abstract: Eosinophils are innate immune cells that are most commonly associated with parasite infection and allergic responses. Recent studies, though, have identified eosinophils as cells with diverse effector functions at baseline and in disease. Eosinophils in specific tissue immune environments are proposed to promote unique and specific effector functions, suggesting these cells have the capacity to differentiate into unique subtypes. The studies here focus on defining these subtypes using functional, molecular, and genetic analysis as well as using novel techniques to image these subtypes in situ. To characterized these subtypes, an in vitro cytokine induced type 1 (E1) and type 2 (E2) eosinophil model was developed that display features and functions of eosinophils found in vivo. For example, E1 eosinophils secrete type 1 mediators (e.g., IL-12, CXCL9 and CXCL10), express iNOS and express increased levels of the surface molecules PDL1 and MHC-I. Conversely, E2 eosinophils release type 2 mediators (e.g., IL4, IL13, CCL17, and CCL22), degranulate and express increased surface molecules CD11b, ST2 and Siglec-F. Completion of differential expression analysis of RNAseq on these subtypes revealed 500 and 655 unique genes were upregulated in E1 and E2 eosinophils, respectively. Functional enrichment studies showed interferon regulatory factor (IRF) transcription factors were uniquely regulated in both mouse and human E1 and E2 eosinophils. These subtypes are sensitive to their environment, modulating their IRF and cell surface expression when stimulated with opposing cytokines, suggesting plasticity. To identify and study these subtypes in situ, chromogenic and fluorescent eosinophil-specific immunostaining protocols were developed. Methods were created and optimized, here, to identify eosinophils by their granule proteins in formalin fixed mouse tissues. Yet, eosinophil-specific antibodies alone are not enough to identify and study the complex interactions eosinophil subtypes perform within a tissue. Therefore, as part of this thesis, a novel highly-multiplexed immunohistochemistry technique was developed utilizing cleavable linkers to address these concerns. This technique is capable of analyzing up to 22 markers within a single biopsy with single-cell resolution. With this approach, eosinophil subtypes can be studied in situ in routine patient biopsies. / Dissertation/Thesis / Doctoral Dissertation Biochemistry 2020

<p>Serum levels of eosinophil granule proteins namely ECP, EPO and EPX, which are stored in the matrix of specific granules, were shown to correlate with the course of disease in disorders involving eosinophils. The concentration of eosinophil proteins in serum is the result of their release <i>in vivo</i> and <i>ex vivo</i> during the sampling procedure. Generally, eosinophils release the content of their specific granules in three ways: exocytosis, piecemeal degranulation (PM) or cytolysis. Which of them is operating in circulating eosinophils has not yet been defined. The aim of this thesis was to study the mechanisms of granule protein release from blood eosinophils in respect of protein subcellular localization and cell ultrastructure.</p><p>In patients with bacterial infections, serum levels of ECP but not EPO increased, while in patients with viral infections both proteins remained within the range of healthy controls. G-CSF is a cytokine involved in the response mechanism to bacterial but not viral infections. Administration of G-CSF to healthy subjects induced an elevation of eosinophil numbers and a preferential increase of serum EPX and ECP in comparison to EPO.</p><p>The model of PM consists of the stepwise transportation of specific granule contents from the granules towards the plasma membrane. We observed that administration of G-CSF to healthy subjects and the allergen exposure of allergic subjects during the pollen season, caused changes in the ultrastructure of eosinophil specific granules such as loosening of the matrix, granule matrix lucency and ragged losses of their core. Similar alterations of morphology had been previously described for eosinophils undergoing PM.</p><p>ECP, EPX and EPO were localized not only in the specific granules but also in extra-granular compartments as shown both by immuno electron microscopy and subcelular fractionations, An extra-granular EPX compartment was present in healthy as well as in allergic and in hypereosinophilic subjects, and there were no significant differences in its size between the groups. The size of the extra-granular compartments of ECP and EPO was increased in allergics during the season, and these compartments were clearly separate from that of EPX. Results of this show the differential mobilization ofgranule proteins in blood stream eosinophils serum and indicates PM as its mechanism.</p>

Serum levels of eosinophil granule proteins namely ECP, EPO and EPX, which are stored in the matrix of specific granules, were shown to correlate with the course of disease in disorders involving eosinophils. The concentration of eosinophil proteins in serum is the result of their release in vivo and ex vivo during the sampling procedure. Generally, eosinophils release the content of their specific granules in three ways: exocytosis, piecemeal degranulation (PM) or cytolysis. Which of them is operating in circulating eosinophils has not yet been defined. The aim of this thesis was to study the mechanisms of granule protein release from blood eosinophils in respect of protein subcellular localization and cell ultrastructure. In patients with bacterial infections, serum levels of ECP but not EPO increased, while in patients with viral infections both proteins remained within the range of healthy controls. G-CSF is a cytokine involved in the response mechanism to bacterial but not viral infections. Administration of G-CSF to healthy subjects induced an elevation of eosinophil numbers and a preferential increase of serum EPX and ECP in comparison to EPO. The model of PM consists of the stepwise transportation of specific granule contents from the granules towards the plasma membrane. We observed that administration of G-CSF to healthy subjects and the allergen exposure of allergic subjects during the pollen season, caused changes in the ultrastructure of eosinophil specific granules such as loosening of the matrix, granule matrix lucency and ragged losses of their core. Similar alterations of morphology had been previously described for eosinophils undergoing PM. ECP, EPX and EPO were localized not only in the specific granules but also in extra-granular compartments as shown both by immuno electron microscopy and subcelular fractionations, An extra-granular EPX compartment was present in healthy as well as in allergic and in hypereosinophilic subjects, and there were no significant differences in its size between the groups. The size of the extra-granular compartments of ECP and EPO was increased in allergics during the season, and these compartments were clearly separate from that of EPX. Results of this show the differential mobilization ofgranule proteins in blood stream eosinophils serum and indicates PM as its mechanism.

This thesis describes studies in patients with inflammatory bowel disease (IBD) and collagenous colitis (CC). We investigated mucosal eosinophil and neutrophil granulocytes and T-cells involved in the inflammatory processes and aimed at determining whether these processes are reflected in the faecal (F) contents of specific proteins secreted by cells in the intestinal mucosa. Thus, we measured eosinophil cationic protein (ECP) and eosinophil protein X (EPX) and the neutrophil derived myeloperoxidase (MPO) and calprotectin (C); and in addition, chromogranin A (CgA), Chromogranin B (CgB) and secretoneurin (SN), derived from EEC cells and cells in the enteric nervous system. We found that a normalised FC level can serve as a surrogate marker for successful treatment in patients with IBD, but persistently high FC levels need further evaluation (study I). Furthermore, FC and F-MPO appear to relate better than F-EPX to treatment outcome in IBD. We evaluated F-ECP, F-EPX, F-MPO and FC as markers of disease activity and treatment outcome in patients with CC (study III) and concluded that F-ECP was the best discriminator of detecting active CC. Normalised F-ECP and F-EPX could serve as markers of successful treatment. We showed that the inflammation in CC is characterised by activated eosinophils, but that there is no neutrophil activity (study II). T-cells have a lower grade of activity in active CC than in control subjects. During budesonide treatment the normal activation of eosinophils and T-cells is restored, with concomitant clinical remission. The findings in studies II and III indicate that the eosinophils have an essential role in the pathophysiology of CC. Markedly higher values of F-CgA, F-CgB and F-SN were found in patients with CC than in those with IBD and controls (study IV) indicating a crucial role for the intestinal neuro-endocrine system in the pathogenesis of collagenous colitis.