractory or relapsed multiple myeloma

2010-022243-38

CNTO 136 (sirukumab, anti-IL-6 monoclonal antibody)

Patients with rheumatoid arthritis

2010-022242-24

CNTO 136 (sirukumab, anti-IL-6 monoclonal antibody)


2011-004763-72

BMS-945429 (monoclonal anti-IL-6 antibody) Clinical efficacy and safety of induction and maintenance therapy in patients with Crohn’s disease

2011-001735-22


Patients with high-risk smoldering multiple myeloma

2006-001904-36


Patients with multiple myeloma in combination with bortezomib


165

2010-023956-99

BMS-945429 (monoclonal anti-IL-6 antibody) Efficacy and safety with or without methotrexate in rheumatoid arthritis and other inflammatory polyarthropathies

2010-020968-38

CNTO 136 (sirukumab, human monoclonal anti-IL-6 antibody)

Patients with lupus nephritis

2006-005704-13


Women with recurrent epithelial ovarian cancer, fallopian tube carcinoma,primary peritoneal carcinoma

2008-004931-39

BMS-945429 (monoclonal anti-IL-6 antibody) Patients with rheumatoid arthritis

2009-010813-57

CDP6038 (olokizumab, monoclonam anti-IL-6 antibody


2006-001897-26


Patients with refractory or relapsed multiple myeloma

2011-000261-12


Combined with best supportive care in anemic patients

2011-004016-29

clazakizumab (monoclonal anti-IL-6 antibody) Patients with active psoriatic arthritis

2008-007157-12


Patients with multiple myeloma

(as of April 1st 2016, www.clinicaltrials.gov and clinicaltrialsregister.eu)


166


7.3.1 Animals

Eight week old OF-1 mice were obtained from Charles River and housed in groups of 6 animals

in standardized conditions (12h light-dark cycle, 21 ± 2°C, 40-60% humidity) with unlimited

access to regular chow and tapwater. Mice were allowed to acclimatize 10 days before the

experiments. All experiments were approved by the Committee for Medical Ethics and the use

of Experimental Animals at the University of Antwerp (file number 2012-42).

7.3.2 Induction of sepsis

In order to induce sepsis, the cecal ligation and puncture procedure was performed as

previously described in Chapter 3 (Buras et al., Nat Rev 2005). In short, mice were anesthetized

and placed in the supine position. A midline laparotomy was performed, and the cecum was

exteriorized, ligated for 50% with a 4/0 silk thread and punctured once through-and-through

with a 25G needle in order to obtain a mild sepsis without mortality. The ligated cecum was

repositioned into the abdominal cavity, and the abdomen was closed in layers. Mice received

fluid resuscitation and pain relief. Sham-operated mice served as controls. The experiments

described below were initiated 48h following the CLP or sham-procedure (figure 7.1). Animals

were prematurely sacrificed when they lost over 15% of their baseline body weight or when

they appeared to be moribund or had a clinical disease score (CDS) > 8 (Heylen et al., 2013).

7.3.3 Experimental design

In a first set of experiments, a preventive protocol was designed in which mice received a single

i.p. injection of anti-IL-6 antibodies (1 mg/kg) or its control compound (an irrelevant IgG1 kappa

isotype control, 1 mg/kg) 5 min prior to the start of the CLP- or sham-procedure. Based upon

data previously published in literature and own preliminary experiments (data not shown), 1

mg/kg was chosen as the optimal dose (Riedemann et al., 2003; Pallua et al., 2003). In a second

set of experiments (preventive protocol with repeated injection), mice did additionally receive

a second injection with the antibodies or the control compound 24h following the CLP-or sham-

procedure. In a third set of experiments, a curative protocol was designed in which mice

received a single i.p. injection of anti-IL-6 antibodies (1 mg/kg) or its control solution (an

irrelevant IgG1 kappa isotype control, 1 mg/kg) 24h following the CLP- or sham-procedure. The

design of the study is depicted in figure 7.1).


167

In each experimental protocol, a first group of animals (n = 7-10/group) was used to assess GI

motility and signs of clinical disease. Following terminal anaesthesia, animals were sacrificed by

cardiac exsanguination while obtaining serum samples. Blood samples were centrifuged and

supernatants were stored until further analysis. Inflammation was also assessed locally at the

colonic level (see methods below).

In each experimental protocol, a second group of animals (n = 9-12/group) was implemented

to assess colonic permeability in septic and control animals under terminal anesthesia 48h

following the CLP- or sham-procedure. Blood and mesenteric lymph nodes were obtained from

all animals for culturing experiments (see below) to study bacterial translocation.

7.3.4 In vivo measurement of gastrointestinal transit

Mice were kept sober overnight with free access to water. A gavage was given 48h following

sham or CLP with 0.5 mL of sterile water containing 25 glass green-colored beads through a

20G flexible catheter. Mice were subsequently sacrificed 2h following the gavage and the GI

tract was resected from the distal esophageal sphincter until the anal verge, and divided into

10 parts (stomach, 5 small bowel segments, cecum, proximal colon, distal colon and feces). The

number of beads in every segment was counted under a stereomicroscope for calculation of

the percentage gastric emptying (%GE) and the geometric center of intestinal transit (GC) as a

marker for GI transit (Seerden et al., 2007).


The concentration of IL-6, TNF-α, IL-10, IL-17A, IL-1α, IL-1β and IL-12p70 in serum (pg/ml) was

determined using the BD CBA Bead Based Immunoassay on an Accuri® flow cytometer (BD

Biosciences) according to the manufacturer’s instructions. Data were processed using FCAP

Array (BD). To assess colonic cytokines at the protein level, whole colons were processed as

previously described in Chapter 3. For cytokine detection at the mRNA level, total RNA was

isolated from a snap-frozen piece of colonic tissue using the Qiagen RNeasy Mini Kit. Total RNA

was treated with DNase to obtain DNA-free RNA and converted to cDNA. Quantitative real-

time PCR was performed using the TaqMan® Universal PCR Master Mix (Life Technologies) and

primers as mentioned in Chapter 3.


168

Figure 7.1 Experimental design of the study. In a first experimental setup (A), mice received anti-interleukin-6 antibodies (1 mg/kg) or the vehicle (IgG1 kappa isotype) 5 min prior to the start of the CLP- or sham-procedure (preventive setup). In a second setup (B), mice furthermore received a repeated injection with the antibodies 24h following the procedure. In a third setup (C), mice only received anti-IL-6 or vehicle 24h following the CLP- or sham-procedure (curative setup). 48h post-CLP, mice received a gavage of colored solid beads and 2h later mice were sacrificed for measurement of gastrointestinal transit, prelevation of serum and tissue samples. A second group was subjected to the Evans blue protocol and sacrificed 1h later for measurement of colonic permeability and quantification of cell adhesion molecules at the mRNA and protein level. CBA: cytometric bead array; CDS: clinical disease score; CLP: cecal ligation and puncture; MLN: mesenteric lymph nodes; mRNA: messenger ribonucleic acid; RT-PCR: reverse transcriptase real-time polymerase chain reaction.

A

CLP or shamExperimental

setup

1 mg/kg anti-interleukin-6 orIgG1 isotype control

Clinical Disease ScoreGastrointestinal motilityCBA serumRT-PCR & CBA colon

Cultures blood and MLNColonic permeability (EB)+

+


setup




+

B


setup




+

C


169

7.3.6 Colonic permeability, tight junction molecules and bacterial

translocation

Colonic permeability was assessed by means of the Evans blue (EB) method (Lange et al., 1994)

as previously described in Chapter 3. The extracted amount of EB from the colon was

determined spectrophotometrically by measuring emission at 610 nm and expressed as µg

EB/100 mg colonic tissue. Real-time RT-PCR was performed on a piece of colon not subjected

to the formamide extraction procedure for different proteins of the tight junction (occludin,

zonulin-1, claudin-1), desmosome (desmoglein-2) and adherens junction (E-cadherin).

Finally, cultures from blood and homogenized mesenteric lymph nodes were obtained in order

to estimate the bacterial load.

7.3.7 Histology

A full thickness segment (0.5x0.5 cm) was taken from the proximal colon immediately adjacent

to the cecum. The segment was fixed for 24h in 4% formaldehyde and subsequently embedded

in paraffin. Transverse sections (5 µm) were stained with hematoxylline and eosin.

Inflammation was scored by assessing the presence and degree of inflammatory infiltrates,

presence of goblet cells, architecture of the crypts and presence of mucosal erosion as

previously published (Heylen et al., 2013), resulting in a cumulative score ranging from 0

(minimum) to 13 (maximum).

7.3.8 Statistical analysis

Data are presented as mean ± SEM, with ‘n’ representing the number of mice. Two-way

ANOVA followed by one-way ANOVA with post hoc Student-Newman-Keuls (SNK) analysis was

applied to compare the results of CDS, rectal temperature, motility parameters, cytokine levels,

PCR and other data as appropriate. A p-value ≤ 0.05 was considered to be statistically

significant. Data were analyzed using SPSS version 20.0 (IBM, Chicago) and visualized using

GraphPad Prism version 5.00.


170

7.4 Results

7.4.1 Effect of preventive administration of anti-IL-6 antibodies on septic

ileus

CLP resulted in a reproducible occurrence of sepsis without significant mortality up until 48h

post-procedure, and no difference in survival was noted between the anti-IL-6 treated and

isotype-treated septic animal group. 24h following CLP, the CDS increased substantially from

0.33±0.21 in sham-operated animals to 6.17±0.32 in septic animals (p<0.001), and the body

weight significantly decreased in septic animals) (figure 7.2A-C). The septic animals that

received anti-IL-6 antibodies demonstrated a significantly lower CDS (4.45±0.47). This

beneficial effect of anti-IL-6 was not apparent anymore 48 h following the procedure,

immediately prior to sacrifice (figure 7.2B). All sham-animals treated with anti-IL-6 antibodies

did not display abnormal behavior or signs of illness. CLP-induced sepsis resulted in an

impairment of GI transit as represented by a significant drop in the intestinal geometric center

and gastric emptying. Administration of anti-IL-6 protected mice from developing an impaired

GI transit, as demonstrated by a significant increase of the intestinal geometric center (figure

7.2E). Of note, the increase in gastric emptying rate in anti-IL-6 treated septic animals was not

significant. CLP significantly increased the colonic wall permeability to Evans blue. Prophylactic

treatment with anti-IL-6 had no effect in controls but prevented the impaired permeability in

septic mice (figure 7.2F).

Following CLP-induced sepsis, we detected a significant rise in the serum levels of

proinflammatory cytokines such as IL-6 and TNF-α, as well as a simultaneous increase in the

serum levels of the anti-inflammatory IL-10 (table 7.3A). In septic mice, administering

antibodies to IL-6 significantly decreased the concentration of serum IL-6 and IL-10. The

decrease in TNF-α concentration did not reach significance. Similar to the serum results, the

concentration of IL-6, TNF-α and IL-1α in the colonic wall rose significantly following CLP-

induced sepsis (table 7.3B). The rise in colonic TNF-α concentration was significantly reduced

by anti-IL-6. These results were confirmed at the mRNA level by RT-PCR on colonic samples

(table 7.3C).


171

Figure 7.2 Effect of preventive treatment with anti-IL-6 antibodies. Effect of preventive treatment with anti-IL-6 antibodies on sepsis-induced clinical signs of disease 24h (A) and 48h (B) following CLP or sham-surgery, percentage of weight loss at day 1 (C), percentage of gastric emptying (D), geometric center of GI transit (E) and colonic permeability as measured by the Evans blue method (F). Two-way ANOVA followed by One-way ANOVA and SNK post-hoc testing when appropriate, or its non-parametric equivalent for ordinal data; n = 7-10/group for A, B and C; n = 9-12/group for D; *p ≤ 0.05, ***p ≤ 0/001, # significant effect of CLP, § significant effect of anti-IL-6.


Sham

CLP d

ay 1

0

5

10

15IgG isotype control

Anti-IL-6 1 mg/kg

****

***

CD

S


Sham

CLP d

ay 2

0

5

10

15

# #

IgG isotype control

Anti-IL-6 1 mg/kg

CD

S

Percentage of initial body weight

Sham

CLP d

ay 1

0204060

90

95

100

# #,§IgG isotype control

Anti-IL-6 1 mg/kg

§

% o

f in

itia

l b

od

y w

eig

ht

% Gastric Emptying

Sham

CLP d

ay 2

0

50

100

IgG isotype control

Anti-IL-6 1 mg/kg# #

%G

E

Geometric Center

Sham

CLP d

ay 2

0

2

4

6

8


Anti-IL-6 1 mg/kg

#

#, §

§

GC

Evans blue extracted from colon(colonic permeability)

Sham

CLP d

ay 2

0

20

40

60


anti-IL-6 1 mg/kg

****

µg

/100m

g c

olo

n t

issu

e

A B

C D

E F


172

Table 7.3 Cytokine levels in serum (A) and supernatants of homogenized colons (B) measured by CBA or ELISA (IL-1α), and determined by RT-PCR in colon (C) during the preventive set-up (administration of anti-IL-6 antibodies simultaneously with the CLP- or sham-procedure).

sham + IgG isotype

sham + anti-IL-6

CLP d2 + IgG isotype

CLP d2 + anti-IL-6

A. Serum cytokines (pg/mL) IL-6 4.56 ± 2.51 1.96 ± 0.55§ 276.61 ± 71.64# 84.48 ± 18.30#,§

TNF-α <L/D 2.28 ± 1.42 46.55 ± 11.86# 22.13 ± 7.56#

IL-10 <L/D <L/D 12.43 ± 7.66 2.77 ± 1.45a

IL-1β 1.02 ± 0.52 1.25 ± 0.50 0.85 ± 0.44 1.65 ± 0.42 IL-17A <L/D <L/D 0.91 ± 0.26 0.89 ± 0.15 IL-1α <L/D <L/D <L/D <L/D

B. Colon cytokines (pg/100 mg colon tissue) IL-6 3.81 ± 2.38 2.18 ± 0.75 209.83 ± 94.59# 52.33 ± 14.06#

TNF-α 2.95 ± 1.94 0.45 ± 0.45 19.19 ± 9.27* 1.83 ± 1.20 IL-10 <L/D <L/D <L/D <L/D IL-1β 6.88 ± 2.27 5.10 ± 1.60 71.91 ± 54.84 8.01 ± 1.36 IL-17A <L/D <L/D <L/D <L/D IL-1α 106.74 ± 17.45 111.05 ± 7.33 334.43 ± 82.70# 202.56 ± 33.93#

C. mRNA colonic tissue IL-6 1.73 ± 0.86 1.61 ± 0.15 136.18 ± 43.85# 70.18 ± 36.12# TNF-α 1.52 ± 0.56 2.05 ± 0.60 4.91 ± 1.17# 3.22 ± 0.77# IL-10 1.88 ± 0.80 3.11 ± 1.06 10.93 ± 4.46 4.00 ± 0.82 IL-1β 1.41 ± 0.42 2.40 ± 1.00 42.53 ± 20.77# 15.21 ± 7.21#

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. # significant effect of CLP, § significant effect of anti-IL-6; for the SNK post-hoc testing: *p < 0.05 as compared to the other three groups; a p < 0.05 compared to CLP d2 + IgG isotype. When sham concentrations were below the theoretical limit of detection, the unpaired Student’s t-test was subsequently applied to compare the CLP d2 + IgG isotype with the CLP d2 + anti-IL-6 group. For the PCR-results, data are expressed as relative expression (2-ΔΔCT method) and the sham + IgG isotype group was chosen as calibrator. Data are presented as mean ± SEM. N = 8-12 animals per group. <L/D, below the limit of detection; CBA, cytometric bead array; ELISA: enzyme-linked immunosorbent assay; IL, interleukin; TNF, tumor necrosis factor.

Blood cultures from sham mice were consistently negative, whereas 78% of blood samples

from septic mice were positive. Anti-IL-6 antibodies however did not significantly reduce the

number of positive blood cultures, nor did it reduce the number of colony forming units or the

number of different pathogens that could be identified in one culture (table 7.4A). Treatment

with anti-IL-6 also had no favorable effect on the outcomes of the MLN cultures, as these were

100% positive in treated as well as untreated septic animals.


173

The mRNA-levels of different cell adhesion proteins were all numerically upregulated during

sepsis, with a significant peak in claudin-1, desmoglein-2 and E-cadherin mRNA levels, a finding

that is consistent with the vehicle-treated septic animals in our previous study (table 7.5A).

No obvious differences were observed on haematoxylin-eosin staining between control and

septic animals, nor did preventive administration of anti-IL-6 result in any histological changes

(score for all four groups: 0.0±0.0, p = NS). Therefore, the beneficial GI motility changes we

observed following anti-IL-6 treatment in the septic animals were not associated with an

amelioration of the gross GI histology (figure 7.4).

Table 7.4 Cultures of blood and homogenized mesenteric lymph nodes.

Sham + IgG isotype

sham + anti-IL-6


CLP d2 + anti-IL-6

A. Preventive setup Number of positive blood cultures (%)

0/6 (0%) 0/8 (0%) 7/9 (77.78%)* 5/9 (55.56%)*

Number of positive MLN cultures (%)

1/6 (16.67%) 2/8 (25%) 10/10 (100%)* 10/10 (100%)*

B. Preventive setup + repeated injection Number of positive blood cultures (%)

0/7 (0%) 1/6 (16.67%) 10/10 (100%)* 8/11 (72.73%)*


0/7 (0%) 0/6 (0%) 10/10 (100%)* 11/11 (100%)*

C. Curative setup Number of positive blood cultures (%)

0/5 (0%) 0/5 (0%) 6/6 (100%)* 6/6 (100%)*


0/5 (0%) 0/5 (0%) 6/6 (100%)* 6/6 (100%)*

Percentage of positive cultures from blood and homogenized mesenteric lymph nodes (24h incubation at 37°C, ambient air supplied with 5% CO2). Pearson’s chi-squared test for the percentage of positive cultures; *p ≤ 0.05 versus the sham groups.

CFU: colony forming units; CLP: caecal ligation and puncture; MLN: mesenteric lymph nodes


174

Table 7.5 RT-PCR of cell adhesion molecules.

Sham + IgG isotype

sham + anti-IL-6


CLP d2 + anti-IL-6

A. Preventive set-up occludin 1.16 ± 0.26 1.97 ± 0.46 1.75 ± 0.26 1.88 ± 0.21 zonulin-1 1.17 ± 0.24 1.63 ± 0.29 1.56 ± 0.19 1.30 ± 0.14 claudin-1 1.44 ± 0.46 1.81 ± 0.81 3.04 ± 0.86* 2.21 ± 0.37 E-cadherin 1.11 ± 0.24 1.57 ± 0.22 2.19 ± 0.27# 1.82 ± 0.15#

desmoglein-2 1.19 ± 0.30 1.73 ± 0.431 2.02 ± 0.29# 2.57 ± 0.25#,1

B. Preventive set-up + repeated injection occludin 1.40 ± 0.58 1.05 ± 0.44§ 1.82 ± 0.44 0.45 ± 0.05§

zonulin-1 1.01 ± 0.06 0.99 ± 0.202 1.35 ± 0.08 0.98 ± 0.092

claudin-1 1.10 ± 0.21 0.98 ± 0.30§ 2.03 ± 0.67 0.49 ± 0.06§

E-cadherin 1.04 ± 0.11 0.92 ± 0.073 1.45 ± 0.09# 1.22 ± 0.12#,3 desmoglein-2 1.05 ± 0.16 1.10 ± 0.33 2.70 ± 0.54* 1.06 ± 0.09

C. Curative set-up occludin 1.36 ± 0.40 2.06 ± 0.34 1.96 ± 0.17* 0.96 ± 0.13 zonulin-1 1.48 ± 0.50 3.61 ± 0.53 3.46 ± 0.30 1.90 ± 0.25a

claudin-1 1.77 ± 0.76 2.56 ± 0.72 3.07 ± 0.46* 1.04 ± 0.24 E-cadherin 1.39 ± 0.42 2.34 ± 0.31 3.99 ± 0.31* 1.78 ± 0.24 desmoglein-2 1.57 ± 0.58 3.16 ± 0.53 3.95 ± 0.29 1.92 ± 0.28a

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. # significant effect of CLP, § significant effect of anti-IL-6; *p < 0.05 compared to the other three groups; a p < 0.05 as compared to CLP d2 + IgG isotype control and sham + anti-IL-6; 1p = 0.08 for the effect of anti-IL-6; 2p = 0.06 for the effect of anti-IL-6; 3p = 0.07 for the effect of anti-IL-6. N = 8-11 animals per group.

7.4.2 Effect of preventive administration of anti-IL-6 antibodies with one

repeated injection 24h later on CLP-induced septic ileus

In a second experiment, animals received a second injection with antibodies specific to IL-6, or

the IgG control compound. 24h following CLP, the CDS was significantly higher in animals

subjected to CLP, whereas septic animals treated with anti-IL-6 antibodies displayed a

significant drop in their CDS (figure 7.3A). This effect on CDS was no longer apparent on day 2

prior to sacrifice (figure 7.3B). The percentage of weight loss was significantly increased in

septic animals; however no difference was noted between treated and untreated animals

(figure 7.3C). CLP-induced sepsis significantly impaired GI motility, as reflected by a drop in

%GE and GC; repeated treatment with anti-IL-6 antibodies significantly increased the GC, whilst

the effect on %GE was not statistically different (figure 7.3D-E). Treatment with the antibodies

normalized the colonic permeability in CLP-animals to that in sham animals, as can be deduced

by the amount of Evans blue extracted from the colon (figure 7.3F). Remarkably, a significant


175

improvement in survival was observed in the septic animals treated twice with the antibodies,

compared to the isotype-treated CLP-group (figure 7.5).

Administering a second injection with anti-IL-6 antibodies to animals that were subjected to

CLP normalized IL-6 levels in both serum and colon to that in control animals (table 7.6A,B).

Colonic IL-6 mRNA levels in septic animals however remained unaltered following treatment

(table 7.6C). TNF-α, IL-1β and IL-1α levels rose following CLP, but were not significantly altered

after targeting interleukin-6. Remarkably, TNF-α levels paradoxically rose significantly in the

colon as well as the serum following the repeated treatment with antibodies; we already

observed this feature in previous research following the reduction of IL-6 to levels observed in

control animals (Nullens S et al., Shock 2015). Blood cultures from control animals were

consistently negative, whereas 100% of cultures from blood as well as MLN from untreated

septic animals yielded positive results. Treating animals with anti-IL-6 levels however did not

significantly influence the outcome of the cultures (table 7.4).

The mRNa levels of E-cadherin and desmoglein-2 were significantly upregulated during in septic

animals, and treatment with anti-IL-6 significantly decreased mRNA concentrations of occludin

and claudin-1 (table 7.5).

7.4.3 Effect of curative administration of anti-IL-6 antibodies on septic

ileus

At day 1 and 2 following the CLP procedure, sepsis significantly increased the CDS. Curative

treatment with anti-IL-6, 48h after the CLP procedure, had a favorable effect on the CDS at the

moment of sacrifice (figure 7.6B). The body weight of septic animals was significantly

decreased in comparison to their non-septic counterparts (figure 7.6C). Sham-animals treated

with anti-IL-6 antibodies in a curative set-up did not display clinical signs of disease.

Sepsis significantly reduced gastrointestinal transit as shown by a reduced intestinal geometric

center. Curative treatment with anti-IL-6 antibodies did no longer result in a significant

amelioration of the GI transit of CLP-treated animals (figure 7.6E).

Similar to our findings in the preventive set-ups, serum and colonic proinflammatory cytokine

levels were significantly augmented following CLP in the curative set-up. Curative treatment

with anti-IL-6 24h following the CLP-procedure significantly decreased serum levels of IL-6 and

IL-10, as well as colonic levels of IL-6 and TNF-α (table 7.7A-B).


176

Figure 7.3 Effect of preventive treatment and repeat injection with anti-IL-6 antibodies. Effect of preventive treatment and repeat injection with anti-IL-6 antibodies on sepsis-induced clinical signs of disease 24h (A) and 48 h (B) following CLP or sham-surgery, percentage of weight loss at day 1 (C), percentage of gastric emptying (D), geometric center of GI transit (E) and colonic permeability as measured by the Evans blue technique (F). Two-way ANOVA followed by One-way ANOVA and SNK post-hoc testing when appropriate, or its non-parametric equivalent for ordinal data; n = 7-10/group for A, B and C; n = 9-12/group for D; *p ≤ 0.05, # significant effect of CLP, § significant effect of anti-IL-6. CDS: clinical disease score; CLP: cecal ligation and puncture; GC: geometric center; %GE: percentage of gastric emptying; GI: gastrointestinal


Sham

CLP d

ay 1

0

5

10



*

CD

S


Sham

CLP d

ay 2

0

5

10

15

# #

IgG isotype control

Anti-IL-6 1 mg/kg

CD

S


Sham

CLP d

ay 1

0204060

90

95



% o

f in

itia

l b

od

y w

eig

ht

Percentage of Gastric Emptying

Sham

CLP d

ay 2

0

20

40

60

80

100

#

IgG isotype control

Anti-IL-6 1 mg/kg

# #

%G

E

Geometric Center

Sham

CLP d

ay 2

0

2

4

6

8


Anti-IL-6 1 mg/kg# #,§§

GC

Evans blue extracted from colon(colonic permeability)

Sham

CLP d

ay 2

0

10

20

30


Anti-IL-6 1 mg/kg

* *

µg

/100m

g c

olo

n t

issu

e

A B

C D

E F


177

Table 7.6 Cytokine levels in serum (A) and supernatants of homogenized colons (B) measured by CBA, and determined by RT-PCR in colon (C) during the preventive set-up with repeated injection (administration of anti-IL-6 antibodies simultaneously with the CLP- or sham-procedure, and repeated once 24h following the procedure).

sham + 2x IgG isotype

sham + 2x anti-IL-6

CLP d2 + 2x IgG isotype

CLP d2 + 2x anti-IL-6

A. Serum cytokines (pg/mL) IL-6 5.19 ± 2.42 0.68 ± 0.29 461.46 ± 99.66* 33.90 ± 3.12 TNF-α 2.96 ± 1.18 3.48 ± 0.34 39.56 ± 7.04# 48.89 ± 6.63#

IL-10 <L/D <L/D 0.89 ± 0.89 2.03 ± 1.64 IL-1β <L/D <L/D 2.06 ± 1.90 1.25 ± 0.87 IL-17A <L/D <L/D 3.67 ± 0.83 1.10 ± 0.30 IL-1α <L/D <L/D 3.82 ± 2.40 2.75 ± 0.68

B. Colon cytokines (pg/100 mg colon tissue) IL-6 6.34 ± 1.29 3.73 ± 0.38 1049.23 ± 260.74* 69.11 ± 17.32 TNF-α 7.15 ± 2.81 7.71 ± 2.16 28.76 ± 4.89# 47.16 ± 10.88#

IL-10 <L/D <L/D <L/D <L/D IL-1β 1.75 ± 1.17 2.77 ± 1.36 28.26 ± 12.82# 36.63 ± 12.61# IL-17A <L/D <L/D 1.61 ± 0.34 1.19 ± 0.46 IL-1α 1.38 ± 0.47 1.68 ± 0.30 21.15 ± 11.73# 20.37 ± 7.50#

C. mRNA colonic tissue IL-6 1.83 ± 0.86 1.19 ± 0.28 13.26 ± 2.85# 11.82 ± 2.05# TNF-α 1.05 ± 0.13 1.12 ± 0.29 6.48 ± 1.51# 5.13 ± 0.64# IL-10 1.07 ± 0.17 0.79 ± 0.13§ 4.07 ± 0.77# 2.13 ± 0.28#,§ IL-1β 1.08 ± 0.18 0.85 ± 0.111 6.86 ± 1.24# 4.09 ± 0.56#,1 IL-1α 1.16 ± 0.27 0.78 ± 0.07 5.69 ± 1.37# 4.64 ± 0.89#

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. # significant effect of CLP, § significant effect of anti-IL-6; for the SNK post-hoc testing *p<0.05 compared to the other three groups, ap < 0.05 compared to CLP d2 + IgG isotype. 1p = 0.055 for the effect of anti-interleukin-6. When sham concentrations were below the theoretical limit of detection, the unpaired Student’s t-test was subsequently applied to compare the CLP d2 with the CLP d2 + anti-IL-6 group. For the PCR-results, data are expressed as relative expression (2-ΔΔCT method) and the sham + IgG isotype group was chosen as calibrator. Data are presented as mean ± SEM. N = 7-10 animals per group. <L/D, below the limit of detection; CBA, cytometric bead array; IL, interleukin; TNF, tumor necrosis factor.

Serum levels of other cytokines however were not significantly altered by the curative injection

with anti-IL-6 antibodies. The same was true for the colonic supernatants. Results at the colonic

level were for the greater part confirmed by RT-PCR at the mRNA level (table 7.7C). Concerning

the assessment of colonic permeability, the significant beneficial effect of the antibodies that

could be observed in the preventive protocols was lost in the curative one (figure 7.6F).


178

Figure 7.4 Hematoxylin-eosin staining in isotype and anti-IL-6 treated animals. Representative haematoxylin-eosin staining in vehicle-treated sham (A) and CLP-animals (C), and animals preventively treated with antibodies to IL-6 (sham (B) and CLP-animals (D)); 200x magnification.

A

C D

B


179

Figure 7.5 Kaplan-Meier survival analysis Kaplan-Meier curve displaying survival of sham and CLP-mice treated with anti-IL-6 antibodies or IgG isotype control up until 48h following the CLP-procedure. Log rank test p = 0.011. The curves of both sham-groups are identical to that of the anti-IL-6 treated CLP-group (no mortality).

The mRNA-levels of claudin-1, occludin and E-cadherin were significantly higher during sepsis,

and septic animals curatively treated with anti-IL-6 did not display this increase in mRNA levels

(table 7.5C). Furthermore, the mRNA level of desmoglein-2 and zonulin-1 was also significantly

lower in curatively treated animals in comparison to their non-treated septic counterparts.

Administering anti-IL-6 had no influence on the outcome of hemocultures or MLN cultures

(100% in IgG isotype control as well as anti-IL-6 treated CLP-animals) (table 7.4C).


180

Figure 7.6 Effect of curative treatment with anti-IL-6 antibodies. Effect of curative treatment with anti-IL-6 antibodies on sepsis-induced clinical signs of disease 24h (A) and 48h (B) following CLP or sham-surgery, percentage of weight loss at day 1 (C), percentage of gastric emptying (D), geometric center of GI transit (E) and colonic permeability as measured by the Evans blue method (F). Two-way ANOVA followed by One-way ANOVA and SNK post-hoc testing when appropriate, or its non-parametric equivalent for ordinal data; n = 7-10/group for A, B and C; n = 9-12/group for D; *p < 0.05, # significant effect of CLP. CDS: clinical disease score; CLP: cecal ligation and puncture; GC: geometric center; %GE: percentage of gastric emptying; GI: gastrointestinal


Sham

CLP d

ay 1

0

5

10


Anti-IL-6 1 mg/kg

# #CD

S


Sham

CLP d

ay 2

0

5

10


Anti-IL-6 1 mg/kg

*

# #

CD

S


Sham

CLP d

ay 1

0204060

90

95



% o

f in

itia

l b

od

y w

eig

ht

Percentage of Gastric Emptying

Sham

CLP d

ay 2

0

20

40

60

80



%G

E

Geometric Center

Sham

CLP d

ay 2

0

2

4

6

8


Anti-IL-6 1 mg/kg

# #

GC

EB extracted from colon (µg/100 mg colon)(colonic permeability)

Sham

CLP d

ay 2

0

5

10

15


Anti-IL-6 1 mg/kg

EB

co

lon

(µ

g/1

00 m

g c

olo

n)

A B

C D

E F


181

Table 7.7 Cytokine levels in serum (A) and supernatants of homogenized colons (B) measured by CBA, and determined by RT-PCR in colon (C) during the curative set-up (administration of anti-IL-6 antibodies 24h following the CLP- or sham-procedure).

sham + IgG isotype

sham + anti-IL-6


CLP d2 + anti-IL-6

A. Serum cytokines (pg/mL) IL-6 23.46 ± 4.46 4.14 ± 1.442 855.20 ± 462.371 54.7 ± 5.541,2

TNF-α 13.85 ± 1.12 19.48 ± 7.19 279.15 ± 133.12# 62.09 ± 10.89#

IL-10 3.36 ± 0.56 3.27 ± 0.72§ 11.74 ± 3.56# 3.90 ± 0.78#,§

IL-1β 1.35 ± 0.50 6.86 ± 5.70 9.23 ± 6.36 3.17 ± 1.11 IL-17A 2.14 ± 0.52 2.49 ± 0.35 4.71 ± 2.02# 5.88 ± 1.29#

IL-1α 1.90 ± 0.50 2.08 ± 0.61 10.99 ± 5.51# 7.46 ± 2.60#

B. Colon cytokines (pg/100 mg colon tissue) IL-6 4.27 ± 1.20 3.28 ± 0.89 432.19 ± 151.88a,b 68.04 ± 7.97c

TNF-α 6.21 ± 0.69 5.36 ± 1.433 31.79 ± 6.54# 18.38 ± 3.63#,3

IL-10 <L/D <L/D <L/D <L/D IL-1β 2.01 ± 0.66 3.99 ± 1.68 39.54 ± 20.25# 10.31 ± 2.05#

IL-17A <L/D <L/D 2.58 ± 1.09 1.17 ± 0.21 IL-1α 2.10 ± 0.19 1.70 ± 0.18 16.96 ± 6.51# 7.72 ± 2.64#

C. mRNA colonic tissue IL-6 1.36 ± 0.38 1.27 ± 0.19§ 117.79 ± 43.89# 25.50 ± 5.54#,§

TNF-α 1.28 ± 0.32 1.46 ± 0.24 7.31 ± 1.30# 5.20 ± 1.08#

IL-10 1.10 ± 0.17 1.65 ± 0.17 4.74 ± 0.51# 4.13 ± 0.50#

IL-1β 1.30 ± 0.33 1.00 ± 0.15 8.91 ± 2.06# 5.89 ± 0.99#

IL-1α 1.99 ± 0.71 1.51 ± 0.25§ 9.61 ± 2.54# 3.28 ± 0.85#,§

Two-way ANOVA followed by one-way ANOVA and SNK post-hoc testing when appropriate. # significant effect of CLP, § significant effect of anti-IL-6; for the SNK post-hoc testing: a <0.05 compared to sham + IgG isotype, b <0.05 compared to sham + anti-IL-6, c <0.05 compared to CLP d2 + IgG isotype. 1p = 0.07 for the effect of CLP, 2p = 0.06 for the effect of anti-IL-6, 3p = 0.07 for the effect of anti-IL-6. When sham concentrations were below the theoretical limit of detection, the unpaired Student’s t-test was subsequently applied to compare the CLP d2 with the CLP d2 + anti-IL-6 group. For the PCR-results, data are expressed as relative expression (2-ΔΔCT method) and the sham + vehicle group was chosen as calibrator. Data are presented as mean ± SEM. N = 10-12 animals per group. <L/D, below the limit of detection; CBA, cytometric bead array; IL, interleukin; TNF, tumor necrosis factor.


182

7.5 Discussion

In the current study, we provide evidence for the beneficial effects of preventive treatment

with antibodies against interleukin-6 on gastrointestinal alterations induced by a polymicrobial

abdominal sepsis, including motility disturbances, systemic and colonic inflammation and

colonic permeability changes. Sepsis was induced by cecal ligation and puncture, one of the

most commonly used animal models in sepsis research (Buras et al., 2005). Increased secretion

of pro- as well as anti-inflammatory cytokines was observed systemically in serum samples as

well as locally in the colon, denoting once again the simultaneous activation of pro-and anti-

inflammatory players during sepsis (Hotchkiss et al., 2013). Preventive administration of anti-IL-

6 antibodies in septic mice resulted in a significant drop of IL-6 and IL-10 levels in serum, as

well as the TNF-α concentration in the colon. Colonic IL-1α and IL-1β levels in septic animals

were also increased during sepsis at the protein and mRNA level, and others showed that IL-1

signaling in enteric glial cells was required for the development of ileus (Stoffels et al., 2013),

and that increased IL-1β levels in the gut muscularis coincided with a delay in intestinal transit

in a mouse model of postoperative ileus (Costes et al., 2014). The clinical disease score was

significantly lower 24 h after the procedure in animals that preventively received anti-IL-6

antibodies compared to vehicle-treated mice, and the impaired GI transit as well as the colonic

permeability were significantly improved. We confirmed a significant rise in claudin-1 levels in

colonic levels, a finding that was also already communicated by others and which probably

represents an attempted repair response (Eadon et al., 2012; Van Itallie et al., 2006). Claudin-1

levels normalized following treatment with anti-IL-6. It should be kept in mind that we utilized

whole organ specimens to assess mRNA levels, which could theoretically be influenced by

infiltration of leukocytes and villus blunting due to cell death. These features however were

very limited at day 2 post-CLP in our animal model.

Our results therefore show that targeting IL-6 prevented the occurrence of impaired GI transit,

local and systemic inflammation and increased permeability induced by the CLP-procedure.

Strikingly, we noticed a discrepancy between the gastric emptying rate and overall GI transit, as

the GC of the solid beads was significantly increased, whereas the %GE was not. This might

represent differential effects of interleukin-6 on gastric compared to small bowel and colonic

motility. Abundant research has been performed concerning the effect of proinflammatory

cytokines on gastric (fundus) motility. IL-1α and TNF-α induced a dose-dependent inhibition of

stomach motility, whereas IL-6 did not exhibit any effect on rat gastric fundus motility in an in


183

vitro model (Montushi et al., 1993). More recently however, Song and colleagues showed in a

rat model of burn-induced impaired gastric motility that gastric emptying was negatively

correlated with IL-6 levels and positively with vagal activity (Song et al., 2013). Injection of IL-6

and TNF-α dose-dependently resulted in anorexigenic effects in rats with a reduced gastric

emptying rate, however the mechanisms of action were not identical (McCarthy et al., 2000).

Finally, Inada et al. showed that administration of IL-6, TNF-α and IL-1β increased gastric

retention, however only TNF-α significantly lowered the GC. Of note, transit in these animals

was only studied 30 min post-cytokine injection (Inada et al., 2006).

Interestingly, these beneficial effects of anti-IL-6 were even more pronounced when the

injection with the antibodies was repeated once 24h following the CLP- or sham-procedure.

Interleukin-6 levels in serum and colon were normalized to control values, with a concomitant

significant improvement in GI transit and colonic barrier function, and a significant

improvement in survival by day 2 post-CLP. However, we were unable to detect a decrease in

the number of positive bacterial cultures from blood and MLN, believed to represent an

important clinical outcome parameter. Of note, the CLP-model in itself is a peritonitis model,

and a major source of bacteria originates not only from the lumen of the GI tract, but directly

from the peritoneum. This presumably explains the discrepancy between the normalized

colonic permeability, and consistent positive outcome of the cultures. Finally, in the human

clinical setting, the outcome of blood cultures not always correlates well with the severity of

sepsis. It should be noted that the favorable effects of antibodies to IL-6 in the preventive

setup on colon permeability could also be due to a concomitant effect of the antibodies on

colonic TNF-α levels, as these were also significantly decreased. Many have shown that TNF-α

has direct deteriorating effects on several epithelial barrier layers (Van Itallie & Anderson,

2006; Keita & Soderholm, 2010; Odenwald & Turner, 2013; Gitter et al., 2000; Han et al., 2003),

making it therefore difficult to disentangle the direct (IL-6) versus indirect (via TNF-α) effects of

anti-IL-6 on colonic permeability.

This is in sharp contrast to the set-up in which anti-IL-6 administration was repeated 24h post-

CLP, in which a reduction of colonic IL-6 levels compared to those observed in control animals

actually resulted in an increase in TNF-α levels. This might be due to the fact that IL-6 also

exerts anti-inflammatory effects on TNF-α via the release of soluble TNF-receptors (Tilg et al.,

1994), an effect which thus is now lost. Furthermore, we and others have also observed this

phenomenon in preceding research (Nullens et al., Shock 2015; Yasukawa et al., 2003;


184

Schindler et al., 1990), but also others have demonstrated an increased TNF-α serum level in

mice treated with anti-IL-6 antibodies (Matthys et al., 1995) or in IL-6 knock-out mice

(Mizuhara et al., 1994), confirming that IL-6 is involved in the regulation of TNF-α levels.

Besides, the anti-inflammatory effect of IL-6 is also reflected by the increased serum-levels of

IL-10 in isotype-treated septic animals that dropped concomitantly with IL-6 following

treatment with the antibodies.

In the animals that were treated with antibodies to IL-6 in a curative fashion, the beneficial

effects were clearly less pronounced. The effects on GI transit and colonic permeability even

failed to reach significance. Septic animals curatively treated with anti-IL-6 however still

demonstrated significantly lower expression levels of IL-6, IL-10 and IL-1α, systemically and/or

locally in the colon. Our results therefore suggest that IL-6 plays a major role in the initiation of

sepsis-related alterations in GI motility, permeability and inflammation, but loses its potency

partially once sepsis-induced alterations are full-blown. This indicates that the earlier the

treatment with anti-IL-6 starts, the more pronounced the beneficial outcome on inflammation

and GI function will be. Our findings also indicate that a specific anti-cytokine strategy may

offer therapeutic possibilities as part of the medical armamentarium in the evaluation of the

patient with a consolidated septic state. In our study design, such a proinflammatory septic

state was characterized by the increased levels of several proinflammatory cytokines in the

serum and colon.

It has been repeatedly suggested that the optimal choice of dosage and time of administration

of anti-IL-6 is pivotal in order to obtain beneficial effects during sepsis. Riedemann and

colleagues demonstrated that a dosage of 1.33 mg/kg of anti-IL-6 injected i.v. at the start of

CLP significantly increased survival, whereas a lower (0.33 mg/kg) or higher dosage (2.66

mg/kg) did not. Delayed infusion (4h post-CLP) resulted in a modest however not significant

increased survival when compared to animals treated with an irrelevant IgG (Riedemann et al.,

2003). In another murine CLP-model, therapy targeting IL-6 however failed to improve survival

whereas early administration of broad spectrum antibiotics was indeed successful (Vyas et al.,

2005). However it should be noted that the authors utilized a very severe form of CLP, which

nearly always results in 100% mortality within 72h (Rittirsch et al., 2009), and IL-6 levels were

substantially higher in the latter study when compared to the former one: the average IL-6

serum level for the untreated septic animals was >14000 pg/ml versus 7500 pg/mL


185

respectively. Combined, these results indicate that the severity of sepsis, as deduced from the

levels of IL-6, and the time of initiation of anti-cytokine strategies will play a pivotal role in the

therapeutic outcome. As human IL-6 levels are indicative of chances of survival during sepsis

(Casey et al., 1993; Giannoudis et al., 2008), a patient-centered decision on whether or not to

administer antibodies to IL-6 is mandatory based upon a broader patient’s immunological

profile. As single biomarkers such as CD14 and procalcitonin, despite being excellent diagnostic

tools, often fail to predict clinical outcome, profiling patients using a combination of these

markers, with the inclusion of IL-6, might represent an attractive approach (Masson et al.,

2014; Cohen et al., 2015).

Remarkably, the number of positive cultures from blood from septic mice did not decrease

significantly following preventive treatment with anti-IL-6, and no effect of anti-IL-6 was seen

at all on the MLN cultures. This is in contrast to our previous findings in which septic mice were

treated with the alpha7 nicotinic acetylcholine receptor agonist GTS-21. By binding onto the

alpha7 nicotinic acetylcholine receptor, GTS-21 prevents macrophages from secreting

proinflammatory cytokines such as IL-6 and we showed that GTS-21 treatment normalized

colonic permeability and significantly reduced the number of positive blood and MLN cultures

from septic mice and normalized colonic and serum IL-6 levels (Nullens et al., Shock 2015). GTS-

21 however can also bind onto the alpha4beta2 receptor on peritoneal macrophages, and

stimulation of this specific receptor will induce the phagocytosis of bacteria (van der Zanden et

al., 2009). These findings together with our current results suggest that the mere targeting of

IL-6, despite resulting in a normalized intestinal permeability, probably does not suffice to halt

bacterial translocation, especially since we were able to completely reduce serum and colonic

IL-6 levels to those measured in control animals. The role of IL-6 in permeability has been

established extensively in in vitro experiments, where it was shown that IL-6 increased

endothelial permeability (Maruo et al., 1992) and it decreased membrane integrity in an

enterocyte cell culture model (Tazuke et al., 2003). Moreover, IL-6 increased claudin-2

expression, which actually decreases the intestinal tight junction barrier (Suzuki et al., 2011).

Some limitations of this study and considerations must be taken into account: a non-lethal CLP-

model was chosen, and thus we refer the interested reader to several other manuscripts that

tackle the clinical endpoint of survival per se (Riedemann et al., 2003; Vyas et al., 2005; Pallua

et al, 2003.). Furthermore, IL-6 treated animals were not completely protected from systemic


186

spread of invading pathogens, as cultures from blood and mesenteric lymph nodes still yielded

positive results following treatment with anti-IL-6.

Anti-cytokine strategies therefore offer us interesting therapeutic possibilities, given the fact

that they are administered in an appropriate dosage, at the suitable moment including a

maintenance regimen if necessary, and in a carefully selected patient along with other

conventional therapies.

In the human setting, there are currently no ongoing clinical trials targeting one or more

cytokines in the septic patient, with the exception of IFN-γ administration in patients suffering

from candidiasis sepsis (clinicaltrials.gov identifier 2009-014600-66 and 2012-002491-14) and

sepsis-induced immune paralysis. However, another trial is ongoing studying the effect of

adsorbing cytokines by means of the ‘CytoSorb’ device in the early stages of septic shock (<48h

from the diagnosis onward) in Hungary (clinicaltrials.gov identifier NCT02288975 and

NCT00559130), or with the ‘Oxiris’ hemofilter that adsorbs endotoxins and cytokines from the

blood during continuous renal replacement therapy (NCT 02600312).

In summary, our data confirm a time-dependent beneficial effect of blocking interleukin-6 via

the administration of specific blocking antibodies in a mouse model of polymicrobial abdominal

sepsis. Beneficial effects were observed upon gastrointestinal motility, inflammation and

permeability, with effects being predominantly pronounced when the antibodies were

administered immediately prior to the induction of sepsis instead of during the course of the

sepsis. This demonstrates that IL-6 is an important initiator not only of inflammation, but it also

affects GI motility and permeability. Whether IL-6 is the main direct acting mediator, and/or

whether it exerts its effects directly or indirectly remains the subject of further research.

Chapter 8

General discussion

Chapter 8 General discussion

188

The pivotal role of the gastrointestinal tract (GI tract) during sepsis is increasingly being

acknowledged (Hassoun, 2001; Deitch, 2011). In short, it has been demonstrated that the

detrimental changes that occur at the GI tract level during sepsis (ranging from injury caused by

ischemia-reperfusion generated molecules, the release of proinflammatory cytokines by

immune cells, and the release of mediators that damage the intricate composition of the GI

wall, among many other instigating processes) will ultimately result in unfavorable changes in

the digestive tract. These will subsequently impair the hallmark functions of the GI tract, such

as the maintenance of a selective barrier against noxious antigens and bacteria, and the

propagating motility function of the digestive tract as a whole. Furthermore, the direct action

of bacterial compounds as well as the indirect deleterious effect of mediators released by

immune cells will impair the barrier function of the mucosa, resulting in disturbed permeability

and further risk of (bacterial) translocation (Zhou et al., 2013; Keita et al., 2010). These noxious

alterations have recently been summarized in what is called the ‘gut-lymph’ or ‘gut-lung’

hypothesis, in which gut-derived mediators (often sterile) during sepsis are able to induce

detrimental effects in distant organs (Deitch et al., 2006; Deitch, 2012; Levy et al., 2013). Of

note, the microbiome could certainly play a determining role in the pathophysiology of these

concepts.

Despite major therapeutic advances that are being made on a near daily basis in medicine, the

development of therapeutic strategies for the septic patients is currently mainly restricted to

supportive therapies in order to sustain vital functions (Dellinger et al., 2013), whereas the

number of therapies that specifically target the eliciting pathogenesis is limited. The translation

of positive results obtained in (murine) preclinical experiments towards main-stream intensive

care medicine is not straightforward, as promising molecules with specific targets often yielded

conflicting results in the human setting (Rittirsch et al., 2007; Dyson & Singer 2009; Angus et

al., 2013; Pene et al., 2015). Many have hypothesized that targeting the impaired GI tract

functions could interrupt the vicious cycle of impaired mucosal barrier function and

subsequent increased risk of translocation of bacteria and inflammatory mediators.


189

Therefore, the aim of this dissertation was to address the following issues:

i. To implement a valid and reproducible animal model of sepsis in our lab, that

displays sufficient similarities with the pathophysiology of the human sepsis

(Chapter 4).

ii. To characterize the neuroimmune players that constitute an important role in the

detrimental effects that occur at the gastrointestinal tract level in a mouse model

of sepsis-induced ileus, at different time-points following the septic insult

(Chapter 4, 5).

iii. To modulate the neuronal players, identified in the second work package, that

play a pivotal role in the pathogenesis of sepsis-induced ileus (Chapter 6).

iv. To modulate the inflammatory immune environment during sepsis-induced ileus

(Chapter 7).

The overall conclusions that can be drawn from this dissertation are as follows:

i. The cecal ligation and puncture (CLP) model of sepsis mimics the human setting of

a perforated appendicitis or diverticulitis, resulting in a polymicrobial, abdominal

peritonitis and subsequent sepsis. Animals subjected to CLP consistently displayed

the development of an abdominal sepsis that approximates the human one. The

occurrence of impaired motility confirms the presence of ileus, and inflammatory

markers are upregulated in the serum as well as at the colonic level.

ii. The immune changes observed at the GI tract level display striking similarities

with those observed by others and, most notably, resemble those that have been

observed in the human setting. This renders the CLP-model (at least partially)

suitable to further study neuroimmune changes at the GI tract level during sepsis.

Besides, the CLP-model could be of merit by the identification of possible

therapeutic targets to tackle sepsis and sepsis-induced GI motility alterations.

iii. The neuronal changes observed at the GI tract level during a prolonged sepsis,

assessed by recording jejunal afferent nerve activity, mRNA analysis of dorsal root

ganglia and PET/CT imaging of the central nervous system, reveal differential

effects on the different relay stations of the nervous system in time and location,

and are not completely in accordance with what is known about the effect of LPS

on afferent and dorsal root ganglion neuronal activity.


190

iv. The modulation of the release of proinflammatory cytokines from immune cells

such as macrophages by targeting the alpha7 nicotinic acetylcholine receptor, the

final endpoint in the so-called vagal anti-inflammatory pathway, yielded several

beneficial effects with clinical endpoints, such as an amelioration in sepsis-

induced GI motility disturbances and mucosal barrier function, and a reduction in

the number of positive hemocultures.

v. The targeting of a single proinflammatory cytokine yielded time- and dose-

dependent favorable effects on inflammation and motility. A reversal of sepsis-

induced changes, however, could not be obtained, suggesting that a single-target-

strategy is probably not the best way to go in sepsis.

For the remainder of this chapter, we will focus on each of these observations separately, and

try to integrate them into what is known in the current clinical setting. We will end by

discussing the potential clinical implications of our research results, and by formulating future

research goals and possible applications of our results in the human setting.

8.1 Characterization of the different neuroimmune players in the digestive tract

involved in the pathogenesis of sepsis-induced ileus in order to identify new

therapeutic targets

Implementation of an animal model of sepsis in our lab

The debate on whether the animal models that are currently in use are adequate or

representative enough to study human sepsis, or to identify possible new therapeutic targets,

is already ongoing for a long time (Rittirsch et al., 2007; Dyson & Singer, 2009; Angus et al.,

2013; Pene et al., 2015). Animal research remains, however, a prerequisite in order to develop

new treatment strategies and sufficient preclinical research needs to be accomplished,

including animal research on different species, in order to obtain permission to perform human

clinical studies in countries that pertain to the European Union or United States.

Others have shown that the CLP-model adequately mimics the cardiovascular, hemodynamic

and metabolic changes that also occur in the human septic patient suffering from shock that

has received adequate fluid resuscitation (Rittirsch et al., 2009).

Until now, the changes at the GI tract level however have not been fully elucidated. Much

research has been performed by the group of Bauer, showing that impaired GI motility as well

as increased inflammation was present in the gut following CLP-induced sepsis. More


191

importantly, they were the first to study the time-dependent effects of CLP-induced sepsis in

the gut, as they performed functional motility as well as molecular experiments during a 4-day

period following CLP (Overhaus et al., 2004).

We opted to study our animals either 48h or 7 days following the septic insult based upon

preliminary data on weight loss and individual signs of clinical disease obtained during the

implementation of the CLP-model in our lab, which included survival analysis. We hypothesized

that these time points could potentially represent the ‘stereotypic’ proinflammatory and anti-

inflammatory immune phase that is often described during human sepsis. Prior research

predominantly focused on the initial pathophysiological changes observed during the earliest

hours following the induction of sepsis. However patients often present themselves to the ICU-

department with a consolidated septic state, and different patients exhibit different immune

phenotypes, based upon their gene profile, comorbidities, immune suppression and

medication use. We therefore opted to further characterize these two different and more

clinically relevant time points.

Based upon the cytokine analysis in the blood and colon, we can fairly state that a profound

proinflammatory phase was observed 48h following the CLP-procedure. A significant increase

in serum and colonic IL-6 and TNF-α levels was detected at protein as well as mRNA level. A

monocytosis and lymphopenia was observed in the blood, which is typically observed in the

septic mouse. Furthermore we did measure a profound decreased thrombocyte count, also

commonly observed in murine sepsis, which is associated with impaired host-defense and

indicative of a worse outcome (de Stoppelaar et al., 2014). Finally, neutropenia was observed

in the blood, reflecting previously described migration of the neutrophils from the blood

towards the locus of inflammation.

Flow cytometric characterization of the cecal ligation and puncture model of sepsis

In accordance with previous observations made by others, we detected an influx of neutrophils

in the colon following the CLP-procedure (Overhaus et al., 2004). Of utmost importance is the

fact that this research group, as many others have done up until now, studied the muscularis

layer of the colon, as their main interest lied with the impairment of colonic motility. We

however focused on the lamina propria as a hallmark ‘antigen-presenting tissue’ in the context

of sepsis. Using a combination of immunohistochemistry and flow cytometry, we were able to

characterize the relative presence of different leukocyte populations in the different tissues


192

that roughly characterize the gastrointestinal tract. Similar to observations made by others, we

could not detect an obvious rise in the relative number of monocytes in the colon in the initial

septic phase (24-48h following CLP). We did, however, observe a significant increase by day 7

during the consolidated septic phase. Performing flow cytometry of various organs pertaining

to the GI tract, whilst simultaneously assessing the total and differential white blood cell count

and differential, allowed us to speculate on how different immune cell populations act during

sepsis. Neutrophils have been clearly shown to migrate from the blood stream towards the GI

tract, including the spleen, during sepsis, resulting in a profound neutropenia by day 7

following CLP. A concomitant drop in lymphocytes in the blood as well as in the organs studied

could be indicative of overall lymphocyte apoptosis, a feature that is commonly described in

human as well as murine sepsis (Hotchkiss et al., 2000; Unsinger et al., 2000; Zou et al., 2013).

Concerning the number of T cells, no increase was noted 48h post-CLP in the relative

percentage of overall T cells or more specifically of T helper cells in the lamina propria of ileum

or colon using flow cytometry. Immunohistochemistry however revealed a significant increase

in CD3+ T cells in the mucosal layer during sepsis, predominantly located between the epithelial

cells. Both techniques therefore should be used in conjunction with one another, while

performing flow cytometric evaluation of the separated epithelium represents another

technical possibility. In sham-operated animals, we confirmed for the greater part data on

different types of gut lamina propria DCs obtained and published by many researchers in the

field (Bekiaris et al., 2014; Varol et al., 2009; Bogunovic et al., 2009), with 21% of DCs being

CD103+CD11b- and 57% being CD103-CD11b+ in the ileum lamina propria (Bekiaris et al.: ±20%

versus ±25% respectively). In the colon lamina propria we detected 16% CD103+CD11b- and

59% CD103-CD11b+ dendritic cells (Bekiaris et al.: ±40% for both subsets). In addition, we

performed the same analysis following CLP-induced sepsis, and found a doubling 48h post-CLP

in the contribution of DCs to the overall leukocyte population in the ileum with identical

subpopulations, whereas colonic numbers remained identical. At day 7 however, the picture

changes dramatically, with a significant increase in the overall number of dendritic cells with a

concomitant explicit drop in the contribution of CD103+ cells in both organs. This specific

subtype of DCs might thus very well represent an interesting target during prolonged sepsis

with characteristics of immunosuppression, but whether they protect against or dispose for

mortality remains to be elucidated. The same holds true for several other cells we

characterized, such as mast cells, of which we observed an astounding rise in numbers at day 2


193

and day 7 in the colonic as well as ileal lamina propria. We can thus fairly state that flow

cytometry represents an interesting technique, in addition to more conventional validated lab

techniques such as immunohistochemistry staining, to assess different immune cells and

potentially their activation status. Some technical limitations and critiques have to be taken

into account. As we had access to a flow cytometer with ‘only’ four optical filters, we could

only quantify the overall percentage of immune cells in relation to the overall leukocyte

population. Future research should certainly assess the absolute number of the different

subsets, as well as their activation status by characterizing cell surface activation markers. In

addition, one could also assess nuclear transcription factors, for example T-bet, GATA-3 and

RORγt, or even the intracellular translation of different cytokines, to further pinpoint the

different immune subtypes (in this case Th1, Th2, and Th17 lymphocyte). It should be noted

that we initially also started off with the intracellular staining in order to assess the different T

helper subsets; different stimulation protocols (e.g. use of the T cell stimulation cocktail

Phorbol 12-Myristate 13-Acetate/ionomycin for 4 to 24h followed by incubation with the

protein transport inhibitor brefeldin) yielded unfortunately no obvious rise in the intracellular

concentrations of interleukins in our initial experiments, necessitating further optimization of

the intracellular staining protocol in our lab. Besides, one must be cautious to interpret results

obtained during an ‘artificial’ in vitro stimulation of isolated polarized T cells, as it does not

necessarily reflect the in vivo human setting during sustained sepsis. Nevertheless, these sorts

of experiments certainly could compliment the data we obtained during the course of this

doctoral thesis.

Gross histological analysis by means of hematoxylin/eosin staining revealed no abundant

alterations in colonic tissue after 48h from animals subjected to the CLP-procedure (Chapter 4).

Overall, subtle villus blunting can be seen, as well as confined infiltration of leukocytes in all

layers. As was already described in Chapter 1, the mouse ileus is of an intense nature but is

primarily due to local cytokine-responses instead of profound influx of leukocytes (Schmidt et

al., 2012), and no notable infiltrate was measured in the small bowel of mice sacrificed 24h

following CLP in a study performed by Maier and colleagues (Maier et al., 2004).


194

Afferent mesenteric nerve recordings and activation status of the dorsal root ganglia and

central nervous system in the CLP-model

Our current knowledge on the effect of sepsis on afferent nerve activity is nowadays solely

derived from studies using pure LPS administered via the intravenous or intraperitoneal route

to the rodent lab animal, with main focus on its immediate effects. However, the long term

effects as well as the alterations induced by a more clinically relevant polymicrobial sepsis were

unknown to date. Many have already elegantly studied the immediate effects of whole or lysed

bacteria, as well as several proinflammatory cytokines, on (mesenteric) afferent nerve activity

or activity in dorsal root ganglia projecting to the colon, and demonstrated primarily a

hypersensitizing effect. Bacteria by themselves can directly activate sensory neurons that

modulate pain and inflammation, and it has been shown that ablating these nerves actually

increased local immune infiltration in a murine model of S. aureus-induced pain, thus indicating

that these nerves counteract inflammation via a feedback mechanism and host-pathogen

interactions (Chiu et al., 2013).

Hypersensitization of afferent nerves thus has been implicated in many pathological conditions,

including irritable bowel syndrome and several other chronic pain conditions (De Winter et al.,

2016; Anand et al., 2007; Buhner et al., 2014; De Winter et al., 2009). We characterized for the

first time, to the best of our knowledge, afferent nerve activity and activation status of the

dorsal root ganglia and central nervous system in an animal model of prolonged abdominal

sepsis. In contrast to the direct hypersensitizing effects of LPS on peripheral afferent nerve

activity, we measured an overt drop in mesenteric nerve activity during ramp distensions,

which could mainly be attributed to impaired high threshold and wide dynamic range fibers

responses. Direct exposure of ‘healthy afferent nerves’ to a septic soup composed of

homogenized GI tissue from septic animals also decreased afferent nerve activity, but this time

decreased activity from low threshold and high threshold fibers was responsible for this

observation, pointing towards a time-dependent effect of CLP-induced inflammation and sepsis

on afferent nerve activity. Recently several research groups are also starting to report

significantly reduced responses on afferent nerves, submucous ganglia or DRGs upon

prolonged incubation with several mediators involved in the pathogenesis of pain. Most

notably, the group of M. Schemann showed that overnight incubation with an ‘IBS cocktail’,

consisting out of histamine, serotonin, tryptase and TNF-α, actually decreased responses of


195

submucous ganglia from IBS patients obtained by biopsy in comparison to healthy control

subjects (Ostertag et al., 2015). They hypothesized that this could be due to the

densensitization to mediators constantly released by mucosal and immune cells in the gut wall

of IBS patients. Most notably, of all the aforementioned four mediators, it has been

demonstrated that direct application actually had an excitatory effect on human enteric

ganglia. The same group has elegantly shown that increased spike discharge of human

submucosal neurons correlated with the levels of serotonin, histamine and tryptase levels in

IBS supernatants that were used to evoke discharge from these neurons (Buhner et al., 2009).

Furthermore, the activation by these supernatants could be inhibited by histamine receptor

antagonists, a 5-HT3 receptor antagonist and protease inhibitors. The most notable difference

is the fact that in the initial paper the authors studied neurons obtained from healthy control

subjects incubated with ‘IBS supernatants’, whereas in their most recent publication by

Ostertag they directly studied neurons obtained from IBS patients. Finally, TNF-α could

increase the mechanosensitivity of DRG neurons (Ozaktay et al., 2006) and colonic afferent

neurons (Hughes et al., 2013; Sommer & Kress, 2004).

In this matter, many have proposed that opioid-release from immunocytes could play a pivotal

role in inhibition of viscerosensory afferents (Hughes et al., 2014). Opioid receptors have been

demonstrated to be present not only on DRGs, but also on peripheral sensory neurons, as has

been nicely summarized by Stein and colleagues (Stein, 2003). These receptors, synthesized in

DRGs and then carried into central and peripheral terminals via axonal transport, are primarily

responsible for the peripheral analgesic effects of locally administered opioid agonists (Zhou et

al., 1998). During inflammation close to the afferent nerve, a downregulation of δ-receptor in

DRGs can be observed, while the level of μ-opioid receptor mRNA remained unaltered (Zhang

et al., 1998; Schäfer et al., 1995). These observations are in accordance with the mRNA levels

we observed in our septic animals. Finally, during inflammation the efficacy of opioids to

decrease the excitability of afferent neurons increases, resulting in enhanced efficacy of opioid

agonists (Antonijevic et al., 1995).

The group of Vanner has demonstrated that the rheobase of neurons increased significantly

following incubation with supernatants from mice suffering from DSS-colitis, and that these

effects were blocked by the opioid-receptor antagonist naloxone (Valdez-Morales et al., 2013).

Moreover they showed increased β-endorphin, tissue-opioid immunoreactivity and CD4+

effector T cells in colons from colitis animals.


196

The release of opioids from T-helper cells is also well established in the context of ‘human’

inflammatory bowel disease (Basso et al., 2014; Basso et al., 2015). Innate immune cells as well

as T cells have the capability to produce endogenous opioids, and it has been demonstrated

that the intensity of visceral pain is inversely correlated with the infiltration of T cells in the

inflamed colonic mucosa, leading to the relieve of pain after a couple of days, as opioids

decrease the excitability of peripheral colonic sensory neurons (Owczarek et al., 2011; Valdez-

Morales et al., 2013; Verma-Gandhu et al., 2007).

It should be noted that, during flow cytometry, we could not observe a numerical difference in

the relative percentage of CD4+ helper T cells in the lamina propria of the colon or ileum from

septic animals at day 2 following the CLP-procedure (Chapter 4); by means of

immunohistochemistry we were able to demonstrate a significant increase in the CD3+ T cell

subset in the overall mucosal layer. Further research is thus warranted whether these mucosal

(epithelial?) T cells are responsible for the effects observed at the peripheral afferent level, and

whether these are indeed mediated via the release of endogenous opioids.

Based on the aforementioned results and what is known in the literature, we therefore

hypothesize that a possible explanatory mechanism concerns the prolonged inflammation at

the colonic level that will lead to the increased release of endogenous opioids by recruited

immune cells in the gut, resulting in decreased afferent mesenteric nerve activity measured in

vitro in our animal model of prolonged polymicrobial sepsis. In order to support this

hypothesis, further research should be conducted studying the effect of (selective) opioid

antagonists on mesenteric activity recordings in the CLP-model, as well as measurements of

opioid levels in colonic tissue from CLP-treated animals. This reasoning might also explain our

finding that incubation of healthy jejunal afferents with supernatants from homogenized septic

colons might immediately decrease afferent nerve activity during ramp distensions, whereas a

predominant effect of proinflammatory interleukins, such as IL-6, TNF-α or IL-1β, would

increase afferent nerve activity (Hughes et al., 2013). Furthermore it would explain the fact

that this decrease was not observed following incubation of control jejunal segments with

supernatants from healthy colons. Of specific interest is the observation that different subunits

of afferent nerves react differently to the presence of ‘sepsis’, as in septic mice the reduction in

afferent discharge was predominantly due to a decrease in high threshold and wide dynamic

range, whereas incubation of healthy jejunum with septic supernatants instead decreased high


197

and low threshold fibers, indicating a time-dependent influence of sepsis on the different

mechanistic subunits in mesenteric afferent fibers.

Several other receptors or channels expressed on peripheral neurons or DRGs have been

shown to play a role in sepsis, afferent nerve hypersensitization and sepsis-induced motility

disturbances. The presence of the NMDA-receptor at the DRG level is mandatory in the

development and maintenance of human visceral pain, and the administration of an NMDA-

receptor antagonist increased the pain threshold in a human trial studying visceral

hypersensitivity following an infusion with hydrochloric acid in the distal esophagus (Willert et

al., 2004). Moreover, the NMDA receptor was shown to be upregulated on colonic afferent

fibers during (neuronal) inflammation in the mouse using an analogue of capsaicin (Suckow &

Caudle, 2009), and in a mouse model of LPS-induced renal insufficiency, blocking NMDA

ameliorated renal failure (Lin et al., 2015). Of note, it has even been demonstrated that sex-

dependent differences in DRGs concerning NMDA-receptor activity are (at least partially)

responsible for the fact that women are more prone to develop chronic visceral pain than men

(McRoberts et al., 2007). We detected an increase of NMDA mRNA-levels at the DRG level in

animals subjected to CLP, pointing towards inflammatory nociceptor sensitization.

CGRP-receptors are involved in the occurrence of sepsis-induced motility disturbances, as

administering a CGRP-receptor antagonist reversed the endotoxin-induced GI motility

impairment in a mouse model of LPS-induced sepsis (De Winter et al., 2009). We however

could not detect an increase in the CGRP mRNA-levels in the DRG. Most notably we did observe

a significant rise in the neuronal activation marker cFos, and in the TRPA1 receptor. TRPA1 is

involved in the hypersensitivity caused by inflammation (Holzer et al., 2011). The TRP channels

are presumed to orchestrate all peripheral afferent input, and they modulate visceral

hypersensitivity by other mediators such as histamine, serotonin and protease activating

receptors (De Winter et al., 2016). As such, TRPA1 has been shown to be involved in the

hypersensitivity caused by inflammation (Holzer et al., 2011), and antagonizing TRPA1 reduced

the inflammatory response in a mouse model of ventilator-associated pneumonia (Wang et al.,

2013). TRPV1 receptors expressed on afferent neurons play a substantial role in the

pathogenesis of gastrointestinal motility disturbances, as antagonists of TRPV1 were able to

counteract the motility disorders in both colitis- as well as endotoxin-induced ileus (De

Schepper et al., 2008; De Winter et al., 2009), appointing an important role to the TRP channels


198

in the alimentary tract (Holzer, 2011). In the current setting we failed however to notice any

difference concerning the TRPV1 mRNA level in DRGs. In summary, we can state that several

receptors involved in nociception or induced during a state of inflammation are upregulated at

the DRG level, pointing towards a ‘hypersensitized’ state of the DRG during CLP-induced sepsis,

whereas in contrast the peripheral afferent nerve activity was decreased. The role of the three

main classes of mediators in sensation and wiring of peripheral information, namely the

opioids, vanilloids and cannabinoids, definitely deserves further study.

It has been gracefully demonstrated by the group of Mayer that not only the hyper- or

desensitization of peripheral afferent nerves plays a substantial role in the experience of pain

or the transmission of sensory information, but that the central nervous should also be taken

into account. Mayer et al demonstrated that significant differences could be measured by

means of cerebral blood flow monitoring and positron emission tomography in central brain

regions, including the amygdala, anterior cingulated cortex, cortices and periaqueductal gray,

regions involved in pain sensation between IBS and ulcerative colitis patients (Mayer et al.,

2005). Based on this study, we designed an experimental setup in which we studied the

activation of different mouse brain regions between septic and control mice by means of [18F]-

FDG PET/CT imaging at the Molecular Imaging Center Antwerp (MICA). An overall profound

decrease in metabolic brain activity was recorded, most pronounced in the amygdala, cortex

and striatum, regions involved in the perception and modulation of visceral sensory

information on pain and inflammation. This could indicate that exposure to a prolonged

polymicrobial milieu decreases overall metabolic brain activity due to concurrent

neuroinflammation (Semmler et al., 2008), or that a negative protective mechanism is present

as was already touched upon earlier (Mayer et al., 2005). Due to the technical limitations of

this technique in mice, several regions of interest such as the periaqueductal gray which

primarily controls descending pain inhibition could unfortunately not be specifically visualized;

so this technique and experimental set-up require further optimization.

8.2 Neuronal modulation of the neuroimmune environment

The discovery of the endogenously present cholinergic anti-inflammatory reflex pathway

marked the start of an abundance of research on several pathologies that are somewhat

related to inflammation, most notably on postoperative ileus and LPS-induced sepsis. The exact


199

anatomical substrate of this cholinergic anti-inflammatory reflex however remains somewhat

of a conundrum, and several investigations have yielded conflicting results on whether or not

the spleen is directly involved in the reflex pathway in the context of postoperative ileus

(Matteoli et al., 2014; Martelli et al., 2014; Cailotto et al., 2014; Costes et al., 2014; Costes et

al., 2014; Bratton et al., Exp Physiol 2012; Pavlov & Tracey, 2015; Bonaz et al., 2016).

The finding that GTS-21, a selective alpha7 nicotinic acetylcholine receptor agonist, dose-

dependently improved survival in the LPS- as well as the CLP-model of sepsis confirms the

therapeutic potential of directly targeting the α7nAChR instead of the more cumbersome

stimulation of the vagal nerve (Pavlov et al., 2007). Furthermore, it remains to be fully

elucidated whether direct VNS might not actually deteriorate the clinical state, as increased

vagal tone could in theory alter the already impaired cardiac function by decreasing heart rate

and blood pressure. It has been shown that decreased vagal tone following vagotomy resulted

in cardiac impairment in a murine model of septic shock (Schulte et al., 2014), and the

administration of sympathicolytic agents such as clonidine or dexmedetomidine to animals

subjected to CLP improved survival and down-regulated proinflammatory mediators (Hofer et

al., 2009). Dexmedetomidine furthermore has been shown to exert its actions via peripheral

vagal- and α7nAChR-dependent mechanisms, as its effects were abolished when animals were

subjected to cervical vagotomy prior to the administration of dexmedetomidine (Xiang et al.,

2014). Research in animal models thus has repetitively shown that VNS is feasible in the human

subject, and several human clinical trials are currently ongoing (see also Chapter 1). It should

be noted however that a recently published preliminary results by Kox and colleagues yielded

no favorable effects on serum levels of proinflammatory cytokines (Kox et al., 2015); the results

of further trials are thus eagerly anticipated.

In our study, the preventive administration of GTS-21, followed by a repeat injection 24h and

48h following the CLP- or sham-procedure, significantly decreased the serum and colonic level

of interleukin-6, normalized impaired gastrointestinal transit, normalized mRNA levels of

colonic cell-cell adhesion molecules, and significantly decreased the number of positive

cultures from blood and mesenteric lymph nodes (Nullens et al., Shock 2015). Our results thus

confirm that the activation of the vagal nerve, e.g. by electrical stimulation, is not a

prerequisite for effective downscaling of inflammation. Furthermore, data demonstrated that

targeting intestinal (more specifically, colonic) inflammation results in positive effects on a


200

relevant clinical endpoint. Our study further elucidated that GTS-21, often postulated to be a

specific α7nAChR agonist, also exhibited α7nAChR-independent effects, as the compound

showed beneficial results on motility and inflammation in α7nAChR-KO mice. We hypothesized

this might be due to its binding to the α4β2 receptor expressed on the cell surface of

macrophages, as activation of this receptor increased their phagocytotic activity (van der

Zanden, 2009). This dual effect of GTS-21 certainly contributed to the significant decrease in

positive cultures from blood and mesenteric lymph nodes, as we were unable to reproduce this

finding in subsequent studies in which only systemic and local colonic inflammation was

downregulated by single anti-cytokine treatment. Attention must be paid to the fact that in this

experimental setup GTS-21 was administered in a preventive manner prior to the induction of

sepsis, which makes the results somewhat difficult to translate to the human setting. Finally,

the role of the spleen merits further investigation, as it is indispensable in the onset of sepsis,

yet also appears to be a target organ of the compound we studied.

8.3 Immunological modulation of the neuroimmune environment

Previous research studying the effect of targeting one specific proinflammatory molecule-of-

interest in murine sepsis has been hampered by the difficulties in translating results obtained

in laboratory animals to the human setting. As such, promising compounds that targeted TNF-α

(Fisher et al., 1996), the IL-1-receptor (Opal et al., 1997), LPS (Angus et al., 2000) or the TLR4-

receptor (Opal et al., 2013) failed to show beneficial results in the first human clinical trials. The

same was true for the administration of high dosages of methylprednisolone, which

theoretically could dampen inflammation (Bone et al., 1987).

Studies focusing on the effect of blocking IL-6 in animal models of sepsis however often are

confined to the study of survival and serum-levels of IL-6. The outcomes have been mixed in

terms of survival, and a bell-shaped effect on survival seems to be present based upon data

published by various research groups, yet the effects on the gastrointestinal tract during sepsis

remain to be fully elucidated (Riedemann et al., 2003; Barkhausen et al., 2011). We opted to

focus on the role of IL-6 on the GI tract during sepsis-induced ileus, as the deleterious effects of

this cytokine on mucosal permeability are well known (Yang et al., 2003). Moreover we

detected a significant rise in colonic IL-6 levels during the immunological characterization study

that coincided with an increased number of positive cultures, whereas a reduction in IL-6 was

observed together with the cultures becoming negative (Nullens et al., 2015).


201

Administration of one injection with antibodies specific to IL-6 did induce some positive effects

in our septic animals. We could, however, not obtain a full normalization of IL-6 levels, and no

difference was noted in the number of positive cultures from blood or MLN. We then repeated

the injection once 24h following the induction of sepsis. This resulted in a normalization of

interleukin-6 levels with normalization of the geometric center of transit and colonic

permeability to the Evans blue dye, but cultures did not become negative. In contrast to data

previously reported by Mees et al, we observed a significant benefit on survival in the animals

that were treated repetitively with anti-IL-6 antibodies (Mees et al., 2009), a major relevant

clinical endpoint. Targeting IL-6 is not irrelevant, as these results have shown in our study, and

they certainly deserve more studies.

In the more clinically relevant, curative treatment of septic mice with anti-IL-6 antibodies (24h

following CLP), some beneficial effects of the antibodies were observed on serum cytokines.

The modest effects on GI motility, colonic permeability and colonic inflammation did not reach

statistical significance, indicating that a curative single-cytokine strategy is probably not the

way to go during sepsis. It should however be noted that the IL-6 profile during CLP in the

serum usually reaches its peak levels during the initial hours following CLP (Rivers et al., 2013),

and that curative administration 24h later might be too late to observe ny beneficial effects on

sepsis-induced deleterious changes.

8.4 Clinical implications of the research presented in this thesis

The gut is obviously involved in the pathogenesis of sepsis, whether one allocates it an

instigating or a mere bystander role. We demonstrated that several pathological phenomena

can be observed at the GI tract level at day 2 following the induction of sepsis, ranging from

increased markers of inflammation to frank impairment of the colonic barrier function,

mesenteric afferent nerve activity and occurrence of widespread gastrointestinal ileus. One

week following the septic insult, we observed concomitant alterations inmarkers of

inflammation (increased colonic levels of IL-17A and IFN-γ) as well as anti-inflammatory

markers such as IL-10 levels and a drop in lymphocyte numbers, presumably due to apoptosis.

Next, we showed that targeting the neuronal and/or immunological players represents an

interesting target in reversing these pathological manifestations during sepsis, as GI motility

and colonic permeability to Evans blue could be normalized. The overall effects of the

compounds we tested in the cecal ligation and puncture model of sepsis are summarized in


202

figure 9.1. Based on the results obtained in the aforementioned studies however, we estimate

that a single-target strategy will probably not suffice in order to ‘cure’ the septic patient.

Targeting only e.g. elevated IL-6 levels did not adequately restore all sepsis-induced

disturbances, especially when it is only administered during the later stages of a consolidated

sepsis. It thus seems prudent to only target interleukin-6 (or for that matter, any other cytokine

or single marker) in patients that display increased IL-6 levels in an early stage of sepsis.

Another important role is reserved for the neuronal players during sepsis, and more specifically

for the vagal anti-inflammatory pathway with the alpha7 nicotinic acetylcholine receptor being

its end target, as modulation of the afferent as well as efferent branches by others (Costantini

et al., 2010) showed promising effects. Finally, in this dissertation we only briefly touched upon

the pivotal role of the barrier function of the gastrointestinal tract. We do believe that also the

GI barrier function is of major importance in the further prevention and treatment of sepsis

and sepsis-induced motility impairments.

8.5 Future challenges and perspectives

Whilst the CLP-model is by many considered to be the gold standard animal model in sepsis

research, it is often difficult to elucidate which effects at the GI tract level are due to sepsis as

such and which effects are due to the local GI effect itself, since the inflammatory insult is

directed explicitly against that same GI tract. Were motility and permeability changes in the GI

tract that we observed due to systemic inflammation, or the simple fact that the inflammatory

focus lied adjacent to this organ tract? Ideally, the effects we observed on GI motility,

inflammation, afferent nerve activity and intestinal permeability ought to be validated in

several other animal models of sepsis and/or inflammation that do not directly target the GI

tract. The entangling of all players involved and their reciprocal effects certainly merits several

more dissertations.


203

Figure 9.1 The different physiological and homeostatic functions we studied in this dissertation and the effects of the compounds studied on these functions.

A major role in future research should be allocated to the study of the GI barrier and its

functions during sepsis. In this dissertation, we specifically focused on motility and

inflammatory changes at the GI tract level, taking the GI barrier less into account. During the

course of our experiments, we came to the understanding that the translocation of whole

bacteria, bacterial compounds and (inflammatory) mediators probably plays a pivotal role.

Compounds directly targeting the barrier and its properties also represent an interesting

therapeutic strategy. In this matter, we are currently studying the effect of a direct mucosal

barrier strategy on sepsis-induced inflammation and motility disturbances.

Inflammation –gastrointestinaltract

Inflammation –systemically

serum

Motilitydisturbances

Neuronal alterations

Permeability disturbances

CURE

GTS-21Anti-IL-6 preventiveAnti-IL-6 curative





204

Furthermore, several immune cells and their mediators involved in the pathogenesis of

increased mucosal permeability during sepsis and other pathologies also deserve more

interest. An important cell in this regard is the mast cell that secretes histamine, tryptase and

other proteases, as it has been demonstrated that these have detrimental effects on the

gastrointestinal barrier and selective permeability.

The impaired mesenteric nerve activity we measured during a prolonged state of inflammation,

has recently also been demonstrated by others following prolonged exposure of submucosal

colonic neurons to an ‘inflammatory soup’, consisting of histamine, tryptase and TNF-α

(Ostertag et al., 2015), and further study is warranted in order to fully elucidate this

mechanism. Our data showing hyposensitivity of the afferent nerves and pain-related brain

centers questions the dogma that LPS can only cause hypersensitization, and it certainly

warrants further detailed research on unraveling why exactly results at first sight seem so

different from one another. In this regard, the further disentangling of the brain-gut axis and

the brain-gut-microbiome axis to further extension during sepsis certainly represents another

important topic of study. Finally, in what way is the microbiome the trigger to increase

gastrointestinal permeability, induce inflammation at the local level and systemically in spleen

and serum, thus ultimately leading to the activation of the vagal anti-inflammatory pathway

and neuro-immunological control of (inflammatory) gastrointestinal processes? Much research

questions are guaranteed in the near future.

Chapter 9

Summary

Chapter 9 Summary

206

The pivotal role of the gastrointestinal tract during sepsis is widely acknowledged and

represents both a target as well as a source of several inflammatory mediators that can

contribute to the further clinical deterioration of the septic patient. As such, several hallmark

functions of the gut are potentially impaired during sepsis, such as the propulsive motility, the

maintenance of a selective barrier against noxious antigens and bacteria, and the uptake of

indispensable nutrients from the lumen of the gastrointestinal tract. Many of these noxious

alterations have been summarized in what is called the ‘gut-lymph’ or ‘gut-lung’ hypothesis, in

which gut-derived mediators (often sterile) during sepsis are able to induce detrimental effects

in other organ tracts at a distance. However, the development of therapeutic strategies for

septic patients is currently mainly restricted to supportive therapies in order to sustain vital

functions, whereas the number of therapies that target the eliciting pathogenesis is limited.

The development and implementation of new compounds in the clinic is hampered by the fact

that beneficial results obtained in (murine) preclinical experiments are difficult to translate to

the human setting, as animal models often do not adequately reflect the average human septic

patient. We hypothesize that targeting the impaired GI tract functions could interrupt the

vicious cycle of impaired mucosal barrier function and subsequent increased risk of

translocation of bacteria and inflammatory mediators.

We therefore aimed to address the aforementioned needs by i) developing and implementing

a valid and reproducible animal model of sepsis, that closely mimics several pathophysiological

properties of ‘human’ sepsis, ii) subsequently characterizing the different neuroimmune

players that are involved in the development of sepsis and sepsis-induced gastrointestinal

motility disturbances and iii) by attempting to modulate these different neuroimmune players

with different compounds that might be of future potential interest in the treatment of septic

patients.

We implemented the cecal ligation and puncture (CLP) model, the gold standard animal

model of sepsis in our lab. We assessed the effect of modulating the severity in this model by

studying survival, signs of clinical disease, gastrointestinal motility, colonic permeability and

inflammation in the serum as well as at the colonic level. We showed that, by performing a

‘mild’ CLP-procedure, we could obtain an animal model with reproducible effects on sickness

behavior, a significant delay in gastrointestinal motility, a significant impairment of the colonic

barrier function and an upregulation of several pro- as well as anti-inflammatory mediators in

the blood as well as the colonic wall. Besides, with this model we were able to induce a

Chapter 9 Summary

207

biphasic disease pattern in our animal, of which we hoped that these phases are representative

of the human systemic inflammatory response syndrome (day 2 post-CLP), and the

compensatory anti-inflammatory response syndrome (day 7 post-CLP).

Next, we aimed to characterize the different neuroimmune players in the development and

maintenance of gastrointestinal tract that were involved during sepsis-induced ileus at both

aforementioned time points. We started by studying the different leukocyte populations in the

spleen, mesenteric lymph nodes, ileal and colonic lamina propria by means of flow cytometry,

and performed additional immunohistochemistry stains of the colonic wall. We demonstrated

that at both time points, severely impaired GI motility could be observed as well as features

from both a pro- as well as anti-inflammatory immune profile. Next, we assessed the activity of

afferent mesenteric nerves, as these represent the first relay station in transmitting signals on

inflammation and disturbed motility and permeability from the gut to the brain. We

demonstrated that exposure to a prolonged polymicrobial sepsis desensitized jejunal afferent

nerves, as they displayed less firing activity during ramp distensions, which is in contrast with

the hypersensitizing effects of pure lipopolysaccharide. Next, we assessed the in vitro

activation status of dorsal root ganglia (DRG) by means of RT-PCR and the in vivo metabolic

activity of the central nervous system by means of PET/CT. DRGs displayed upregulation of

several neuronal markers of inflammation, such as cFos and NMDA, and brain imaging

demonstrated overall decreased metabolic activity after CLP. This reduced metabolic uptake

was most pronounced in the amygdala, cortex and striatum, regions involved in the processing

of visceral information and pain.

The impaired gastrointestinal functions can be targeted in several ways. First, we administered

GTS-21, a compound that blocks the release of proinflammatory cytokines, such as interleukin-

6, from macrophages in the spleen and the gastrointestinal tract. GTS-21 is presumed to be a

selective agonist of the alpha7 nicotinic acetylcholine receptor, a receptor that represents the

final endpoint in the so-called cholinergic anti-inflammatory pathway. Activation of this

endogenously present reflex pathway results in the dampening of inflammation. As such, we

were able to demonstrate that GTS-21 prevented the occurrence of ileus, ameliorated the

increased mucosal permeability, decreased proinflammatory cytokine levels in the serum as

well as in the colon, and significantly decreased the number of positive cultures from blood and

mesenteric lymph nodes.

Chapter 9 Summary

208

In a following experiment, we administered antibodies specific for interleukin-6 to septic

animals in either a preventive (with or without repeat injection) or curative protocol. We were

able to show that both routes of administration had benefits on inflammation, however only by

administering a preventive as well as one repeat injection we could fully prevent the

occurrence of ileus and mortality, but not that of bacterial translocation.

In summary, the CLP-model provides us with insights on the several neuronal and

immunological players that are altered in the GI tract during sepsis. Since many features are

similar to that of human septic patient, the CLP-model should be taken into account when

performing preclinical drug testing with new therapeutic compounds. We can furthermore

conclude by stating that the aforementioned compounds studied in this dissertation are an

interesting target in the battle against sepsis-induced ileus, provided that they are given in a

tailored fashion, in a timely manner, in the correct dosage and to the suitable patient.

Chapter 10

Samenvatting

Chapter 10 Samenvatting

210

Aan de maagdarmtractus wordt door veel specialisten een onmiskenbare rol toegeschreven in

de pathogenese van sepsis, aangezien dit orgaanstelsel zowel een bron als een doelorgaan is

van verschillende mediatoren die de klinische uitkomst van de septische patiënt kunnen

beïnvloeden. Verschillende functies van de gastro-intestinale tractus aangetast zullen zijn

gedurende sepsis, waaronder de propulsieve motiliteit, het onderhouden van een selectieve

barrière tegen schadelijke antigenen en bacteriën, alsmede de opname van onmisbare

voedingsstoffen uit het lumen van de maagdarmtractus. Vele van deze pathologische

veranderingen werden recent ondergebracht in de zogenaamde ‘darm-lymfe’ hypothese of

‘darm-long’ hypothese, in dewelke mediatoren (vaak steriele) afkomstig uit de darm tijdens

sepsis de functie van andere orgaansystemen op afstand kunnen aantasten. Ondanks grote

medische vooruitgang, blijft de behandeling van de septische patiënt vaak beperkt tot louter

supportieve therapie teneinde de orgaanfuncties te ondersteunen, zonder daarbij de

onderliggende mechanismen te behandelen. De ontwikkeling en implementatie van nieuwe

strategieën wordt gehinderd door de moeilijke translatie van gunstige effecten verkregen in

diermodellen naar de kliniek, daar deze diermodellen vaak niet de immunofenotypische

kenmerken vertonen van de septische patiënt. We stellen hierbij dan ook de hypothese vooraf

dat het behandelen van de aangetaste functies van de maagdarmtractus, die een vicieuze cirkel

voeroorzaken door een verstoorde mucosale barrièrefunctie met verhoogd risico op

translocatie van luminele bacteriën en inflammatoire mediatoren, de verdere deterioratie van

de patiënt kan tegengaan.

We poogden daarom de bovenvermelde tekortkomingen aan te pakken door i) een valabel en

reproduceerbaar diermodel van sepsis-geïnduceerde ileus te implementeren in ons labo, dat

verscheidene immunologische en pathologische kenmerken vertoont van de ‘humane’ sepsis,

ii) vervolgens in dit diermodel de verschillende neuro-immune spelers te identificeren die

betrokken zijn in de ontwikkeling van sepsis en sepsis-geïnduceerde motiliteitsproblemen, om

vervolgens iii) deze verscheidene neuro-immune spelers te moduleren aan de hand van

molecules die in de toekomst potentieel aangewend kunnen worden voor de behandeling van

sepsis-patiënten.

We kozen ervoor het cecale ligatie en punctie (CLP) model te implementeren in ons labo, een

model dat door velen beschouwd wordt als het gouden standaard diermodel inzake sepsis-

onderzoek. We bestudeerden het effect van het moduleren van de procedure qua ernst door

de klinische tekens, de overleving, de gastro-intestinale motiliteit, de permeabiliteit van het


211

colon alsmede de inflammatie in het serum en ter hoogte van het colon te bestuderen. We

konden aantonen dat door middel van een ‘milde’ CLP-procedure, we een reproduceerbaar

diermodel verkregen met een beperkte mortaliteit, een significante vertraging van de motiliteit

van de maagdarmtractus, een verstoorde barrièrefunctie van het colon en de inductie van

verscheidene pro- en anti-inflammatoire cytokines, waaronder interleukine-6. Bovendien

verkregen we met dit diermodel een bifasisch verloop van de ziekte, waarvan we

veronderstellen dat deze fasen corresponderen met het humane systemische inflammatoire

respons syndroom (dag 2 na CLP) en het compensatoire anti-inflammatoire respons syndroom

(dag 7 na CLP).

Vervolgens poogden we de verschillende neuro-immune spelers in de maagdarmtractus te

karakteriseren gedurende sepsis op beide voorvermelde tijdspunten. We analyseerden de

verschillende witte bloedcel subtypes in de milt, de mesenterische lymfeklieren en de lamina

propria van ileum en colon aan de hand van flowcytometrie, en voerden bijkomende

immuunhistochemische kleuringen uit op colonsegmenten. Vervolgens karakteriseerden we de

activiteit van de mesenterische afferente neuronen, aangezien deze het eerste station

vertegenwoordigen in het communiceren van signalen met betrekking tot inflammatie en pijn

van de periferie naar het centraal zenuwstelsel. We toonden aan dat de activiteit van jejunale

afferenten sterk onderdrukt was na blootstelling aan een polymicrobiële abdominale sepsis,

wat in scherp contrast is met de directe sensitiserende effecten van puur lipopolysachharide.

Vervolgens bestudeerden we de activatiestatus van de dorsale wortelganglia aan de hand van

RT-PCR, en de in vivo metabole activiteit van het centrale zenuwstelsel aan de hand van

PET/CT. Dorsale wortelganglia vertoonden een verhoogde expressie van verschillende

neuronale merkers van inflammatie, zoals cFos en NMDA. De beeldvorming van de hersenen

toonde een globaal sterk verminderde metabole activiteit; deze was het meest uitgesproken

ter hoogte van de amygdala, cortex en striatum, regio’s betrokken in de perceptie van viscerale

inflammatie en pijn.

De aangetaste gastro-intestinale functies kunnen op verschillende manieren aangepakt

worden. In een eerste experiment bestudeerden we het effect van GTS-21, een molecule die

de vrijstelling van pro-inflammatoire cytokines, waaronder interleukine-6, uit de macrofaag

inhibeert. GTS-21 is een agonist van de alfa7 nicotinerge acetylcholine receptor, een receptor

die het eindpunt is van de zogenaamde cholinerge anti-inflammatoire reflex. Activatie van deze

endogeen aanwezige reflex resulteert in de onderdrukking van inflammatie. We konden


212

aantonen dat GTS-21 het optreden van ileus verhinderde, de toegenomen mucosale

permeabiliteit van het colon verbeterde, de pro-inflammatoire cytokines in het serum en het

colon deed dalen, en tenslotte significant het aantal positieve culturen van blood en

mesenterische lymfeknopen verminderde.

In een daarop volgend experiment bestudeerden we het effect van de toediening van

specifieke antilichamen tegen interleukine-6 hetzij na preventieve toediening (al dan niet met

herhaling), hetzij na een curatieve toediening. We konden aldus aantonen dat we in beide

protocols gunstige effecten verkregen op het gebied van inflammatie. Enkel aan de hand van

preventieve toediening van de antilichamen (met één herhalingsinjectie) konden we het

optreden van ileus echter volledig tegengaan, alsmede de overleving van onze proefdieren

verbeteren. Er werd evenwel geen gunstig effect gemeten op bacteriële translocatie.

Samengevat kunnen we stellen dat het CLP-model inzicht biedt in de alteraties van de

verschillende neuronale en immunologische spelers in de maagdarmtractus gedurende sepsis.

We kunnen bovendien concluderen dat de voorvermelde molecules zeker een interessant

therapeutisch potentieel hebben in de kliniek, op voorwaarde dat ze ‘op maat’ toegediend

worden, in een correct tijdsvenster, in de correcte hoeveelheid en aan de geschikte patiënt.

References

References

214

Abbas A, Lichtman A, Pallai S. Cellular and Molecular Immunology. 19 ed. 2013.

Abraham E, Laterre PF, Garg R, Levy H, Talwar D, Trzaskoma BL, et al. Drotrecogin alfa

(activated) for adults with severe sepsis and a low risk of death. N Engl J Med 2005 Sep

29;353(13):1332-41.

Alfonsi P, Slim K, Chauvin M, Mariani P, Faucheron JL, Fletcher D. French guidelines for

enhanced recovery after elective colorectal surgery. J Visc Surg 2014 Feb;151(1):65-79.

Al-Hashimi M, Scott SW, Thompson JP, Lambert DG. Opioids and immune modulation: more

questions than answers. Br J Anaesth 2013 Jul;111(1):80-8.

Allan AL, Keeney M. Circulating tumor cell analysis: technical and statistical considerations for

application to the clinic. J Oncol 2010;2010:426218. doi: 10.1155/2010/426218. Epub;%2009

Dec 13.:426218.

Alpaerts K, Buckinx R, Cools N, Heylen M, Nullens S, Berneman Z, et al. Effect of schistosomiasis

on CX3CR1-expressing mononuclear phagocytes in the ileum and mesenteric lymph nodes of

the mouse. Neurogastroenterol Motil 2015 Nov;27(11):1587-99.

Amancha PK, Hong JJ, Rogers K, Ansari AA, Villinger F. In vivo blockade of the programmed cell

death-1 pathway using soluble recombinant PD-1-Fc enhances CD4+ and CD8+ T cell responses

but has limited clinical benefit. J Immunol 2013 Dec 15;191(12):6060-70.

Anand P, Aziz Q, Willert R, Van Oudenhove L. Peripheral and central mechanisms of visceral

sensitization in man. Neurogastroenterol Motil 2007 Jan;19(1 Suppl):29-46.

Andersson U, Tracey KJ. Neural reflexes in inflammation and immunity. J Exp Med 2012 Jun

4;209(6):1057-68.

Angus DC, Birmingham MC, Balk RA, Scannon PJ, Collins D, Kruse JA, et al. E5 murine

monoclonal antiendotoxin antibody in gram-negative sepsis: a randomized controlled trial. E5

Study Investigators. JAMA 2000 Apr 5;283(13):1723-30.

Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of

severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care.

Crit Care Med 2001 Jul;29(7):1303-10.

Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013 Aug 29;369(9):840-

51.

Antonijevic I, Mousa SA, Schafer M, Stein C. Perineurial defect and peripheral opioid analgesia

in inflammation. J Neurosci 1995 Jan;15(1 Pt 1):165-72.

Arendash GW, Sengstock GJ, Sanberg PR, Kem WR. Improved learning and memory in aged rats

with chronic administration of the nicotinic receptor agonist GTS-21. Brain Res 1995

Mar;%20;674(2):252-9.

Artis D, Spits H. The biology of innate lymphoid cells. Nature 2015 Jan 15;517(7534):293-301.

References

215

Babu D, Motterlini R, Lefebvre RA. CO and CO-releasing molecules (CO-RMs) in acute gastrointestinal inflammation. Br J Pharmacol 2015 Mar;172(6):1557-73. Bansal V, Costantini T, Ryu SY, Peterson C, Loomis W, Putnam J, et al. Stimulating the central

nervous system to prevent intestinal dysfunction after traumatic brain injury. J Trauma 2010

May;68(5):1059-64.

Barkhausen T, Tschernig T, Rosenstiel P, van Griensven M, Vonberg RP, Dorsch M, et al.

Selective blockade of interleukin-6 trans-signaling improves survival in a murine polymicrobial

sepsis model. Crit Care Med 2011 Jun;39(6):1407-13.

Basso L, Boue J, Bourreille A, Dietrich G. Endogenous regulation of inflammatory pain by T-cell-

derived opioids: when friend turns to foe. Inflamm Bowel Dis 2014 Oct;20(10):1870-7.

Basso L, Bourreille A, Dietrich G. Intestinal inflammation and pain management. Curr Opin

Pharmacol 2015 Dec;25:50-5. doi: 10.1016/j.coph.2015.11.004. Epub;%2015 Nov 26.:50-5.

Bauer AJ, Boeckxstaens GE. Mechanisms of postoperative ileus. Neurogastroenterol Motil 2004

Oct;16 Suppl 2:54-60.:54-60.

Bauer AJ, Schwarz NT, Moore BA, Turler A, Kalff JC. Ileus in critical illness: mechanisms and

management. Curr Opin Crit Care 2002 Apr;8(2):152-7.

Bekiaris V, Persson EK, Agace WW. Intestinal dendritic cells in the regulation of mucosal

immunity. Immunol Rev 2014 Jul;260(1):86-101.

Berthoud HR, Neuhuber WL. Functional and chemical anatomy of the afferent vagal system.

Auton Neurosci 2000 Dec;%20;85(1-3):1-17.

Biswas SK, Lopez-Collazo E. Endotoxin tolerance: new mechanisms, molecules and clinical

significance. Trends Immunol 2009 Oct;30(10):475-87.

Blackshaw LA, Brookes SJ, Grundy D, Schemann M. Sensory transmission in the gastrointestinal

tract. Neurogastroenterol Motil 2007 Jan;19(1 Suppl):1-19.

Blackshaw LA, Gebhart GF. The pharmacology of gastrointestinal nociceptive pathways. Curr

Opin Pharmacol 2002 Dec;2(6):642-9.

Boeckxstaens GE, de Jonge WJ. Neuroimmune mechanisms in postoperative ileus. Gut 2009

Sep;58(9):1300-11.

Boeckxstaens GE, Hirsch DP, Kodde A, Moojen TM, Blackshaw A, Tytgat GN, et al. Activation of

an adrenergic and vagally-mediated NANC pathway in surgery-induced fundic relaxation in the

rat. Neurogastroenterol Motil 1999 Dec;11(6):467-74.

Bogunovic M, Dave SH, Tilstra JS, Chang DT, Harpaz N, Xiong H, et al. Enteroendocrine cells

express functional Toll-like receptors. Am J Physiol Gastrointest Liver Physiol 2007

Jun;292(6):G1770-G1783.

References

216

Bogunovic M, Ginhoux F, Helft J, Shang L, Hashimoto D, Greter M, et al. Origin of the lamina

propria dendritic cell network. Immunity 2009 Sep 18;31(3):513-25.

Bollaert PE, Charpentier C, Levy B, Debouverie M, Audibert G, Larcan A. Reversal of late septic

shock with supraphysiologic doses of hydrocortisone. Crit Care Med 1998 Apr;26(4):645-50.

Bonaz B, Sinniger V, Pellissier S. Vagal tone: effects on sensitivity, motility, and inflammation.

Neurogastroenterol Motil 2016 Apr;28(4):455-62.

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 Jun;101(6):1644-55.

Bone RC, Grodzin CJ, Balk RA. Sepsis: a new hypothesis for pathogenesis of the disease process.

Chest 1997 Jul;112(1):235-43.

Bone RC. Immunologic dissonance: a continuing evolution in our understanding of the systemic

inflammatory response syndrome (SIRS) and the multiple organ dysfunction syndrome (MODS).

Ann Intern Med 1996 Oct 15;125(8):680-7.

Bone RC. Why sepsis trials fail. JAMA 1996 Aug 21;276(7):565-6.

Booth CE, Shaw J, Hicks GA, Kirkup AJ, Winchester W, Grundy D. Influence of the pattern of

jejunal distension on mesenteric afferent sensitivity in the anaesthetized rat.

Neurogastroenterol Motil 2008 Feb;20(2):149-58.

Borovikova LV, Ivanova S, Zhang M, Yang H, Botchkina GI, Watkins LR, et al. Vagus nerve

stimulation attenuates the systemic inflammatory response to endotoxin. Nature 2000 May

25;405(6785):458-62.

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 Aug;88(2):233-40.

Brandtzaeg P. The gut as communicator between environment and host: immunological

consequences. Eur J Pharmacol 2011 Sep;668 Suppl 1:S16-32. doi:

10.1016/j.ejphar.2011.07.006. Epub;%2011 Jul 28.:S16-S32.

Bratton BO, Martelli D, McKinley MJ, Trevaks D, Anderson CR, McAllen RM. Neural regulation

of inflammation: no neural connection from the vagus to splenic sympathetic neurons. Exp

Physiol 2012 Nov;97(11):1180-5.

Brenn D, Richter F, Schaible HG. Sensitization of unmyelinated sensory fibers of the joint nerve

to mechanical stimuli by interleukin-6 in the rat: an inflammatory mechanism of joint pain.

Arthritis Rheum 2007 Jan;56(1):351-9.

References

217

Briegel J, Forst H, Haller M, Schelling G, Kilger E, Kuprat G, et al. Stress doses of hydrocortisone

reverse hyperdynamic septic shock: a prospective, randomized, double-blind, single-center

study. Crit Care Med 1999 Apr;27(4):723-32.

Brierley SM, Jones RC, III, Gebhart GF, Blackshaw LA. Splanchnic and pelvic mechanosensory

afferents signal different qualities of colonic stimuli in mice. Gastroenterology 2004

Jul;127(1):166-78.

Briggs CA, Anderson DJ, Brioni JD, Buccafusco JJ, Buckley MJ, Campbell JE, et al. Functional

characterization of the novel neuronal nicotinic acetylcholine receptor ligand GTS-21 in vitro

and in vivo. Pharmacol Biochem Behav 1997 May;57(1-2):231-41.

Brookes SJ, Spencer NJ, Costa M, Zagorodnyuk VP. Extrinsic primary afferent signalling in the

gut. Nat Rev Gastroenterol Hepatol 2013 May;10(5):286-96.

Buchholz BM, Chanthaphavong RS, Billiar TR, Bauer AJ. Role of interleukin-6 in hemopoietic and

non-hemopoietic synergy mediating TLR4-triggered late murine ileus and endotoxic shock.

Neurogastroenterol Motil 2012 Jul;24(7):658-69, e294.

Bueno L, Fioramonti J, Ruckebusch Y. Postoperative intestinal motility in dogs and sheep. Am J

Dig Dis 1978 Aug;23(8):682-9.

Bueno L, Fioramonti J. Protease-activated receptor 2 and gut permeability: a review.

Neurogastroenterol Motil 2008 Jun;20(6):580-7.

Buhner S, Braak B, Li Q, Kugler EM, Klooker T, Wouters M, et al. Neuronal activation by mucosal

biopsy supernatants from irritable bowel syndrome patients is linked to visceral sensitivity. Exp

Physiol 2014 Oct;99(10):1299-311.

Buhner S, Li Q, Berger T, Vignali S, Barbara G, de Giorgio R, et al. Submucous rather than

myenteric neurons are activated by mucosal biopsy supernatants from irritable bowel

syndrome patients. Neurogastroenterol Motil 2012 Dec;24(12):1134-e572.

Buhner S, Li Q, Vignali S, Barbara G, de Giorgio R, Stanghellini V, et al. Activation of human

enteric neurons by supernatants of colonic biopsy specimens from patients with irritable bowel

syndrome. Gastroenterology 2009 Oct;137(4):1425-34.

Buras JA, Holzmann B, Sitkovsky M. Animal models of sepsis: setting the stage. Nat Rev Drug

Discov 2005 Oct;4(10):854-65.

Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, et al. The MIQE guidelines:

minimum information for publication of quantitative real-time PCR experiments. Clin Chem

2009 Apr;55(4):611-22.

Cahill CM, Taylor AM, Cook C, Ong E, Moron JA, Evans CJ. Does the kappa opioid receptor

system contribute to pain aversion? Front Pharmacol 2014 Nov 17;5:253. doi:

10.3389/fphar.2014.00253. eCollection;%2014.:253.

References

218

Cailotto C, Gomez-Pinilla PJ, Costes LM, van der Vliet J, Di Giovangiulio M, Nemethova A, et al.

Neuro-anatomical evidence indicating indirect modulation of macrophages by vagal efferents

in the intestine but not in the spleen. PLoS One 2014 Jan 29;9(1):e87785.

Camilleri M, Madsen K, Spiller R, Greenwood-Van MB, Verne GN. Intestinal barrier function in

health and gastrointestinal disease. Neurogastroenterol Motil 2012 Jun;24(6):503-12.

Camilleri M, Malagelada JR, Brown ML, Becker G, Zinsmeister AR. Relation between antral

motility and gastric emptying of solids and liquids in humans. Am J Physiol 1985 Nov;249(5 Pt

1):G580-G585.

Camilleri M, Nadeau A, Lamsam J, Nord SL, Ryks M, Burton D, et al. Understanding

measurements of intestinal permeability in healthy humans with urine lactulose and mannitol

excretion. Neurogastroenterol Motil 2010 Jan;22(1):e15-e26.

Carroll J, Alavi K. Pathogenesis and management of postoperative ileus. Clin Colon Rectal Surg

2009 Feb;22(1):47-50.

Casey LC, Balk RA, Bone RC. Plasma cytokine and endotoxin levels correlate with survival in

patients with the sepsis syndrome. Ann Intern Med 1993 Oct 15;119(8):771-8.

Celes MR, Torres-Duenas D, ves-Filho JC, Duarte DB, Cunha FQ, Rossi MA. Reduction of gap and

adherens junction proteins and intercalated disc structural remodeling in the hearts of mice

submitted to severe cecal ligation and puncture sepsis. Crit Care Med 2007 Sep;35(9):2176-85.

Cenac N, Chin AC, Garcia-Villar R, Salvador-Cartier C, Ferrier L, Vergnolle N, et al. PAR2 activation alters colonic paracellular permeability in mice via IFN-gamma-dependent and -independent pathways. J Physiol 2004 Aug 1;558(Pt 3):913-25. Cerra FB. The systemic septic response: multiple systems organ failure. Crit Care Clin 1985

Nov;1(3):591-607.

Chan JK, Roth J, Oppenheim JJ, Tracey KJ, Vogl T, Feldmann M, et al. Alarmins: awaiting a

clinical response. J Clin Invest 2012 Aug;122(8):2711-9.

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 Jan 4;18(1):R3.

Chang KC, Burnham CA, Compton SM, Rasche DP, Mazuski R, Smcdonough J, et al. Blockade

ofthe negative co-stimulatory molecules PD-1 and CTLA-4 improves survival in primary and

secondary fungal sepsis. Crit Care 2013 May 11;17(3):R85.

Chatterjee PK, Al-Abed Y, Sherry B, Metz CN. Cholinergic agonists regulate JAK2/STAT3

signaling to suppress endothelial cell activation. Am J Physiol Cell Physiol 2009

Nov;297(5):C1294-C1306.

Chiu IM, Heesters BA, Ghasemlou N, Von Hehn CA, Zhao F, Tran J, et al. Bacteria activate

sensory neurons that modulate pain and inflammation. Nature 2013 Sep 5;501(7465):52-7.

References

219

Clowes GH, Jr., Zuschneid W, Turner M, Blackburn G, Rubin J, Toala P, et al. Observations on

the pathogenesis of the pneumonitis associated with severe infections in other parts of the

body. Ann Surg 1968 May;167(5):630-50.

Coelho-Aguiar JM, Bon-Frauches AC, Gomes AL, Verissimo CP, Aguiar DP, Matias D, et al. The

enteric glia: identity and functions. Glia 2015 Jun;63(6):921-35.

Cohen J, Vincent JL, Adhikari NK, Machado FR, Angus DC, Calandra T, et al. Sepsis: a roadmap

for future research. Lancet Infect Dis 2015 May;15(5):581-614.

Colpaert EE, Timmermans JP, Lefebvre RA. Immunohistochemical localization of the antioxidant

enzymes biliverdin reductase and heme oxygenase-2 in human and pig gastric fundus. Free

Radic Biol Med 2002 Apr 1;32(7):630-7.

Costantini TW, Bansal V, Krzyzaniak M, Putnam JG, Peterson CY, Loomis WH, et al. Vagal nerve

stimulation protects against burn-induced intestinal injury through activation of enteric glia

cells. Am J Physiol Gastrointest Liver Physiol 2010 Dec;299(6):G1308-G1318.

Costantini TW, Bansal V, Peterson CY, Loomis WH, Putnam JG, Rankin F, et al. Efferent vagal

nerve stimulation attenuates gut barrier injury after burn: modulation of intestinal occludin

expression. J Trauma 2010 Jun;68(6):1349-54.

Costes LM, van der Vliet J, Farro G, Matteoli G, Van Bree SH, Olivier BJ, et al. The spleen

responds to intestinal manipulation but does not participate in the inflammatory response in a

mouse model of postoperative ileus. PLoS One 2014 Jul 10;9(7):e102211.

Costes LM, van der Vliet J, Van Bree SH, Boeckxstaens GE, Cailotto C. Endogenous vagal

activation dampens intestinal inflammation independently of splenic innervation in

postoperative ileus. Auton Neurosci 2014 Oct;185:76-82. doi: 10.1016/j.autneu.2014.07.006.

Epub;%2014 Jul 23.:76-82.

Cosyns SM, Shiva S, Lefebvre RA. Protective effect of exogenous nitrite in postoperative ileus.

Br J Pharmacol 2015 Oct;172(20):4864-74.

Dann SM, Spehlmann ME, Hammond DC, Iimura M, Hase K, Choi LJ, et al. IL-6-dependent

mucosal protection prevents establishment of a microbial niche for attaching/effacing lesion-

forming enteric bacterial pathogens. J Immunol 2008 May 15;180(10):6816-26.

De Backer O, Lefebvre RA. Investigation of a possible interaction between the heme

oxygenase/biliverdin reductase and nitric oxide synthase pathway in murine gastric fundus and

jejunum. Eur J Pharmacol 2008 Aug;%20;590(1-3):369-76.

De Backer O, Elinck E, Blanckaert B, Leybaert L, Motterlini R, Lefebvre RA. Water-soluble CO-

releasing molecules reduce the development of postoperative ileus via modulation of

MAPK/HO-1 signalling and reduction of oxidative stress. Gut 2009 Mar;58(3):347-56.

References

220

De Filippis D, Esposito G, Cirillo C, Cipriano M, De Winter BY, Scuderi C, et al. Cannabidiol

reduces intestinal inflammation through the control of neuroimmune axis. PLoS One

2011;6(12):e28159.

De Filippis D, Iuvone T, d'amico A, Esposito G, Steardo L, Herman AG, et al. Effect of cannabidiol

on sepsis-induced motility disturbances in mice: involvement of CB receptors and fatty acid

amide hydrolase. Neurogastroenterol Motil 2008 Aug;20(8):919-27.

De Herdt V, Bogaert S, Bracke KR, Raedt R, De Vos M, Vonck K, et al. Effects of vagus nerve

stimulation on pro- and anti-inflammatory cytokine induction in patients with refractory

epilepsy. J Neuroimmunol 2009 Sep 29;214(1-2):104-8.

de Jonge WJ, The FO, van der Coelen D, Bennink RJ, Reitsma PH, van Deventer SJ, et al. Mast

cell degranulation during abdominal surgery initiates postoperative ileus in mice.

Gastroenterology 2004 Aug;127(2):535-45.

De Schepper HU, De Man JG, Ruyssers NE, Deiteren A, Van Nassauw L, Timmermans JP, et al.

TRPV1 receptor signaling mediates afferent nerve sensitization during colitis-induced motility

disorders in rats. Am J Physiol Gastrointest Liver Physiol 2008 Jan;294(1):G245-G253.

De Schepper HU, De Man JG, Van Nassauw L, Timmermans JP, Herman AG, Pelckmans PA, et al.

Acute distal colitis impairs gastric emptying in rats via an extrinsic neuronal reflex pathway

involving the pelvic nerve. Gut 2007 Feb;56(2):195-202.

De Schepper HU, De Winter BY, Van Nassauw L, Timmermans JP, Herman AG, Pelckmans PA, et

al. TRPV1 receptors on unmyelinated C-fibres mediate colitis-induced sensitization of pelvic

afferent nerve fibres in rats. J Physiol 2008 Nov 1;586(Pt 21):5247-58.

de Stoppelaar SF, van ’t Veer C, Claushuis TA, Albersen BJ, Roelofs JJ, van der PT.

Thrombocytopenia impairs host defense in gram-negative pneumonia-derived sepsis in mice.

Blood 2014 Dec 11;124(25):3781-90.

De Winter BY, Boeckxstaens GE, De Man JG, Moreels TG, Herman AG, Pelckmans PA. Effect of

adrenergic and nitrergic blockade on experimental ileus in rats. Br J Pharmacol 1997

Feb;120(3):464-8.

De Winter BY, Bredenoord AJ, De Man JG, Moreels TG, Herman AG, Pelckmans PA. Effect of

inhibition of inducible nitric oxide synthase and guanylyl cyclase on endotoxin-induced delay in

gastric emptying and intestinal transit in mice. Shock 2002 Aug;18(2):125-31.

De Winter BY, Bredenoord AJ, Van Nassauw L, De Man JG, De Schepper HU, Timmermans JP, et

al. Involvement of afferent neurons in the pathogenesis of endotoxin-induced ileus in mice:

role of CGRP and TRPV1 receptors. Eur J Pharmacol 2009 Aug 1;615(1-3):177-84.

De Winter BY, De Man JG. Interplay between inflammation, immune system and neuronal

pathways: effect on gastrointestinal motility. World J Gastroenterol 2010 Nov 28;16(44):5523-

35.

References

221

De Winter BY, Deiteren A, De Man JG. Novel nervous system mechanisms in visceral pain.

Neurogastroenterol Motil 2016 Mar;28(3):309-15.

De Winter BY, van den Wijngaard RM, de Jonge WJ. Intestinal mast cells in gut inflammation

and motility disturbances. Biochim Biophys Acta 2012 Jan;1822(1):66-73.

De Winter BY, Van Nassauw L, De Man JG, de JF, Bredenoord AJ, Seerden TC, et al. Role of

oxidative stress in the pathogenesis of septic ileus in mice. Neurogastroenterol Motil 2005

Apr;17(2):251-61.

Deitch EA, Rutan R, Waymack JP. Trauma, shock, and gut translocation. New Horiz 1996

May;4(2):289-99.

Deitch EA, Xu D, Kaise VL. Role of the gut in the development of injury- and shock induced SIRS

and MODS: the gut-lymph hypothesis, a review. Front Biosci 2006 Jan 1;11:520-8.:520-8.

Deitch EA. Gut lymph and lymphatics: a source of factors leading to organ injury and

dysfunction. Ann N Y Acad Sci 2010 Oct;1207 Suppl 1:E103-11. doi: 10.1111/j.1749-

6632.2010.05713.x.:E103-E111.

Deitch EA. Gut-origin sepsis: evolution of a concept. Surgeon 2012 Dec;10(6):350-6.

Deiteren A, De Man JG, Keating C, Jiang W, De Schepper HU, Pelckmans PA, et al. Mechanisms

contributing to visceral hypersensitivity: focus on splanchnic afferent nerve signaling.

Neurogastroenterol Motil 2015 Dec;27(12):1709-20.

Deiteren A, De Winter BY, Nullens S, Pelckmans PA, De Man JG. Role of tachykinin receptors in

the modulation of colonic peristaltic activity in mice. Eur J Pharmacol 2011 Sep 30;667(1-

3):339-47.

del Rio ML, Bernhardt G, Rodriguez-Barbosa JI, Forster R. Development and functional

specialization of CD103+ dendritic cells. Immunol Rev 2010 Mar;234(1):268-81.

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 Apr;30(4):536-55.

Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, et al. Surviving Sepsis

Campaign: international guidelines for management of severe sepsis and septic shock: 2008.

Intensive Care Med 2008 Jan;34(1):17-60.

Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving sepsis

campaign: international guidelines for management of severe sepsis and septic shock, 2012.

Intensive Care Med 2013 Feb;39(2):165-228.

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

Mar 26;14:166. doi: 10.1186/1471-2334-14-166.:166-14.

References

222

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 Jun;3(6):678-81.

Dofferhoff AS, Bom VJ, de Vries-Hospers HG, van Ingen J, vd Meer J, Hazenberg BP, et al.

Patterns of cytokines, plasma endotoxin, plasminogen activator inhibitor, and acute-phase

proteins during the treatment of severe sepsis in humans. Crit Care Med 1992 Feb;20(2):185-

92.

Doi K, Leelahavanichkul A, Yuen PS, Star RA. Animal models of sepsis and sepsis-induced kidney

injury. J Clin Invest 2009 Oct;119(10):2868-78.

Donnenberg AD, Donnenberg VS. Rare-event analysis in flow cytometry. Clin Lab Med 2007

Sep;27(3):627-52, viii.

Donovan J, Grundy D. Endocannabinoid modulation of jejunal afferent responses to LPS.

Neurogastroenterol Motil 2012 Oct;24(10):956-e465.

Drewry AM, Fuller BM, Skrupky LP, Hotchkiss RS. The presence of hypothermia within 24 hours

of sepsis diagnosis predicts persistent lymphopenia. Crit Care Med 2015 Jun;43(6):1165-9.

Dyson A, Singer M. Animal models of sepsis: why does preclinical efficacy fail to translate to the

clinical setting? Crit Care Med 2009 Jan;37(1 Suppl):S30-S37.

Eadon MT, Hack BK, Xu C, Ko B, Toback FG, Cunningham PN. Endotoxemia alters tight junction

gene and protein expression in the kidney. Am J Physiol Renal Physiol 2012 Sep 15;303(6):F821-

F830.

Echtenacher B, Mannel DN, Hultner L. Critical protective role of mast cells in a model of acute

septic peritonitis. Nature 1996 May 2;381(6577):75-7.

Engel DR, Koscielny A, Wehner S, Maurer J, Schiwon M, Franken L, et al. T helper type 1

memory cells disseminate postoperative ileus over the entire intestinal tract. Nat Med 2010

Dec;16(12):1407-13.

Erbil Y, Berber E, Seven R, Calis A, Eminoglu L, Kocak M, et al. The effect of intestinal transit

time on bacterial translocation. Acta Chir Belg 1998 Dec;98(6):245-9.

Eskandari MK, Kalff JC, Billiar TR, Lee KK, Bauer AJ. Lipopolysaccharide activates the muscularis

macrophage network and suppresses circular smooth muscle activity. Am J Physiol 1997

Sep;273(3 Pt 1):G727-G734.

Eskandari MK, Kalff JC, Billiar TR, Lee KK, Bauer AJ. LPS-induced muscularis macrophage nitric

oxide suppresses rat jejunal circular muscle activity. Am J Physiol 1999 Aug;277(2 Pt 1):G478-

G486.

Farquhar MG, Palade GE. Junctional complexes in various epithelia. J Cell Biol 1963

May;17:375-412.:375-412.

References

223

Feighery L, Smyth A, Keely S, Baird AW, O'Connor WT, Callanan JJ, et al. Increased intestinal

permeability in rats subjected to traumatic frontal lobe percussion brain injury. J Trauma 2008

Jan;64(1):131-7.

Feng B, Gebhart GF. In vitro functional characterization of mouse colorectal afferent endings. J

Vis Exp 2015 Jan 21;(95):52310.

Fernandez R, Nardocci G, Navarro C, Reyes EP, Acuña-Castillo C, Cortes PP. Neural reflex

regulation of systemic inflammation: potential new targets for sepsis therapy. Front Physiol

2014 Dec 15;5:489. doi: 10.3389/fphys.2014.00489. eCollection;%2014.:489.

Fink MP. Sepsis, ghrelin, the cholinergic anti-inflammatory pathway, gut mucosal

hyperpermeability, and high-mobility group box 1. Crit Care Med 2009 Aug;37(8):2483-5.

Fisher CJ, Jr., Agosti JM, Opal SM, Lowry SF, Balk RA, Sadoff JC, et al. Treatment of septic shock

with the tumor necrosis factor receptor:Fc fusion protein. The Soluble TNF Receptor Sepsis

Study Group. N Engl J Med 1996 Jun 27;334(26):1697-702.

Fishman JE, Sheth SU, Levy G, Alli V, Lu Q, Xu D, et al. Intraluminal nonbacterial intestinal

components control gut and lung injury after trauma hemorrhagic shock. Ann Surg 2014

Dec;260(6):1112-20.

Fleshner M, Goehler LE, Schwartz BA, McGorry M, Martin D, Maier SF, et al. Thermogenic and

corticosterone responses to intravenous cytokines (IL-1beta and TNF-alpha) are attenuated by

subdiaphragmatic vagotomy. J Neuroimmunol 1998 Jun 15;86(2):134-41.

Flores-Langarica A, Meza-Perez S, Calderon-Amador J, Estrada-Garcia T, Macpherson G,

Lebecque S, et al. Network of dendritic cells within the muscular layer of the mouse intestine.

Proc Natl Acad Sci U S A 2005 Dec 27;102(52):19039-44.

Floto RA, Smith KG. The vagus nerve, macrophages, and nicotine. Lancet 2003 Mar

29;361(9363):1069-70.

Gakis G, Mueller MH, Hahn J, Glatzle J, Grundy D, Kreis ME. Neuronal activation in the nucleus

of the solitary tract following jejunal lipopolysaccharide in the rat. Auton Neurosci 2009 Jun

15;148(1-2):63-8.

Galley HF. Oxidative stress and mitochondrial dysfunction in sepsis. Br J Anaesth 2011

Jul;107(1):57-64.

Garcia-Lopez MA, Sanchez-Madrid F, Rodriguez-Frade JM, Mellado M, Acevedo A, Garcia MI, et

al. CXCR3 chemokine receptor distribution in normal and inflamed tissues: expression on

activated lymphocytes, endothelial cells, and dendritic cells. Lab Invest 2001 Mar;81(3):409-18.

Gebhart GF. Descending modulation of pain. Neurosci Biobehav Rev 2004 Jan;27(8):729-37.

Gennari R, Alexander JW, Pyles T, Hartmann S, Ogle CK. Effects of antimurine interleukin-6 on

bacterial translocation during gut-derived sepsis. Arch Surg 1994 Nov;129(11):1191-7.

References

224

Gennari R, Alexander JW. Anti-interleukin-6 antibody treatment improves survival during gut-

derived sepsis in a time-dependent manner by enhancing host defense. Crit Care Med 1995

Dec;23(12):1945-53.

Giannoudis PV, Harwood PJ, Loughenbury P, van GM, Krettek C, Pape HC. Correlation between

IL-6 levels and the systemic inflammatory response score: can an IL-6 cutoff predict a SIRS

state? J Trauma 2008 Sep;65(3):646-52.

Giebelen IA, van Westerloo DJ, Larosa GJ, de Vos AF, van der Poll T. Stimulation of alpha 7

cholinergic receptors inhibits lipopolysaccharide-induced neutrophil recruitment by a tumor

necrosis factor alpha-independent mechanism. Shock 2007 Apr;27(4):443-7.

Gitter AH, Bendfeldt K, Schulzke JD, Fromm M. Leaks in the epithelial barrier caused by

spontaneous and TNF-alpha-induced single-cell apoptosis. FASEB J 2000 Sep;14(12):1749-53.

Glatzle J, Leutenegger CM, Mueller MH, Kreis ME, Raybould HE, Zittel TT. Mesenteric lymph

collected during peritonitis or sepsis potently inhibits gastric motility in rats. J Gastrointest Surg

2004 Sep;8(6):645-52.

Glowka TR, Steinebach A, Stein K, Schwandt T, Lysson M, Holzmann B, et al. The novel CGRP

receptor antagonist BIBN4096BS alleviates a postoperative intestinal inflammation and

prevents postoperative ileus. Neurogastroenterol Motil 2015 Jul;27(7):1038-49.

Goehler LE, Gaykema RP, Hammack SE, Maier SF, Watkins LR. Interleukin-1 induces c-Fos

immunoreactivity in primary afferent neurons of the vagus nerve. Brain Res 1998 Sep

7;804(2):306-10.

Gomez-Pinilla PJ, Farro G, Di GM, Stakenborg N, Nemethova A, de Vries A, et al. Mast cells play

no role in the pathogenesis of postoperative ileus induced by intestinal manipulation. PLoS One

2014 Jan 9;9(1):e85304.

Green KJ, Jones JC. Desmosomes and hemidesmosomes: structure and function of molecular

components. FASEB J 1996 Jun;10(8):871-81.

Grivennikov SI, Tumanov AV, Liepinsh DJ, Kruglov AA, Marakusha BI, Shakhov AN, et al. Distinct

and nonredundant in vivo functions of TNF produced by t cells and macrophages/neutrophils:

protective and deleterious effects. Immunity 2005 Jan;22(1):93-104.

Grundy D, Schemann M. Enteric nervous system. Curr Opin Gastroenterol 2007 Mar;23(2):121-

6.

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.

Guilliams M, Ginhoux F, Jakubzick C, Naik SH, Onai N, Schraml BU, et al. Dendritic cells,

monocytes and macrophages: a unified nomenclature based on ontogeny. Nat Rev Immunol

2014 Aug;14(8):571-8.

References

225

Han X, Fink MP, Delude RL. Proinflammatory cytokines cause NO*-dependent and -

independent changes in expression and localization of tight junction proteins in intestinal

epithelial cells. Shock 2003 Mar;19(3):229-37.

Hara N, Chijiiwa M, Yara M, Ishida Y, Ogiwara Y, Inazu M, et al. Metabolomic Analyses of Brain

Tissue in Sepsis Induced by Cecal Ligation Reveal Specific Redox Alterations--Protective Effects

of the Oxygen Radical Scavenger Edaravone. Shock 2015 Dec;44(6):578-84.

Hassoun HT, Kone BC, Mercer DW, Moody FG, Weisbrodt NW, Moore FA. Post-injury multiple

organ failure: the role of the gut. Shock 2001 Jan;15(1):1-10.

Hedley BD, Keeney M. Technical issues: flow cytometry and rare event analysis. Int J Lab

Hematol 2013 Jun;35(3):344-50.

Heffernan DS, Monaghan SF, Chung CS, Cioffi WG, Gravenstein S, Ayala A. A divergent response

of innate regulatory T-cells to sepsis in humans: circulating invariant natural killer T-cells are

preserved. Hum Immunol 2014 Mar;75(3):277-82.

Helft J, Bottcher J, Chakravarty P, Zelenay S, Huotari J, Schraml BU, et al. GM-CSF Mouse Bone

Marrow Cultures Comprise a Heterogeneous Population of CD11c(+)MHCII(+) Macrophages

and Dendritic Cells. Immunity 2015 Jun 16;42(6):1197-211.

Heylen M, Ruyssers NE, De Man JG, Timmermans JP, Pelckmans PA, Moreels TG, et al. Worm

proteins of Schistosoma mansoni reduce the severity of experimental chronic colitis in mice by

suppressing colonic proinflammatory immune responses. PLoS One 2014 Oct

14;9(10):e110002.

Heylen M, Ruyssers NE, Gielis EM, Vanhomwegen E, Pelckmans PA, Moreels TG, et al. Of

worms, mice and man: an overview of experimental and clinical helminth-based therapy for

inflammatory bowel disease. Pharmacol Ther 2014 Aug;143(2):153-67.

Heylen M, Ruyssers NE, Nullens S, Schramm G, Pelckmans PA, Moreels TG, et al. Treatment

with Egg Antigens of Schistosoma mansoni Ameliorates Experimental Colitis in Mice Through a

Colonic T-cell-dependent Mechanism. Inflamm Bowel Dis 2014 Nov 27.

Hofer S, Steppan J, Wagner T, Funke B, Lichtenstern C, Martin E, et al. Central sympatholytics

prolong survival in experimental sepsis. Crit Care 2009;13(1):R11.

Holzer P, Danzer M, Schicho R, Samberger C, Painsipp E, Lippe IT. Vagal afferent input from the

acid-challenged rat stomach to the brainstem: enhancement by interleukin-1beta.

Neuroscience 2004;129(2):439-45.

Holzer P, Lippe IT, Heinemann A, Bartho L. Tachykinin NK1 and NK2 receptor-mediated control

of peristaltic propulsion in the guinea-pig small intestine in vitro. Neuropharmacology

1998;37(1):131-8.

Holzer P. Opioid receptors in the gastrointestinal tract. Regul Pept 2009 Jun 5;155(1-3):11-7.

References

226

Holzer P. Transient receptor potential (TRP) channels as drug targets for diseases of the

digestive system. Pharmacol Ther 2011 Jul;131(1):142-70.

Hotchkiss RS, Chang KC, Swanson PE, Tinsley KW, Hui JJ, Klender P, et al. Caspase inhibitors

improve survival in sepsis: a critical role of the lymphocyte. Nat Immunol 2000 Dec;1(6):496-

501.

Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003 Jan

9;348(2):138-50.

Hotchkiss RS, Monneret G, Payen D. Immunosuppression in sepsis: a novel understanding of

the disorder and a new therapeutic approach. Lancet Infect Dis 2013 Mar;13(3):260-8.

Hotchkiss RS, Monneret G, Payen D. Sepsis-induced immunosuppression: from cellular

dysfunctions to immunotherapy. Nat Rev Immunol 2013 Dec;13(12):862-74.

Hotchkiss RS, Opal S. Immunotherapy for sepsis--a new approach against an ancient foe. N Engl

J Med 2010 Jul 1;363(1):87-9.

Hotchkiss RS, Tinsley KW, Swanson PE, Schmieg RE, Jr., Hui JJ, Chang KC, et al. Sepsis-induced

apoptosis causes progressive profound depletion of B and CD4+ T lymphocytes in humans. J

Immunol 2001 Jun 1;166(11):6952-63.

Huang J, Wang Y, Jiang D, Zhou J, Huang X. The sympathetic-vagal balance against

endotoxemia. J Neural Transm 2010 Jun;117(6):729-35.

Hughes PA, Harrington AM, Castro J, Liebregts T, Adam B, Grasby DJ, et al. Sensory neuro-

immune interactions differ between irritable bowel syndrome subtypes. Gut 2013

Oct;62(10):1456-65.

Hughes PA, Moretta M, Lim A, Grasby DJ, Bird D, Brierley SM, et al. Immune derived opioidergic

inhibition of viscerosensory afferents is decreased in Irritable Bowel Syndrome patients. Brain

Behav Immun 2014 Nov;42:191-203. doi: 10.1016/j.bbi.2014.07.001. Epub;%2014 Jul 23.:191-

203.

Hugunin KM, Fry C, Shuster K, Nemzek JA. Effects of tramadol and buprenorphine on select

immunologic factors in a cecal ligation and puncture model. Shock 2010 Sep;34(3):250-60.

Hulspas R, O'Gorman MR, Wood BL, Gratama JW, Sutherland DR. Considerations for the control

of background fluorescence in clinical flow cytometry. Cytometry B Clin Cytom 2009

Nov;76(6):355-64.

Hunter CA, Jones SA. IL-6 as a keystone cytokine in health and disease. Nat Immunol 2015

May;16(5):448-57.

Huston JM, Gallowitsch-Puerta M, Ochani M, Ochani K, Yuan R, Rosas-Ballina M, et al.

Transcutaneous vagus nerve stimulation reduces serum high mobility group box 1 levels and

improves survival in murine sepsis. Crit Care Med 2007 Dec;35(12):2762-8.

References

227

Huston JM, Ochani M, Rosas-Ballina M, Liao H, Ochani K, Pavlov VA, et al. Splenectomy

inactivates the cholinergic antiinflammatory pathway during lethal endotoxemia and

polymicrobial sepsis. J Exp Med 2006 Jul 10;203(7):1623-8.

Huston JM, Wang H, Ochani M, Ochani K, Rosas-Ballina M, Gallowitsch-Puerta M, et al.

Splenectomy protects against sepsis lethality and reduces serum HMGB1 levels. J Immunol

2008 Sep 1;181(5):3535-9.

Hutchins NA, Unsinger J, Hotchkiss RS, Ayala A. The new normal: immunomodulatory agents

against sepsis immune suppression. Trends Mol Med 2014 Apr;20(4):224-33.

Imamura M, Clowes GH, Jr. Hepatic blood flow and oxygen consumption in starvation, sepsis

and septic shock. Surg Gynecol Obstet 1975 Jul;141(1):27-34.

Inada T, Hamano N, Yamada M, Shirane A, Shingu K. Inducible nitric oxide synthase and tumor

necrosis factor-alpha in delayed gastric emptying and gastrointestinal transit induced by

lipopolysaccharide in mice. Braz J Med Biol Res 2006 Nov;39(11):1425-34.

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 Jul;36(1):38-44.

Izcue A, Coombes JL, Powrie F. Regulatory T cells suppress systemic and mucosal immune

activation to control intestinal inflammation. Immunol Rev 2006 Aug;212:256-71.:256-71.

Izzo AA, Sharkey KA. Cannabinoids and the gut: new developments and emerging concepts.

Pharmacol Ther 2010 Apr;126(1):21-38.

Jacob C, Yang PC, Darmoul D, Amadesi S, Saito T, Cottrell GS, et al. Mast cell tryptase controls

paracellular permeability of the intestine. Role of protease-activated receptor 2 and beta-

arrestins. J Biol Chem 2005 Sep 9;280(36):31936-48.

Ji H, Rabbi MF, Labis B, Pavlov VA, Tracey KJ, Ghia JE. Central cholinergic activation of a vagus

nerve-to-spleen circuit alleviates experimental colitis. Mucosal Immunol 2014 Mar;7(2):335-47.

Jiang LN, Yao YM, Sheng ZY. The role of regulatory T cells in the pathogenesis of sepsis and its

clinical implication. J Interferon Cytokine Res 2012 Aug;32(8):341-9.

Jiang Q, Yi M, Guo Q, Wang C, Wang H, Meng S, et al. Protective effects of polydatin on

lipopolysaccharide-induced acute lung injury through TLR4-MyD88-NF-kappaB pathway. Int

Immunopharmacol 2015 Dec;29(2):370-6.

Jiang W, Kirkup AJ, Grundy D. Mast cells drive mesenteric afferent signalling during acute

intestinal ischaemia. J Physiol 2011 Aug 1;589(Pt 15):3867-82.

Johansson-Lindbom B, Svensson M, Pabst O, Palmqvist C, Marquez G, Forster R, et al.

Functional specialization of gut CD103+ dendritic cells in the regulation of tissue-selective T cell

homing. J Exp Med 2005 Oct 17;202(8):1063-73.

References

228

Johnston DG, Corr SC. Toll-Like Receptor Signalling and the Control of Intestinal Barrier

Function. Methods Mol Biol 2016;1390:287-300. doi: 10.1007/978-1-4939-3335-8_18.:287-300.

Kalff JC, Buchholz BM, Eskandari MK, Hierholzer C, Schraut WH, Simmons RL, et al. Biphasic

response to gut manipulation and temporal correlation of cellular infiltrates and muscle

dysfunction in rat. Surgery 1999 Sep;126(3):498-509.

Kalff JC, Carlos TM, Schraut WH, Billiar TR, Simmons RL, Bauer AJ. Surgically induced leukocytic

infiltrates within the rat intestinal muscularis mediate postoperative ileus. Gastroenterology

1999 Aug;117(2):378-87.

Kalff JC, Schraut WH, Billiar TR, Simmons RL, Bauer AJ. Role of inducible nitric oxide synthase in

postoperative intestinal smooth muscle dysfunction in rodents. Gastroenterology 2000

Feb;118(2):316-27.

Kalff JC, Schraut WH, Simmons RL, Bauer AJ. Surgical manipulation of the gut elicits an

intestinal muscularis inflammatory response resulting in postsurgical ileus. Ann Surg 1998

Nov;228(5):652-63.

Kang S, Tanaka T, Kishimoto T. Therapeutic uses of anti-interleukin-6 receptor antibody. Int

Immunol 2015 Jan;27(1):21-9.

Kawahara R, Yasuda M, Hashimura H, Amagase K, Kato S, Takeuchi K. Activation of alpha7

nicotinic acetylcholine receptors ameliorates indomethacin-induced small intestinal ulceration

in mice. Eur J Pharmacol 2011 Jan 10;650(1):411-7.

Keating C, Beyak M, Foley S, Singh G, Marsden C, Spiller R, et al. Afferent hypersensitivity in a

mouse model of post-inflammatory gut dysfunction: role of altered serotonin metabolism. J

Physiol 2008 Sep 15;586(Pt 18):4517-30.

Keita AV, Soderholm JD. The intestinal barrier and its regulation by neuroimmune factors.

Neurogastroenterol Motil 2010 Jul;22(7):718-33.

Kessler W, Diedrich S, Menges P, Ebker T, Nielson M, Partecke LI, et al. The role of the vagus

nerve: modulation of the inflammatory reaction in murine polymicrobial sepsis. Mediators

Inflamm 2012;2012:467620. Epub;%2012 Mar 21.:467620.

Khan MA, Farkhondeh M, Crombie J, Jacobson L, Kaneki M, Martyn JA. Lipopolysaccharide

upregulates alpha7 acetylcholine receptors: stimulation with GTS-21 mitigates growth arrest of

macrophages and improves survival in burned mice. Shock 2012 Aug;38(2):213-9.

Kianian M, Al-Banna NA, Kelly ME, Lehmann C. Inhibition of endocannabinoid degradation in

experimental endotoxemia reduces leukocyte adhesion and improves capillary perfusion in the

gut. J Basic Clin Physiol Pharmacol 2013;24(1):27-33.

Kimura A, Kishimoto T. IL-6: regulator of Treg/Th17 balance. Eur J Immunol 2010

Jul;40(7):1830-5.

References

229

Kinnebrew MA, Buffie CG, Diehl GE, Zenewicz LA, Leiner I, Hohl TM, et al. Interleukin 23

production by intestinal CD103(+)CD11b(+) dendritic cells in response to bacterial flagellin

enhances mucosal innate immune defense. Immunity 2012 Feb 24;36(2):276-87.

Knowles CH, Aziz Q. Basic and clinical aspects of gastrointestinal pain. Pain 2009

Feb;141(3):191-209.

Konigsrainer I, Turck MH, Eisner F, Meile T, Hoffmann J, Kuper M, et al. The Gut is not only the

Target but a Source of Inflammatory Mediators Inhibiting Gastrointestinal Motility During

Sepsis. Cell Physiol Biochem 2011;28(4):753-60.

Koscielny A, Boerner T, Wehner S, Kurts C, Kalff JC. The role of dendritic cells in the

gastrointestinal field effect. Transplant Proc 2006 Jul;38(6):1815-7.

Koscielny A, Engel D, Maurer J, Wehner S, Kurts C, Kalff JC. The role of lymphoid tissue in the

attenuation of the postoperative ileus. Am J Physiol Gastrointest Liver Physiol 2013 Feb

15;304(4):G401-G412.

Koscielny A, Kalff JC. T-helper cell type 1 memory cells and postoperative ileus in the entire gut.

Curr Opin Gastroenterol 2011 Oct;27(6):509-14.

Kox M, Pompe JC, Gordinou de Gouberville MC, van der Hoeven JG, Hoedemaekers CW,

Pickkers P. Effects of the alpha7 nicotinic acetylcholine receptor agonist GTS-21 on the innate

immune response in humans. Shock 2011 Jul;36(1):5-11.

Kox M, Pompe JC, Peters E, Vaneker M, van der Laak JW, van der Hoeven JG, et al. alpha7

nicotinic acetylcholine receptor agonist GTS-21 attenuates ventilator-induced tumour necrosis

factor-alpha production and lung injury. Br J Anaesth 2011 Oct;107(4):559-66.

Kox M, van Eijk LT, Verhaak T, Frenzel T, Kiers HD, Gerretsen J, et al. Transvenous vagus nerve

stimulation does not modulate the innate immune response during experimental human

endotoxemia: a randomized controlled study. Arthritis Res Ther 2015 Jun 7;17:150. doi:

10.1186/s13075-015-0667-5.:150-0667.

Kreiss C, Toegel S, Bauer AJ. Alpha2-adrenergic regulation of NO production alters

postoperative intestinal smooth muscle dysfunction in rodents. Am J Physiol Gastrointest Liver

Physiol 2004 Sep;287(3):G658-G666.

Krzyzaniak M, Peterson C, Loomis W, Hageny AM, Wolf P, Reys L, et al. Postinjury vagal nerve

stimulation protects against intestinal epithelial barrier breakdown. J Trauma 2011

May;70(5):1168-75.

Lange S, Delbro DS, Jennische E. Evans blue permeation of intestinal mucosa in the rat. Scand J

Gastroenterol 1994 Jan;29(1):38-46.

Ledeganck KJ, Boulet GA, Bogers JJ, Verpooten GA, De Winter BY. The TRPM6/EGF pathway is

downregulated in a rat model of cisplatin nephrotoxicity. PLoS One 2013;8(2):e57016.

References

230

Lee JA, Spidlen J, Boyce K, Cai J, Crosbie N, Dalphin M, et al. MIFlowCyt: the minimum

information about a Flow Cytometry Experiment. Cytometry A 2008 Oct;73(10):926-30.

Lefrancois L, Lycke N. Isolation of mouse small intestinal intraepithelial lymphocytes, Peyer's

patch, and lamina propria cells. Curr Protoc Immunol 2001 May;Chapter 3:Unit 3.19. doi:

10.1002/0471142735.im0319s17.:Unit.

Leng FY, Liu JL, Liu ZJ, Yin JY, Qu HP. Increased proportion of CD4(+)CD25(+)Foxp3(+) regulatory

T cells during the early-stage sepsis in ICU patients. J Microbiol Immunol Infect 2012 Aug 23.

Levy G, Fishman JE, Xu D, Chandler BT, Feketova E, Dong W, et al. Parasympathetic stimulation

via the vagus nerve prevents systemic organ dysfunction by abrogating gut injury and lymph

toxicity in trauma and hemorrhagic shock. Shock 2013 Jan;39(1):39-44.

Levy MM, Dellinger RP, Townsend SR, Linde-Zwirble WT, Marshall JC, Bion J, et al. The Surviving

Sepsis Campaign: results of an international guideline-based performance improvement

program targeting severe sepsis. Intensive Care Med 2010 Feb;36(2):222-31.

Li YY, Li YN, Ni JB, Chen CJ, Lv S, Chai SY, et al. Involvement of cannabinoid-1 and cannabinoid-2

receptors in septic ileus. Neurogastroenterol Motil 2010 Mar;22(3):350-e88.

Lin CS, Hung SF, Huang HS, Ma MC. Blockade of the N-Methyl-D-Aspartate Glutamate Receptor

Ameliorates Lipopolysaccharide-Induced Renal Insufficiency. PLoS One 2015 Jul

2;10(7):e0132204.

Liu C, Li A, Weng YB, Duan ML, Wang BE, Zhang SW. Changes in intestinal mucosal immune

barrier in rats with endotoxemia. World J Gastroenterol 2009 Dec 14;15(46):5843-50.

Liu CY, Jiang W, Muller MH, Grundy D, Kreis ME. Sensitization of mesenteric afferents to

chemical and mechanical stimuli following systemic bacterial lipopolysaccharide.

Neurogastroenterol Motil 2005 Feb;17(1):89-101.

Liu CY, Mueller MH, Grundy D, Kreis ME. Vagal modulation of intestinal afferent sensitivity to

systemic LPS in the rat. Am J Physiol Gastrointest Liver Physiol 2007 May;292(5):G1213-G1220.

Liu CY, Mueller MH, Rogler G, Grundy D, Kreis ME. Differential afferent sensitivity to mucosal

lipopolysaccharide from Salmonella typhimurium and Escherichia coli in the rat jejunum.

Neurogastroenterol Motil 2009 Dec;21(12):1335-e129.

Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative

PCR and the 2(-Delta Delta C(T)) Method. Methods 2001 Dec;25(4):402-8.

Livingston EH, Passaro EP, Jr. Postoperative ileus. Dig Dis Sci 1990 Jan;35(1):121-32.

Lomax AE, Fernandez E, Sharkey KA. Plasticity of the enteric nervous system during intestinal

inflammation. Neurogastroenterol Motil 2005 Feb;17(1):4-15.

Luettig J, Rosenthal R, Barmeyer C, Schulzke JD. Claudin-2 as a mediator of leaky gut barrier

during intestinal inflammation. Tissue Barriers 2015 Apr 3;3(1-2):e977176.

References

231

MacDonald TT, Monteleone I, Fantini MC, Monteleone G. Regulation of homeostasis and

inflammation in the intestine. Gastroenterology 2011 May;140(6):1768-75.

MacFie J, O'Boyle C, Mitchell CJ, Buckley PM, Johnstone D, Sudworth P. Gut origin of sepsis: a

prospective study investigating associations between bacterial translocation, gastric microflora,

and septic morbidity. Gut 1999 Aug;45(2):223-8.

MacFie J, Reddy BS, Gatt M, Jain PK, Sowdi R, Mitchell CJ. Bacterial translocation studied in 927

patients over 13 years. Br J Surg 2006 Jan;93(1):87-93.

MacGlashan D, Jr. IgE receptor and signal transduction in mast cells and basophils. Curr Opin

Immunol 2008 Dec;20(6):717-23.

Maecker HT, McCoy JP, Jr., Amos M, Elliott J, Gaigalas A, Wang L, et al. A model for harmonizing

flow cytometry in clinical trials. Nat Immunol 2010 Nov;11(11):975-8.

Maier S, Traeger T, Entleutner M, Westerholt A, Kleist B, Huser N, et al. Cecal ligation and

puncture versus colon ascendens stent peritonitis: two distinct animal models for polymicrobial

sepsis. Shock 2004 Jun;21(6):505-11.

Maldonado RA, von Andrian UH. How tolerogenic dendritic cells induce regulatory T cells. Adv

Immunol 2010;108:111-65. doi: 10.1016/B978-0-12-380995-7.00004-5.:111-65.

Mangodt TC, Van Herck MA, Nullens S, Ramet J, De Dooy JJ, Jorens PG, et al. The role of Th17

and Treg responses in the pathogenesis of RSV infection. Pediatr Res 2015 Nov;78(5):483-91.

Mann ER, Landy JD, Bernardo D, Peake ST, Hart AL, Al-Hassi HO, et al. Intestinal dendritic cells:

their role in intestinal inflammation, manipulation by the gut microbiota and differences

between mice and men. Immunol Lett 2013 Feb;150(1-2):30-40.

Mann ER, Landy JD, Bernardo D, Peake ST, Hart AL, Al-Hassi HO, et al. Intestinal dendritic cells:

Their role in intestinal inflammation, manipulation by the gut microbiota and differences

between mice and men. Immunol Lett 2013 Jan 23;150(1-2):30-40.

Markov AG, Veshnyakova A, Fromm M, Amasheh M, Amasheh S. Segmental expression of

claudin proteins correlates with tight junction barrier properties in rat intestine. J Comp Physiol

B 2010 Apr;180(4):591-8.

Marshall JC. Such stuff as dreams are made on: mediator-directed therapy in sepsis. Nat Rev

Drug Discov 2003 May;2(5):391-405.

Martelli D, McKinley MJ, McAllen RM. The cholinergic anti-inflammatory pathway: A critical

review. Auton Neurosci 2014 May;182:65-9. doi: 10.1016/j.autneu.2013.12.007. Epub;%2013

Dec 24.:65-9.

Martelli D, Yao ST, McKinley MJ, McAllen RM. Reflex control of inflammation by sympathetic

nerves, not the vagus. J Physiol 2014 Apr 1;592(Pt 7):1677-86.

References

232

Maruo N, Morita I, Shirao M, Murota S. IL-6 increases endothelial permeability in vitro.

Endocrinology 1992 Aug;131(2):710-4.

Masson S, Caironi P, Spanuth E, Thomae R, Panigada M, Sangiorgi G, et al. Presepsin (soluble

CD14 subtype) and procalcitonin levels for mortality prediction in sepsis: data from the

Albumin Italian Outcome Sepsis trial. Crit Care 2014 Jan 7;18(1):R6.

Matteoli G, Gomez-Pinilla PJ, Nemethova A, Di GM, Cailotto C, VAN Bree SH, et al. A distinct

vagal anti-inflammatory pathway modulates intestinal muscularis resident macrophages

independent of the spleen. Gut 2014 Jun;63(6):938-48.

Matthys P, Mitera T, Heremans H, Van DJ, Billiau A. Anti-gamma interferon and anti-

interleukin-6 antibodies affect staphylococcal enterotoxin B-induced weight loss,

hypoglycemia, and cytokine release in D-galactosamine-sensitized and unsensitized mice. Infect

Immun 1995 Apr;63(4):1158-64.

Maurer M, Wedemeyer J, Metz M, Piliponsky AM, Weller K, Chatterjea D, et al. Mast cells

promote homeostasis by limiting endothelin-1-induced toxicity. Nature 2004 Nov

25;432(7016):512-6.

Mayer EA, Aziz Q, Coen S, Kern M, Labus JS, Lane R, et al. Brain imaging approaches to the

study of functional GI disorders: a Rome working team report. Neurogastroenterol Motil 2009

Jun;21(6):579-96.

Mayer EA, Berman S, Suyenobu B, Labus J, Mandelkern MA, Naliboff BD, et al. Differences in

brain responses to visceral pain between patients with irritable bowel syndrome and ulcerative

colitis. Pain 2005 Jun;115(3):398-409.

Mayer EA. Gut feelings: the emerging biology of gut-brain communication. Nat Rev Neurosci

2011 Jul 13;12(8):453-66.

McCarthy DO. Tumor necrosis factor alpha and interleukin-6 have differential effects on food

intake and gastric emptying in fasted rats. Res Nurs Health 2000 Jun;23(3):222-8.

McRoberts JA, Li J, Ennes HS, Mayer EA. Sex-dependent differences in the activity and

modulation of N-methyl-d-aspartic acid receptors in rat dorsal root ganglia neurons.

Neuroscience 2007 Sep 21;148(4):1015-20.

Mees ST, Toellner S, Marx K, Faendrich F, Kallen KJ, Schroeder J, et al. Inhibition of interleukin-

6-transsignaling via gp130-Fc in hemorrhagic shock and sepsis. J Surg Res 2009 Dec;157(2):235-

42.

Merad M, Sathe P, Helft J, Miller J, Mortha A. The dendritic cell lineage: ontogeny and function

of dendritic cells and their subsets in the steady state and the inflamed setting. Annu Rev

Immunol 2013;31:563-604. doi: 10.1146/annurev-immunol-020711-074950.:563-604.

References

233

Miller SM, Reed D, Sarr MG, Farrugia G, Szurszewski JH. Haem oxygenase in enteric nervous

system of human stomach and jejunum and co-localization with nitric oxide synthase.

Neurogastroenterol Motil 2001 Apr;13(2):121-31.

Mizock BA. The multiple organ dysfunction syndrome. Dis Mon 2009 Aug;55(8):476-526.

Mizuhara H, O'Neill E, Seki N, Ogawa T, Kusunoki C, Otsuka K, et al. T cell activation-associated

hepatic injury: mediation by tumor necrosis factors and protection by interleukin 6. J Exp Med

1994 May 1;179(5):1529-37.

Monneret G, Debard AL, Venet F, Bohe J, Hequet O, Bienvenu J, et al. Marked elevation of

human circulating CD4+CD25+ regulatory T cells in sepsis-induced immunoparalysis. Crit Care

Med 2003 Jul;31(7):2068-71.

Monneret G, Venet F. A rapidly progressing lymphocyte exhaustion after severe sepsis. Crit

Care 2012 Jul 23;16(4):140.

Montuschi P, Preziosi P, Navarra P. Interleukin-1 alpha and tumour necrosis factor inhibit rat

gastric fundus motility in vitro. Eur J Pharmacol 1993 Mar 23;233(2-3):303-4.

Moore FA, Moore EE, Poggetti R, McAnena OJ, Peterson VM, Abernathy CM, et al. Gut bacterial

translocation via the portal vein: a clinical perspective with major torso trauma. J Trauma 1991

May;31(5):629-36.

Mosinska P, Zielinska M, Fichna J. Expression and physiology of opioid receptors in the

gastrointestinal tract. Curr Opin Endocrinol Diabetes Obes 2016 Feb;23(1):3-10.

Mostafa Anower AK, Shim JA, Choi B, Sohn S. Pretreatment with interleukin-6 small interfering

RNA can improve the survival rate of polymicrobial cecal ligation and puncture mice by down

regulating interleukin-6 production. Eur J Pharmacol 2012 Aug 5;688(1-3):76-83.

Munford RS, Pugin J. Normal responses to injury prevent systemic inflammation and can be

immunosuppressive. Am J Respir Crit Care Med 2001 Feb;163(2):316-21.

Munyaka P, Rabbi MF, Pavlov VA, Tracey KJ, Khafipour E, Ghia JE. Central muscarinic cholinergic

activation alters interaction between splenic dendritic cell and CD4+CD25- T cells in

experimental colitis. PLoS One 2014 Oct 8;9(10):e109272.

Murray PJ, Wynn TA. Protective and pathogenic functions of macrophage subsets. Nat Rev

Immunol 2011 Oct 14;11(11):723-37.

Nadel S, Goldstein B, Williams MD, Dalton H, Peters M, Macias WL, et al. Drotrecogin alfa

(activated) in children with severe sepsis: a multicentre phase III randomised controlled trial.

Lancet 2007 Mar 10;369(9564):836-43.

Nance DM, Sanders VM. Autonomic innervation and regulation of the immune system (1987-

2007). Brain Behav Immun 2007 Aug;21(6):736-45.

References

234

Nardi GM, Bet AC, Sordi R, Fernandes D, Assreuy J. Opioid analgesics in experimental sepsis:

effects on physiological, biochemical, and haemodynamic parameters. Fundam Clin Pharmacol

2013 Aug;27(4):347-53.

Nemethova A, Michel K, Gomez-Pinilla PJ, Boeckxstaens GE, Schemann M. Nicotine attenuates

activation of tissue resident macrophages in the mouse stomach through the beta2 nicotinic

acetylcholine receptor. PLoS One 2013 Nov 1;8(11):e79264.

Neunlist M, Rolli-Derkinderen M, Latorre R, Van LL, Coron E, Derkinderen P, et al. Enteric glial

cells: recent developments and future directions. Gastroenterology 2014 Dec;147(6):1230-7.

Nijhuis LE, Olivier BJ, de Jonge WJ. Neurogenic regulation of dendritic cells in the intestine.

Biochem Pharmacol 2010 Dec 15;80(12):2002-8.

Ochoa-Cortes F, Ramos-Lomas T, Miranda-Morales M, Spreadbury I, Ibeakanma C, Barajas-

Lopez C, et al. Bacterial cell products signal to mouse colonic nociceptive dorsal root ganglia

neurons. Am J Physiol Gastrointest Liver Physiol 2010 Sep;299(3):G723-G732.

Odenwald MA, Turner JR. Intestinal permeability defects: is it time to treat? Clin Gastroenterol

Hepatol 2013 Sep;11(9):1075-83.

Olincy A, Freedman R. Nicotinic mechanisms in the treatment of psychotic disorders: a focus on

the alpha7 nicotinic receptor. Handb Exp Pharmacol 2012;(213):211-32.

Olincy A, Harris JG, Johnson LL, Pender V, Kongs S, Allensworth D, et al. Proof-of-concept trial of

an alpha7 nicotinic agonist in schizophrenia. Arch Gen Psychiatry 2006 Jun;63(6):630-8.

Opal SM, Fisher CJ, Jr., Dhainaut JF, Vincent JL, Brase R, Lowry SF, et al. Confirmatory

interleukin-1 receptor antagonist trial in severe sepsis: a phase III, randomized, double-blind,

placebo-controlled, multicenter trial. The Interleukin-1 Receptor Antagonist Sepsis Investigator

Group. Crit Care Med 1997 Jul;25(7):1115-24.

Opal SM, Laterre PF, Francois B, LaRosa SP, Angus DC, Mira JP, et al. Effect of eritoran, an

antagonist of MD2-TLR4, on mortality in patients with severe sepsis: the ACCESS randomized

trial. JAMA 2013 Mar;%20;309(11):1154-62.

Ostertag D, Buhner S, Michel K, Pehl C, Kurjak M, Gotzberger M, et al. Reduced Responses of

Submucous Neurons from Irritable Bowel Syndrome Patients to a Cocktail Containing

Histamine, Serotonin, TNFalpha, and Tryptase (IBS-Cocktail). Front Neurosci 2015 Dec 16;9:465.

doi: 10.3389/fnins.2015.00465. eCollection;%2015.:465.

Overhaus M, Moore BA, Barbato JE, Behrendt FF, Doering JG, Bauer AJ. Biliverdin protects

against polymicrobial sepsis by modulating inflammatory mediators. Am J Physiol Gastrointest

Liver Physiol 2006 Apr;290(4):G695-G703.

Overhaus M, Toegel S, Bauer AJ. Interaction of hemorrhagic shock and subsequent

polymicrobial sepsis on gastrointestinal motility. Shock 2009 Apr;31(4):382-9.

References

235

Overhaus M, Togel S, Pezzone MA, Bauer AJ. Mechanisms of polymicrobial sepsis-induced ileus.

Am J Physiol Gastrointest Liver Physiol 2004 Sep;287(3):G685-G694.

Owczarek D, Cibor D, Mach T, Ciesla A, Pierzchala-Koziec K, Salapa K, et al. Met-enkephalins in

patients with inflammatory bowel diseases. Adv Med Sci 2011;56(2):158-64.

Ozaktay AC, Kallakuri S, Takebayashi T, Cavanaugh JM, Asik I, DeLeo JA, et al. Effects of

interleukin-1 beta, interleukin-6, and tumor necrosis factor on sensitivity of dorsal root

ganglion and peripheral receptive fields in rats. Eur Spine J 2006 Oct;15(10):1529-37.

Pacheco R, Riquelme E, Kalergis AM. Emerging evidence for the role of neurotransmitters in the

modulation of T cell responses to cognate ligands. Cent Nerv Syst Agents Med Chem 2010

Mar;10(1):65-83.

Pachot A, Monneret G, Voirin N, Leissner P, Venet F, Bohe J, et al. Longitudinal study of

cytokine and immune transcription factor mRNA expression in septic shock. Clin Immunol 2005

Jan;114(1):61-9.

Pallua N, Low JF, von Heimburg D. Pathogenic role of interleukin-6 in the development of

sepsis. Part II: Significance of anti-interleukin-6 and anti-soluble interleukin-6 receptor-alpha

antibodies in a standardized murine contact burn model. Crit Care Med 2003 May;31(5):1495-

501.

Pallua N, von Heimburg D. Pathogenic role of interleukin-6 in the development of sepsis. Part I:

Study in a standardized contact burn murine model. Crit Care Med 2003 May;31(5):1490-4.

Patierno S, Raybould HE, Sternini C. Abdominal surgery induces mu opioid receptor endocytosis

in enteric neurons of the guinea-pig ileum. Neuroscience 2004;123(1):101-9.

Paul R, Koedel U, Winkler F, Kieseier BC, Fontana A, Kopf M, et al. Lack of IL-6 augments

inflammatory response but decreases vascular permeability in bacterial meningitis. Brain 2003

Aug;126(Pt 8):1873-82.

Pavlov VA, Ochani M, Gallowitsch-Puerta M, Ochani K, Huston JM, Czura CJ, et al. Central

muscarinic cholinergic regulation of the systemic inflammatory response during endotoxemia.

Proc Natl Acad Sci U S A 2006 Mar 28;103(13):5219-23.

Pavlov VA, Ochani M, Yang LH, Gallowitsch-Puerta M, Ochani K, Lin X, et al. Selective alpha7-

nicotinic acetylcholine receptor agonist GTS-21 improves survival in murine endotoxemia and

severe sepsis. Crit Care Med 2007 Apr;35(4):1139-44.

Pavlov VA, Parrish WR, Rosas-Ballina M, Ochani M, Puerta M, Ochani K, et al. Brain

acetylcholinesterase activity controls systemic cytokine levels through the cholinergic anti-

inflammatory pathway. Brain Behav Immun 2009 Jan;23(1):41-5.

Pavlov VA, Tracey KJ. Neural circuitry and immunity. Immunol Res 2015 Dec;63(1-3):38-57.

Pavlov VA, Wang H, Czura CJ, Friedman SG, Tracey KJ. The cholinergic anti-inflammatory

pathway: a missing link in neuroimmunomodulation. Mol Med 2003 May;9(5-8):125-34.

References

236

Pavlov VA. Cholinergic modulation of inflammation. Int J Clin Exp Med 2008;1(3):203-12.

Pene F, it-Oufella H, Taccone FS, Monneret G, Sharshar T, Tamion F, et al. Insights and limits of

translational research in critical care medicine. Ann Intensive Care 2015 Apr 30;5:8. doi:

10.1186/s13613-015-0050-3. eCollection;%2015.:8-0050.

Persson EK, Scott CL, Mowat AM, Agace WW. Dendritic cell subsets in the intestinal lamina

propria: ontogeny and function. Eur J Immunol 2013 Dec;43(12):3098-107.

Peterson CY, Costantini TW, Loomis WH, Putnam JG, Wolf P, Bansal V, et al. Toll-like receptor-4

mediates intestinal barrier breakdown after thermal injury. Surg Infect (Larchmt ) 2010

Apr;11(2):137-44.

Piliponsky AM, Chen CC, Nishimura T, Metz M, Rios EJ, Dobner PR, et al. Neurotensin increases

mortality and mast cells reduce neurotensin levels in a mouse model of sepsis. Nat Med 2008

Apr;14(4):392-8.

Pinsky MR, Vincent JL, Deviere J, Alegre M, Kahn RJ, Dupont E. Serum cytokine levels in human

septic shock. Relation to multiple-system organ failure and mortality. Chest 1993

Feb;103(2):565-75.

Platt AM, Mowat AM. Mucosal macrophages and the regulation of immune responses in the

intestine. Immunol Lett 2008 Aug 15;119(1-2):22-31.

Platzer C, Meisel C, Vogt K, Platzer M, Volk HD. Up-regulation of monocytic IL-10 by tumor

necrosis factor-alpha and cAMP elevating drugs. Int Immunol 1995 Apr;7(4):517-23.

Plourde V, Wong HC, Walsh JH, Raybould HE, Tache Y. CGRP antagonists and capsaicin on celiac

ganglia partly prevent postoperative gastric ileus. Peptides 1993 Nov;14(6):1225-9.

Poehlmann H, Schefold JC, Zuckermann-Becker H, Volk HD, Meisel C. Phenotype changes and

impaired function of dendritic cell subsets in patients with sepsis: a prospective observational

analysis. Crit Care 2009;13(4):R119.

Ponda PP, Mayer L. Mucosal epithelium in health and disease. Curr Mol Med 2005

Sep;5(6):549-56.

Rahman SH, Ammori BJ, Holmfield J, Larvin M, McMahon MJ. Intestinal hypoperfusion

contributes to gut barrier failure in severe acute pancreatitis. J Gastrointest Surg 2003

Jan;7(1):26-35.

Ranieri VM, Thompson BT, Barie PS, Dhainaut JF, Douglas IS, Finfer S, et al. Drotrecogin alfa

(activated) in adults with septic shock. N Engl J Med 2012 May 31;366(22):2055-64.

Reddy BS, MacFie J, Gatt M, farlane-Smith L, Bitzopoulou K, Snelling AM. Commensal bacteria

do translocate across the intestinal barrier in surgical patients. Clin Nutr 2007 Apr;26(2):208-

15.

References

237

Remick DG, Ward PA. Evaluation of endotoxin models for the study of sepsis. Shock 2005

Dec;24 Suppl 1:7-11.:7-11.

Riedemann NC, Neff TA, Guo RF, Bernacki KD, Laudes IJ, Sarma JV, et al. Protective effects of IL-

6 blockade in sepsis are linked to reduced C5a receptor expression. J Immunol 2003 Jan

1;170(1):503-7.

Riley TP, Neal-McKinney JM, Buelow DR, Konkel ME, Simasko SM. Capsaicin-sensitive vagal

afferent neurons contribute to the detection of pathogenic bacterial colonization in the gut. J

Neuroimmunol 2013 Apr 15;257(1-2):36-45.

Rittirsch D, Hoesel LM, Ward PA. The disconnect between animal models of sepsis and human

sepsis. J Leukoc Biol 2007 Jan;81(1):137-43.

Rittirsch D, Huber-Lang MS, Flierl MA, Ward PA. Immunodesign of experimental sepsis by cecal

ligation and puncture. Nat Protoc 2009;4(1):31-6.

Rivers EP, Jaehne AK, Nguyen HB, Papamatheakis DG, Singer D, Yang JJ, et al. Early biomarker

activity in severe sepsis and septic shock and a contemporary review of immunotherapy trials:

not a time to give up, but to give it earlier. Shock 2013 Feb;39(2):127-37.

Rong W, Hillsley K, Davis JB, Hicks G, Winchester WJ, Grundy D. Jejunal afferent nerve

sensitivity in wild-type and TRPV1 knockout mice. J Physiol 2004 Nov 1;560(Pt 3):867-81.

Rosas-Ballina M, Goldstein RS, Gallowitsch-Puerta M, Yang L, Valdes-Ferrer SI, Patel NB, et al.

The selective alpha7 agonist GTS-21 attenuates cytokine production in human whole blood and

human monocytes activated by ligands for TLR2, TLR3, TLR4, TLR9, and RAGE. Mol Med 2009

Jul;15(7-8):195-202.

Rosas-Ballina M, Olofsson PS, Ochani M, Valdes-Ferrer SI, Levine YA, Reardon C, et al.

Acetylcholine-synthesizing T cells relay neural signals in a vagus nerve circuit. Science 2011 Oct

7;334(6052):98-101.

Rosas-Ballina M, Tracey KJ. Cholinergic control of inflammation. J Intern Med 2009

Jun;265(6):663-79.

Rotstein OD. Pathogenesis of multiple organ dysfunction syndrome: gut origin, protection, and

decontamination. Surg Infect (Larchmt ) 2000;1(3):217-23.

Rumio C, Besusso D, Arnaboldi F, Palazzo M, Selleri S, Gariboldi S, et al. Activation of smooth

muscle and myenteric plexus cells of jejunum via Toll-like receptor 4. J Cell Physiol 2006

Jul;208(1):47-54.

Ruyssers NE, De Winter BY, De Man JG, Loukas A, Pearson MS, Weinstock JV, et al. Therapeutic

potential of helminth soluble proteins in TNBS-induced colitis in mice. Inflamm Bowel Dis 2009

Apr;15(4):491-500.

References

238

Ruyssers NE, De Winter BY, De Man JG, Ruyssers ND, Van Gils AJ, Loukas A, et al. Schistosoma

mansoni proteins attenuate gastrointestinal motility disturbances during experimental colitis in

mice. World J Gastroenterol 2010 Feb 14;16(6):703-12.

Samel S, Lanig S, Lux A, Keese M, Gretz N, Nichterlein T, et al. The gut origin of bacterial

pancreatic infection during acute experimental pancreatitis in rats. Pancreatology

2002;2(5):449-55.

Schafer M, Imai Y, Uhl GR, Stein C. Inflammation enhances peripheral mu-opioid receptor-

mediated analgesia, but not mu-opioid receptor transcription in dorsal root ganglia. Eur J

Pharmacol 1995 Jun 12;279(2-3):165-9.

Scheller J, Chalaris A, Schmidt-Arras D, Rose-John S. The pro- and anti-inflammatory properties

of the cytokine interleukin-6. Biochim Biophys Acta 2011 May;1813(5):878-88.

Schindler R, Mancilla J, Endres S, Ghorbani R, Clark SC, Dinarello CA. Correlations and

interactions in the production of interleukin-6 (IL-6), IL-1, and tumor necrosis factor (TNF) in

human blood mononuclear cells: IL-6 suppresses IL-1 and TNF. Blood 1990 Jan 1;75(1):40-7.

Schlitzer A, McGovern N, Teo P, Zelante T, Atarashi K, Low D, et al. IRF4 transcription factor-

dependent CD11b+ dendritic cells in human and mouse control mucosal IL-17 cytokine

responses. Immunity 2013 May 23;38(5):970-83.

Schmidt J, Stoffels B, Chanthaphavong RS, Buchholz BM, Nakao A, Bauer AJ. Differential

molecular and cellular immune mechanisms of postoperative and LPS-induced ileus in mice and

rats. Cytokine 2012 Jul;59(1):49-58.

Schulte A, Lichtenstern C, Henrich M, Weigand MA, Uhle F. Loss of vagal tone aggravates

systemic inflammation and cardiac impairment in endotoxemic rats. J Surg Res 2014 May

15;188(2):480-8.

Schulz O, Jaensson E, Persson EK, Liu X, Worbs T, Agace WW, et al. Intestinal CD103+, but not

CX3CR1+, antigen sampling cells migrate in lymph and serve classical dendritic cell functions. J

Exp Med 2009 Dec 21;206(13):3101-14.

Scirocco A, Matarrese P, Petitta C, Cicenia A, Ascione B, Mannironi C, et al. Exposure of Toll-like

receptors 4 to bacterial lipopolysaccharide (LPS) impairs human colonic smooth muscle cell

function. J Cell Physiol 2010 May;223(2):442-50.

Scott CL, Aumeunier AM, Mowat AM. Intestinal CD103+ dendritic cells: master regulators of

tolerance? Trends Immunol 2011 Sep;32(9):412-9.

Scumpia PO, Delano MJ, Kelly KM, O'Malley KA, Efron PA, McAuliffe PF, et al. Increased natural

CD4+CD25+ regulatory T cells and their suppressor activity do not contribute to mortality in

murine polymicrobial sepsis. J Immunol 2006 Dec 1;177(11):7943-9.

Scumpia PO, Moldawer LL. Biology of interleukin-10 and its regulatory roles in sepsis

syndromes. Crit Care Med 2005 Dec;33(12 Suppl):S468-S471.

References

239

Seeley EJ, Sutherland RE, Kim SS, Wolters PJ. Systemic mast cell degranulation increases

mortality during polymicrobial septic peritonitis in mice. J Leukoc Biol 2011 Sep;90(3):591-7.

Seerden TC, De Man JG, Holzer P, Van den Bossche RM, Herman AG, Pelckmans PA, et al.

Experimental pancreatitis disturbs gastrointestinal and colonic motility in mice: effect of the

prokinetic agent tegaserod. Neurogastroenterol Motil 2007 Oct;19(10):856-64.

Semmler A, Hermann S, Mormann F, Weberpals M, Paxian SA, Okulla T, et al. Sepsis causes

neuroinflammation and concomitant decrease of cerebral metabolism. J Neuroinflammation

2008 Sep 15;5:38. doi: 10.1186/1742-2094-5-38.:38-5.

Seng P, Drancourt M, Gouriet F, La Scola B, Fournier PE, Rolain JM, et al. Ongoing revolution in

bacteriology: routine identification of bacteria by matrix-assisted laser desorption ionization

time-of-flight mass spectrometry. Clin Infect Dis 2009 Aug 15;49(4):543-51.

Senthil M, Brown M, Xu DZ, Lu Q, Feketeova E, Deitch EA. Gut-lymph hypothesis of systemic

inflammatory response syndrome/multiple-organ dysfunction syndrome: validating studies in a

porcine model. J Trauma 2006 May;60(5):958-65.

Setiawan T, Metwali A, Blum AM, Ince MN, Urban JF, Jr., Elliott DE, et al. Heligmosomoides

polygyrus promotes regulatory T-cell cytokine production in the murine normal distal intestine.

Infect Immun 2007 Sep;75(9):4655-63.

Sharkey KA. Emerging roles for enteric glia in gastrointestinal disorders. J Clin Invest 2015 Mar

2;125(3):918-25.

Sharma A, Yang WL, Matsuo S, Wang P. Differential alterations of tissue T-cell subsets after

sepsis. Immunol Lett 2015 Nov;168(1):41-50.

Shen L, Weber CR, Raleigh DR, Yu D, Turner JR. Tight junction pore and leak pathways: a

dynamic duo. Annu Rev Physiol 2011;73:283-309. doi: 10.1146/annurev-physiol-012110-

142150.:283-309.

Sheridan BS, Lefrancois L. Isolation of mouse lymphocytes from small intestine tissues. Curr

Protoc Immunol 2012 Nov;Chapter 3:Unit3.19. doi: 10.1002/0471142735.im0319s99.:Unit3.

Silva E, de Figueiredo LF, Colombari F. Prowess-shock trial: a protocol overview and

perspectives. Shock 2010 Sep;34 Suppl 1:48-53. doi: 10.1097/SHK.0b013e3181e7e97b.:48-53.

Singleton KD, Wischmeyer PE. Distance of cecum ligated influences mortality, tumor necrosis

factor-alpha and interleukin-6 expression following cecal ligation and puncture in the rat. Eur

Surg Res 2003 Nov;35(6):486-91.

Smith PD, Ochsenbauer-Jambor C, Smythies LE. Intestinal macrophages: unique effector cells of

the innate immune system. Immunol Rev 2005 Aug;206:149-59.:149-59.

Snoek SA, Dhawan S, Van Bree SH, Cailotto C, van Diest SA, Duarte JM, et al. Mast cells trigger

epithelial barrier dysfunction, bacterial translocation and postoperative ileus in a mouse

model. Neurogastroenterol Motil 2012 Feb;24(2):172-84, e91.

References

240

Sommer C, Kress M. Recent findings on how proinflammatory cytokines cause pain: peripheral

mechanisms in inflammatory and neuropathic hyperalgesia. Neurosci Lett 2004 May 6;361(1-

3):184-7.

Song J, Yin J, Sallam HS, Bai T, Chen Y, Chen JD. Electroacupuncture improves burn-induced

impairment in gastric motility mediated via the vagal mechanism in rats. Neurogastroenterol

Motil 2013 Oct;25(10):807-e635.

Song M, Kellum JA. Interleukin-6. Crit Care Med 2005 Dec;33(12 Suppl):S463-S465.

Sonnenberg GF, Artis D. Innate lymphoid cells in the initiation, regulation and resolution of

inflammation. Nat Med 2015 Jul;21(7):698-708.

Souza-Fonseca-Guimaraes F, Parlato M, Fitting C, Cavaillon JM, dib-Conquy M. NK cell tolerance

to TLR agonists mediated by regulatory T cells after polymicrobial sepsis. J Immunol 2012 Jun

15;188(12):5850-8.

Sprung CL, Annane D, Keh D, Moreno R, Singer M, Freivogel K, et al. Hydrocortisone therapy for

patients with septic shock. N Engl J Med 2008 Jan 10;358(2):111-24.

Stearns-Kurosawa DJ, Osuchowski MF, Valentine C, Kurosawa S, Remick DG. The pathogenesis

of sepsis. Annu Rev Pathol 2011;6:19-48. doi: 10.1146/annurev-pathol-011110-130327.:19-48.

Stein C. Opioid Receptors. Annu Rev Med 2016 Jan 14;67:433-51. doi: 10.1146/annurev-med-

062613-093100. Epub;%2015 Aug 26.:433-51.

Stein C. The control of pain in peripheral tissue by opioids. N Engl J Med 1995 Jun

22;332(25):1685-90.

Sternini C, Patierno S, Selmer IS, Kirchgessner A. The opioid system in the gastrointestinal tract.

Neurogastroenterol Motil 2004 Oct;16 Suppl 2:3-16.:3-16.

Stoffels B, Hupa KJ, Snoek SA, van BS, Stein K, Schwandt T, et al. Post-operative Ileus Involves

Interleukin-1 Receptor Signaling in Enteric Glia. Gastroenterology 2013 Sep 22;(13):10.

Stoutenbeek CP, Van Saene HK, Miranda DR, Zandstra DF. A new technique of infection

prevention in the intensive care unit by selective decontamination of the digestive tract. Acta

Anaesthesiol Belg 1983 Sep;34(3):209-21.

Stoutenbeek CP, Van Saene HK, Miranda DR, Zandstra DF. The effect of selective

decontamination of the digestive tract on colonisation and infection rate in multiple trauma

patients. Intensive Care Med 1984;10(4):185-92.

Suckow SK, Caudle RM. NMDA receptor subunit expression and PAR2 receptor activation in

colospinal afferent neurons (CANs) during inflammation induced visceral hypersensitivity. Mol

Pain 2009 Sep 22;5:54. doi: 10.1186/1744-8069-5-54.:54-5.

References

241

Sutherland RE, Olsen JS, McKinstry A, Villalta SA, Wolters PJ. Mast cell IL-6 improves survival

from Klebsiella pneumonia and sepsis by enhancing neutrophil killing. J Immunol 2008 Oct

15;181(8):5598-605.

Suzuki T, Yoshinaga N, Tanabe S. Interleukin-6 (IL-6) regulates claudin-2 expression and tight

junction permeability in intestinal epithelium. J Biol Chem 2011 Sep 9;286(36):31263-71.

Szurszewski JH, Ermilov LG, Miller SM. Prevertebral ganglia and intestinofugal afferent

neurones. Gut 2002 Jul;51 Suppl 1:i6-10.:i6-10.

Takeuchi O, Akira S. Pattern recognition receptors and inflammation. Cell 2010

Mar;%19;140(6):805-20.

Tazuke Y, Drongowski RA, Teitelbaum DH, Coran AG. Interleukin-6 changes tight junction

permeability and intracellular phospholipid content in a human enterocyte cell culture model.

Pediatr Surg Int 2003 Jul;19(5):321-5.

The FO, Boeckxstaens GE, Snoek SA, Cash JL, Bennink R, Larosa GJ, et al. Activation of the

cholinergic anti-inflammatory pathway ameliorates postoperative ileus in mice.

Gastroenterology 2007 Oct;133(4):1219-28.

The FO, Bennink RJ, Ankum WM, Buist MR, Busch OR, Gouma DJ, et al. Intestinal handling-

induced mast cell activation and inflammation in human postoperative ileus. Gut 2008

Jan;57(1):33-40.

The FO, Buist MR, Lei A, Bennink RJ, Hofland J, van den Wijngaard RM, et al. The role of mast

cell stabilization in treatment of postoperative ileus: a pilot study. Am J Gastroenterol 2009

Sep;104(9):2257-66.

Tilg H, Trehu E, Atkins MB, Dinarello CA, Mier JW. Interleukin-6 (IL-6) as an anti-inflammatory

cytokine: induction of circulating IL-1 receptor antagonist and soluble tumor necrosis factor

receptor p55. Blood 1994 Jan 1;83(1):113-8.

Toscano MG, Ganea D, Gamero AM. Cecal ligation puncture procedure. J Vis Exp 2011 May

7;(51):2860.

Tracey KJ. Immune cells exploit a neural circuit to enter the CNS. Cell 2012 Feb 3;148(3):392-4.

Tracey KJ. Physiology and immunology of the cholinergic antiinflammatory pathway. J Clin

Invest 2007 Feb;117(2):289-96.

Tracey KJ. The inflammatory reflex. Nature 2002 Dec;%19-26;420(6917):853-9.

Tsukamoto T, Chanthaphavong RS, Pape HC. Current theories on the pathophysiology of

multiple organ failure after trauma. Injury 2010 Jan;41(1):21-6.

Turler A, Schnurr C, Nakao A, Togel S, Moore BA, Murase N, et al. Endogenous endotoxin

participates in causing a panenteric inflammatory ileus after colonic surgery. Ann Surg 2007

May;245(5):734-44.

References

242

Turler A, Schwarz NT, Turler E, Kalff JC, Bauer AJ. MCP-1 causes leukocyte recruitment and

subsequently endotoxemic ileus in rat. Am J Physiol Gastrointest Liver Physiol 2002

Jan;282(1):G145-G155.

Turner JR. Intestinal mucosal barrier function in health and disease. Nat Rev Immunol 2009

Nov;9(11):799-809.

Ulloa L, Tracey KJ. The "cytokine profile": a code for sepsis. Trends Mol Med 2005 Feb;11(2):56-

63.

Unsinger J, McGlynn M, Kasten KR, Hoekzema AS, Watanabe E, Muenzer JT, et al. IL-7 promotes

T cell viability, trafficking, and functionality and improves survival in sepsis. J Immunol 2010 Apr

1;184(7):3768-79.

Valdez-Morales E, Guerrero-Alba R, Ochoa-Cortes F, Benson J, Spreadbury I, Hurlbut D, et al.

Release of endogenous opioids during a chronic IBD model suppresses the excitability of

colonic DRG neurons. Neurogastroenterol Motil 2013 Jan;25(1):39-46.

Van Bree SH, Nemethova A, van Bovenkamp FS, Gomez-Pinilla P, Elbers L, Di Giovangiulio M, et

al. Novel method for studying postoperative ileus in mice. Int J Physiol Pathophysiol Pharmacol

2012;4(4):219-27.

Van Craenenbroeck AH, Van Ackeren K, Hoymans VY, Roeykens J, Verpooten GA, Vrints CJ, et

al. Acute exercise-induced response of monocyte subtypes in chronic heart and renal failure.

Mediators Inflamm 2014;2014:216534. doi: 10.1155/2014/216534. Epub;%2014 Dec

22.:216534.

Van den Broeck W, Derore A, Simoens P. Anatomy and nomenclature of murine lymph nodes:

Descriptive study and nomenclatory standardization in BALB/cAnNCrl mice. J Immunol

Methods 2006 May 30;312(1-2):12-9.

van der Zanden EP, Snoek SA, Heinsbroek SE, Stanisor OI, Verseijden C, Boeckxstaens GE, et al.

Vagus nerve activity augments intestinal macrophage phagocytosis via nicotinic acetylcholine

receptor alpha4beta2. Gastroenterology 2009 Sep;137(3):1029-39, 1039.

Van Itallie CM, Anderson JM. Claudins and epithelial paracellular transport. Annu Rev Physiol

2006;68:403-29.:403-29.

Van Nassauw L, Adriaensen D, Timmermans JP. The bidirectional communication between

neurons and mast cells within the gastrointestinal tract. Auton Neurosci 2007 Apr

30;133(1):91-103.

Vanneste G, Van NL, Kalfin R, Van C, I, Elinck E, Van CK, et al. Jejunal cholinergic, nitrergic, and soluble guanylate cyclase activity in postoperative ileus. Surgery 2008 Sep;144(3):410-26. Bansal V, Costantini T, Kroll L, Peterson C, Loomis W, Eliceiri B, et al. Traumatic brain injury and

intestinal dysfunction: uncovering the neuro-enteric axis. J Neurotrauma 2009 Aug;26(8):1353-

9.

References

243

van Westerloo DJ, Giebelen IA, Florquin S, Bruno MJ, Larosa GJ, Ulloa L, et al. The vagus nerve

and nicotinic receptors modulate experimental pancreatitis severity in mice. Gastroenterology

2006 May;130(6):1822-30.

Vanheel H, Farre R. Changes in gastrointestinal tract function and structure in functional

dyspepsia. Nat Rev Gastroenterol Hepatol 2013 Mar;10(3):142-9.

Varol C, Vallon-Eberhard A, Elinav E, Aychek T, Shapira Y, Luche H, et al. Intestinal lamina

propria dendritic cell subsets have different origin and functions. Immunity 2009 Sep

18;31(3):502-12.

Venet F, Chung CS, Kherouf H, Geeraert A, Malcus C, Poitevin F, et al. Increased circulating

regulatory T cells (CD4(+)CD25 (+)CD127 (-)) contribute to lymphocyte anergy in septic shock

patients. Intensive Care Med 2009 Apr;35(4):678-86.

Venet F, Filipe-Santos O, Lepape A, Malcus C, Poitevin-Later F, Grives A, et al. Decreased T-cell

repertoire diversity in sepsis: a preliminary study. Crit Care Med 2013 Jan;41(1):111-9.

Venet F, Lepape A, Monneret G. Clinical review: flow cytometry perspectives in the ICU - from

diagnosis of infection to monitoring of injury-induced immune dysfunctions. Crit Care 2011 Oct

7;15(5):231.

Venet F, Lukaszewicz AC, Payen D, Hotchkiss R, Monneret G. Monitoring the immune response

in sepsis: a rational approach to administration of immunoadjuvant therapies. Curr Opin

Immunol 2013 Aug;25(4):477-83.

Verma-Gandhu M, Verdu EF, Bercik P, Blennerhassett PA, Al-Mutawaly N, Ghia JE, et al. Visceral

pain perception is determined by the duration of colitis and associated neuropeptide

expression in the mouse. Gut 2007 Mar;56(3):358-64.

Vermeulen W, De Man JG, Pelckmans PA, De Winter BY. Neuroanatomy of lower

gastrointestinal pain disorders. World J Gastroenterol 2014 Jan 28;20(4):1005-20.

Vida G, Pena G, Deitch EA, Ulloa L. alpha7-cholinergic receptor mediates vagal induction of

splenic norepinephrine. J Immunol 2011 Apr 1;186(7):4340-6.

Vilz TO, Overhaus M, Stoffels B, Websky M, Kalff JC, Wehner S. Functional assessment of

intestinal motility and gut wall inflammation in rodents: analyses in a standardized model of

intestinal manipulation. J Vis Exp 2012 Sep 11;(67):4086.

Vincent JL, Bernard GR, Beale R, Doig C, Putensen C, Dhainaut JF, et al. Drotrecogin alfa

(activated) treatment in severe sepsis from the global open-label trial ENHANCE: further

evidence for survival and safety and implications for early treatment. Crit Care Med 2005

Oct;33(10):2266-77.

Vincent JL, Marshall JC, Namendys-Silva SA, Francois B, Martin-Loeches I, Lipman J, et al.

Assessment of the worldwide burden of critical illness: the intensive care over nations (ICON)

audit. Lancet Respir Med 2014 May;2(5):380-6.

References

244

Vyas D, Javadi P, Dipasco PJ, Buchman TG, Hotchkiss RS, Coopersmith CM. Early antibiotic

administration but not antibody therapy directed against IL-6 improves survival in septic mice

predicted to die on basis of high IL-6 levels. Am J Physiol Regul Integr Comp Physiol 2005

Oct;289(4):R1048-R1053.

Wan W, Wetmore L, Sorensen CM, Greenberg AH, Nance DM. Neural and biochemical

mediators of endotoxin and stress-induced c-fos expression in the rat brain. Brain Res Bull

1994;34(1):7-14.

Wang B, Glatzle J, Mueller MH, Kreis M, Enck P, Grundy D. Lipopolysaccharide-induced changes

in mesenteric afferent sensitivity of rat jejunum in vitro: role of prostaglandins. Am J Physiol

Gastrointest Liver Physiol 2005 Aug;289(2):G254-G260.

Wang F, Graham WV, Wang Y, Witkowski ED, Schwarz BT, Turner JR. Interferon-gamma and

tumor necrosis factor-alpha synergize to induce intestinal epithelial barrier dysfunction by up-

regulating myosin light chain kinase expression. Am J Pathol 2005 Feb;166(2):409-19.

Wang H, Yu M, Ochani M, Amella CA, Tanovic M, Susarla S, et al. Nicotinic acetylcholine

receptor alpha7 subunit is an essential regulator of inflammation. Nature 2003 Jan

23;421(6921):384-8.

Wang Q, Fang CH, Hasselgren PO. Intestinal permeability is reduced and IL-10 levels are

increased in septic IL-6 knockout mice. Am J Physiol Regul Integr Comp Physiol 2001

Sep;281(3):R1013-R1023.

Wang S, Shi P, Wang Y. TRPA1 ion channels in vagal afferent nerves contribute to ventilator-

induced lung injury in a rat model. Gen Physiol Biophys 2013 Sep;32(3):389-94.

Wang X, Soltesz V, Axelson J, Andersson R. Cholecystokinin increases small intestinal motility

and reduces enteric bacterial overgrowth and translocation in rats with surgically induced

acute liver failure. Digestion 1996;57(1):67-72.

Ward SM. Interstitial cells of Cajal in enteric neurotransmission. Gut 2000 Dec;47 Suppl 4:iv40-

3; discussion iv52.:iv40-iv43.

Weigmann B, Tubbe I, Seidel D, Nicolaev A, Becker C, Neurath MF. Isolation and subsequent

analysis of murine lamina propria mononuclear cells from colonic tissue. Nat Protoc

2007;2(10):2307-11.

Weinstein WM, Onderdonk AB, Bartlett JG, Gorbach SL. Experimental intra-abdominal

abscesses in rats: development of an experimental model. Infect Immun 1974 Dec;10(6):1250-

5.

Wichterman KA, Baue AE, Chaudry IH. Sepsis and septic shock--a review of laboratory models

and a proposal. J Surg Res 1980 Aug;29(2):189-201.

References

245

Willert RP, Woolf CJ, Hobson AR, Delaney C, Thompson DG, Aziz Q. The development and

maintenance of human visceral pain hypersensitivity is dependent on the N-methyl-D-

aspartate receptor. Gastroenterology 2004 Mar;126(3):683-92.

Wintersteller S, Hahnhaussen J, Kofler B, Emmanuel K. Molecular mediators of polymicrobial

sepsis. Front Biosci (Elite Ed) 2012 Jun 1;4:2584-604.:2584-604.

Wirthlin DJ, Cullen JJ, Spates ST, Conklin JL, Murray J, Caropreso DK, et al. Gastrointestinal

transit during endotoxemia: the role of nitric oxide. J Surg Res 1996 Feb 1;60(2):307-11.

Xiang H, Hu B, Li Z, Li J. Dexmedetomidine controls systemic cytokine levels through the

cholinergic anti-inflammatory pathway. Inflammation 2014 Oct;37(5):1763-70.

Yang R, Han X, Uchiyama T, Watkins SK, Yaguchi A, Delude RL, et al. IL-6 is essential for

development of gut barrier dysfunction after hemorrhagic shock and resuscitation in mice. Am

J Physiol Gastrointest Liver Physiol 2003 Sep;285(3):G621-G629.

Yasukawa H, Ohishi M, Mori H, Murakami M, Chinen T, Aki D, et al. IL-6 induces an anti-

inflammatory response in the absence of SOCS3 in macrophages. Nat Immunol 2003

Jun;4(6):551-6.

Yokoo H, Chiba S, Tomita K, Takashina M, Sagara H, Yagisita S, et al. Neurodegenerative

evidence in mice brains with cecal ligation and puncture-induced sepsis: preventive effect of

the free radical scavenger edaravone. PLoS One 2012;7(12):e51539.

Zador F, Wollemann M. Receptome: Interactions between three pain-related receptors or the

"Triumvirate" of cannabinoid, opioid and TRPV1 receptors. Pharmacol Res 2015 Dec;102:254-

63. doi: 10.1016/j.phrs.2015.10.015. Epub;%2015 Oct 28.:254-63.

Zhang Q, Schaffer M, Elde R, Stein C. Effects of neurotoxins and hindpaw inflammation on

opioid receptor immunoreactivities in dorsal root ganglia. Neuroscience 1998 Jul;85(1):281-91.

Zhang SC, Wang W, Ren WY, He BM, Zhou K, Zhu WN. Effect of cisapride on intestinal bacterial

and endotoxin translocation in cirrhosis. World J Gastroenterol 2003 Mar;9(3):534-8.

Zhang X, Mosser DM. Macrophage activation by endogenous danger signals. J Pathol 2008

Jan;214(2):161-78.

Zhou L, Zhang Q, Stein C, Schafer M. Contribution of opioid receptors on primary afferent

versus sympathetic neurons to peripheral opioid analgesia. J Pharmacol Exp Ther 1998

Aug;286(2):1000-6.

Zhuo M. Neural Mechanisms Underlying Anxiety-Chronic Pain Interactions. Trends Neurosci

2016 Mar;39(3):136-45.

Zittel TT, Reddy SN, Plourde V, Raybould HE. Role of spinal afferents and calcitonin gene-

related peptide in the postoperative gastric ileus in anesthetized rats. Ann Surg 1994

Jan;219(1):79-87.

References

246

Zittel TT, Rothenhofer I, Meyer JH, Raybould HE. Small intestinal capsaicin-sensitive afferents

mediate feedback inhibition of gastric emptying in rats. Am J Physiol 1994 Dec;267(6 Pt

1):G1142-G1145.

Zou Y, Tao T, Tian Y, Zhu J, Cao L, Deng X, et al. Ginsenoside Rg1 improves survival in a murine

model of polymicrobial sepsis by suppressing the inflammatory response and apoptosis of

lymphocytes. J Surg Res 2013 Feb 26;(13):10.

Zuo DC, Choi S, Shahi PK, Kim MY, Park CG, Kim YD, et al. Inhibition of pacemaker activity in

interstitial cells of Cajal by LPS via NF-kappaB and MAP kinase. World J Gastroenterol 2013 Feb

28;19(8):1210-8.

Scientific Curriculum Vitae


248

Scientific Curriculum Vitae Sara Nullens

Particulars

Name Sara Nullens

Date of Birth February 13th, 1987

Place of Birth Rumst, Belgium

Nationality Belgian

Marital status living together

Education

2012 – 2016 PhD in Medical Sciences

Laboratory for Experimental Medicine and Pediatrics – Gastroenterology

University of Antwerp, Antwerp, Belgium

Dissertation: Characterization and modulation of the enteric neuroimmune environment

during sepsis-induced ileus.

(Laboratory for Experimental Medicine and Peadiatrics; University of Antwerp, Belgium;

Promotors: Prof. dr. Benedicte De Winter & Prof. dr. Sven Francque)

2008 – 2012 Master of Medicine – summa cum laude, 3rd prize of the Jury


Thesis: Effect of venom specific immunotherapy on histamine degranulation in basophils of

wasp allergic patients (Laboratory of Immunology, Allergology and Rheumatology; University of

Antwerp, Belgium; promotor: prof. dr. Didier Ebo)

2005 – 2008 Bachelor of Medicine – magna cum laude


1999 – 2005 Latin – Mathematics – Science

Sint-Ritacollege, Kontich, Belgium


249

Research Experience – Participation in Research (National and Abroad)

LEMP – pre-doctoral lab experience (2008-2012)

Experiments in the field of organ baths, in vitro colonic peristalsis measurements and

TNBS-induced colitis, resulting in the publication of two papers in peer-reviewed

journals.

Laboratory of Immunology and Rheumatology – masterthesis (2008-2012)

Experience with flow cytometry, intracellular staining and the basophil activation test,

resulting in the publication of two papers in peer-reviewed journals.

Exchange student at the Royal Children’s Hospital (Melbourne) under the

supervision of Prof. T. Catto-Smith (pediatric gastroenterology)(March 2011)

Research Trainee at Mayo Clinic under the supervision of Prof. M. Camilleri (2011)

A two month research stay conducted at the Clinical Enteric Neuroscience

Translational and Epidemiological Research unit (C.E.N.T.E.R) at Mayo Clinic

(Rochester, USA) under the supervision of Professor Michael Camilleri, resulting in the

publication of three papers in peer-reviewed journals.

Topics:

o Study of the enteroendocrine and neuronal mechanisms in the

pathophysiology of acute infectious diarrhea

o Study on the regional (colonic) transit characteristics in patients with

dyssynergic defecation and slow transit constipation

Research stay at the Laboratory of Biomedical Sciences under the supervision of

Prof. D. Grundy (2013).

A two week research stay to familiarize with the isolation and recordings of jejunal

and colonic afferent nerves in the mouse at the Laboratory of Biomedical Sciences

(Sheffield, UK) under the supervision of Professor David Grundy

– implementation of the afferent recording technique in our laboratory in Antwerp,

Belgium, of which two papers are currently in preparation.

Additional certificates

Diploma of Electrocardiography (June 2009) - summa cum laude

United States Medical Licensing Examination Step 1 (March 2014)

United States Medical Licensing Examination Step 2 Clinical Knowledge (October 2014)

United States Medical Licensing Examination Step 2 Clinical Skills (May 2016)

FELASA type C certificate (Federation of European Laboratory Animal Science

Associations) – principle investigator in conducting animal research (October 2012)


250

Grants & Scientific Awards

Flemish Fund for Scientific Research (FWO) Aspirant Mandate (2012) & Aspirant

Renewal (2014) – Grant number 11G7415N

Flemish Society of Gastroenterology (VVGE) Travel Grant 2013

International Sepsisforum 2014 ‘Best Abstract Award’ at the 2014 Sepsis Meeting in

Paris, France.

Flemish Society of Gastroenterology (VVGE) Travel Grant 2015

Belgian Week of Gastroenterology Research Grant in basic research 2015

Distinguished Neurogastroenterology and Motility Abstract Award at the 2015 DDW

meeting in Washington DC, USA.

Lionel Bueno Award for Best Young Investigator Basic Science Presentation at the

2015 NeuroGASTRO meeting in Istanbul, Turkey.

2015 Rosa Blanckaert Award by the University of Antwerp (Aanmoedigingsbeurs voor

jonge vorsers).

Travel Grant for Young Investigators in Basic Science at the 2015 United European

Gastroenterology Week in Barcelona, Spain

Elected contestant for the upcoming 2016 Neurogastroenterology and Motility Young

Investigator poster prize at the upcoming Digestive Disease Week in San Diego.

Elected Young Investigator at the upcoming 2016 Young Investigator Forum at the

Federal Neurogastroenterology and Motility / American Neurogastroenterology and

Motility Meeting in San Fransisco (Top 15 young investigators elected around the

globe).


251

List of publications

Papers related to the dissertation

1. Nullens S, Staessens M, Peleman C, Schrijvers DM, Malhotra-Kumar S, Francque S, Matteoli G, Boeckxstaens GE, De Man JG, De Winter BY. Effect of GTS-21, a selective alpha7 nicotinic acetylcholine receptor agonist, on CLP-induced inflammatory, gastrointestinal motility and colonic permeability changes. Shock. 2016 Apr;45(4):450-9. doi: 10.1097/SHK.0000000000000519.

2. Nullens S, Staessens M, Peleman C, Plaeke P, Malhotra-Kumar S, Francque S, De Man JG, De Winter BY. Beneficial effects of anti-interleukin-6 antibodies on impaired gastrointestinal motility, inflammation and increased colonic permeability in a murine model of sepsis are most pronounced when administered in a preventive setup. PLoS One. 2016 Apr 4;11(4):e0152914. doi: 10.1371/journal.pone.0152914.

3. Nullens S, Deiteren A, Jiang W, Keating C, Ceuleers H, Francque SF, Grundy D, De Man

JG, De Winter BY. In vitro recording of mesenteric afferent nerve activity in mouse jejunal and colonic segments (Accepted for publication, Journal of Visualized Experiments, May 2016).

4. Nullens S, Deleye S, Deiteren A, De Man JG, Jiang W, Keating C, Francque SF, Staelens

S, De Winter BY. Brain-gut signaling in a mouse model of polymicrobial abdominal sepsis (Submitted to Annals of Surgery, July 2016).

5. Nullens S, De Man JG, Bridts C, Ebo DG, Francque SF, De Winter BY. Identifying new

therapeutic possibilities for sepsis research with a focus on the gastrointestinal tract: a characterization study of the cecal ligation and puncture model (Submitted to Journal of Leukocyte Biology, July 2016).

6. Nullens S, Staessens M, Peleman C, Jorens P, Francque SF, De Man JG, De Winter BY.

Interplay between the gut and the immune system during sepsis – from characterizing the neuroimmune environment to the development of new therapeutic agents (In preparation).

7. Nullens S, Plaeke P, Dua C, Hens S, Francque SF, De Man JG, De Winter BY. Impaired permeability of the gastrointestinal tract as a new therapeutic point of action in sepsis (In preparation).


252

Papers unrelated to the dissertation

1. Deiteren A, De Winter BY, Nullens S, Pelckmans PA, De Man JG. Role of tachykinin receptors in the modulation of colonic peristaltic acitvity in mice. Eur J Pharmacol. 2011 Sep 30; 667(1-3):339-347.

2. Vermeulen W, De Man JG, Nullens S, Pelckmans PA, De Winter BY, Moreels TG. The use of colonoscopy to follow the inflammatory time course of TNBS colitis in rats. Acta Gastroenterol Belg. 2011;74(2):304-11.

3. Camilleri M, Nullens S, Nelsen T. Enteroendocrine and neuronal mechanisms in pathophysiology of acute infectious diarrhea. Dig Dis Sci. 2012 Jan;57(1):19-27.

4. Verweij MM, Sabato V, Nullens S, Bridts CH, De Clerck LS, Stevens WJ, Ebo DG. STAT5 in human basophils: IL-3 is required for its FcεRI-mediated phosphorylation. Cytometry B Clin Cytom. 2012 Mar;82(2):101-6.

5. Nullens S, Nelsen T, Camilleri M, Burton D, Eckert D, Iturrino J, Vazquez-Roque M, Zinsmeister AR. Regional colon transit in patients with dys-synergic defaecation or slow transit in patients with constipation. Gut 2012 Aug;61(8):1132-9.

6. Shin A, Camilleri M, Nadeau A, Nullens S, Rhee JC, Jeong ID, Burton D. Interpretation of overall colonic transit in defecation disorders in males and females. Neurogastroenterol Motil 2013;25(6):502-508

7. Nullens S, Sabato V, Faber M, Leysen J, Bridts C, De Clerck L, Falcone F, Maurer M, Ebo D. Basophilic histamine content and release during venom immunotherapy: insights by flow cytometry. Cytometry B Clin Cytom 2013; 84(3):173-8

8. Heylen M, Ruyssers NE, Nullens S, Schramm G, Pelckmans PA, Moreels TG, De Man JG, De Winter BY.Treatment with Egg Antigens of Schistosoma mansoni ameliorates experimental colitis in mice through a colonic T-cell-dependent mechanism. Inflamm Bowel Dis 2015; 21(1):48-59

9. Wolf M, Saenen J, Bories W, Miljoen H, Nullens S, Vrints C, Sarkozy A. Superior efficacy of pulmonary vein isolation with online contact force measurement persists after the learning period: a prospective case control study. J Interv Card Electrophysiol 2015;43(3):287-96.

10. Alpaerts K, Buckinx R, Cools N, Heylen M, Nullens S, Berneman Z, De Winter B, Adriaensen D, Van Nassauw L, Timmermans JP. Effect of schistosomiasis on CX3CR1-expressing mononuclear phagocytes in the ileum and mesenteric lymph nodes of the mouse. Neurogastroenterol Motil 2015; 27(11):1587-99.


253

Abstracts related to the dissertation

1. Nullens S, De Man JG, Pelckmans PA, De Winter BY. Impaired gastrointestinal motility in animal models of sepsis. Autumn meeting of the Belgian Society of Physiology and Pharmacology 2013; oral presentation.

2. Nullens S, Heylen M, Ruyssers NE, De Man JG, Pelckmans PA, Van Brussel I, Schrijvers D, De Winter BY. Systemic and local immune fingerprint during septic ileus induced by cecal ligation and puncture. Acta Gastroenterol Belg 2014;77(1):B05. 26th Belgian Week of Gastroenterology meeting 2014; oral presentation.

3. Nullens S, De Man JG, Pelckmans PA, De Winter BY. GTS-21, a selective α7 nicotinic acetylcholine receptor agonist, ameliorates septic ileus in a mouse cecal ligation and puncture model. Neurogastroenterol Motil 2014;26(Suppl 1):23 PP42. First Federation of Neurogastroenterology and Motility Meeting in Guangzhou, China, September 5th-7th 2014; poster presentation.

4. Nullens S, De Man JG, Pelckmans PA, De Winter BY. GTS-21, a selective α7 nicotinic

acetylcholine receptor agonist, ameliorates septic ileus in a mouse cecal ligation and puncture model. J Neurogastroenterol Motil 2014;20(4):Suppl 2 PP42. First Federation of Neurogastroenterology and Motility Meeting 2014, Guangzhou, China; poster presentation.

5. Nullens S, Schrijvers DM, De Man JG, Pelckmans PA, De Winter BY. In vitro effects of GTS-21, a selective alpha7 nicotinic acetylcholine receptor agonist, on cytokine release in bone marrow derived macrophages and conventional dendritic cells. Autumn Meeting of the Belgian Society of Physiology and Pharmacology 2014; oral presentation.

6. Nullens S, De Man JG, Pelckmans PA, De Winter BY. Effects of splenectomy and GTS-21, a selective α7 nicotinic acetylcholine receptor agonist, on the development of septic ileus in mice. Crit Care 2014; 18(Suppl 2):P40. International 2014 Sepsisforum, Paris, France; oral and poster presentation. Winner of the ‘Sepsis 2014 Best Abstract Award’.

7. Nullens S, Peleman C, Staessens M, Pelckmans PA, Lammens C, Malhotra-Kumar S, De Man JG, De Winter BY. Beneficial effects of the selective alpha7 nicotinic acetylcholine receptor agonist GTS-21 on sepsis-induced ileus and impaired mucosal barrier function. Acta Gastroenterol Belg 2015;78(1):B07. Belgian Week of Gastroenterology 2015; oral presentation. Winner of the ‘Belgian Week of Gastroenterology Research Grant in basic research’.

8. Nullens S, Staessens M, Peleman C, Pelckmans PA, Lammens C, Malhotra-Kumar S, De Man JG, De Winter BY. GTS-21, a selective alpha7 nicotinic acetylcholine receptor agonist, ameliorates impaired mucosal barrier function and inhibits bacterial translocation in a mouse cecal ligation and puncture model. Gastroenterol 2015; 148(4):S141-S142. Digestive Disease Week 2015; oral presentation. Winner of the


254

‘Distinguished Neurogastroenterology and Motility Abstract’.

9. Nullens S, Deiteren A, De Man JG, Francque S, Keating C, Jiang W, Grundy D, De Winter BY. The effect of CLP-induced sepsis on mesenteric afferent nerve activity in mice. Neurogastroenterol Motil 2015; 27 (Suppl 2): 22. Neurogastroenterology and Motility Meeting 2015; oral presentation. Winner of the Lionel Bueno Award for Best Young Investigator Basic Science Presentation.

10. Nullens S, Deiteren A, De Man JG, Francque S, Keating C, Jiang W, Grundy D, De Winter BY. Jejunal afferent nerve activity is impaired in a mouse model of abdominal polymicrobial sepsis. Poster presentation Onderzoeksdag van de Faculteit Geneeskunde en Gezondheidswetenschappen, University of Antwerp.

11. Nullens S, Staessens M, Peleman C, Francque S, Lammens C, Malhotra-Kumar S, De Man JG, De Winter BY. Administration of anti-mouse interleukin-6 antibodies ameliorates caecal ligation and puncture-induced gastrointestinal motility disturbances and colonic permeability changes. UEG Journal 2015; 3(55):P1015. United European Gastroenterology Week 2015; poster presentation.

12. Nullens S, Deleye S, De Man JG, Francque SM, Staelens S, De Winter BY. Activation status of the central nervous system and lumbar dorsal root ganglia in a mouse model of polymicrobial abdominal septic ileus. Digestive Disease Week 2016; poster presentation.

13. Nullens S, Staessens M, Peleman C, Plaeke P, Francque SM, Lammens C, Malhotra-Kumar S, De Man JG, De Winter BY. Effect of gastrointestinal barrier protection on sepsis-induced changes of intestinal motility, inflammation and colonic permeability. Digestive Disease Week 2016; poster presentation.

14. Nullens S, Peleman C, Staessens M, Plaeke P, Francque SM, Lammens C, Malhotra-Kumar S, De Man JG, De Winter BY. Therapeutic and curative administration of antibodies to interleukin-6 ameliorates attenuated gastrointestinal motility, colonic inflammation and impaired mucosal barrier function in a mouse model of polymicrobial sepsis. Digestive Disease Week 2016; poster presentation.

15. Nullens S, Peleman C, Staessens M, Plaeke P, Francque SM, Lammens C, Malhotra-Kumar S, De Man JG, De Winter BY. The effect of anti-interleukin-6 antibodies on gastrointestinal motility, colonic inflammation and mucosal barrier function during in a mouse model of caecal ligation and puncture induced septic ileus. Belgian Week of Gastroenterology 2016, oral presentation.

16. Nullens S, De Man JG, Francque SF, De Winter BY. Flow cytometric identification of immune players in the enteric tract during sepsis-induced impaired gastrointestinal motility (abstract to be presented as a poster presentation at the 2016 Federation of Neurogastroenterology and Motility Meeting (ANMS) in San Fransisco, USA)


255

17. Nullens S, Deiteren A, Deleye S, De Man JG, Francque SF, Staelens S, De Winter BY. The effect of prolonged abdominal sepsis on mesenteric nerve activity, dorsal root ganglia and the central nervous system in a murine model of cecal ligation and puncture induced sepsis (abstract to be presented as an oral presentation at the 2016 Federation of Neurogastroenterology and Motility Meeting (ANMS) in San Fransisco, USA)

Abstract unrelated to the dissertation

1. Vermeulen W, De Man J, Moreels T, Nullens S, Pelckmans P, De Winter B. Long-lasting visceral hypersensitivity following acute TNBS-colitis in rats. Acta Gastroenterol. Belg., 2010, 73 (1): B03. Belgian Week of Gastroenterology 2010; oral presentation.

2. Deiteren A, De Winter B, Nullens S, Pelckmans P, De Man J. Endogenous tachykinins modulate distension-induced peristalsis in the mouse colon mainly through activation of NK1 and NK2 but not NK3 receptors. Acta Gastroenterol. Belg., 2010, 73 (1): B11. Belgian Week of Gastroenterology 2010; oral presentation.

3. Vermeulen W, De Man JG, Nullens S, Pelckmans PA, De Winter BY, Moreels TG. Colonoscopy to assess the time course of inflammation after TNBS colitis in rats. Autumn meeting of the Belgian Society of Fundamental and Clinical Physiology and Pharmacology 2010; oral presentation.

4. Sabato V, Verweij M, Bridts C, Leysen J, Nullens S, De Clerck L, Ebo D. Wasp venom immunotherapy promotes the spontaneous expression and venom-induced upregulation of CD300a on human basophils. Allergy 2012; 67: 1-2. Annual congress of the European Academy of Allergy and Immunology 2012; oral presentation.

5. De Schepper H, Nullens S, Moreels T, Pelckmans P, Van Outryve M. Manometric and transrectal ultrasound characteristics in patients with coexisting fecal incontinence and constipation: a retrospective study. Gut 2012;61 (Suppl 3): A305. 20th United Gastroenterology Week 2012; poster presentation.

6. Shin A, Camilleri M, Nadeau A, Nullens S, Rhee J, Jeong I, Burton D. Characterization of anorectal functions and overall colonic transit in female and male patients with dyssynergic defecation. 2012 Annual congress of the American College of Gastroenterology; poster presentation.

7. Boita M, Nullens S, Sabato V, Faber M, Leysen J, Bridts CH, De Clerck LD, Falcone F, Maurer M, Ebo DG. Basophilic histamine content and release during venom immunotherapy: insights by flow cytometry. International Mast Cell and Basophil Meeting 2013; poster presentation.

8. Heylen M, Ruyssers NE, Nullens S, De Man JG, Schramm G, Pelckmans PA, Moreels TG, De Winter B. Helminth egg antigens and their protective effect on colitis in a murine


256

transfer model. Acta Gastroenterol Belg 2014;77(1): I03. 26th Belgian Week of Gastroenterology meeting 2014, oral presentation.

9. Heylen M, Ruyssers NE, Nullens S, De Man JG, Schrijvers D, Van Brussel I, Schramm G, Pelckmans PA, Moreels TG, De Winter BY. Schistosoma mansoni helminth egg antigens (SEA) and their protective effect on colitis in a murine transfer model. Gastroenterol 2014; 146(5):S639-S640. Digestive Disease Week 2014, oral presentation.

10. Alpaerts K, Buckinx R, Cools N, Heylen M, Nullens S, Berneman Z, De Winter BY, Van Nassauw L and Timmermans J.-P. Annual Dendritic Cell conference 2014; poster presentation.

11. Alpaerts K, Buckinx R, Cools N, Heylen M, Nullens S, Berneman Z, De Winter B, Van Nassauw L, Timmermans J. The effect of schistosomiasis on the distribution of CX3CR1-positive dendritic cells and mononuclear phagocytes in the ileum and mesenteric lymph nodes of the mouse. Acta Gastroenterol Belg 2015;78(1):B06. Belgian Week of Gastroenterology 2015; oral presentation.

12. Alpaerts K, Buckinx R, Cools N, Heylen M, Nullens S, Berneman Z, De Winter B, Van Nassauw L, Timmermans J-P. Schistosomiasis affects the distribution of CX3CR1+ dendritic cells and mononuclear phagocytes in the ileum and mesenteric lymph nodes of the mouse. Neurogastroenterol Motil 2015; 27 (Suppl 2): 40. Neurogastroenterology and Motility Meeting 2015; poster presentation.

13. Deiteren A, De Man JG, Keating C, Jiang W, Nullens S, Francque S, Grundy D, De Winter BY. Neurogastroenterology and Motility Meeting 2015; Neurogastroenterol Motil 2015; 27 (Suppl 2): 53. Neurogastroenterology and Motility Meeting 2015; poster presentation.

14. Van Spaendonk H, Nullens S, Ceuleers H, Schrijvers D, Francque SM, De Man JG, De Winter BY. The effect of a protease inhibitor in a chronic colitis transfer model. Digestive Disease Week 2016; poster presentation.

Dankwoord

Zij die mij kennen, weten dat ik dit hoofdstuk altijd als eerste ga lezen gedurende een

doctoraatsverdediging! En dus heb ik het bij deze ook als eerste hoofdstuk geschreven in mijn

eigenste thesis, reeds meer dan een jaar geleden.

Er waren velen die mij, met verbaasde blik en stem, gevraagd hebben waarom ik in godsnaam

ging doctoreren na mijn studies Geneeskunde, en niet meteen ging verder specialiseren in de

Gastro-enterologie & Hepatologie. En toen ik daar niet meteen een duidelijk antwoord op kon

geven, werden die mensen hun ergste vermoedens bevestigd…

Wel nu, ik geloof er ten stelligste in dat je als arts niet enkel ‘geneest’, maar ook ‘innoveert’ en

‘onderwijst’. Stilstaan is achteruitgaan. Het totaalpakket van arts-onderzoeker sprak mij aan, en

na de opleiding tot basisarts was het ideale moment dan ook aangebroken om in een labo te

kruipen en mezelf vier jaar te wijden aan het bestuderen van de rol van de maagdarmtractus

gedurende sepsis-gemedieerde ileus.

Mijn keuze voor de Gastro-enterologie was niet zo vanzelfsprekend. Het duurde maar liefst vier

jaar vooraleer ik ‘het licht’ zag, en dat was dan nog wel tijdens een duistere les betreffende

anale pathologie… Ik weet nog zeer goed dat ik bij mezelf dacht dat “iedereen nu eens goed

kon lachen met mij en mijn roeping”.

Het examen gastro ging nogal heel vlot… zo vlot dat Prof. Pelckmans mij aanspoorde om “eens

een kijkje te gaan nemen in het labo gastro”. En aldus ben ik erin gerold.

Prof. Paul Pelckmans, bij deze dan ook van harte bedankt om mij ‘op te pikken’ uit die grote

groep van 120 nerveuze vierdejaars studenten Geneeskunde. Mede dankzij uw steun kon ik

tevens twee maanden doorbrengen aan de Mayo Clinic voor een stage op de C.E.N.T.E.R. unit

onder supervisie van Prof. Michael Camilleri, één van de meest innemende artsen (en mensen)

die ik ooit mocht ontmoeten. Het heeft mij voor een stukje mee gevormd en mijn beeld van de

ideale arts veranderd.

Het is wel een eer om uw ‘laatste’ doctoraatsstudente te mogen zijn!

Alzo ging ik een kijkje nemen op het labo gastro in de T-blok op de campus. Blijkbaar werd dat

gerund door Prof. Benedicte De Winter, waarvan ik dacht dat ze enkel studentjes drilde op

gebied van medisch-technische vaardigheden. De ontvangst was hartelijk, en ik ben er dan

maar blijven plakken.

Benedicte, er zijn weinig mensen die zo gedreven en meticuleus werken, en tegelijkertijd zo

benaderbaar zijn. Bedankt om een ideale promotor te zijn gedurende de afgelopen vier jaar.

Het is een grote luxe om om 23u een manuscript in te sturen, en dat de volgende ochtend

reeds verbeterd in je mailbox terug te vinden. Je bent kritisch en tezelfdertijd optimistisch, een

ideale ingesteldheid om kwaliteitsvol onderzoek te verrichten. Als strong women was u voor

menig X-chromosoom een groot voorbeeld, en ik hoop later dan ook even vlot te kunnen

jongleren met mijn professionele en privé-leven.

258

Mijn eerste stapjes als student op het labo werd ik geduldig begeleid door Ing. Joris De Man.

Het labo gastro zou het labo gastro niet zijn zonder ‘onze Joris’! Iedereen is het er unaniem

over eens dat het labo mede groot is geworden door jouw inzet en kennis (en ook een beetje

door je oneindige voorraad aan wisselstukken!). Je kon geduldig luisteren en mijn innerlijke

rust doen weerkeren als ik aan het flippen was om welke reden dan ook, en uw sarcastisch

humoristisch trekje was hilarisch om te mogen aanschouwen, zeker als anderen er het

‘slachtoffer’ van waren!

Maar nog belangrijker dan dat ben je een fijn mens. Als iedereen zoals jou was, dan was de

wereld een betere plaats. Het lot was ons de laatste jaren niet zo gunstig gezind, hopelijk

kunnen we wat rust vinden. Ik ga u missen.

Benedicte en Joris, verander niet te veel aan het lab (behalve misschien de aankoop van een

paar Ussing chambers en een automatische carroussel voor de flowcytometer... en graag ook

nog een gekoelde centrifuge enzo), want het is er prima!

Prof. Sven Francque, bedankt om mijn werkstukjes steeds met vlijt na te lezen, en de

spreektaal om te vormen in deftige schrijftaal. Het aantal howevers is dankzij u drastisch

gereduceerd. Tot binnen twee jaar op de gastro!

Bij deze wens ik tevens de leden van de jury te bedanken om mijn werk door te lezen en ze van

kritische commentaren te voorzien. Prof. Nathalie Cools en Prof. Romain Lefebvre voor het

meticuleus nakijken van elke p-waarde. Prof. Jörg Kalff, many thanks for commenting on my

thesis and joining us for the public defense. Prof. Guy Hubens, voor het nalezen van dit werk,

het vinden van een geknipte opvolger voor mijn nalatenschap, en voor de goede zorgen voor

onze pa; Prof. Philippe Jorens voor de steun aan en het vertrouwen in mijn opvolger Philip.

Mijn eerste dagen op het labo als student verrichtte ik proefjes samen met Wim Vermeulen;

bedankt voor de toffe babbels terwijl we voor het eerst in het lab een coloscoop in een rat

staken! Nathalie, bedankt voor het shock-effect door mij op mijn eerste dag meteen tien

muizen te laten opofferen (slik).

Mijn bureaugenoten waren zeer belangrijk voor mij gedurende mijn doctoraat. Annemie, je

was zo’n beetje mijn voorganger (zowel in de States als in het lab), maar ik kan nooit tippen

aan jouw wetenschappelijke ingesteldheid en gedrevenheid. Veel succes daar Down Under, je

hebt er hard voor gewerkt en je hebt het zeker en vast verdiend!

Marthe, je nuchtere kijk op de dingen zette mij met beide voetjes op de grond. Bedankt om me

te helpen bij mijn eerste flowcytometrische ongemakjes, ook jij nog veel succes gewenst!

Johanna, je was een lieve aanwezigheid in onze bureau, veel succes met de oprichting van je

eigenste bedrijf!

Wilco, vanaf dag 1 hebben we lief en leed gedeeld gedurende onze doctoraten. En soms ging

dat eerder ver: we hebben samen onze wijsheidstanden laten trekken, en we gingen samen

naar de autokeuring! Bedankt om een luisterend oor te zijn en aanstekelijk optimistisch te zijn

259

gedurende de donkere momenten. De lever ligt ver van de darm, en tegelijk ook weer niet.

Veel succes op uw post-doc avonturen! We komen mekaar later sowieso nog tegen, in welke

toestand that I don’t know, maar ik kijk er alleszins naar uit!

Hannah en Hanne, de jonge dames van onze bureau. Het klikte meteen met jullie bij de start op

T2, en de sfeer in de bureau heeft er zeker niet onder geleden! Het was fijn om een beetje jullie

‘mentor/mama’ te zijn (dat was ik toch hé?), en verscheidene edities van onze Komen Eten zijn

nu al legendarisch door het optreden van Wolkje of de keuken-die-net-niet-in-brand-vloog. Ons

LA-San Diego tripje was zeker de moeite, ik heb immens veel respect voor de hoeveelheid eten

en drank (maar geen Tequila) die jullie kunnen binnenspelen of het gemak waarmee jullie op

elk openbaar vervoermiddel kunnen maffen. Jullie zijn waardige opvolsters van Annemie en

Marthe, en de resultaten zijn reeds navenant.

Aan Philip, mijn eigenste opvolger: ge zijt ne crème van ne chirugische vent, en ik geef ‘mijn’

CLP-model dan ook met het volste vertrouwen door aan jou. Doe dat goed! I’ll be watching

you! No pressure!

De gastro-assistenten aan de overkant: leuk om samen de Manama en congressen met jullie te

trotseren! Ook een bijzondere dank aan Wim Verlinden voor de humoristische momenten, en

aan Heiko De Schepper: ik heb mij goed geamuseerd als uw ‘persoonlijke’ co-assistent tijdens

je laatste jaar als assistent in opleiding, en nu heb je het toch maar mooi tot Prof. geschopt!

Goed bezig!

Veel praktisch werk werd mede verricht door ‘mijn’ Masterproefstudenten Geneeskunde,

Michael Staessens en Cédric Peleman. Veel succes met jullie verdere carrière. Tevens was het

ook fijn samenwerken met Mikhaïl Van Herck, Thomas Mangodt, Sebastiaan Hens en Cedric

Dua. Ook aan julie nog veel succes gewenst.

De dagelijkse sfeer op de werkvloer werd verzorgd door alle medewerkers en

doctoraatsstudenten uit maar liefst 3 labo’s, namelijk de fysio’s, farmaco’s en de collega’s van

het LEMP. Dankuwel aan Mandy om mij in tijden van hoofdpijn van een oneindige voorraad

frisse cola te voorzien, Miche om mij wegwijs te maken in de oneindige wondere wereld van de

dendritische cel, Lynn voor het aangename gezelschap tijdens de oneindige opofferingen in de

operatiezaal, Zarha voor de oneindige humor en vastberadenheid, en Dorien DM, Isabelle,

Carole, Cé, Hadise, Dries, Ammar, Bieke, Arthur, Lindsey, Leni, Guido, Wim, Gilles, Cor, Vincent,

Sonja en Maria’ke, Anne-Elise, Min, Katrien L, Katrien DM, Chrisje, Annie en Inn, Rachid, Ilse,

Angelika, Petra, Marleen, Lieve, Jessy, Annelies, Els, Stijn, Mikhaïl en Kristien. Paul, bedankt om

samen met mij mijn laatste wapenfeit op T2 uit te werken! Calcium ist alles!

Een speciale dankuwel aan Dorien Schrijvers voor het eeuwige enthousiasme en de harde

werklust (én voor het nalezen van mijn dankwoord!), het ga je goed op je nieuwe werkplek, en

aan Rita voor de immer vrolijke begroetingen (de hele dag door!) en het nooit aflatend

produceren van microscopische kunstwerkjes.

260

Het labo Microbiologie aan de Universiteit Antwerpen, in het bijzonder Prof. Surbhi Malhotra,

Christine Lammens en Mandy Andriessen wens ik te bedanken voor het verrichten van de

culturen.

Het Molecular Imaging Center Antwerp (MICA), in het bijzonder Prof. Steven Staelens, Steven

Deleye en Caroline Berghmans wens ik te bedanken voor de vlotte samenwerking, het zijn heel

schoon beeldjes geworden.

Ook buiten de grenzen van onze universiteit zijn er mensen die bedankt dienen te worden:

Prof. Gianluca Matteoli en Prof. Guy Boeckxstaens voor de Chrna7-/- muizen, en Dafne

Balemans en Mira Wouters om mij de isolatie van de DRGs aan te leren. Het was daags na de

begrafenis van mijn pa een welgekome afleiding.

Van het labo immuunhistochemie: Roeland Buckinx, voor de mooie beeldjes die je voor ons

gemaakt hebt en de leuke verhalen op het Neurogastro congres in Istanbul, en Katrien Alpaerts

voor het nooit aflatende enthousiasme, de steun en het doorzettingsvermogen.

De werknemers van het Animalarium wens ik te bedanken voor de goede zorg voor onze

proefdiertjes.

Onderzoek en ontwikkeling van nieuwe therapieën zou niet mogelijk zijn zonder de

implementatie van proefdieren. Ik denk dat het dan ook belangrijk is om hier zeker bij stil te

staan. Zij hebben hun leven letterlijk gegeven voor de wetenschap, opdat wij er beter van

zouden worden. Vergeet dat nooit, en benader hen steeds met respect! Een bijzondere ode

ook nog aan Stakker en Bully.

Ik ben er ook wel een beetje trots op dat ik doorheen de jaren her en der echt goede vrienden

heb verzameld, op wie ik ten allen tijde kan rekenen. Ik prijs mezelf dan ook gelukkig met de

volgende collectie : Julie en Bart, Jolien en Jurgen, Christine, Tess, Nele, Lotje en Makke, Eva

en Dieter, Marjan en Patrick, Wim en Alexis, Dries en Frida, Babsie en Bart, Sandrine en Nicolas,

Bert en Marijn.

Voorts een speciale dank aan alle mensen (possé-people rule!) die mij gedurende de voorbije

vier jaren voorzien hebben van WC-rollekes als kooiverrijking voor mijn proefdieren. Ettelijke

kilometers WC-papier hebben jullie lustig verbruikt om die rollekes toch maar op tijd ‘af’ te

krijgen, en ik kan jullie met de hand op het hart verzekeren dat mijn diertjes het ten zeerste

wisten te appreciëren! Fotomateriaal wordt nog apart verzonden!

2015 was ongetwijfeld het jaar met de meeste hoogtes en laagtes tot nu toe in mijn leven. Op

professioneel vlak liep alles gesmeerd (Een eerste paper gepubliceerd! Prijzen! Heel veel

prijzen!), maar dat alles weegt nooit op tegen de moeilijke persoonlijke problemen die mijn

gezin, partner en ikzelf hebben doorgemaakt door de ziekte en het overlijden van onze pa. Ik

wens dan ook iedereen te bedanken die op tijd en stond voor wat afleiding gezorgd heeft,

merci!

261

Het heeft er alleen maar voor gezorgd dat we eens te meer beseffen wat écht belangrijk is, en

wat eigenlijk overbodig is…

Ma’ke, je bent de sterkste persoon die ik ooit ben tegengekomen, en je bent dan ook een groot

voorbeeld voor mij! Danku voor het nooit aflatende vertrouwen en de steun.

Pa, hopelijk is het hierboven een beetje plezant, met veel treinen en fietsen. Je zei nooit veel,

maar het was voor mij een zeer pakkend moment toen je die laatste weken vertelde dat je zeer

trots was op je kinderen en er volledig voor gegaan bent het laatste jaar omdat je geloofde in

mijn hoop en kennis. Het is spijtig genoeg anders uitgedraaid.

Sofie (Schwessie!), ge zijt daar goed bezig in uw Deurganckdoksluis als vrouwelijke werfleider

terwijl al die mannen naar u moeten luisteren! Ha! We zijn heeeeeeel verschillend

(understatement), maar ik ben trots op u!

Noureddine, ge zijt ne goeie gast voor ons Sofie, laat u niet doen hé!

Xena (zaliger) en Bobientje, om mij op tijd en stond uit te laten en (letterlijk) een frisse wind

door mijn hoofd te laten waaien.

Michael, mijn schattie/Snoopy/skreitje/moppie! Het was officieel ‘aan’ daags voordat ik mijn

doctoraat ging starten. Dus als er mensen zijn die zich afvragen waarom ik de eerste weken een

beetje afgeleid was, nu weten jullie waarom ! We hebben al heel wat watertjes

doorzwommen de afgelopen jaren, maar samen staan we nu sterker dan ooit. En we gaan

weeral beiden een nieuw hoofdstuk starten in onze carrière! Ik ben trots op je

doorzettingsvermogen en toewijding, maar vooral ben ik blij dat je mij zo graag ziet en mij

onvoorwaardelijk steunt! Het doet deugd om jou op moeilijke momenten te horen zeggen dat

ik goed bezig ben. Vooral als je zegt dat “uw vrouwke uw rots is” . Ik prijs mezelf gelukkig dat

we het eigenlijk best wel goed hebben, en ben zeer benieuwd naar alle zotte dingen die we

samen nog gaan meemaken!

Hopelijk wordt 2016 beter; tot nu toe is dat reeds een vaststaand feit!

En dan nu het moment waar jullie allemaal op gewacht hebben – de quote die volgens jullie

perfect op mijn lijf geschreven is (in positieve zin!):

“Intelligence without ambition is a bird without wings”

Salvador Dali

doc.anet.be faculty of medicine and health sciences characterization and modulation of the enteric...

Documents