Immunomodulatory and Regenerative Effects of Mesenchymal Stromal Cells in Trauma

Ulm University Medical Center

Institute of Transfusion Medicine

Prof. Dr. med. Hubert Schrezenmeier

Immunomodulatory and regenerative effects of mesenchymal stromal cells in trauma

Dissertation

to obtain the Doctoral Degree of Human Biology

at the Medical Faculty of Ulm University

Elisa Maria Amann

Juliaca (Peru)

2018

Present Dean: Prof. Dr. rer. nat. Thomas Wirth 1st reviewer: Prof. Dr. med. Hubert Schrezenmeier 2nd reviewer: Prof. Dr. med. Rolf Brenner Date of graduation: November 9th, 2018

Parts of this thesis were already published in

Amann EM, Rojewski MT, Rodi S, Fürst D, Fiedler J, Palmer A, Braumüller S, Huber-Lang M, Schrezenmeier H, Brenner RE (2018) Systemic recovery and therapeutic effects of transplanted allogenic and xenogenic mesenchymal stromal cells in a rat blunt chest trauma model. Cytotherapy 20: 218-231. DOI: 10.1016/j.jcyt.2017.11.005. Epub 2017 Dec 7. CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/

Amann EM, Groß A, Rojewski MT, Kestler HA, Kalbitz M, Brenner RE, Huber- Lang M, Schrezenmeier H (2019) Inflammatory response of mesenchymal stromal cells after in vivo exposure with selected trauma-related factors and polytrauma serum. PLoS ONE 14(5):e0216862. DOI: 10.1371/journal.pone.0216862. eCollection 2019 CC BY 4.0 https://creativecommons.org/licenses/by/4.0/

List of contents

LIST OF ABBREVIATIONS III

1. INTRODUCTION 1

1.1. Polytrauma 1

1.2. Blunt chest trauma and acute lung injury 2

1.3. Human multipotent mesenchymal stromal cells 5

1.4. Aims of the study 14

2. MATERIAL AND METHODS 16

2.1. Tools, machines, consumables and software 16

2.2. Chemicals and kits 19

2.3. Antibodies 22

2.4. Cells 23

2.5. Cultivation of mesenchymal stromal cells, NR8383 cells and C2C12 cells 23

2.6. Mesenchymal stromal cells and polytrauma 26

2.7. Effects of manipulating MSC for in vivo experiments 32

2.8. Animal experiments 40

2.9. Statistics 51

3. RESULTS 52

3.1. Mesenchymal stromal cells and polytrauma 52

3.2. Characterization of rat mesenchymal stromal cells 72

3.3. Effects of manipulating MSC for in vivo experiments 73

3.4. Animal experiments 96

List of contents

4. DISCUSSION 135

4.1. The role of interleukin 1 beta in polytrauma 135

4.2. Characterization of rat mesenchymal stromal cells 142

4.3. Effects of manipulating MSC for in vivo experiments 143

4.4. Mesenchymal stromal cells in blunt chest trauma 150

4.5. Conclusion 160

5. SUMMARY 162

6. REFERENCE LIST 164

APPENDIX 186

ACKNOWLEDGEMENTS 198

CURRICULUM VITAE: ELISA MARIA AMANN 199

List of abbreviations

AIS Abbreviated Injury Scale ALI Acute lung injury Alpha MEM Minimal Essential Medium Eagle, Alpha modification ANOVA Analysis of variance APC Allophycocyanin APC-Cy7 Allophycocyanin coupled with the cyanine dye Cy7 ARDS Acute respiratory distress syndrome ASC Adipose-derived mesenchymal stromal cells

BAL Broncho-alveolar lavage fluid BM-MSC Bone marrow derived mesenchymal stromal cell(s) BSA Bovine serum albumin

C3a Complement component 3a C5a Complement component 5a CARS Compensatory anti-inflammatory response syndrome CCL C-C motif chemokine ligand CCR C-C motif chemokine receptor CD Cluster of differentiation CFSE Carboxyfluorescein diacetate succinimidyl ester CFU Colony forming units cm Centimetre cm2 Square centimetre CSF Colony stimulating factor CXCL C-X-C motif chemokine ligand CXCR C-X-C motif chemokine receptor d Days DAMP Damage-associated molecular pattern DGU Deutsche Gesellschaft für Unfallchirurgie DMSO Dimethyl sulfoxide DNA Deoxyribonucleic acid DPBS Dulbecco´s phosphate buffered saline e.g. Exempli gratia EDTA Ethylenediaminetetraacetic acid ELISA Enzyme-linked immunosorbent assay

FACS Fluorescence activated cell sorting

III

List of abbreviations

FBS Fetal bovine serum FGF Fibroblast growth factor FITC Fluorescein isothiocyanate FSC-H Forward scatter height g Gram g Standard gravity (9.8 m/s2) GMP Good manufacturing practice Gy Gray [absorbed radiation dose] h Hours HLA Human leucocyte antigen HMGB1 High mobility group box 1 hMSC Human multipotent mesenchymal stromal cell(s) hMSC (IL1B) IL1B pre-stimulated hMSC HSA Human serum albumin

ICAM Intercellular adhesion molecule ICAM1 Synonym: CD54 IDO Indoleamine 2,3-dioxygenase IFNG Interferon gamma Ig Immunoglobulin IGF1 Insulin-like growth factor 1 IL Interleukin IL1A Interleukin 1 alpha IL1B Interleukin 1 beta IL1RN Interleukin 1 receptor antagonist ISCT International Society for Cellular Therapy ISS Injury Severity Score IU International units kg Kilogram l Litre LPS Lipopolysaccharide

M Molar, mol/l MAPK Mitogen-activated protein kinases MEM Minimum essential media mg Milligram µg Microgram IV

List of abbreviations min Minutes µl Microlitre ml Millilitre µm Micrometre mM Millimolar, mmol/l mmHg Millimetre of mercury MMP Matrix metallopeptidase MODS Multiple organ dysfunction syndrome MOF Multiple organ failure MPO Myeloperoxidase MSC Mesenchymal stromal cells

NF-κB Nuclear factor kappa-light-chain-enhancer of activated B cells ng Nanogram nm Nanometre

P Probability PCR Polymerase chain reaction PDGF-BB Homodimer of platelet-derived growth factor PE Phycoerythrin PerCP Peridinin-chlorophyll-protein PGE2 Prostaglandin E2 PKH+ Events in flow cytometry analysis being in PKH26 gate, PKH26- fluorescent, PKH26-stained PKH- Events in flow cytometry analysis not being in PKH26 gate PL Platelet lysate PML Polymorphonuclear leucocytes PTCH Polytrauma cocktail high PTCL Polytrauma cocktail low PTGS2 Prostaglandin-endoperoxide synthase 2

RFU Relative fluorescence units rMSC Rat multipotent mesenchymal stromal cell(s) rMSC (IL1B) IL1B pre-stimulated rMSC RNA Ribonucleic acid s Seconds SD Standard deviation SIRS Systemic inflammatory response syndrome SSC-H Sideward scatter height

List of abbreviations

TGFB1 Transforming growth factor beta 1 THBD Thrombomodulin THPO Thrombopoietin TLR Toll-like receptor TMB 3,3′,5,5′-Tetramethylbenzidine TNF Tumor necrosis factor (alpha) TNFAIP6 Tumor necrosis factor alpha inducible protein 6 TSG-6 TNFAIP6 TxT Blunt chest trauma TxT+hMSC Blunt chest trauma and treatment with human MSC TxT+hMSC (IL1B) Blunt chest trauma and treatment with IL1B-primed human MSC TxT+rMSC Blunt chest trauma and treatment with rat MSC TxT+rMSC (IL1B) Blunt chest trauma and treatment with IL1B-primed rat MSC

U Units UC-MSC Umbilical cord-derived mesenchymal stromal cells

VCAM1 Vascular cell adhesion molecule 1, synonym: CD106 VEGFA Vascular endothelial growth factor A

Introduction

1. Introduction

1.1. Polytrauma

Road injuries are a leading cause of death worldwide [96]. The Trauma Register DGU® (Deutsche Gesellschaft für Unfallchirurgie) published in their Annual Report 2017 that the most common causes of trauma are traffic accidents (51 %) and falls from various heights (42 %) [85]. Polytrauma is a life-threatening disease with a hospital mortality of 11 % (Trauma Register DGU® 2016: 3507 patients) [85]. Furthermore, 30 % of surviving patients have disabilities at time of discharge (Trauma Register DGU® 2016: 9588 patients; [85]). After discharge from hospital life expectancy of severely injured patients is significantly lower than that of normal population [178]. The current definition of polytrauma is that at least two body regions must be significantly injured and at least one physiological problem (high age, unconsciousness, hypotension, acidosis or coagulopathy) must exist [188]. To define polytrauma and to assess the risk of death the Abbreviated Injury Scale (AIS) and the Injury Severity Score (ISS) are used. AIS and ISS indicate the number and severity of injuries [92]. The AIS divides the body in seven regions (external, head, neck, thorax, abdomen, spine and extremities) and evaluates the severity of the injuries in the body regions with a scale from one to six [3,92]. The ISS is calculated by summing up the squares of the highest AIS score in each of the three most severely injured areas [9]. In polytrauma the ISS is usually greater than 15 [84,186]. Mortality of polytrauma patients is composed of early mortality (within 24 hours (h): 5.1 %; for example due to haemorrhagic shock) and late mortality (within 30 days (d): 10.8 %; due to brain injuries or host defence failure) [85,118]. Fractures and soft tissue injuries (first hits) induce local and systemic inflammation. Second hits like metabolic acidosis, ischaemia/reperfusion injuries, infections or massive transfusions increase this systemic inflammatory reaction and distant organs and tissues are affected [84,118,215]. Polytrauma is often accompanied by a systemic inflammatory response syndrome (SIRS) and a compensatory anti-inflammatory response syndrome (CARS). The imbalance of these immune responses, epithelial cell damage, parenchymal cell damage and cell death lead to sepsis, multiple organ dysfunction syndrome (MODS) or multiple organ failure (MOF) [84,118]. At the development of SIRS, CARS, sepsis, MODS

Introduction and MOF numerous cells, especially the innate immune system with phagocytic activity (monocytes, macrophages and neutrophils), protection mechanisms (coagulation cascade, kallikrein–kinin system, complement system, acute phase reaction) and mediators (damage-associated molecular pattern (DAMP), pathogen-associated molecular pattern, cytokine, chemokine, neuromediators, heat shock proteins and oxygen radicals) are involved [84,118]. Until today, there is an effort to define markers that have been linked to an increased disease severity and poorer clinical outcomes in severely injured patients [54,218,236]. The spectrum of biomarkers includes inflammatory cytokines [54,167,218], complement components [273], coagulation proteins [219] and markers that reflect cell injury like DAMPs [39]. Therefore, these factors can be used to study severity and potential pathomechanisms in polytrauma [56,99]. The early treatment of polytrauma patients proceeds according to the rule “treat first, what kills first”. Effective therapeutic approaches to avoid complications and to improve long-term survival are still to be identified. The potential role of mesenchymal stromal cells (MSC) as therapeutic approach for severely injured patients is discussed [106,163,175] and the role of MSC in polytrauma is under investigation [99,224,258]. Allogeneic MSC may support the anti-inflammatory [181,198], anti-apoptotic [114,199,270], anti-microbial [127,238], anti-oxidative [199,254], pro-angiogenic [270] and regenerative signalling pathways (see section 1.3.) after trauma and thus they may reduce disability and late mortality in traumatized patients [106,163,175].

1.2. Blunt chest trauma and acute lung injury

The Trauma Register DGU® published that 96 % of trauma patients had a blunt trauma [85]. Polytrauma patients (n=82648), from 2014 to 2016, usually suffered from head injuries (AIS score ≥ 3, 34 %) and thoracic injuries (AIS score ≥ 3, 38 %) [85]. The severity of thoracic injury has an impact on the ventilation period, time of intensive care unit stay and complication rates (sepsis, pneumonia, acute lung injury (ALI), MOF) of severely injured patients [13]. It is still under discussion if thoracic injury has an impact on the mortality of severely injured patients [13,38]. One complication of severe blunt chest trauma is ALI or its more severe form

Introduction acute respiratory distress syndrome (ARDS) with a mortality rate of > 30 % [16,155]. The definition of blunt chest wall trauma is a blunt chest injury leading to chest wall contusion or rib fractures with or without non-immediate life-threatening lung injury [12]. ALI is defined by acute onset, by diffuse bilateral infiltrate in the lungs on X- ray, by pulmonary oedema and by absence of left arterial hypertension or pulmonary wedge pressure of > 18 millimetre of mercury (mmHg) that might explain pulmonary oedema [17,200]. The Horovitz quotient generally describes the lung capacity to saturate blood with oxygen and is the ratio of oxygen partial pressure in the arterial blood and the fraction of inspired oxygen. It defines and distinguishes ALI (≤ 300 mmHg) from ARDS (≤ 200 mmHg) [17].

1.2.1. Pathophysiology of blunt chest trauma and acute lung injury

The pathophysiological mechanisms of ALI have not yet been fully clarified. To understand the pathophysiological mechanisms of ALI, the anatomy of alveoli is here introduced briefly. Gas exchange takes place in the alveolar space. The alveoli consist of alveolar macrophages, bronchial epithelium (40 % of pneumocyte type I and 60 % of pneumocyte type II), basal lamina and capillaries [257]. Pneumocyte type II produce surfactant that reduces surface tension and is important for re-stretching of alveoli [257]. Compression and rebound of elastic lung tissue leads to pulmonary contusion [220]. Damage occurs by transfer of kinetic energy to lung tissue consisting of different densities (heavier hilar structures versus lighter alveolar tissue, gas versus liquid) and by implosion of gas bubbles [83,220]. Lung contusion results in damage of bronchial epithelium and vascular endothelium. Impairment of alveolo- capillary integrity leads to inflow of blood and protein-rich liquid from capillary regions into the alveolar spaces [83,220]. Furthermore, interstitial and intra- alveolar hemorrhage results in interstitial or perivascular oedema [220]. Thickening of septum and loss of compliance results in respiratory distress with hypoxemia and hypercarbia [40,83,183]. Apoptosis of pneumocytes type II leads to impaired surfactant production [91,229]. The increase of surface tension makes breathing difficult and impairs gas exchange [183]. Similar to polytrauma, inflammatory reactions play a critical role in the development of ALI after blunt chest trauma

Introduction

[83,97,220]. Activated alveolar macrophages recruit neutrophils and monocytes [97]. There is an early influx of neutrophils in the pulmonary alveoli and a delayed monocyte infiltration [122,220]. Activated inflammatory cells release pro- inflammatory molecules (interleukin 1 beta (IL1B), tumor necrosis factor alpha (TNF), interleukin (IL) 6, C-X-C motif chemokine ligand (CXCL) 8) that are elevated in broncho-alveolar lavage fluid (BAL) and plasma of ARDS patients [167,236]. Neutrophils further damage epithelium and endothelium by release of reactive oxygen species and proteases [260]. In spite of accumulation of neutrophils bacterial clearance in ALI is impaired [57].

1.2.2. Therapy of acute lung injury

Therapeutic approaches for ALI/ARDS are mechanical ventilation, prone positioning, anti-inflammatory therapy, antimicrobial therapy, neuromuscular blocking agents and conservative fluid management [14]. However, causal therapeutic approaches to improve regeneration or to reduce morbidity and mortality are still to be identified. MSC as treatment option for ALI are being discussed increasingly [97,162,176]. It could be shown that MSC have antimicrobial [127,238], anti-inflammatory [181,198] and pro-angiogenic effects [270]. In pre-clinical models for ALI the effect of MSC has already been tested for infection- and chemical-induced lung injury in different species [50,210,264]. Furthermore, there are some clinical studies of MSC for lung injury [180]. The search terms “lung injury” for disease and “mesenchymal” for other terms result in 21 clinical trials. Most of the studies are dealing with lung injury induced by another primary disease (10/21 bronchopulmonary dysplasia, 2/21 pneumoconiosis), a chemical substance (1/21) or a viral disease (1/21). One study deals with MSC treatment in ARDS patients receiving hematopoietic stem cell transplantation. Studies on MSC for ALI/ARDS are listed in table 1. Most of these studies address allogeneic transplantation of MSC and the safety and efficacy of treatment with MSC (table 1, [271]).

Introduction

Table 1 Clinical trials dealing with acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) and MSC therapy. Data based on [180,271]. UC-MSC: umbilical cord-derived mesenchymal stromal cells; BM-MSC: bone marrow derived mesenchymal stromal cells; ASC: adipose-derived mesenchymal stromal cells.

Title Identifier Recruitment Conditions Interventions Comments Human Umbilical-Cord- NCT02444455 Unknown ALI, ARDS Allogeneic  Phase I-II, Safety Derived Mesenchymal UC-MSC and Efficacy Stem Cell Therapy in  i.v. 5x105 cells/kg Acute Lung Injury once a day, total of three times Human Mesenchymal NCT01775774 Completed ARDS Allogeneic  Phase I-II, dose Stem Cells For Acute BM-MSC escalation, safety Respiratory Distress  i.v. 1/5/10x106 Syndrome (START) cells/kg Human Mesenchymal NCT02097641 Active, not ARDS Allogeneic  Phase II Stem Cells For Acute recruiting BM-MSC or  Extension of the Respiratory Distress Plasma-Lyte A phase I pilot study Syndrome (START) (NCT01775774)  i.v. 10x106 cells/kg Adipose-derived NCT01902082 Unknown ARDS Allogeneic  Phase I, Safety Mesenchymal Stem ASC or  i.v. 1x106 cells/kg Cells in Acute Placebo Respiratory Distress (saline) Syndrome Mesenchymal Stem Cell NCT02112500 Unknown ARDS BM-MSC  Phase II, Safety and in Patients With Acute Efficacy Severe Respiratory Failure (STELLAR) Repair of Acute NCT03042143 Not yet ARDS UC-derived  Phase I-II Respiratory Distress recruiting CD362+ ve Syndrome by Stromal MSC or Cell Administration Placebo (REALIST) (Plasma-Lyte)

1.3. Human multipotent mesenchymal stromal cells

The presence and regenerative potential of MSC was first suggested in 1867 by Cohnheim. He hypothesized that non-hematopoietic bone marrow derived cells could be recruited to injured sites and that they could take part in tissue regeneration [41]. Friedenstein established the research about MSC in 1970. He reported about the fibroblast-like shape of MSC and called them colony forming unit-fibroblasts [78]. Furthermore, MSC are regarded as a heterogeneous multipotent cell population [42,212]. Crisan and colleagues established the assumption that MSC are of peri-vascular origin and are distributed in tissues of the whole body [44,49]. They used surface marker expression to show that

Introduction perivascular cells (cluster of differentiation (CD) 146high, CD34-, CD45-, CD56-) also express MSC-characteristic markers (CD44+, CD73+, CD90+, and CD105+) under cell culture condition or within intact tissue. Furthermore, tri-lineage differentiation of cultured perivascular cells could be demonstrated [44]. However, Guimarães-Camboa et al. used an inducible Tbx18-CreERT2 line in mice and immunohistology to demonstrate that pericytes and vascular smooth muscle cells did not transdifferentiate during aging or pathological status [94]. They suggest therefore that the regenerative potential of MSC arose during the cell expansion process. Cell expansion is necessary to obtain clinical doses because MSC are present in low frequency [27]. Different isolation and expansion protocols are available and used in clinical studies [18,71,170]. The tissue source, isolation process, culture conditions, flow cytometry phenotype and application field of MSC varied a lot and may influence the functional characteristics [169]. The International Society for Cellular Therapy (ISCT) already published three position statements to make MSC research more uniform and clearer. First, the name mesenchymal stromal cell was established instead of mesenchymal stem cell because stemness could not finally be proven [105]. Minimal criteria to define MSC were published in 2006. MSC have to be plastic adherent, positive (≥ 95 %) for the markers CD105, CD73 and CD90 and negative (≥ 98 %) for the markers CD45, CD34, CD14, CD11b, CD79A, CD19, human leucocyte antigen (HLA)-DP, HLA-DQ, HLA-DR. Furthermore, they should exhibit at least tri-lineage differentiation in vitro in adipocytes, chondrocytes and osteoblasts [58]. Clinical studies using MSC deal with the regenerative (replacement therapy) or with the immunomodulatory potential of these cells [235]. The immunomodulatory function of MSC is of increasing interest and functional markers of potency were discussed. The ISCT has published two papers to support the establishment of common guidelines for functional assays and to respond to the requirements from the Regulatory Authorities [81,126]. Functional assays should have predictive power to in vivo outcome. The ISCT recommended that in vitro analysis of resting and activated (exempli gratia (e.g.) by pro-inflammatory factors like interferon gamma (IFNG) and IL1B) MSC might be helpful to have an outlook into the in vivo effect of MSC (due to the immunological status of the patient) or rather their immunomodulatory potential [81,126]. The ISCT listed several genes which are important for the immunomodulatory function of MSC like indoleamine 2,3-

Introduction dioxygenase (IDO), CXCL8, CXCL9, CXCL10, CXCL11, CXCL12, intercellular adhesion molecule (ICAM) 1, vascular cell adhesion molecule 1 (VCAM1), C-C motif chemokine ligand (CCL) 2, CCL5, CCL7, Toll-like receptor (TLR) 3, TLR4, HLA-DR, HLA-G 5, hepatocyte growth factor, IL6, C-C motif chemokine receptor (CCR) 7, CCR10, vascular endothelial growth factor A (VEGFA), CD55, CD274, cyclooxygenase-2, TNF superfamily member 10, tumor necrosis factor alpha inducible protein 6 (TNFAIP6), fibroblast growth factor (FGF) 7, CD46, transforming growth factor beta 1 (TGFB1), C-X-C motif chemokine receptor (CXCR) 1, CXCR4, CXCR6, TIMP metallopeptidase inhibitor (TIMP) 1, TIMP2, angiopoietin 2, galectin 1, heme oxygenase 1 and interleukin 1 receptor antagonist (IL1RN) [81]. The focus in the present study lies on IL6, CXCL8, VEGFA, CCL2, IL1RN and TNFAIP6. Since 1867 until today it is assumed that MSC play an important role in physiological remodelling and tissue homeostasis [67]. Several publications showed that MSC are involved in response to injury and post damage repair [37,124,131,152,198]. Thus, there are 827 studies listed on ClinicalTrials.gov for the search term “mesenchymal stem cells” [180].

1.3.1. Mesenchymal stromal cells in cell therapy

MSC are a therapeutic option due to their multi-lineage differentiation potential [23], secretion of active molecules [181,185,198] and formation of apoptotic bodies [82,247]. These promote regeneration and prevent excessive inflammation. There are further reasons for the use of MSC in regenerative medicine. Extraction of MSC is ethically justifiable, isolation is convenient and the cells grow fast in contrast to embryonic stem cells and induced pluripotent stem cell [120,168]. Furthermore, MSC are often described as immune privileged due to their low expression of major histocompatibility complex, class II proteins [71,169] and costimulatory molecules like CD40, CD80 and CD86 [169] on their cell surfaces. Additionally, it could be shown that the allogeneic transplantation of MSC can evoke humoral and cellular immune response [8,62,174]. Moll and colleagues showed that this immune reaction depends on cell dose and cell passage [174]. According to this, early passages of MSC should not evoke an immune reaction [174]. Additionally, MSC have immunosuppressive effects which enable a

Introduction prolonged persistence. Also, MSC have the attribute to be immune-evasive (see section 1.3.2.). Another advantage of MSC is that they migrate to sites of injury and fulfil their therapeutic mission (see section 1.3.3.). Furthermore, it is possible to improve MSC functions by priming the cells before therapeutic application. Thus, there might be an increase in the therapeutic effect of MSC in several diseases (see section 1.3.4.). MSC can be used both in autologous transplantation as well as in allogeneic transplantation. The benefits of the allogeneic transplantation are the possibility of extensive pre-testing (characteristic tests and functionality tests) of the cells and the donors. This kind of transplantation is also important for emergency medicine since allogeneic cells can be expanded in advance and are then available as off- the-shelf product for immediate treatment of injured patients. This is also advantageous as MSC from elderly or sick patients might be not as functional as MSC from young healthy donors and such patients might profit especially from bio- banking [123,150,206]. In clinical trials bone marrow and adipose-tissue are often used as starting material because of easy accessibility [170]. Up to now, medium supplemented with xenogeneic fetal bovine serum (FBS) has usually been used for the expansion of MSC. However, alternative medium supplements like platelet lysate (PL) become more popular [18,70,221] and were also used in clinical studies [87,170]. Human MSC used in the present and already published animal experiments [5,117] were isolated, characterized and expanded according to good manufacturing practice (GMP) protocols used in clinical trials for bone regeneration (clinicaltrials.gov, NCT02751125, NCT02065167 and NCT01842477). Human MSC used in the present animal experiment already showed significant therapeutic effects in pre-clinical studies dealing with wound healing and long bone defects [23,87,198].

1.3.2. Immunomodulatory properties of mesenchymal stromal cells

MSC interact with immune cells via soluble factors, cell-to-cell interaction and extracellular vesicles (reviewed in [140]) . Figure 1 depicts the effect of MSC on immune cells and endothelial cells and lists some mechanisms which are involved.

Introduction

In the following, only the interaction of macrophages and neutrophils with MSC will be presented due to their role in ALI.

Figure 1 MSC influence immune cells and endothelial cells by soluble factors, extracellular vesicles (EV) and by cell-to-cell contact (CCC). Overview of soluble factors and effects of MSC on immune cells and endothelial cells. Data based on [6,140,246]. MSC: Mesenchymal stromal cells; NK cell: natural killer cell; TGFB1: transforming growth factor beta 1; HGF: hepatocyte growth factor; HLA-G: human leucocyte antigen G; IDO: indoleamine 2,3-dioxygenase; IL10: interleukin 10; IL6: interleukin 6; LIF: LIF, interleukin 6 family cytokine; HMOX1: heme oxygenase 1; PGE2: prostaglandin E2; CCL2: C-C motif chemokine ligand 2; IL1RN: interleukin 1 receptor antagonist; C3: complement C3; TNFAIP6: tumor necrosis factor alpha inducible protein 6; CSF1: colony stimulating factor 1; CSF2: colony stimulating factor 2; MIF: macrophage migration inhibitory factor; CXCL8: C-X-C motif chemokine ligand 8; SOD3: superoxide dismutase 3; Th1: Type 1 helper T cell; Th17: helper T cell producing interleukin 17; Treg: regulatory T cell; DC: dendritic cell; M1: macrophage phenotype M1, classically activated; M2: macrophage phenotype M2, alternatively activated.

Neutrophils take part in acute inflammation. They are antimicrobial by phagocytosis of pathogens, secretion of molecules and neutrophil extracellular traps. It could be shown that MSC recruit neutrophils by secretion of CXCL8 and macrophage migration inhibitory factor [22]. Jiang and colleagues revealed that MSC engulf apoptotic neutrophils in an ICAM1-dependent manner and inhibit

Introduction neutrophil death, trap formation and release of destructive enzymes by secretion of superoxide dismutase 3 and MSC reduce tissue damage in a murine vasculitis model [113]. Furthermore, MSC suppress apoptosis of neutrophils by secretion of IL6, IFNG and colony stimulating factor (CSF) 2, even if neutrophils were activated by CXCL8 [30,199]. In addition, microvesicles of MSC show therapeutic effect in an endotoxin-induced lung injury model by reducing the influx of neutrophils into the lung [272].

Macrophages can be polarized into classically activated (secretion of IL1 and TNF, anti-microbial activity) or alternatively activated macrophages (secretion of IL10, focusing on tissue repair). MSC recruit macrophages to site of injury and promote generation of alternatively activated macrophages [77,151,268]. Németh and colleagues revealed that prostaglandin E2 (PGE2) secretion of MSC leads to an increase of IL10-producing macrophages in a mouse sepsis model [181]. IL1RN from MSC also plays a role in the generation of alternatively activated macrophages [134,158]. Furthermore, IL1RN and TNFAIP6 from MSC reduce macrophage activation (reduction of TNF secretion) and therefore improve healing in bleomycin-induced lung injury [185], wound healing [198] and peritonitis [37].

The possibility of species-specific immunomodulation of MSC has to be considered. Ren and colleagues showed that suppression of T cell proliferation is mediated by IDO in humans and mediated by nitric oxide in mice [202]. Furthermore, Menard and colleagues showed that the immunomodulatory function of MSC depends on the source of MSC isolation and the cell culture condition [169].

1.3.3. Homing of mesenchymal stromal cells

Systemically injected MSC have to migrate to site of injury and transmigrate through the endothelial barrier to reach their site of action. Homing of MSC was demonstrated in several publications [63,104,244]. However, the underlying mechanisms are not fully identified yet. After systemic administration of MSC they follow a chemokine gradient. Several soluble factors are involved in homing of MSC like CXCL12 [143,159,160,263],

Introduction platelet-derived growth factor [73,194], FGF2 [223], VEGFA [74], placental growth factor [74,189], CCL3 [234], CCL2 [60], CXCL8 [207] and complement component 3a (C3a) [99]. Concentration of these factors and activation state of MSC has an impact on the chemoattractive potential of these factors [28,194,223]. MSC extravasation is composed of capturing of MSC, rolling along and adhering to endothelium, followed by a transmigration step. Inflammatory cytokines (IL1B, TNF) activate the endothelium to upregulate adhesion molecules (selectin P, selectin E, ICAM1, VCAM1) [216]. MSC might bind to these molecules by glycoproteins [267] and galectin 1 [237]. Furthermore, integrin subunit alpha 4 [130] and integrin subunit beta 1 [2] on MSC seem to be important for rolling and adhesion. C-C chemokine receptors like CXCR4 and CCR2 play a role in homing of MSC [15,159,160,263]. Belema-Bedada and colleagues showed that CCR2 plays a role in cytoskeletal reorganization/polarization in MSC which is necessary for trans-endothelial migration [15]. Collagenases and matrix metallopeptidases (MMP2, MMP9) enable MSC to cross basal membrane by degrading collagen type IV [52,233]. After that MSC follow a gradient of chemoattractive molecules to site of action. Mobilization of MSC takes a further step. Cells are recruited from bone marrow to blood circulation before they migrate to site of injury [35]. In the literature there are contrary reports about the increase of mobilized MSC in the blood after trauma or injury [5,208,258].

1.3.4. Licensing/priming/stimulation of mesenchymal stromal cells

Activation of MSC before their therapeutic application has several advantages compared to application of resting MSC. Migratory abilities or MSC are improved after stimulation with TNF or IL1B [28,79,99,194]. Fu and colleagues demonstrated that TNF significantly upregulates ICAM1 in rat bone marrow- derived mesenchymal stromal cells (BM-MSC) from Wistar rats and thus enhance migration ability of MSC [79]. Furthermore, they described an activation of p38 and extracellular signal-regulated kinase by TNF [79]. Ponte and colleagues showed that TNF stimulation of BM-MSC for 24 h increases migration of MSC toward chemokines [194]. IL1B stimulation of human BM-MSC for 24 h induces ICAM1,

Introduction

ICAM4, MMP1 and MMP3 expression, partially mediated by nuclear factor kappa- light-chain-enhancer of activated B cells (NF-κB) pathway [28,99].

The immunomodulatory properties of MSC are increased after pre-stimulation of MSC with IFNG [125,169], TNF [169,172,198] and IL1B [28,68,99]. Menard and colleagues described that priming of MSC significantly increases the expression of costimulatory and adhesion molecules (e.g. CD40, ICAM1, VCAM1) and immunomodulatory cytokines like IDO1, nitric oxide synthase 2, prostaglandin- endoperoxide synthase 2 (PTGS2) and TNFAIP6. Furthermore, they pointed out that the inhibition of T cell, B cell and natural killer cell proliferation is induced or enhanced by priming. In addition, the susceptibility to natural killer cell-mediated lysis was reduced after priming [169]. Polchert and colleagues presented an increased MSC suppression of Graft versus host disease after IFNG exposure [192]. Fan and colleagues proved a therapeutic benefit of IL1B priming of human umbilical cord-derived mesenchymal stromal cells (UC-MSC) in a dextran sulfate sodium-induced colitis model in C57BL/6J mice [68]. Furthermore, several publications revealed that TNFAIP6 can be induced by the pro-inflammatory cytokines IL1B and TNF [37,99,198,261]. TNFAIP6 is known as anti-inflammatory protein and plays an important role in tissue remodelling. In animal models of myocardial infarction [135], peritonitis [37], wound healing [198] and lipopolysaccharide (LPS)-induced lung injury [50] therapeutic effects of MSC were explained by TNFAIP6 secretion or increased after TNFAIP6 overexpression. Thus, priming might be one option to increase therapeutic effect of MSC. However, some aspects have to be considered for priming of MSC to increase immunomodulatory properties. First of all, effects of priming can differ in MSC derived from different tissues [169]. Secondly, MSC have a pro- and anti- inflammatory phenotype depending on the microenvironment [142]. Concentrations of factors [33,34,192,214], timing [265], stimulation of TLRs [30,255] play a role in formation of the secretion phenotype of MSC and MSC function.

Another type of pre-treatment is the induction of differentiation in MSC before therapeutic application [10,29,51]. In C-CURE (Cardiopoietic stem Cell therapy in heart failURE) Multicenter Randomized Trial BM-MSC were treated with a cytokine

Introduction cocktail to induce cardiogenic differentiation [10]. This trial investigated the safety and benefit of cardiopoietic stem cell therapy in patients with chronic heart failure and concluded with positive results.

With regard to lung injury, it has already been shown that pre-treatment of MSC with oxygen, N-acetylcysteine or TGFB1 had an increased therapeutic benefit in pre-clinical studies of oxygen-induced neonatal lung injury [254], bleomycin- induced lung injury [251] and LPS-induced lung injury [139].

Introduction

1.4. Aims of the study

Administration of MSC is a therapeutic option for polytrauma patients. In this thesis, the overarching aim was to investigate a potential benefit of BM-MSC and their mechanism of action in a rat blunt chest trauma (TxT) model.

The work was structured in the following parts: a potential effect of trauma factors on MSC in vitro should be investigated since this might help to understand changes in the proliferation, migration and secretion profile of administered exogenous resting MSC. Single factors as well as a cocktail of trauma factors were used to stimulate MSC. Proliferation, migration, expression profile and secretion profile of MSC were analysed to determine the potential role of MSC in trauma and to find trauma factors which have a significant effect on MSC. This work shall provide the basis for further research to analyse the effect of sera from polytrauma patients on MSC allowing the comparison of the effects of polytrauma cocktail on MSC to the effects of polytrauma sera on MSC.

Since Wistar rats were used in the TxT model, MSC from Wistar rats had to be isolated and characterized. The fate of human multipotent mesenchymal stromal cells (hMSC) and rat multipotent mesenchymal stromal cells (rMSC) as well as IL1B-primed MSC (IL1B pre-stimulated hMSC (hMSC (IL1B)) and IL1B pre- stimulated rMSC (rMSC (IL1B))) were analysed in co-culture with the rat alveolar macrophage cell line NR8383 and after in vivo application to the TxT model. Species-specific IL1B priming was performed to pre-stimulate MSC. To perform these experiments, labelling of hMSC and rMSC with a vital dye (PKH26) has to be established. Aims of this set of experiments were:  to isolate and characterize rMSC  to establish the PKH26 staining of hMSC and rMSC  to establish priming of hMSC and rMSC with IL1B  to perform in vitro co-culture experiments using the rat alveolar macrophage cell line NR8383 and rMSC or hMSC.

Thoracic injuries play a major role in polytrauma patients. We hypothesized that a rat TxT model will be a suitable model for pre-clinical studies and analysed the effect of a single dose of systemically administered MSC in this model after 24 h 14

Introduction and 5 d. Since this work shall provide pre-clinical data for the use of MSC in humans, we decided to use hMSC, i.e. the therapeutic product in development. On the other hand, allogeneic rMSC served as a control to the xenogeneic transplantation. Aims of the in vivo study were:  to clarify if hMSC or rMSC improve lung injury after TxT  to identify if IL1B priming of hMSC or rMSC improves the therapeutic effect of MSC after lung injury  to analyse the recovery and characteristics of administered hMSC or rMSC from blood and BAL fluids from rats.

Material and methods

2. Material and methods

2.1. Tools, machines, consumables and software

Tools and machines Axioskop 2 Mot Plus Carl Zeiss Microscopy GmbH, Göttingen

Axiovert 40 CFL Carl Zeiss Microscopy GmbH, Göttingen

Biobeam 8000 Gamma-Service Medical GmbH, Leipzig

Dewar Carrying Flasks KGW-Isotherm, Karlsruher Glastechnisches Werk - Schieder GmbH, Karlsruhe

Eclipse TS100 Inverted Routine Microscope Nikon GmbH, Düsseldorf

Eppendorf Research® pipette (0.1-2.5/0.5-10/2-20/10-100/100- Eppendorf Instrumente GmbH, Hamburg 1000 microlitre (µl))

Eppendorf Research® pipette 12-channel (30-300 µl) Eppendorf Instrumente GmbH, Hamburg

FACSAria BD Bioscience, Heidelberg

Fluorescence microscopy AF6000 Leica Mikrosysteme Vertrieb GmbH, Wetzlar

HERAcell 240i CO2 Incubator Thermo Fisher Scientific, Waltham, Massachusetts, USA

HERAsafe® HSP safety cabinets Thermo Fisher Scientific, Waltham, Massachusetts, USA

Leucodiff 700 Instrumentation Laboratory, Bedford, Massachusetts, USA

Liqui Chip for protein-based suspension arrays Qiagen, Hilden

Modified Boyden chamber, 48 well Neuro Probe, Gaithersburg, Maryland, USA

Multipette® plus Eppendorf Eppendorf Instrumente GmbH, Hamburg

NanoDrop1000 Thermo Fisher Scientific, Waltham, Massachusetts, USA Neubauer improved brightline Laboroptik, Lancing, UK

Pipetus® Hirschmann Laborgeräte GmbH & Co. KG, Eberstadt

Polarstar Omega Reader BMG Labtech GmbH, Ortenberg

Rotixa 50 RS Andreas Hettich GmbH & Co.KG, Tuttlingen

Shaker IKA®-Werke GmbH & Co. KG, Staufen

Shandon Elliott Cytospin centrifuge Shandon, London, UK

StepOnePlus Real-Time PCR system Thermo Fisher Scientific, Waltham, Massachusetts, USA Sunrise, spectrophotometer Tecan Group Ltd., Männedorf, Switzerland

ULTRA-TURRAX® T 25 IKA®-Werke GmbH & Co. KG, Staufen

Vortex-Genie 2 Scientific Industries, Inc., Bohemia, New York , USA

Water bath with circulation Köttermann GmbH & Co KG, Uetze/Hänigsen

Material and methods

Consumables 96 well cell culture plate, F-bottom, with lid, sterile Greiner Bio-One GmbH, Frickenhausen

96 Well Clear Polystyrene High Bind Stripwell™ Microplate Corning GmbH HQ, Wiesbaden

96 well microtiter plate Thermo Fisher Scientific, Waltham, Massachusetts, USA

Cell Strainer, 100 micrometre (µm) Nylon, sterile Becton Dickinson, Franklin Lakes, New Jersey, USA

Chamber Slide Flask Thermo Fisher Scientific, Waltham, Massachusetts, USA

Combi Stopper B. Braun Melsungen AG, Melsungen

Combitips advanced® 10 ml Eppendorf Vertrieb Deutschland GmbH, Wesseling- Berzdorf

Combitips plus 2.5 ml Eppendorf Vertrieb Deutschland GmbH, Wesseling- Berzdorf

Costar® 12 well clear tissue culture treated multiple well plates, Corning GmbH HQ, Wiesbaden Polystyrene

Costar® 6 well clear tissue culture treated multiple well plates, Corning GmbH HQ, Wiesbaden Polystyrene

Cryo tubes, 2 ml, Polypropylen Greiner Bio-One GmbH, Frickenhausen

Easy Flask, NunclonTM Delta Surface, 175 square centimetre Thermo Fisher Scientific, Waltham, Massachusetts, (cm2) USA

Easy Flask, NunclonTM Delta Surface, 25 cm2 Thermo Fisher Scientific, Waltham, Massachusetts, USA

Easy Flask, NunclonTM Delta Surface, 75 cm2 Thermo Fisher Scientific, Waltham, Massachusetts, USA

Enzyme-linked immunosorbent assays (ELISA) plate sealers for R&D systems GmbH, Wiesbaden-Nordenstadt 96 well plates

Eppendorf Safe-Lock Tubes, 0.5 ml Eppendorf Vertrieb Deutschland GmbH, Wesseling- Berzdorf epT.I.P.S. Reloads (20-300 µl, 50-1250 µl) Eppendorf Instrumente GmbH, Hamburg

Falcon® 15mL High Clarity Polypropylene Centrifuge Tube, Corning GmbH HQ, Wiesbaden Conical Bottom, with Dome Seal Screw Cap, Sterile

Falcon® 5 ml/10 ml/ 25 ml/ 50 ml/ 100 ml Serological Pipet, Corning GmbH HQ, Wiesbaden Polystyrene

Falcon® 50mL High Clarity Polypropylene Centrifuge Tube, Corning GmbH HQ, Wiesbaden Conical Bottom, Sterile

Falcon® 5mL Round Bottom Polystyrene Test Tube, with Cell Corning GmbH HQ, Wiesbaden Strainer Snap Cap

Filter papers Macherey-Nagel, Düren

FlipTube® - Polypropylene (PP), 1.5 ml Semadeni (Europe) AG, Düsseldorf

HTS FluoroBlokTM 96 well multiwell permeable support system Corning GmbH HQ, Wiesbaden with 8 µm high density PET membrane

Intravascular catheter Insyte-WTM Becton Dickinson Infusion Therapy Systems, Sandy, Utah, USA

Leukofix®, 9.2 m x 1.25 centimetre (cm) BSN medical GmbH, Hannover 17

Material and methods

Leukosilk®, 9.2 m x 2.5 cm BSN medical GmbH, Hannover

Manual filter pipette tips (2 µl, 10 µl, 100 µl, 1250 µl) San Diego, California, USA

Microscope cover glasses Marienfeld GmbH & Co. KG, Lauda-Königshofen

Microscope slides, frosted end Thermo Fisher Scientific, Waltham, Massachusetts, USA

Mylar® Polyester foil, 150A Du Pont de Nemours, Contern, Luxembourg

NOBATOP® 12 non-woven swabs NOBAMED Paul Danz AG, Wetter

Safety Multifly® cannula 21G with long tube and assembled Sarstedt AG & Co, Nümbrecht multi-adapter

Shandon™ Single Cytoslides™; noncoated, circle on back Thermo Fisher Scientific, Waltham, Massachusetts, USA

Single-use needle (20G, 26G) B. Braun Melsungen AG, Melsungen

S-Monovette® haematology, EDTA (ethylenediaminetetraacetic Sarstedt AG & Co, Nümbrecht acid) K3, 7.5 ml

Surgical disposable carbon steel scalpel, #11, #21 Aesculap AG, Tuttlingen

Syringe, 1 ml Becton Dickinson, Franklin Lakes, New Jersey, USA

Syringe, 20 ml, Luer-LokTM Tip Becton Dickinson, Franklin Lakes, New Jersey, USA

Syringe, 3 ml, Luer-LokTM Tip Becton Dickinson, Franklin Lakes, New Jersey, USA

Test tubes, 1.5 ml, colourless, sterile Biozym Scientific GmbH, Hessisch Oldendorf

Transwell® with 0.4µm Pore Polycarbonate Membrane Insert, Corning, Kennebunk, Maine, USA Sterile

Triple Flask, NunclonTM Delta Surface, 500 cm2 Thermo Fisher Scientific, Waltham, Massachusetts, USA

VICRYL surgical suture material, absorbable Johnson & Johnson Medical GmbH, Norderstedt

White Filter Cards Thermo Fisher Scientific, Waltham, Massachusetts, USA

Z Serum Sep Clot Activator Greiner Bio-One GmbH, Frickenhausen

Software

AxioVision V4.82 Carl Zeiss Microscopy GmbH, Göttingen

Bio-Plex Manager Software 4.1.1 Bio-Rad Laboratories GmbH, Munich

FACSDiva V6.1.3 BD Bioscience, Heidelberg

Graph Pad Prism 7 GraphPad Software, Inc., La Jolla, California, USA

Leica Application Suite AF Leica Mikrosysteme Vertrieb GmbH, Wetzlar, Germany

MARS Omega Data Analysis 1.11 BMG Labtech GmbH, Ortenberg

StepOne Software V2.32 Thermo Fisher Scientific, Waltham, Massachusetts, USA

Material and methods

2.2. Chemicals and kits

Chemicals

2-Propanol VWR international, Fontenay-sous-Bois, France

3,3′,5,5′-Tetramethylbenzidine (TMB) Sigma-Aldrich, Munich

3-Isobutyl-1-methylxanthine Sigma-Aldrich, Saint Louis, Missouri, USA

AB-serum IKT Ulm, Ulm

Albumin Standard Ampules, 2 milligram (mg) per ml Thermo Fisher Scientific, Waltham, Massachusetts, USA

Albumin, bovine Sigma-Aldrich, Saint Louis, Missouri, USA

Minimal Essential Medium Eagle, Alpha modification (Alpha Lonza, Verviers, Belgium MEM)

Annexin Binding Buffer (5x) Thermo Fisher Scientific, Waltham, Massachusetts, USA

Annexin V, Allophycocyanin (APC) conjugate Thermo Fisher Scientific, Waltham, Massachusetts, USA

Aqua iniectabilia B. Braun Melsungen AG, Melsungen

BD FACS™ Lysing Solution (10X) BD Biosciences, San Jose, California, USA

β-Glycerophosphate disodium salt hydrate Sigma-Aldrich Chemie GmbH, Steinheim

Bicinchoninic acid solution Sigma-Aldrich, Munich

Biocoll separating solution Biochrom GmbH, Berlin

Buprenorphin Temgesic® ESSEX PHARMA GmbH, Munich

Ciprofloxacin Kabi 200 mg/100 ml Fresenius Kabi Deutschland GmbH, Bad Homburg

Compressed air for respiration MTI IndustrieGase AG, Neu-Ulm

Copper(II) sulfate anhydrous (CuSO₄) Merck Millipore, Darmstadt Dexamethasone Sigma-Aldrich, Saint Louis, Missouri, USA

Dimethyl sulfoxide (DMSO), ≥ 99.9% Sigma-Aldrich, Saint Louis, Missouri, USA

DMSO Fluka, Munich

Dinatriumhydrogenphosphat (Na2HPO4) Merck, Darmstadt

Dipotassium hydrogen phosphate (K₂HPO₄) Merck Millipore, Darmstadt Dulbecco's Modified Eagle's Medium, high glucose, pyruvate Thermo Fisher Scientific, Waltham, Massachusetts, USA

Dulbecco´s phosphate buffered saline (DPBS) without Ca++ and Lonza, Verviers, Belgium Mg++

Ethanol Sigma-Aldrich, Saint Louis, Missouri, USA

EUKITT® mounting medium ORSAtec GmbH, Bobingen

FBS, heat inactivated Thermo Fisher Scientific, Waltham, Massachusetts, USA

Formaldehyde, 37 % AppliChem, Darmstadt

Giemsa´s azur eosin methylene blue solution Merck, Darmstadt

GlutaMAX™ Supplement (100x) Thermo Fisher Scientific, Waltham, Massachusetts, USA

Ham's F-12K (Kaighn's) Medium Thermo Fisher Scientific, Waltham, Massachusetts, USA 19

Material and methods

Hematoxilin Solution, Harris Modified Sigma-Aldrich Chemie GmbH, Steinheim

Heparin-Natrium 25000 I.U. ratiopharm GmbH, Ulm

Hexadecyltrimethylammonium bromide Sigma-Aldrich, Munich

Humanalbin®, human serum albumin (HSA, 50 g/l) CSL Behring GmbH, Marburg

Hydrochloric acid (HCl) Sigma-Aldrich, Saint Louis, Missouri, USA

Hydrogen peroxide (H2O2), 30 % Fischar, Saarbrücken Immersion oil 518C Carl Zeiss Microscopy GmbH, Göttingen

Indomethacin Sigma-Aldrich, Saint Louis, Missouri, USA kodan® tincture colourless Schülke & Mayr GmbH, Norderstedt

L-Ascorbic acid 2-phosphate sesquimagnesium salt hydrate Sigma-Aldrich, Saint Louis, Missouri, USA

LPS from Escherichia coli Sigma-Aldrich, Saint Louis, Missouri, USA

Löffler's methylene blue solution Merck Millipore, Darmstadt

May-Grünwald´s eosine methylene blue solution modified for Merck, Darmstadt microscopy

Medical oxygen MTI IndustrieGase AG, Neu-Ulm

Methanol Merck, Darmstadt

Myeloperoxidase (MPO), human polymorphonuclear leucocytes Merck Millipore, Darmstadt (PML)

Natriumdihydrogenphosphat (NaH2PO4 H2O) Merck, Darmstadt Neo-Clear® Merck KGaA, Darmstadt

Nuclear fast red Sigma-Aldrich, Saint Louis, Missouri, USA

Oil Red O Sigma-Aldrich Chemie GmbH, Steinheim

Penicillin Streptomycin (10,000 units (U) per mL) Thermo Fisher Scientific, Waltham, Massachusetts, USA

Perchloric acid Sigma-Aldrich, Munich

Platelet lysate IKT Ulm, Ulm

Potassium dihydrogen phosphate (KH₂PO₄) Merck Millipore, Darmstadt

Potassium hexacyanoferrate(II) trihydrate (K4Fe(CN)6 3H2O) Sigma-Aldrich, Saint Louis, Missouri, USA Potassium phosphate monobasic Sigma-Aldrich Chemie GmbH, Steinheim

Protease inhibitor cocktail Sigma-Aldrich, Saint Louis, Missouri, USA

Quantikine Wash Buffer 1 R&D systems GmbH, Wiesbaden-Nordenstadt

Reagent Diluent Concentrate 2 R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant human C3a EMD Chemicals, San Diego, California, USA

Recombinant human complement component 5a (C5a) des-Arg EMD Chemicals, San Diego, California, USA

Recombinant human FGF2 R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant human high mobility group box 1 (HMGB1) /HMG1 R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant human insulin-like growth factor 1 (IGF1) R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant human IL1B PeproTech Germany, Hamburg

Recombinant human IL6 BIOMOL GmbH, Hamburg

Recombinant human insulin, zinc solution Thermo Fisher Scientific, Waltham, Massachusetts, USA

Recombinant human homodimer of platelet-derived growth factor R&D systems GmbH, Wiesbaden-Nordenstadt (PDGF-BB)

Material and methods

Recombinant human TGFB1 R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant human thrombomodulin (THBD)/BDCA-3 R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant human thrombopoietin (THPO) R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant Human TSG-6 Protein R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant human/ mouse/ rat activin A R&D systems GmbH, Wiesbaden-Nordenstadt

Recombinant rat IL1B PeproTech Germany, Hamburg

SevoraneTM (Sevoflurane) Abbott GmbH & Co. KG, Wiesbaden

SIGMAFASTTM BCIP®/NBT tablet Sigma-Aldrich, Saint Louis, Missouri, USA

Sodium hydroxide Sigma-Aldrich Chemie GmbH, Steinheim

Stop Solution 2N Sulfuric Acid R&D systems GmbH, Wiesbaden-Nordenstadt

Streptavidin-HRP R&D systems GmbH, Wiesbaden-Nordenstadt

Substrate Reagent Pack R&D systems GmbH, Wiesbaden-Nordenstadt

SYBR® Green Master Mix Thermo Fisher Scientific, Waltham, Massachusetts, USA

SYTOX Green dead cell stain Thermo Fisher Scientific, Waltham, Massachusetts, USA

Trypan blue solution (0.4 %) Sigma-Aldrich Chemie GmbH, Steinheim

Trypsin 2.5 % (10x) Thermo Fisher Scientific, Waltham, Massachusetts, USA

Trypsin-EDTA (0.05 %), phenol red Thermo Fisher Scientific, Waltham, Massachusetts, USA

Water for chromatography, LiChrosolv® water Merck Millipore, Molsheim

Kits CellTraceTM carboxyfluorescein diacetate succinimidyl ester Thermo Fisher Scientific, Waltham, Massachusetts, (CFSE) Cell Proliferation Kit USA

CyQuant® Cell Proliferation Assay Kit for cells in culture Thermo Fisher Scientific, Waltham, Massachusetts, USA

Human Albumin ELISA Kit Abcam, Cambridge, UK

Human Cytokine/Chemokine Magnetic Bead Panel - Immunology Merck Millipore, Molsheim Multiplex Assay

Human MMP Magnetic Bead Panel 2 Merck Millipore, Molsheim

Mouse IL-1ra/IL-1F3 Quantikine ELISA Kit R&D systems GmbH, Wiesbaden-Nordenstadt

Mouse/Rat FGF basic Quantikine ELISA Kit R&D systems GmbH, Wiesbaden-Nordenstadt

PKH26 Red Fluorescent Cell Linker Mini Kit for General Cell Sigma-Aldrich, Saint Louis, Missouri, USA Membrane Labelling

Prostaglandin E2 Parameter Assay Kit R&D systems GmbH, Wiesbaden-Nordenstadt

QIAamp deoxyribonucleic acid (DNA) Mini Kit Qiagen GmbH, Hilden

Rat C5a ELISA Kit MyBioSource, San Diego, California, USA

Rat expanded Cytokine/Chemokine Magnetic Bead Panel - Merck Millipore, Molsheim Immunology Multiplex Assay

Rat IL17A ELISA Ready-SET-Go!® Thermo Fisher Scientific, Waltham, Massachusetts, USA

Material and methods

2.3. Antibodies

Human TSG-6 Biotinylated Antibody R&D systems GmbH, Wiesbaden-Nordenstadt

TSG-6 Antibody (A38.1.20) Santa Cruz Biotechnology, Heidelberg

Antigen Isotype Label Clone Company

Anti-human CD105 Mouse IgG1 FITC SN6 AbD Serotec, Puchheim

Anti-human CD105 Mouse IgG1, APC 266 BD Bioscience, Heidelberg kappa Anti-human CD3 Mouse BALB/c PerCP SK7 BD Bioscience, Heidelberg IgG1, kappa Anti-human CD45 Mouse IgG1, PerCP 2D1 BD Bioscience, Heidelberg kappa Anti-human CD73 Mouse IgG1, APC AD2 BD Bioscience, Heidelberg kappa Anti-human CD90 Mouse IgG1, FITC 5E10 BD Bioscience, Heidelberg kappa Anti-human CD110 Mouse IgG2a APC 167639 R&D systems GmbH, Wiesbaden-Nordenstadt

Anti-mouse/rat CD29 Hamster IgG APC- HMβ1-1 Biolegend, San Diego, California, USA Cy7 Anti-mouse/rat CD49e Hamster IgG PE HMα5-1 Biolegend, San Diego, California, USA

Anti-rat CD103 Mouse IgG1 FITC OX-62 Biolegend, San Diego, California, USA

Anti-rat CD106 Mouse IgG1, PE MR106 Biolegend, San Diego, California, USA kappa Anti-rat CD11b/c Mouse IgG2a, APC OX-42 Biolegend, San Diego, California, USA kappa Anti-rat CD172a Mouse IgG2a, PE OX-41 Biolegend, San Diego, California, USA kappa Anti-rat CD3 Mouse IgM, kappa APC 1F4 Biolegend, San Diego, California, USA

Anti-rat CD31 Mouse IgG1 FITC TLD- AbD Serotec, Puchheim 3A12 Anti-rat CD44H Mouse IgG2a, FITC OX-49 Biolegend, San Diego, California, USA kappa Anti-rat CD45 Mouse IgG1, APC- OX-1 Biolegend, San Diego, California, USA kappa Cy7 Anti-rat CD45, Mouse IgG1, APC OX-1 Thermo Fisher Scientific, Waltham, eBioscienceTM kappa Massachusetts, USA Anti-rat CD49d Mouse IgG2a, FITC MRα4-1 Biolegend, San Diego, California, USA kappa Anti-rat CD54 Mouse IgG1 FITC 1A29 AbD Serotec, Puchheim

Anti-rat CD54 Mouse IgG1, PE 1A29 Biolegend, San Diego, California, USA kappa Anti-rat CD90/mouse Mouse IgG1, PerCP OX-7 Biolegend, San Diego, California, USA CD90.1 kappa Isotype control Mouse IgG2a, FITC MOPC- Biolegend, San Diego, California, USA kappa 173 Isotype control Mouse IgG1, PerCP MOPC- Biolegend, San Diego, California, USA kappa 21

Material and methods

Isotype control Mouse IgG2a, APC MOPC- Biolegend, San Diego, California, USA kappa 173 Isotype control Mouse IgG2a APC 20102 R&D systems GmbH, Wiesbaden-Nordenstadt

Isotype control Hamster IgG APC- HTK888 Biolegend, San Diego, California, USA Cy7 Isotype control Mouse IgG1, FITC X40 BD Bioscience, Heidelberg kappa Isotype control Mouse IgG1, PerCP X40 BD Bioscience, Heidelberg kappa Isotype control Mouse IgG1, APC MOPC- BD Bioscience, Heidelberg kappa 21 Isotype control Mouse IgG1, APC- MOPC- Biolegend, San Diego, California, USA kappa Cy7 21 Isotype control Mouse IgG1, PE MOPC- Biolegend, San Diego, California, USA kappa 21 Isotype control Hamster IgG PE HTK888 Biolegend, San Diego, California, USA

2.4. Cells

C2C12 (Mouse C3H muscle myoblast cell line) p14 Sigma-Aldrich, Saint Louis, Missouri, USA NR8383 [AgC11x3A, NR8383.1] (ATCC® CRL-2192™) ATCC LGC Standards GmbH, Wesel

2.5. Cultivation of mesenchymal stromal cells, NR8383 cells and C2C12 cells

Human mesenchymal stromal cells Human MSC were obtained from bone marrow aspirates (iliac crests) of volunteer healthy donors. The following section was already partly published in Amann et al. on page 3 [4]. Declaration of consent of healthy donors was obtained according to the Declaration of Helsinki and collection of this material has been approved by the Ethics Committee of Ulm University (Ethical approval numbers: 32/10, 24/11). Isolation, characterization and expansion of hMSC were performed according to standardized GMP-compliant protocols as described previously [71]. For cryopreservation, cells were harvested with 0.5 % trypsin and cell pellet was suspended in Alpha MEM supplemented with 8 % PL, 1 international units (IU) per ml heparin and 10 % DMSO. For in vitro experiments hMSC were used at passage one to four. For rat experiments cryopreserved hMSC were used at passage one and cultured in Alpha MEM supplemented with 8 % PL and 1 IU/ml heparin.

Material and methods

Rat mesenchymal stromal cells Rat MSC were obtained from the femur of male Wistar rats as described previously [244]. After euthanasia of the animals, femurs were prepared from surrounding tissue with a scalpel. The bones were disinfected by a short ethanol bath. The ends of the bones were opened sterile by a bone cutter. Bone marrow was rinsed with 1 millilitre (ml) Alpha MEM and centrifuged at 400 standard gravity (g) for 10 minutes (min). The cell pellet was suspended in Alpha MEM supplemented with 15 % FBS, 2 millimolar (mM) GlutaMAX, 100 U/ml Penicillin and 100 microgram (µg) per ml Streptomycin. At the beginning of culturing the medium was once additionally supplemented with 12 µg/ml ciprofloxacin. Bone marrow of one rat was seeded in 75 cm² cell culture flask and incubated at 37 °C in an atmosphere of 5 % humidified CO2. After 3 d the cells, which did not adhere, were removed by washing the cell layer three times with DPBS. 40 % of the medium was replaced every third to fourth days. The cells were harvested by using 0.05 % Trypsin-EDTA. For cryopreservation, cell pellet was suspended in 90 % FBS and 10 % DMSO. Cyropreserved rMSC were used at passages three or four for cell culture experiments and intravenous injection in rat experiments. Rat MSC were characterized by tri-lineage differentiation and flow cytometry. Adipogenic, chondrogenic, osteogenic differentiation assays were performed as described previously [65]. In brief, rMSC at passage one to four were seeded in Chamber Slide Flasks and were either stimulated with differentiation medium or cultured in standard rat cell culture medium (control) for a different length of time. Medium was exchanged every third to fourth day. Induction of adipogenic differentiation was started after rMSC grow confluent. Induction medium of adipogenic differentiation medium was supplemented with 3-Isobutyl-1- methylxanthine, dexamethasone, recombinant human insulin and indomethacin. Maintenance medium of adipogenic differentiation medium was only supplemented with insulin. Osteogenic differentiation medium was supplemented with dexamethasone, L-ascorbic acid 2-phosphate and β-glycerophosphate. Chondrogenic differentiation medium was supplemented with L-ascorbic acid 2- phosphate, TGFB1 and insulin. After different length of time cells were fixed with methanol or formaldehyde and stained with Oil RedO/haematoxylin staining, methylene blue staining or treated with SIGMAFASTTM BCIP®/NBT tablet for detection of alkaline phosphatase.

Material and methods

Table 2 shows the staining panel for the characterization of rMSC. Flow cytometry analysis was done using a FACSAria with FACSDiva Software.

Table 2 Flow cytometry staining panels for characterization of rMSC. Staining approaches included isotype control (1) and combination of positive and negative rat CD markers of MSC (2-6) with different fluorochromes (FITC: fluorescein isothiocyanate; PE: phycoerythrin; PerCP: peridinin- chlorophyll-protein; APC: allophycocyanin; APC-Cy7: allophycocyanin coupled with the cyanine dye Cy7). All of them were mouse anti-rat antibodies, except for §-labelled, which were hamster anti-rat antibodies. Ig: immunoglobulin; CD: cluster of differentiation.

Fluorochrome/ FITC PE PerCP APC APC-Cy7 Approaches

IgG IgG IgG IgG IgG 1 5 µl 20 µl 5 µl 5 µl 20 µl

CD44H CD106 CD90 CD11b/c CD45 2 2 µl 5 µl 2 µl 5 µl 20 µl

CD31 CD54 IgG § 3 5 µl 5 µl 5 µl

CD49d CD172a CD29 § 4 5 µl 5 µl 5 µl

CD103 IgG § 5 5 µl 5 µl

CD49e § 6 5 µl

Cultivation of CRL-2192 NR8383 rat alveolar macrophage cells Cell cultivation was performed according to manufacturer’s protocol. Cell culture medium was Ham's F12K medium supplemented with 15 % FBS.

Cultivation of C2C12 cell line Cell cultivation was performed according to manufacturer’s protocol. Cell culture medium was Dulbecco's Modified Eagle's Medium supplemented with 10 % FBS and 100 µg/ml Streptomycin. C2C12 cells were harvested and seeded in 25 cm² cell culture flask (1.2 x 106 MSC). After overnight incubation, C2C12 cells were irradiated with 60 Gray (Gy) by the use of Biobeam 8000. Two days after seeding medium of the C2C12 cells was replaced with rat cell culture medium (Alpha MEM supplemented with 15 % FBS, 2 mM GlutaMAX, 100 U/ml Penicillin, 100 µg/ml Streptomycin and 25

Material and methods

12 µg/ml ciprofloxacin). The C2C12 cells were then ready to use as feeder layer for cell culture of sorted cells (see section 2.8.8.).

2.6. Mesenchymal stromal cells and polytrauma

To investigate the role of MSC in trauma, MSC were stimulated with signal molecules and proliferation, migration, gene expression and cytokine profile were analysed. Furthermore, sera from polytrauma patients were analysed by Multiplex assay.

2.6.1. Composition of polytrauma cocktail

The polytrauma cocktail (table 3) included inflammatory cytokines (IL1B, IL6, HMGB1), complement components (C3a, C5a), THBD as an indicator for endothelial damage and THPO as a regulator of platelet count and activation.

Table 3 Composition of polytrauma cocktail low (PTCL) and polytrauma cocktail high (PTCH) in the in vitro cell culture experiments. IL1B: interleukin 1 beta; IL6: interleukin 6; C3a: complement component 3a; C5a: complement component 5a; HMGB1: high mobility group box 1; THBD: thrombomodulin; THPO: thrombopoietin.

Cytokine/Complement factor PTCL PTCH IL1B 300 pg/ml 10 ng/ml IL6 300 pg/ml 10 ng/ml C3a 500 ng/ml 500 ng/ml C5a 10 ng/ml 100 ng/ml HMGB1 10 ng/ml 60 ng/ml THBD 100 ng/ml 300 ng/ml THPO 150 pg/ml 450 pg/ml

Polytrauma cocktail low (PTCL) included pathophysiological concentrations of cytokines and other trauma factors in serum or plasma (see table 4). Polytrauma cocktail high (PTCH) included higher concentrations of these factors. The trauma factors IL1B, IL6 and C5a had supraphysiological concentrations in the PTCH.

Material and methods

Table 4 Plasma or serum level of trauma factors after multiple injuries as described in literature. Time refers to admission to hospital or time of diagnosis. ALI: Acute Lung Injury; ARDS: Acute Respiratory Distress Syndrome; ISS: Injury Severity Score; IL1B: interleukin 1 beta; IL6: interleukin 6; C3a: complement component 3a; C5a: complement component 5a; HMGB1: high mobility group box 1; THBD: thrombomodulin; THPO: thrombopoietin.

Patient population Serum/ Analyte Concentration Time Reference Plasma ARDS, survivors (n=9) plasma IL1B 313 pg/ml 24 h [167] IL6 407 pg/ml ARDS, survivors (n=6) serum IL6 213 pg/ml 0 h [21] Polytraumatized patients, serum C3a 20 ng/mg serum 0 h or [26] mean ISS of 30 (n=40) protein 24 h C5a 0.25 ng/mg serum protein 0.1 ng/mg serum protein ARDS (n=12) plasma C3a 1000 ng/ml 4-12 h [173] Polytraumatized patients with plasma C3a 430 ng/ml 0 h [273] or without ARDS (n=30), 230 ng/ml 24 h ISS>30 Severe injury (n=168), mean plasma HMGB1 10 ng/ml 0 h [39] ISS 17 (range 9-26) (ISS>25) Trauma patients with serum HMGB1 54 ng/ml 0-6 h [266] Haemorrhagic shock patients 57 ng/ml 24 h (n=25) ALI or ARDS (n=45) plasma THBD 250 ng/ml 0 h [253] ARDS (n=449) plasma THBD 123 ng/ml ~ 0 h [219] Polytraumatized patients serum THPO 207 pg/ml 0 h [101] (n=17), ISS>30 517 pg/ml 4 d

Recombinant factors were reconstituted and stored according to manufacturer’s instructions.

2.6.2. Proliferation assay

The effect of trauma factors on MSC proliferation was analysed by CyQuant Cell Proliferation Assay. MSC at passage two or three were harvested and 200 cells per well were seeded in 96 well cell culture plate. Each well contained 100 µl Alpha MEM supplemented with different percentages (0 %, 0.6 %, 1.2 %, 8 %) of PL. Due to the fact, that PL supplementation had been shown to have a stronger 27

Material and methods influence on proliferation of MSC than for example FBS-supplementation [72], the percentage of PL was reduced to 0.6 % or 1.2 % in proliferation assay. MSC adhered for about 4 h, afterwards 10 µl DPBS without and with factors (PTCH, PTCH without THBD, PTCL, IL1B, HMGB1, THBD, THPO) were added to each well. A growth factor cocktail [72] including activin A (5 nanogram (ng)/ml), FGF2 (20 ng/ml), PDGF-BB (20 ng/ml), IGF1 (400 ng/ml) and dexamethasone (1µmol/l) was also used as a control. Alpha MEM supplemented with 8 % PL was used as positive control. Alpha MEM without any supplementation was used as negative control. All experiments were performed in triplicates. Stimulation was repeated two times on two consecutive days. After cell growth for 6 to 7 d, cell culture medium was removed and adherent cells were washed once with DPBS. The cell culture plate was frozen at -70 °C for at least for one day. The following section was already published in Amann et al. on page 5 [4]. MSC were lysed and stained using CyQuant Cell Proliferation Assay according to manufacturer’s instructions. A cell standard with 0 to 5 x 104 cells (frozen on the first day of the experiment) was pipetted into same plate. The plate was measured with bottom optic settings by the Polarstar Omega Reader (485/520 nanometre (nm)).

2.6.3. Migration assays and flow cytometry analysis of CD110

Transwell migration assay Chemotaxis was analysed using the BD Falcon HTS FluoroBlok Insert System for 96 well plates. The cells were separated from chemoattractants by 8.0 µm light- tight polyethylene terephthalate (PET) membrane which blocks the transmission of light between 400 nm to 700 nm. MSC (1.5 x 106 MSC/ml) at passage one or two were harvested and stained with 10 µM CFSE in 0.1 % HSA/DPBS at 37 °C for 10 min. After a centrifugation step, the MSC were washed at 37 °C for 30 min in the 4.0 to 5.0-fold of the staining volume. After a second centrifugation step, 7.5 x 104 cells in 50 µl of 0.1 % HSA/DPBS were added to the upper chamber and incubated for 15 min. 225 µl of 0.1 % HSA/DPBS without (negative control) or with recombinant factors (PTCH, PTCH without THBD, PTCL, IL1B, C3a, HMGB1, THBD, THPO) were put in the lower chamber. Positive controls were 10 % FBS or 8 % PL. The experiments were performed in triplicates. Kinetics of migration

Material and methods

(0 min, 15 min, 30 min, 45 min, 60 min, 75 min, 90 min) were measured at 37 °C with bottom optic settings by the Polarstar Omega Reader (485/520 nm).

Modified Boyden chamber To validate the transwell migration assay, two experiments were performed with the modified Boyed chamber in the Division for Biochemistry of Joint and Connective Tissue Diseases as described previously [99]. Polycarbonate filters with 8 µm pores separated the MSC (1 x 104 cells/well in 50 µl Alpha MEM) at passage three (upper chamber) from the chemoattractants (28.5 µl in lower chamber). Negative control was Alpha MEM. Positive control was the cell culture supernatant. The experiments were performed with four-fold determination. After an incubation time of 4 h, MSC on the lower side of the filter were fixed, stained with Giemsa solution and counted at 20-fold magnification.

Flow cytometry analysis of CD110 Furthermore, MSC at passage two to four were analysed by flow cytometry to determine the incidence of CD110 on cell surface. Staining protocol was performed according to manufacturer’s instructions.

2.6.4. Ribonucleic acid sequencing and cytokine analysis in supernatant

To determine the effect of PTCH and PTCL on the migratory ability and secretion profile of MSC, ribonucleic acid (RNA) sequencing Multiplex assay and ELISA were performed. In addition, the effect of IL1B stimulation (10 ng/ml IL1B) was analysed. MSC were seeded at passage two with 2.5 x 104 cells per cm² in 1 ml Alpha MEM supplemented with 8 % PL and 1 IU/ml heparin in 12 well cell culture multidishes. After overnight incubation medium of cells was changed. Five different experimental approaches including five different cell culture media: 1. Alpha MEM supplemented with 8 % PL and 1 IU/ml heparin 2. Alpha MEM supplemented with 20 % AB-serum 3. Alpha MEM supplemented with 20 % AB-serum and 10 ng/ml IL1B 4. Alpha MEM supplemented with 20 % AB-serum and PTCH 5. Alpha MEM supplemented with 20 % AB-serum and PTCL

Material and methods

For statistical reasons the second approach was carried out twice. After 6 h and 24 h, supernatants were collected and stored at -70 °C until cytokine analysis. Cells were washed with 500 µl DPBS and harvested with 250 µl 0.5 % trypsin. Cells were transferred in 1.5 ml sterile test tubes, centrifuged and cell pellet was frozen at -70 °C until RNA isolation was performed.

RNA isolation and RNA sequencing of coding RNA of MSC were performed by the Genomics-Core Facility of Ulm University Medical Center. Differential gene expression of biological triplicates and two different times were analysed by the Institute for Medical Systems Biology of Ulm University. Criteria for differential analysis were a fold change of at least three for up or down regulation, a false discovery rate less than 0.001 and a read count of at least 10 for the respective target gene.

Supernatants were analysed by Human MMP Magnetic Bead Panel 2 (assay sensitivity for MMP1: 3.0 pg/ml; MMP2: 200 pg/ml; MMP9: 2.0 pg/ml; MMP10: 5.0 pg/ml) and by Human Cytokine/Chemokine Magnetic Bead Panel (assay sensitivity for TNF: 0.7 pg/ml; IL10: 1.1 pg/ml; IL1RN: 8.3 pg/ml; IL1B: 0.8 pg/ml; IL6: 0.9 pg/ml; CXCL8: 0.4 pg/ml; CXCL1: 9.9 pg/ml; CCL2: 1.9 pg/ml; VEGFA: 26.3 pg/ml; FGF2: 7.6 pg/ml; IL17A: 0.7 pg/ml; CCL4: 3.0 pg/ml). Supernatants were diluted 1:4 and 1:8 for MMP assay and were used undiluted for cytokine/chemokine assay. Assay procedures were performed according to manufacturer´s instructions. Controls were included. Analysis was performed with the Liqui Chip device and Bio-Plex Manager Software. To analyse the secretion profile of MSC, cytokine levels of cell culture medium without MSC were subtracted from cytokine levels of supernatant from cell culture.

The following section was already partly published in Amann et al. on page 5 [4]. TNFAIP6 ELISA was performed as previously described [99]. Briefly, High Bind Stripwell™ Microplate was coated with TSG-6 antibody. After overnight incubation, microplate was washed four times with Quantikine Wash Buffer and blocking buffer (Reagent Diluent) was added for 1 h. After a second washing, step standard and samples were added in duplicate and incubated on plate shaker for 2 h. Alpha MEM without and with supplementation was measured on every plate and used as

Material and methods a control. Samples were diluted 1:30 with blocking buffer. After a third washing step, TSG-6 Biotinylated Antibody was incubated on plate shaker for 2 h. After a fourth washing step, Streptavidin-HRP was added and incubated for 20 min in the dark. After a fifth washing step, Substrate Reagent was added for about 30 min. Afterwards Stop Solution was added and microplate was measured immediately at 450 nm with wavelength correction at 570 nm. Analysis of ELISA was performed with Polarstar Omega Reader, MARS Omega Data Analysis 1.11 and linear regression. To analyse the secretion profile of MSC, cytokine level of cell culture medium without MSC was subtracted from cytokine level of supernatant from cell culture.

2.6.5. Cytokine analysis of serum

Serum from polytrauma patients The study was performed in collaboration with the Institute for Clinical and Experimental Trauma-Immunology of Ulm University. The following section was already published in Amann et al. on page 2 and 3 [4]. The study was done in accordance with the Declaration of Helsinki, with approval of the Ethics Committee of Ulm University (polytrauma patients: vote number 244/11 and 94/14), and with informed consent of all patients (age: 49 years, 33 years, 47 years, 49 years, 26 years). The inclusion criterion for the study was an ISS ≥ 32. After admission to hospital (0 h) blood was taken from patients at certain times (0 h, 4 h, 12 h, 24 h, 48 h, 120 h, 240 h), then the blood clotted at 4°C for 2 h. Afterwards, tubes were centrifuged at 1560 g for 10 min and stored in 0.5 ml Safe-Lock tubes at -80°C.

Serum from healthy volunteers The following section was already published in Amann et al. on page 3 [4]. As control for cytokine measurement, healthy volunteers (age: 22 years, 24 years, 30 years, 49 years, 52 years) had donated blood. The study was done in accordance with the Declaration of Helsinki, with approval of the Ethics Committee of Ulm University (vote number 209/17) and with informed consent of all volunteers. Their blood was treated as described above.

Material and methods

Analysis of AB-serum, serum from healthy volunteers and from polytrauma patients Different cytokines were analysed in serum from healthy volunteers, commercially available AB-serum (serum pooled from 20 donors) and serum from polytrauma patients. Human cytokines and chemokines (assay sensitivity see section 2.6.4., CCL4 level was not determined, further assay sensitivities are IFNG: 0.8 pg/ml; IL12p40: 7.4 pg/ml; IL12p70: 0.6 pg/ml; IL13: 1.3 pg/ml; IL15: 1.2 pg/ml; interleukin 1 alpha (IL1A): 9.4 pg/ml; IL2: 1.0 pg/ml; IL4: 4.5 pg/ml; IL5: 0.5 pg/ml; CCL3: 2.9 pg/ml) were analysed using a Merck Millipore Multiplex assay platform. Assay procedure was performed according to manufacturer´s instructions. Serum from healthy volunteers and AB-serum were analysed for most of the cytokines described above, except for IL2, IL4, IL12p40, IL12p70 and IL17A. Human Albumin ELISA Kit was performed according to manufacturer´s instructions. Samples were diluted 1:10000 and analysis of ELISA was performed with Polarstar Omega Reader.

2.7. Effects of manipulating MSC for in vivo experiments

Previous results showed that MSC could be activated by IL1B stimulation [28,99]. MSC were stained with PKH26 and primed with IL1B in a rat TxT model (see section 2.8.1.). Characteristics of PKH26-stained MSC and IL1B-primed MSC are described below in detail.

2.7.1. Fluorescence intensity and cell growth of PKH26-stained mesenchymal stromal cells

MSC were harvested and stained with the PKH26 red fluorescent cell linker kit. Comparison of the manufacturer protocol (PKH26 dye concentration 2.0 x 10-6 molar (M) in 100 µl staining solution) and the staining protocol which was ultimately used in rat experiment (PKH26 dye concentration 1.6 x 10-5 M in 100 µl staining solution) was performed. Staining was performed according to the manufacturer instruction with two different dye concentrations. The higher PKH26 dye concentration is based on PKH26 staining protocol from Deak et al. [53], though cell concentration in our protocol was 10-fold lower. Unstained Cells were used as control. Flow cytometry analysis was performed to verify PKH26 staining. 32

Material and methods

Comparison included determination of fluorescence intensity over time (0 h, 24 h, 48 h and 120 h after staining), colony forming units (CFU) and doubling time. For these experiments hMSC were used at passage three or four and rMSC were used at passage four.

Fluorescence intensity over time Unstained and PKH26-stained cells were seeded with the same seeding density. Human MSC were seeded with a cell density of 2.0 x 104 cells per cm2. Rat MSC were seeded with a cell density of 2.5 x 104 cells per cm². Cells were cultured for 24 h, 48 h and 120 h. For analysis of fluorescence intensity, cells were harvested and about 30000 events were measured by flow cytometry (FACSAria). Analysis of data was performed with FACSDiva software.

Doubling time Human MSC were seeded with a cell density of 240 cells per cm2. Rat MSC were seeded with a cell density of 1.5 x 104 cells per cm². Medium change occurred two times within 7 d of cell culture. After that time, MSC were harvested and counted in Neubauer counting chamber. Doubling time was calculated with the following formula. ℎ ln(2) × 푑푎푦푠 표푓 푐푢푙푡푢푟푒 [푑] × 24 [ ] 푑표푢푏푙𝑖푛푔 푡𝑖푚푒 [ℎ] = 푑 푡표푡푎푙 푛푢푚푏푒푟 표푓 ℎ푎푟푣푒푠푡푒푑 푐푒푙푙푠 [푐푒푙푙푠] ln ( ) 푡표푡푎푙 푛푢푚푏푒푟 표푓 푀푆퐶 푠푒푒푑푒푑 푐푒푙푙푠 [푐푒푙푙푠] Calculation of the number of population doublings was performed by using the following formula: ℎ 푑푎푦푠 표푓 푐푢푙푡푢푟푒 [푑] × 24 [ ] 푛푢푚푏푒푟 표푓 푝표푝푢푙푎푡𝑖표푛 푑표푢푏푙𝑖푛푔푠 = 푑 푑표푢푏푙𝑖푛푔 푡𝑖푚푒 [ℎ]

Colony forming units For determination of CFU of unstained and PKH26-stained hMSC 8 cells per cm² or 16 cells per cm² were seeded in duplicate in 6 well cell culture multidish. For determination of CFU of unstained and PKH26-stained rMSC 16 cells per cm² were seeded in duplicate in 6 well cell culture multidish. Medium was changed after 3 or 4 d. After 9 to 11 d cells were washed two times with DPBS and fixed with methanol for 5 min. After an air-dry step, cells were stained with Giemsa´s

Material and methods solution for 5 min. Cells were washed two times with distilled water and air-dried. Counting of CFU was blinded and colonies had to include at least 50 cells.

2.7.2. Surface marker expression of PKH26-stained mesenchymal stromal cells

Surface marker expression of MSC used in rat experiment was analysed using the following staining protocols (table 5 and 6). Human MSC at passage one or two and rMSC at passage three or four were harvested, washed with DPBS and 0.5 x 106 cells per approach were stained for 20 min. After that, cells were washed once with DPBS and suspended DPBS for flow cytometry analysis. In addition, Annexin V and SYTOX staining were performed according manufacturer´s instructions. Analysis was performed with FACSAria and FACSDiva Software.

Table 5 Flow cytometry staining panels for characterization of rMSC. Staining approaches included unstained control (1), isotype control (2) and CD markers of rat cells (3) with different fluorochromes (FITC: fluorescein isothiocyanate; PerCP: peridinin-chlorophyll-protein; APC: allophycocyanin; APC-Cy7: allophycocyanin coupled with the cyanine dye Cy7). All of them were mouse anti-rat antibodies, except for §-labelled, which were hamster anti-rat antibodies. Ig: immunoglobulin; CD: cluster of differentiation.

Fluorochrome/ FITC PerCP APC APC-Cy7 Approaches 1

IgG IgG IgG IgG § 2 5 µl 5 µl 5 µl 5 µl CD3 2 µl CD54 CD90 CD11b/c CD29 § 3 1.5 µl 2 µl 5 µl 5 µl CD45 1 µl

Material and methods

Table 6 Flow cytometry staining panels for characterization of hMSC. Staining approaches included unstained control (1), isotype control (2) and CD markers of human cells (3,4) with different fluorochromes (FITC: fluorescein isothiocyanate; PerCP: peridinin-chlorophyll-protein; APC: allophycocyanin). All of them were mouse anti-human antibodies. Ig: immunoglobulin; CD: cluster of differentiation.

Fluorochrome/ FITC PerCP APC Approaches 1

IgG IgG IgG 2 20 µl 20 µl 5 µl

CD90 CD45 CD73 3 1 µl 20 µl 5 µl

CD90 CD3 CD105 4 1 µl 20 µl 5 µl

2.7.3. Effect of interleukin 1 beta stimulation on secretion profile of human mesenchymal stromal cells

Cell culture experiment was performed to figure out the secretion profile over time of IL1B-stimulated hMSC. Additionally, secretion profiles of IL1B pre-stimulated MSC and PKH26-stained MSC were analysed. Human MSC at passage three or four were used in these experiments. Human MSC were harvested and one part was stained with PKH26 (see section 2.7.1., protocol used in rat experiment). Cells were seeded with a seeding density of 2.5 x 104 cells per cm² and 263 µl cell culture medium per cm². Two wells of a 12 well plate were used for stimulation of hMSC with 10 ng/ml IL1B for 24 h. After 24 h, PKH26-stained cells as well as IL1B-primed cells were stimulated with IL1B (10 ng/ml) for 24 h. To analyse time course of IL1B stimulation of hMSC four different concentrations of IL1B were used: 0.1 ng/ml, 1 ng/ml, 10 ng/ml and 100 ng/ml. IL1B stimulation was performed after overnight incubation and after adherence of cells. Therefore, cell culture medium (Alpha MEM supplemented with 8 % PL and 1 IU/ml heparin) with or without IL1B was changed completely. Unstimulated hMSC were used as negative control. After different points in time (4 h, 8 h, 12 h, 24 h, 48 h) supernatants were collected and stored at -70 °C until cytokine analysis.

Material and methods

Human cytokines and chemokines (assay sensitivity see section 2.6.4., CCL4 level was not determined) were analysed using a Merck Millipore Multiplex assay platform. Assay procedure was performed according to manufacturer´s instructions. Controls were included. Analysis was performed with the Liqui Chip device and Bio-Plex Manager Software. To analyse the secretion profile of MSC, cytokine levels of cell culture medium without MSC were subtracted from cytokine levels of supernatant from cell culture. TNFAIP6 ELISA was performed as previously described [99] (see section 2.6.4.). Samples were diluted 1:20 with blocking buffer. To analyse the secretion profile of MSC, cytokine level of cell culture medium without MSC was subtracted from cytokine level of supernatant from cell culture.

2.7.4. Co-culture of human mesenchymal stromal cells and the rat cell line NR8383

Human MSC were used at passage three or four. Human MSC were harvested and seeded with a seeding density of 2.5 x 104 cells per cm². After overnight incubation and after adherence of cells, culture medium was discarded and cells were washed once with DPBS. NR8383 were harvested and were seeded with a 5.0-fold higher seeding density in direct co-culture or co-culture with a transwell insert in 6 well cell culture plate. Controls included cell culture of hMSC or NR8383 alone. Cell culture medium was an equal mixture of hMSC minimum cell culture medium (Alpha MEM supplemented with 0.6 % PL and 1 IU/ml heparin) and NR8383 cell culture medium (Ham’s F12K supplemented with 15 % FBS, 100 U/ml Penicillin and 100 µg/ml Streptomycin). Heparin, Penicillin and Streptomycin were added according to the final volume of co-culture medium. Co- culture included unstimulated cells, IL1B-stimulated cells (10 ng/ml human IL1B) and LPS-stimulated cells (100 ng/ml). After 24 h supernatants were collected and stored at -70 °C until cytokine analysis. Human cytokines and chemokines were analysed using a Merck Millipore Multiplex assay platform (as described in section 2.6.4). Rat cytokines and chemokines (assay sensitivity for TNF: 1.9 pg/ml; IL10: 2.7 pg/ml; IL1B: 2.8 pg/ml; IL6: 30.7 pg/ml; CXCL1: 19.7 pg/ml; CCL2: 9.0 pg/ml; VEGFA: 2.6 pg/ml; CXCL2: 11.3 pg/ml) were analysed using a Merck Millipore Multiplex assay platform. Assay

Material and methods procedure was performed according to manufacturer´s instructions. Controls were included. Analysis was performed with the Liqui Chip device and Bio-Plex Manager Software. Since the manufacturer did not provide any information about cross reactivity, it was tested by using human cytokine/chemokine standard in rat cytokine/chemokine assay and vice versa. The manufacturer only told us that the rat-specific CCL2 capture antibody is 100 % cross reactive to human but there is no human-cross-reactivity-information listed for the detection antibody. The rat- specific CXCL1 capture antibody is 100 % cross reactive to human but the detection antibody is less than 10 % cross reactive. The rat-specific VEGFA capture antibody is 20 % cross reactive to human but there is no information available for the detection antibody. The rat-specific IL6 and rat-specific IL10 detection antibodies are not cross reactive to human but there is no information on the capture. In a single experiment no cross reactivity of rat-specific antibodies against human cytokine IL6 and TNF could be observed. A minor cross reactivity (0.1 % - 23.0 %) of rat-specific antibodies against human cytokines could only be detected for CXCL1, VEGFA, IL6, IL1B, CCL2 and IL10. In contrast to that, cross reactivity of human-specific antibodies to rat cytokines could not be detected. This is in line with four test measurements of the manufacturer. TNFAIP6 ELISA was performed as previously described [99] (see section 2.6.4.). Samples were diluted 1:2 with blocking buffer. Cell culture medium was measured without dilution. To analyse the secretion profile of MSC, cytokine levels of cell culture medium without MSC were subtracted from cytokine levels of supernatant from cell culture.

2.7.5. Co-culture of rat mesenchymal stromal cells and the rat cell line NR8383

Rat MSC were used at passage four. Experimental procedure and experimental approaches are depicted in figure 2. One day before starting the co-culture experiment, rMSC were harvested and a part of them were stained with PKH26 (see section 2.7.1, protocol used in rat experiment). 2.5 x 104 rMSC per cm2 were seeded in 6 well culture plate in rat cell culture medium (Alpha MEM with 15 % FBS, 2 mM GlutaMAX, 100 U/ml Penicillin and 100 µg/ml Streptomycin). Some of the unstained rMSC were stimulated with 10 ng/ml rat IL1B for 24 h. One day after

Material and methods seeding of rMSC, cell culture medium was discarded and cells were washed once with DPBS. Co-culture experiment with a cell ratio of 1:5 was started by adding 12.5 x 104 NR8383 cells per cm2 direct to the well or in the insert of the well. Cell culture medium in co-culture experiment was a 1:1 mixture of rat cell culture medium (Alpha MEM with 15 % FBS, 2 mM GlutaMAX, 100 U/ml Penicillin and 100 µg/ml Streptomycin) and NR8383 cell culture medium (Ham’s F12K with 15 % FBS, 100 U/ml Penicillin and 100 µg/ml Streptomycin). Supernatant was collected 24 h after stimulation and stored at -70 °C until cytokine analysis.

-1 d: seeding/PKH26 staining/priming of rMSC

0 d: start of co-culture Seeding of B, C, D) rMSC Experimental approaches: +1 d: collecting of supernatant E) PKH26-stained rMSC A) NR8383 B) NR8383, rMSC F) rMSC, primed with stored at -70 °C IL1B for 24 h C) NR8383, rMSC (transwell) D) rMSC E) rMSC (PKH26-stained) F) rMSC (IL1B-primed) Cell culture medium additive - nothing or - 10 ng/ml IL1B or - 100 ng/ml LPS

Figure 2 Experimental procedure and experimental approaches (A-F) consisted of unstimulated cells and stimulated cells (10 ng/ml IL1B or 100 ng/ml LPS). rMSC: rat mesenchymal stromal cells; IL1B: interleukin 1 beta; LPS: lipopolysaccharide.

Rat cytokines and chemokines were analysed using a Merck Millipore Multiplex assay platform (as described in section 2.7.4). TNFAIP6 ELISA was performed as previously described [99] (see section 2.6.4.). Samples were diluted 1:2 with blocking buffer.

2.7.6. Unspecific binding of human mesenchymal stromal cells-specific antibodies to rat mesenchymal stromal cells and vice versa

Human MSC at passage three and rat MSC at passage four were harvested, washed with DPBS and 0.5 x 106 cells per approach were stained for 20 min by the use of the following schemes (table 7 and 8). Afterwards, cells were washed once with DPBS and analysed by flow cytometry.

Material and methods

Table 7 Flow cytometry staining panels to exclude cross-reactivity of hMSC. Staining approaches included unstained control (1), isotype control (2) and CD markers of rat cells (3) with different fluorochromes (FITC: fluorescein isothiocyanate; PerCP: peridinin-chlorophyll-protein; APC: allophycocyanin; APC-Cy7: allophycocyanin coupled with the cyanine dye Cy7). All of them were mouse anti-rat antibodies, except for §-labelled, which were hamster anti-rat antibodies. Ig: immunoglobulin; CD: cluster of differentiation.

Fluorochrome/ FITC PerCP APC APC-Cy7 Approaches IgG IgG IgG IgG $ 1 5 µl 5 µl 5 µl 5 µl CD3 2 µl CD54 CD90 CD11b/c CD29$ 2 1.5 µl 2 µl 5 µl 5 µl CD45 1 µl IgG1 3 20µl

CD49d CD90 CD11b/c CD45 4 5 µl 2 µl 5 µl 20 µl

Table 8 Flow cytometry staining panels to exclude cross-reactivity of rMSC. Staining approaches included isotype control (1) and CD markers of human cells (2,3) with different fluorochromes (FITC: fluorescein isothiocyanate; PerCP: peridinin-chlorophyll-protein; APC: allophycocyanin). All of them were mouse anti-human antibodies. Ig: immunoglobulin; CD: cluster of differentiation.

Fluorochrome/ FITC PerCP APC Approaches IgG IgG IgG 1 20 µl 20 µl 5 µl

CD90 CD45 CD73 2 1 µl 20 µl 5 µl

CD90 CD3 CD105 3 1 µl 20 µl 5 µl

Unspecific binding was calculated by subtraction of isotype control from specific staining.

Material and methods

2.8. Animal experiments

2.8.1. Study protocol

To determine if administration of xenogeneic or allogeneic MSC have an effect on the inflammatory and regenerative processes, there were six experimental groups of animals: (i) sham, (ii) traumatized (TxT) rats and rats receiving (iii) hMSC or (iv) rMSC after TxT. Further groups included IL1B-primed (v) hMSC or (vi) rMSC. To recover the injected MSC and to distinguish between endogenous and the administered MSC, PKH26 staining was performed. 24 h or 5 d after TxT, blood, BAL, lung, liver, spleen, kidney, brain and gonads were collected. Flow cytometry analysis and fluorescence microscopy were performed to detect and characterize the mobilized MSC and the administered MSC in blood and BAL. Blood and BAL cells were partially cultured to verify the criteria of MSC like plastic adherence, CFU and expression of specific cell surface proteins. Plasma and BAL were analysed by Multiplex assay and ELISAs. Paraffin sections of the lungs were scored for the histological pulmonary damage. Additionally, distribution of hMSC could be also detected by Alu sequence analysis of collected rat tissue.

2.8.2. Animals

The following section was already partly published in Amann et al. on page 220 [5]. The animal experiment was reviewed and approved by the regional authority (Regierungspräsidium Tübingen, Germany; registration number 1182). The male Wistar rats were 7 to 8 weeks old and ranged in weight from 248 gram (g) to 363 g (Charles River, Sulzfeld, Germany). The cages were equipped with wood pulp and softwood granulates. All animals had free access to water and food (V1124-30; ssniff Spezialdiäten GmbH, Soest, Germany). The husbandry conditions were two rats per cage at maximum, a day-night cycle of 14/10 h, mean room temperature of 22.2 °C and 50 % to 60 % humidity.

2.8.3. Priming and PKH26 staining of mesenchymal stromal cells

IL1B priming was performed by stimulating the MSC with human or rat IL1B (10 ng/ml) for 24 h before harvesting.

Material and methods

The cells were harvested and stained with the PKH26 red fluorescent cell linker kit. Staining was performed according to the manufacturer instruction with a dye concentration of 1.6 x 10-5 M per 1 x 106 cells. After three washing steps, the cells were passed through a sterile nylon cell strainer to separate cell aggregates and counted in Neubauer counting chamber with trypan blue staining. Cells were suspended in 0.1 % bovine serum albumin (BSA) in DPBS with a cell concentration of 5 x 106 cells per ml. The cell suspension was split in 1 ml per 3 ml syringe, closed with Combi Stopper and stored at room temperature up to 3 h, protected from light. Rat MSC of two donors were pooled at the end of PKH26 staining. Afterward fluorescence intensity was measured by flow cytometry. Unstained cells served as a control. Delta fluorescence intensity was calculated with the following formula:

훥푓푙푢표푟푒푠푐푒푛푐푒 𝑖푛푡푒푛푠𝑖푡푦 = 푚푒푑𝑖푎푛 푓푙푢표푟푒푠푐푒푛푐푒 𝑖푛푡푒푛푠𝑖푡푦 표푓 푃퐾퐻26 푠푡푎𝑖푛푒푑 푐푒푙푙푠 − 푚푒푑𝑖푎푛 푓푙푢표푟푒푠푐푒푛푐푒 𝑖푛푡푒푛푠𝑖푡푦 표푓 푢푛푠푡푎𝑖푛푒푑 푐푒푙푙푠

2.8.4. Blunt chest trauma and administration of mesenchymal stromal cells

Blunt chest trauma model was used to investigate local and systemic immunological changes after lung contusion. Isolated lung damage without collateral damage (for example rib fractures, liver ruptures, intestinal wall necrosis) should be achieved. First of all, the rats were anesthetized with rinsing mask anaesthesia and a mixture of 4 % sevoflurane and 96 % oxygen (2 l/min). Animals were fixed in a supine position and extremities were splayed out. To distinguish between the rats, the tail of the rat was marked, chest and abdominal region were shaved and the Processus xiphoideus, as well as costal arches and the centre line of the sternum were marked. The rats were aligned under the blast wave generator with the junction between the Processus xiphoideus and the body of the sternum. The distance between cylinder and sternum was 2 cm. Anaesthesia was still achieved by a mixture of 2.5 % sevoflurane and 97.5 % oxygen. The TxT was induced by a single blast wave centred on the thorax as described previously [108,111]. Knöferl et al. and Seitz et al. have established this model in the Ulm University [122,225]. The reproducibility is based on Mylar® polyester membrane clamped in the 41

Material and methods cylinder and ruptures at a constant pressure. Following, apnoea phase was recorded. Buprenorphin 0.03 mg per kilogram (kg) body weight was administered as subcutaneous depot after spontaneous breathing had been reinstated. 5 x 106 cells or 1 ml 0.1 % BSA/DPBS solution alone (sham and TxT animals) was intravenously administered (Vena dorsalis penis). Buprenorphin administration was repeated within 8 h. Sham rats got the same treatment without blast wave application. Figure 3 depicts a representative thoracic region from sham and TxT rat.

Figure 3 Thoracic region of sham and blunt chest trauma (TxT) rats. Blunt chest trauma led to irregular hemorrhage in surrounding tissue, especially the lung.

2.8.5. Blood, tissue and broncho-alveolar lavage fluid cell harvesting procedure

The rats were humanely killed in deep anaesthesia either 24 h or 5 d after TxT or sham procedure by exsanguination (Vena cava) plus setting a bilateral pneumothorax. Blood was collected in EDTA-Monovette®. After performing a manual blood smear on microscope slide, blood samples were centrifuged at 340 g for 10 min and plasma was stored at -70 °C. The following section was already published in Amann et al. on page 220 [5]. The pulmonary circulation was flushed with 10 ml DPBS. The lower right lung lobe and the left lung were clamped while the rest of the lung was flushed with 2.5 ml cold DPBS by cannulated trachea. It was mixed with 50 µl protease inhibitor cocktail and was used for cell counting, cytospin and cytokine measurements. After that, the lung was flushed two times with 5 ml DPBS for flow cytometry analysis. The 42

Material and methods left lung was flushed with 4 % formaldehyde solution (Lillie) and stored. The lower lung lobe, liver, spleen, kidney, brain and gonads were frozen in liquid nitrogen for Alu polymerase chain reaction (PCR) analysis.

Neutral buffered formalin (Lillie)

4 g Natriumdihydrogenphosphat (NaH2PO4 H2O)

6.5 g Dinatriumhydrogenphosphat (Na2HPO4) The components were solved in 900 ml distilled water. 100 ml Formaldehyde, 37 %

2.8.6. Flow cytometry and cell sorting

The following section was already partly published in Amann et al. on page 221 [5]. Flow cytometry analysis was done using a FACSAria with FACSDiva V6.1.3 Software. BAL and DPBS-diluted blood was used to detect events in flow cytometry analysis being in PKH26 gate (PKH+) or not (PKH-). About 170000 events of peripheral rat blood were measured with a range of 25071 to 501914 events. About 340000 events of BAL were measured with a range of 12300 to 1000000 events. Furthermore, DPBS-diluted blood was exposed to a Ficoll gradient centrifugation at 1500 g for 20 min and isolation of PKH+ events was done by fluorescence activated cell sorting (FACS). Cytospin was performed as previously described. Table 9 shows the staining panel for the characterization of hMSC in rat blood. Table 10 shows the staining panel for the detection of administered or endogenous rMSC in rat blood. Blood was diluted 1:1 with DPBS, 50-100 µl were stained for 15-20 min and erythrocytes were lysed by addition of 900 µl BD FACS™ lysing solution for 15-30 min according to manufacturer´s instructions and washed twice with DPBS. Life gate was defined in forward-sideward scatter diagram to exclude debris and dead cells.

Material and methods

Table 9 Flow cytometry staining panels for characterization of hMSC in rat blood. Staining approaches included unstained control (1), isotype control (2) and CD markers of human (CD90, CD105 were mouse anti-human antibodies) and rat (CD45 was mouse anti-rat antibodies) cells (3,4) with different fluorochromes (FITC: fluorescein isothiocyanate; APC: allophycocyanin; APC- Cy7: allophycocyanin coupled with the cyanine dye Cy7). Ig: immunoglobulin; CD: cluster of differentiation.

Fluorochrome/ FITC APC APC-Cy7 Approaches 1

IgG IgG IgG 2 20 µl 5 µl 20 µl

CD90 CD105 CD45 4 1 µl 5 µl 20 µl

Table 10 Flow cytometry staining panels for characterization of rMSC in rat blood. Staining approaches included unstained control (1), isotype control (2) and CD markers of rat cells (3) with different fluorochromes (FITC: fluorescein isothiocyanate; PerCP: peridinin-chlorophyll-protein; APC: allophycocyanin; APC-Cy7: allophycocyanin coupled with the cyanine dye Cy7). All of them were mouse anti-rat antibodies, except for §-labelled, which were hamster anti-rat antibodies. Ig: immunoglobulin; CD: cluster of differentiation.

Fluorochromes/ FITC PerCP APC APC-Cy7 Approaches 1 2 IgG IgG IgG IgG $ 5 µl 5 µl 5 µl 5 µl

3 CD54 CD90 CD3 CD29$ 1.5 µl 2 µl 2 µl 5 µl CD11b/c 5 µl CD45 1 µl

2.8.7. Incubation of mesenchymal stromal cells with rat blood for 24 hours

The following section was already partly published in Amann et al. on page [5]. Residual blood from sham animals was pooled from two to four animals and used for in vitro experiments. In this study, 270270 PKH26-stained cells and 54 µl 0.1 % BSA/DPBS were added per 1 ml blood. After 24 h incubation, flow cytometry analysis was performed. PKH26-stained cells of the same batch, incubated at 4 °C for 24 h, served as positive control. Storage condition for hMSC was established 44

Material and methods and described previously [245]. Subsequently, blood was exposed to a Ficoll gradient centrifugation and PKH+ or PKH- cells were detected and isolated by FACS (see section 2.8.6.). Cytospin was performed as described in section 2.8.10..

2.8.8. Cell culture of sorted cells

PKH+ or PKH- events of blood were sorted by the use of FACSAria. Sorted and unsorted peripheral blood and BAL events were cultured with or without feeder cells C2C12 (see section 3.4.8). Blood and BAL events were cultured in rMSC culture medium as described in section 2.5.. Further 1 IU/ml heparin was added. PKH+ or PKH- events of blood and BAL from hMSC-treated animals were cultured in hMSC culture medium (see section 2.5.). Events from animals that were not treated with hMSC were cultured in rMSC culture medium. After one month cells were harvested and characterized by flow cytometry analysis as described in section 2.8.6. The MSC surface marker panel was modified and is depicted in table 11.

Table 11 Flow cytometry staining panels for characterization of hMSC. Staining approaches included isotype control (1) and CD markers of human cells (2,3) with different fluorochromes (FITC: fluorescein isothiocyanate; PerCP: peridinin-chlorophyll-protein; APC: allophycocyanin). All of them were mouse anti-human antibodies.

Fluorochrome/ FITC PerCP APC Approaches IgG IgG IgG 1 20 µl 20 µl 5 µl

CD105 CD45 CD73 2 10 µl 20 µl 5 µl

CD90 CD3 CD105 3 1 µl 20 µl 5 µl

2.8.9. Alu polymerase chain reaction analysis

The Alu analysis was performed in the Division for Biochemistry of Joint and Connective Tissue Diseases of Ulm University. The following section was already published in Amann et al. on page 221 [5]. The presence of human cells was

Material and methods tested in lung, liver, kidney, brain, gonad, and spleen tissue, which were collected after euthanasia of the rats and subsequently stored in liquid nitrogen. The presence of human cells was proved by the detection of Alu sequences using real- time PCR method according to Nicklas and Buel [182]. About 100 mg of each cryopreserved tissue was digested and DNA isolated using QIAamp DNA mini kit following manufactures instructions. Blood smears were air-dried and DNA was isolated. Per real-time PCR probe, 10 ng of purified DNA was used for a 20 µl reaction mix, consisting of 10 µl Power SYBR™ Green Master Mix, and 0.5 µl per primer (forward; 5’-GTCAGGAGATCGAGACCATCCC-3’; reverse; 5’- TCCTGCCTCAGCCTCCCAAG-3’; biomers.net GmbH, Ulm). Finally, water was added to the end volume. The PCR was done using a StepOne Plus with the following settings: 95 °C for 2 min, 35 cycles 95 °C for 15 seconds (s), 68 °C for 30 s, 72 °C for 30 s. PCR products were proved by a melt curve from 50 °C up to 90 °C. Expression levels were analysed by using the StepOne Software V2.32. The detection limit by this method was comparable to at least the DNA of one human cell per sample.

2.8.10. Broncho-alveolar lavage fluid cell count, differential analysis and fluorescence microscopy

BAL was centrifuged at 450 g for 10 min (4 °C). Supernatant was stored at -70 °C. The cell pellet was suspended in 1 ml DPBS and cells were counted in a Neubauer counting chamber partly by using 0.01 % crystal violet. A cytospin was performed at 600 g for 3 min with 50-100 µl cell suspension. BAL cytospins and blood smears were analysed by fluorescence microscopy AF6000. Pictures were taken by the use of Leica Application Suite AF software. Phase contrast and TX2 ET filter cube were sequentially switched and uniform settings regarding to exposure time, gain and intensity of TX2 ET were used for all samples. Red fluorescent cells were manually counted and related to 300 total events. Additionally, after air-drying, BAL cytospins were stained by Pappenheim staining as described below.

Pappenheim buffer

4.08 g Potassium phosphate monobasic (KH2PO4) 12.5 ml 1 M Sodium hydroxide (NaOH) 46

Material and methods

The components were filled up to 5 litre (l) with aqua iniectabilia and a final pH of 6.8 was achieved.

Pappenheim staining protocol Incubation time Staining steps Air drying of samples 10 min Methanol 10 min 1:1 diluted May-Grünwald´s eosine methylene blue solution modified for microscopy with Pappenheim buffer 30-35 min 22.5 % Giemsa´s azur eosin methylene blue solution in Pappenheim buffer 5 min Washing with Pappenheim buffer Last step was repeated once.

300 nucleated cells were manually counted by means of Leucodiff 700 for differential analysis (lymphocytes, basophils, eosinophils, neutrophils, alveolar macrophages and monocytes).

2.8.11. Measurement of plasma and broncho-alveolar lavage fluid cytokines and protein

Multiplex assay was performed in cooperation with the Department of Transplantation Immunology of the Institute of Clinical Transfusion Medicine and Immunogenetics (IKT) Ulm of the German Red Cross Blood Donor Service Baden- Württemberg–Hessia. Rat cytokines and chemokines (assay sensitivity for TNF, IL10, IL1B, IL6, CXCL1, CCL2, VEGFA see section 2.7.4.) were analysed using a Merck Millipore Multiplex assay platform. Assay procedure was performed according to manufacturer´s instructions. Controls were included. Analysis was performed with the Liqui Chip device and Bio-Plex Manager Software. TNFAIP6 ELISA was performed as previously described [99]. Supernatant from LPS (20 ng/ml) stimulated rMSC in rMSC culture medium was used as controls. Linear regression was performed. PGE2 (assay sensitivity: 41.4 pg/ml), rat FGF2 (assay sensitivity: 3.68 pg/ml), mouse IL1RN (assay sensitivity: 8.5 pg/ml; cross- reactivity of rat IL1RN: 12 %) and rat IL17A (assay sensitivity: 2.0 pg/ml) were measured by ELISA according to manufacturer´s instructions and by use of Polarstar Omega Reader and MARS Omega Data Analysis software. Rat C5a ELISA Kit (assay sensitivity: 37.5 pg/ml) was measured in cooperation with the Department of Orthopedic Trauma, Hand, Plastic and Reconstructive Surgery, 47

Material and methods

Ulm University Medical Center, Ulm. Assay was measured according to manufacturer´s instructions with a plasma dilution of 1:10 or 1:15. Total amount of protein in BAL was measured by NanoDrop1000.

2.8.12. Measurement of myeloperoxidase activity

MPO is an enzyme, localized in the azurophil granules, (phago-) lysosomes and secretory granules of PML. It catalyses the oxidation of halide ions and thiocyanate via H2O2; thereby it has antimicrobial activity but also contributes to inflammatory tissue injury [80]. PML contain MPO, which is involved in killing bacteria. It is an essential enzyme for normal PML function. Measurement of MPO activity was performed in cooperation with the Institute for Clinical and Experimental Trauma-Immunology of Ulm University. The assay procedure was performed according to a modified method protocol from Suzuki et al. [239] and normalised to albumin content of lung tissue. MPO activity was measured by the H2O2-dependent oxidation of TMB, the enzyme substrate of MPO.

Buffer 1: 0.08 M potassium phosphate buffer, pH 5.4

10.35 g Potassium dihydrogen phosphate (KH2PO4) 950 ml Distilled water

0.91 g Dipotassium hydrogen phosphate (K2HPO4) 50 ml Distilled water

Buffer 2: 0.05 M potassium phosphate buffer, pH 6.0

6.12 g Potassium dihydrogen phosphate (KH2PO4) 900 ml Distilled water

1.14 g Dipotassium hydrogen phosphate (K2HPO4) 100 ml Distilled water

Buffer 3: 0.50 g Hexadecyltrimethylammonium bromide 100 ml Buffer 2

TMB-DMSO-solution 10 mg 3,3',5,5'-Tetramethylbenzidine (TMB) 10 ml Dimethylsulfoxide (DMSO)

Material and methods

H2O2-solution 1:

100 µl Hydrogen peroxide (H2O2), 30 % 30 ml Distilled water

H2O2-solution 2:

670 µl H2O2-solution 1 40 ml Buffer 1

Reagent C (preparation occurred immediately before use): 1.80 ml Copper(II) sulfate anhydrous (CuSO₄), 4 % 90 ml Bicinchoninic acid solution

On the first day, cryopreserved right lung lobe was mechanically crushed and 20 mg of lung tissue was weighed. It was transferred into a reaction tube and 1.5 ml of buffer 3 was added to dissolve cell membrane. Lung tissue was homogenized by the lab homogenizer ULTRA TURRAX T25 (level 4). Afterwards, samples were incubated at 60 °C in a water bath for 2 h to inactivate catalase. Samples were stored at 4 °C overnight. On the second day, samples were mixed thoroughly and centrifuged at 3950 g for 15 min. Samples were diluted 1:10 in buffer 3. MPO standard was prepared by serial dilution of MPO in buffer 3 (1000 to 15.6 U/l). 25 µl of standard or sample dilution was pipetted per well, in duplicate, into a 96 well microtiter plate. Blank wells contain buffer 3. 25 µl of TMB-DMSO- solution and 200 µl of H2O2-solution 2 were added per well. After an incubation period at 37 °C for 5 min, enzyme reaction was stopped by adding 50 µl of 2 M sulfuric acid (H2SO4) to each well. Absorbance was measured using the sunrise spectrophotometer and a wave length of 450 nm. After the measurement of MPO activity, the protein content of the samples was determined. Protein precipitation was reached by adding 100 µl of perchloric acid

(HClO4) to the remaining homogenisate and mixing thoroughly. Samples were centrifuged at 3950 g for 10 min, supernatant was discarded, 3 ml of distilled water was added and mixed thoroughly. This washing step was repeated two times. Afterwards, 2 ml of 0.5 M sodium hydroxide (NaOH) was added to the pellet. Samples were mixed thoroughly and incubated at 56 °C in water bath for 30 min. Samples were diluted 1:5 in distilled water. BSA standard was prepared by serial dilution in distilled water (Albumin Standard Ampules, 2000 to 31.25 µg/ml). 15 µl 49

Material and methods of standard or diluted sample was pipetted per well, in duplicate, into a 96 well microtiter plate. 300 µl of reagent C was added to all wells. After incubation period at 37 °C for 30 min absorbance was measured at 562 nm.

2.8.13. Histological score of the lung

The histological score of the lung was performed in the Division for Biochemistry of Joint and Connective Tissue Diseases of Ulm University according to Matute-Bello [164]. The following section was already published in Amann et al. on page 221 [5]. Collected lung tissue was cut into five samples (figure 4). Paraffin embedded samples were deparaffined by standard methods. Afterwards, Prussian blue staining of lung tissue was performed (20 min in 2 % potassium hexacyanoferrate trihydrate and 1 % hydrochloric acid). After washing with distilled water a nuclear fast red stain was performed. Thereafter a second washing was done and the samples were embedded in EUKITT®. Dissection was done using a Zeiss AxioMot digital system with Zeiss AxioVision V4.82 software. The histological lung injury score was assessed according to Matute-Bello et al. [164] by hemorrhage, granulocytes, macrophages and oedema, with a range of zero to three. Septum thickening was valued with zero or one. Slides were evaluated by an observer who was blinded to the treatment group.

Figure 4 Paraffin sections (1-5) of left lung of rat. Courtesy of Division for Biochemistry of Joint and Connective Tissue Diseases of Ulm University.

Material and methods

2.9. Statistics

Statistical analysis was performed with GraphPad Prism 7. Results are shown as mean ± standard deviation (SD). Normal distribution was verified by D´Agostino and Pearson normality test (in vivo experiments) or Shapiro-Wilk test (in vitro experiments). If the measured values were normal distributed statistical analysis was performed using Welch’s t-test, one or two-way analysis of variance (ANOVA) followed by the Bonferroni test as a post hoc test for multiple comparisons to determine significant differences between experimental means. If normal distribution was not valid in all comparing groups Wilcoxon test, Mann-Whitney test or Kruskal-Wallis test with Dunn’s multiple comparisons test was used. Threshold for significant difference was probability (P) < 0.05, P < 0.01, P < 0.001, P < 0.0001. General remarks: Non-significant differences/trends are described in the result part without P-values. Differences more than 2.0-fold or less than 0.5-fold are described.

Results

3. Results

3.1. Mesenchymal stromal cells and polytrauma

To investigate the role of MSC in trauma, MSC were stimulated with different stimuli. Proliferation, migration, gene expression and the cytokine profile were analysed.

3.1.1. Influence of trauma factors on proliferation of mesenchymal stromal cells

As already described in previous publications of the Institute of Transfusion Medicine [72], lower concentrations of PL than those used for large-scale MSC expansion were sufficient to stimulate MSC growth. To investigate the effect of trauma factors on proliferation of MSC these low concentrations were used. Figure 5 depicts results of CyQuant Cell Proliferation Assay.

Figure 5 Influence of trauma factors on proliferation of human MSC. CyQuant Cell Proliferation Assay was performed 6 to 7 d after stimulation. Alpha MEM was supplemented with 0.6 % or 1.2 % PL. Alpha MEM supplemented with 8 %PL was used as a positive control. MSC were either unstimulated (-), stimulated with polytrauma cocktail high (PTCH), stimulated with polytrauma cocktail high without thrombomodulin (PTCH-THBD), stimulated with polytrauma cocktail low (PTCL) or stimulated with single trauma factors (IL1B, HMGB1, THBD, THPO). Stimulation of MSC with growth factors (activin A (5ng/ml), fibroblast growth factor 2 (20 ng/ml), platelet-derived growth factor BB (20 ng/ml), insulin-like growth factor 1 (400ng/ml) and dexamethasone (1µmol/l)) was used as a positive control as well. The cell count in the beginning of the experiment was 200 cells (dotted line). Mean and SD of n=1-5 are depicted. IL1B: interleukin 1 beta; HMGB1: high mobility group box 1; THBD: thrombomodulin; THPO: thrombopoietin.

Results

Using one-way ANOVA (stimulated cells versus unstimulated cells) no significant effect of trauma factors on cell growth could be detected. However, stimulation of MSC with PTCH increased proliferation 2.5-fold compared to unstimulated MSC in 1.2 % PL-supplemented medium. Stimulation of MSC with PTCH-THBD increased proliferation 2.3-fold compared to unstimulated MSC in 1.2 % PL-supplemented medium. It seemed that PTCH led to a higher increase in proliferation than PTCL (0.6 % PL PTCH versus PTCL: 1.5-fold; 1.2 % PL PTCH versus PTCL: 2.0-fold). Stimulation with single factors like IL1B, HMGB1, THBD and THPO in high concentrations (same as in PTCH) did not seem to have the same effect on proliferation than PTCH. THPO had the least effect on proliferation of all measured individual factors.

3.1.2. Chemoattractive potential of trauma factors for mesenchymal stromal cells

To analyse the chemoattractive potential of selected trauma factors, transwell migration assays were performed. Results are depicted in figure 6 and 7.

Figure 6 Migration of human MSC toward polytrauma cocktail high (PTCH), polytrauma cocktail high without thrombomodulin (PTCH-THBD), polytrauma cocktail low (PTCL) and single trauma factors (IL1B, C3a, HMGB1, THBD, THPO). Migration of carboxyfluorescein diacetate succinimidyl ester (CFSE)-labelled MSC (RFU: relative fluorescence units) over time was measured in transwell assay with Polarstar Omega Reader. Values obtained from measurement at time 0 min were subtracted from values obtained from other points in time. Negative control was 0.1 % HSA/DPBS without factors. Positive controls were 10 % FBS and 8 % PL in HSA/DPBS solution (controls are not depicted). Mean and SD of n=3-4 are depicted. IL1B: interleukin 1 beta; IL6: interleukin 6; C3a: complement component 3a; HMGB1: high mobility group box 1; THBD: thrombomodulin; THPO: thrombopoietin. 53

Results

Relative fluorescence units (RFU) of negative control (0.1 % HSA/DPBS without factors) were 5191 ± 3980 after a migration time of 45 min. RFU of positive controls (10 % FBS or 8 % PL) were 7728 ± 1909 or 6056 ± 1235 after a migration time of 45 min. The PTCH had a 2.5-fold to 3.5-fold stronger effect on migration of MSC than the PTCL at any time (except for the measurement 15 min after starting experiment). No difference in stimulation of chemotaxis of MSC could be observed between IL1B and C3a at any time (ratio: 0.8 - 0.9). THPO seemed to be a stronger chemotactic stimulus than HMGB1 and THBD. The chemotactic activity of 1 µg/ml HMGB1 was already described previously by the Institute of Transfusion Medicine [153].

Figure 7 Concentration-dependent chemoattractive potential of thrombopoietin (THPO). (A) Migration of carboxyfluorescein diacetate succinimidyl ester (CFSE)-labelled human MSC after 30 min (RFU: relative fluorescence units) was measured in transwell assay with Polarstar Omega Reader. Values obtained from measurement at time 0 min were subtracted from values obtained from the 30 min measurement (n=3). Significance was calculated with one-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05. (B) Migration of MSC relative to negative control (Alpha MEM without THPO; n=2) was analysed in modified Boyden chamber 4 h after addition of THPO. Mean and SD are depicted.

Transwell migration assays revealed a concentration-dependent chemoattractive potential of THPO. Starting at 150 pg/ml of THPO, the chemoattractive potential of THPO decreased with increasing THPO concentration. 150 pg/ml of THPO led to a significantly higher (3.2-fold, P=0.0353) migratory activity of MSC than 4500 pg/ml of THPO. Modified Boyden chamber experiments showed a comparable trend. 150 pg/ml of THPO led to a 3.9-fold higher migratory activity of MSC than 4500 pg/ml of THPO.

Results

Furthermore, 89.37 % of MSC (n=3) expressed the THPO receptor CD110 on their cell surface.

3.1.3. Gene expression analysis of untreated and stimulated mesenchymal stromal cells

To determine the effect of PTCH and PTCL on the gene expression of MSC, RNA sequencing was performed. In addition, the effect of IL1B stimulation (10 ng/ml IL1B) on MSC was analysed, since it has been shown that IL1B has the highest potential to mimic the effect of the polytrauma cocktail used by Hengartner and colleagues [99], consisting of IL1B, IL6, CXCL8, C3a and C5a. Since PL itself contains many growth factors [70], the MSC were cultured in AB- serum instead of PL to exclude this PL effect. As a base for this set of experiments, the difference between the two culture conditions (PL supplementation or AB-serum supplementation) of MSC was analysed. Afterwards, MSC treated with IL1B, PTCH or PTCL were compared to MSC in AB- serum alone. Differential gene expression of biological triplicates and two different points in time after stimulation (6 h, 24 h) were analysed. Interpretation of the results was performed by the Institute for Medical Systems Biology of Ulm University (see table 12 for scheme of 13 comparisons). Results are depicted in figure 8 to 10 and table 13.

Results

Table 12 Matrix of analysis of gene expression. Biological triplicates, different medium supplementation (platelet lysate (PL), AB-serum (AB) without and with cytokines (IL1B: interleukin 1 beta; PTCH: polytrauma cocktail high; PTCL: polytrauma cocktail low) were included in the 36 samples. 13 comparisons were performed (I-XIII) taking into account differences due to cell preparation (donor), incubation time (treated for) and medium. Courtesy of the Institute for Medical Systems Biology of Ulm University.

Medium Donor TreatedFor I II III IV V VI VII VIII IX X XI XII XIII Control1 Donor Donor Donor Donor Donor Donor Donor Donor Donor Donor Donor Donor Donor Control2 TreatedFor TreatedFor TreatedFor TreatedFor Medium Contrast Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium TreatedFor 1 PL 7537 6 h TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE 2 AB 7537 6 h TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE 3 AB 7537 6 h TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE 4 AB+PTCH 7537 6 h FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE 5 AB+PTCL 7537 6 h FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE 6 AB+IL1B 7537 6 h FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE 7 PL 7554 6 h TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE 8 AB 7554 6 h TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE 9 AB 7554 6 h TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE 10 AB+PTCH 7554 6 h FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE 11 AB+PTCL 7554 6 h FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE 12 AB+IL1B 7554 6 h FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE 13 PL 7559 6 h TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE 14 AB 7559 6 h TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE 15 AB 7559 6 h TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE 16 AB+PTCH 7559 6 h FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE 17 AB+PTCL 7559 6 h FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE 18 AB+IL1B 7559 6 h FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE 19 PL 7537 24 h FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE 20 AB 7537 24 h FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 21 AB 7537 24 h FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 22 AB+PTCH 7537 24 h FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE 23 AB+PTCL 7537 24 h FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE 24 AB+IL1B 7537 24 h FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE 25 PL 7554 24 h FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE 26 AB 7554 24 h FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 27 AB 7554 24 h FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 28 AB+PTCH 7554 24 h FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE 29 AB+PTCL 7554 24 h FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE 30 AB+IL1B 7554 24 h FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE 31 PL 7559 24 h FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE 32 AB 7559 24 h FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 33 AB 7559 24 h FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 34 AB+PTCH 7559 24 h FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE 35 AB+PTCL 7559 24 h FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE 36 AB+IL1B 7559 24 h FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE

The heat maps (figure 8) depict that the similarity of the gene expression of all samples was very high (> 99 %, see colour keys). Heat map I shows that the impact of cell preparation/donor was stronger than the impact of PL supplementation compared to AB-serum supplementation. However, heat map XII depicts that the impact of stimulation with IL1B or with PTCH was stronger than the impact of the cell preparation/donor.

Results

Figure 8 Heat map I depicts the clustering of MSC cultured in 8 % platelet lysate-supplemented medium (PLP, n=3) and of MSC cultured in 20 % AB-serum-supplemented medium (AB, three biological replicates and two technical replicates) for 6 h. Heat map XII depicts the clustering of MSC cultured in 20 % AB-serum-supplemented medium (AB, three biological replicates and two technical replicates) without and with 10 ng/ml interleukin 1 beta (AB+IL1B, n=3) for 6 h and 24 h. The title indicates the method of analysis (vst cor, clusters by ward: variance-stabilizing transformation correlation, hierarchical cluster analysis by Ward's method). This figure was already published in Amann et al. (figure 1 on page 6) [4], CC BY 4.0 https://creativecommons.org/licenses/by/4.0/.

As shown in figure 9, treatment of MSC with PL, IL1B, PTCH or PTCL had different effects on gene expression. There were only minimal changes in gene expression comparing MSC cultured in PL-supplemented cell culture medium and MSC cultured in AB-serum supplemented cell culture medium: at 6 h: 31 differentially expressed genes, at 24 h: 168 differentially expressed genes and 63 genes which were differentially expressed both after 6 h and 24 h. The stimulation of MSC with PTCH had the strongest effect on gene expression regardless of the incubation time (6 h: 470 differentially expressed genes compared to unstimulated cells; 24 h: 409 differentially expressed genes compared to unstimulated cells; both at 6 h and 24 h: 457 differentially expressed genes compared to unstimulated cells). IL1B stimulation had similar effects to PTCH (6 h: 469 differentially expressed genes compared to unstimulated cells; 24 h: 402 differentially expressed genes compared to unstimulated cells; both at 6 h and 24 h: 433 differentially expressed genes compared to unstimulated cells). Stimulation of

Results

MSC with PTCL changed gene expression in a more moderate way (6 h: 183 differentially expressed genes compared to unstimulated cells; 24 h: 174 differentially expressed genes compared to unstimulated cells; both at 6 h and 24 h: 211 differentially expressed genes compared to unstimulated cells).

6 h/24 h 20 191 24 h 28 146 PTCL 6 h 21 162 6 h/24 h 74 383 24 h 111 298

PTCH 6 h 104 366 6 h/24 h 75 358 - 24 h 110 292

IL1B + 6 h 103 366 6 h/24 h 34 29

24 h 76 92 PL 6 h 22 9 0 100 200 300 400 500 number of genes regulated differentially compared to AB-serum medium

Figure 9 Differentially expressed genes depending on the type of treatment. Unstimulated MSC cultured in AB-serum were compared to MSC cultured in AB-serum with polytrauma cocktail low (PTCL, n=3) or polytrauma cocktail high (PTCH, n=3) or 10 ng/ml interleukin 1 beta (IL1B, n=3). Additionally MSC cultured in AB-serum were compared to MSC cultured in platelet lysate (PL, n=3). Genes were either downregulated (-) or upregulated (+) compared to cell culture in AB- serum. Different times (6 h, 24 h) after starting cell culture experiment are depicted. Furthermore, numbers of differentially expressed genes after 6 h and 24 h of cell culture are depicted. This figure was already published in Amann et al. (figure 2 (A) on page 7) [4], CC BY 4.0 https://creativecommons.org/licenses/by/4.0/.

Intersections of differentially expressed genes in the PTCH (II), PTCL (III) and IL1B (IV) group are exemplarily depicted for the time 6 h (figure 10). The intersection of differentially expressed genes in the PTCH group (MSC treated with AB-serum+PTCH versus MSC treated with AB-serum) and the IL1B group (MSC treated with AB-serum+PTCH versus MSC treated with AB-serum) was high (6 h: 408 differentially expressed genes compared to unstimulated cells; 24 h: 342 differentially expressed genes compared to unstimulated cells; both at 6 h and 24 h: 378 differentially expressed genes compared to unstimulated cells). The intersection of differentially expressed genes in the PTCH group (MSC treated with AB-serum+PTCH versus MSC treated with AB-serum) and the PTCL group (MSC treated with AB-serum+PTCL versus MSC treated with AB-serum) was high (6 h:

Results

171 differentially expressed genes compared to unstimulated cells; 24 h: 163 differentially expressed genes compared to unstimulated cells; both at 6 h and 24 h: 202 differentially expressed genes compared to unstimulated cells).

Figure 10 Intersection of differentially expressed genes in the PTCH (II), PTCL (III) and IL1B (IV) group 6 h after stimulation of MSC. PTCH: polytrauma cocktail high; PTCL: polytrauma cocktail low; IL1B: interleukin 1 beta. Modified figure according to Amann et al. (figure 2 (B) on page 7) [4], CC BY 4.0 https://creativecommons.org/licenses/by/4.0/.

These results indicated that the effects of PTCH can be simulated mainly by IL1B stimulation and that higher concentrations of factors have more effects on gene expression than lower concentrations of factors. Furthermore, referring to the components of PTCH (IL1B, IL6, C3a, C5a, HMGB1, THBD, THPO) and to protein analysis (IL1B, IL6, TNF, CXCL1, CCL2, CXCL8, CCL4, VEGFA, FGF2, MMP1, MMP2, MMP10, IL10, IL17A, IL1RN, TNFAIP6) below, the intersection of differentially expressed genes (6 h and 24 h) in the PTCH group and the IL1B group contained an upregulated expression of C3a, IL1B, IL6, CXCL1, CCL2, CXCL8, FGF2, MMP1, MMP10, IL1RN and TNFAIP6. However, IL1RN was only differentially expressed 6 h after stimulation and not anymore 24 h after stimulation with PTCH or IL1B (detailed results are shown in appendix, table 26).

Table 13 shows regulated pathways in MSC after stimulation with PTCH or IL1B. Expressed genes of the intersection of both groups (PTCH and IL1B) were assigned to pathways by the Institute for Medical Systems Biology. The 30 most frequently involved signalling pathways of 198 assigned pathways are listed here

Results

(see appendix for complete list of assigned signalling pathways). More than half of expressed genes could not be assigned to any signalling pathway.

Table 13 Expressed genes of the intersection of the polytrauma cocktail high group and interleukin 1 beta group 6 h, 24 h after stimulation or both times taken together (6 h and 24 h) were assigned to signalling pathways. The 30 most frequently involved signalling pathways are listed here. In some cases, several signalling pathways had been assigned to one gene. Upregulated (+) and downregulated (-) genes compared to unstimulated MSC are listed. TNF: tumor necrosis factor (alpha); PI3K: phosphatidylinositol-4,5-bisphosphate 3-kinase; NOD: nucleotide-binding oligomerization domain; JAK: Janus kinase; STAT: signal transducer and activator of transcription; NF-kappaB: nuclear factor kappa-light-chain-enhancer of activated B cells; MAPK: mitogen- activated protein kinases; HTLV: human T-lymphotropic virus.

KEGG pathway (Homo sapiens) Regulation 6h 24h 6h Total number compared to and of unstimulated 24h differentially MSC expressed genes in pathway - 1 0 3 Cytokine-cytokine receptor interaction 106 + 35 28 39 - 0 0 0 TNF signalling pathway 66 + 22 17 27 - 1 4 3 Metabolic pathways 57 + 16 14 19 - 2 2 5 Pathways in cancer 56 + 17 11 19 - 0 0 0 Chemokine signalling pathway 52 + 19 13 20 - 0 0 0 Rheumatoid arthritis 48 + 16 15 17 Transcriptional misregulation in - 0 2 1 42 cancer + 14 10 15 - 2 2 3 PI3K-Akt signalling pathway 41 + 10 10 14 - 0 0 0 Influenza A 35 + 12 7 16 - 0 0 0 NOD-like receptor signalling pathway 34 + 12 9 13 - 0 0 0 Jak-STAT signalling pathway 33 + 10 10 13 - 0 0 0 NF-kappa B signalling pathway 32 + 11 7 14 - 1 1 6 MAPK signalling pathway 31 + 6 7 10 - 1 1 4 Ras signalling pathway 29 + 6 7 10 Neuroactive ligand-receptor - 5 10 3 29 interaction + 4 2 5

Table 13 is continued on page 61.

Results

Table 13, continued. KEGG pathway (Homo sapiens) Regulation 6h 24h 6h Total number compared to and of unstimulated 24h differentially MSC expressed genes in pathway - 0 0 0 Legionellosis 27 + 9 7 11 Chagas disease (American - 0 0 0 27 trypanosomiasis + 10 6 11 - 0 0 0 Salmonella infection 27 + 9 8 10 - 0 0 0 Amoebiasis 27 + 9 9 9 - 0 0 1 HTLV-I infection 26 + 8 4 13 - 0 0 0 Pertussis 25 + 7 8 10 - 0 1 0 Tuberculosis 25 + 8 5 11 - 0 0 0 Hematopoietic cell lineage 24 + 8 7 9 - 3 6 3 Calcium signalling pathway 24 + 4 3 5 - 0 0 0 Toll-like receptor signalling pathway 23 + 8 5 10 - 0 1 0 Herpes simplex infection 22 + 7 6 8 - 2 2 3 Regulation of actin cytoskeleton 22 + 5 4 6 - 1 2 1 Proteoglycans in cancer 22 + 6 6 6 - 0 0 0 Malaria 21 + 7 6 8 - 0 0 0 Apoptosis 20 + 7 3 10 - 44 58 62 Pathway not assigned 677 + 189 146 178 - 12 32 19 Pathway assigned 451 + 133 106 149

Results

3.1.4. Matrix metallopeptidase and cytokine secretion of mesenchymal stromal cells after stimulation with trauma factors

To determine the effect of PTCH, PTCL and IL1B stimulation (10 ng/ml IL1B) on the migratory ability and secretion profile of MSC a broad range of factors was measured using either Multiplex assay or ELISA. Firstly, the background of factors in the media with supplements that were used for the MSC cultures was determined: Alpha MEM supplemented with 20 % AB- serum (without MSC) had a higher TNFAIP6 (2.7-fold), MMP2 (15-fold), MMP9 (2.6-fold), MMP10 (4.3-fold), CCL2 (3.3-fold) and CXCL8 (4.5-fold) level, but a 79 % lower CXCL1 level than Alpha MEM supplemented with 8 % PL (without MSC). Figure 11 to 13 depict the level of factors in the supernatant. To depict only the secretion of factors by MSC, cytokine level of cell culture medium without MSC was subtracted from cytokine level of supernatant from cell culture with MSC for each cytokine. Alpha MEM supplemented with 8 % PL contained more TNFAIP6, IL10, IL1RN, IL1B and FGF2 compared to the amount produced by the cells themselves. Cell culture medium supplemented either with PL or AB-serum contained more MMP9 compared to the amount produced by the cells themselves. MSC cultured for 6 h or 24 h in Alpha MEM supplemented with 20 % AB-serum without stimulation produced significantly higher TNFAIP6, IL10 and IL1RN levels than MSC cultured in Alpha MEM supplemented with 8 % PL. FGF2 secretion by MSC in Alpha MEM supplemented with 20 % AB-serum was significantly higher than FGF2 secretion by MSC cultured in Alpha MEM supplemented with 8 % PL after 24 h of incubation. CCL4 level from MSC in Alpha MEM supplemented with 20 % AB-serum was significantly higher than CCL4 level of supernatant from MSC cultured in Alpha MEM supplemented with 8 % PL after 6 h. IL1B level was highest in supernatant from IL1B-stimulated MSC and PTCH- stimulated MSC, as already expected. IL6 secretion by MSC was increased by IL1B stimulation (3.3-fold), PTCH stimulation (4.3-fold) and PTCL stimulation (4.4- fold) compared to unstimulated MSC after 6 h of incubation. TNF secretion was significantly influenced at an early point in time (6 h) by stimulation of MSC with IL1B (4.4-fold increase compared to unstimulated MSC, P=0.0023), PTCH (5.0- fold increase compared to unstimulated MSC, P=0.0004) and PTCL (5.8-fold increase compared to unstimulated MSC, P<0.0001). Both IL6 and TNF levels 62

Results were only increased for a brief period in early stage and levels were almost 50 % lower 24 h after stimulation compared to 6 h after stimulation. CXCL1 secretion was increased by stimulation with IL1B (6 h: 12-fold; 24 h: 6.4-fold), PTCH (6 h: 18-fold, P=0.0063; 24 h: 6.7-fold) and PTCL (6 h: 14-fold; 24 h: 6.7-fold), especially after 6 h of incubation. CCL2 secretion was 2.1-fold higher after PTCH stimulation of MSC compared to unstimulated MSC after 6 h of incubation. CXCL8 level of supernatant from IL1B-stimulated MSC (2.0-fold), PTCH-stimulated MSC (6.4-fold) and PTCL-stimulated MSC (4.2-fold) was also increased compared to unstimulated MSC 6 h after stimulation. VEGFA secretion was 2.5-fold higher after IL1B stimulation and PTCH stimulation compared to unstimulated MSC after 24 h of incubation.

Results

Figure 11 Cytokine secretion of human MSC. MSC were cultured either in Alpha MEM supplemented with 8 % platelet lysate and 1 IU/ml heparin (8 % PL) or in Alpha MEM supplemented with 20 % AB-serum. MSC in Alpha MEM supplemented with 20 % AB-serum were either unstimulated (-) or stimulated with interleukin 1 beta (IL1B), polytrauma cocktail high (PTCH) or polytrauma cocktail low (PTCL). Cytokine level of cell culture medium without MSC was subtracted from cytokine level of the supernatant from cell culture. Levels of cytokines after 6 h and 24 h were analysed. Mean and SD of n=3 are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, ** P < 0.01, *** P < 0.001, **** P < 0.0001, ## P < 0.01 versus 8 % PL 6 h, ### P < 0.001 versus 8 % PL 6 h, #### P < 0.0001 versus 8 % PL 6 h, §§§§ P < 0.0001 versus 8 % PL 24 h. IL6: interleukin 6; TNF: tumor necrosis factor (alpha); CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; CXCL8: C-X-C motif chemokine ligand 8. Parts of the data were published in Amann et al. (figure 5 on page 12) [4], CC BY 4.0 https://creativecommons.org/licenses/by/4.0/.

Results

Figure 12 Cytokine secretion of human MSC. MSC were cultured either in Alpha MEM supplemented with 8 % platelet lysate and 1 IU/ml heparin (8 % PL) or in Alpha MEM supplemented with 20 % AB-serum. MSC in Alpha MEM supplemented with 20 % AB-serum were either unstimulated (-) or stimulated with interleukin 1 beta (IL1B), polytrauma cocktail high (PTCH) or polytrauma cocktail low (PTCL). Protein level of cell culture medium without MSC was subtracted from protein level of the supernatant from cell culture. Levels of cytokines after 6 h and 24 h were analysed. Mean and SD of n=3 are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, **** P < 0.0001, # P < 0.05 versus 8 % PL 6 h, ## P < 0.01 versus 8 % PL 6 h, ### P < 0.001 versus 8 % PL 6 h, §§§ P < 0.001 versus 8 % PL 24 h, §§§§ P < 0.0001 versus 8 % PL 24 h. MMP1: matrix metallopeptidase 1; MMP2: matrix metallopeptidase 2; MMP10: matrix metallopeptidase 10; IL10: interleukin 10; IL17A: interleukin 17A; IL1RN: interleukin 1 receptor antagonist. Parts of the data were published in Amann et al. (figure 5 on page 12) [4], CC BY 4.0 https://creativecommons.org/licenses/by/4.0/.

Results

Figure 13 Cytokine secretion of human MSC. MSC were cultured either in Alpha MEM supplemented with 8 % platelet lysate and 1 IU/ml heparin (8 % PL) or in Alpha MEM supplemented with 20 % AB-serum. MSC in Alpha MEM supplemented with 20 % AB-serum were either unstimulated (-) or stimulated with interleukin 1 beta (IL1B), polytrauma cocktail high (PTCH) or polytrauma cocktail low (PTCL). Protein level of cell culture medium without MSC was subtracted from protein level of the supernatant from cell culture. Levels of cytokines after 6 h and 24 h were analysed. Mean and SD of n=3 are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, ** P < 0.01, # P < 0.05 versus 8 % PL 6 h, #### P < 0.0001 versus 8 % PL 6 h, §§ P < 0.01 versus 8 % PL 24 h, §§§ P < 0.001 versus 8 % PL 24 h, §§§§ P < 0.0001 versus 8 % PL 24 h. VEGFA: vascular endothelial growth factor-A; FGF2: fibroblast growth factor 2; CCL4: C-C motif chemokine ligand 4; TNFAIP6: tumor necrosis factor alpha inducible protein 6. Parts of the data were published in Amann et al. (figure 5 on page 12) [4], CC BY 4.0 https://creativecommons.org/licenses/by/4.0/.

There was a 2.5-fold increase of MMP1 secretion after stimulation of MSC with PTCH for 6 h compared to unstimulated MSC. However, 24 h after IL1B or PTCH stimulation of MSC MMP1 secretion was 4.5-fold higher compared to unstimulated MSC. Stimulation of MSC with PTCL did not have such a strong effect on MMP1 secretion, there was only a 2.6-fold increase of MMP1 secretion compared to unstimulated MSC after an incubation time of 24 h. IL1B stimulation and PTCH stimulation increased MMP10 secretion by MSC 5.0- fold or 4.0-fold compared to unstimulated MSC after an incubation time of 24 h. However, PTCL stimulation of MSC reduced MMP10 secretion by half compared to unstimulated MSC after 24 h. Furthermore, stimulation of MSC with IL1B led to 66

Results a significantly higher MMP10 secretion (10-fold, P=0.0287) than stimulation of MSC with PTCL 24 h after stimulation. IL10 level in supernatant from IL1B-stimulated MSC was reduced by half compared to IL10 level in supernatant from unstimulated MSC. In addition, one-way ANOVA (stimulated (8 % PL, IL1B, PTCH, PTCL) versus unstimulated cells (-)) revealed significant differences with regard to IL6, CXCL1 and CXCL8 levels in supernatant. Levels of these cytokines differed significantly comparing supernatant from unstimulated MSC to supernatant from stimulated MSC (IL1B, PTCH or PTCL) after 24 h of incubation. In general, IL1B stimulation and PTCH stimulation induced secretion of IL6, CXCL1, CXCL8, VEGFA, MMP1, MMP2 and MMP10 in a similar way. Most cytokine levels of supernatants were higher 24 h after stimulation than 6 h after stimulation, except for IL6, TNF, CXCL1, CCL2, CXCL8, CCL4 and TNFAIP6.

3.1.5. Cytokine analysis of AB-serum, serum from healthy volunteers and from polytrauma patients

We measured different cytokines/chemokines in serum from polytrauma patients at various points in time to describe the pattern of these factors after trauma. As a control serum from healthy volunteers and commercially available AB-serum (serum pooled from 20 donors) were analysed. Figure 14 depicts the content of albumin and the cytokine level of IL6, TNF, IL10, CXCL1, CCL2, CXCL8 and IL1RN in these sera. Table 14 depicts cytokine levels which were detectable only in individual samples. IL2, IL4, IL12p40, IL12p70 and IL17A were not detectable in sera from polytrauma patients at any time. IL1A and IL1B levels were close to or below the detection limit (IL1A: 9.4 pg/ml; IL1B: 0.8 pg/ml) in sera from healthy volunteers, in AB-serum as well as in sera from polytrauma patients. Serum levels of the following factors were lower in polytrauma patients compared to healthy controls: IL5 (58 % to 95 %), IL13 (0 h: 84 %; 12 h to 10 d: 56 % to 98 %), IL15 (0 h to 48 h: 76 % to 84 %; 10 d: 52 %) and CCL3 (48 h to 10 d: 57 % to 76 %). However, it must be considered critically, that the cytokines were only detectable in individual patients. Serum level of IFNG was 23.3 pg/ml in one patient at time 0 h, whereas the IFNG level was not detectable in other patients at any time. In most cases IFNG was not detectable in sera from polytrauma patients. Serum

Results levels of the growth factors VEGFA (52 % to 80 %) and FGF2 (4 h to 10 d: 95 %) were lower in trauma patients compared to healthy volunteers. Both factors were only detectable in individual patients at certain times. In general, AB-serum had similar cytokine levels as sera from healthy volunteers. Trauma patients received fluid resuscitation which leads to dilution of proteins in the serum. In order to quantify this effect concentration of albumin was determined. The albumin level in serum from polytrauma patients was significantly lower (e.g. 48 h: 51 %, P=0.0003; 5 d and 10 d: 69 %, P<0.0001) at any time compared to the albumin level in serum from healthy persons. Furthermore, there was a significant decrease of 52 % of albumin level 5 d or 10 d after trauma compared to 4 h (4 h versus 5 d: P=0.0444; 4 h versus 5 d: P=0.0434). The IL6 levels in serum from polytrauma patients was 47-fold higher (P=0.0027) at 0 h and 43-fold higher 24 h after admission to hospital compared to IL6 level in serum from controls. IL6 level in serum from polytrauma patients 4 h and 12 h after admission was 25-fold or 23-fold (P=0.0270) higher compared to controls. A gradual reduction of cytokine increase after polytrauma could be observed at 48 h (9.0-fold higher compared to control), 5 d (6.6-fold higher compared to control) and 10 d (3.8-fold higher compared to control). The TNF level was higher in polytrauma patients compared to controls (0 h: 6.9-fold; 4 h: 2.6-fold; 5 d: 2.1-fold; 10 d: 3.5- fold). There was an increase of IL10 in serum from polytrauma patients up to 12 h after polytrauma compared to the IL10 level of healthy persons (0 h: 25-fold; 4 h: 5.0-fold; 12 h: 1.4-fold). However, thereafter IL10 level of polytrauma patients decreased below the level of controls (85 % at 24 h, 90 % at 48 h, 93 % at 5 d and 76 % at 10 d). The CXCL1 level was significantly increased (4.6-fold, P=0.0008) in polytrauma patients after admission to hospital (0 h) compared to controls. The CXCL1 level decreased to the CXCL1 level of healthy controls between 4 h and 5 d after admission to hospital. 10 d after admission, the increase was 2.4-fold higher compared to controls. The CCL2 level in serum from polytrauma patients was increased compared to serum level of healthy controls. The increase of CCL2 level was strongest directly after admission to hospital (0 h: 14-fold, P=0.0009) compared to healthy controls, 4 h after admission the increase of CCL2 level was only 5.5-fold compared to healthy controls and 12 h to 10 d the level of CCL2 was 3.1-fold to 3.7-fold higher compared to healthy controls. The CXCL8 level was increased in polytrauma patients especially between 0 h (3.8-fold) and 24 h (2.5-

Results fold) after admission compared to the CXCL8 level of controls. The IL1RN level was only increased at time 0 h (3.4-fold) and 4 h (3.8-fold) compared to controls. After these two points in time, the IL1RN level in polytrauma patients was lower (64 % to 88 %) compared to controls.

Results

pg/ml; pg/ml; IFNG:

pg/ml; CCL3: 2.9

15: 1.2

A; FGF2: fibroblastA; growthfactor 2.

pg/ml; pg/ml; IL

pg/ml; pg/ml; IL13: 1.3

serum (n=1) and in serum from polytrauma patients (n=5). Mean and

pg/ml; IL5: 0.5

amma; VEGFA:vascular endothelial growthfactor

pg/ml; IL1B: 0.8

pg/ml. IL1A: interleukin 1 alpha; IL1B: interleukin 1 beta; IL5: interleukin 5; IL13: interleukin 13; IL15: interleukin

pg/ml; pg/ml; FGF2: 7.6

shown. Assay sensitivities: IL1A: 9.4

are are

C motif C chemokine ligand3; IFNG: interferong

Analysis of cytokines in serum from healthy volunteers (control; n=5), in AB

in in pg/ml

pg/ml; VEGFA: 26.3

Table SD 0.8 CCL3: 15; C

Results

Figure 14 Analysis of cytokines in serum from healthy volunteers (control; n=5), in AB-serum (n=1) and in serum from polytrauma patients (n=5). Mean and SD are shown. Significance was calculated with one-way ANOVA with Bonferroni’s multiple comparisons test (albumin and CXCL1) and one-way ANOVA followed by Dunn’s multiple comparisons test. IL6: interleukin 6; TNF: tumor necrosis factor (alpha); IL10: interleukin 10; CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; CXCL8: C-X-C motif chemokine ligand 8; IL1RN: interleukin 1 receptor antagonist. Parts of the data were published in Amann et al. (figure 3 (A) on page 8) [4], CC BY 4.0 https://creativecommons.org/licenses/by/4.0/.

Results

In the present study, the effects of MSC in vivo were investigated in a rat TxT model, both rMSC (as allogeneic cells) and hMSC (as xenogeneic cells) were used. Since this served as a pre-clinical model for potential use of human MSC in polytrauma patients, hMSC were used. In the following sections characteristics of rMSC (see section 3.2., 3.3.2., 3.3.5.) and hMSC (see section 3.3.2., 3.3.3, 3.3.4.) are described, which were used for the TxT model.

3.2. Characterization of rat mesenchymal stromal cells

The isolated rMSC were characterized via measurement of expression of surface markers and tri-lineage differentiation. Since there is no clear definition of rMSC by unique surface markers, several markers described for commercially available rMSC were tested including markers that should be expressed on rMSC (e.g. CD29, CD90) and markers which are specific for other haematopoietic cell types co-isolated with rMSC.

Figure 15 Flow cytometric analysis of surface marker antigens of rat MSC at passage one to four (n=16). Percentage of positive (stained) cells is shown. Unspecifically stained cells were determined by isotype control. Mean and SD of the percentage of positive cells are shown. CD: cluster of differentiation.

Figure 15 shows that about 90.00 % of the isolated rMSC expressed the surface markers CD90 (89.34 % ± 6.38 %), CD54 (91.06 % ± 9.26 %) and CD29

Results

(92.40 % ± 4.36 %), but only a few expressed the surface markers CD31 (-0.08 % ± 0.56 %), CD103 (0.00 % ± 0.67 %), CD49d (2.06 % ± 2.66 %) and CD106 (2.39 % ± 4.82 %). Furthermore, 9.84 % ± 7.72 % of the rMSC were positive for the surface marker CD11b/c and 11.94 % ± 7.16 % for the marker CD45.

Figure 16 Differentiation capacity of rat MSC at passage one to four (n=16). Adipogenic (Oil RedO/haematoxylin staining), chondrogenic (methylene blue staining) and osteogenic (detection of alkaline phosphatase) differentiation assays are shown. Rat MSC were cultured in differentiation media containing differentiation factors. Control assays were performed in Alpha MEM supplemented with 10 % FBS.

Figure 16 shows the differentiation capacity of rMSC. In general, adipogenic differentiation was seen in a lower percentage of cells than chondrogenic and osteogenic differentiation.

3.3. Effects of manipulating MSC for in vivo experiments

Previous results have shown that MSC could be activated by IL1B stimulation [28,99]. MSC were labelled with the stable membrane dye PKH26 to study the distribution of MSC in vivo in the rat TxT model. Therefore, characteristics of PKH26-stained MSC and IL1B-primed MSC were investigated and described below.

Results

3.3.1. Fluorescence intensity and cell growth of PKH26-stained human and rat mesenchymal stromal cells

MSC were stained with the PKH26 red fluorescent cell linker kit. Two different PKH26 dye concentrations were used (manufacturer protocol: 2.0 x 10-6 M; modified protocol used in rat experiment (see section 3.4.): 1.6 x 10-5 M). Unstained cells were used as control. Fluorescence intensity over time, CFU and doubling time of unstained and PKH26-stained cells were analysed.

Decrease of fluorescence intensity over time Figure 17 depicts the mean of the median fluorescence intensity of unstained and PKH26-stained MSC. Table 15 shows the P-values for the different staining protocols (unstained, stained according to manufacturer’s protocol and according to a modified protocol used in rat experiment). MSC which were stained with PKH26 according to manufacturer’s protocol did not significantly differ from unstained MSC. However, MSC which were stained according to the modified protocol used in the rat experiment differed significantly from unstained MSC. Fluorescence intensity of unstained rMSC was 3.3-fold higher than fluorescence intensity of hMSC, however fluorescence intensity of both cell types was different in reverse direction (0.5-fold) in rat experiment. Therefore, these variations may be based more on differences in flow cytometry measurement. Human MSC stained with PKH26 according to protocol used in rat experiment had a 647-fold higher initial (0 h) mean fluorescence intensity compared to unstained MSC and a 12-fold higher initial mean fluorescence intensity than MSC stained with PKH26 according to manufacturer’s protocol. Rat MSC stained with PKH26 according to protocol used in rat experiment had a 163-fold higher initial mean fluorescence intensity compared to unstained MSC and a 5.2-fold higher initial mean fluorescence intensity than MSC stained with PKH26 according to manufacturer’s protocol. Mean fluorescence intensity of hMSC stained according to manufacturer’s protocol was 26 % of the initial fluorescence intensity after 5 d. Mean fluorescence intensity of hMSC stained according to protocol used in rat experiment was 14 % of the initial fluorescence intensity after 5 d. Mean fluorescence intensity of rMSC stained according to manufacturer’s protocol was 75 % of the initial fluorescence intensity

Results after 5 d. Mean fluorescence intensity of rMSC stained according to protocol used in rat experiment was 65 % of the initial fluorescence intensity after 5 d. Fluorescence intensity of MSC stained according to the protocol used in rat experiment was significantly different to fluorescence intensity of MSC stained according to manufacturer’s protocol throughout the entire measurement period.

Figure 17 Fluorescence intensity of human MSC (hMSC) and rat MSC (rMSC) after PKH26 staining. Two PKH26 staining protocols (manufacturer´s protocol and protocol used in rat experiment) were performed and fluorescence intensity was measured 0 h, 24 h, 48 h and 120 h after staining (n=3). Mean and SD of median fluorescence intensities of unstained (black line) and PKH26-stained MSC (grey or red line) are depicted. For statistical analysis see table 15.

Table 15 Statistical analysis of fluorescence intensity of human MSC (hMSC) and rat MSC (rMSC) after PKH26 staining. Four different times were analysed. P-values were determined by one-way ANOVA followed by Bonferroni’s multiple comparisons test.

0 h 24 h 48 h 120 h

hMSC rMSC hMSC rMSC hMSC rMSC hMSC rMSC Unstained versus >0.9999 0.9852 0.4978 0.2574 0.1341 0.1775 0.4160 >0.9999 manufacturer’s protocol Unstained versus protocol used 0.0259 0.0038 0.0004 <0.0001 <0.0001 <0.0001 <0.0001 0.0153 in rat experiment manufacturer’s protocol versus 0.0373 0.0106 0.0011 0.0002 0.0003 0.0002 0.0001 0.0445 protocol used in rat experiment

Results

Doubling time Table 16 depicts the doubling time of hMSC and rMSC without and with PKH26 staining. Human MSC had seven to eight population doublings within 7 d of cell culture. Rat MSC hat two to three population doublings within 7 d.

Table 16 Doubling time of unstained and PKH26-stained human MSC (hMSC) and rat MSC (rMSC). Doubling time (h) was calculated after 7 d of cell culture. Two different staining protocols were used (n=3). Mean and SD are depicted.

hMSC rMSC Doubling time (h) Unstained 22 ± 0 72 ± 21 PKH26: manufacturer’s protocol 22 ± 1 68 ± 13 PKH26: protocol used in rat experiment 23 ± 1 66 ± 13

Colony forming units Table 17 shows CFU counts of unstained and PKH26-stained hMSC and rMSC.

Table 17 Counts of colony forming units of unstained and PKH26-stained human MSC (hMSC) and rat MSC (rMSC). Colonies with more than 50 cells were counted manually in a 6 well cell culture plate. Two different seeding densities and two different staining protocols were used (hMSC: n=5; rMSC: n=3). Mean and SD are depicted.

hMSC rMSC Seeding density 8 cells/cm² 16 cells/cm² 16 cells/cm² Unstained 14 ± 6 23 ± 5 7 ± 4 PKH26: manufacturer’s protocol 23 ± 20 33 ± 27 17 ± 8 PKH26: protocol used in rat experiment 14 ± 12 26 ± 19 14 ± 6

There are no significant differences between the CFU counts of unstained, of PKH26-stained MSC according to manufacturer’s protocol and of PKH26-stained MSC according to protocol used in rat experiment. Thus, the PKH26 staining of MSC does not negatively impact proliferation or clonogenicity of rMSC.

Results

3.3.2. Surface marker expression of PKH26-stained human and rat mesenchymal stromal cells

Change in surface marker expression was analysed using the same antibodies and at least the MSC preparations used in rat experiment. Although there are some other characteristic surface markers of MSC, for rat experiments three positive markers and at least two negative markers were chosen. Gating strategies and surface marker expression were published in Amann et al. [5]. It was shown that PKH26 staining did not have significant effects on surface marker expression (figure 18) and the proportion of dead cells (figure 19-21).

Figure 18 Percentage of positive events in life gate of unstained (A) rat (n=8) and (B) human (n=3- 4) MSC and PKH26-stained (PKH+) rat (n=6) and human (n=5) MSC for the (A) rMSC marker combination CD3, CD11b/c and CD45 and for the markers CD90, CD54, CD29 as well as the (B) hMSC marker CD3, CD45, CD73, CD90 and CD105. These figures were published in Amann et al. (figure 1 B on page 222 and supplement figure 3 on page 5) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

Percentage of CD3- or CD45-positive cells in flow cytometry analysis was less than 5.00 % in hMSC with or without PKH26 staining or IL1B stimulation (figure 1 B). Percentage of CD73-, CD90- or CD105-positive cells was ≥ 90 % in PKH26- stained hMSC with or without IL1B priming (data not shown). These cells fulfilled release criteria regarding purity and identity as specified for clinical trials of hMSC [262]. Marker expression of hMSC was not significantly influenced by PKH26 staining. Rat MSC showed expression of CD90, CD54 and CD29 (figure 18 A). IL1B priming of unstained rMSC (n=4; 80.03 % ± 9.85 %) led to a significantly

Results higher expression of CD54 compared to unstained (n=4; 46.08 % ± 20.45 %; P=0.0004) and to PKH26-stained rMSC. (n=4; 47.83 % ± 22.91 %; P=0.0009). Figure 18 A includes these data of PKH26-unstained and –stained IL1B-primed rMSC. Percentage of cell death of hMSC and rMSC is presented in figure 19 and 20. Percentage of the cell death marker “SYTOX or Annexin V” in hMSC were 21.33 % ± 3.79 % and in PKH26-stained hMSC 19.36 % ± 9.92 %. IL1B-primed hMSC contained 16.30 % ± 4.47 % SYTOX- or Annexin V-positive cells and IL1B- primed and PKH26-stained hMSC contained 22.40 % ± 7.79 % SYTOX- or Annexin V-positive cells. Percentage of the cell death marker SYTOX or Annexin V in rMSC was 26.00 % ± 8.58 % and in PKH26-stained rMSC it was 22.28 % ± 5.87 %. IL1B-primed rMSC contained 24.83 % ± 2.37 % SYTOX- or Annexin V-positive cells and IL1B-primed and PKH26-stained rMSC contained 21.70 % ± 9.33 % SYTOX- or Annexin V-positive cells. These results are in line with the results of trypan blue staining after cell harvest with or without PKH26 staining (figure 21). Percentage of living cells was not significantly different between hMSC and rMSC with or without IL1B pre-treatment.

Figure 19 Cell death marker of rat MSC (rMSC) which were used in the rat experiment. Unstained rMSC (n=4), PKH26-stained (PKH+) rMSC (n=4), IL1B-primed unstained rMSC (n=4) and IL1B- primed, PKH+ rMSC (n=2, four donors and two measurements represented) were analysed. Mean and SD of percentages of positive cells in life gate are depicted. CD: cluster of differentiation.

Results

Figure 20 Cell death marker of human MSC (hMSC). Two of these hMSC preparations were used in the rat experiment. Unstained hMSC (n=3-4), PKH26-stained (PKH+) hMSC (n=4), IL1B-primed hMSC (n=5), and IL1B-primed, PKH+ hMSC (n=5) were analysed. Mean and SD of percentage of positive cells in life gate are depicted. CD: cluster of differentiation.

Figure 21 Analysis of trypan blue staining of cell preparations which were used in the rat experiment. Mean and SD of percentages of living cells of unstained and PKH26-stained (PKH+) human MSC (hMSC) and rat MSC (rMSC) without and with IL1B pre-treatment (hMSC: n=6-7; hMSC (IL1B): n=5-6; rMSC: n=4; rMSC (IL1B): n=4) are depicted. Significance was calculated with Welch’s t-test, * P < 0.05.

3.3.3. Effect of interleukin 1 beta stimulation on secretion profile of human mesenchymal stromal cells

IL1B stimulation of MSC was already described previously [28,99]. In contrast to previously published studies, the cell culture medium used in this study was

Results supplemented with human PL instead of FBS. The presented cell culture experiments were performed to describe the secretion profile over time of IL1B- stimulated hMSC. Additionally, secretion profiles of IL1B pre-stimulated hMSC and PKH26-stained hMSC were analysed. Cytokine level of cell culture medium without hMSC was subtracted from cytokine level of supernatant from cell culture. Figures 22, 23 and 24 show the effect of a single IL1B stimulation on secretion profile of MSC after 4 h, 8 h, 12 h, 24 h and 48 h. Concentration of IL1B in supernatant was measured at different points in time to analyse consumption or degradation of added IL1B. At a starting concentration of 0.1 ng/ml and 1 ng/ml IL1B a continuous decrease of IL1B concentration in supernatant could be observed with increasing time. 100 ng/ml IL1B was above the detection limit (10000 pg/ml), except for 48 h after incubation with hMSC. With regards to the 10 ng/ml and 100 ng/ml starting concentration of IL1B a decline in IL1B concentration could be observed comparing the 4 h and the 48 h points in time (10 ng/ml IL1B: 4.4 ng/ml ± 1.7 ng/ml at 4 h versus 2.9 ng/ml ± 0.9 ng/ml at 48 h; 100 ng/ml IL1B: 9.8 pg/ml ± 8.6 ng/ml at 4 h versus 3.8 ng/ml ± 1.4 ng/ml at 48 h). IL6, CXCL1, CCL2 and CXCL8 concentrations in supernatant from IL1B- stimulated hMSC increased compared to cytokine concentration in supernatant from unstimulated hMSC from 4 h to 24 h after stimulation. Of note, concentration of chemokines increased earlier than of regenerative factors (VEGFA, TNFAIP6). TNFAIP6 level increased 48 h after IL1B stimulation. IL10, IL1RN, FGF2 and TNF are components of the cell culture medium or rather of PL [70] and were not produced by the cells themselves in these cell culture experiments. FGF2 concentration in supernatant decreased during cell expansion from 0 h to 48 h after stimulation. IL17A level in supernatant was below the detection limit. IL6 was produced by hMSC. Concentration of IL6 in supernatant increased 3.5-fold to 5.7- fold after IL1B stimulation (0.1 ng/ml, 1 ng/ml, 10 ng/ml, 100 ng/ml) compared to concentration of IL6 in supernatant from unstimulated hMSC from 4 h to 24 h after incubation. The increase of IL6 level was only 2.1-fold in supernatant from IL1B- stimulated (0.1 ng/ml or 100 ng/ml) hMSC compared to supernatant from unstimulated hMSC 48 h after incubation.

Results

Figure 22 Cytokine secretion of human MSC after stimulation with interleukin 1 beta (IL1B). Cytokine level of cell culture medium without MSC was subtracted from cytokine level of supernatant from cell culture. Cell culture supernatants were analysed 4 h, 8 h, 12 h, 24 h and 48 h after stimulation with 0 ng/ml, 0.1 ng/ml, 1 ng/ml, 10 ng/ml or 100 ng/ml IL1B. Additionally, cytokine secretion profile of PKH26-stained MSC (PKH+) and IL1B-primed MSC (IL1B-primed) 24 h after IL1B stimulation (0 ng/ml, 10 ng/ml) are shown. Mean and SD of cytokine levels in supernatant (n=5) are displayed. Significant changes after IL1B stimulation compared to unstimulated MSC were calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test or Welch’s t-test (PKH26-stained MSC, IL1B-primed MSC), * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. IL6: interleukin 6; TNFAIP6: tumor necrosis factor alpha inducible protein 6.

Results

The TNFAIP6 level increased constantly after stimulation with high IL1B concentration (10 ng/ml, 100 ng/ml). The TNFAIP6 level increased significantly 48 h after IL1B stimulation compared to unstimulated hMSC (0.1 ng/ml: 2.9-fold, P=0.0090; 1 ng/ml: 3.5-fold, P=0.0003; 10 ng/ml: 3.7-fold, P=0.0001; 100 ng/ml: 3.9-fold, P<0.0001). CXCL1 secretion (supernatant without background/cell culture medium) by unstimulated hMSC was 242 pg/ml ± 234 pg/ml 4 h after incubation and 485 pg/ml ± 367 pg/ml 48 h after incubation. CXCL1 secretion by IL1B- stimulated (0.1 ng/ml, 1 ng/ml, 10 ng/ml, 100 ng/ml) hMSC increased 6.6-fold to 14-fold compared to CXCL1 secretion by unstimulated hMSC 8 h to 24 h after incubation. The increase of CXCL1 level was only 2.7-fold to 3.7-fold in supernatant from IL1B-stimulated (0.1 ng/ml, 1 ng/ml, 10 ng/ml, 100 ng/ml) hMSC compared to supernatant from unstimulated hMSC 48 h after incubation. CCL2 secretion by IL1B-stimulated (0.1 ng/ml, 1 ng/ml, 10 ng/ml, 100 ng/ml) hMSC was 2.6-fold to 5.2-fold (24 h after stimulation with 100 ng/ml IL1B, P=0.0091) higher compared to CCL2 secretion by unstimulated hMSC from 4 h to 24 h after stimulation. CXCL8 secretion by hMSC was upregulated early (4 h) after IL1B stimulation (0.1 ng/ml, 1 ng/ml, 10 ng/ml) compared to unstimulated hMSC. CXCL8 secretion by stimulated hMSC declined over time (4 h to 48 h after incubation). CXCL8 secretion by IL1B-stimulated hMSC (0.1 ng/ml, 1 ng/ml, 10 ng/ml, 100 ng/ml) increased 19-fold to 21-fold compared to CXCL8 secretion by unstimulated hMSC 4 h after stimulation. 8 h after stimulation the increase of CXCL8 concentration was 8.9-fold (100 ng/ml IL1B) to 14-fold (1 ng/ml IL1B), 12 h after stimulation 7.1-fold (1 ng/ml IL1B) to 8.3-fold (100 ng/ml IL1B), 24 h after stimulation 4.2-fold (100 ng/ml IL1B) to 9.2-fold (0.1 ng/ml IL1B) and 48 h after stimulation 2.5-fold (100 ng/ml IL1B) to 3.2-fold (1 ng/ml IL1B). The VEGFA level in supernatant increased during cell expansion. Stimulation with 10 ng/ml IL1B significantly increased (P=0.0049, 1.5-fold) VEGFA secretion 48 h after stimulation compared to unstimulated hMSC. The FGF2 level in supernatant decreased during cultivation. Stimulation of hMSC with 10 ng/ml IL1B led to a 54 % decrease of FGF2 concentration in the supernatant compared to the supernatant from unstimulated hMSC after 24 h incubation. Stimulation of hMSC with 100 ng/ml IL1B led to a significant 55 % (P=0.0412) or 65 % (P=0.0247) decrease of FGF2 concentration in supernatant compared to supernatant from unstimulated hMSC after 4h or 24 h incubation.

Results

Figure 23 Cytokine secretion of human MSC after stimulation with interleukin 1 beta (IL1B). Cytokine level of cell culture medium without MSC was subtracted from cytokine level of supernatant from cell culture. Cell culture supernatants were analysed 4 h, 8 h, 12 h, 24 h and 48 h after stimulation with 0 ng/ml, 0.1 ng/ml, 1 ng/ml, 10 ng/ml or 100 ng/ml IL1B. Additionally, cytokine secretion profile of PKH26-stained MSC (PKH+) and IL1B-primed MSC (IL1B-primed) 24 h after IL1B stimulation (0 ng/ml, 10 ng/ml) are shown. Mean and SD of cytokine levels in supernatant (n=4-5) are displayed. Significant changes after IL1B stimulation compared to unstimulated MSC were calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test or Welch’s t-test (PKH26-stained MSC), * P < 0.05, ** P < 0.01. CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; CXCL8: C-X-C motif chemokine ligand 8. 83

Results

Figure 24 Cytokine secretion of human MSC after stimulation with interleukin 1 beta (IL1B). Cytokine level of cell culture medium without MSC was subtracted from cytokine level of supernatant from cell culture. Cell culture supernatants were analysed 4 h, 8 h, 12 h, 24 h and 48 h after stimulation with 0 ng/ml, 0.1 ng/ml, 1 ng/ml, 10 ng/ml or 100 ng/ml IL1B. Additionally, cytokine secretion profile of PKH26-stained MSC (PKH+) and IL1B-primed MSC (IL1B-primed) 24 h after IL1B stimulation (0 ng/ml, 10 ng/ml) are shown. Mean and SD of cytokine levels in supernatant (n=5) are displayed. Significant changes after IL1B stimulation compared to unstimulated MSC were calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. VEGFA: vascular endothelial growth factor-A; FGF2: fibroblast growth factor 2.

There was no significant difference between the four stimulation concentrations (0.1 ng/ml to 100 ng/ml) with regard to the induction of cytokine secretion. The secretion profile of PKH26-stained hMSC and IL1B-primed hMSC did not significantly differ from the secretion profile of unstained and unprimed hMSC, whether IL1B (10 ng/ml) was added for stimulation or not. In general, a second IL1B stimulation did not have the same strong effects on cytokine secretion than the first IL1B stimulation for 24 h. 84

Results

3.3.4. Co-culture of human mesenchymal stromal cells and the rat alveolar macrophage cell line NR8383

To investigate the interaction of human and rat cells, we performed co-culture of hMSC and the rat alveolar macrophage cell line NR8383 for 24 h. Previous publications had shown that IL1B activates MSC [28,99] and that LPS activates macrophages [198,205]. Both stimulants were used to investigate the secretion profile of hMSC and rat alveolar macrophages in individual cell culture or co- culture experiments. Human IL1B was used to activate hMSC. Cell culture medium was a 1:1 mixture of minimal culture medium of hMSC and cell culture medium of NR8383 cells. Cell culture supernatant was analysed by human- and rat-specific Multiplex assay. The Multiplex assay specific for human factors revealed that co-culture medium without cells contained 38 pg/ml CXCL1, the levels of the remaining cytokines (CCL2, IL6, VEGFA, IL10, IL1RN, CXCL8, IL1B, FGF2, TNF, IL17A) were below the detection limit. The Multiplex assay specific for rat factors revealed that co-culture medium without cells contained 31 pg/ml CCL2, the levels of the remaining cytokines (CXCL1, VEGFA, IL6, IL1B, TNF, IL10, CXCL2) were below the detection limit. TNFAIP6 assay showed that co-culture medium without cells contained 586 pg/ml TNFAIP6. In general, in the Multiplex assay specific for human factors, no cytokines were detected in the supernatant from NR8383 cells (except for very low concentrations of VEGFA, IL10 and FGF2), showing a high species-specificity for human protein. In the Multiplex assay specific for rat factors, no cytokines were detected in supernatant from hMSC (except for low concentrations of CXCL1, VEGFA and CCL2), showing a high species-specificity for rat protein. Cross-reactivity of antibodies is explained in detail in section 2.7.4.. Concentrations of Human IL1RN and human IL17A were below the detection limit of the human Multiplex assay in all supernatant specimens. Cytokine levels of cell culture medium without cells were subtracted from cytokine levels of supernatant from cell culture to analyse the cytokine secretion of the cells. Figure 25, 26 and 27 depict the human and rat cytokine levels in supernatant 24 h after starting co-culture.

Results

Figure 25 Cytokine level of supernatant from human MSC (hMSC), rat cell line NR8383 (NR8383) and co-culture of both of them (hMSC NR8383: direct co-culture; hMSC NR8383 (TW): co-culture with transwell insert). Cell culture supernatants were analysed 24 h after interleukin 1 beta (10 ng/ml, IL1B) or lipopolysaccharide (100 ng/ml, LPS) stimulation or cell culture without stimulation (-). Mean and SD of cytokine levels in supernatant (n=3) are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. *** P < 0.001, **** P < 0.0001. IL6: interleukin 6; TNF: tumor necrosis factor (alpha).

Results

Figure 26 Cytokine level of supernatant from human MSC (hMSC), rat cell line NR8383 (NR8383) and co-culture of both of them (hMSC NR8383: direct co-culture; hMSC NR8383 (TW): co-culture with transwell insert). Cell culture supernatants were analysed 24 h after interleukin 1 beta (10 ng/ml, IL1B) or lipopolysaccharide (100 ng/ml, LPS) stimulation or cell culture without stimulation (-). Mean and SD of cytokine levels in supernatant (n=3) are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. *** P < 0.001, **** P < 0.0001. CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; IL10: interleukin 10.

Results

Figure 27 Cytokine level of supernatant from human MSC (hMSC), rat cell line NR8383 (NR8383) and co-culture of both of them (hMSC NR8383: direct co-culture; hMSC NR8383 (TW): co-culture with transwell insert). Cell culture supernatants were analysed 24 h after interleukin 1 beta (10 ng/ml, IL1B) or lipopolysaccharide (100 ng/ml, LPS) stimulation or cell culture without stimulation (-). Mean and SD of cytokine levels in supernatant (n=3) are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. *** P < 0.001, **** P < 0.0001. VEGFA: vascular endothelial growth factor-A; CXCL8: C-X-C motif chemokine ligand 8; CXCL2: C-X-C motif chemokine ligand 2; FGF2: fibroblast growth factor 2; TNFAIP6: tumor necrosis factor alpha inducible protein 6.

Rat cytokine secretion of NR8383 cells was increased after LPS stimulation (CXCL1: 19-fold; IL6: 131-fold, P<0.0001; TNF: 93-fold, P<0.0001; CCL2: 3.5-fold,

Results

P=0.0100; IL10: 54-fold, P<0.0001; CXCL2: 7.8-fold, P<0.0001; TNFAIP6: 6.3- fold) compared to unstimulated NR8383 cells. In general, human IL1B did not increase cytokine secretion by NR8383 cells after an incubation time of 24 h compared to unstimulated NR8383 cells. Only TNFAIP6 level in supernatant from NR8383 cells increased 2.0-fold after IL1B stimulation compared to supernatant from unstimulated NR8383 cells. Human MSC secreted more TNFAIP6 (3.4-fold, P=0.0001), CXCL1 (67-fold, P<0.0001), IL6 (14-fold, P=0.0002), TNF (3.3-fold), CCL2 (4.5-fold) and CXCL8 (32-fold, P=0.0001) after IL1B stimulation compared to unstimulated hMSC analysed by ELISA and human-specific Multiplex assay. The effect of IL1B stimulation on the cytokine profile of hMSC was already described in section 3.1.4 and 3.3.3. However, in section 3.3.3., the cell culture medium had a higher PL supplementation. The effect of IL1B on the cytokine profile of hMSC was similar in both experiments, however the effects in the co-culture medium (FBS/PL supplementation) were stronger than in 8 % PL-supplemented medium (CXCL1 increase was 67-fold instead of 7.0-fold, CXCL8 increase was 32-fold instead of 7.9-fold, IL6 increase was 14-fold instead of 4.8-fold, TNFAIP6 increase was 3.4- fold instead of 2.6-fold, CCL2 increase was 4.5-fold instead of 3.7-fold) compared to the unstimulated hMSC. Stimulation of hMSC with LPS did not have significant effects on cytokine secretion analysed by human-specific Multiplex assay. However, this assay revealed non-significant increase of the CXCL1 (3.3-fold), IL6 (3.0-fold) and CXCL8 (6.4-fold) level in supernatant from hMSC after LPS stimulation compared to supernatant from unstimulated hMSC. IL1B-stimulated hMSC secreted more CXCL1 (39-fold, P=0.0001) and VEGFA (2.9-fold) compared to IL1B-stimulated NR8383 cells analysed in rat-specific Multiplex assay. Furthermore, hMSC secreted more TNFAIP6 (without stimulation: 24-fold; IL1B stimulation: 40-fold, P<0.0001; LPS stimulation: 4.0-fold) compared to the NR8383 cell line regardless of the activation status (without/IL1B/LPS stimulation). Rat-specific cytokine assay showed that supernatants from unstimulated co- culture of hMSC and NR8383 cells (independent of direct of transwell co-culture) had higher CXCL1 (direct co-culture: 3.2-fold compared to single cultures; transwell co-culture: 4.1-fold compared to single cultures) and VEGFA (2.8-fold compared to hMSC, 3.5-fold compared to NR8383 cells) levels compared to

Results supernatants from unstimulated single cultures of hMSC or NR8383 cells. Rat VEGFA level in supernatant was also higher in IL1B- (6.2-fold compared to NR8383 cells, 2.2-fold compared to hMSC) or LPS-stimulated (4.7-fold compared to NR8383 cells, 4.5-fold compared to hMSC) direct co-culture compared to IL1B- or LPS-stimulated single cultures of hMSC or NR8383 cells. Rat IL10 concentration in supernatant was higher (2.1-fold compared to NR8383 cells, 3.9- fold compared to hMSC) in IL1B-stimulated direct co-culture compared to IL1B- stimulated single cultures. TNFAIP6 level was significantly higher in supernatant from LPS-stimulated direct co-culture compared to LPS-stimulated single cultures (14-fold compared to NR8383 cells (P<0.0001), 3.6-fold compared to hMSC (P<0.0001)). Human-specific cytokine assay revealed that supernatant from co- culture of hMSC and NR8383 cells (independent of direct of transwell co-culture) without stimulation had an increased level of CXCL1 (direct: 3320-fold compared to NR8383 cells (P=0.0039), 33-fold compared to hMSC (P=0.0057); transwell: 3241-fold compared to NR8383 cells (P=0.0036), 33-fold compared to hMSC (P=0.0053)), IL6 (direct: 2245-fold compared to NR8383 cells, 6.9-fold compared to hMSC; transwell: 2147-fold compared to NR8383 cells, 6.6-fold compared to hMSC), CCL2 (direct: 887-fold compared to NR8383 cells, 2.4-fold compared to hMSC; transwell: 945-fold compared to NR8383 cells, 2.5-fold compared to hMSC) and CXCL8 (direct: 2826-fold compared to NR8383 cells, 16-fold compared to hMSC; transwell: 2609-fold compared to NR8383 cells, 15-fold compared to hMSC) compared to supernatant from unstimulated single cultures of hMSC or NR8383 cells. LPS stimulation of co-culture increased concentration of CXCL1 (direct: significant compared to NR8383 cells (P<0.0001), 15-fold compared to hMSC (P<0.0001); transwell: significant compared to NR8383 cells (P<0.0001), 14-fold compared to hMSC (P=0.0001)), VEGFA (direct: 4.5-fold compared to NR8383 cells, 2.1-fold compared to hMSC), IL6 (direct: 3406-fold compared to NR8383 cells (P=0.0023), 3.5-fold compared to hMSC; transwell: 2867-fold compared to NR8383 cells (P=0.0176), 2.9-fold compared to hMSC), CCL2 (direct: 1668-fold compared to NR8383 cells, 2.5-fold compared to hMSC) and CXCL8 (direct: 4780-fold compared to NR8383 cells (P=0.0006), 4.3-fold compared to hMSC (P=0.017); transwell: 4429-fold compared to NR8383 cells (P=0.0017), 4.0-fold compared to hMSC (P=0.0478)) in supernatant compared to

Results supernatant from LPS-stimulated single cultures measured by human-specific Multiplex assay. There were significant differences between direct and transwell co-culture regarding the TNFAIP6 (P=0.0057), rat TNF (P=0.0013) and rat IL10 (P=0.0042) level in LPS-simulated co-culture. Direct co-culture had a significantly higher cytokine level than transwell co-culture meaning that cell contact of hMSC and NR8383 cells is essential for a higher level of anti-inflammatory cytokines like TNFAIP6 and IL10.

3.3.5. Co-culture of rat mesenchymal stromal cells and the rat alveolar macrophage cell line NR8383

The rat alveolar macrophage cell line NR8383 from the lung was cultured alone or with rMSC for 24 h. The secretion profile of PKH26-stained or IL1B-primed rMSC was analysed and compared to unstained or unprimed rMSC cultured simultaneously. Cell culture medium was a 1:1 mixture of culture medium of rMSC and cell culture medium of NR8383 cells. Co-culture medium without cells was analysed by Multiplex assays and cytokine concentration was below the detection limit for all measured cytokines (TNF, IL10, IL1B, IL6, CXCL1, CCL2, VEGFA and CXCL2). TNFAIP6 level was analysed by ELISA method. The cell culture medium without cells contained 587 pg/ml TNFAIP6. TNFAIP6 level of cell culture medium without cells was subtracted from TNFAIP6 level of supernatant from cell culture to analyse the cytokine secretion of the cells. Figure 28 and 29 depict the cytokine level in supernatant 24 h after starting co-culture or stimulation. Unstimulated rMSC secreted VEGFA (457 pg/ml ± 182 pg/ml), CCL2 (81 pg/ml ± 54 pg/ml) and TNFAIP6 (364 pg/ml ± 388 pg/ml). Stimulation with IL1B increased CXCL1 (460-fold, P<0.0001), VEGFA (3.1-fold), IL6 (131-fold), CCL2 (86-fold, P<0.0001), IL10 (5.5-fold), CXCL2 (37-fold) and TNFAIP6 (2.5-fold) secretion of rMSC compared to unstimulated rMSC after 24 h. Stimulation with LPS increased CXCL1 (70-fold), CCL2 (21-fold), IL10 (3.2-fold) and CXCL2 (8.8- fold) secretion of rMSC compared to unstimulated rMSC. In general, the induction of secretion was weaker in LPS-stimulated rMSC than in IL1B-stimulated rMSC.

Results

Figure 28 Cytokine secretion of co-culture of rat MSC and the rat cell line NR8383. Cell culture supernatants were analysed 24 h after IL1B (10 ng/ml IL1B) or LPS (100 ng/ml LPS) stimulation or cell culture without stimulation (-). Mean and SD of cytokine levels in supernatant (n=3) are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. *** P < 0.001, **** P < 0.0001. IL1B: interleukin 1 beta; IL6: interleukin 6; TNF: tumor necrosis factor (alpha); CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; CXCL2: C-X-C motif chemokine ligand 2. 92

Results

Figure 29 Cytokine secretion of co-culture of rat MSC and the rat cell line NR8383. Cell culture supernatants were analysed 24 h after IL1B (10 ng/ml IL1B) or LPS (100 ng/ml LPS) stimulation or cell culture without stimulation (-). Mean and SD of cytokine levels in supernatant (n=3) are shown. Significance was calculated with two-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. *** P < 0.001, **** P < 0.0001, §§ P < 0.01 versus unstained IL1B- stimulated rMSC. IL10: interleukin 10; TNFAIP6: tumor necrosis factor alpha inducible protein 6; VEGFA: vascular endothelial growth factor-A.

Unstimulated NR8383 cells secreted IL1B (5.5 pg/ml ± 1.9 pg/ml), CCL2 (2197 pg/ml ± 295 pg/ml), CXCL2 (170 pg/ml ± 18 pg/ml) and TNFAIP6 (165 pg/ml ± 90 pg/ml). Stimulation with IL1B did not induce increased cytokine secretion by NR8383 cells compared to unstimulated NR8383 cells. Stimulation with LPS induced CXCL1 (9.4-fold), IL6 (122-fold), IL1B (2.1-fold), TNF (196-fold, P<0.0001), CCL2 (2.9-fold, P=0.0276), IL10 (42-fold, P<0.0001) and CXCL2 (24- fold, P=0.0003) secretion compared to unstimulated NR8383 cells after 24 h. In general, LPS stimulation of NR8383 cells induced secretion of inflammatory cytokines. Rat MSC and NR8383 cells in co-culture without stimulation increased cytokine secretion compared to unstimulated rMSC or unstimulated NR8383 cells alone (CXCL1: 20-fold in direct co-culture, 12-fold in transwell co-culture; IL10: 3.6-fold in direct co-culture or 2.6-fold in transwell co-culture compared to unstimulated 93

Results rMSC, 3.0-fold in direct co-culture or 2.2-fold in transwell co-culture compared to unstimulated NR8383 cells). There was no significant difference between direct co-culture and transwell co-culture. Stimulation with LPS had a significantly stronger effect on cytokine secretion (IL6, TNF, CCL2, IL10, CXCL2) in both direct and transwell co-culture compared to IL1B stimulation. CXCL1 level in supernatant from IL1B-stimulated co-culture was 168-fold higher than CXCL1 level in supernatant from IL1B-stimulated NR8383 cell culture and significantly lower (64 %, P=0.0031) than CXCL1 level in supernatant from IL1B-stimulated rMSC cell culture. Supernatant from IL1B-stimulated transwell co-culture had a significantly higher (244-fold, P=0.0260) CXCL1 level than supernatant from IL1B-stimulated NR8383 cells. Stimulation with LPS further increased CXCL1 (30-fold compared to NR8383 cells (direct co-culture: P=0.0063, transwell: P=0.0078) and 4.0-fold compared to rMSC with or without transwell), IL6 (direct co-culture: 6.0-fold compared to NR8383 cells (P=0.0002) or 451-fold compared to rMSC (P<0.0001); transwell: 3.7-fold compared to NR8383 cells or 280-fold compared to rMSC (P=0.0102)) and CCL2 (direct co-culture: 2.4-fold compared to NR8383 cells (P<0.0001) or 9.0-fold compared to rMSC (P<0.0001); transwell: 2.0-fold (P=0.0002) compared to NR8383 cells or 7.4-fold compared to rMSC (P<0.0001)) secretion in both direct co-culture and transwell co-culture compared to LPS- stimulated NR8383 cells and LPS-stimulated rMSC alone. VEGFA level in supernatant from LPS-stimulated direct co-culture was higher than in supernatant from LPS-stimulated NR8383 cells (VEGFA: 331-fold, P=0.0027) and LPS- stimulated rMSC (VEGFA: 2.1-fold) alone. Supernatant form LPS-stimulated transwell co-culture had a similar VEGFA level as the supernatant from LPS- stimulated rMSC. However, TNF level in supernatant from LPS-stimulated co- culture was 54 % lower (direct co-culture: P<0.0001) or 75 % lower (transwell co- culture: P<0.0001) compared to TNF level in supernatant from LPS-stimulated NR8383 cells. However, TNF level in supernatant from LPS-stimulated co-culture was 101-fold (direct co-culture: P=0.0007) or 78-fold (transwell co-culture: P=0.0163) higher than in supernatant from LPS-stimulated rMSC. Unstimulated PKH26-stained rMSC had a lower VEGFA (49 %), CCL2 (87 %) and TNFAIP6 (67 %) level compared to unstimulated unstained rMSC. IL1B-stimulated PKH26-stained rMSC secreted less VEGFA (57 %, P=0.0055) and TNFAIP6 (54 %) compared to IL1B-stimulated unstained rMSC. However, the relative

Results

VEGFA or TNFAIP6 level (level of IL1B or LPS-stimulated cells divided by level of unstimulated cells) of supernatant from PKH26-stained rMSC did not change compared to unstained rMSC. Relative CCL2 level in supernatant from PKH26- stained cells increased 7.0-fold compared to CCL2 level in supernatant from unstained cells. Supernatant from IL1B-primed rMSC had a higher chemokine level (CXCL1, CCL2) even after cultivation for additional 24 h in culture medium without stimulant compared to supernatant from unprimed, unstimulated rMSC. Repetition of IL1B stimulation decreased CXCL2 (54 %) and TNFAIP6 (51 %) levels in supernatant 24 h after the second stimulation compared to unprimed, IL1B-stimulated rMSC. In general, a second IL1B stimulation did not have the same strong effects on cytokine secretion than the first IL1B stimulation for 24 h.

3.3.6. Cross-reactivity of human mesenchymal stromal cells-specific antibodies to rat mesenchymal stromal cells and vice versa

We planned to study the recovery of rMSC and hMSC in the rat TxT model after systemic administration of MSC (see section 3.4.2). In order to assess whether species-specific detection of rMSC and hMSC is possible, we studied the cross- reactivity of antibodies for markers which were characteristic of rMSC and hMSC. Previously characterized hMSC were stained with rat-specific antibodies and previously characterized rMSC were stained with human-specific antibodies. Percentage of positive cells was calculated by subtraction of the percentage of unspecifically stained cells (isotype control) from the percentage of cells stained with antigen-specific antibodies. Results of flow cytometry analysis are shown in table 18 and 19. There was a small amount of hMSC that was positively stained by rat-specific antibodies. The percentage for these cells was below 5.0 %, except for CD45 (see table 18). As already mentioned the release criteria for hMSC in clinical studies include that negative markers of MSC are less than 5.0 % [262]. Therefore, these values were accepted.

Results

Table 18 Human MSC were stained with rat-specific antibodies (n=4). Mean and SD of percentages of cross-reactivity. CD: cluster of differentiation.

CD45, CD29 CD49d CD54 CD90 CD11b/c CD45 CD11b/c, CD3 -0.75±1.52 1.28±1.46 2.35±1.91 -0.14±0.34 0.43±0.64 4.80±3.69 0.35±0.58

The amount of rMSC positively stained with the human-specific antibodies for CD3, CD45, CD73, CD90 and CD105 were very low (<1.0 %).

Table 19 Rat MSC were stained with human-specific antibodies (n=3). Mean and SD of percentages of cross-reactivity. CD: cluster of differentiation.

CD3 CD45 CD73 CD90 CD105 -0.03±0.06 0.10±0.10 0.33±0.32 0.70±0.16 0.43±0.25

3.4. Animal experiments

Blunt chest trauma model was used to investigate local and systemic immunological changes after lung contusion. To determine if administration of xenogeneic or allogeneic MSC has an effect on inflammatory and regenerative processes, there were six experimental groups of animals: (i) sham, (ii) traumatized (TxT) rats and rats receiving (iii) hMSC or (iv) rMSC after TxT (TxT+hMSC/TxT+rMSC). Further groups included (v) IL1B-primed hMSC or (vi) IL1B-primed rMSC (TxT+hMSC (IL1B)/TxT+rMSC (IL1B)). Two points in time after trauma application were analysed (24 h, 5 d). Study design is depicted in figure 30.

Results

Figure 30 Time course of cell preparation, blunt chest trauma and MSC treatment. IL1B: stimulation of MSC with IL1B for 24 h; TxT: setting the blunt chest trauma; MSC: intravenous injection of MSC; † animal sacrificed; BAL: broncho-alveolar lavage fluid. This figure was already published in Amann et al. (supplemental figure 1 on page 3) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

3.4.1. Fluorescence intensity of PKH26-stained mesenchymal stromal cells before intravenous injection in rats

MSC were stained with the PKH26 dye enabling tracing and recovery of the injected cells. Figure 31 depicts the definition of flow cytometry gates for PKH26 staining of MSC. The median fluorescence intensity of PKH26-stained hMSC (24699 ± 9418) was significantly higher (285-fold; Wilcoxon: P=0.0078) compared to unstained hMSC (87 ± 11). Furthermore, IL1B-stimulated hMSC (92 ± 10) had a significant increase in fluorescence intensity after PKH26 staining (46131 ± 31712; 504-fold; P=0.0156). Unstimulated PKH26-stained rMSC had a fluorescence intensity of 4080 ± 723, which was higher (88-fold) than unlabelled rMSC (46 ± 9). IL1B- stimulated PKH26-stained rMSC (6152 ± 1107) had 117-fold higher fluorescence intensity than stimulated unlabelled rMSC (53 ± 5).

Results

Figure 31 Flow cytometry gating strategy for phycoerythrin (PE) fluorescence intensity of MSC. Fluorescence intensity of unstained and PKH26-stained MSC with or without IL1B priming was analysed in flow cytometry. Dot plot and histogram of unstained and PKH26-stained hMSC and rMSC without and with IL1B pre-treatment for 24 h are depicted. Definition of life gate and proportion of PKH26-positive cells are shown. FSC-H: forward scatter height; SSC-H: sideward scatter height. This figure was already published in Amann et al. (supplemental figure 2 on page 4) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

Figure 32 shows the median fluorescence intensity of the injected cells corrected by the subtraction of the unstained control. Fluorescence intensity of PKH+ hMSC was on average 6.1-fold higher than of PKH+ rMSC. Fluorescence intensity of PKH+ hMSC (IL1B) was 7.6-fold higher than of PKH+ rMSC (IL1B). Furthermore, the fluorescence intensity of PKH+ hMSC (IL1B) was significantly 11-fold higher (P=0.0211) than of PKH+ rMSC, while the fluorescence intensity of PKH+ hMSC was 4.0-fold higher than of PKH+ rMSC (IL1B). In summary, the PKH+ hMSC had higher fluorescence intensity than PKH+ rMSC.

Results

Figure 32 Flow cytometry analysis of PKH26 staining of MSC before injection in rat. Human and rat MSC without (hMSC/rMSC) or with IL1B pre-stimulation (hMSC (IL1B)/rMSC (IL1B)) are depicted. Mean and SD of the differences between median fluorescence intensity of stained and unstained MSC (hMSC: n=8; hMSC (IL1B): n=7; rMSC: n=4; rMSC (IL1B): n=2) are shown. P- value was calculated by one-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05. This figure was already published in Amann et al. (figure 1 A on page 222) [5], CC BY- NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

3.4.2. PKH26-fluorescent cells in peripheral rat blood and broncho-alveolar lavage fluid

To measure recovery of MSC, peripheral blood and BAL was analysed via flow cytometry 24 h and 5 d after TxT. Figure 33 and 35 shows the definition of the flow cytometry gates used for the recovery of MSC.

Results

Figure 33 Definition of flow cytometry gates for the recovery of MSC in peripheral blood from rat. Definition of life gate and PKH26-positive events are shown. FSC-H: forward scatter height; SSC- H: sideward scatter height; FITC: fluorescein isothiocyanate; sham: sham rats; TxT: rats receiving a blunt chest trauma; TxT+hMSC: traumatized rats receiving human MSC; TxT+hMSC (IL1B): traumatized rats receiving IL1B-treated human MSC; TxT+rMSC: traumatized rats receiving rat MSC; TxT+rMSC (IL1B): traumatized rats receiving IL1B-treated rat MSC. This figure was already published in Amann et al. (supplemental figure 6 on page 8) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

Figure 34 shows the analysis of the recovery of the injected MSC in rat blood 24 h (figure 34 A and C) and 5 d (figure 34 B and D) after application of trauma. The percentage of PKH+ events in life gate in blood from sham animals was 0.02 % ± 0.03 % and in blood from TxT rats 0.00 % ± 0.01 % 24 h after TxT. The percentage of PKH+ events in life gate in the blood from TxT+hMSC rats was 64- fold higher than in the blood from sham animals and 362-fold significantly higher (P=0.0270) than in the blood from TxT animals. Blood from TxT+ hMSC (IL1B) rats had a significantly higher percentage of PKH+ events in life gate than blood from control rats (85-fold compared to sham rats, P=0.0246; 479-fold compared to TxT rats, P=0.0081). The percentage of PKH+ events in life gate of TxT+rMSC rats was significantly higher compared to sham (61-fold, P=0.0163) and compared to TxT rats (343-fold, P=0.0052). The percentage of PKH+ events in life gate of TxT+rMSC (IL1B) rats was significantly higher compared to sham rats (93-fold, P=0.0016) and compared to TxT rats (530-fold, P=0.0005).

100

Results

Figure 34 Percentage of PKH26-fluorescent (PKH+) events in life gate in rat peripheral blood after 24 h (A) and 5 d (B). Median fluorescence intensity of events in PKH26 gate of peripheral blood after 24 h (C) and 5 d (D) are shown (n=7-8). Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Significance was calculated with one-way ANOVA followed by Dunn’s multiple comparisons test, § P < 0.05 versus sham rats, §§ P < 0.01 versus sham rats, §§§ P < 0.001 versus sham rats, # P < 0.05 versus TxT rats, ## P < 0.01 versus TxT rats, ### P < 0.001 versus TxT rats. Mean and SD are depicted. These figures were already published in Amann et al. (figure 1 C-F on page 222) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

The percentage of PKH+ events in life gate in blood from sham animals after 5 d was 0.00 % ± 0.00 % and in blood from TxT rats 0.00 % ± 0.01 %. The percentage of PKH+ events in life gate of TxT+hMSC rats was significantly higher (705-fold, P=0.0037) compared to sham animals and significantly higher (353-fold, P=0.0082) compared to TxT animals. The percentage of PKH+ events in life gate of TxT+hMSC (IL1B) rats was significantly higher (838-fold, P=0.0007) compared

101

Results to sham animals and significantly higher (419-fold, P=0.0016) compared to TxT animals. Figure 34 C and D shows the fluorescence intensity of PKH+ events for the experimental groups 24 h and 5 d after application of trauma (animals treated with unstimulated or stimulated MSC). One-way ANOVA test did not show any significant difference between the four groups regarding the fluorescence intensity in the PKH26 gate. The fluorescence intensity of PKH+ events was significantly different between 24 h and 5 d with regard to the TxT+hMSC group and the TxT+hMSC (IL1B) group. After 5 d the mean of the median fluorescence intensity of blood from the TxT+hMSC animals decreased to 81 % (Mann-Whitney test, P=0.0499) of the 24 h values. Fluorescence intensity of blood from the TxT+hMSC (IL1B) decreased to 76 % (Welch’s t-test, P<0.0001) of the 24 h values. The following section was already partly published in Amann et al. on page 223 [5]. Additionally, Mann-Whitney test revealed that there was a decrease in fluorescence intensity when comparing the PKH+ cells before and 24 h after injection (compare figure 32 and 34 C and D). The fluorescence intensity of the PKH+ cells in the blood from TxT+hMSC rats was 16-fold significantly lower (P=0.0002) compared to PKH+ hMSC. PKH+ cells in the blood from TxT+hMSC (IL1B) group had a 26-fold significantly lower (P=0.0003) fluorescence intensity than PKH+ hMSC (IL1B). In the TxT+rMSC group and the TxT+rMSC (IL1B) group the fluorescence intensity of PKH+ events was significantly (P=0.0040) 2.5-fold or non-significantly 3.9-fold lower after recovery compared to the non-injected PKH+ control (rMSC or rMSC (IL1B)). The lowest fluorescence intensity of PKH26- stained hMSC after 24 h of cell culturing was already 4.8-fold higher than the highest fluorescence intensity of PKH+ cells of the blood from TxT+hMSC rats or TxT+rMSC (IL1B) rats (compare figure 17 and 34 C).

102

Results

Figure 35 Definition of flow cytometry gates for the recovery of MSC in broncho-alveolar lavage fluid from rat. Definition of life gate and PKH26-positive events are shown. FSC-H: forward scatter height; SSC-H: sideward scatter height; FITC: fluorescein isothiocyanate; sham: sham rats; TxT: rats receiving a blunt chest trauma; TxT+hMSC: traumatized rats receiving human MSC; TxT+hMSC (IL1B): traumatized rats receiving IL1B-treated human MSC; TxT+rMSC: traumatized rats receiving rat MSC; TxT+rMSC (IL1B): traumatized rats receiving IL1B-treated rat MSC. This figure was already published in Amann et al. (supplemental figure 7 on page 9) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

Figure 36 shows the percentage of PKH+ events in life gate in BAL 24 h (figure 36 A) and 5 d (figure 36 B) after sham or trauma treatment. The percentage of PKH+ cells in BAL from sham animals was 0.04 % ± 0.05 % and from TxT rats 0.00 % ± 0.00 %. There was no significant difference between the control groups (sham, TxT) and the cell treated groups (TxT+hMSC, TxT+hMSC (IL1B), TxT+rMSC), except for the TxT+rMSC (IL1B) group 24 h after treatment. BAL from the TxT+rMSC (IL1B) rats included significantly higher (P=0.0162) counts of PKH+ events than BAL from TxT rats 24 h after application of trauma. The percentage of PKH+ events in BAL from sham rats and TxT rats was 0.00 ± 0.00 % 5 d after sham or trauma treatment. It was significantly higher in hMSC-treated rats than in sham (TxT+hMSC versus sham: P=0.0032; TxT+hMSC (IL1B) versus sham: P=0.0004) and TxT rats (TxT+hMSC versus TxT: 224-fold, P=0.0072; TxT+hMSC (IL1B) versus TxT: 268-fold, P=0.0011). Significant differences between the two points in time could be observed. After 5 d the mean recovery rate of injected cells increased in the TxT+hMSC group (Welch’s t-test, P<0.0001) and the TxT+hMSC (IL1B) group (Mann-Whitney test, P=0.0019) compared to the 24 h point in time.

103

Results

Figure 36 Percentage of PKH26-fluorescent (PKH+) events in life gate in rat broncho-alveolar lavage fluid (BAL) after 24 h (A) and 5 d (B). Median fluorescence intensity of events in PKH26 gate of BAL after 24 h (C) and 5 d (D) are shown (n=8–10). Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre- stimulation (TxT+rMSC/TxT+rMSC (IL1B)). Significance was calculated with one-way ANOVA followed by Dunn multiple comparisons test (A, B) or by Bonferroni’s multiple comparisons test (C), § P < 0.05 versus TxT+hMSC rats, §§ P < 0.01 versus sham rats, §§§ P < 0.001 versus sham rats, # P < 0.05 versus TxT rats ## P < 0.01 versus TxT rats (B) and versus TxT+hMSC (IL1B) (C). Mean and SD are depicted. These figures were already published in Amann et al. (figure 2 on page 223) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

Figure 36 shows the fluorescence intensity of PKH+ events 24 h (figure 36 C) and 5 d (figure 36 D) after sham or trauma treatment. The mean of the median fluorescence intensity of PKH+ events in the BAL from the TxT+rMSC (IL1B) group was significantly higher compared to the TxT+hMSC group (1.9-fold, P=0.0228) and TxT+hMSC (IL1B) group (2.3-fold, P=0.0076) 24 h after TxT.

104

Results

Fluorescence intensity of PKH+ events was significantly higher after 5 d compared to 24 h (TxT+hMSC: 3.5-fold, P=0.0035; TxT+hMSC (IL1B): 3.7-fold, P=0.0142). Welch’s t-test revealed that the mean of the median fluorescence intensity of PKH+ events in the BAL from TxT+hMSC (94 %, P=0.0002), TxT+hMSC (IL1B; 97 %, P=0.0096) and TxT+rMSC (50 %, P=0.0038) rats was significantly lower than the fluorescence intensity of the injected PKH+ MSC 24 h after TxT. The mean of the median fluorescence intensity of PKH+ events in the BAL from TxT+hMSC (80 %, P=0.0005) and TxT+hMSC (IL1B; 90 %, P=0.0132) rats was significantly lower than the fluorescence intensity of the respective injected PKH+ MSC 5 d after TxT. Furthermore, PKH+ events in the BAL had a minimum fluorescence intensity of 235, so that they were still about 2.3-fold higher than the maximum fluorescence intensity of unstained MSC (used as control for flow cytometry analysis of PKH26 staining).

Apart from the fact, that blood and BAL are completely different liquids which included different amounts of PKH+ events due to their origin and extraction, the percentage of PKH+ events in peripheral blood in MSC-treated animals was at least 20-fold higher than in BAL 24 h after TxT. However, the fluorescence intensity of PKH+ events was similar in peripheral blood and BAL. The percentage of PKH+ events in peripheral blood from hMSC-treated rats was 3.2-fold (TxT+hMSC) or 3.1-fold (TxT+hMSC (IL1B)) higher than in BAL 5 d after application of trauma. In contrast, the fluorescence intensity of PKH+ events in BAL was 4.1-fold (TxT+hMSC) or 3.3-fold (TxT+hMSC (IL1B)) higher compared to PKH+ events in peripheral blood from hMSC-treated animals.

3.4.3. Fluorescence microscopy of broncho-alveolar lavage cells

Cytospins of BAL cells were counted manually and the percentage of red fluorescent cells was calculated. Results are depicted in figure 37. BAL cytospins of TxT+hMSC (IL1B) rats had 76 % lower percentage of fluorescent cells than BAL cytospins of sham rats, 71 % lower percentage of fluorescent cells than BAL cytospins of TxT rats and 68 % lower percentage of fluorescent cells than BAL cytospins of TxT+hMSC rats 24 h after sham procedure or TxT. BAL cytospins of TxT+rMSC rats had a 2.1-fold higher percentage of fluorescent cells

105

Results than BAL cytospins of TxT+hMSC rats and a significantly higher (6.5-fold, P=0.0148) percentage of fluorescent cells than BAL cytospins of TxT+hMSC (IL1B) rats 24 h after sham procedure or TxT. BAL cytospins of TxT+rMSC (IL1B) rats had a 2.1-fold higher percentage of fluorescent cells than BAL cytospins of TxT rats, a 2.3-fold higher percentage of fluorescent cells than BAL cytospins of TxT+hMSC rats and a significantly higher (7.3-fold, P=0.0043) percentage of fluorescent cells than BAL cytospins of TxT+hMSC (IL1B) rats 24 h after sham procedure or TxT. 5 d after sham procedure or TxT, BAL cytospins of TxT+hMSC (IL1B) rats had 51 % lower percentage of fluorescent cells than BAL cytospins of sham rats and 52 % lower percentage of fluorescent cells than BAL cytospins of TxT rats. BAL cytospins of TxT+hMSC (IL1B) rats 5 d after sham procedure or TxT had 2.6-fold higher percentage of fluorescent cells than BAL cytospins of TxT+hMSC (IL1B) rats 24 h after sham procedure or TxT.

Figure 37 Red fluorescent cells of broncho-alveolar lavage fluid (BAL) were analysed by fluorescence microscopy 24 h after blunt chest trauma. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre- stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of percentage of red fluorescent cells in rat BAL 24 h (n=8-13) and 5 d (n=8) after blunt chest trauma are shown. Significance for 24 h measurement calculated by one-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01.

106

Results

3.4.4. Description of PKH26-positive events and endogenous mobilization of rat mesenchymal stromal cells

Subsequently, a more detailed analysis of the PKH+ events was performed. Human and rat MSC markers were used to further characterize the PKH+ events. Figure 38 B shows the definition of flow cytometry gates for PKH+ events and for the hMSC markers CD90 and CD105.

Figure 38 (A) Percentage of PKH26-positive events (PKH+) events in rat blood which were positive for the human (h) MSC markers CD90 and CD105 24 h post trauma. Sham rats (n=2), blunt chest trauma rats (TxT, n=2), traumatized rats receiving hMSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B), n=3). Mean and SD are depicted. (B) Flow cytometry gating strategy for detection of PKH26-stained hMSC in rat peripheral blood. FSC-H: forward scatter height; SSC-H: sideward scatter height; FITC: fluorescein isothiocyanate; PE: phycoerythrin; APC: allophycocyanin; Ig: immunoglobulin; CD: cluster of differentiation. These figures were already published in Amann et al. (supplemental figure 9 on page 11) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

The percentage of PKH+, hCD90+ and hCD105+ cells in blood from hMSC and hMSC (IL1B)-treated rats was not higher than in sham rats 24 h (figure 38 A) and 5 d (data not shown, n=1-2) after treatment of rats. Furthermore, almost all PKH+ events were positive for rat CD45, meaning that hMSC characteristics had disappeared from PKH+ signals after 24 h (figure 39).

107

Results

Figure 39 Definition of flow cytometry gate for life gate, PKH26-positive gate, rat CD90 and rat CD45. FSC-H: forward scatter height; SSC-H: sideward scatter height; PE: phycoerythrin; PerCP: peridinin-chlorophyll-protein; APC-Cy7: allophycocyanin coupled with the cyanine dye Cy7; CD: cluster of differentiation; trauma+hMSC: blunt chest trauma rats receiving human MSC.

Figure 40 Definition of flow cytometry gates for PKH26-positive (PKH+) cells, cells which were negative for the rat markers CD3, CD11b/c and CD45 and positive for the rat MSC (rMSC) markers CD29, CD54 and CD90 in rat peripheral blood. FSC-H: forward scatter height; SSC-H: sideward scatter height; FITC: fluorescein isothiocyanate; PE: phycoerythrin; PerCP: peridinin-chlorophyll- protein; APC: allophycocyanin; APC-Cy7: allophycocyanin coupled with the cyanine dye Cy7; Ig: immunoglobulin; CD: cluster of differentiation. This figure was already published in Amann et al. (supplemental figure 10 on page 12) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

108

Results

Figure 40 defines the flow cytometry gates for PKH+, leucocyte marker negative and rMSC marker positive cells. Figure 41 A depicts the percentage of events in the PKH26 gate which were negative for the leucocyte markers CD3, CD11b/c, CD45 and positive for the three rMSC markers CD54, CD90 and CD29. No single event for sham group and TxT group could be detected. Only the TxT+rMSC (mean event number of 22, P<0.0001 compared to sham and TxT rats) and the TxT+rMSC (IL1B) group (mean event number of 13, P=0.0004 compared to sham and TxT rats) contained cells in the PKH26 gate which were positive for the rMSC markers but not for the leucocyte markers.

Figure 41 Percentage of PKH26-positive (PKH+) events (A) or PKH26-negative (PKH-) events (B) in non-leucocyte gate (CD3-, CD11b/c-, CD45-), which were positive for the rat-specific MSC markers CD29, CD54 and CD90. Sham rats (n=4), blunt chest trauma rats (TxT, n=4), traumatized rats receiving rMSC with or without IL1B pre-stimulation (TxT+rMSC, n=8/TxT+rMSC (IL1B), n=7) are shown. Mean and SD are depicted. Significance was calculated with one-way ANOVA followed by Bonferroni’s multiple comparisons test, §§§ P < 0.001 versus sham rats, §§§§ P < 0.0001 versus sham rats, ### P < 0.001 versus TxT rats, #### P < 0.0001 versus TxT rats, * P < 0.05. CD: cluster of differentiation. These figures were already published in Amann et al. (figure 3 B on page 224 and supplement figure 12 on page 14) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by- nc-nd/4.0/legalcode.

To detect an endogenous mobilization of rMSC, the definition of flow cytometry gates was similar to figure 40, however the focus had shifted to the PKH- cells. The non-leucocyte gate as well as the gate for the rMSC markers were the same as mentioned before. Figure 41 B shows the percentage of the endogenous rMSC after 24 h in the blood. Taking all three rMSC markers into account, only the TxT and TxT+rMSC 109

Results

(IL1B) group contained cells which fulfilled the criteria. The TxT group (0.125 % ± 0.096 %, corresponding to a mean event number of 6) contained a higher (1.8-fold) amount of these cells compared to the TxT+rMSC (IL1B) group (0.071 % ± 0.049 %, corresponding to a mean event number of 4).

3.4.5. Analysis of Alu sequence in rat organs

The presence of human cells was tested in blood smear, lung, liver, kidney, brain, gonad, and spleen tissue. Only in the lung of the 24 h animals 1.05 pg to 3.75 pg human DNA could be detected in five of sixteen hMSC-treated rats (two TxT+hMSC rats, three TxT+hMSC (IL1B) rats), which is less than the DNA content of one single cell [86]. Sham and TxT rats did not contain human DNA in all analysed tissues meaning that cycle threshold values were beyond the detection limit.

3.4.6. Incubation of mesenchymal stromal cells with rat blood for 24 hours

To investigate the reason for the low recovery of hMSC after administration to traumatized rats, in vitro experiments were performed. Human MSC were incubated in vitro with rat pooled blood for 24 h. Afterwards flow cytometry analysis and FACS were performed. Sorted (gating strategy see figure 42) and unsorted cells were cultivated for one month and flow cytometry analysis was performed again. Incubation of rMSC with rat blood was investigated as well. Cell culture of 270270 PKH26-stained hMSC incubated with 1 ml blood from sham rats revealed two colonies, but they could not be trypsinized and analysed. In the second experiment incubating hMSC with rat blood for 24 h, 70.00 % of the PKH+- sorted cells, cultured for one month on plastic surface, expressed the marker combination hCD45-, hCD73+ and hCD105+. Furthermore, 95.10 % of these PKH+ cells expressed the marker combination hCD3-, hCD90+ and hCD105+. These percentages were similar to those of hMSC of a parallel culture from the same donor, which were stained at the same time as the spiked hMSC and seeded again at the same time as PKH+ fraction of spiked hMSC. Percentages of above mentioned markers on these hMSC were 72.20 % or 97.50 %. PKH+-sorted cells cultured on irradiated C2C12 feeder cells did not express hMSC surface markers like hCD73 and hCD105 and only 0.30 % of these cells presented hCD90. 110

Results

PKH- fraction of spiked hMSC contained only 0.10 % cells which were hCD45-, hCD73+ and CD105+ after one month. Zero percent of PKH- fraction of spiked hMSC presented hCD3-, hCD90+ and CD105+. Incubation of rMSC with rat blood and reseeding of sorted and unsorted blood did not lead to cell growth of rMSC (see table 20 and 22). After one month of cell culture, 0.10 % of unsorted cells cultured on plastic surface and of PKH+ fraction of spiked rMSC cultured on feeder layer had the following marker combination: CD3-, CD11b/c-, CD45-, CD29+, CD54+ and CD90+. 0.20 % of PKH--sorted cells cultured on feeder cells were CD3-, CD11b/c-, CD45-, CD29+, CD54+ and CD90+.

3.4.7. Morphology of mesenchymal stromal cells before and after injection

Peripheral blood from rats was sorted via FACSAria (gating strategy see figure 42). The PKH+ events were partly applied on slides via the cytospin method and stained with the Pappenheim staining.

Figure 42 Gating strategy for sorting PKH26-positive cells in rat peripheral blood. Flow cytometry gate distinguished between PKH26-positive and -negative cell fraction. This figure was already published in Amann et al. (supplemental figure 8 on page 10) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode. 111

Results

The comparison of morphology of MSC before and 24 h after injection (figure 43 and 44) as well as the comparison of the sideward scatter height (SSC-H) and the forward scatter height (FSC-H) gate (figure 31, 33 and 35) showed that there is a difference in size and morphology between MSC and PKH+ events of rat peripheral blood. Figure 43 shows that MSC differ in size. Nevertheless, PKH+ events were about 2.0-fold smaller than MSC on cytospin slides. Cytoplasma was reduced in PKH+ events compared to MSC and nucleus in PKH+ events was more circular than in MSC. The 24 h-incubation of MSC with rat blood did not have the same effect.

Figure 43 Comparison of morphology of rat MSC (rMSC) and human MSC (hMSC) by means of Cytospin and Pappenheim staining. (Panel 1 + 2) MSC before (unstained) and (panel 3 + 4) after PKH26 staining (stained), (panel 5 + 6) PKH+ events after ex vivo incubation of MSC with rat peripheral blood of sham animals for 24 h and isolation by fluorescence activated cell sorting (FACS) (ex vivo) and (panel 7 + 8) 24 h after injection of MSC in traumatized rats and isolation by FACS (in vivo). This figure was already published in Amann et al. (figure 3 A on page 224) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

112

Results

Figure 44 Pictures of confocal microscopy of PKH26-positive (PKH+) cells. Cultured human MSC (hMSC), PKH+ cells isolated from peripheral blood (PB) of blunt chest trauma rats receiving human or rat MSC (TxT+hMSC/TxT+rMSC) are depicted 24 h after PKH26 staining or trauma. Different magnifications are shown.

Furthermore, confocal microscopy showed that PKH26 fluorescence dye was intracellular 24 h after PKH26 staining (figure 44).

3.4.8. Culturing of rat blood and broncho-alveolar lavage cells

To get more information about the injected MSC and MSC mobilization after trauma, rat blood and BAL cells were cultured. Rat peripheral blood or BAL was sorted by FACS. PKH+ cells, PKH- cells, as well as unsorted cells were cultured with or without feeder cells (C2C12) for one month. After expansion of the cells, they were harvested and analysed regarding MSC marker expression. Table 20 to 23 show the results of rat cell expansion.

Cell culture of rat blood 24 h after blunt chest trauma or sham procedure Culture of PKH+, PKH- or unsorted cells derived from rat blood failed (see table 20). However, rat blood did not contain plastic adherent cells. Figure 45 depicts exemplarily the spherical PKH+ events of blood from TxT+hMSC rats. Unsorted cells were difficult to cultivate because of partial coagulation, although the medium contained heparin. PKH+ fraction of TxT+hMSC rats on feeder layer did not grow, however, could be analysed after one month. 93.00 % of these cells were positive for the rat leucocyte markers CD3, CD11b/c, CD45 and zero percent of these cells were in the non-leucocyte gate and concurrently positive for the rat MSC marker CD29, CD54 and CD90. Furthermore, these cells did not have the combination of surface markers hCD45-, hCD105+ and hCD73+, as well as hCD3-, hCD90+ and hCD105+. The PKH- fraction of TxT+hMSC rats on the feeder layer had the same 113

Results characteristics. 93.20 % of these cells were positive for the rat leucocyte markers CD3, CD11b/c, CD45. Furthermore, zero percent of C2C12 (n=2) and x-rayed C2C12 (n=3) were hCD45-, hCD105+ and hCD73+, as well as hCD3-, hCD90+ and hCD105+. Alu sequence analysis of this PKH- fraction confirmed the results. Human DNA could not be determined. PKH+ and PKH- fraction of pooled blood from TxT+rMSC (n=2) and TxT+rMSC (IL1B) (n=1) rats did not present the marker combination of CD3-, CD11b/c-, CD45-, CD29+, CD54+ and CD90+, after one month of cell culture on feeder layer. Furthermore, these cells could not be expanded.

Table 20 Rat blood from 24 h animals was cultured for one month. The six rat groups are listed: sham rats, blunt chest trauma rats (TxT) and traumatized rats receiving human MSC (hMSC) with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)), as well as traumatized rats receiving rat MSC (rMSC) with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)). Blood was partly pooled from up to five rats (1-5) and sorted to isolate the PKH26-fluorescent (PKH+) fraction and the non-fluorescent fraction (PKH-). The cells were seeded on plastic surface or on feeder layer (FL). Table rows represent the replicates. Blood from four TxT rats were pooled to incubate with hMSC or rMSC for 24 h. Red line means that the cells did not grow; green cross means that the cells grew.

114

Results

Figure 45 Cell culture of PKH26-positive (PKH+) events isolated from rat peripheral blood (PB) 24 h after blunt chest trauma and human MSC (hMSC) application (TxT+hMSC) were compared to cell culture of unstained hMSC. Human MSC were stored at 4 °C for 24 h before seeding. Cell culture was performed for 3 d in both cases. Human MSC were plastic adherent and had fibroblastoid morphology. PKH+ events stayed spherical. Three different magnifications are shown.

Cell culture of rat broncho-alveolar lavage fluid 24 h after blunt chest trauma or sham procedure Unsorted BAL cells adhered to plastic surface and grew on it (see table 21). Unsorted BAL cells from TxT+hMSC rats were cultured on feeder layer. Alu sequence analysis revealed that they did not contain human DNA. These cells did not present the hMSC markers (hCD73+, hCD90+ and CD105+) as well. The expression profile of cell surface markers of the other cell cultures is depicted in figure 46. Percentage of cells from TxT (15.47 %) and TxT+rMSC rats (18.37 %) which were negative for leucocyte markers (CD3-, CD11b/c-, CD45-) and positive for the MSC marker combination CD54+, CD90+ and CD29+ were similar and distinguished themselves significantly from feeder cells (C2C12 versus TxT+rMSC: P=0.0422; C2C12 x-rayed versus TxT+rMSC: P=0.0313). Percentage of BAL cells from TxT rats which were negative for leucocyte markers and positive for CD54, CD90 and CD29 was 2.4-fold higher than the percentage of PKH+ cells of BAL from TxT rats. BAL from TxT+rMSC and TxT+rMSC (IL1B) rats contained more PKH+ cells (TxT+rMSC: 25.90 %; TxT+rMSC (IL1B): 33.90 %) than cells which were negative for leucocyte markers and positive for CD54, CD90 and CD29 (TxT+rMSC: 18.37 %; TxT+rMSC (IL1B): 15.50 %).

115

Results

Table 21 Rat broncho-alveolar lavage fluid (BAL) from 24 h animals was cultured for one month. The six rat groups are listed: sham rats, blunt chest trauma rats (TxT) and traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)), as well as traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)). BAL was partly pooled from up to five rats (1-5) and sorted to isolate the PKH26- fluorescent (PKH+) fraction and the non-fluorescent fraction (PKH-). The cells were seeded on plastic surface or on feeder layer (FL). Table rows represent the replicates. Red line means that the cells did not grow; green cross means that the cells grew.

Figure 46 Flow cytometry analysis of cell culture of broncho-alveolar lavage fluid (BAL) cells after one month. Percentage of PKH26-positive (PKH+) cells (A). Percentage of cells which were in non- leucocyte gate (CD3-, CD11b/c-, CD45-) and positive for rMSC marker (CD29+, CD54+, CD90+) are shown (B). C2C12 cells (n=4), x-rayed C2C12 cells (n=5), BAL cells of traumatized rats (TxT) (n=3), BAL cells of human MSC and rat MSC-treated animals with or without IL1B pre-stimulation (TxT+hMSC (n=1)/TxT+rMSC (n=3)/TxT+rMSC (IL1B, n=1)) are shown. BAL cells of TxT+hMSC rats grew on x-rayed C2C12 cells. Mean and SD are depicted. Significance was calculated with one-way ANOVA with Bonferroni’s multiple comparisons test, ** P < 0.01, § P < 0.05 versus C2C12 cells, §§§§ P < 0.0001 versus C2C12 cells, # P < 0.05 versus x-rayed C2C12 cells, #### P < 0.0001 versus x-rayed C2C12 cells.

116

Results

Cell culture of rat blood and broncho-alveolar lavage fluid after 5 d Table 22 and 23 summarize the cell culture experiments performed with blood and BAL from rats 5 d after induction of trauma. Blood from sham and TxT rats was used for in vitro experiments (section 2.8.7. and 3.4.6.). Single experiments showed that expansion of blood and BAL cells was not possible.

Table 22 Rat blood from 5 d animals was cultured for one month. Four rat groups are listed: sham rats, blunt chest trauma rats (TxT) and traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)). Blood was pooled from two rats (2) and sorted to isolate the PKH26-positive fraction. The cells were seeded on plastic surface. Table rows represent the replicates. Blood from up to four TxT rats were pooled to incubate with hMSC or rat MSC (rMSC) for 24 h. One half of the cells were grown on feeder layer and the other half on plastic surface (*). Red line means that the cells did not grow; green cross means that the cells grow.

Table 23 Rat broncho-alveolar lavage fluid (BAL) from 5 d animals was cultured for one month. Four rat groups are listed: sham rats, blunt chest trauma rats (TxT) and traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)). BAL was partly pooled from up to three (1, 3). The cells were seeded on plastic surface. Table rows represent the replicates. Red line means that the cells did not grow.

3.4.9. Measurements of signal molecules in plasma and broncho-alveolar lavage fluid

Systemic and local reaction in Wistar rats on TxT and treatment with MSC was analysed by Multiplex assay and ELISA. Cytokine, chemokine and growth factor levels in plasma and BAL are depicted in figure 47 to 50 and table 24.

117

Results

Figure 47 and 48 show the systemic reaction in rats on different treatments after 24 h and 5 d. Interestingly, there was no significant difference in plasma concentration of the analysed (rat-specific) factors between sham and TxT animals neither at 24 h nor at 5 d post trauma. Nonetheless, PGE2 level in plasma of TxT rats was 52 % lower than of sham rats 24 h after induction of TxT. TxT+hMSC and TxT+hMSC (IL1B) rats had higher rat cytokine levels (IL6, TNF) compared to sham and TxT rats, 24 h post trauma. IL6 level in plasma from TxT+hMSC rats increased 2.2-fold compared to sham rats and 2.1-fold compared to TxT rats. Plasma level of TNF increased 2.0-fold in TxT+hMSC rats and 2.7-fold in TxT+hMSC (IL1B) rats compared to sham rats. Furthermore, TNF concentration increased 2.1-fold in plasma from TxT+hMSC (IL1B) rats compared to TxT rats. IL10 level in blood was 2.2-fold higher in TxT+hMSC (IL1B) rats compared to sham rats and 2.1-fold higher compared to TxT rats. After 5 d, the TxT+hMSC (IL1B) rats had still a 2.2-fold higher TNF plasma level compared to TxT rats and a 2.1-fold higher IL10 plasma level compared to sham rats. TxT+rMSC rats had a 54 % lower TNF plasma level compared to the TxT rats and a 53 % lower PGE2 plasma level compared to sham rats, 24 h after trauma. Plasma concentration of CXCL1 from TxT+rMSC (IL1B) rats was 59 % lower than from sham rats and 54 % lower than from TxT rats 24 h post trauma. In contrast, IL10 level was increased 2.1-fold in plasma from TxT+rMSC (IL1B) compared to sham rats and was increased 2.0-fold compared to TxT rats, 24 h post trauma. TxT+rMSC and TxT+rMSC (IL1B) rats had a lower IL1B, IL6, TNF and CXCL1 plasma level than TxT+hMSC and TxT+hMSC (IL1B) rats 24 h after TxT. TxT+rMSC rats had a 54 % lower plasma level of IL1B than TxT+hMSC rats and a 51 % lower plasma level of IL1B than TxT+hMSC (IL1B) rats. TxT+rMSC rats had a 69 % lower IL6 plasma level than TxT+hMSC rats and TxT+rMSC (IL1B) rats had a 65 % lower IL6 level than TxT+hMSC (IL1B) rats. TxT+rMSC rats had a 71 % lower TNF plasma level than TxT+hMSC rats and TxT+rMSC (IL1B) rats had a 72 % lower TNF level than TxT+hMSC (IL1B) rats. TxT+rMSC rats had a 59 % lower CXCL1 plasma level than TxT+hMSC rats and TxT+rMSC (IL1B) rats had a 73 % lower CXCL1 level than TxT+hMSC (IL1B) rats.

118

Results

Figure 47 Concentration of cytokines (IL1B, IL6, TNF, IL10) and chemokines (CXCL1, CCL2) in plasma from rats 24 h and 5 d after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre- stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=7-10 are shown. IL1B: interleukin 1 beta; IL6: interleukin 6; CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; IL10: interleukin 10. This figure was already published in Amann et al. (supplemental figure 11 on page 13) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

119

Results

Furthermore, TxT+rMSC (IL1B) rats had a 2.2-fold higher plasma level of IL1B compared to TxT+rMSC rats, whereas pre-stimulation of hMSC with human IL1B did not change plasma level of rat IL1B from TxT+hMSC (IL1B) rats compared to TxT+hMSC rats, 24 h after trauma. Concentration of the measured molecules in the plasma did not differ significantly at the time of 24 h and 5 d in all groups (Mann-Whitney test), except for FGF2. However, plasma concentration of IL1B (2.1-fold) IL6 (2.1-fold) and IL10 (2.3-fold) increased in TxT rats after 5 d compared to 24 h. Plasma concentration of FGF2 was significantly higher (2.1-fold, P=0.0205) in TxT+hMSC rats after 5 d compared to 24 h.

Figure 48 Concentration of TNF, growth factors (VEGFA, FGF2) and PGE2 in plasma from rats 24 h and 5 d after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=7-10 are shown. Significance between FGF2 level after 24 h and 5 d was calculated with Mann-Whitney test, § P < 0.05. TNF: tumor necrosis factor (alpha); PGE2: prostaglandin E2; VEGFA: vascular endothelial growth factor-A; FGF2: fibroblast growth factor 2.

120

Results

Figure 49 Concentration of cytokines (IL1B, IL6, IL10), chemokines (CXCL1, CCL2), and inflammatory molecules (TNFAIP6, IL1RN) in broncho-alveolar lavage fluid (BAL) from rats 24 h after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=8-13 are shown. Significance was calculated with one-way ANOVA followed by Dunn’s multiple comparisons test, except for CXCL1, which was calculated with one-way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. IL1B: interleukin 1 beta; IL6: interleukin 6; CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; IL10: interleukin 10; TNFAIP6: tumor necrosis factor alpha inducible protein 6; IL1RN: interleukin 1 receptor antagonist. This figure was already published in Amann et al. (figure 5 C in on page 226) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

121

Results

Table 24 Concentration of cytokines (IL1B, IL6, IL10), chemokines (CXCL1, CCL2), and inflammatory molecules (TNFAIP6, IL1RN) in broncho-alveolar lavage fluid (BAL) from rats 5 d after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=8 are shown. IL1B: interleukin 1 beta; IL6: interleukin 6; CXCL1: C-X-C motif chemokine ligand 1; CCL2: C-C motif chemokine ligand 2; IL10: interleukin 10; TNFAIP6: tumor necrosis factor alpha inducible protein 6; IL1RN: interleukin 1 receptor antagonist.

TxT+hMSC BAL 5 d sham TxT TxT+hMSC (IL1B) IL1B 60.2 ± 29.3 61.9 ± 50.9 32.9 ± 20.8 27.7 ± 10.7 IL6 1213.6 ± 1070.0 1484.0 ± 1227.6 427.3 ± 200.5 453.8 ± 266.2 IL10 40.7 ± 25.7 44.0 ± 29.7 22.2 ± 19.8 15.8 ± 10.6 CXCL1 340.3 ± 196.6 363.1 ± 239.1 156.5 ± 57.9 156.6 ± 23.7 CCL2 287.8 ± 217.8 333.5 ± 181.4 163.9 ± 178.8 172.2 ± 227.9 TNFAIP6 213.9 ± 224.0 198.5 ± 225.7 7.6 ± 21.5 9.9 ± 17.9 IL1RN 19.5 ± 8.5 22.2 ± 8.6 19.4 ± 10.9 25.1 ± 11.4

The local reaction showed more discrepancies between the six animal groups 24 h after trauma (see table 24, figure 49 and 50). There was an increase of the pro- inflammatory rat cytokines (IL1B: 2.1-fold, TNF: 3.6-fold), chemokines (CXCL1: 4.2-fold and CCL2: 8.1-fold) and the anti-inflammatory IL1RN (34-fold) in BAL from TxT rats compared to sham rats at 24 h time. However, there was no significant difference between the BAL cytokine level of sham and TxT rats, except for the CCL2 (P=0.0225) level and the IL1RN (P=0.0026) level. IL17A could only be detected in BAL from two of eight TxT rats (2.65 pg/ml, 2.03 pg/ml) and in one of thirteen TxT+hMSC rats (2.34 pg/ml) . In all cases concentration was near the detection limit of rat IL17A quantification (2.00 pg/ml). In general, the treatment with hMSC and hMSC (IL1B) reduced the pro- inflammatory cytokine level (IL1B, IL6, TNF), the chemokine level (CXCL1, CCL2) and the IL1RN level, but increased the TNFAIP6 level in the BAL compared to TxT rats 24 h after TxT. The TxT+hMSC or TxT+hMSC (IL1B) group had a 51 % or a 55 % lower IL1B level compared to the TxT group. Both groups also had a 54 % (TxT+hMSC) or a 66 % (TxT+hMSC (IL1B)) lower IL6 level compared to the TxT rats. The TxT+hMSC rats had even a 55 % lower IL6 level than the sham rats. The TxT+hMSC and TxT+hMSC (IL1B) rats further had a 72 % or 62 % lower TNF level compared to the TxT rats. The chemokine level in TxT+hMSC and TxT+hMSC (IL1B) rats was higher (CXCL1: 2.9-fold or 3.7-fold; CCL2: 3.8-fold or 2.2-fold) than in sham rats but lower than in TxT rats (CCL2: 53 % or 73 %). The TxT+hMSC and TxT+hMSC (IL1B) rats had a 8.8-fold (TxT+hMSC: P=0.0143) or 122

Results

6.6-fold higher IL1RN level compared to the sham rats, but a 74 % (TxT+hMSC) or a 80 % (TxT+hMSC (IL1B)) lower IL1RN level compared to the TxT rats. In contrast, the TxT+hMSC animals had a 2.6-fold higher TNFAIP6 level compared to the sham animals and a 2.3-fold higher compared to the TxT rats.

Figure 50 Concentration of TNF, growth factors (VEGFA, FGF2) and PGE2 in broncho-alveolar lavage fluid (BAL) from rats 24 h and 5 d after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre- stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=8-13 are shown. Significance for VEGFA (5 d) and PGE2 (5 d) was calculated with one-way ANOVA followed by Bonferroni’s multiple comparisons test. Significance for the remaining cytokine levels was calculated with one-way ANOVA followed by Dunn’s multiple comparisons test, * P < 0.05, ** P < 0.01. Significance between the two points in time (24 h and 5 d) was calculated with Welch’s t-test or Mann-Whitney test, § P < 0.05, §§ P < 0.01. TNF: tumor necrosis factor (alpha); PGE2: prostaglandin E2; VEGFA: vascular endothelial growth factor-A; FGF2: fibroblast growth factor 2.

In contrast to the hMSC treatment, the treatment with rMSC and rMSC (IL1B) did not decrease the pro-inflammatory cytokine levels (IL1B, IL6), the level of the anti- inflammatory cytokine IL10 and the chemokine CXCL1 in the BAL compared to the 123

Results

TxT rats 24 h after induction of TxT. The TxT+rMSC rats rather had a significant increase (7.6-fold, P=0.0013) of CXCL1 level in BAL compared to the sham rats. The CCL2 level of BAL was significantly higher (6.8-fold, P=0.0110) in TxT+rMSC rats compared to sham rats. The TxT+rMSC rats had a significant increase (16- fold, P=0.0067) of IL1RN compared to the sham rats. The TxT+rMSC or TxT+rMSC (IL1B) rats had a 53 % or 80 % lower IL1RN level compared to the TxT rats but a 4.8-fold or 3.4-fold (TxT+rMSC: P=0.0439) higher TNFAIP6 level compared to TxT rats. TxT+rMSC rats also had a significant 5.4-fold (P=0.0199) higher TNFAIP6 level than the sham rats. Furthermore, the TxT+rMSC had a 55 % lower PGE2 level compared to the sham rats. Cytokine and chemokine levels of rMSC/rMSC (IL1B)-treated rats were higher compared to hMSC/hMSC (IL1B)-treated rats. IL1B level in TxT+rMSC rats was 2.3-fold higher compared to TxT+hMSC rats. The TxT+rMSC (IL1B) rats had a significantly higher (2.2-fold, P=0.0423) IL1B level compared to TxT+hMSC (IL1B) rats. Regarding the IL6 level, the treatment with rMSC or rMSC (IL1B) differed in a similar way from the hMSC treatment (2.6-fold) and significantly differed from the hMSC (IL1B) treatment (TxT+rMSC (IL1B)/TxT+hMSC (IL1B): 3.8-fold, P=0.0086). The TxT+rMSC rats had a 2.2-fold higher TNF level compared to sham and hMSC-treated rats. Furthermore, the TxT+rMSC group had a significantly higher (2.6-fold, P=0.0224) CXCL1 level compared to TxT+hMSC rats. The TxT+rMSC (IL1B) rats had a 2.0-fold higher CCL2 level compared to TxT+hMSC (IL1B) rats. The TxT+rMSC rats had a significantly higher (1.9-fold, P=0.0223) BAL concentration of IL10 compared to TxT+hMSC rats. The TxT+rMSC or TxT+rMSC (IL1B) rats had a 2.1-fold or 2.8-fold higher TNFAIP6 level compared to TxT+hMSC or TxT+hMSC (IL1B) rats. In contrast, treatment with rMSC reduced the PGE2 level compared to the remaining groups. Furthermore, in general significant differences between hMSC/rMSC with or without IL1B pre-stimulation could not be detected in BAL and plasma cytokine levels.

Five days after induction of trauma or sham procedure, sham and TxT rats had similar cytokine levels. TxT+hMSC and TxT+hMSC (IL1B) rats had lower cytokine levels in BAL compared to control rats (sham and TxT rats). TxT+hMSC rats had a 65 % or 71 % lower IL6 level and a 54 % or 57 % lower CXCL1 level compared to control rats. They also had a 50 % lower CCL2 and IL10 level than TxT rats and a

124

Results significantly lower (54 %, P=0.0212) VEGFA level than sham rats. TNFAIP6 level was reduced to 4 % of TNFAIP6 level in BAL from control rats. However, the PGE2 level was significantly higher (1.9-fold, P=0.0423) in hMSC-treated animals compared to sham rats. TxT+hMSC (IL1B) rats had a 54 % or 55 % lower IL1B level, 63 % or 69 % lower IL6 level, 54 % or 57 % CXCL1 level, 61 % (P=0.0067) or 53 % lower VEGFA level and a 61 % or 64 % lower IL10 level than control rats. TNFAIP6 level was reduced to 5 % of TNFAIP6 level in BAL from control rats. Treatment of TxT with rMSC was not observed after 5 d.

Concentration of the measured molecules in BAL differed significantly between the two points in time (24 h and 5 d) within each group (TxT, TxT+hMSC and TxT+hMSC (IL1B); Mann-Whitney test). Sham rats had the least variations regarding the two points in time. The concentration of IL1B decreased about 50 % in TxT, TxT+hMSC (Welch’s t- test, P=0.0440) and TxT+hMSC (IL1B) rats after 5 d compared to 24 h. The TxT rats had a significantly lower (70 %, P=0.0354) TNF level after 5 d compared to 24 h. The level of CXCL1 decreased after 5 d compared to 24 h (TxT: 75 %, P=0.0041 (Welch’s t-test); TxT+hMSC: 85 %, P=0.0023 (Welch’s t-test); TxT+hMSC (IL1B): 88 %). The CCL2 level also decreased after 5 d compared to 24 h (TxT: 85 %; TxT+hMSC: 85 %, P=0.0152 (Mann-Whitney test); TxT+hMSC (IL1B): 72 %). In TxT+hMSC (P=0.0025) and TxT+hMSC (IL1B; P=0.0132) rats the VEGFA level decreased about 50 % at 5 d compared to 24 h. The TxT+hMSC (IL1B) had a significantly lower (63 %, Welch’s t-test: P=0.0241) IL10 level after 5 d compared to 24 h. The TNFAIP6 level was 98 % lower (Mann-Whitney test, P=0.0045) in TxT+hMSC rats and 95 % lower in TxT+hMSC (IL1B) rats at 5 d compared to the 24 h measurement. IL1RN level was also 98 % lower (Mann- Whitney test, P=0.0003) in TxT rats, 92 % lower (Mann-Whitney test, P<0.0001) in TxT+hMSC rats and 86 % lower (Mann-Whitney test, P=0.0006) in TxT+hMSC (IL1B) rats after 5 d compared to the 24 h value. PGE2 level was significantly higher (1.7-fold, P=0.0099) in TxT+hMSC rats after 5 d compared to 24 h.

125

Results

3.4.10. Complement component 5a in rat plasma

Plasma concentration of C5a (figure 51) was measured in cooperation with the Department of Orthopedic Trauma, Hand, Plastic and Reconstructive Surgery, Ulm University Medical Center, Ulm. Plasma concentration of C5a from sham and TxT rats did not vary significantly 24 h and 5 d after induction of trauma. However, TxT+hMSC (IL1B; 1.4-fold, P=0.0204) and TxT+rMSC (IL1B; 1.5-fold, P=0.0019) rats had a significantly higher C5a plasma level than TxT rats 24 h after trauma. TxT+rMSC (IL1B; 1.5- fold, P=0.0065) rats had even a significantly higher C5a plasma level than sham rats 24 h after trauma. Plasma level of C5a from TxT+hMSC (43 % reduction, P=0.0011) and from TxT+hMSC (IL1B; 40 % reduction, P=0.0007) rats decreased significantly 5 d after trauma compared to 24 h measurement.

Figure 51 Concentration of complement component 5a (C5a) in plasma from rats 24 h and 5 d after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=7-8 are shown. Significance between the different groups was calculated with one- way ANOVA followed by Bonferroni’s multiple comparisons test, * P < 0.05, ** P < 0.01. Significance between the two points in time (24 h and 5 d) was calculated with Welch’s t-test, §§ P < 0.01, §§§ P < 0.001.

126

Results

3.4.11. Total protein content in broncho-alveolar lavage fluid

Figure 52 depicts total protein in rat BAL.

Figure 52 Total amount of protein in broncho-alveolar lavage fluid from rats 24 h and 5 d after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=8-13 are shown. Significance between the two points in time (24 h and 5 d) was calculated with Mann-Whitney test, § P < 0.05.

The mean of total protein in the BAL was slightly increased in all groups receiving a TxT compared to sham rats 24 h after trauma or sham procedure. There was a significant reduction of 22 % (P=0.0260) of total protein in BAL from TxT rats 5 d after trauma compared to 24 h after trauma.

127

Results

3.4.12. Cell count in broncho-alveolar lavage fluid

Cell count of BAL was counted after 24 h and 5 d (figure 53).

Figure 53 Cell counts of broncho-alveolar lavage fluid (BAL) per ml 24 h (A) and 5 d (B) after treatment of rats. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=8-13 are shown. Significance was calculated with one-way ANOVA followed by Dunn’s multiple comparisons test, * P < 0.05. This figure was already published in Amann et al. (figure 5 A and B on page 226) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by- nc-nd/4.0/legalcode.

The range after 24 h was between 35000 and 2215000 cells/ml. One-way ANOVA test followed by Dunn’s multiple comparisons test did not reveal significant differences after 24 h comparing the six groups. After TxT the BAL cell count increased 4.2-fold compared to sham rats. MSC treatment reduced cell count to a similar level as observed in sham rats, except for TxT+rMSC rats. The TxT+hMSC rats had a 68 % lower BAL cell count number than TxT rats. The TxT+hMSC (IL1B) and the TxT+rMSC (IL1B) rats had a 75 % lower BAL cell count number than TxT rats. The BAL cell count of TxT+rMSC rats was 2.8-fold higher than of sham rats. The TxT+rMSC rats had a 2.1-fold higher BAL cell count number than TxT+hMSC rats. The TxT+rMSC (IL1B) rats had a 63 % lower BAL cell count number than TxT+rMSC rats. While the cell count of BAL was increased in TxT rats compared to sham rats 24 h after TxT, it was significantly decreased (P=0.0306) by 51 % 5 d after TxT. The BAL cell count differed significantly in TxT rats between these two points in time (one-way ANOVA followed by Dunn´s test, P=0.0327 or Mann-Whitney test, 128

Results

P=0.0207). It was 6.4-fold higher in TxT rats after 24 h compared to the 5 d point in time. BAL cell count in TxT+hMSC rats at 24 h was 1.7-fold higher (Welch’s t- test, P=0.0057) compared to BAL cell count at 5 d.

3.4.13. Differential analysis of cells in broncho-alveolar lavage fluid

Inflammatory cells play an important role in the TxT model. Pappenheim staining of BAL cytospins after 24 h and 5 d are depicted in figure 54 and 55 A. Histological analysis of BAL cytospins are shown in table 25 and figure 55 B.

Figure 54 Representative images of Pappenheim staining of cytospins with broncho-alveolar lavage (BAL) cells. BAL cells of sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and of traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) 24 h after treatment are depicted.

129

Results

Table 25 Histological analyses of broncho-alveolar lavage (BAL) cytospins. Percentage of alveolar macrophages, neutrophils and lymphocytes after 24 h of sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre- stimulation (TxT+rMSC/TxT+rMSC (IL1B); n=8-13). Two-way ANOVA with Bonferroni’s multiple comparisons test was used to calculate significance. P-values for analysis against sham and TxT rats. Mean and SD are depicted. Basophils, eosinophils and monocytes were included in statistical analysis; data not shown because of small number. This table was already published in Amann et al. (supplemental table 1 on page 15) ) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

Basophils, eosinophils and monocytes were only detected every so often in a few samples. Percentage of these cells did not differ between the six groups (not shown in table 25 and figure 55 B). The proportion of lymphocytes in TxT+rMSC (IL1B) rats was 52 % lower compared to sham rats and 53 % lower compared to TxT rats. Furthermore, the TxT+rMSC (IL1B) rats had a 50 % lower lymphocyte proportion than TxT+rMSC rats. Neutrophils were not detected in BAL cytospins of sham rats. There was a significant increase (P<0.0001) in neutrophils in traumatized animals with or without treatment, except for rMSC (IL1B) treated animals 24 h post trauma. The proportion of neutrophils in TxT+rMSC (IL1B) rats was significantly lower (62 %, P=0.0009) compared to TxT rats. Furthermore, the TxT+rMSC (IL1B) rats had a significantly lower (54 %, P=0.0272) neutrophil proportion compared to TxT+hMSC and a 50 % lower neutrophil proportion compared to TxT+hMSC (IL1B) rats. Coherently, percentage of alveolar macrophages differed significantly between sham rats and traumatized animals with or without treatment, except for TxT+rMSC (IL1B) animals. The TxT (35 %, P<0.0001), TxT+hMSC (19 %, P=0.0003), TxT+hMSC (IL1B, 17 %, P=0.0073) and TxT+rMSC (22 %, P<0.0001) rats had a significantly lower macrophage proportion compared to sham rats. The TxT+hMSC (1.2-fold, P=0.0072), TxT+hMSC (IL1B, 1.3-fold, P=0.0065), TxT+rMSC (IL1B, 1.5-fold, P<0.0001) rats had a significantly higher macrophage proportion compared to TxT rats.

130

Results

Furthermore, the TxT+rMSC (IL1B) rats had a significantly higher (1.2-fold, P=0.0035) macrophage proportion compared to TxT+rMSC rats.

Figure 55 (A) Representative images of Pappenheim staining of cytospins with broncho-alveolar lavage (BAL) cells 5 d after treatment. (B) Percentage of lymphocytes, neutrophils and alveolar macrophages on BAL cytospins 5 d after treatment are depicted. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) are depicted. Mean and SD of n=8 are shown.

After 5 d no differences between the treatment groups could be observed (figure 55). Mann-Whitney test (TxT+hMSC (IL1B) neutrophils) and Welch’s t-test (remaining groups) revealed that the percentage of neutrophils and alveolar macrophages differed significantly between the 24 h and 5 d point in time in TxT (neutrophils: P=0.0056; alveolar macrophages: P=0.0007), TxT+hMSC (neutrophils: P=0.0001; alveolar macrophages: P=0.0001) and TxT+hMSC (IL1B) rats (neutrophils: P=0.0002; alveolar macrophages: P=0.0007).

3.4.14. Myeloperoxidase content in the lung

Measurement of MPO activity was performed in cooperation with the Institute for Clinical and Experimental Trauma-Immunology of Ulm University. The MPO activity in lung tissue correlates with the infiltration of neutrophils [154,197]. MPO

131

Results content in cryopreserved lung tissue 24 h and 5 d after treatment is depicted in figure 56.

Figure 56 Myeloperoxidase (MPO) content in cryopreserved lung tissue 24 h and 5 d after treatment. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)) are depicted. Mean and SD of n=7-13 are shown.

One day after TxT MPO activity in TxT rats was 3.306 U/mg serum albumin and 2.8-fold higher compared to sham rats. Similar increase of MPO activity could be observed in rats treated with hMSC, rMSC and pre-stimulated rMSC. Only lungs from TxT+hMSC (IL1B) revealed a 67 % reduction of MPO activity compared to TxT rats and a 63 % reduction compared to TxT+hMSC rats.

Five days after TxT MPO activity was still 2.9-fold higher in TxT rats than in sham rats. The treatment with hMSC (73 %) and pre-stimulated hMSC (62 %) reduced MPO activity in lung tissue compared to TxT rats. Comparing the two points in time, the 24 h measurement of MPO activity was not significantly different from the 5 d measurement. Only MPO content in the TxT+hMSC group was 2.1-fold higher 24 h after trauma compared to 5 d after trauma.

3.4.15. Histological lung injury score

The histological score of the lung was performed in the Division for Biochemistry of Joint and Connective Tissue Diseases of Ulm University. The histological lung 132

Results injury score considers hemorrhage, granulocytes, macrophages, oedema and septum thickening. Representative pictures of Prussian blue staining of lung slices and lung injury score 24 h and 5 d after treatment are depicted in figure 57.

Figure 57 (A) Representative pictures of paraffin sections from lung of sham and TxT rats after 24 h. Pictures were performed with 20-fold magnification. Histological lung injury score after 24 h (B) and 5 d (C) with n=8. Sham rats, blunt chest trauma rats (TxT), traumatized rats receiving human MSC with or without IL1B pre-stimulation (TxT+hMSC/TxT+hMSC (IL1B)) and traumatized rats receiving rat MSC with or without IL1B pre-stimulation (TxT+rMSC/TxT+rMSC (IL1B)). Significance was calculated with one-way ANOVA followed by Dunn multiple comparisons test, * P < 0.05. Mean and SD are depicted. This figure was already published in Amann et al. (figure 4 on page 225) [5], CC BY-NC-ND 4.0, https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode.

Accumulation of lymphocytes could be observed in almost all samples, except for two single samples (TxT+hMSC 24 h and TxT+hMSC (IL1B) 5 d). However, accumulation of lymphocytes was not a criterion in the lung injury score. Hemorrhage was observed once in sham rats 24 h after operation and twice 5 d after operation. Blunt chest trauma often led to moderate hemorrhage (seven out of eight rats), accumulation of macrophages (seven out of eight rats) and granulocytes (six out of eight rats). However, granulocytes could not be observed in lung sections of TxT+hMSC rats and TxT+hMSC (IL1B) rats 24 h post injury. In

133

Results contrast, TxT+rMSC rats (six out of eight rats) and TxT+rMSC (IL1B) rats (four out of eight rats) contained granulocytes accumulation.

The lung injury score was significantly increased (2.3-fold, P=0.0272) in TxT rats compared to sham rats 24 h after operation. The treatment with hMSC and hMSC (IL1B) reduced lung injury score to 48 % or significantly (P=0.0272) to 43 % of TxT rats. Treatment with rMSC did hardly change the lung injury score compared to TxT rats. The TxT+rMSC (IL1B) rats had a 2.1-fold higher lung injury score than sham rats and TxT+hMSC (IL1B) rats.

In all cases, 5 d after treatment, the lung injury score did not differ significantly from the values at 24 h (Mann-Whitney test). Furthermore, the lung injury score did not differ significantly between the individual groups after 5 d. Notably, the lung injury score of TxT rats was still 1.8-fold higher compared to sham rats.

134

Discussion

4. Discussion

Polytrauma including thoracic injury is a life-threatening disease and often accompanied by complications like ARDS, sepsis, MODS and MOF. Causal therapeutic approaches to improve regeneration or to reduce morbidity and mortality are still to be identified. MSC transplantation is a therapeutic option for trauma patients. To investigate the immunomodulatory and regenerative effects of MSC in trauma, in vitro and in vivo experiments using trauma models were performed. The role of single trauma factors (IL1B, IL6, C3a, C5a, HMGB1, THPO and THBD) was investigated in cell culture experiments. Furthermore, IL1B-primed and unprimed MSC were administered to rats receiving a TxT.

4.1. The role of interleukin 1 beta in polytrauma

Previous results depicted a major role of IL1B within a polytrauma cocktail consisting of IL1B, IL6, CXCL8, C3a and C5a [99]. IL1B had the strongest effect on cytokine secretion, when dissecting the components of the polytrauma cocktail. It was shown that IL1B had similar effects as a combination of factors (IL1B, IL6, CXCL8, C3a and C5a) on expression of MMP1, immunomodulatory proteins (PTGS2, prostaglandin E synthase, TNFAIP6), chemokines and cytokines in MSC [99]. Furthermore, advantages of primed MSC were discussed in many publications [28,68,169,194,251,254]. IL1B priming of hMSC had shown significant effects in DSS-induced mice colitis [127]. Previous publications revealed that different sources, isolation process and cultivation of MSC influence their immunomodulatory properties [169]. The MSC used in this thesis differed from previously investigated MSC by manufacturing process (according to GMP- compliant protocols) and by supplementing medium in the presence of human PL instead of FBS, which was used in most of the studies investigating the effect of trauma factors on MSC. Therefore, one focus was set on priming when analysing the effect of several factors on MSC. The effects of a combination of trauma factors on proliferation, migration, gene expression and cytokine secretion of MSC were investigated in the present thesis. Pathophysiological (PTCL) and supraphysiological (PTCH) concentrations of IL1B, IL6, C3a, C5a, HMGB1, THPO and THBD were used [21,26,39,101,167,173,219,253,273].

135

Discussion

4.1.1. Proliferation

Stimulation of MSC with PTCH increased proliferation 2.5-fold compared to unstimulated MSC. IL1B did not influence proliferation after 7 d of cell culture (figure 5). However, FGF2 consumption of MSC increased with increasing IL1B concentration (figure 24, 4 h and 24 h). FGF2 is one of the essential components for proliferation of MSC and part of the growth factor cocktail [72]. Carrero and colleagues observed an effect of IL6 (20 ng/ml) on proliferation of MSC in MTT assay after 24 h of incubation, but did not observe an effect of IL1B (25 ng/ml) [28]. However, there were other data showing that IL1B stimulation increases the MSC cell count compared to unstimulated MSC [19,161]. These studies used multiple stimulation protocols, lower IL1B concentrations and observed effects in MSC of higher passages [19,161]. Lotfi and colleagues suggested that PL induced proliferation of MSC by DAMPs [153]. In cell culture medium supplemented with 1 % human serum, HMGB1 significantly induced cell proliferation of MSC from a concentration of 25 ng/ml to at least 100 ng/ml [153]. In the present study, no effect of HMGB1 (60 ng/ml) on proliferation of MSC could be detected. Different methods (medium supplementation) or MSC preparations might be the reason for different outcomes. For culture of MSC, 8 % PL, but in proliferation assay 0.6 % to 1.2 % PL was used. The variation of results in the present study compared to the study of Lotfi and colleagues might be due to homologous desensitization of TLR4. HMGB1 can bind to advanced glycosylation end-product specific receptor and TLR4 expressed on MSC [148]. In macrophages, it could be shown that HMGB1 binds to advanced glycosylation end-product specific receptor leading to the internalization and desensitization of TLR4 [144]. Reasons for the different outcomes compared to previous literature might be due to the use of different proliferation assays, stimulation protocols, MSC preparation (passage number, isolation and cultivation protocol, medium supplement, donor variability).

4.1.2. Migration

PTCH contained some chemotactic factors (C3a, HMGB1, THPO). Significant effects of combined (PTCH, PTCL) and single factors (IL1B, C3a, HMGB1, THBD,

136

Discussion

THPO) on migration of MSC compared to control (without chemotactic gradient) could not be observed. However, we observed a significant three to four-fold increase of migration with 150 pg/ml THPO compared to 4500 pg/ml THPO or control by two different migration assays (transwell assay, Boyden chamber). In contrast to our data, Hengartner and colleagues demonstrated that the same concentration of C3a as used in the present study, significantly increased the number of migrated MSC about two-fold in a modified Boyden chamber [99]. They investigated the effect of a polytrauma cocktail (200 pg/ml IL1B, 500 pg/ml IL6, 150 pg/ml CXCL8, 500 ng/ml C3a, 10 ng/ml C5a) and its single components on migration of MSC and showed that the polytrauma cocktail and C3a induced migration equally strong. But IL1B, IL6, CXCL8 and C5a did not induce migration [99]. Previous publications showed the chemotactic activity of HMGB1 on MSC. Lotfi and colleagues demonstrated a significant chemotactic effect of 1 µg/ml HMGB1 compared to unstimulated MSC [153]. Lin and colleagues observed a significantly enhanced migration already at 50 ng/ml HMGB1 [148]. In contrast to these studies, no significant effect of HMGB1 on migration of MSC, neither of 60 ng/ml HMGB1 nor of 1 µg/ml HMGB1, could be observed. Different outcomes in the present study might be due to different cultivation protocol of MSC before starting the migration assay or due to donor variability. The effect of THPO on MSC migration has not yet been investigated. A concentration-dependent chemotactic effect of THPO was observed in human umbilical cord vein-derived endothelial cells (1-10 ng/ml, [24]) and in hepatoblastoma cells (5-250 ng/ml, [213]) compared to unstimulated cells. Migration of these cells increased with increasing THPO concentrations. In the present thesis, also lower concentrations of THPO were investigated and revealed that 0.15 ng/ml THPO had a 3.2-fold or 3.9-fold higher chemotactic effect on MSC than 4.5 ng/ml THPO. Furthermore, it was shown that MSC express the thrombopoietin receptor (CD110) on their cell surface. Literature search revealed publications investigating the effect of THPO on fibrogenesis, proliferation and mobilization of MSC [31,32,232]. Other literature revealed that MSC secrete THPO to support adherence and proliferation of hematopoietic stem cells [36,248]. THPO plays a role in the differentiation of hematopoietic stem cells (megakaryocytopoiesis and thrombopoiesis, [36]). Furthermore, THPO plasma

137

Discussion level increased in septic patients compared to healthy volunteers, it enhanced platelet activation and possibly thereby contributes to MOF [156]. In addition, THPO (0.05-10 ng/ml) regulates the activation of PML [25]. Different concentrations of THPO may target different cells and produce different effects; these may help to maintain homeostasis. Different results compared to previous literature might be due to the use of different migration assays, stimulation protocols, MSC preparation (passage number, isolation and cultivation protocol, donor variability) and the large SD in some experiments of the present study.

4.1.3. Gene expression analysis

MSC were cultured in AB-serum for future experiments. Investigation of effects of serum from individual polytrauma patients on the expression profile of MSC is ongoing. Furthermore, the comparison of the effect of the polytrauma cocktail (see section 2.6.1.) on MSC with the effect of polytrauma serum on MSC is under investigation. Therefore, AB-serum instead of PL was used as medium supplement in cell culture experiments for RNA sequencing. AB-serum from healthy volunteers (pooled from 20 donors) served as a control in the present and following experiments. AB-serum (pooled from 20 donors) contained only higher levels of albumin and CXCL8 compared to the five representative sera from individual healthy volunteers. Level of the remaining analysed cytokines was similar in AB-serum and the five representative sera from healthy volunteers. Therefore, AB-serum (pooled from 20 donors) instead of single serum form healthy individuals was used as medium supplement in cell culture experiments for RNA sequencing. RNA sequencing was performed and revealed that change of cell culture medium had minor effects after 6 h (31 differentially expressed genes). IL1B, PTCH and PTCL had more effects on gene expression of MSC after 6 h than after 24 h of incubation. At 6 h, the overlap of differentially expressed genes between the IL1B group and PTCH group was 87 %. Genes of this intersection were assigned to pathways, but 59 % of these genes could not be assigned to any pathway because some genes did not code for a protein. The four major pathways which are upregulated after stimulation for 6 h were cytokine-cytokine receptor

138

Discussion interaction, TNF signalling pathway, chemokine signalling pathway and metabolic pathways. 26 % of differentially expressed genes in intersection of the IL1B and the PTCH group play a role in these pathways. At 24 h, still 21 % of differentially expressed genes in intersection play a role in these pathways. Hengartner and colleagues investigated the effect of a polytrauma cocktail (IL1B, IL6, CXCL8, C3a and C5a) in physiological conditions and IL1B on MSC by gene expression array (RT2 Profiler PCR Array Human Inflammatory Response and Autoimmunity) after 24 h of incubation. Their criteria for differentially expressed genes were at least two-fold upregulation and Student’s t-test [99]. In the present study, the criteria for differentially expressed genes after RNA sequencing were at least three-fold difference for up or down regulation, a false discovery rate less than 0.001 and a read count of at least 10 for the respective target gene. Hengartner and colleagues detected that the expression profile of MSC stimulated with their polytrauma cocktail and IL1B (0.2 ng/ml) was similar [99]. In line with this study, the present study showed similar effects using supraphysiological concentrations of cytokines in PTCH and of IL1B. Similar to our results they found significant effects of IL1B on cytokine and chemokine expression profile [99]. In contrast to our results, they did not find downregulated genes [99]. Our results showed that 22 % (at 6 h) or 27 % (at 24 h) of differentially expressed genes were downregulated after stimulation of MSC with IL1B or PTCH. Downregulated genes in intersection mainly affected (25 % of downregulated genes) neuroactive ligand- receptor interaction, calcium signalling pathway, pathways in cancer, mitogen- activated protein kinases (MAPK) signalling pathway and metabolic pathways after stimulation for 6 h and 24 h. These differences in results exist due to the difference in the extent of gene analysis. Hengartner and colleagues did not detect a major concentration-dependent effect of IL1B (0.2 ng/ml versus 10 ng/ml) on cytokine and chemokine expression [99]. In contrast to that, PTCH revealed 2.5-fold higher amount of differentially expressed genes compared to PTCL. This might be due to a concentration-dependent effect of IL1B or due to the other factors in PTCL (IL6, HMGB1, THBD, THPO). Carrero and colleagues investigated the effect of a supraphysiological concentration of IL1B (25 ng/ml) on MSC cultivated in FBS-supplemented media by gene expression analysis (microarray and real time quantitative PCR). In line with the present study, they highlighted the effect of IL1B on expression of

139

Discussion chemokines (like CXCL1, CCL2), interleukins (IL1, IL6, CXCL8, TNF) and MMPs (MMP1, MMP3) [28]. Furthermore, in line with the present study they revealed an impact on Toll-like receptor signalling and NF-kappa B signalling pathway [28]. Interleukin-1 signalling pathway consists of the NF-κB, c-Jun N-terminal kinase, and the p38 MAPK pathways which mediate the IL1 effects on gene expression [256]. Responsive genes are NF-κB inhibitor alpha, CCL2, IL6, CXCL8 and PTGS2 [256]. This general principle is reflected also in the present study. The effect of IL1B on gene expression of MSC was investigated in several studies using different MSC preparations (isolation and cultivation protocol, donor variability) and methods to analyse gene expression. All studies revealed that IL1B activates MSC and increases the expression of factors which might be helpful in the therapeutic application of MSC. Furthermore, the present study confirmed that IL1B can mimic the effect of PTCH and PTCL on MSC. Therefore, IL1B stimulation of MSC was used in the investigation of the role of MSC in polytrauma and to prime MSC for therapeutic application in the rat TxT model.

4.1.4. Secretion profile of mesenchymal stromal cells

Concentrations of cytokines, chemokines, MMPs and growth factors in supernatant from MSC culture (supplementation with 20 % AB-serum) after stimulation with IL1B, PTCH and PTCL was analysed 6 h and 24 h after incubation. There was a significant increase of IL6, CXCL1 and CXCL8 in supernatant from MSC 24 h after stimulation with IL1B, PTCH and PTCL. TNF level increased already 6 h after stimulation with IL1B, PTCH and PTCL. No significant differences between physiological and supraphysiological concentrations of trauma factors on secretion profile of MSC, except for MMP10 (24 h: IL1B versus PTCL) could be observed. Wight and colleagues reviewed the role of MMP10 in lung inflammation [259]. MMP10 drives the conversion of classically activated macrophages into alternatively activated macrophages [166]. Furthermore, MMP10 influenced collagenolytic activity of macrophages thus reducing fibrosis and scar formation [209]. Induction of MMP10 secretion after IL1B stimulation might be a therapeutic advantage of primed MSC.

140

Discussion

Cytokine analysis of supernatant revealed that medium exchange (20 % AB-serum instead of 8 % PL) influenced cytokine secretion significantly. Medium supplementation with AB-serum induced higher secretion level (TNFAIP6, IL10, IL1RN, FGF2, CCL4) than medium supplementation with PL. Ulivi and colleagues investigated the role of MSC in wound healing [242]. They described that PL induced an inflammatory response (PGE2, IL6) in MSC which maintained macrophages in a pro-inflammatory state (secretion of TNF) [242]. MSC enhanced the initial inflammatory response to injury and thus supported the activation of wound healing [242]. They showed that 100 U/ml IL1A and 5 % PL had significant additive effects on NF-κB activity, IL6 expression and PGE2 synthesis in mouse BM-MSC after 16 h or 24 h compared to MSC stimulated only with either IL1A or PL [242]. They used conditioned medium from untreated and pre-treated MSC to investigate the effect on macrophages [242]. Several publications revealed the impact of TNF-secreting macrophages and the effect of TNF on MSC to increase TNFAIP6 expression and to induce the switch of macrophages to a pro-resolving phenotype [181,198,242]. The secretion profile of MSC depends on the cultivation method. Furthermore, subtraction of background might disguise low cytokine levels and changes in these low cytokine levels. Analysis of the secretion profile of MSC after IL1B stimulation confirmed that IL1B mimics the effect of PTCH.

4.1.5. Sera from polytrauma patients

Ongoing experiments investigate the effect of serum from polytrauma patients on the expression profile of MSC and compare the effect of the polytrauma cocktail (see section 2.6.1.) on MSC with the effect of polytrauma serum on MSC. The level of 21 inflammatory factors and albumin in five sera of polytrauma patients was investigated by Multiplex assay or ELISA. Not every factor included in PTCL was addressed in these assays, due to small sample amount of serum from patients with polytrauma and to other priorities. Serum albumin level was significantly reduced in polytrauma patients compared to healthy controls, especially 5 d and 10 d after polytrauma due to fluid resuscitation. There was a significant increase of IL6 (47-fold), CXCL1 (4.6-fold) and CCL2 (14-fold) level in serum from polytrauma patients at admission to hospital (0 h) compared to serum

141

Discussion from healthy volunteers. Furthermore, IL10 level was also not significantly increased. IL1B could not be detected in serum of polytrauma patients and of healthy controls. IL1B did not play a role in these patients, but studies with a higher number of patients revealed an increase of IL1B in plasma from ARDS patients compared to healthy volunteers [167]. According to Meduri and colleagues persistent elevation of IL1B and IL6 levels in plasma from ARDS patients is a predictor of poor outcome [167]. Sapan and colleagues investigated mRNA expression in polytrauma patients and used IL6 as a mediator of SIRS and IL10 as a mediator of CARS [218]. High IL6/IL10 ratio indicated pro-inflammatory response, while low IL6/IL10 ratio indicated a malfunction of immune system [218]. Both are predictors for MODS and death [218]. The composition and levels of cytokines in sera from polytrauma patients differed from the ones of PTCL. This may due to the heterogeneity of trauma, different analysis methods and underlying diseases of patients. To obtain more significant results a larger and more homogenous trauma group would be needed. Future experiments will show the comparability of polytrauma sera and PTCL or PTCH.

4.2. Characterization of rat mesenchymal stromal cells

Allogeneic rMSC transplantation was used as control to xenogeneic hMSC transplantation in rat blunt chest model. Therefore, one aim of the present study was to isolate, characterize and investigate the rat-specific IL1B priming of rMSC from Wistar rats. Rat MSC used in the present study were isolated, cultured and characterized similar to established protocols in the Institute of Transfusion Medicine [65,244]. Merck Millipore characterizes their BM-MSC from Fisher 344 rats by expression of CD90 and the absence of CD45 and CD31 [171]. Cyagen Biosciences characterizes their Wistar MSC by tri-lineage differentiation and by flow cytometry marker [48]. The MSC are negative for CD11b/c, CD34 and CD45 with a minimum of 5 %. The marker CD90, CD44 and CD29 are positive for more than 70 % of the cells [48]. In the present study, rMSC fulfilled these criteria, but only 13.71 % ± 23.50 % rMSC were positive for CD44H. CD44H is the standard form of CD44 and did not show time-dependent increase on MSC [65]. Since only early passages of rMSC were used in this thesis, low CD44 expression is in line with the

142

Discussion observation of Barzilay and colleagues who compared different rat-specific MSC markers on four rat strains (Fisher, Lewis, Sprague Dawley and Wistar) and showed expression of CD90 and CD29 as well as absence of expression of CD45 and CD11b on MSC from all rat strains at least until passage seven, whereas CD44 and CD106 expression depended of the rat strain [11]. Time-dependent increase of CD44 mainly was observed for MSC isolated from Fisher and Lewis rats, whereas a decrease of CD44 expression occurred on MSC from Sprague Dawley and Wistar rats [11]. Wistar rats were shown to express CD44 on about 10 % of MSC of passage two with a decrease of expression to 1 % at passage seven [11]. In the present study rMSC were successfully isolated and characterized and comparable results could be obtained with regard to literature and commercially available rMSC.

4.3. Effects of manipulating MSC for in vivo experiments

Aims of the thesis were to establish the PKH26 staining for tracking MSC in in vivo experiments and to analyse the effect of priming of hMSC and rMSC with IL1B in in vitro and in vivo trauma models. Furthermore, in vitro co-culture experiments using the rat alveolar macrophage cell line NR8383 and rMSC or hMSC were performed to analyse the effect of allogeneic and xenogeneic co-cultivation. The in vitro experiments included the same MSC cell preparations which were used in the in vivo experiment. The in vitro experiments therefore can be seen as a kind of potency assays. In order to assess whether PKH26 staining or IL1B priming of MSC changes the properties of MSC, in vitro studies were performed.

4.3.1. PKH26 staining of mesenchymal stromal cells

Doubling time, CFU, and cell death marker of unstained and PKH26-stained MSC were analysed to exclude cytotoxicity of staining concentration of PKH26. Fluorescence intensity of PKH26-stained MSC was measured for up to 5 d to compare fluorescence intensity with recovered MSC from rat blood. Furthermore, the effect of PKH26 staining on selected surface markers of MSC and secretion of selected cytokines was investigated.

143

Discussion

PKH26 staining of MSC (hMSC and rMSC) according to manufacturer’s protocol (final PKH26 dye concentration of 2.0 x 10-6 M) did not produce MSC with significantly higher fluorescence intensity compared to unstained MSC. Therefore, higher concentrations of PKH26 (final PKH26 dye concentration of 1.6 x 10-5 M) dye were used. Deak and colleagues used a final PKH26 dye concentration of 1.7 x 10-5 M for the staining of mouse MSC [53]. Kim and colleagues used even a final PKH26 dye concentration of 10-3 M for rat BM-MSC and investigated engraftment of PKH26-positive cells in Sprague-Dawley rats 6 weeks post injury [121]. Median fluorescence intensity of PKH26-stained hMSC and PKH26-stained rMSC fluctuated strongly immediately after staining. No significant differences between median fluorescence intensity of PKH26-stained hMSC and of PKH26-stained rMSC could be observed. However, rate of decrease in fluorescence intensity differed between PKH26-stained hMSC and PKH26-stained rMSC. Fluorescence intensity of hMSC decreased 4.7-fold faster than that of rMSC, due to differences in doubling time. CFU and doubling time did not differ between PKH26-stained and unstained MSC. However, doubling time of rMSC was 3.1-fold higher on average than that of hMSC. Annexin V and SYTOX staining of hMSC and rMSC were analysed by flow cytometry and did not reveal significant differences between unstained and PKH26-stained MSC. MSC stained with trypan blue were manually counted and results revealed significant differences between unstained and PKH26-stained rMSC. Therefore, for in vivo studies, trypan blue staining after PKH26 staining was performed to adjust cell count number that had to be injected in rats. The present study revealed that PKH26 staining did not significantly influence expression of selected surface markers which were used to analyse recovery of MSC in TxT model. In addition, PKH26 did not significantly influence secretion of selected relevant cytokines. PKH26 was often used to stain MSC and to track these cells in animal models [53,121,217]. Morphology, cell cycle analysis, proliferation, viability and differentiation capacity of MSC after staining with PKH26 were analysed and investigators concluded their results with the safe and efficient use of PKH26 to label and to track MSC [141,187,231]. Polzer and colleagues compared three different vital staining methods (PKH26, CFSE, enhanced green fluorescent protein transduction). They analysed metabolic activity and relative fluorescence

144

Discussion intensity of stained immortalized hMSC in vitro and in vivo [193]. They observed a significant decrease of metabolic activity of hMSC after staining with PKH26 and CFDA-SE, but not of eGFP cells compared to control [193]. They indicated that eGFP was the only one which is suitable for long-term detection (up to 6 weeks) of hMSC [193]. However, they also revealed, in in vivo experiments with athymic nude mice, that relative fluorescence intensity of eGFP cells decreased over the first 21 d as it is the case with PKH26 and CFDA-SE labelling [193]. They guessed that the decrease of fluorescence intensity is due to cell death after transplantation [193]. We investigated the fate of transplanted MSC only up to 5 d after TxT and considered PKH26 staining as suitable. For in vivo experiments we kept in mind that PKH26-labelled cell debris could transfer to host cells and display them red fluorescence which was determined in in vitro and in vivo assays [141]. We therefore confirmed recovery of injected cells in TxT model by cell adherence, surface marker expression and Alu analysis. In line with other publications, the present study showed that the use of PKH26 to label and to track MSC is safe and efficient. PKH26 staining did not influence properties of MSC (doubling time, CFU, surface marker expression, proportion of dead cells).

4.3.2. Interleukin 1 beta priming of mesenchymal stromal cells

Several publications revealed that pro-inflammatory signals are necessary to induce immunosuppressive potential of MSC [6,99,169]. Trauma induces the release of pro-inflammatory cytokines which might induce immunosuppressive and regenerative activity of MSC. We hypothesized that priming of MSC could accelerate this process. Therefore, unprimed and primed MSC were compared in in vitro and in vivo experiments.

Effect on adhesion molecules Flow cytometry analysis of rMSC revealed significant increase of CD54 (ICAM1) surface marker expression after stimulation with 10 ng/ml IL1B for 24 h compared to unstimulated rMSC. Wang and colleagues also observed a significant increase of VCAM1 expression (messenger RNA and protein) using the same protocol of stimulation [249]. Furthermore, they observed a significant increase of cell

145

Discussion adhesion ability in stimulated rMSC compared to unstimulated rMSC [249]. RNA sequencing of IL1B-stimulated hMSC revealed an increase of ICAM1, VCAM1 and CXCR4 compared to unstimulated hMSC after 6 h of incubation. 24 h after incubation with IL1B only ICAM1 was increased compared to the unstimulated control.

Effect on cytokine secretion Furthermore, the effect of IL1B stimulation (0.1 ng/ml, 1 ng/ml, 10 ng/ml, 100 ng/ml) on cytokine secretion of hMSC after different incubation times (4 h, 8 h, 12 h, 24 h, 48 h) was analysed in the present study. We chose several inflammatory cytokines (IL1B, IL6, IL10, IL17A, TNF), chemokines (CXCL1, CCL2), pro-angiogenic (CXCL8, VEGFA) and regenerative factors (IL1RN, FGF2, TNFAIP6). Measurement of IL1B in supernatant revealed that IL1B was added in abundance to cell culture. Even the lowest concentration of IL1B was not consumed after 48 h. We did not observe any increasing effects by stimulation of MSC with 100 ng/ml IL1B and decided to use 10 ng/ml IL1B for in vivo experiments. There were early rising factors (4 h to 24 h: IL6, CXCL1 and CXCL8) and late rising factors (VEGFA, TNFAIP6) after stimulation of MSC with IL1B compared to unstimulated MSC. IL10, IL1RN, FGF2 and TNF are components of the cell culture medium or rather of PL [70] and were not produced by the cells itself in this cell culture experiment. High levels of cytokines in PL may mask the effect of IL1B stimulation. The elevation of IL1RN expression and secretion in MSC after IL1B stimulation mentioned before (see section 3.1.3. and 3.1.4.) could not be observed in medium supplemented with PL. However, we decided to use PL as supplement, because this protocol has been approved as protocol to manufacture MSC, e.g. for the clinical trials OrthoCT-1 (EudraCT number: 2011- 005441-13, ClinicalTrials.gov Identifier: NCT01842477), OrthoCT-2 (EudraCT number: 2012-002010-39, ClinicalTrials.gov Identifier: NCT02065167) and MaxilloCT-1 (EudraCT number: 2012-003139-50, ClinicalTrials.gov Identifier: NCT02751125) by the national competent authorities in Germany, France, Italy, Spain (OrthoCT-1, OrthoCT-2) and Norway (MaxilloCT-1) [71,87,180]. This protocol has already been successfully used in pre-clinical models dealing with bone regeneration and wound healing as well [23,198].

146

Discussion

Furthermore we confirmed that the induction of cytokine and chemokine secretion continues 24 h after priming with 10 ng/ml of IL1B for 24 h. A second IL1B stimulation did not have the same effects on cytokine secretion of hMSC and rMSC. In contrast to that, IL1B stimulation for 48 h without medium/microenvironment change revealed a further increase of some factors (TNFAIP6, VEGFA).

We investigated the effect of IL1B stimulation (10 ng/ml IL1B for 24 h) of hMSC on secretion of TNFAIP6, IL6, TNF, CCL2, CXCL1, CXCL8, VEGFA, FGF2, IL10 and IL1RN in three different media (Supplementation with 20 % AB-serum, 8 % PL or a mixture of 0.6 % PL and 15 % FBS). The effect of cytokine secretion comparing stimulated with unstimulated hMSC is very different between these three media. The lowest effect on cytokine secretion could be observed with medium containing AB-serum. Only the secretion of CXCL1 and VEGFA was elevated more than 2.0- fold. Supplementation with 8 % PL revealed some additional effects. Secretion of TNFAIP6, IL6, CCL2, CXCL1 and CXCL8 was elevated more than 2.0-fold. The most and strongest effects could be observed in the co-culture medium containing FBS. Secretion of TNFAIP6, IL6, TNF, CCL2, CXCL1 and CXCL8 was elevated more than 2.0-fold. Secretion of CXCL1 was elevated 6.4-fold in medium containing AB-serum, 7.0-fold in medium containing PL and 67-fold in medium containing FBS compared to unstimulated hMSC. Secretion of CXCL8 was elevated 1.8-fold in medium containing AB-serum, 7.9-fold in medium containing PL and 32-fold in medium containing FBS compared to unstimulated hMSC. TNFAIP6 secretion did not increase in medium containing AB-serum, but increased 2.6-fold in medium containing PL and 3.4-fold in medium containing FBS compared to unstimulated hMSC. Background of IL1B stimulation seemed to play a role in outcome analysis. In contrast to the present study, Hengartner and colleagues revealed a 25-fold increase of TNFAIP6 secretion after stimulation BM- MSC with 10 ng/ml IL1B [100]. They used different isolation and cultivation protocols including FBS for supplementation of cell culture medium which might be the reason for difference in outcome [99,100].

Cytokine panel in human-specific and rat-specific Multiplex analysis differed due to manufacturer's availability. Relative cytokine secretion of rMSC after IL1B

147

Discussion stimulation for 24 h was similar (TNFAIP6), slightly increased (VEGFA, IL10) or strongly increased (IL6, CCL2, CXCL1) compared to relative cytokine secretion of hMSC. With regard to the above explanations, it should be noted that medium containing a xenogeneic supplement was used for the analysis of rat cytokine secretion pattern after IL1B priming. Optimal cell culture condition for rMSC was established in a previous work of the Institute of Transfusion Medicine, Ulm University [65].

In summary, there are several hints that IL1B priming of MSC has advantages for treatment of TxT rats. Adhesion molecules (ICAM1), MMPs (MMP1) and several cytokines (IL6, CXCL1, CXCL8, TNFAIP6) were upregulated after stimulation of MSC with IL1B. Chemokines and adhesion molecules are important for the interaction of MSC with target cells [64,184,203]. Chemokine-mediated migration of MSC to these target cells shortens the distance between the cells and makes paracrine interactions possible [95,203,204]. CXCL8 secretion by MSC recruits neutrophils [22]. IL6 secretion by MSC suppressed apoptosis of neutrophils [30,199]. Furthermore, Jiang and colleagues proved an ICAM1-dependent engulfment of apoptotic neutrophils by MSC avoiding further tissue damage by destructive enzymes and factors [114]. Pricola and colleagues reported that IL6 maintained the undifferentiated state of MSC [196]. Furthermore, IL6 seems to be important for PGE2 secretion by MSC [20]. Co-culture experiments of endothelial cells and hMSC highlighted the ability of IL6 to reduce infiltration of leucocytes [157]. CXCL1 and CXCL8 are pro-angiogenic and may promote tissue repair [250,252]. Lee and colleagues revealed a CCL2-dependent recruitment of myeloid-derived suppressor cells [132] which might influence T cell response after TxT [107]. The role of TNFAIP6 was already mentioned in the introduction, its therapeutic effect was shown in several publications [37,50,135,198].

4.3.3. Co-culture of human or rat mesenchymal stromal cells and the rat alveolar macrophage cell line NR8383

To investigate the interaction of human and rat cells, co-culture of hMSC and rat alveolar macrophage cell line NR8383 was performed. NR8383 cells were also used to study ALI [55]. Previous publications have shown that IL1B activates MSC

148

Discussion

[28,99] and that LPS activates macrophages [198,205]. Stimulation of co-culture with human IL1B did not reveal many effects. However, LPS stimulation of co- culture revealed that hMSC interacted with the rat cell line NR8383 resulting in a significantly higher level of TNFAIP6 in cell culture supernatant compared to single cultures and a lower level of rat TNF in cell culture supernatant compared to NR8383 cells. In addition, we observed that cell to cell contact is essential to reach a higher level of TNFAIP6, rat TNF and rat IL10 in co-culture supernatant compared to transwell co-culture. Stimulation of co-culture of rMSC and NR8383 cells with LPS had more effects than stimulation with rat IL1B, except for TNFAIP6 and VEGFA secretion. There was no significant difference between direct co-culture and transwell co-culture. LPS stimulation of co-culture revealed that rMSC interacted with the NR8383 cell line resulting in a significantly higher level of CXCL1, CCL2 and IL6 in cell culture supernatant compared to single cultures. Furthermore, TNF secretion in NR8383 single culture was higher and lower in rMSC single culture than in co-cultures. The effect of rMSC and hMSC was different in co-culture experiments, except for reduction of TNF secretion of NR8383 cells after LPS stimulation. The mechanism of interaction seems to be different but both hMSC and rMSC influence secretion profile of the rat alveolar macrophage cell line NR8383 (reduction of TNF secretion). However, the key to reduce TNF secretion by macrophages lies rather in the activation of the macrophages than the MSC as already explained in section 4.1.4.. Differences in co-culture analysis might be caused by different culture conditions or species-specific mechanisms. Interspecies differences of macrophages have been investigated by several groups. Alveolar macrophages from humans and rats have similar morphology [102], but differed in cell diameter [128], phagocytic activity [222] and inflammatory mediators [112,243]. Furthermore, humans have a greater number of alveolar macrophages in the airways compared to rats [43]. In addition, it seemed that there are different mechanisms of immunomodulation of MSC from different sources. Co-culture experiments investigated the anti-inflammatory effect of human BM-MSC, adipose- derived mesenchymal stromal cells (ASC), UC-MSC in LPS response of NR8383 cells [115]. UC-MSC had the strongest anti-inflammatory effects measured by IL1A, IL6 and CXCL8 level in supernatant [115]. UC-MSC secreted significantly higher amount of angiopoietin 1 in co-culture compared to BM-MSC and ASC and

149

Discussion

UC-MSC is the only one that had significant effect on all three cytokine levels [115]. Different cultivation protocols for MSC expansion, cytotoxic LPS concentration [138] and other readout analysis [115] were chosen compared to our study. Further experiments are necessary to obtain information about MSC, their potential role in inflammation process and their interaction with other cells. Preliminary experiments are the basis for in vivo studies. However, the effect in in vivo studies might be different since outcome depends on microenvironment, species-specific mechanisms and the injury model.

4.4. Mesenchymal stromal cells in blunt chest trauma

Inflammatory and fibrotic pathways are induced by lung injury. There are several animal models of ALI (reviewed in [165]). We investigated the effect of a single dose of allogeneic or xenogeneic MSC in a rat TxT model. This model was originally established for rats [108,111] and used to study early cardiopulmonary and histological changes as well as local and systemic inflammatory reactions after lung contusion [117,122,227,229]. This pre-clinical model revealed that lung injury is driven by activation of alveolar macrophages, neutrophil infiltration and apoptosis of pneumocyte type II [147,190,191,226,227,229]. Furthermore, there is also evidence that these cells play a role in the development of ALI in humans as already described in section 1.2.1.. We analysed the recovery of MSC, local and systemic inflammatory reaction as well as histological lung injury score 24 h after TxT or sham treatment. In addition, we have chosen a second point in time (5 d) for the investigation of the effect of hMSC in the rat TxT model.

We injected 5 x 106 MSC in 1 ml 0.1 % BSA/DPBS intravenously into rats. This is in line with other publications [98,179,216,269]. Taking into account the bodyweight and surface area of rat and human, this dose corresponds to about 2.6 x 106 MSC per kg bodyweight in clinical studies [241]. The clinical studies, described in the introduction part, used up to 10 x 106 MSC per kg bodyweight, thus our pre-clinical study is in accordance with current clinical studies [180].

150

Discussion

Median fluorescence intensity of PKH26-labelled hMSC used for in vivo experiments was not significantly higher compared to median fluorescence intensity of PKH26-labelled rMSC prior to injection of MSC in rats, but pilot in vitro experiments in this study showed contrasting results. Furthermore, IL1B-primed and PKH26-labelled MSC showed a non-significant increase of median florescence intensity compared to unprimed, PKH26-labelled MSC. Change in cell membrane composition, cell morphology and cell size might be the reason for the difference in PKH26 uptake [116,211]. For analysis of rat peripheral blood and BAL, we defined life gate in forward- sideward scatter diagram to exclude debris and dead cells. Gating strategy for PKH+ events used in the present study was similar to Kim and colleagues [121].

There were several differences in recovery rate between allogeneic and xenogeneic MSC administration. Percentages of PKH+ events in rat blood from TxT+rMSC and TxT+rMSC (IL1B) rats were significantly higher compared to controls 24 h after TxT. Recovery of injected rMSC was observed in rat blood but not in BAL 24 h after TxT. Fluorescence intensity of PKH+ events in rat blood was moderately reduced in TxT+rMSC rats compared to fluorescence intensity of originally injected PKH26-labelled rMSC. However, there was no significant reduction of fluorescence intensity between PKH26-labelled rMSC (IL1B) and PKH+ events in blood and BAL from rMSC (IL1B)-treated animals. About 50 % of PKH+ events in rat blood derived from TxT+rMSC and TxT+rMSC (IL1B) rats showed rMSC-specific markers (CD3-, CD11b/c-, CD45-, CD29+, CD54+, CD90+). We concluded that rMSC could be recovered from rats 24 h after rMSC administration. While fluorescence intensity of PKH+ events decreased significantly over time in blood from hMSC-treated rats, it increased in BAL after 5 d compared to 24 h. Percentage of PKH+ events in rat blood and BAL was significantly higher in TxT+hMSC and TxT+hMSC (IL1B) rats compared to controls 5 d after TxT. Late increase of fluorescence intensity and recovery of PKH+ cells might be due to transfer of PKH26 to rat cells [141]. Fluorescence intensity of PKH+ events in rat blood was reduced severely in TxT+hMSC and TxT+hMSC (IL1B) rats compared to fluorescence intensity of originally injected hMSC with or without IL1B priming. An analysis of PKH+ events in blood derived from TxT+hMSC and TxT+hMSC

151

Discussion

(IL1B) rats in more detail did not reveal hMSC-specific markers (CD105, CD90). Seven different tissues were analysed by Alu sequence analysis. Only very low amounts of human DNA were detected in lung tissue of few TxT+hMSC and TxT+hMSC (IL1B) rats 24 h after trauma. Size and morphology of hMSC changed after systemic injection into rat, but it did not change if cells were incubated with rat blood ex vivo. PKH+ cells isolated from blood were smaller than the originally stained MSC. Diameter of isolated PKH+ events from rat blood was about 5 to 10 µm on cytospin slide and in confocal microscopy analysis. The reduction of fluorescence intensity of PKH+ cells before and after injection into rats might be explained by smaller cell size. Furthermore, PKH+ cells of blood from TxT+hMSC rats were not plastic adherent and expressed leucocyte-specific surface markers (CD3, CD11b/c, CD45). We concluded that hMSC could not be recovered from rats 24 h after hMSC administration.

Leibacher and colleagues investigated the intravenous administration of PKH67- labelled hMSC (2 x 106) from male donors in immunocompetent female mice [136]. Analysis of expression of the sex-determining region Y revealed that 5 min after transplantation about half of the MSC were detected in the lung and decreased dramatically within 24 h. In contrast to our results, they did not find PKH67-positive signals in blood by flow cytometry analysis 24 h after MSC administration. However, they found 0.2 % of injected cells in minced lung by flow cytometry analysis 24 h after MSC administration. They revealed similar results to the present animal study. They observed reduction of cell size and absence of nucleus of PKH67-positive signals. They used calibrated microbeads to analyse cell size of PKH67-positive cells before and after injection in mice and showed an increase of percentage of smaller particles with a size between 10 and 29.6 µm 24 h after MSC administration. They showed an early interaction (within 30 min) of PKH67- labelled MSC with phagocytic cells by surface marker staining. In addition, they showed an increase of apoptosis marker (Annexin V, Caspase-3) on PKH67- positive cells [136]. MSC particles described by our study and in the study of Leibacher and colleagues had a bigger size than microvesicles and exosomes. However, size of apoptotic bodies (< 4 µm) is close to the observed particle size with regard to extracellular vesicles [1]. The importance of apoptotic cells and phagocytic cells for therapy of ALI is discussed in section 4.4.1..

152

Discussion

Bioluminescence imaging, intravital microscopy, donor DNA analysis and donor RNA analysis revealed that the persistence of hMSC (in mice with severe combined immunodeficiency), mouse MSCs (in syngeneic mice) and rMSC (in allogeneic rats) was limited, with the majority of cells dying within 48 h after systemic infusion [119,135,240]. Furthermore, Vernikouskaya and colleagues used the same expansion protocol for hMSC and rMSC [244]. They used magnetic resonance imaging and observed signal loss of iron-marked hMSC already 3 h after intramuscular injection of hMSC into Wistar rats [244]. However, they observed redistribution effects of rMSC up to 48 h after rMSC injection [244]. Until now, it is not fully elucidated if fragmentation of MSC occurred due to shear stress in lung capillaries, innate immune system or generation of extracellular vesicles [137].

In addition to flow cytometry analysis, cytospins of BAL were performed and analysed with fluorescence microscopy in the present study. In line with flow cytometry analysis of BAL, non-significant increase of red fluorescent events in MSC-treated animals could be observed compared to controls 24 h post trauma or sham procedure. Recovery of MSC from BAL as described in cell culture experiments (TxT+rMSC) implies that MSC might be transmigrated through the alveolar wall or reached the alveolar space by leakage of blood wall as a consequence of TxT.

Analysis of endogenous mobilization of MSC in sham, TxT, TxT+rMSC, TxT+rMSC (IL1B) rats by cell culturing and flow cytometry revealed a non- significant increase of cells which fulfilled MSC criteria in TxT, TxT+rMSC and TxT+rMSC (IL1B) rats compared to sham rats or mouse cell line (feeder cells). A combination of markers was used to detect exogenous and endogenous MSC by flow cytometry. This approach is not common and resulted in few event numbers. Release criteria for MSC do not include combination of markers in flow cytometry analysis, just percentages of single markers are utilized [262]. It is assumed that stem cells are mobilized during tissue injury to increase regeneration. Hannoush and colleagues showed that CSF3 pre-treatment of rats and local administration of CXCL12 immediately after lung contusion significantly increased the homing of bone marrow derived cells to sites of injury and

153

Discussion significantly reduced lung injury score 5 d after lung contusion compared to untreated traumatized rats in a unilateral lung contusion rat model [98]. In contrast to the present study, they used shredded lung tissue, performed colony assays and observed significant more colonies 24 h after lung contusion and administration of both factors (CXCL12 and CSF3) compared to lung contusion with or without administration of single factors. Rochefort and colleagues revealed that hypoxia selectively increased circulating MSC number but not hematopoietic progenitor cell number in Wistar rats [208]. In contrast to our data, they observed circulating MSC also in normoxic rats. They normalized CFU to seeded cells like it was done in other studies [98,208]. We investigated plastic-adherence cells of cells from blood and BAL. Rochefort and colleagues analysed MSC-specific surface markers after cell culture of circulating MSC. CD3, CD11b/c and CD29 were not included in their MSC panel [208]. In humans, studies using blood samples from patients with organ injury did not reveal an increase of circulating MSC compared to healthy controls [103,258], however, there are conflicting findings with regard to trauma patients with fractures [103,258]. Different surface marker panel, limited animal/patient number and different regeneration potential of rats and humans might be the reason for different outcomes.

We did not observe significant changes in systemic cytokine levels between the six animal groups, except for the C5a plasma level. TxT+rMSC (IL1B) rats had a significantly higher C5a level compared to sham and TxT rats. TxT+hMSC (IL1B) rats had a significantly higher C5a level compared to TxT rats. The advantages of anti-C5a treatment immediately after induction of TxT in Wistar rats were already shown in a previous publication [75]. The significant increase of the C5a level after administration of primed MSC revealed the hyperacute rejection reaction [129]. It was shown that IFNG or TNF priming of BM-MSC increased the expression of major histocompatibility complex, class I proteins [195]. Furthermore, IFNG and TNF priming induced expression of the co-stimulatory molecule CD40 especially in BM-MSC which were cultivated in PL-supplemented medium [169]. This might also be the case for IL1B priming, resulting in hyperacute rejection reaction. Local inflammatory reaction might also be masked by high variations within each animal group. Only CCL2 and IL1RN level in BAL differed significantly between sham and TxT rats 24 h after trauma. TxT+hMSC and TxT+hMSC (IL1B) rats had

154

Discussion a non-significantly lower cytokine level (IL1B, IL6, TNF), chemokine level (CXCL1, CCL2) and IL1RN level, but higher TNFAIP6 level in BAL compared to TxT rats at 24 h. In contrast, the treatment with rMSC and rMSC (IL1B) did not ameliorate inflammatory reaction after TxT at 24 h. IL1B, IL6, IL10, CXCL1 and TNFAIP6 level in BAL was significantly higher in rMSC-treated rats compared to hMSC- treated rats 24 h after trauma. TNFAIP6 levels in BAL from TxT+rMSC rats were significantly higher compared to sham, TxT and TxT+hMSC (IL1B) rats 24 h after trauma. The use of a rat-specific cytokine assay without detection of human cytokines or different effects of xenogeneic and allogeneic transplantation might be the reason for these differences. As already mentioned, in vitro experiments showed that relative cytokine secretion of rMSC after IL1B stimulation for 24 h was similar (TNFAIP6), slightly increased (VEGFA, IL10) or much higher (IL6, CCL2, CXCL1) but never lower compared to relative cytokine secretion of hMSC, thus in vivo experiments might reflect in vitro experiments. Furthermore, it was shown that hMSC mediated anti-inflammatory effects by secretion of TNFAIP6 in a mouse model of LPS-induced ALI [50]. Our results showed that TNFAIP6 secretion might also be a therapeutic mechanism of rMSC. VEGFA level in BAL was significantly reduced in TxT+hMSC and TxT+hMSC (IL1B) rats compared to sham rats 5 d after trauma. PGE2 level in BAL was significantly increased in TxT+hMSC rats compared to sham rats 5 d after trauma. It could be an indication of consumption and increased production of these regenerative factors by lung cells (endothelium and epithelium) triggered by MSC administration. Local and systemic reaction after TxT in Wistar rats were analysed in several previous studies [61,75,145–147]. Recknagel and colleagues observed a significant increase of IL6 level in serum from TxT rats with femur osteotomy compared to untreated rats and rats with femur osteotomy 6 h and 24 h post treatment [201]. However, systemic inflammatory reaction was not induced 24 h after TxT as shown in several studies [75,146,147]. Blunt chest trauma was often combined with a second injury like fracture [201], sepsis [190] or haemorrhagic shock [228]. These two hit models revealed significant increase of systemic cytokine level compared to single injury models [190,201]. Flierl and colleagues as well as Liener and colleagues depicted an early increase of TNF (1 h, 6 h and 12 h), CXCL1 (peak at 6 h) and CCL2 (peak at 6 h) in plasma from TxT rats compared to sham rats [75,147]. However another publication did not show

155

Discussion significant upregulation of CCL2, IL10 and PGE2 in plasma from TxT rats compared to sham rats 6 h and 24 h post injury [146]. A local inflammatory reaction after TxT was observed in Wistar rats in several studies [75,146,147]. Liener and colleagues showed that TxT rats had significantly higher IL1B, IL6, CXCL1 and CCL2 levels in BAL compared to sham rats 24 h after trauma [147]. However, it was shown that PGE2 was an early marker in local inflammatory reaction with a peak at 10 min after TxT [75,147]. Furthermore, BAL cytokine levels vary between different studies using the same TxT model and rat strain [75,147]. IL6, CXCL1 and CCL2 levels from BAL of sham rats were higher in our study compared to previous publication [147]. This might be due to several reasons, like different methods to obtain the BAL (flushing a part or the entire lung, single or multiple flushing and volume of flushing), different analysis methods to determine cytokine level (assay sensitivity of Multiplex and ELISA) or the immunological status of the rats before application of TxT. BAL cell count and BAL cytokine level in our study differed partly from previously published values [75]. This might be due to the use of different controls (sham procedure [147] with or without intravenous administration of 0.1 %BSA/DPBS [5] or rabbit IgG in DPBS [75]), timing of outcome analysis, sampling condition and high variability in each group.

TxT and TxT+rMSC rats had a not significantly higher BAL cell count compared to sham rats 24 h after application of trauma. TxT+hMSC, TxT+hMSC (IL1B) and TxT+rMSC (IL1B) rats had similar cell count to sham rats. In addition, TxT+rMSC (IL1B) rats had significantly reduced neutrophil proportion in BAL compared to TxT rats 24 h after trauma. In contrast, TxT+hMSC (IL1B) rats had the lowest MPO content in lung at 24 h. In line with this observation, lung sections of TxT+hMSC and TxT+hMSC (IL1B) rats did not contain granulocyte accumulations 24 h post trauma. Therefore, reduction of neutrophil infiltration seemed to be a therapeutic mechanism of primed MSC. In contrast to previous publications [99,169,198] we did not observe significant effects of primed MSC compared to unprimed MSC in the rat TxT model. As already described, the in vitro experiments revealed that a second IL1B stimulation did not have the same strong effects on cytokine secretion of MSC than the first IL1B stimulation. Therefore, priming of MSC might reduce subsequent responsiveness to pro-inflammatory factors in vivo.

156

Discussion

The antagonization of chemokines as well as the reduction of neutrophil transmigration or function can contribute to the attenuation of organ damage in experimental organ failure after trauma and sepsis [75,190]. Allogeneic BM-MSC transplantation after bleomycin-induced lung injury in mice revealed a reduction of neutrophil proportion in BAL compared to injured mice without MSC administration [185]. Curley and colleagues also showed that human UC-MSC decreased neutrophil content in BAL in immunocompetent lung-injured rats compared to untreated lung-injured rats [47].

We did not observe significant effects on lung injury score between TxT rats and TxT+hMSC or TxT+hMSC (IL1B) rats 5 d after trauma, but there were significantly different effects 24 h after trauma. Furthermore, rMSC-treatment of TxT rats did not reduce lung injury score 24 h after trauma compared to TxT rats. In contrast to that, Hannoush and colleagues used a unilateral lung contusion rat model and the same rMSC dose but a different outbred rat strain and rMSC characterization marker and observed significant reduction of the lung injury score compared to rats with lung injury 5 d post injury, but not 24 h after lung contusion [98]. Gore and colleagues used a double hit model (lung injury accompanied by chronic stress or haemorrhagic shock) and the same dose of rMSC and observed significant therapeutic effects of rMSC 7 d after application of lung injury [89,90]. The different outcome might be caused by the use of different rat strains (Sprague Dawley versus Wistar rats). However, it might be that rMSC had late effect on lung injury score but in our opinion this effect would be irrelevant due to the fact that we did not observe significant effects between sham and TxT rats 5 d after trauma.

As already described in our publication [5], different outcomes between hMSC- treated rats and rMSC-treated rats could be due to different source of MSC (human or rat), different expansion protocols (GMP-compliant or standard laboratory protocol) and different expansion media supplement (PL or FBS). There are publications showing therapeutic effect of rMSC in lung injury. However, these publications used different rat strains [90,98], different lung injury models (unilateral [98], radiation-induced lung injury [114] and ventilator-induced lung injury [46]), different dosage [46] and different endpoints [46,90]. Therefore, we

157

Discussion concluded that for the first time we investigated the effect of allogeneic and xenogeneic MSC in a bilateral blast injury model with the main emphasis on the early inflammation response [5].

4.4.1. Mechanisms of therapeutic effects

Significant changes of local inflammatory reaction (cytokine level and neutrophil distribution) in MSC-treated animals revealed an impact of MSC on the rat immune system. These effects could be mediated by soluble factors, mitochondrial transfer and/or apoptotic bodies. Previous studies used hMSC [37,68], extracellular vesicles derived from hMSC [45,59] or human recombinant protein [37,76] in immunocompetent rodents and showed significant therapeutic effects in colitis [68], peritonitis [37] and lung injury model [76]. Fan and colleagues pre-treated human UC-MSC with murine IL1B for 48 h and observed increased therapeutic effects in colitis model compared to untreated UC-MSC [68]. Furthermore, it could be shown that human CXCL8 binds to murine CXCR1 [69]. Therefore, soluble factors might also be a therapeutic mechanism of xenogeneic MSC. The immunomodulatory effect of soluble factors derived from MSC is described in section 1.3.2.. Several publications revealed that the therapeutic effect of MSC in ALI models is mediated by soluble factors like TNFAIP6 [50], IL1RN [185] and fibroblast growth factor 7 [133] rather than by the engraftment of administered MSC. PKH+ events detected by flow cytometry and fluorescence microscopy did not carry MSC-specific markers. This might indicate the exchange of cell membrane from PKH26-labelled MSC to rat cells. Extracellular vesicles or mitochondrial transfer as therapeutic mechanism of MSC is also discussed in some studies dealing with ALI/ARDS. This enhanced bioenergetics in macrophages, increase their phagocytic activity and polarize them into alternatively activated macrophages [109,110,177]. This mechanism could be observed in animal studies using hMSC [110] or mouse MSC [109]. Furthermore, our results and results of Leibacher and colleagues revealed a fragmentation of hMSC [5,136]. Gonçalves and colleagues performed in vitro experiments and revealed immunomodulatory effect of membrane particles derived from human ASC by induction of selective apoptosis of pro-inflammatory monocytes [88]. Furthermore, they showed that

158

Discussion membrane particles derived from IFNG pre-treated MSC increased percentage of anti-inflammatory monocytes [88]. In contrast to rMSC, hMSC could not be detected 24 h after TxT. Nevertheless, hMSC seemed to reveal therapeutic effects on local inflammatory reaction leading to a significant impact on histological lung injury score. It was already mentioned that PKH+ particles might be apoptotic bodies. Galleu and colleagues depicted that apoptotic MSC therapeutically influenced the immune system in a mouse model of graft-versus-host disease and in in vitro experiments by the use of samples derived from patients with graft-versus-host disease [82]. They showed that activated cytotoxic cells were necessary for the induction of apoptosis in MSC. These were then phagocyted and induced IDO expression in phagocytes [82]. Xenogeneic MSC might be recognized by the immune system of the rat and might be phagocytosed by macrophages [7,93]. Seitz and colleagues induced TxT in Sprague-Dawley rats and Wistar rats and observed enhanced phagocytic activity of alveolar macrophages after TxT [147,227]. In vitro experiments revealed that phagocytosis of apoptotic cells by human macrophages [66] and rat alveolar macrophages [227] altered release of cytokines including the reduced release of IL1B. Furthermore, Liu and colleagues showed that apoptotic human UC-MSC had a therapeutic impact in a rat model of bleomycin-induced ALI [149]. Immune reaction of the host may play a role in the therapeutic effect of hMSC in the rat TxT model. Immune reaction of polytrauma patients receiving an allogeneic transplantation of these immunogenic MSC may also play a role in the clinic. In line with other publications we did not observe long-term engraftment of intact cells in rat TxT model. Human MSC may rather exert “touch-and-go” effects.

4.4.2. Clinical implications

The TxT model revealed some clinical aspects like inflammatory response, neutrophilic infiltration, endothelial and epithelial damage, hypoxia due to hemorrhage or oedema. We showed that these aspects could be therapeutically influenced by hMSC therapy. One aim of the study was to prove the safety and efficacy of hMSC in the TxT model. Three animals died before the end of the experiment. Two (TxT, TxT+rMSC) of them died immediately after trauma

159

Discussion induction. Only one TxT rat died with a time delay of less than 24 h. Therefore, we conclude that the injection of hMSC was safe and efficient. However, further pre-clinical studies in models closer to humans are needed. There are several differences in immune response of humans and rats [165,230]. As already mentioned above, several studies revealed that nitric oxide as mediator in inflammatory response plays a greater role in rodents than in humans [112,202,243]. In addition to the obvious differences between hMSC and rMSC, such as doubling time, we suggested that therapeutic mechanisms differed between hMSC and rMSC (see section 4.3 and 4.4.1). Furthermore, Matute-Bello and colleagues concluded that not all characteristics of ALI/ARDS in humans could be reproduced in a single animal model [165]. It is a challenging imperative to investigate mechanisms in ALI in more detail and to improve or to develop therapeutic strategies.

4.5. Conclusion

Several publications showed that priming increased therapeutic effect of MSC [68,249]. To prime MSC, they were often exposed to the environmental conditions of the disease or to single factors involved in disease progression [5,68,139,254]. We showed that IL1B had a major role in trauma cocktail. We stimulated MSC with IL1B, thereby enhancing the expression of adhesion molecules, MMPs, and increasing secretion of several cytokines and chemokines which might enhance the therapeutic effect. The rat TxT model revealed new insights into the therapeutic approach of MSC administration in case of lung injury. We observed significant therapeutic effects of xenogeneic and allogeneic MSC 24 h after TxT. These effects were partially increased by priming of MSC. In addition, we observed different effects and recovery rates of xenogeneic and allogeneic MSC. Furthermore, new insights in the fate of PKH26-labelled allogeneic and xenogeneic MSC 24 h and 5 d after trauma were shown. Nevertheless, some limitations of the study should be considered. Firstly, Wistar rats are an outbred strain, thus the heterogeneity of the genetic background might mask the effects on plasma and BAL cytokine levels. Some readout parameters of the sham group showed a very high inter-individual variation. Secondly, the rat

160

Discussion cytokines and chemokines were measured; however the human factors were not analysed. Only PGE2 and TNFAIP6 assay included the detection of human molecules. Thirdly, only one dosage, one route of administration and one time of administration of MSC were chosen. The optimal dosage and time of administration has yet to be found. Fourth, the presented study deals with the safety and efficacy of MSC in a pre-clinical model of ALI. The underlying mechanisms of therapeutic efficacy have still to be elucidated. Our results give hints to a therapeutic role of MSC in ALI. Future experiments will reveal the underlying mechanisms of BM-MSC and ASC cultured in PL in inflammatory diseases and organ damage.

161

Summary

5. Summary

Polytrauma is a life-threatening disease and is accompanied by release of several factors indicating the inflammatory reaction and tissue damage. Inflammatory reaction after trauma can cause late mortality. Thoracic injury influences recovery process of polytrauma patients. Allogeneic transplantation of mesenchymal stromal cells (MSC) might be a therapeutic option for polytrauma patients due to their multi-lineage differentiation potential, secretion of active molecules and formation of apoptotic bodies. These promote regeneration and prevent excessive inflammation. Previous publications showed that activation of MSC can enhance migratory abilities and immunomodulatory properties. One aim of this work was to investigate the effect of several single trauma factors and a mix of these factors on MSC in vitro by proliferation and migration assays, sequencing of ribonucleic acid, Multiplex assays, enzyme-linked immunosorbent assays (ELISA) and flow cytometry analysis. A second aim was to investigate a potential benefit of a single dose of systemically administered allogeneic and xenogeneic human bone marrow-derived MSC (hMSC) in a rat blunt chest trauma model 24 hours and 5 days after treatment. Rat MSC (rMSC) were isolated from Wistar rats and characterized by flow cytometry and differentiation assays. To investigate the fate of MSC in vivo they were labelled with the vital dye PKH26. Species-specific interleukin 1 beta (IL1B) priming for hMSC and rMSC was established since these primed cells were used in in vivo experiments. Co-culture experiments using the rat alveolar macrophage cell line NR8383 and rMSC or hMSC were performed as pilot tests for in vivo experiments. Methods of in vivo experiments included thorax trauma induced by a blast wave and the harvesting of blood and broncho-alveolar lavage fluid (BAL) for cell culture experiments, Multiplex assays and ELISA. Further methods included flow cytometry analysis and cell sorting, fluorescence microscopy, differential analysis, detection of Alu sequences, measurement of MPO activity in lung tissue and analysis of the histological score of the lung. No significant effect of high concentration of single trauma factors on proliferation of MSC was detected. For the first time it could be shown that MSC carry the thrombopoietin (THPO) receptor and that THPO influences MSC migration in a concentration-dependent manner. Gene expression analysis revealed a high intersection of differentially expressed genes in MSC stimulated with a mix of 162

Summary trauma factors and IL1B alone 6 and 24 hours after stimulation. Analysis of cell culture supernatant of IL1B-stimulated MSC showed that protein secretion is time- dependent (early factors like interleukin 6 (IL6), C-X-C motif chemokine ligand 1 (CXCL1); late factors like vascular endothelial growth factor A, tumor necrosis factor alpha inducible protein 6 (TNFAIP6), matrix metallopeptidase 1). Macrophages influenced the secretion of factors from MSC by soluble factors and cell-cell contact. Prior to the use of MSC in in vivo experiments it was proven that PKH26 labelling of MSC did not influence doubling time, colony forming units, surface marker expression and portion of dead cells. Animal experiments revealed that rMSC, but not hMSC could be recovered from rat blood and BAL. hMSC lost their characteristic phenotype (size, surface marker, plastic adherence) after injection into rats. Nevertheless, the lung injury score of traumatized rats receiving unprimed or IL1B-primed hMSC was at least 50 % reduced compared to traumatized rats without MSC treatment 24 hours post trauma. This reduction might partially be due to a reduction of the number of granulocytes in lung tissue. In general, treatment of traumatized rats with unprimed or IL1B-primed hMSC reduced pro-inflammatory cytokines (IL1B, IL6, CXCL1) and total cell count in BAL compared to traumatized rats without MSC treatment at 24 hours. The treatment of traumatized rats with rMSC did not reveal such therapeutic effects on the lung injury score and cytokine levels, however significantly increased TNFAIP6 level and significantly reduced neutrophil counts in BAL compared to traumatized rats without MSC treatment at 24 hours. Different outcomes of allogeneic and xenogeneic transplantation could be due to species-specific differences, different expansion protocols and medium supplements. Furthermore, differences to some of the literature might lie in different sources of MSC, passage number of MSC, stimulation protocols, trauma model, analysis methods or timing. hMSC might be recognized by the immune system of the rat. The therapeutic effect of hMSC in the rat blunt chest trauma model is rather a “touch-and-go” effect than a long-term engraftment of intact cells. A positive effect on healing by the immune response of the recipient to the transplant is currently discussed. Future studies will have to show the underlying therapeutic mechanism of administered MSC after trauma and further investigate species-specific mechanisms, optimal dose and route of administration.

163

Reference List

6. Reference List

1. Akers JC, Gonda D, Kim R, Carter BS, Chen CC (2013) Biogenesis of extracellular vesicles (EV): exosomes, microvesicles, retrovirus-like vesicles, and apoptotic bodies. J Neurooncol 113: 1-11. 2. Aldridge V, Garg A, Davies N, Bartlett DC, Youster J, Beard H, Kavanagh DP, Kalia N, Frampton J, Lalor PF, Newsome PN (2012) Human mesenchymal stem cells are recruited to injured liver in a beta1-integrin and CD44 dependent manner. Hepatology 56: 1063-1073. 3. AMA Committee on Medical Aspects of Automotive Safety (1971) Rating the severity of tissue damage. I. The abbreviated scale. JAMA 215: 277-280. 4. Amann EM, Gross A, Rojewski MT, Kestler HA, Kalbitz M, Brenner RE, Huber-Lang M, Schrezenmeier H (2019) Inflammatory response of mesenchymal stromal cells after in vivo exposure with selected trauma-related factors and polytrauma serum. PLoS One 14: e0216862. 5. Amann EM, Rojewski MT, Rodi S, Furst D, Fiedler J, Palmer A, Braumuller S, Huber-Lang M, Schrezenmeier H, Brenner RE (2018) Systemic recovery and therapeutic effects of transplanted allogenic and xenogenic mesenchymal stromal cells in a rat blunt chest trauma model. Cytotherapy 20: 218-231. 6. Andreeva E, Bobyleva P, Gornostaeva A, Buravkova L (2017) Interaction of multipotent mesenchymal stromal and immune cells: Bidirectional effects. Cytotherapy 19: 1152-1166. 7. Arutyunyan I, Elchaninov A, Fatkhudinov T, Makarov A, Kananykhina E, Usman N, Bolshakova G, Glinkina V, Goldshtein D, Sukhikh G (2015) Elimination of allogeneic multipotent stromal cells by host macrophages in different models of regeneration. Int J Clin Exp Pathol 8: 4469-4480. 8. Badillo AT, Beggs KJ, Javazon EH, Tebbets JC, Flake AW (2007) Murine bone marrow stromal progenitor cells elicit an in vivo cellular and humoral alloimmune response. Biol Blood Marrow Transplant 13: 412-422. 9. Baker SP, O'Neill B, Haddon W, Jr., Long WB (1974) The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 14: 187- 196. 10. Bartunek J, Behfar A, Dolatabadi D, Vanderheyden M, Ostojic M, Dens J, El NB, Banovic M, Beleslin B, Vrolix M, Legrand V, Vrints C, Vanoverschelde JL, Crespo-Diaz R, Homsy C, Tendera M, Waldman S, Wijns W, Terzic A (2013) Cardiopoietic stem cell therapy in heart failure: the C-CURE (Cardiopoietic stem Cell therapy in heart failURE) multicenter randomized trial with lineage-specified biologics. J Am Coll Cardiol 61: 2329-2338. 11. Barzilay R, Sadan O, Melamed E, Offen D (2009) Comparative characterization of bone marrow-derived mesenchymal stromal cells from four different rat strains. Cytotherapy 11: 435-442. 12. Battle CE, Hutchings H, Evans PA (2012) Risk factors that predict mortality in patients with blunt chest wall trauma: a systematic review and meta-analysis. Injury 43: 8-17.

164

Reference List

13. Bayer J, Lefering R, Reinhardt S, Kuhle J, Zwingmann J, Sudkamp NP, Hammer T (2017) Thoracic trauma severity contributes to differences in intensive care therapy and mortality of severely injured patients: analysis based on the TraumaRegister DGU(R). World J Emerg Surg 12: 43. 14. Bein T, Grasso S, Moerer O, Quintel M, Guerin C, Deja M, Brondani A, Mehta S (2016) The standard of care of patients with ARDS: ventilatory settings and rescue therapies for refractory hypoxemia. Intensive Care Med 42: 699-711. 15. Belema-Bedada F, Uchida S, Martire A, Kostin S, Braun T (2008) Efficient homing of multipotent adult mesenchymal stem cells depends on FROUNT-mediated clustering of CCR2. Cell Stem Cell 2: 566-575. 16. Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban A, Gattinoni L, van HF, Larsson A, McAuley DF, Ranieri M, Rubenfeld G, Thompson BT, Wrigge H, Slutsky AS, Pesenti A (2016) Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA 315: 788-800. 17. Bernard GR, Artigas A (2016) The definition of ARDS revisited: 20 years later. Intensive Care Med 42: 640-642. 18. Bieback K, Hecker A, Kocaomer A, Lannert H, Schallmoser K, Strunk D, Kluter H (2009) Human alternatives to fetal bovine serum for the expansion of mesenchymal stromal cells from bone marrow. Stem Cells 27: 2331-2341. 19. Bigildeev AE, Zezina EA, Shipounova IN, Drize NJ (2015) Interleukin-1 beta enhances human multipotent mesenchymal stromal cell proliferative potential and their ability to maintain hematopoietic precursor cells. Cytokine 71: 246-254. 20. Bouffi C, Bony C, Courties G, Jorgensen C, Noel D (2010) IL-6-dependent PGE2 secretion by mesenchymal stem cells inhibits local inflammation in experimental arthritis. PLoS One 5: e14247. 21. Bouros D, Alexandrakis MG, Antoniou KM, Agouridakis P, Pneumatikos I, Anevlavis S, Pataka A, Patlakas G, Karkavitsas N, Kyriakou D (2004) The clinical significance of serum and bronchoalveolar lavage inflammatory cytokines in patients at risk for Acute Respiratory Distress Syndrome. BMC Pulm Med 4: 6. 22. Brandau S, Jakob M, Hemeda H, Bruderek K, Janeschik S, Bootz F, Lang S (2010) Tissue- resident mesenchymal stem cells attract peripheral blood neutrophils and enhance their inflammatory activity in response to microbial challenge. J Leukoc Biol 88: 1005-1015. 23. Brennan MA, Renaud A, Amiaud J, Rojewski MT, Schrezenmeier H, Heymann D, Trichet V, Layrolle P (2014) Pre-clinical studies of bone regeneration with human bone marrow stromal cells and biphasic calcium phosphate. Stem Cell Res Ther 5: 114. 24. Brizzi MF, Battaglia E, Montrucchio G, Dentelli P, Del SL, Garbarino G, Pegoraro L, Camussi G (1999) Thrombopoietin stimulates endothelial cell motility and neoangiogenesis by a platelet-activating factor-dependent mechanism. Circ Res 84: 785-796. 25. Brizzi MF, Battaglia E, Rosso A, Strippoli P, Montrucchio G, Camussi G, Pegoraro L (1997) Regulation of polymorphonuclear cell activation by thrombopoietin. J Clin Invest 99: 1576- 1584. 165

Reference List

26. Burk AM, Martin M, Flierl MA, Rittirsch D, Helm M, Lampl L, Bruckner U, Stahl GL, Blom AM, Perl M, Gebhard F, Huber-Lang M (2012) Early complementopathy after multiple injuries in humans. Shock 37: 348-354. 27. Caplan AI (2007) Adult mesenchymal stem cells for tissue engineering versus regenerative medicine. J Cell Physiol 213: 341-347. 28. Carrero R, Cerrada I, Lledo E, Dopazo J, Garcia-Garcia F, Rubio MP, Trigueros C, Dorronsoro A, Ruiz-Sauri A, Montero JA, Sepulveda P (2012) IL1beta induces mesenchymal stem cells migration and leucocyte chemotaxis through NF-kappaB. Stem Cell Rev 8: 905- 916. 29. Carvalho JL, Braga VB, Melo MB, Campos AC, Oliveira MS, Gomes DA, Ferreira AJ, Santos RA, Goes AM (2013) Priming mesenchymal stem cells boosts stem cell therapy to treat myocardial infarction. J Cell Mol Med 17: 617-625. 30. Cassatella MA, Mosna F, Micheletti A, Lisi V, Tamassia N, Cont C, Calzetti F, Pelletier M, Pizzolo G, Krampera M (2011) Toll-like receptor-3-activated human mesenchymal stromal cells significantly prolong the survival and function of neutrophils. Stem Cells 29: 1001-1011. 31. Chai HX, Cheng FJ, Liu QH, Tang JM, Yang JY, Wang JN (2008) Effects of different doses of thrombopoietin on proliferation of bone marrow mesenchymal stem cells in mice. Zhongguo Shi Yan Xue Ye Xue Za Zhi 16: 859-862. 32. Chai HX, Wang F, Cheng FJ, Luo Q, Yao H, Zheng S (2016) Effect of Thrombopoietin on the Mobilization of Bone Marrow Mesenchymal Stem Cells. Military Medical Journal of South China 149-151. 33. Chan JL, Tang KC, Patel AP, Bonilla LM, Pierobon N, Ponzio NM, Rameshwar P (2006) Antigen-presenting property of mesenchymal stem cells occurs during a narrow window at low levels of interferon-gamma. Blood 107: 4817-4824. 34. Chan WK, Lau AS, Li JC, Law HK, Lau YL, Chan GC (2008) MHC expression kinetics and immunogenicity of mesenchymal stromal cells after short-term IFN-gamma challenge. Exp Hematol 36: 1545-1555. 35. Chen Y, Xiang LX, Shao JZ, Pan RL, Wang YX, Dong XJ, Zhang GR (2010) Recruitment of endogenous bone marrow mesenchymal stem cells towards injured liver. J Cell Mol Med 14: 1494-1508. 36. Cheng L, Qasba P, Vanguri P, Thiede MA (2000) Human mesenchymal stem cells support megakaryocyte and pro-platelet formation from CD34(+) hematopoietic progenitor cells. J Cell Physiol 184: 58-69. 37. Choi H, Lee RH, Bazhanov N, Oh JY, Prockop DJ (2011) Anti-inflammatory protein TSG-6 secreted by activated MSCs attenuates zymosan-induced mouse peritonitis by decreasing TLR2/NF-kappaB signaling in resident macrophages. Blood 118: 330-338. 38. Chrysou K, Halat G, Hoksch B, Schmid RA, Kocher GJ (2017) Lessons from a large trauma center: impact of blunt chest trauma in polytrauma patients-still a relevant problem? Scand J Trauma Resusc Emerg Med 25: 42.

166

Reference List

39. Cohen MJ, Brohi K, Calfee CS, Rahn P, Chesebro BB, Christiaans SC, Carles M, Howard M, Pittet JF (2009) Early release of high mobility group box nuclear protein 1 after severe trauma in humans: role of injury severity and tissue hypoperfusion. Crit Care 13: R174. 40. Cohn SM, Dubose JJ (2010) Pulmonary contusion: an update on recent advances in clinical management. World J Surg 34: 1959-1970. 41. Cohnheim J (1867) Ueber Entzuendung und Eiterung. J Arch Path Anat Physiol Klin Med 40: 1-79. 42. Colter DC, Sekiya I, Prockop DJ (2001) Identification of a subpopulation of rapidly self- renewing and multipotential adult stem cells in colonies of human marrow stromal cells. Proc Natl Acad Sci U S A 98: 7841-7845. 43. Crapo JD, Barry BE, Gehr P, Bachofen M, Weibel ER (1982) Cell number and cell characteristics of the normal human lung. Am Rev Respir Dis 126: 332-337. 44. Crisan M, Yap S, Casteilla L, Chen CW, Corselli M, Park TS, Andriolo G, Sun B, Zheng B, Zhang L, Norotte C, Teng PN, Traas J, Schugar R, Deasy BM, Badylak S, Buhring HJ, Giacobino JP, Lazzari L, Huard J, Peault B (2008) A perivascular origin for mesenchymal stem cells in multiple human organs. Cell Stem Cell 3: 301-313. 45. Cruz FF, Borg ZD, Goodwin M, Sokocevic D, Wagner DE, Coffey A, Antunes M, Robinson KL, Mitsialis SA, Kourembanas S, Thane K, Hoffman AM, McKenna DH, Rocco PR, Weiss DJ (2015) Systemic Administration of Human Bone Marrow-Derived Mesenchymal Stromal Cell Extracellular Vesicles Ameliorates Aspergillus Hyphal Extract-Induced Allergic Airway Inflammation in Immunocompetent Mice. Stem Cells Transl Med 4: 1302-1316. 46. Curley GF, Hayes M, Ansari B, Shaw G, Ryan A, Barry F, O'Brien T, O'Toole D, Laffey JG (2012) Mesenchymal stem cells enhance recovery and repair following ventilator-induced lung injury in the rat. Thorax 67: 496-501. 47. Curley GF, Jerkic M, Dixon S, Hogan G, Masterson C, O'Toole D, Devaney J, Laffey JG (2017) Cryopreserved, Xeno-Free Human Umbilical Cord Mesenchymal Stromal Cells Reduce Lung Injury Severity and Bacterial Burden in Rodent Escherichia coli-Induced Acute Respiratory Distress Syndrome. Crit Care Med 45: e202-e212. 48. Cyagen Biosciences (2018) Wistar Rat Mesenchymal Stem Cells. https://www.cyagen.com/us/en/product/wistar-mesenchymal-stem-cells.html. Access Date: 18-2-2018. 49. da Silva ML, Chagastelles PC, Nardi NB (2006) Mesenchymal stem cells reside in virtually all post-natal organs and tissues. J Cell Sci 119: 2204-2213. 50. Danchuk S, Ylostalo JH, Hossain F, Sorge R, Ramsey A, Bonvillain RW, Lasky JA, Bunnell BA, Welsh DA, Prockop DJ, Sullivan DE (2011) Human multipotent stromal cells attenuate lipopolysaccharide-induced acute lung injury in mice via secretion of tumor necrosis factor- alpha-induced protein 6. Stem Cell Res Ther 2: 27. 51. de Oliveira LF, Almeida TR, Ribeiro Machado MP, Cuba MB, Alves AC, da Silva MV, Rodrigues J, V, Dias da Silva VJ (2015) Priming Mesenchymal Stem Cells with Endothelial Growth Medium Boosts Stem Cell Therapy for Systemic Arterial Hypertension. Stem Cells Int 2015: 685383. 167

Reference List

52. De BA, Van HP, Bakkus M, Vande B, I, De WJ, De WM, Van R, I (2007) Migration of culture- expanded human mesenchymal stem cells through bone marrow endothelium is regulated by matrix metalloproteinase-2 and tissue inhibitor of metalloproteinase-3. Haematologica 92: 440-449. 53. Deak E, Ruster B, Keller L, Eckert K, Fichtner I, Seifried E, Henschler R (2010) Suspension medium influences interaction of mesenchymal stromal cells with endothelium and pulmonary toxicity after transplantation in mice. Cytotherapy 12: 260-264. 54. Dekker AB, Krijnen P, Schipper IB (2016) Predictive value of cytokines for developing complications after polytrauma. World J Crit Care Med 5: 187-200. 55. Deng Y, Yang Z, Gao Y, Xu H, Zheng B, Jiang M, Xu J, He Z, Wang X (2013) Toll-like receptor 4 mediates acute lung injury induced by high mobility group box-1. PLoS One 8: e64375. 56. Denk S, Taylor RP, Wiegner R, Cook EM, Lindorfer MA, Pfeiffer K, Paschke S, Eiseler T, Weiss M, Barth E, Lambris JD, Kalbitz M, Martin T, Barth H, Messerer DAC, Gebhard F, Huber-Lang MS (2017) Complement C5a-Induced Changes in Neutrophil Morphology During Inflammation. Scand J Immunol 86: 143-155. 57. Dolgachev VA, Yu B, Reinke JM, Raghavendran K, Hemmila MR (2012) Host susceptibility to gram-negative pneumonia after lung contusion. J Trauma Acute Care Surg 72: 614-622. 58. Dominici M, Le BK, Mueller I, Slaper-Cortenbach I, Marini F, Krause D, Deans R, Keating A, Prockop D, Horwitz E (2006) Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 8: 315- 317. 59. Drommelschmidt K, Serdar M, Bendix I, Herz J, Bertling F, Prager S, Keller M, Ludwig AK, Duhan V, Radtke S, de MK, Horn PA, van de Looij Y, Giebel B, Felderhoff-Muser U (2017) Mesenchymal stem cell-derived extracellular vesicles ameliorate inflammation-induced preterm brain injury. Brain Behav Immun 60: 220-232. 60. Dwyer RM, Potter-Beirne SM, Harrington KA, Lowery AJ, Hennessy E, Murphy JM, Barry FP, O'Brien T, Kerin MJ (2007) Monocyte chemotactic protein-1 secreted by primary breast tumors stimulates migration of mesenchymal stem cells. Clin Cancer Res 13: 5020-5027. 61. Ehrnthaller C, Flierl M, Perl M, Denk S, Unnewehr H, Ward PA, Radermacher P, Ignatius A, Gebhard F, Chinnaiyan A, Huber-Lang M (2015) The molecular fingerprint of lung inflammation after blunt chest trauma. Eur J Med Res 20: 70. 62. Eliopoulos N, Stagg J, Lejeune L, Pommey S, Galipeau J (2005) Allogeneic marrow stromal cells are immune rejected by MHC class I- and class II-mismatched recipient mice. Blood 106: 4057-4065. 63. Erices AA, Allers CI, Conget PA, Rojas CV, Minguell JJ (2003) Human cord blood-derived mesenchymal stem cells home and survive in the marrow of immunodeficient mice after systemic infusion. Cell Transplant 12: 555-561. 64. Escobar P, Bouclier C, Serret J, Bieche I, Brigitte M, Caicedo A, Sanchez E, Vacher S, Vignais ML, Bourin P, Genevieve D, Molina F, Jorgensen C, Lazennec G (2015) IL-1beta

168

Reference List

produced by aggressive breast cancer cells is one of the factors that dictate their interactions with mesenchymal stem cells through chemokine production. Oncotarget 6: 29034-29047. 65. Espig, Sandy (5-2-0016) Isolation and characterization of rat bone-marrow derived mesenchymal stromal cells [dissertation]. Dissertation of the Medical Faculty Ulm University. 66. Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY, Henson PM (1998) Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J Clin Invest 101: 890-898. 67. Fajardo-Orduna GR, Mayani H, Montesinos JJ (2015) Hematopoietic Support Capacity of Mesenchymal Stem Cells: Biology and Clinical Potential. Arch Med Res 46: 589-596. 68. Fan H, Zhao G, Liu L, Liu F, Gong W, Liu X, Yang L, Wang J, Hou Y (2012) Pre-treatment with IL-1beta enhances the efficacy of MSC transplantation in DSS-induced colitis. Cell Mol Immunol 9: 473-481. 69. Fan X, Patera AC, Pong-Kennedy A, Deno G, Gonsiorek W, Manfra DJ, Vassileva G, Zeng M, Jackson C, Sullivan L, Sharif-Rodriguez W, Opdenakker G, Van DJ, Hedrick JA, Lundell D, Lira SA, Hipkin RW (2007) Murine CXCR1 is a functional receptor for GCP-2/CXCL6 and interleukin-8/CXCL8. J Biol Chem 282: 11658-11666. 70. Fekete N, Gadelorge M, Furst D, Maurer C, Dausend J, Fleury-Cappellesso S, Mailander V, Lotfi R, Ignatius A, Sensebe L, Bourin P, Schrezenmeier H, Rojewski MT (2012) Platelet lysate from whole blood-derived pooled platelet concentrates and apheresis-derived platelet concentrates for the isolation and expansion of human bone marrow mesenchymal stromal cells: production process, content and identification of active components. Cytotherapy 14: 540-554. 71. Fekete N, Rojewski MT, Furst D, Kreja L, Ignatius A, Dausend J, Schrezenmeier H (2012) GMP-compliant isolation and large-scale expansion of bone marrow-derived MSC. PLoS One 7: e43255. 72. Fekete N, Rojewski MT, Lotfi R, Schrezenmeier H (2014) Essential components for ex vivo proliferation of mesenchymal stromal cells. Tissue Eng Part C Methods 20: 129-139. 73. Fiedler J, Etzel N, Brenner RE (2004) To go or not to go: Migration of human mesenchymal progenitor cells stimulated by isoforms of PDGF. J Cell Biochem 93: 990-998. 74. Fiedler J, Leucht F, Waltenberger J, Dehio C, Brenner RE (2005) VEGF-A and PlGF-1 stimulate chemotactic migration of human mesenchymal progenitor cells. Biochem Biophys Res Commun 334: 561-568. 75. Flierl MA, Perl M, Rittirsch D, Bartl C, Schreiber H, Fleig V, Schlaf G, Liener U, Brueckner UB, Gebhard F, Huber-Lang MS (2008) The role of C5a in the innate immune response after experimental blunt chest trauma. Shock 29: 25-31. 76. Foskett AM, Bazhanov N, Ti X, Tiblow A, Bartosh TJ, Prockop DJ (2014) Phase-directed therapy: TSG-6 targeted to early inflammation improves bleomycin-injured lungs. Am J Physiol Lung Cell Mol Physiol 306: L120-L131.

169

Reference List

77. Francois M, Romieu-Mourez R, Li M, Galipeau J (2012) Human MSC suppression correlates with cytokine induction of indoleamine 2,3-dioxygenase and bystander M2 macrophage differentiation. Mol Ther 20: 187-195. 78. Friedenstein AJ, Chailakhjan RK, Lalykina KS (1970) The development of fibroblast colonies in monolayer cultures of guinea-pig bone marrow and spleen cells. Cell Tissue Kinet 3: 393- 403. 79. Fu X, Han B, Cai S, Lei Y, Sun T, Sheng Z (2009) Migration of bone marrow-derived mesenchymal stem cells induced by tumor necrosis factor-alpha and its possible role in wound healing. Wound Repair Regen 17: 185-191. 80. Furtmuller PG, Burner U, Obinger C (1998) Reaction of myeloperoxidase compound I with chloride, bromide, iodide, and thiocyanate. Biochemistry 37: 17923-17930. 81. Galipeau J, Krampera M, Barrett J, Dazzi F, Deans RJ, DeBruijn J, Dominici M, Fibbe WE, Gee AP, Gimble JM, Hematti P, Koh MB, LeBlanc K, Martin I, McNiece IK, Mendicino M, Oh S, Ortiz L, Phinney DG, Planat V, Shi Y, Stroncek DF, Viswanathan S, Weiss DJ, Sensebe L (2016) International Society for Cellular Therapy perspective on immune functional assays for mesenchymal stromal cells as potency release criterion for advanced phase clinical trials. Cytotherapy 18: 151-159. 82. Galleu A, Riffo-Vasquez Y, Trento C, Lomas C, Dolcetti L, Cheung TS, von BM, Barbieri L, Halai K, Ward S, Weng L, Chakraverty R, Lombardi G, Watt FM, Orchard K, Marks DI, Apperley J, Bornhauser M, Walczak H, Bennett C, Dazzi F (2017) Apoptosis in mesenchymal stromal cells induces in vivo recipient-mediated immunomodulation. Sci Transl Med 9. 83. Ganie FA, Lone H, Lone GN, Wani ML, Singh S, Dar AM, Wani NU, Wani SN, Nazeer NU (2013) Lung Contusion: A Clinico-Pathological Entity with Unpredictable Clinical Course. Bull Emerg Trauma 1: 7-16. 84. Gebhard F, Huber-Lang M (2008) Polytrauma--pathophysiology and management principles. Langenbecks Arch Surg 393: 825-831. 85. German Trauma Society (DGU) (2017) TraumaRegister DGU® - Annual Report 2017. Sektion NIS of the German Trauma Society (DGU) / AUC -Akademie der Unfallchirurgie GmbH. Available: http://www.traumaregister-dgu.de/fileadmin/user_upload/traumaregister- dgu.de/docs/Downloads/TR-DGU_Annual_Report_2017.pdf. Access Date: 10-2-2018. 86. Gillooly JF, Hein A, Damiani R (2015) Nuclear DNA Content Varies with Cell Size across Human Cell Types. Cold Spring Harb Perspect Biol 7: a019091. 87. Gomez-Barrena E, Rosset P, Gebhard F, Hernigou P, Baldini N, Rouard H, Sensebe L, Gonzalo-Daganzo RM, Giordano R, Padilla-Eguiluz N, Garcia-Rey E, Cordero-Ampuero J, Rubio-Suarez JC, Stanovici J, Ehrnthaller C, Huber-Lang M, Flouzat-Lachaniette CH, Chevallier N, Donati DM, Ciapetti G, Fleury S, Fernandez MN, Cabrera JR, Avendano-Sola C, Montemurro T, Panaitescu C, Veronesi E, Rojewski MT, Lotfi R, Dominici M, Schrezenmeier H, Layrolle P (2018) Feasibility and safety of treating non-unions in tibia, femur and humerus with autologous, expanded, bone marrow-derived mesenchymal stromal

170

Reference List

cells associated with biphasic calcium phosphate biomaterials in a multicentric, non- comparative trial. Biomaterials . 88. Goncalves FDC, Luk F, Korevaar SS, Bouzid R, Paz AH, Lopez-Iglesias C, Baan CC, Merino A, Hoogduijn MJ (2017) Membrane particles generated from mesenchymal stromal cells modulate immune responses by selective targeting of pro-inflammatory monocytes. Sci Rep 7: 12100. 89. Gore AV, Bible LE, Livingston DH, Mohr AM, Sifri ZC (2015) Can mesenchymal stem cells reverse chronic stress-induced impairment of lung healing following traumatic injury? J Trauma Acute Care Surg 78: 767-772. 90. Gore AV, Bible LE, Song K, Livingston DH, Mohr AM, Sifri ZC (2015) Mesenchymal stem cells increase T-regulatory cells and improve healing following trauma and hemorrhagic shock. J Trauma Acute Care Surg 79: 48-52. 91. Greene KE, Wright JR, Steinberg KP, Ruzinski JT, Caldwell E, Wong WB, Hull W, Whitsett JA, Akino T, Kuroki Y, Nagae H, Hudson LD, Martin TR (1999) Serial changes in surfactant- associated proteins in lung and serum before and after onset of ARDS. Am J Respir Crit Care Med 160: 1843-1850. 92. Greenspan L, McLellan BA, Greig H (1985) Abbreviated Injury Scale and Injury Severity Score: a scoring chart. J Trauma 25: 60-64. 93. Grinnemo KH, Mansson A, Dellgren G, Klingberg D, Wardell E, Drvota V, Tammik C, Holgersson J, Ringden O, Sylven C, Le BK (2004) Xenoreactivity and engraftment of human mesenchymal stem cells transplanted into infarcted rat myocardium. J Thorac Cardiovasc Surg 127: 1293-1300. 94. Guimaraes-Camboa N, Cattaneo P, Sun Y, Moore-Morris T, Gu Y, Dalton ND, Rockenstein E, Masliah E, Peterson KL, Stallcup WB, Chen J, Evans SM (2017) Pericytes of Multiple Organs Do Not Behave as Mesenchymal Stem Cells In Vivo. Cell Stem Cell 20: 345-359. 95. Guo J, Zhang H, Xia J, Hou J, Wang Y, Yang T, Wang S, Zhang X, Chen X, Wu X (2018) Interleukin-1beta induces intercellular adhesion molecule-1 expression, thus enhancing the adhesion between mesenchymal stem cells and endothelial progenitor cells via the p38 MAPK signaling pathway. Int J Mol Med 41: 1976-1982. 96. Haagsma JA, Graetz N, Bolliger I, Naghavi M, Higashi H, Mullany EC, Abera SF, Abraham JP, Adofo K, Alsharif U, Ameh EA, Ammar W, Antonio CA, Barrero LH, Bekele T, Bose D, Brazinova A, Catala-Lopez F, Dandona L, Dandona R, Dargan PI, De LD, Degenhardt L, Derrett S, Dharmaratne SD, Driscoll TR, Duan L, Petrovich ES, Farzadfar F, Feigin VL, Franklin RC, Gabbe B, Gosselin RA, Hafezi-Nejad N, Hamadeh RR, Hijar M, Hu G, Jayaraman SP, Jiang G, Khader YS, Khan EA, Krishnaswami S, Kulkarni C, Lecky FE, Leung R, Lunevicius R, Lyons RA, Majdan M, Mason-Jones AJ, Matzopoulos R, Meaney PA, Mekonnen W, Miller TR, Mock CN, Norman RE, Orozco R, Polinder S, Pourmalek F, Rahimi-Movaghar V, Refaat A, Rojas-Rueda D, Roy N, Schwebel DC, Shaheen A, Shahraz S, Skirbekk V, Soreide K, Soshnikov S, Stein DJ, Sykes BL, Tabb KM, Temesgen AM, Tenkorang EY, Theadom AM, Tran BX, Vasankari TJ, Vavilala MS, Vlassov VV, Woldeyohannes SM, Yip P, Yonemoto N, Younis MZ, Yu C, Murray CJ, Vos T (2016) The 171

Reference List

global burden of injury: incidence, mortality, disability-adjusted life years and time trends from the Global Burden of Disease study 2013. Inj Prev 22: 3-18. 97. Han S, Mallampalli RK (2015) The acute respiratory distress syndrome: from mechanism to translation. J Immunol 194: 855-860. 98. Hannoush EJ, Sifri ZC, Elhassan IO, Mohr AM, Alzate WD, Offin M, Livingston DH (2011) Impact of enhanced mobilization of bone marrow derived cells to site of injury. J Trauma 71: 283-289. 99. Hengartner NE, Fiedler J, Schrezenmeier H, Huber-Lang M, Brenner RE (2015) Crucial role of IL1beta and C3a in the in vitro-response of multipotent mesenchymal stromal cells to inflammatory mediators of polytrauma. PLoS One 10: e0116772. 100. Hengartner, Nina Emily (17-11-2015) Immunomodulatory effects and migratory activity of multipotent mesenchymal stromal cells in the posttraumatic inflammatory response [dissertation]. Medical Faculty Ulm University. 101. Hobisch-Hagen P, Jelkmann W, Mayr A, Wiedermann FJ, Fries D, Herold M, Klingler A, Schobersberger W (2000) Low platelet count and elevated serum thrombopoietin after severe trauma. Eur J Haematol 64: 157-163. 102. Hoffman E, Patel A, Ball D, Klapwijk J, Millar V, Kumar A, Martin A, Mahendran R, Dailey LA, Forbes B, Hutter V (2017) Morphometric Characterization of Rat and Human Alveolar Macrophage Cell Models and their Response to Amiodarone using High Content Image Analysis. Pharm Res 34: 2466-2476. 103. Hoogduijn MJ, Verstegen MM, Engela AU, Korevaar SS, Roemeling-van RM, Merino A, Franquesa M, de JJ, Ijzermans JN, Weimar W, Betjes MG, Baan CC, van der Laan LJ (2014) No evidence for circulating mesenchymal stem cells in patients with organ injury. Stem Cells Dev 23: 2328-2335. 104. Horwitz EM, Gordon PL, Koo WK, Marx JC, Neel MD, McNall RY, Muul L, Hofmann T (2002) Isolated allogeneic bone marrow-derived mesenchymal cells engraft and stimulate growth in children with osteogenesis imperfecta: Implications for cell therapy of bone. Proc Natl Acad Sci U S A 99: 8932-8937. 105. Horwitz EM, Le BK, Dominici M, Mueller I, Slaper-Cortenbach I, Marini FC, Deans RJ, Krause DS, Keating A (2005) Clarification of the nomenclature for MSC: The International Society for Cellular Therapy position statement. Cytotherapy 7: 393-395. 106. Huber-Lang M, Wiegner R, Lampl L, Brenner RE (2016) Mesenchymal Stem Cells after Polytrauma: Actor and Target. Stem Cells Int 2016: 6289825. 107. Husecken Y, Muche S, Kustermann M, Klingspor M, Palmer A, Braumuller S, Huber-Lang M, Debatin KM, Strauss G (2017) MDSCs are induced after experimental blunt chest trauma and subsequently alter antigen-specific T cell responses. Sci Rep 7: 12808. 108. Irwin RJ, Lerner MR, Bealer JF, Brackett DJ, Tuggle DW (1997) Cardiopulmonary physiology of primary blast injury. J Trauma 43: 650-655. 109. Islam MN, Das SR, Emin MT, Wei M, Sun L, Westphalen K, Rowlands DJ, Quadri SK, Bhattacharya S, Bhattacharya J (2012) Mitochondrial transfer from bone-marrow-derived stromal cells to pulmonary alveoli protects against acute lung injury. Nat Med 18: 759-765. 172

Reference List

110. Jackson MV, Morrison TJ, Doherty DF, McAuley DF, Matthay MA, Kissenpfennig A, O'Kane CM, Krasnodembskaya AD (2016) Mitochondrial Transfer via Tunneling Nanotubes is an Important Mechanism by Which Mesenchymal Stem Cells Enhance Macrophage Phagocytosis in the In Vitro and In Vivo Models of ARDS. Stem Cells 34: 2210-2223. 111. Jaffin JH, McKinney L, Kinney RC, Cunningham JA, Moritz DM, Kraimer JM, Graeber GM, Moe JB, Salander JM, Harmon JW (1987) A laboratory model for studying blast overpressure injury. J Trauma 27: 349-356. 112. Jesch NK, Dorger M, Enders G, Rieder G, Vogelmeier C, Messmer K, Krombach F (1997) Expression of inducible nitric oxide synthase and formation of nitric oxide by alveolar macrophages: an interspecies comparison. Environ Health Perspect 105 Suppl 5: 1297- 1300. 113. Jiang D, Muschhammer J, Qi Y, Kugler A, de Vries JC, Saffarzadeh M, Sindrilaru A, Beken SV, Wlaschek M, Kluth MA, Ganss C, Frank NY, Frank MH, Preissner KT, Scharffetter- Kochanek K (2016) Suppression of Neutrophil-Mediated Tissue Damage-A Novel Skill of Mesenchymal Stem Cells. Stem Cells 34: 2393-2406. 114. Jiang X, Jiang X, Qu C, Chang P, Zhang C, Qu Y, Liu Y (2015) Intravenous delivery of adipose-derived mesenchymal stromal cells attenuates acute radiation-induced lung injury in rats. Cytotherapy 17: 560-570. 115. Jin HJ, Bae YK, Kim M, Kwon SJ, Jeon HB, Choi SJ, Kim SW, Yang YS, Oh W, Chang JW (2013) Comparative analysis of human mesenchymal stem cells from bone marrow, adipose tissue, and umbilical cord blood as sources of cell therapy. Int J Mol Sci 14: 17986-18001. 116. Joos H, Albrecht W, Laufer S, Reichel H, Brenner RE (2008) IL-1beta regulates FHL2 and other cytoskeleton-related genes in human chondrocytes. Mol Med 14: 150-159. 117. Kalbitz M, Amann EM, Bosch B, Palmer A, Schultze A, Pressmar J, Weber B, Wepler M, Gebhard F, Schrezenmeier H, Brenner R, Huber-Lang M (2017) Experimental blunt chest trauma-induced myocardial inflammation and alteration of gap-junction protein connexin 43. PLoS One 12: e0187270. 118. Keel M, Trentz O (2005) Pathophysiology of polytrauma. Injury 36: 691-709. 119. Kidd S, Spaeth E, Dembinski JL, Dietrich M, Watson K, Klopp A, Battula VL, Weil M, Andreeff M, Marini FC (2009) Direct evidence of mesenchymal stem cell tropism for tumor and wounding microenvironments using in vivo bioluminescent imaging. Stem Cells 27: 2614-2623. 120. Kim HJ, Park JS (2017) Usage of Human Mesenchymal Stem Cells in Cell-based Therapy: Advantages and Disadvantages. Dev Reprod 21: 1-10. 121. Kim JW, Ha KY, Molon JN, Kim YH (2013) Bone marrow-derived mesenchymal stem cell transplantation for chronic spinal cord injury in rats: comparative study between intralesional and intravenous transplantation. Spine (Phila Pa 1976 ) 38: E1065-E1074. 122. Knoferl MW, Liener UC, Seitz DH, Perl M, Bruckner UB, Kinzl L, Gebhard F (2003) Cardiopulmonary, histological, and inflammatory alterations after lung contusion in a novel mouse model of blunt chest trauma. Shock 19: 519-525.

173

Reference List

123. Koci Z, Turnovcova K, Dubsky M, Baranovicova L, Holan V, Chudickova M, Sykova E, Kubinova S (2014) Characterization of human adipose tissue-derived stromal cells isolated from diabetic patient's distal limbs with critical ischemia. Cell Biochem Funct 32: 597-604. 124. Koning JJ, Kooij G, de Vries HE, Nolte MA, Mebius RE (2013) Mesenchymal stem cells are mobilized from the bone marrow during inflammation. Front Immunol 4: 49. 125. Krampera M, Cosmi L, Angeli R, Pasini A, Liotta F, Andreini A, Santarlasci V, Mazzinghi B, Pizzolo G, Vinante F, Romagnani P, Maggi E, Romagnani S, Annunziato F (2006) Role for interferon-gamma in the immunomodulatory activity of human bone marrow mesenchymal stem cells. Stem Cells 24: 386-398. 126. Krampera M, Galipeau J, Shi Y, Tarte K, Sensebe L (2013) Immunological characterization of multipotent mesenchymal stromal cells--The International Society for Cellular Therapy (ISCT) working proposal. Cytotherapy 15: 1054-1061. 127. Krasnodembskaya A, Song Y, Fang X, Gupta N, Serikov V, Lee JW, Matthay MA (2010) Antibacterial effect of human mesenchymal stem cells is mediated in part from secretion of the antimicrobial peptide LL-37. Stem Cells 28: 2229-2238. 128. Krombach F, Munzing S, Allmeling AM, Gerlach JT, Behr J, Dorger M (1997) Cell size of alveolar macrophages: an interspecies comparison. Environ Health Perspect 105 Suppl 5: 1261-1263. 129. Kroshus TJ, Rollins SA, Dalmasso AP, Elliott EA, Matis LA, Squinto SP, Bolman RM, III (1995) Complement inhibition with an anti-C5 monoclonal antibody prevents acute cardiac tissue injury in an ex vivo model of pig-to-human xenotransplantation. Transplantation 60: 1194-1202. 130. Kumar S, Ponnazhagan S (2007) Bone homing of mesenchymal stem cells by ectopic alpha 4 integrin expression. FASEB J 21: 3917-3927. 131. Kumar S, Ponnazhagan S (2012) Mobilization of bone marrow mesenchymal stem cells in vivo augments bone healing in a mouse model of segmental bone defect. Bone 50: 1012- 1018. 132. Lee HJ, Ko JH, Jeong HJ, Ko AY, Kim MK, Wee WR, Yoon SO, Oh JY (2015) Mesenchymal stem/stromal cells protect against autoimmunity via CCL2-dependent recruitment of myeloid- derived suppressor cells. J Immunol 194: 3634-3645. 133. Lee JW, Krasnodembskaya A, McKenna DH, Song Y, Abbott J, Matthay MA (2013) Therapeutic effects of human mesenchymal stem cells in ex vivo human lungs injured with live bacteria. Am J Respir Crit Care Med 187: 751-760. 134. Lee KC, Lin HC, Huang YH, Hung SC (2015) Allo-transplantation of mesenchymal stem cells attenuates hepatic injury through IL1Ra dependent macrophage switch in a mouse model of liver disease. J Hepatol 63: 1405-1412. 135. Lee RH, Pulin AA, Seo MJ, Kota DJ, Ylostalo J, Larson BL, Semprun-Prieto L, Delafontaine P, Prockop DJ (2009) Intravenous hMSCs improve myocardial infarction in mice because cells embolized in lung are activated to secrete the anti-inflammatory protein TSG-6. Cell Stem Cell 5: 54-63.

174

Reference List

136. Leibacher J, Dauber K, Ehser S, Brixner V, Kollar K, Vogel A, Spohn G, Schafer R, Seifried E, Henschler R (2017) Human mesenchymal stromal cells undergo apoptosis and fragmentation after intravenous application in immune-competent mice. Cytotherapy 19: 61- 74. 137. Leibacher J, Henschler R (2016) Biodistribution, migration and homing of systemically applied mesenchymal stem/stromal cells. Stem Cell Res Ther 7: 7. 138. Li B, Zhang H, Zeng M, He W, Li M, Huang X, Deng DY, Wu J (2015) Bone marrow mesenchymal stem cells protect alveolar macrophages from lipopolysaccharide-induced apoptosis partially by inhibiting the Wnt/beta-catenin pathway. Cell Biol Int 39: 192-200. 139. Li D, Liu Q, Qi L, Dai X, Liu H, Wang Y (2016) Low levels of TGF-beta1 enhance human umbilical cord-derived mesenchymal stem cell fibronectin production and extend survival time in a rat model of lipopolysaccharide-induced acute lung injury. Mol Med Rep 14: 1681- 1692. 140. Li N, Hua J (2017) Interactions between mesenchymal stem cells and the immune system. Cell Mol Life Sci 74: 2345-2360. 141. Li P, Zhang R, Sun H, Chen L, Liu F, Yao C, Du M, Jiang X (2013) PKH26 can transfer to host cells in vitro and vivo. Stem Cells Dev 22: 340-344. 142. Li W, Ren G, Huang Y, Su J, Han Y, Li J, Chen X, Cao K, Chen Q, Shou P, Zhang L, Yuan ZR, Roberts AI, Shi S, Le AD, Shi Y (2012) Mesenchymal stem cells: a double-edged sword in regulating immune responses. Cell Death Differ 19: 1505-1513. 143. Li Y, Yu X, Lin S, Li X, Zhang S, Song YH (2007) Insulin-like growth factor 1 enhances the migratory capacity of mesenchymal stem cells. Biochem Biophys Res Commun 356: 780- 784. 144. Li Z, Scott MJ, Fan EK, Li Y, Liu J, Xiao G, Li S, Billiar TR, Wilson MA, Jiang Y, Fan J (2016) Tissue damage negatively regulates LPS-induced macrophage necroptosis. Cell Death Differ 23: 1428-1447. 145. Liener UC, Bartl C, Knöferl MW, Kinzl L, Brückner UB, et al. (2005) Ein stumpfes Thoraxtrauma induziert ausgeprägte Veränderungen der Immunlage der Lunge. In: Rothmund M, Jauch K-W, Bauer H., editors. Chirurgisches Forum 2005 . Springer Berlin Heidelberg. pp. 331-332. 146. Liener UC, Knöferl MW, Perl M, Kurz M, Brückner UB, et al. (2003) Induziert das stumpfe Thoraxtrauma eine Aktivierung zirkulierender Monozyten? In: Menger MD, Haas NP, Neugebauer E, Bauer H, editors. 120. Kongress der Deutschen Gesellschaft für Chirurgie München, 29. 04. – 02.05.2003. Springer Berlin Heidelberg. pp. 273-275. 147. Liener UC, Perl M, Huber-Lang MS, Seitz DH, Bruckner UB, Gebhard F, Knoferl MW (2011) Is the function of alveolar macrophages altered following blunt chest trauma? Langenbecks Arch Surg 396: 251-259. 148. Lin F, Zhang W, Xue D, Zhu T, Li J, Chen E, Yao X, Pan Z (2016) Signaling pathways involved in the effects of HMGB1 on mesenchymal stem cell migration and osteoblastic differentiation. Int J Mol Med 37: 789-797.

175

Reference List

149. Liu FB, Lin Q, Liu ZW (2016) A study on the role of apoptotic human umbilical cord mesenchymal stem cells in bleomycin-induced acute lung injury in rat models. Eur Rev Med Pharmacol Sci 20: 969-982. 150. Liu M, Lei H, Dong P, Fu X, Yang Z, Yang Y, Ma J, Liu X, Cao Y, Xiao R (2017) Adipose- Derived Mesenchymal Stem Cells from the Elderly Exhibit Decreased Migration and Differentiation Abilities with Senescent Properties. Cell Transplant 26: 1505-1519. 151. Liu W, Zhang S, Gu S, Sang L, Dai C (2015) Mesenchymal stem cells recruit macrophages to alleviate experimental colitis through TGFbeta1. Cell Physiol Biochem 35: 858-865. 152. Liu Y, Yang X, Jing Y, Zhang S, Zong C, Jiang J, Sun K, Li R, Gao L, Zhao X, Wu D, Shi Y, Han Z, Wei L (2015) Contribution and Mobilization of Mesenchymal Stem Cells in a mouse model of carbon tetrachloride-induced liver fibrosis. Sci Rep 5: 17762. 153. Lotfi R, Eisenbacher J, Solgi G, Fuchs K, Yildiz T, Nienhaus C, Rojewski MT, Schrezenmeier H (2011) Human mesenchymal stem cells respond to native but not oxidized damage associated molecular pattern molecules from necrotic (tumor) material. Eur J Immunol 41: 2021-2028. 154. Lundberg C, Arfors KE (1983) Polymorphonuclear leukocyte accumulation in inflammatory dermal sites as measured by 51Cr-labeled cells and myeloperoxidase. Inflammation 7: 247- 255. 155. Luo L, Shaver CM, Zhao Z, Koyama T, Calfee CS, Bastarache JA, Ware LB (2016) Clinical predictors of hospital mortality differ between direct and indirect acute respiratory distress syndrome. Chest . 156. Lupia E, Bosco O, Mariano F, Dondi AE, Goffi A, Spatola T, Cuccurullo A, Tizzani P, Brondino G, Stella M, Montrucchio G (2009) Elevated thrombopoietin in plasma of burned patients without and with sepsis enhances platelet activation. J Thromb Haemost 7: 1000- 1008. 157. Luu NT, McGettrick HM, Buckley CD, Newsome PN, Rainger GE, Frampton J, Nash GB (2013) Crosstalk between mesenchymal stem cells and endothelial cells leads to downregulation of cytokine-induced leukocyte recruitment. Stem Cells 31: 2690-2702. 158. Luz-Crawford P, Djouad F, Toupet K, Bony C, Franquesa M, Hoogduijn MJ, Jorgensen C, Noel D (2016) Mesenchymal Stem Cell-Derived Interleukin 1 Receptor Antagonist Promotes Macrophage Polarization and Inhibits B Cell Differentiation. Stem Cells 34: 483-492. 159. Marquez-Curtis LA, Gul-Uludag H, Xu P, Chen J, Janowska-Wieczorek A (2013) CXCR4 transfection of cord blood mesenchymal stromal cells with the use of cationic liposome enhances their migration toward stromal cell-derived factor-1. Cytotherapy 15: 840-849. 160. Marquez-Curtis LA, Janowska-Wieczorek A (2013) Enhancing the migration ability of mesenchymal stromal cells by targeting the SDF-1/CXCR4 axis. Biomed Res Int 2013: 561098. 161. Matsumura E, Tsuji K, Komori K, Koga H, Sekiya I, Muneta T (2017) Pretreatment with IL- 1beta enhances proliferation and chondrogenic potential of synovium-derived mesenchymal stem cells. Cytotherapy 19: 181-193.

176

Reference List

162. Matthay MA (2015) Therapeutic potential of mesenchymal stromal cells for acute respiratory distress syndrome. Ann Am Thorac Soc 12 Suppl 1: S54-S57. 163. Matthay MA, Pati S, Lee JW (2017) Concise Review: Mesenchymal Stem (Stromal) Cells: Biology and Preclinical Evidence for Therapeutic Potential for Organ Dysfunction Following Trauma or Sepsis. Stem Cells 35: 316-324. 164. Matute-Bello G, Downey G, Moore BB, Groshong SD, Matthay MA, Slutsky AS, Kuebler WM (2011) An official American Thoracic Society workshop report: features and measurements of experimental acute lung injury in animals. Am J Respir Cell Mol Biol 44: 725-738. 165. Matute-Bello G, Frevert CW, Martin TR (2008) Animal models of acute lung injury. Am J Physiol Lung Cell Mol Physiol 295: L379-L399. 166. McMahan RS, Birkland TP, Smigiel KS, Vandivort TC, Rohani MG, Manicone AM, McGuire JK, Gharib SA, Parks WC (2016) Stromelysin-2 (MMP10) Moderates Inflammation by Controlling Macrophage Activation. J Immunol 197: 899-909. 167. Meduri GU, Headley S, Kohler G, Stentz F, Tolley E, Umberger R, Leeper K (1995) Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Plasma IL- 1 beta and IL-6 levels are consistent and efficient predictors of outcome over time. Chest 107: 1062-1073. 168. Medvedev SP, Shevchenko AI, Zakian SM (2010) Induced Pluripotent Stem Cells: Problems and Advantages when Applying them in Regenerative Medicine. Acta Naturae 2: 18-28. 169. Menard C, Pacelli L, Bassi G, Dulong J, Bifari F, Bezier I, Zanoncello J, Ricciardi M, Latour M, Bourin P, Schrezenmeier H, Sensebe L, Tarte K, Krampera M (2013) Clinical-grade mesenchymal stromal cells produced under various good manufacturing practice processes differ in their immunomodulatory properties: standardization of immune quality controls. Stem Cells Dev 22: 1789-1801. 170. Mendicino M, Bailey AM, Wonnacott K, Puri RK, Bauer SR (2014) MSC-based product characterization for clinical trials: an FDA perspective. Cell Stem Cell 14: 141-145. 171. Merck (2018) SCR027 | Rat Mesenchymal Stem Cells (Bone Marrow). http://www.merckmillipore.com/DE/de/product/Rat-Mesenchymal-Stem-Cells-Bone- Marrow,MM_NF-SCR027?bd=1. Access Date: 18-2-2018 172. Miettinen JA, Pietila M, Salonen RJ, Ohlmeier S, Ylitalo K, Huikuri HV, Lehenkari P (2011) Tumor necrosis factor alpha promotes the expression of immunosuppressive proteins and enhances the cell growth in a human bone marrow-derived stem cell culture. Exp Cell Res 317: 791-801. 173. Modig J, Samuelsson T, Hallgren R (1986) The predictive and discriminative value of biologically active products of eosinophils, neutrophils and complement in bronchoalveolar lavage and blood in patients with adult respiratory distress syndrome. Resuscitation 14: 121- 134. 174. Moll G, Rasmusson-Duprez I, von BL, Connolly-Andersen AM, Elgue G, Funke L, Hamad OA, Lonnies H, Magnusson PU, Sanchez J, Teramura Y, Nilsson-Ekdahl K, Ringden O, Korsgren O, Nilsson B, Le BK (2012) Are therapeutic human mesenchymal stromal cells compatible with human blood? Stem Cells 30: 1565-1574. 177

Reference List

175. Monsel A, Zhu YG, Gennai S, Hao Q, Liu J, Lee JW (2014) Cell-based therapy for acute organ injury: preclinical evidence and ongoing clinical trials using mesenchymal stem cells. Anesthesiology 121: 1099-1121. 176. Monsel A, Zhu YG, Gudapati V, Lim H, Lee JW (2016) Mesenchymal stem cell derived secretome and extracellular vesicles for acute lung injury and other inflammatory lung diseases. Expert Opin Biol Ther 16: 859-871. 177. Morrison TJ, Jackson MV, Cunningham EK, Kissenpfennig A, McAuley DF, O'Kane CM, Krasnodembskaya AD (2017) Mesenchymal Stromal Cells Modulate Macrophages in Clinically Relevant Lung Injury Models by Extracellular Vesicle Mitochondrial Transfer. Am J Respir Crit Care Med 196: 1275-1286. 178. Mutschler W, Mutschler M, Graw M, Lefering R (2016) [Long-term survival after severe trauma]. Unfallchirurg 119: 546-553. 179. Nagaya N, Fujii T, Iwase T, Ohgushi H, Itoh T, Uematsu M, Yamagishi M, Mori H, Kangawa K, Kitamura S (2004) Intravenous administration of mesenchymal stem cells improves cardiac function in rats with acute myocardial infarction through angiogenesis and myogenesis. Am J Physiol Heart Circ Physiol 287: H2670-H2676. 180. National Library of Medicine at the National Institutes of Health (2017) https://clinicaltrials.gov/. Access Date: 25.4-2018. 181. Nemeth K, Leelahavanichkul A, Yuen PS, Mayer B, Parmelee A, Doi K, Robey PG, Leelahavanichkul K, Koller BH, Brown JM, Hu X, Jelinek I, Star RA, Mezey E (2009) Bone marrow stromal cells attenuate sepsis via prostaglandin E(2)-dependent reprogramming of host macrophages to increase their interleukin-10 production. Nat Med 15: 42-49. 182. Nicklas JA, Buel E (2003) Development of an Alu-based, real-time PCR method for quantitation of human DNA in forensic samples. J Forensic Sci 48: 936-944. 183. Obertacke, U, Redl, H., Schmit-Neuerburg, K. P., and Schlag, G. (1994) Lokale und systemische Reaktionen nach Lungenkontusion. Springer-Verlag Berlin Heidelberg. 184. Oh JY, Roddy GW, Choi H, Lee RH, Ylostalo JH, Rosa RH, Jr., Prockop DJ (2010) Anti- inflammatory protein TSG-6 reduces inflammatory damage to the cornea following chemical and mechanical injury. Proc Natl Acad Sci U S A 107: 16875-16880. 185. Ortiz LA, Dutreil M, Fattman C, Pandey AC, Torres G, Go K, Phinney DG (2007) Interleukin 1 receptor antagonist mediates the antiinflammatory and antifibrotic effect of mesenchymal stem cells during lung injury. Proc Natl Acad Sci U S A 104: 11002-11007. 186. Palmer C (2007) Major trauma and the injury severity score--where should we set the bar? Annu Proc Assoc Adv Automot Med 51: 13-29. 187. Pan H, Lan J, Luo X, Gao J, Xie X, Guo H (2014) Biologic properties of gadolinium diethylenetriaminepentaacetic acid-labeled and PKH26-labeled human umbilical cord mesenchymal stromal cells. Cytotherapy 16: 74-83. 188. Pape HC, Lefering R, Butcher N, Peitzman A, Leenen L, Marzi I, Lichte P, Josten C, Bouillon B, Schmucker U, Stahel P, Giannoudis P, Balogh Z (2014) The definition of polytrauma revisited: An international consensus process and proposal of the new 'Berlin definition'. J Trauma Acute Care Surg 77: 780-786. 178

Reference List

189. Park SJ, Kim KJ, Kim WU, Cho CS (2014) Interaction of mesenchymal stem cells with fibroblast-like synoviocytes via cadherin-11 promotes angiogenesis by enhanced secretion of placental growth factor. J Immunol 192: 3003-3010. 190. Perl M, Hohmann C, Denk S, Kellermann P, Lu D, Braumuller S, Bachem MG, Thomas J, Knoferl MW, Ayala A, Gebhard F, Huber-Lang MS (2012) Role of activated neutrophils in chest trauma-induced septic acute lung injury. Shock 38: 98-106. 191. Perl M, Kieninger M, Huber-Lang MS, Gross HJ, Bachem MG, Braumuller S, Gebhard F, Ayala A, Knoferl MW (2012) Divergent effects of activated neutrophils on inflammation, Kupffer cell/splenocyte activation, and lung injury following blunt chest trauma. Shock 37: 210-218. 192. Polchert D, Sobinsky J, Douglas G, Kidd M, Moadsiri A, Reina E, Genrich K, Mehrotra S, Setty S, Smith B, Bartholomew A (2008) IFN-gamma activation of mesenchymal stem cells for treatment and prevention of graft versus host disease. Eur J Immunol 38: 1745-1755. 193. Polzer H, Volkmer E, Saller MM, Prall WC, Haasters F, Drosse I, Anz D, Mutschler W, Schieker M (2012) Long-term detection of fluorescently labeled human mesenchymal stem cell in vitro and in vivo by semi-automated microscopy. Tissue Eng Part C Methods 18: 156- 165. 194. Ponte AL, Marais E, Gallay N, Langonne A, Delorme B, Herault O, Charbord P, Domenech J (2007) The in vitro migration capacity of human bone marrow mesenchymal stem cells: comparison of chemokine and growth factor chemotactic activities. Stem Cells 25: 1737- 1745. 195. Prasanna SJ, Gopalakrishnan D, Shankar SR, Vasandan AB (2010) Pro-inflammatory cytokines, IFNgamma and TNFalpha, influence immune properties of human bone marrow and Wharton jelly mesenchymal stem cells differentially. PLoS One 5: e9016. 196. Pricola KL, Kuhn NZ, Haleem-Smith H, Song Y, Tuan RS (2009) Interleukin-6 maintains bone marrow-derived mesenchymal stem cell stemness by an ERK1/2-dependent mechanism. J Cell Biochem 108: 577-588. 197. Pulli B, Ali M, Forghani R, Schob S, Hsieh KL, Wojtkiewicz G, Linnoila JJ, Chen JW (2013) Measuring myeloperoxidase activity in biological samples. PLoS One 8: e67976. 198. Qi Y, Jiang D, Sindrilaru A, Stegemann A, Schatz S, Treiber N, Rojewski M, Schrezenmeier H, Vander BS, Wlaschek M, Bohm M, Seitz A, Scholz N, Durselen L, Brinckmann J, Ignatius A, Scharffetter-Kochanek K (2014) TSG-6 released from intradermally injected mesenchymal stem cells accelerates wound healing and reduces tissue fibrosis in murine full-thickness skin wounds. J Invest Dermatol 134: 526-537. 199. Raffaghello L, Bianchi G, Bertolotto M, Montecucco F, Busca A, Dallegri F, Ottonello L, Pistoia V (2008) Human mesenchymal stem cells inhibit neutrophil apoptosis: a model for neutrophil preservation in the bone marrow niche. Stem Cells 26: 151-162. 200. Ragaller M, Richter T (2010) Acute lung injury and acute respiratory distress syndrome. J Emerg Trauma Shock 3: 43-51.

179

Reference List

201. Recknagel S, Bindl R, Kurz J, Wehner T, Ehrnthaller C, Knoferl MW, Gebhard F, Huber- Lang M, Claes L, Ignatius A (2011) Experimental blunt chest trauma impairs fracture healing in rats. J Orthop Res 29: 734-739. 202. Ren G, Su J, Zhang L, Zhao X, Ling W, L'huillie A, Zhang J, Lu Y, Roberts AI, Ji W, Zhang H, Rabson AB, Shi Y (2009) Species variation in the mechanisms of mesenchymal stem cell- mediated immunosuppression. Stem Cells 27: 1954-1962. 203. Ren G, Zhang L, Zhao X, Xu G, Zhang Y, Roberts AI, Zhao RC, Shi Y (2008) Mesenchymal stem cell-mediated immunosuppression occurs via concerted action of chemokines and nitric oxide. Cell Stem Cell 2: 141-150. 204. Ren G, Zhao X, Zhang L, Zhang J, L'Huillier A, Ling W, Roberts AI, Le AD, Shi S, Shao C, Shi Y (2010) Inflammatory cytokine-induced intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 in mesenchymal stem cells are critical for immunosuppression. J Immunol 184: 2321-2328. 205. Ren W, Wang Z, Hua F, Zhu L (2015) Plasminogen activator inhibitor-1 regulates LPS- induced TLR4/MD-2 pathway activation and inflammation in alveolar macrophages. Inflammation 38: 384-393. 206. Rennert RC, Sorkin M, Januszyk M, Duscher D, Kosaraju R, Chung MT, Lennon J, Radiya- Dixit A, Raghvendra S, Maan ZN, Hu MS, Rajadas J, Rodrigues M, Gurtner GC (2014) Diabetes impairs the angiogenic potential of adipose-derived stem cells by selectively depleting cellular subpopulations. Stem Cell Res Ther 5: 79. 207. Ringe J, Strassburg S, Neumann K, Endres M, Notter M, Burmester GR, Kaps C, Sittinger M (2007) Towards in situ tissue repair: human mesenchymal stem cells express chemokine receptors CXCR1, CXCR2 and CCR2, and migrate upon stimulation with CXCL8 but not CCL2. J Cell Biochem 101: 135-146. 208. Rochefort GY, Delorme B, Lopez A, Herault O, Bonnet P, Charbord P, Eder V, Domenech J (2006) Multipotential mesenchymal stem cells are mobilized into peripheral blood by hypoxia. Stem Cells 24: 2202-2208. 209. Rohani MG, McMahan RS, Razumova MV, Hertz AL, Cieslewicz M, Pun SH, Regnier M, Wang Y, Birkland TP, Parks WC (2015) MMP-10 Regulates Collagenolytic Activity of Alternatively Activated Resident Macrophages. J Invest Dermatol 135: 2377-2384. 210. Rojas M, Cardenes N, Kocyildirim E, Tedrow JR, Caceres E, Deans R, Ting A, Bermudez C (2014) Human adult bone marrow-derived stem cells decrease severity of lipopolysaccharide-induced acute respiratory distress syndrome in sheep. Stem Cell Res Ther 5: 42. 211. Rojewski MT, Baldus C, Knauf W, Thiel E, Schrezenmeier H (2002) Dual effects of arsenic trioxide (As2O3) on non-acute promyelocytic leukaemia myeloid cell lines: induction of apoptosis and inhibition of proliferation. Br J Haematol 116: 555-563. 212. Rojewski MT, Weber BM, Schrezenmeier H (2008) Phenotypic Characterization of Mesenchymal Stem Cells from Various Tissues. Transfus Med Hemother 35: 168-184. 213. Romanelli RG, Petrai I, Robino G, Efsen E, Novo E, Bonacchi A, Pagliai G, Grossi A, Parola M, Navari N, Delogu W, Vizzutti F, Rombouts K, Gentilini P, Laffi G, Marra F (2006) 180

Reference List

Thrombopoietin stimulates migration and activates multiple signaling pathways in hepatoblastoma cells. Am J Physiol Gastrointest Liver Physiol 290: G120-G128. 214. Romieu-Mourez R, Francois M, Boivin MN, Stagg J, Galipeau J (2007) Regulation of MHC class II expression and antigen processing in murine and human mesenchymal stromal cells by IFN-gamma, TGF-beta, and cell density. J Immunol 179: 1549-1558. 215. Rotstein OD (2003) Modeling the two-hit hypothesis for evaluating strategies to prevent organ injury after shock/resuscitation. J Trauma 54: S203-S206. 216. Ruster B, Gottig S, Ludwig RJ, Bistrian R, Muller S, Seifried E, Gille J, Henschler R (2006) Mesenchymal stem cells display coordinated rolling and adhesion behavior on endothelial cells. Blood 108: 3938-3944. 217. Sadeghi Z, Isariyawongse J, Kavran M, Izgi K, Marini G, Molter J, Daneshgari F, Flask CA, Caplan A, Hijaz A (2016) Mesenchymal stem cell therapy in a rat model of birth-trauma injury: functional improvements and biodistribution. Int Urogynecol J 27: 291-300. 218. Sapan HB, Paturusi I, Islam AA, Yusuf I, Patellongi I, Massi MN, Pusponegoro AD, Arief SK, Labeda I, Rendy L, Hatta M (2017) Interleukin-6 and interleukin-10 plasma levels and mRNA expression in polytrauma patients. Chin J Traumatol 20: 318-322. 219. Sapru A, Calfee CS, Liu KD, Kangelaris K, Hansen H, Pawlikowska L, Ware LB, Alkhouli MF, Abbott J, Matthay MA (2015) Plasma soluble thrombomodulin levels are associated with mortality in the acute respiratory distress syndrome. Intensive Care Med 41: 470-478. 220. Sattler S, Maier RV (2002) Pulmonary Contusion. In: Thoracic Trauma and Critical Care. Boston, MA: Springer Science+Business Media New York. pp. 235-236. 221. Schallmoser K, Bartmann C, Rohde E, Reinisch A, Kashofer K, Stadelmeyer E, Drexler C, Lanzer G, Linkesch W, Strunk D (2007) Human platelet lysate can replace fetal bovine serum for clinical-scale expansion of functional mesenchymal stromal cells. Transfusion 47: 1436-1446. 222. Schlesinger RB, Fine JM, Chen LC (1992) Interspecies differences in the phagocytic activity of pulmonary macrophages subjected to acidic challenge. Fundam Appl Toxicol 19: 584-589. 223. Schmidt A, Ladage D, Schinkothe T, Klausmann U, Ulrichs C, Klinz FJ, Brixius K, Arnhold S, Desai B, Mehlhorn U, Schwinger RH, Staib P, Addicks K, Bloch W (2006) Basic fibroblast growth factor controls migration in human mesenchymal stem cells. Stem Cells 24: 1750- 1758. 224. Seebach C, Henrich D, Tewksbury R, Wilhelm K, Marzi I (2007) Number and proliferative capacity of human mesenchymal stem cells are modulated positively in multiple trauma patients and negatively in atrophic nonunions. Calcif Tissue Int 80: 294-300. 225. Seitz DH, Froba JS, Niesler U, Palmer A, Veltkamp HA, Braumuller ST, Wagner F, Wagner K, Bader S, Wachter U, Calzia E, Radermacher P, Huber-Lang MS, Zhou S, Gebhard F, Knoferl MW (2012) Inhaled hydrogen sulfide induces suspended animation, but does not alter the inflammatory response after blunt chest trauma. Shock 37: 197-204. 226. Seitz DH, Niesler U, Palmer A, Sulger M, Braumuller ST, Perl M, Gebhard F, Knoferl MW (2010) Blunt chest trauma induces mediator-dependent monocyte migration to the lung. Crit Care Med 38: 1852-1859. 181

Reference List

227. Seitz DH, Palmer A, Niesler U, Froba JS, Heidemann V, Rittlinger A, Braumuller ST, Zhou S, Gebhard F, Knoferl MW (2011) Alveolar macrophage phagocytosis is enhanced after blunt chest trauma and alters the posttraumatic mediator release. Shock 36: 621-627. 228. Seitz DH, Perl M, Liener UC, Tauchmann B, Braumuller ST, Bruckner UB, Gebhard F, Knoferl MW (2011) Inflammatory alterations in a novel combination model of blunt chest trauma and hemorrhagic shock. J Trauma 70: 189-196. 229. Seitz DH, Perl M, Mangold S, Neddermann A, Braumuller ST, Zhou S, Bachem MG, Huber- Lang MS, Knoferl MW (2008) Pulmonary contusion induces alveolar type 2 epithelial cell apoptosis: role of alveolar macrophages and neutrophils. Shock 30: 537-544. 230. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker HV, Xu W, Richards DR, McDonald- Smith GP, Gao H, Hennessy L, Finnerty CC, Lopez CM, Honari S, Moore EE, Minei JP, Cuschieri J, Bankey PE, Johnson JL, Sperry J, Nathens AB, Billiar TR, West MA, Jeschke MG, Klein MB, Gamelli RL, Gibran NS, Brownstein BH, Miller-Graziano C, Calvano SE, Mason PH, Cobb JP, Rahme LG, Lowry SF, Maier RV, Moldawer LL, Herndon DN, Davis RW, Xiao W, Tompkins RG (2013) Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A 110: 3507-3512. 231. Shao-Fang Z, Hong-Tian Z, Zhi-Nian Z, Yuan-Li H (2011) PKH26 as a fluorescent label for live human umbilical mesenchymal stem cells. In Vitro Cell Dev Biol Anim 47: 516-520. 232. Shen JL, Huang YZ, Yin WJ, Cen J, Zheng PH, Gong LZ, Liu Y (2011) [The effects of thrombopoietin on the fibrogenesis of bone marrow stromal cells in absence of megakaryocytes]. Zhongguo Ying Yong Sheng Li Xue Za Zhi 27: 163-166. 233. Son BR, Marquez-Curtis LA, Kucia M, Wysoczynski M, Turner AR, Ratajczak J, Ratajczak MZ, Janowska-Wieczorek A (2006) Migration of bone marrow and cord blood mesenchymal stem cells in vitro is regulated by stromal-derived factor-1-CXCR4 and hepatocyte growth factor-c-met axes and involves matrix metalloproteinases. Stem Cells 24: 1254-1264. 234. Sordi V, Malosio ML, Marchesi F, Mercalli A, Melzi R, Giordano T, Belmonte N, Ferrari G, Leone BE, Bertuzzi F, Zerbini G, Allavena P, Bonifacio E, Piemonti L (2005) Bone marrow mesenchymal stem cells express a restricted set of functionally active chemokine receptors capable of promoting migration to pancreatic islets. Blood 106: 419-427. 235. Squillaro T, Peluso G, Galderisi U (2016) Clinical Trials With Mesenchymal Stem Cells: An Update. Cell Transplant 25: 829-848. 236. Stormann P, Lustenberger T, Relja B, Marzi I, Wutzler S (2017) Role of biomarkers in acute traumatic lung injury. Injury 48: 2400-2406. 237. Suila H, Hirvonen T, Kotovuori A, Ritamo I, Kerkela E, Anderson H, Natunen S, Tuimala J, Laitinen S, Nystedt J, Rabina J, Valmu L (2014) Human umbilical cord blood-derived mesenchymal stromal cells display a novel interaction between P-selectin and galectin-1. Scand J Immunol 80: 12-21. 238. Sung DK, Chang YS, Sung SI, Yoo HS, Ahn SY, Park WS (2016) Antibacterial effect of mesenchymal stem cells against Escherichia coli is mediated by secretion of beta- defensin- 2 via toll- like receptor 4 signalling. Cell Microbiol 18: 424-436.

182

Reference List

239. Suzuki K, Ota H, Sasagawa S, Sakatani T, Fujikura T (1983) Assay method for myeloperoxidase in human polymorphonuclear leukocytes. Anal Biochem 132: 345-352. 240. Toma C, Wagner WR, Bowry S, Schwartz A, Villanueva F (2009) Fate of culture-expanded mesenchymal stem cells in the microvasculature: in vivo observations of cell kinetics. Circ Res 104: 398-402. 241. U.S.Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research (2005) Guidance for Industry Estimating the Maximum Safe Starting Dose in Initial Clinical Trials for Therapeutics in Adult Healthy Volunteers. U.S. Food and Drug Administration. Available: https://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidance s/ucm078932.pdf. Access Date: 15-4-2018. 242. Ulivi V, Tasso R, Cancedda R, Descalzi F (2014) Mesenchymal stem cell paracrine activity is modulated by platelet lysate: induction of an inflammatory response and secretion of factors maintaining macrophages in a proinflammatory phenotype. Stem Cells Dev 23: 1858-1869. 243. Upham JW, Strickland DH, Bilyk N, Robinson BW, Holt PG (1995) Alveolar macrophages from humans and rodents selectively inhibit T-cell proliferation but permit T-cell activation and cytokine secretion. Immunology 84: 142-147. 244. Vernikouskaya I, Fekete N, Bannwarth M, Erle A, Rojewski M, Landfester K, Schmidtke- Schrezenmeier G, Schrezenmeier H, Rasche V (2014) Iron-loaded PLLA nanoparticles as highly efficient intracellular markers for visualization of mesenchymal stromal cells by MRI. Contrast Media Mol Imaging 9: 109-121. 245. Veronesi E, Murgia A, Caselli A, Grisendi G, Piccinno MS, Rasini V, Giordano R, Montemurro T, Bourin P, Sensebe L, Rojewski MT, Schrezenmeier H, Layrolle P, Ginebra MP, Panaitescu CB, Gomez-Barrena E, Catani F, Paolucci P, Burns JS, Dominici M (2014) Transportation conditions for prompt use of ex vivo expanded and freshly harvested clinical- grade bone marrow mesenchymal stromal/stem cells for bone regeneration. Tissue Eng Part C Methods 20: 239-251. 246. Volarevic V, Gazdic M, Simovic MB, Jovicic N, Djonov V, Arsenijevic N (2017) Mesenchymal stem cell-derived factors: Immuno-modulatory effects and therapeutic potential. Biofactors 43: 633-644. 247. Voll RE, Herrmann M, Roth EA, Stach C, Kalden JR, Girkontaite I (1997) Immunosuppressive effects of apoptotic cells. Nature 390: 350-351. 248. Walenda T, Bokermann G, Ventura Ferreira MS, Piroth DM, Hieronymus T, Neuss S, Zenke M, Ho AD, Muller AM, Wagner W (2011) Synergistic effects of growth factors and mesenchymal stromal cells for expansion of hematopoietic stem and progenitor cells. Exp Hematol 39: 617-628. 249. Wang CM, Guo Z, Xie YJ, Hao YY, Sun JM, Gu J, Wang AL (2014) Co-treating mesenchymal stem cells with IL1beta and TNF-alpha increases VCAM-1 expression and improves post-ischemic myocardial function. Mol Med Rep 10: 792-798. 250. Wang J, Wang Y, Wang S, Cai J, Shi J, Sui X, Cao Y, Huang W, Chen X, Cai Z, Li H, Bardeesi AS, Zhang B, Liu M, Song W, Wang M, Xiang AP (2015) Bone marrow-derived 183

Reference List

mesenchymal stem cell-secreted IL-8 promotes the angiogenesis and growth of colorectal cancer. Oncotarget 6: 42825-42837. 251. Wang Q, Zhu H, Zhou WG, Guo XC, Wu MJ, Xu ZY, Jiang JF, Shen C, Liu HQ (2013) N- acetylcysteine-pretreated human embryonic mesenchymal stem cell administration protects against bleomycin-induced lung injury. Am J Med Sci 346: 113-122. 252. Wang Y, Liu J, Jiang Q, Deng J, Xu F, Chen X, Cheng F, Zhang Y, Yao Y, Xia Z, Xu X, Su X, Huang M, Dai L, Yang Y, Zhang S, Yu D, Zhao RC, Wei Y, Deng H (2017) Human Adipose-Derived Mesenchymal Stem Cell-Secreted CXCL1 and CXCL8 Facilitate Breast Tumor Growth By Promoting Angiogenesis. Stem Cells 35: 2060-2070. 253. Ware LB, Fang X, Matthay MA (2003) Protein C and thrombomodulin in human acute lung injury. Am J Physiol Lung Cell Mol Physiol 285: L514-L521. 254. Waszak P, Alphonse R, Vadivel A, Ionescu L, Eaton F, Thebaud B (2012) Preconditioning enhances the paracrine effect of mesenchymal stem cells in preventing oxygen-induced neonatal lung injury in rats. Stem Cells Dev 21: 2789-2797. 255. Waterman RS, Tomchuck SL, Henkle SL, Betancourt AM (2010) A new mesenchymal stem cell (MSC) paradigm: polarization into a pro-inflammatory MSC1 or an Immunosuppressive MSC2 phenotype. PLoS One 5: e10088. 256. Weber A, Wasiliew P, Kracht M (2010) Interleukin-1 (IL-1) pathway. Sci Signal 3. 257. Welsch U (2006) Atmungsorgane. In: Lehrbuch Histologie. München: Elsevier. pp. 305-332. 258. Wiegner R, Rudhart NE, Barth E, Gebhard F, Lampl L, Huber-Lang MS, Brenner RE (2017) Mesenchymal stem cells in peripheral blood of severely injured patients. Eur J Trauma Emerg Surg . 259. Wight TN, Frevert CW, Debley JS, Reeves SR, Parks WC, Ziegler SF (2017) Interplay of extracellular matrix and leukocytes in lung inflammation. Cell Immunol 312: 1-14. 260. Windsor AC, Mullen PG, Fowler AA, Sugerman HJ (1993) Role of the neutrophil in adult respiratory distress syndrome. Br J Surg 80: 10-17. 261. Wisniewski HG, Vilcek J (1997) TSG-6: an IL-1/TNF-inducible protein with anti-inflammatory activity. Cytokine Growth Factor Rev 8: 143-156. 262. Wuchter P, Bieback K, Schrezenmeier H, Bornhauser M, Muller LP, Bonig H, Wagner W, Meisel R, Pavel P, Tonn T, Lang P, Muller I, Renner M, Malcherek G, Saffrich R, Buss EC, Horn P, Rojewski M, Schmitt A, Ho AD, Sanzenbacher R, Schmitt M (2015) Standardization of Good Manufacturing Practice-compliant production of bone marrow-derived human mesenchymal stromal cells for immunotherapeutic applications. Cytotherapy 17: 128-139. 263. Wynn RF, Hart CA, Corradi-Perini C, O'Neill L, Evans CA, Wraith JE, Fairbairn LJ, Bellantuono I (2004) A small proportion of mesenchymal stem cells strongly expresses functionally active CXCR4 receptor capable of promoting migration to bone marrow. Blood 104: 2643-2645. 264. Xu YL, Liu YL, Wang Q, Li G, Lu XD, Kong B (2012) Intravenous transplantation of mesenchymal stem cells attenuates oleic acid induced acute lung injury in rats. Chin Med J (Engl ) 125: 2012-2018.

184

Reference List

265. Yang C, Chen Y, Li F, You M, Zhong L, Li W, Zhang B, Chen Q (2018) The biological changes of umbilical cord mesenchymal stem cells in inflammatory environment induced by different cytokines. Mol Cell Biochem . 266. Yang R, Harada T, Mollen KP, Prince JM, Levy RM, Englert JA, Gallowitsch-Puerta M, Yang L, Yang H, Tracey KJ, Harbrecht BG, Billiar TR, Fink MP (2006) Anti-HMGB1 neutralizing antibody ameliorates gut barrier dysfunction and improves survival after hemorrhagic shock. Mol Med 12: 105-114. 267. Yilmaz G, Vital S, Yilmaz CE, Stokes KY, Alexander JS, Granger DN (2011) Selectin- mediated recruitment of bone marrow stromal cells in the postischemic cerebral microvasculature. Stroke 42: 806-811. 268. Zhang QZ, Su WR, Shi SH, Wilder-Smith P, Xiang AP, Wong A, Nguyen AL, Kwon CW, Le AD (2010) Human gingiva-derived mesenchymal stem cells elicit polarization of m2 macrophages and enhance cutaneous wound healing. Stem Cells 28: 1856-1868. 269. Zhao Y, Yang C, Wang H, Li H, Du J, Gu W, Jiang J (2013) Therapeutic effects of bone marrow-derived mesenchymal stem cells on pulmonary impact injury complicated with endotoxemia in rats. Int Immunopharmacol 15: 246-253. 270. Zhen G, Xue Z, Zhao J, Gu N, Tang Z, Xu Y, Zhang Z (2010) Mesenchymal stem cell transplantation increases expression of vascular endothelial growth factor in papain-induced emphysematous lungs and inhibits apoptosis of lung cells. Cytotherapy 12: 605-614. 271. Zheng G, Huang L, Tong H, Shu Q, Hu Y, Ge M, Deng K, Zhang L, Zou B, Cheng B, Xu J (2014) Treatment of acute respiratory distress syndrome with allogeneic adipose-derived mesenchymal stem cells: a randomized, placebo-controlled pilot study. Respir Res 15: 39. 272. Zhu YG, Feng XM, Abbott J, Fang XH, Hao Q, Monsel A, Qu JM, Matthay MA, Lee JW (2014) Human mesenchymal stem cell microvesicles for treatment of Escherichia coli endotoxin-induced acute lung injury in mice. Stem Cells 32: 116-125. 273. Zilow G, Sturm JA, Rother U, Kirschfink M (1990) Complement activation and the prognostic value of C3a in patients at risk of adult respiratory distress syndrome. Clin Exp Immunol 79: 151-157.

185

Appendix

Table 26 Expressed genes of the intersection of the polytrauma cocktail high group and interleukin 1 beta group 6 h, 24 h after stimulation or both times taken together (6 h and 24 h) were assigned to signalling pathways. The signalling pathways are listed here. In some cases, several signalling pathways had been assigned to one gene. Upregulated (+) and downregulated (-) genes compared to unstimulated MSC are listed. ABC: ATP binding cassette; DNA: deoxyribonucleic acid; ECM: extracellular matrix; ErbB protein family: also known as epidermal growth factor receptor (EGFR) family; GnRH: gonadotropin-releasing hormone; HIF-1: hypoxia-inducible factor 1; HTLV: human T- lymphotropic virus; IgA: immunoglobulin A; JAK: Janus kinase; STAT: signal transducer and activator of transcription; MAPK: mitogen-activated protein kinases; RNA: ribonucleic acid; NF- kappaB: nuclear factor kappa-light-chain-enhancer of activated B cells; NOD: nucleotide-binding oligomerization domain; PI3K: phosphatidylinositol-4,5-bisphosphate 3-kinase; PPAR: peroxisome proliferator-activated receptor; CoA: coenzyme A; TGF-beta: transforming growth factor-beta; TNF: tumor necrosis factor; VEGF: vascular endothelial growth factor; mTOR: mammalian target of rapamycin.

KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 0 0 3 ABC transporters hsa02010 + 1 1 1 - 0 0 0 4 Acute myeloid leukemia hsa05221 + 1 0 3 Adipocytokine signaling - 0 0 0 8 hsa04920 pathway + 3 1 4 - 0 1 0 13 African trypanosomiasis hsa05143 + 4 3 5 Alanine, aspartate and - 0 0 0 6 hsa00250 glutamate metabolism + 2 2 2 - 0 0 0 4 Alcoholism hsa05034 + 1 1 2 Aldosterone-regulated sodium - 0 0 0 5 hsa04960 reabsorption + 2 1 2 - 0 0 0 4 Alzheimer's disease hsa05010 + 1 2 1 Amino sugar and nucleotide - 0 0 0 3 hsa00520 sugar metabolism + 1 1 1 - 0 0 0 27 Amoebiasis hsa05146 + 9 9 9 - 0 0 0 2 Amphetamine addiction hsa05031 + 0 1 1

Table 26 is continued on page 185.

186

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 0 0 4 Amyotrophic lateral sclerosis hsa05014 + 1 2 1 Antigen processing and - 0 0 0 3 hsa04612 presentation + 1 1 1 - 0 0 0 20 Apoptosis hsa04210 + 7 3 10 - 0 0 1 8 Arachidonic acid metabolism hsa00590 + 2 2 3 - 0 1 0 4 Arginine and proline metabolism hsa00330 + 1 0 2 Arrhythmogenic right ventricular - 0 1 0 5 hsa05412 cardiomyopathy + 2 1 1 - 0 0 0 2 Asthma hsa05310 + 1 0 1 - 0 1 0 6 Axon guidance hsa04360 + 2 2 1 B cell receptor signaling - 0 0 0 5 hsa04662 pathway + 2 0 3 Bacterial invasion of epithelial - 0 0 1 3 hsa05100 cells + 1 0 1 - 1 0 1 10 Basal cell carcinoma hsa05217 + 3 2 3 - 0 0 0 3 Bile secretion hsa04976 + 1 1 1 - 0 0 0 6 Bladder cancer hsa05219 + 2 2 2 - 3 6 3 24 Calcium signaling pathway hsa04020 + 4 3 5 Carbohydrate digestion and - 0 0 0 2 hsa04973 absorption + 1 0 1 - 0 0 0 2 Cardiac muscle contraction hsa04260 + 1 1 0 - 0 0 0 13 Cell adhesion molecules hsa04514 + 4 2 7

Table 26 is continued on page 186.

187

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway Chagas disease (American - 0 0 0 27 hsa05142 trypanosomiasis) + 10 6 11 - 0 0 0 8 Chemical carcinogenesis hsa05204 + 3 3 2 - 0 0 0 52 Chemokine signaling pathway hsa04062 + 19 13 20 - 0 1 0 4 Cholinergic synapse hsa04725 + 1 0 2 - 0 0 1 7 Chronic myeloid leukemia hsa05220 + 2 0 4 - 0 1 0 7 Circadian entrainment hsa04713 + 2 2 2 - 0 1 0 4 Circadian rhythm hsa04710 + 1 1 1 - 0 0 0 3 Cocaine addiction hsa05030 + 0 1 2 - 0 0 0 1 Collecting duct acid secretion hsa04966 + 0 0 1 - 0 0 0 2 Colorectal cancer hsa05210 + 1 0 1 Complement and coagulation - 2 2 1 14 hsa04610 cascades + 3 2 4 Cysteine and methionine - 0 0 0 3 hsa00270 metabolism + 1 1 1 Cytokine-cytokine receptor - 1 0 3 106 hsa04060 interaction + 35 28 39 - 0 0 0 17 Cytosolic DNA-sensing pathway hsa04623 + 6 3 8 - 0 1 0 9 Dilated cardiomyopathy hsa05414 + 3 3 2 - 0 0 0 2 Dopaminergic synapse hsa04728 + 0 1 1 - 1 1 1 3 Dorso-ventral axis formation hsa04320 + 0 0 0

Table 26 is continued on page 187.

188

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 0 0 9 ECM-receptor interaction hsa04512 + 3 2 4 Endocrine and other factor- - 0 0 1 3 hsa04961 regulated calcium reabsorption + 1 0 1 - 1 1 3 9 Endocytosis hsa04144 + 1 1 2 - 0 0 0 2 Endometrial cancer hsa05213 + 1 0 1 Epithelial cell signaling in - 0 0 0 12 hsa05120 Helicobacter pylori infection + 4 3 5 - 0 0 0 15 Epstein-Barr virus infection hsa05169 + 5 3 7 - 0 0 0 8 ErbB signaling pathway hsa04012 + 4 0 4 - 0 0 0 3 Estrogen signaling pathway hsa04915 + 1 0 2 - 0 0 1 5 Ether lipid metabolism hsa00565 + 1 1 2 - 0 0 1 5 Fat digestion and absorption hsa04975 + 1 2 1 - 0 0 0 2 Fatty acid biosynthesis hsa00061 + 1 1 0 - 0 0 0 5 Fatty acid elongation hsa00062 + 2 1 2 - 0 0 0 2 Fatty acid metabolism hsa01212 + 1 1 0 - 0 0 0 6 Fc epsilon RI signaling pathway hsa04664 + 2 1 3 Fc gamma R-mediated - 0 0 0 6 hsa04666 phagocytosis + 2 1 3 - 0 0 1 16 Focal adhesion hsa04510 + 5 4 6 - 0 0 0 3 Folate biosynthesis hsa00790 + 1 1 1

Table 26 is continued on page 188.

189

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway Fructose and mannose - 0 0 0 1 hsa00051 metabolism + 0 1 0 - 0 0 0 1 Galactose metabolism hsa00052 + 0 1 0 - 0 0 1 6 Gap junction hsa04540 + 1 3 1 - 0 0 0 3 Gastric acid secretion hsa04971 + 1 1 1 - 0 0 1 7 Glioma hsa05214 + 3 1 2 - 0 0 0 5 Glutamatergic synapse hsa04724 + 1 2 2 - 0 0 0 5 Glycerolipid metabolism hsa00561 + 1 2 2 Glycerophospholipid - 0 0 1 5 hsa00564 metabolism + 1 1 2 Glycine, serine and threonine - 0 0 0 3 hsa00260 metabolism + 1 1 1 Glycosaminoglycan - 0 0 0 2 biosynthesis - chondroitin hsa00532 sulfate / dermatan sulfate + 1 1 0 Glycosaminoglycan - 2 2 1 5 biosynthesis - heparan sulfate / hsa00534 heparin + 0 0 0 Glycosaminoglycan - 0 0 0 2 hsa00533 biosynthesis - keratan sulfate + 1 0 1 Glycosphingolipid biosynthesis - - 0 0 0 2 hsa00604 ganglio series + 1 0 1 Glycosphingolipid biosynthesis - - 0 0 1 4 hsa00603 globo series + 1 0 2 Glycosphingolipid biosynthesis - - 0 1 1 3 hsa00601 lacto and neolacto series + 0 0 1 Glycosylphosphatidylinositol(GP - 0 0 0 1 hsa00563 I-anchor biosynthesis) + 0 0 1 - 0 0 0 4 GnRH signaling pathway hsa04912 + 1 1 2

Table 26 is continued on page 189.

190

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 0 0 9 Graft-versus-host disease hsa05332 + 3 3 3 - 0 0 0 9 HIF-1 signaling pathway hsa04066 + 3 2 4 - 0 0 1 26 HTLV-I infection hsa05166 + 8 4 13 - 1 0 1 11 Hedgehog signaling pathway hsa04340 + 3 2 4 - 0 0 0 24 Hematopoietic cell lineage hsa04640 + 8 7 9 - 0 0 0 16 Hepatitis B hsa05161 + 5 3 8 - 0 0 0 11 Hepatitis C hsa05160 + 4 2 5 - 0 1 0 22 Herpes simplex infection hsa05168 + 7 6 8 - 1 0 2 17 Hippo signaling pathway hsa04390 + 5 3 6 - 0 1 0 7 Huntington's disease hsa05016 + 2 1 3 - 0 1 0 11 Hypertrophic cardiomyopathy hsa05410 + 4 3 3 - 0 0 0 18 Inflammatory bowel disease hsa05321 + 6 4 8 - 0 0 0 35 Influenza A hsa05164 + 12 7 16 - 2 2 1 7 Insulin secretion hsa04911 + 0 1 1 - 0 0 0 7 Insulin signaling pathway hsa04910 + 3 1 3 Intestinal immune network for - 0 0 0 6 hsa04672 IgA production + 2 1 3 - 0 0 0 33 Jak-STAT signaling pathway hsa04630 + 10 10 13

Table 26 is continued on page 190.

191

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 0 0 27 Legionellosis hsa05134 + 9 7 11 - 0 0 0 15 Leishmaniasis hsa05140 + 5 4 6 Leukocyte transendothelial - 0 0 0 9 hsa04670 migration + 3 2 4 - 0 0 1 2 Linoleic acid metabolism hsa00591 + 0 0 1 - 0 0 0 7 Long-term depression hsa04730 + 2 2 3 - 0 0 0 1 Long-term potentiation hsa04720 + 0 1 0 - 0 0 0 5 Lysosome hsa04142 + 2 1 2 - 1 1 6 31 MAPK signaling pathway hsa04010 + 6 7 10 - 0 0 0 21 Malaria hsa05144 + 7 6 8 - 0 0 0 19 Measles hsa05162 + 6 4 9 - 0 0 0 7 Melanogenesis hsa04916 + 3 1 3 - 1 1 2 17 Melanoma hsa05218 + 4 5 4 - 1 4 3 57 Metabolic pathways hsa01100 + 16 14 19 Metabolism of xenobiotics by - 0 0 0 7 hsa00980 cytochrome P450 + 3 2 2 - 0 0 1 12 MicroRNAs in cancer hsa05206 + 3 4 4 - 0 0 0 18 Mineral absorption hsa04978 + 7 8 3 - 0 0 0 4 Morphine addiction hsa05032 + 1 1 2

Table 26 is continued on page 191.

192

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway Mucin type O-Glycan - 0 0 0 2 hsa00512 biosynthesis + 1 0 1 - 0 1 0 3 N-Glycan biosynthesis hsa00510 + 1 0 1 - 0 0 0 32 NF-kappa B signaling pathway hsa04064 + 11 7 14 NOD-like receptor signaling - 0 0 0 34 hsa04621 pathway + 12 9 13 Natural killer cell mediated - 0 0 0 9 hsa04650 cytotoxicity + 3 2 4 Neuroactive ligand-receptor - 5 10 3 29 hsa04080 interaction + 4 2 5 - 0 0 0 11 Neurotrophin signaling pathway hsa04722 + 4 1 6 Nicotinate and nicotinamide - 0 0 0 3 hsa00760 metabolism + 1 1 1 - 0 0 0 1 Nicotine addiction hsa05033 + 0 1 0 - 0 0 0 1 Nitrogen metabolism hsa00910 + 0 1 0 - 0 0 0 15 Non-alcoholic fatty liver disease hsa04932 + 5 4 6 - 0 0 0 3 Non-small cell lung cancer hsa05223 + 2 0 1 - 0 0 0 3 Oocyte meiosis hsa04114 + 1 1 1 - 0 1 0 15 Osteoclast differentiation hsa04380 + 4 3 7 Other types of O-glycan - 0 1 0 1 hsa00514 biosynthesis + 0 0 0 - 0 0 0 10 Ovarian steroidogenesis hsa04913 + 3 3 4 - 2 2 3 41 PI3K-Akt signaling pathway hsa04151 + 10 10 14

Table 26 is continued on page 192.

193

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 1 0 7 PPAR signaling pathway hsa03320 + 1 1 4 - 0 0 0 7 Pancreatic cancer hsa05212 + 3 1 3 - 1 1 1 6 Pancreatic secretion hsa04972 + 1 1 1 Pantothenate and CoA - 0 0 0 1 hsa00770 biosynthesis + 1 0 0 Pathogenic Escherichia coli - 0 1 0 4 hsa05130 infection + 1 1 1 - 2 2 5 56 Pathways in cancer hsa05200 + 17 11 19 Pentose and glucuronate - 0 0 0 1 hsa00040 interconversions + 0 1 0 - 0 0 0 3 Peroxisome hsa04146 + 1 1 1 - 0 0 0 25 Pertussis hsa05133 + 7 8 10 - 0 0 0 8 Phagosome hsa04145 + 2 3 3 - 0 0 0 3 Phenylalanine metabolism hsa00360 + 1 1 1 Phenylalanine, tyrosine and - 0 0 0 3 hsa00400 tryptophan biosynthesis + 1 1 1 Phosphatidylinositol signaling - 0 0 0 2 hsa04070 system + 1 0 1 Porphyrin and chlorophyll - 0 0 0 2 hsa00860 metabolism + 1 1 0 - 0 0 0 5 Primary bile acid biosynthesis hsa00120 + 2 1 2 - 0 0 0 1 Primary immunodeficiency hsa05340 + 0 0 1 - 0 0 0 12 Prion diseases hsa05020 + 4 4 4

Table 26 is continued on page 193.

194

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway Progesterone-mediated oocyte - 0 0 0 5 hsa04914 maturation + 2 1 2 - 0 0 0 5 Prolactin signaling pathway hsa04917 + 1 1 3 - 0 0 2 15 Prostate cancer hsa05215 + 5 2 6 Protein digestion and - 1 2 1 10 hsa04974 absorption + 2 4 0 - 1 2 1 22 Proteoglycans in cancer hsa05205 + 6 6 6 - 0 0 0 11 Purine metabolism hsa00230 + 3 4 4 - 0 0 0 2 Pyrimidine metabolism hsa00240 + 0 1 1 - 0 0 0 1 Pyruvate metabolism hsa00620 + 0 1 0 RIG-I-like receptor signaling - 0 0 0 7 hsa04622 pathway + 2 1 4 - 1 1 4 29 Ras signaling pathway hsa04014 + 6 7 10 - 2 2 3 22 Regulation of actin cytoskeleton hsa04810 + 5 4 6 - 0 0 0 3 Renal cell carcinoma hsa05211 + 2 0 1 - 0 0 0 2 Retinol metabolism hsa00830 + 1 1 0 Retrograde endocannabinoid - 0 1 0 4 hsa04723 signaling + 1 1 1 - 0 0 0 48 Rheumatoid arthritis hsa05323 + 16 15 17 SNARE interactions in vesicular - 0 0 0 2 hsa04130 transport + 0 1 1 - 1 3 1 8 Salivary secretion hsa04970 + 1 1 1

Table 26 is continued on page 194.

195

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 0 0 27 Salmonella infection hsa05132 + 9 8 10 - 1 1 0 7 Serotonergic synapse hsa04726 + 1 2 2 - 0 0 0 6 Shigellosis hsa05131 + 2 1 3 - 0 0 0 12 Small cell lung cancer hsa05222 + 5 1 6 Staphylococcus aureus - 0 0 0 7 hsa05150 infection + 2 3 2 - 0 0 0 9 Steroid hormone biosynthesis hsa00140 + 4 2 3 - 0 1 1 3 Synaptic vesicle cycle hsa04721 + 0 1 0 - 0 0 0 5 Systemic lupus erythematosus hsa05322 + 1 3 1 T cell receptor signaling - 0 0 0 12 hsa04660 pathway + 4 2 6 - 1 1 2 11 TGF-beta signaling pathway hsa04350 + 1 2 4 - 0 0 0 66 TNF signaling pathway hsa04668 + 22 17 27 - 0 1 0 1 Thyroid cancer hsa05216 + 0 0 0 - 0 0 0 1 Thyroid hormone synthesis hsa04918 + 0 0 1 - 0 0 0 5 Tight junction hsa04530 + 2 2 1 Toll-like receptor signaling - 0 0 0 23 hsa04620 pathway + 8 5 10 - 0 0 0 8 Toxoplasmosis hsa05145 + 3 1 4 Transcriptional misregulation in - 0 2 1 42 hsa05202 cancer + 14 10 15

Table 26 is continued on page 195.

196

Appendix

Table 26, continued. KEGG pathway (Homo KEGG Regulation 6h 24h 6h Total sapiens) gene compared and number of (Homo to 24h differentially sapiens) unstimulate expressed d MSC genes in pathway - 0 0 0 9 Tryptophan metabolism hsa00380 + 3 3 3 - 0 1 0 25 Tuberculosis hsa05152 + 8 5 11 - 0 0 0 7 Type I diabetes mellitus hsa04940 + 2 3 2 - 0 0 0 3 Type II diabetes mellitus hsa04930 + 1 1 1 - 0 0 0 3 Tyrosine metabolism hsa00350 + 1 1 1 - 0 0 0 6 Ubiquitin mediated proteolysis hsa04120 + 2 1 3 - 0 0 0 6 VEGF signaling pathway hsa04370 + 2 1 3 Valine, leucine and isoleucine - 0 0 0 3 hsa00280 degradation + 1 1 1 Vascular smooth muscle - 0 3 2 11 hsa04270 contraction + 2 1 3 Vasopressin-regulated water - 0 0 0 1 hsa04962 reabsorption + 0 0 1 - 0 0 0 13 Viral carcinogenesis hsa05203 + 4 1 8 - 0 0 0 3 Viral myocarditis hsa05416 + 1 1 1 Vitamin digestion and - 0 0 0 3 hsa04977 absorption + 1 1 1 - 0 1 0 8 Wnt signaling pathway hsa04310 + 3 1 3 - 0 0 1 2 alpha-Linolenic acid metabolism hsa00592 + 0 0 1 - 0 0 0 5 mTOR signaling pathway hsa04150 + 2 1 2 - 0 0 0 6 p53 signaling pathway hsa04115 + 2 2 2

- 44 58 62 Pathway not determined + 189 146 178

- 12 32 19 Pathway determined + 133 106 149 197

Acknowledgements

The acknowledgements were removed for reasons of data protection.

198

Curriculum Vitae: Elisa Maria Amann

Curriculum Vitae: Elisa Maria Amann The curriculum vitae was removed for reasons of data protection.

199

Curriculum Vitae: Elisa Maria Amann

The curriculum vitae was removed for reasons of data protection.

200