doc.anet.be faculty of medicine and health sciences characterization and modulation of the enteric...
TRANSCRIPT
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
XII
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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).
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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).
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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.
Chapter 1 Introduction
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).
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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.
Chapter 1 Introduction
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-
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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).
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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).
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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).
Chapter 1 Introduction
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
Chapter 1 Introduction
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).
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 3 Materials and Methods
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
Chapter 3 Materials and Methods
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.
Chapter 3 Materials and Methods
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
Chapter 3 Materials and Methods
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.
Chapter 3 Materials and Methods
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.
Chapter 3 Materials and Methods
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.
Chapter 3 Materials and Methods
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.
Chapter 3 Materials and Methods
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,
Chapter 3 Materials and Methods
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
Chapter 3 Materials and Methods
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
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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).
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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).
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
82
Chapter 4 Characterizing the cecal ligation and puncture model
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).
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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).
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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.
sham CLP day 2 CLP day 7 p-value
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.
Chapter 4 Characterizing the cecal ligation and puncture model
89
Table 4.8 Flow cytometric staining of T and B cells in the mesenteric lymph nodes.
sham CLP day 2 CLP day 7 p-value
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).
sham CLP day 2 CLP day 7 p-value
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).
sham CLP day 2 CLP day 7 p-value
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
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 = 6 to 7/group.
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
sham CLP day 2 CLP day 7
T helper cells
spleen
MLN
ileum LPMC
colon LPMC
-100,00%
0,00%
100,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
cytotoxic T cells
spleen
MLN
ileum LPMC
colon LPMC
-100,00%
0,00%
100,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
regulatory T cells
spleen
MLN
ileum LPMC
colon LPMC
-100,00%
0,00%
100,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
B cells
spleen
MLN
ileum LPMC
colon LPMC
Chapter 4 Characterizing the cecal ligation and puncture model
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.
sham CLP day 2 CLP day 7 p-value
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
Chapter 4 Characterizing the cecal ligation and puncture model
92
Table 4.12 Flow cytometric staining of dendritic cell markers in the mesenteric lymph nodes.
sham CLP day 2 CLP day 7 p-value
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
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 4.13 Flow cytometric staining of dendritic cell markers in the ileum (LPMC).
sham CLP day 2 CLP day 7 p-value
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
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 = 6 to 7/group.
Table 4.14 Flow cytometric staining of dendritic cell markers in the colon (LPMC).
sham CLP day 2 CLP day 7 p-value
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
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 = 6 to 7/group.
Chapter 4 Characterizing the cecal ligation and puncture model
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
sham CLP day 2 CLP day 7
CD11b+ dendritic cells
spleen
MLN
ileum LPMC
colon LPMC
-100,00%
0,00%
100,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
CD103+ dendritic cells
spleen
MLN
ileum LPMC
colon LPMC
-100,00%
0,00%
100,00%
200,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
dendritic cells
spleen
MLN
ileum LPMC
colon LPMC
Chapter 4 Characterizing the cecal ligation and puncture model
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
sham CLP day 2 CLP day 7 p-value
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
sham CLP day 2 CLP day 7 p-value
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.
Chapter 4 Characterizing the cecal ligation and puncture model
95
Table 4.17 Flow cytometric staining of other leukocyte populations in the ileum (LPMC)
sham CLP day 2 CLP day 7 p-value
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
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 = 6 to 7/group. MC: mast cell; Mφ: macrophage; Mo: monocytes; NK: natural killer cel; NPh: neutrophil.
Table 4.18 Flow cytometric staining of other leukocyte populations in the colon (LPMC)
sham CLP day 2 CLP day 7 p-value
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
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 = 6 to 7/group. MC: mast cell; Mφ: macrophage; Mo: monocytes; NK: natural killer cel; NPh: neutrophil.
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
sham CLP day 2 CLP day 7
macrophages
spleen
MLN
ileum LPMC
colon LPMC
0,00%
100,00%
200,00%
300,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
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
sham CLP day 2 CLP day 7
natural killer cells
spleen
MLN
ileum LPMC
colon LPMC
-100,00%
0,00%
100,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
monocytes
spleen
MLN
ileum LPMC
colon LPMC
0,00%
100,00%
200,00%
300,00%
Pe
rce
nta
ge c
han
ge
sham CLP day 2 CLP day 7
neutrophils
spleen
MLN
ileum LPMC
colon LPMC
Chapter 4 Characterizing the cecal ligation and puncture model
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.
sham CLP day 2 CLP day 7 p-value
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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.,
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 4 Characterizing the cecal ligation and puncture model
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.
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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.
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
113
5.3 Material and methods
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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.
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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.
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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).
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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)
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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.
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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)
Multi-unit recording / #units
0 10 20 30 40 50 60
0
5
10
15 Sham (n = 11)
Sham + supernatans CLP-colon (n = 11)
Pressure (mmHg)
Aff
ere
nt
dis
ch
arg
e (
imp
/s)
A B
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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)
Sham + supernatans CLP-colon (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)
Sham + supernatans CLP-colon (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)
Sham + supernatans CLP-colon (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)
Multi-unit recording / #units
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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*
* ** * * * *
*
* * **
* *
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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%)
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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.
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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,
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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.,
Chapter 5 Afferent nerve activity and CNS imaging in the CLP model
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
Parts of this chapter were published as:
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
139
6.3 Material and methods
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.,
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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).
6.3.6 Cytokine measurements
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).
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Clinical Disease Score day 2
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
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Chapter 6 Neuronal modulation of septic ileus: Effect of 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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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).
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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.
Chapter 6 Neuronal modulation of septic ileus: Effect of GTS-21
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
Parts of this chapter were published as:
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.
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Cecal ligation and puncture
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
Cecal ligation and puncture
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
Cecal ligation and puncture
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
Cecal ligation and puncture
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
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
Castleman’s disease in combination with best supportive care (versus best supportive care)
2009-017237-22
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
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)
Patients with rheumatoid arthritis
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
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
Patients with high-risk smoldering multiple myeloma
2006-001904-36
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
Patients with multiple myeloma in combination with bortezomib
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
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
Patients with rheumatoid arthritis
2006-001897-26
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
Patients with refractory or relapsed multiple myeloma
2011-000261-12
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
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
CNTO 328 (siltuximab, chimeric monoclonal anti-IL-6 antibody)
Patients with multiple myeloma
(as of April 1st 2016, www.clinicaltrials.gov and clinicaltrialsregister.eu)
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
166
7.3 Material and methods
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).
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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).
7.3.5 Cytokine measurements
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.
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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)+
+
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)+
+
B
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)+
+
C
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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.
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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).
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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.
Clinical Disease Score day 1
Sham
CLP d
ay 1
0
5
10
15IgG isotype control
Anti-IL-6 1 mg/kg
****
***
CD
S
Clinical Disease Score day 2
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
10IgG isotype control
Anti-IL-6 1 mg/kg
#
#, §
§
GC
Evans blue extracted from colon(colonic permeability)
Sham
CLP d
ay 2
0
20
40
60
80IgG isotype control
anti-IL-6 1 mg/kg
****
µg
/100m
g c
olo
n t
issu
e
A B
C D
E F
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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.
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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 + IgG isotype
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%)*
Number of positive MLN cultures (%)
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%)*
Number of positive MLN cultures (%)
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
174
Table 7.5 RT-PCR of cell adhesion molecules.
Sham + IgG isotype
sham + anti-IL-6
CLP d2 + IgG isotype
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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).
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Clinical Disease Score day 1
Sham
CLP d
ay 1
0
5
10
15IgG isotype control
Anti-IL-6 1 mg/kg# #
*
CD
S
Clinical Disease Score day 2
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
100IgG isotype control
Anti-IL-6 1 mg/kg# #
% 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
10IgG isotype control
Anti-IL-6 1 mg/kg# #,§§
GC
Evans blue extracted from colon(colonic permeability)
Sham
CLP d
ay 2
0
10
20
30
40IgG isotype control
Anti-IL-6 1 mg/kg
* *
µg
/100m
g c
olo
n t
issu
e
A B
C D
E F
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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).
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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).
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Clinical Disease Score day 1
Sham
CLP d
ay 1
0
5
10
15IgG isotype control
Anti-IL-6 1 mg/kg
# #CD
S
Clinical Disease Score day 2
Sham
CLP d
ay 2
0
5
10
15IgG isotype control
Anti-IL-6 1 mg/kg
*
# #
CD
S
Percentage of initial body weight
Sham
CLP d
ay 1
0204060
90
95
100IgG isotype control
Anti-IL-6 1 mg/kg# #
% 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
100IgG isotype control
Anti-IL-6 1 mg/kg# #
%G
E
Geometric Center
Sham
CLP d
ay 2
0
2
4
6
8
10IgG isotype control
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
20IgG isotype control
Anti-IL-6 1 mg/kg
EB
co
lon
(µ
g/1
00 m
g c
olo
n)
A B
C D
E F
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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 + IgG isotype
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.
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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;
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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
Chapter 7 Immunological modulation of septic ileus: effect of anti-IL-6
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.
Chapter 8 General discussion
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.
Chapter 8 General discussion
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
Chapter 8 General discussion
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
Chapter 8 General discussion
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
Chapter 8 General discussion
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).
Chapter 8 General discussion
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
Chapter 8 General discussion
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.
Chapter 8 General discussion
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
Chapter 8 General discussion
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
Chapter 8 General discussion
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
Chapter 8 General discussion
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
Chapter 8 General discussion
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).
Chapter 8 General discussion
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
Chapter 8 General discussion
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.
Chapter 8 General discussion
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
GTS-21Anti-IL-6 preventiveAnti-IL-6 curative
GTS-21Anti-IL-6 preventiveAnti-IL-6 curative
GTS-21Anti-IL-6 preventiveAnti-IL-6 curative
Chapter 8 General discussion
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
Chapter 10 Samenvatting
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
Chapter 10 Samenvatting
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
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
University of Antwerp, Antwerp, Belgium
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
University of Antwerp, Antwerp, Belgium
1999 – 2005 Latin – Mathematics – Science
Sint-Ritacollege, Kontich, Belgium
Scientific Curriculum Vitae
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)
Scientific Curriculum Vitae
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).
Scientific Curriculum Vitae
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).
Scientific Curriculum Vitae
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.
Scientific Curriculum Vitae
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
Scientific Curriculum Vitae
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)
Scientific Curriculum Vitae
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
Scientific Curriculum Vitae
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