Embed Size (px)
Citation preview
Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 971
_____________________________ _____________________________
Cytokine-Regulated Eosinophil Migration in Inflammatory Disorders
Clinical and Experimental Studies
BY
MARIA LAMPINEN
ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2000
2
Dissertation for the Degree of Doctor of Philosophy (Faculty of Medicine) presented at UppsalaUniversity in 2000
ABSTRACT
Lampinen, M. 2000.Cytokine-regulated eosinophil migration in inflammatory disorders. Clinical andexperimental studies. Acta Universitatis Upsaliensis. Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine 971. 60 pp. Uppsala. ISBN 91-554-4849-6.
The accumulation of eosinophil granulocytes (EOS) at sites of inflammation is acommon feature of asthma, allergic rhinitis and inflammatory bowel disease. The aimof the present investigation was to study the mechanisms involved in thisaccumulation.Bronchoalveolar lavage (BAL) fluid obtained from patients with birch-pollen allergyduring the season exhibited increased eosinophil chemotactic activity compared withpre-season BAL fluid from the same patients. We identified IL-5, IL-8 and RANTESas the main eosinophil chemotactic agents in the BAL fluid. Only EOS from allergicdonors responded to IL-8. IL-2 inhibited albumin-stimulated eosinophil migrationtowards buffer or chemoattractants. EOS from allergic subjects were less sensitive tothis inhibition than EOS from normal subjects, and in vitro priming of the EOS withIL-5 prevented the inhibitory effect of IL-2. We therefore hypothesise that IL-2 actsas an autocrine regulator of EOS migration, and that this inhibitory effect may bedown-regulated in allergy, resulting in increased migration of EOS towardschemotactic factors. The stimulation of eosinophil migration by albumin is mediatedby PI3 kinase. Decreased expression of CD49d and CD49f caused by albumin maydecrease the adhesiveness of the EOS, which in turn may facilitate migration. Wefound higher chemotactic activity in perfusion fluids from patients with ulcerativecolitis than from control patients. The chemotactic activity correlated with theconcentrations of eosinophil granule proteins in the perfusion fluids. IL-5 and TNF-αwere identified as two of the chemotactic agents in the perfusion fluid that wereinhibited by steroid treatment. Agents with steroid-insensitive chemotactic activityremain to be identified.
Key words: Eosinophils, chemotaxis, birch-pollen allergy, asthma, cytokines, migration, interleukin-2,albumin, adhesion molecules, ulcerative colitis, inflammatory bowel disease.
Maria Lampinen, Department of Medical Sciences, Clinical Chemistry, University Hospital, SE-751 85Uppsala, Sweden
© Maria Lampinen 2000
ISSN 0282-7476ISBN 91-554-4849-6
Printed in Sweden by University Printers, Uppsala 2000
3
To Pablo and Clara
4
This thesis is based on the following papers, which will be referred to inthe text by their Roman numerals:
I. Venge J, Lampinen M, Håkansson L, Rak S, Venge PIdentification of IL-5 and RANTES as the major eosinophilchemoattractants in the asthmatic lung.Journal of Allergy and Clinical Immunology 1996, 97, 1110-1115
II. Lampinen M, Rak S, Venge PThe role of interleukin-5, interleukin-8 and RANTES in thechemotactic attraction of eosinophils to the allergic lung.Clinical and Experimental Allergy 1999, 29, 314-322
III. Lampinen M, Håkansson L, Venge PInterleukin-2 inhibits eosinophil migration but is counteracted byIL-5 priming.Clinical and Experimental Allergy (In press)
IV. Lampinen M, Carlson M, Sangfelt P, Taha Y, Thörn M, Lööf L,Raab Y, Venge PIL-5 and TNF-α participate in the recruitment of eosinophils to theintestinal mucosa in ulcerative colitis.Submitted for publication
V. Lampinen M, Håkansson L, Venge PMechanisms behind albumin-stimulated eosinophil migration.Submitted for publication
Reprints were made with the permission of the publishers.
5
CONTENTS
ABBREVIATIONS………………………………………… 6
INTRODUCTION……………………………………….. 7
Basic mechanisms of allergic inflammation…………… 7Bronchial asthma���������������� 8
Ulcerative colitis………………………………………… 9
The eosinophil granulocyte…………………………… 10
Cytokines and eosinophil function……………………… 10Interleukins and GM-CSF������������� 11Chemokines������������������... 11Interferons������������������� 12Tumour necrosis factors�������������.. 12
Eosinophil surface receptors……………………………. 13Adhesion molecules���������������. 13Cytokine receptors���������������... 15Other cell surface receptors������������.. 16
Eosinophil recruitment………………………………… 17Eosinophil priming���������������. 17Eosinophil rolling and adhesion����������. 17Eosinophil diapedesis and chemotaxis�������� 18
AIMS OF THE PRESENT INVESTIGATION……… 21
PATIENTS AND REFERENCE SUBJECTS………… 22
METHODS………………………………………………. 25
RESULTS………………………………………………... 29
GENERAL DISCUSSION……………………………… 39
SUMMARY…………………………………………………. 45
ACKNOWLEDGEMENTS…………………………… 46
REFERENCES…………………………………………... 48
6
ABBREVIATIONS
Ig ImmunoglobulinPAF Platelet activating factorLT LeukotrieneTNF Tumour necrosis factorIL InterleukinGM-CSF Granulocyte macrophage-colony stimulating factorTh T helperECP Eosinophil cationic proteinEPO Eosinophil peroxidaseEPX/EDN Eosinophil protein x / eosinophil derived neurotoxinHES Hypereosinophilic syndromeCLC Charcot Leyden crystalIFN Interferon gammaRANTES Regulated on activation, normal T-cell-expressed, and secretedMCP Monocyte chemotactic proteinMIP Macrophage inflammatory proteinVCAM Vascular cell adhesion moleculeICAM Intercellular adhesion moleculeVLA Very late antigenJAK Janus family of tyrosine kinasesSTAT Signal transducers and activators of transcriptionIP3 Inositol 1,4,5-triphosphatePKC Protein kinase CGTP Guanosine triphosphateBAL BronchoalveolarBHR Bronchial hyperresponsivenessHSA Human serum albuminRIA RadioimmunoassayLIF Leukaemia inhibitory factorFS Forward scatterSS Side scatter
7
INTRODUCTIONThe eosinophil granulocyte was discovered by Paul Erlich in 1879 [1],and its involvement in the defence against parasites was soon recognised.Not until almost 100 years later did investigators begin to consider theeosinophil as a cell playing an active role in inflammatory processes inthe body. The central role of eosinophils in the pathogenesis of asthmaand allergic rhinitis is now established [2-5], and there is increasingevidence of the involvement of these cells in inflammatory bowel disease[6;7].The accumulation of eosinophils at sites of inflammation is one of thehallmarks of these diseases, and the studies presented in this thesisfocused on the highly regulated and complex process of eosinophilrecruitment. The general introduction will include a brief discussion ofthe mechanisms involved in mucosal inflammation, particularly inbronchial asthma and ulcerative colitis. It will also deal with theeosinophil, with a short description of its general features, membranereceptors and migratory functions.
Basic mechanisms of allergic inflammationAllergic inflammation is initiated by IgE-related mechanisms and ischaracterized by infiltration of inflammatory cells into the target organ[8;9]. The initial allergic reaction mediated by IgE is the sensitisationphase, in which specific IgE directed against a particular antigen isformed in genetically predisposed individuals. The effector phase mayoccur years later, and is triggered by exposure to antigen. Thisinflammatory reaction may consist of an immediate and a late-phasereaction [9;10]. The immediate reaction is induced by mast cells, whichrelease mediators in response to cross-linking of surface-bond IgE byantigen. The mediators, such as histamine and leukotrienes, cause anincrease in vascular permeability, oedema and muscle contraction, aprocess that lasts a few hours and does not cause tissue destruction. Insome cases a late-phase reaction takes place 4 to 24 hours after theimmediate response. The late phase is characterized by tissue infiltration
8
by inflammatory cells, predominantly T lymphocytes and eosinophils[11;12]. The interaction of these cells with resident cells such as mastcells, macrophages, epithelial cells and endothelial cells generates acascade of events that contributes to chronic inflammation and tissuedestruction [13;14]. Mediators involved in the inflammatory reactions arehistamine, platelet activating factor (PAF), prostaglandins, cysteinylleukotrienes (LTC4, LTD4 and LTE4), toxic oxygen radicals, andcytokines[15;16]. Macrophages are able to produce a variety ofproinflammatory cytokines, e.g. tumour necrosis factor (TNF),interleukin (IL)-1β, IL-2, IL-6 and granulocyte macrophage-colonystimulating factor (GM-CSF) [17]. Endothelial and epithelial cellsstimulated by cytokines express increased numbers of adhesion moleculesimportant for the accumulation of inflammatory cells. They are also ableto produce proinflammatory mediators. Hence, these cells not only aretargets for tissue destruction, but also take an active part in theinflammatory process [18-20]. The T lymphocytes, especially T helper(Th)-2 type cells, have a prominent role in allergic responses, producingcytokines such as IL-3, IL-4, IL-5 and GM-CSF. IL-4 enhances synthesisof IgE by antigen-stimulated B lymphocytes, and also favours productionof Th2-type cytokines. IL-3, IL-5 and GM-CSF promote eosinophilaccumulation in the tissue [21-24].The role of eosinophil granulocytes in the inflammatory reaction isassociated with the release of many different mediators, which will bediscussed later in this chapter. Briefly, toxic proteins, oxygen radicals andlipid mediators as well as an array of proinflammatory cytokines areproduced and released by activated eosinophils [25].
Bronchial asthmaChronic mucosal inflammation plays an important role in thepathogenesis of asthma, which is a disease characterized by reversibleairway obstruction and hyperresponsiveness to external and endogenousstimuli [5;9]. Histopathological findings include massive infiltration byactivated lymphocytes and eosinophils in and around the bronchial
9
epithelium, dilatation of blood vessels, loss of surface lining epithelium,collagen deposition beneath the basement membrane of the epithelium,mucosal oedema, and hypertrophy of both submucosal glands andbronchial smooth muscle [5]. Activation of T lymphocytes andsubsequent eosinophil recruitment and secretion may contribute toepithelial cell damage and possibly also to bronchial hyper-responsiveness in asthma. The presence of eosinophils and their productsin the airways has been found to be positively correlated to the diseaseseverity and the development of airway hyper-reactivity [25;26].
Ulcerative colitisUlcerative colitis is an inflammatory disease, the aetiology of which hasnot yet been established. The inflammation is manifested as oedema,congestion, spontaneous bleeding, and erosions of the colorectal mucosa.Infiltration of the mucosa by activated lymphocytes, macrophages,neutrophils and eosinophils is a characteristic feature of the inflammatorycondition, reminiscent of allergy. Furthermore, the inflammation inulcerative colitis is recurrent and paroxysmal, another similarity toinflammatory diseases with an allergic component, such as bronchialasthma [6;27]. An association has actually been suggested betweenallergy and ulcerative colitis on the basis of the finding of a higherprevalence of allergic symptoms in patients with the latter disease than incontrol subjects [28]. The neutrophil infiltration is, however, morepronounced in ulcerative colitis than in allergy, where eosinophilinfiltration predominates. The involvement of neutrophils in thepathogenesis of ulcerative colitis has been studied extensively [29-32],but there is also increasing interest in the role of eosinophils in thisdisease. Morphological studies have provided evidence of eosinophilactivation in inflammatory bowel disease [33;34], and increased releaseof eosinophil cationic protein (ECP), eosinophil peroxidase (EPO) andeosinophil protein X / eosinophil derived neurotoxin (EPX/EDN) hasbeen found in ulcerative colitis [35].
10
The eosinophil granulocyteThe eosinophil granulocyte is a white blood cell whose main function isactually to kill invading parasites. In the absence of parasites, however,activated eosinophils may cause tissue destruction and inflammation[36;37]. Indeed, a number of inflammatory diseases are associated witheosinophilia, for example asthma, allergic rhinitis, atopic skin diseases,idiopathic hypereosinophilic syndrome (HES) and inflammatory boweldisease [6;38].The eosinophils are produced in the bone marrow, where they developfrom primitive stem cells, which divide and differentiate until theyacquire the typical features of eosinophils, namely a bilobed nucleus anddistinct granules of varying sizes [39]. The mature eosinophils enter theblood stream, where they remain for about 25 hours before migrating intothe tissue [40]. In the circulation, they constitute 1-3 % of all white bloodcells, but the eosinophils are primarily tissue cells, and there areapproximately 500 times as many eosinophils in tissues as in the blood[37].Activated eosinophils are able to phagocytose particles, such as bacteria,but their main killing mechanism is the release of toxic granule proteinsand production of oxygen free radicals [39]. There are five majoreosinophil granule proteins, namely ECP, MBP, EPX/EDN, EPO and theCharcot-Leyden crystal (CLC) protein. MBP and CLC protein are alsofound in basophils, whereas ECP, EPO and EPX/EDN are specific foreosinopils. All of these proteins are able to kill both mammalian cells andnon-mammalian cells with varying efficiency. Eosinophils also producecytokines and mediators such as leukotriene C4 (LTC4) and PAF [3].
Cytokines and eosinophil functionCytokines are low-molecular-weight glycoproteins that stimulate orinhibit the differentiation, proliferation or function of immune cells. Eachcytokine has multiple functions, many of which are carried out by morethan one cytokine [16;21]. This redundancy of the system means that thenet effect of the cytokines is dependent on the balance of cytokine
11
expression. Cytokines may also interact both synergistically andantagonistically [16].Cytokines can be divided into:
• interleukins• colony-stimulating factors• chemokines• interferons (IFNs) and• tumour necrosis factors
Interleukins and GM-CSFAmong the interleukins, IL-4 and IL-5 are derived from a subset of T-helper cells called Th2, which have been implicated in the pathogenesisof asthma and allergy. IL-4 induces IgE production and IL-5 has a crucialrole in the proliferation, differentiation, survival and activation ofeosinophils. Another subset of T-helper cells, Th1, produce IL-2 andIFN-γ, but not IL-4 or IL-5. The Th1 cytokines are involved inmacrophage activation and cytotoxicity, and are thus active in the defenceagainst bacterial infections. Both subsets produce IL-3 and GM-CSF [22-24]. IL-3 and GM-CSF stimulate the development of eosinophils as wellas of other leucocytes, whereas IL-5 acts specifically on eosinophils [41].
ChemokinesThe chemokines constitute a superfamily of chemoattractant cytokines.They have been categorised on the basis of whether the first two cysteinesare adjacent to each other or separated by an amino acid. The two mostimportant subgroups are the CXC (α-chemokines) and the CC (β-chemokines). The only one of the CXC chemokines that has beenassociated with eosinophil function is IL-8 [42;43]. In contrast, a numberof CC chemokines have been found to be related to eosinophils:• RANTES (regulated on activation, normal T-cell-expressed, and
secreted)• Eotaxin• Eotaxin-2
12
• Monocyte chemotactic protein (MCP)-1 and-3• Macrophage inflammatory protein (MIP)-1αThese chemokines stimulate the migration of eosinophils, basophils, Tcells and/or monocytes, but not of neutrophils [44;45].
InterferonsIFNs are produced during viral or bacterial infection; IFN-α mainly byleucocytes, IFN-β by fibroblasts and IFN-γ by T cells. IFN-γ plays acritical role in the immune response and is one of the Th1-specificcytokines that promote Th1 responses and inhibit Th2 responses. IFN-γalso stimulates macrophages to produce other proinflammatory cytokines[45].
Tumour necrosis factorsTNFs are endotoxin-induced serum factors that are able to cause necrosisof tumours. TNF-α is produced by monocytes / macrophages, and TNF-βby T cells [45]. TNF-α increases the ability of eosinophils to kill parasites[46] and has also been shown to stimulate eosinophil migration [47] andto up-regulate the expression of vascular cell adhesion molecule(VCAM)-1, which is important for eosinophil recruitment to sites ofinflammation [16].
13
Eosinophil cell surface receptorsEosinophils express a variety of receptors on their external plasmamembranes, including receptors involved in adhesion. They also expressreceptors for cytokines, complement and immunoglobulins [48].
Adhesion moleculesAdhesion of leucocytes to the blood vessel endothelium by means ofadhesion molecules is a prerequisite for the accumulation of these cells inthe tissue. Receptors involved in eosinophil adhesion can be divided intoselectins, integrins and immunoglubulin-like structures [49].The selectin family consists of L-selectin (CD62L), P-selectin (CD62P)and E-selectin (CD62E). They mediate "rolling", i.e. the initial andreversible contact between eosinophils and the endothelial cells of theblood vessel [48]. Selectins are characterised by a Ca2+-dependent lectindomain which is able to bind to glycoproteins that contain thecarbohydrate structure sialyl Lewis X. L-selectin is constitutivelyexpressed on eosinophils and is shed upon activation during eosinophilrecruitment, which may explain the reversible nature of the binding. L-selectin interacts with MAdCAM-1, CD13 and with the other selectins. P-and E-selectins are expressed on activated endothelial cells. The mainligand for P-selectin is CD162 (P-selectin glycoprotein ligand-1 orPSGL-1), and E-selectin binds to sialyl Lewis X on the eosinophil [48].During rolling the eosinophils become activated and the expression ofintegrins is up-regulated. These adhesion molecules enable theeosinophils to bind firmly to extracellular matrix components or to theimmunoglobulin-like structures on other cells [48]. They also mediateintracellular signalling [50]. Integrins are composed of α and β subunits.They are subdivided according to the β chain, which are linked todifferent α chains. Eosinophils express the β1 integrins α4β1
(CD49d/CD29), a heterodimer called very late antigen (VLA)-4, and α6β1
(VLA-6), enabling them to interact with the extracellular matrix proteinslaminin, fibronectin and fibrinogen. VLA-4 also binds to VCAM-1 andmediates selective eosinophil recruitment [48;50;51]. The common β
14
chain of the β2 integrins is CD18. Among the β2 integrins, eosinophilsexpress αLβ2 (CD11a/CD18), αMβ2 (CD11b/CD18), αXβ2 (CD11c/CD18)and αDβ2. Their counterstructures are intercellular adhesion molecule(ICAM)-1, -2 and -3 as well as VCAM-1, fibrinogen and the complementfragment C3bi [48].Immunoglobulin-like structures involved in adhesion have been observedon eosinophils. For example ICAM-1 has been found to be upregulatedon eosinophils from asthmatic subjects or in vitro by stimulation withTNF-α together with IL-3, IL-5 or GM-CSF [52]. Eosinophils alsoexpress ICAM-3 and PECAM (CD31).
Table I. Eosinophil adhesion molecules and their ligands.Eosinophil receptor Ligand
Selectins L-selectin MAdCAM-1, CD34
Integrins VLA-4 (CD49d/CD29) VCAM-1, fibronectinVLA-6 (CD49f/CD29) LamininCD11a/CD18 ICAM-1,-2,-3CD11b/CD18 ICAM-1, fibrinogen,
C3biCD11c/CD18 Fibrinogen, C3biα4β7 MAdCAM-1, VCAM-
1, fibronectin
Immunoglobulin-like ICAM-1/-3, PECAM PECAM-1, Αvβ3
Carbohydrate PSGL-1 P-selectinSialyl Lewis X E-selectin
Others CD44 hyaluronate
15
Cytokine receptorsMost cytokine receptors consist of two or more subunits, one of which isligand-specific. The other subunits are associated with signal generationand transduction. Several cytokine receptors use a common signaltransducer, which may explain the redundancy of the cytokine functions.With few exceptions (e.g. the IL-8 receptor), cytokine receptors do notcontain intrinsic tyrosine kinases. Instead, dimerisation of the receptorcomponents results in activation of cytoplasmic protein tyrosine kinases,so-called JAKs (Janus family of tyrosine kinases), which in turn activatemembers of the STAT family (signal transducers and activators oftranscription) or induce the Ras-mitogen-activated protein kinase cascade[21;45;53]. Cytokine receptors expressed on human eosinophils are listedin Table II. IL-3, IL-5 and GM-CSF share a common β chain and havetheir own cytokine-specific α chains. The IL-5 receptor is fairly specific,since it is only expressed on eosinophils and basophils, while receptorsfor IL-3 and GM-CSF are present on many haematopoietic cells [54]. Thereceptor for IL-2 consists of three subunits (α, β and γ), of which the αchain (CD25) has been detected on eosinophils [55;56]. The γ chain thatis a common subunit shared by the receptors for IL-2, IL-4 and IL-7 isalso expressed on eosinophils [57]. The chemokine receptors are namedCCR or CXCR, according to what kind of chemokines they bind. Atpresent, five CXC receptors and eight CC receptors are known [58].Among the CXC receptors, the receptor for IL-8 (CXCR-2) has beenfound on activated eosinophils [59]. CCR-1 and CCR-3 are expressed oneosinophils, basophils and activated lymphocytes. CCR-1 is also presenton macrophages. Chemokines binding to CCR-1 include MIP-1α,RANTES and MCP-3. CCR-3 is the principal receptor for eotaxin andeotaxin-2, and also binds RANTES and MCP-2, -3 and -4 [58;60].
16
Table II: cytokine receptors (R) on human eosinophilsIL-2R α (CD25) and γ chains
IL-3R α chainIL-5R α chain + common β chainGM-CSFR α chain
IL-4R α and γ chainsIL-9RIL-13RIFN-γRTNF-αRCXCR-2 (on activated eosinophils)CCR-1CCR-3TGF-β
Other cell surface receptorsEosinophils express receptors for chemoattractants such as PAFleukotriene (LT) B4, D4 and E4, complement factors C5a and C3a and f-Met-Leu-Phe [48]. Among the immunoglobulin receptors, eosinophilsexpress Fc receptors for IgG (CD32), IgA (CD89) and IgE (CD23). Thesereceptors are involved in processes such as degranulation, respiratoryburst activation and phagocytosis [61]. Furthermore, eosinophils expressapoptosis-inducing structures such as CD95 (fas) and CD69, signallingstructures with different functions and cell surface enzymes. Thedescription of all these surface molecules is beyond the scope of thisintroduction, but they are excellently reviewed elsewhere [48].
17
Eosinophil recruitmentThe recruitment of eosinophils to inflamed tissue is a multi-stage processthat can be divided into the following events:1. Priming of the eosinophils in the circulation2. Rolling along the endothelial cells3. Firm adhesion to the endothelium4. Trans-endothelial diapedesis5. Chemotaxis into the inflammatory site
Eosinophil primingPeripheral blood eosinophils of normal, healthy subjects are relativelyrefractory to activation. However, they can change into an activation-sensitive, or primed, phenotype after interaction with certain cytokines.Eosinophils from patients with allergy and asthma exhibit increasedmigratory responses [62;63], adhesiveness [64] and degranulation [65]compared with eosinophils from normal subjects. The priming is likely tobe caused by IL-3, IL-5 and GM-CSF in the peripheral blood. Thesecytokines have been detected in the blood of allergic subjects [66;67], andpriming has been mimicked in vitro by use of recombinant IL-3, IL-5 andGM-CSF, resulting in increased migratory responses to chemoattractants[68;69] and increased degranulation [70]. Priming of eosinophils can alsoinduce responses to factors such as IL-8, which have no effect onunprimed eosinophils [69]. One mechanism underlying this change inresponsiveness may be the induction of receptors on the cell surface [59].Another function of IL-3, IL-5 and GM-CSF is to prolong the survival ofeosinophils by postponing apoptosis [71]. This may contribute to theincreased numbers of eosinophils in the tissue.
Eosinophil rolling and adhesionRolling, the initial contact of circulating eosinophils with the blood vesselwall prior to extravasation, is mediated by selectins. The reversibleadhesion between the selectins and their ligands makes the eosinophilsmove slowly along the vascular endothelium. The expression of E-and P-
18
selectins on the endothelial cells can be up-regulated by IL-1 and TNF-α,while L-selectin is constantly expressed on the eosinophils [72;73]. Theslow movement along the endothelium helps the eosinophils to adherefirmly by interaction of integrins and their ligands. Rolling is alsobelieved to activate the integrin receptors on the eosinophils, whichresults in high-affinity binding to their ligands [72]. The molecules thatare most important for firm adhesion of eosinophils to the endothelialcells are CD11b/CD18 and VLA-4 on the eosinophils, and their ligandsICAM-1 and VCAM-1 on the endothelial cells. Expression of adhesionmolecules is stimulated by cytokines. For example, IL-1 and TNF-αinduce ICAM-1 and VCAM-1 expression on endothelial cells. Somecytokines are selective inducers for certain adhesion molecules, such asIL-4 for VCAM-1 and IFN-γ for ICAM-1 [74]. Increased expression ofICAM-1 and VCAM-1 has been found in biopsy samples both frompatients with allergic asthma [74] and from patients with inflammatorybowel disease [75]. Thus, the adhesion of eosinophils to endothelial cellsin inflammatory diseases is promoted by activation both of theendothelial cells and of the eosinophils.
Eosinophil diapedesis and chemotaxisEosinophils from the blood of normal individuals can adhere to IL-1- orTNF-α -activated endothelial cells, but they are unable to migrate throughthis layer. In contrast, eosinophils from allergic or asthmatic individualsdo not only adhere to the endothelial cells, but also pass spontaneouslythrough the cell layer. This capacity to transmigrate can be induced bypretreatment of eosinophils from normal individuals with IL-3, IL-5 orGM-CSF [76]. Thus, priming of the eosinophils is a prerequisite fordiapedesis. The expression of VCAM-1 and ICAM-1 on the endothelialcell is also necessary for eosinophil transmigration.Eosinophils then migrate into the tissue in response to chemotactic factorsproduced locally at the inflammatory site [27;77-82]. This migration is amovement directed towards chemoattractant gradients, and ischaracterised by adhesion/de-adhesion to extracellular matrix proteins. In
19
response to a chemoattractant, the concentration of cytoplasmic Ca2+ risesand the cell undergoes polarisation. A flat extension of the cytoplasm,called a lamellipod, is formed at the front of the moving cell, and a tailcalled uropod is formed at the rear. The nucleus is located closer to therear than to the front [83;84]. In polarised locomoting cells there isalways a calcium gradient with higher Ca2+ in the tail than in the front.When cells turn, the intracellular Ca2+ concentration rises transiently andthen falls most rapidly in the new direction of locomotion. These Ca2+
signals arise from internal Ca2+ stores released in response to inositol1,4,5-triphosphate (IP3). Diacylglycerol (DAG), which is co-producedwith IP3, has an inhibitory effect on Ca2+ signals, probably throughprotein kinase C (PKC). The eosinophil moves by extending thelamellipod that adheres to the substrate, and constricting the uropod whilede-adhering from the substrate [83-85]. The actin system comprises themotor of the cell, and actin polymerisation and depolymerisation providethe cell with contractile forces. These forces are probably induced byphosphorylated myosin II interacting with actin filaments [86]. Manysignal transduction mechanisms, e.g. calcium transients, pH changes,protein phosphorylation reactions, phospholipid turnover, lipoxygenaseactivity, GTP-binding proteins, and others, have been proposed asmediators of these events [83].Cell migration may be divided into chemotaxis and chemokinesis.Chemotaxis is a directed movement towards a gradient ofchemoattractant. Chemokinesis, on the other hand, is an undirected, orrandom, movement that represents cell activation and subsequentmovement without the need of a chemotactic gradient [87]. Chemotaxisand chemokinesis can be differentiated by means of a chequerbordanalysis [88].At the time of the start of this study, several factors had been suggested asbeing chemotactic for eosinophils. PAF, complement factor C5a, LTB4
and f-Met-Leu-Phe were some of the first-identified eosinophilchemoattractants [89]. All of these mediators, however, act on botheosinophils and neutrophils. Recently, PAF and C5a have been found to
20
be important for the transmigration of eosinophils across epithelial cells[90]. When the present study began, increasing interest had already beenpaid to the chemotactic properties of cytokines. IL-3, IL-5 and GM-CSFhad been recognised as activators of eosinophil function, includingmigration [89;91]. Chemokines such as RANTES [92] and IL-8 [93] hadalso been found to be eosinophil chemoattractants, although at the time ofthose observations human eotaxin had not yet been discovered. IL-2 hadalso been claimed to induce eosinophil migration [55], an opinion thatwill be challenged in this thesis.
21
AIMS OF THE PRESENT INVESTIGATION
The overall aim of this investigation was to study mechanisms involvedin the attraction of eosinophil granulocytes to sites of inflammation. Thespecific aims were:
• to identify the eosinophil chemotactic agents in broncho-alveolarlavage (BAL) fluid obtained from pollen-allergic patients duringthe birch-pollen season;
• to compare eosinophils from normal and allergic individualsregarding the responsiveness to these chemotactic agents;
• to study interactions between different proinflammatory cytokines;
• to investigate the effect of interleukin-2 on eosinophil migration;
• to analyse intestinal perfusion fluid from patients with ulcerativecolitis regarding eosinophil chemotactic activity;
• to identify mechanisms by which human serum albumin stimulateseosinophil migration.
22
PATIENTS AND REFERENCE SUBJECTS
Papers I and II
Patients (for BAL-fluid studies):Eight birch pollen-allergic patients, four men and four women, with amean age of 29 years (range 21-37), participated in these studies. Twopatients were occasional smokers. The diagnosis of allergy was based ona history of rhino-conjunctivitis and wheezing during the birch-pollenseason, a positive skin prick test and RAST. All patients exhibitedincreased bronchial hyper-reactivity (BHR), within the �asthmatic range�,as measured by means of histamine sensitivity at the time of admission tothe study. BHR was further increased during the pollen season. Thepatients underwent BAL twice; before the birch-pollen season (inFebruary and March 1988) and during the birch-pollen season (in May1988).
Blood donors:Only healthy, non-allergic donors were included in study I. In study II,blood from either normal healthy donors or from donors with presumablyin vivo primed eosinophils was used. The normal group consisted of eightmen and five women, mean age 37 years (range 24 to 61). The "primed",or allergic group comprised four men and 12 women, mean age 39 years(range 25 to 76). Blood from two pollen-allergic blood donors wasobtained during the birch-pollen season. Five of the allergic subjects wereat the beginning of a hyposensitization treatment and were receivingallergen injections, and the remaining nine had perennial allergy withrhino-conjunctivitis, asthma or atopic eczema. None of the asthmaticsubjects were recieving inhaled corticosteroids at doses higher than 800µg Budesonide daily.
23
Paper III
Blood donors:Blood from either normal healthy donors or from allergic donors wasused. The normal group consisted of ten men and eight women, mean age36 years (range 22 to 62 years). The allergic group comprised four menand 17 women, mean age 33 years (range 21 to 50). Blood from ninepollen-allergic individuals was obtained during the birch-pollen season.Six of the allergic patients were at the beginning of a hyposensitizationtreatment and were receiving allergen injections, and the remaining sixhad perennial allergy with rhinoconjunctivitis, asthma or atopic eczema.None of the asthmatic patients were receiving inhaled corticosteroids atdoses higher than 800 µg Budesonide daily.
Paper IV
Patients (for studies on perfusion fluid):Eleven patients with ulcerative colitis, six women and five men, meanage 45 years (range 30-69), participated in the study. Six of the patientshad proctitis and five had distal colitis. The clinical disease activity wasmild to intermediate, with diarrhoea in most cases but with blood in thestools less frequently. All patients had endoscopically active disease, witha score of 3-4 according to the Binder system [94]. Nine patients receivedno treatment for at least eight weeks prior to inclusion in the study andtwo patients were receiving a stable oral mesalazine preparation. Onepatient was taking atenolol 50 mg once a day for arterial hypertension.Intestinal perfusion fluid was collected on three occasions from allpatients; first before starting treatment with steroid enema (Pred-Clysma ) (day 0), secondly on the 7th day of treatment, and thirdly on the28th day of treatment.
24
Reference subjects:Seven patients without a history of or current inflammatory boweldisease, five women and two men, mean age 57 years (range 45-71), wereincluded in the study as controls. They underwent total colonoscopy forinvestigation of gastrointestinal bleeding or for follow-up afterpolypectomy.
Blood donors:Only blood from normal, healthy donors was used in this study.
Paper V
Blood donors:Only blood from normal, healthy donors was used in this study.
25
METHODS
Isolation of granulocytes (BAL experiments in studies I and II)Heparinised blood was obtained from normal or allergic donors.Granulocytes were isolated by dextran sedimentation [95]. Remainingerythrocytes were lysed, and the granulocytes were washed withphysiological saline. The granulocytes were diluted to a concentration of
1.5 x 109/L in Gey's solution with human serum albumin (HSA; 1g/L).
Purification of eosinophilsEosinophils were isolated using an immunomagnetic bead technique aspreviously described [96]. Briefly, 80 ml of heparinised blood wasobtained from normal, healthy blood donors or allergic donors, andgranulocytes were isolated by Percoll (1.082 g/ml) gradientcentrifugation. Red blood cells were removed by hypotonic lysis, andeosinophils were subsequently isolated using a method based on the factthat eosinophils, in contrast to neutrophils, lack the cell surface antigenCD16. As a result, highly purified eosinophils can be isolated byremoving neutrophils bound to immunomagnetic beads coated with anti-CD16. The eosinophil purity was 99 ± 1%, and their viability 99 ± 1%, asdetermined by staining with Trypan Blue.
Cell counting and quantitationTotal blood cell counts were made with a Technicon H1 cell counter(Tournai, Belgium) (papers I-III) or a Coulter STKS (Coulter Electronics,Hialeah, FL, USA) (papers IV-V). Cells stained with Türks dye werecounted under a light microscope. Differential counts were obtained byuse of a cytospin preparation (Cytospin, Shandon Southern Instruments,Sewickley, USA). The cells were stained with May Grünewald andGiemsa stain and examined under a light microscope.
26
Chemotaxis assayEosinophil chemotaxis was measured by a modification of the Boydenchamber technique [97], using a 48-well microchemotaxis chamber(Neuro Probe, Cabin John, MD, USA). Chemotaxins or Gey's solution(28 µl) was added to the lower compartments. A nitrocellulose filter(Millipore Corporation, Bedford, MA, USA) with a pore size of 5µm wasplaced between the upper and lower compartment. The cell suspensions,50 µl per well, were added to each upper well. All tests were performedin duplicate. The incubation time was 60 min and the temperature 37°C.Migration was assayed according to the leading front technique [88].Eosinophils in granulocyte mixture were identified by staining withChromotrope 2R [98].
Adherence assay (paper III)The adherence of eosinophil granulocytes to E-selectin and ICAM-1 wasmeasured by a recently described method [99] utilising cell lines ofhamster kidney fibroblasts that selectively express E-selectin and ICAM-1. As a reference the adhesion to the hamster fibroblasts without anytransfected adhesion molecules was also measured, and referred to asbasal adhesion.
Priming of eosinophils with IL-5To mimic the primed state of eosinophils from allergic and asthmaticdonors, normal eosinophils were preincubated with IL-5.In study II, purified eosinophils diluted in Gey's solution with human
serum albumin (1g/L) to a concentration of 1 x 109/L were incubatedwith recombinant human IL-5 at different concentrations for 12 min atroom temperature and the cell suspensions were transferred to chambersfor measurement of migration.In study III, purified eosinophils were incubated for 30 min at 37ûC withIL-5 (10-10M) or Gey�s solution before incubation of the cells with IL-2(10-11M) and subsequent migration assay.
27
Intestinal perfusion technique (paper IV)Subjects were prepared as for endoscopy: that is, they drank 2 to 4 litres(until a clear fluid stool was passed) of a non-irritative isotonic fluid withan osmotic effect (Laxabon�, Tika Läkemedel Ltd, Lund, Sweden).Meperidine (25-50 mg) and/or diazepam (5-10 mg) was givenintravenously as premedication, to those who so desired. The perfusiontechnique has been described in detail elsewhere [100]. In the presentstudy the perfusion tube was modified and only a rectal segment was usedfor perfusion. Channels in the wall of the tube were used either to inflatethe balloons with air, to perfuse and drain the rectal segment, or toadminister a dye marker at the tip of the tube for detection of any leakageinto the rectal segment from the sigmoid colon. The open main channel ofthe tube allowed passage of gas and fluids from the intestine above to thetube during perfusion. The perfusion tube, with the endoscope acting as aguide, was introduced into the rectum to the predetermined position,whereafter the balloons were inflated with air and the endoscope waswithdrawn. In the control group, the perfusion segment was attached to acolonoscope and the rectal perfusion was performed before the rest of thecolon was examined. The segment was first filled with 60 mL (estimatedvolume of the segment) of a buffer (120 mM NaCl, 5.4 mM KCL, 2 mMNa2HPO4, 10 mM glucose, 35 mM mannitol and 1g/L polyethyleneglycol (Mw 4 kD), pH 8.16-8.50 and osmolality 273 mosm/L) at 37°Cand then perfused with the same buffer at a rate of 3 ml per min. After arinsing period the perfusion fluid was collected for 20 min by gravitydrainage. Ten millilitres of aprotinin (10, 000 KIU/mL) was added toevery litre of perfusion buffer to inhibit proteolytic activity.
Radioimmunoassay (RIA) of ECP, EPX and EPO (paper IV):The concentrations of ECP and EPX in the perfusion fluids were assayedby means of a specific RIA (Pharmacia & Upjohn Diagnostics AB,Uppsala, Sweden). EPO was measured with an EPO CAP FEIA prototype(Pharmacia & Upjohn). The inter- and intra-assay coefficients ofvariation were < 10 % for all tests.
28
Flow cytometric analysis (paper V)Flow cytometric analyses were performed on an EPICS-XL (CoulterCompany Inc., Miami, Fla., USA). The instrument was calibrated daily inaccordance with the recommendations of the manufacturer. A gate wasset on the eosinophil population in the forward scatter-side scatter (FS-SS) histogram and the mean FS and the staining of the cells by FITC-conjugated antibodies, measured as mean fluorescence intensity (MFI),was evaluated. The control antibody was used to set the backgroundfluorescence level. An average of 10, 000 eosinophils in each tube wereanalysed.
Statistical evaluationPaired non-parametric statistics were used to evaluate the statisticalsignificance of differences in chemotactic activity, forward scatter andreceptor expression. Correlation coefficients were obtained by Pearsonproduct-moment correlation (paper II) or the Spearman rank correlationtest (paper IV). The level of statistical significance adopted was p<0.05.All calculations were performed on a personal computer by means of thestatistical software Statistica (Statsoft, USA).
29
RESULTS
Eosinophil chemotactic activity in BAL fluid from pollen-allergicpatients during the birch-pollen season (papers I and II)BAL fluid from pollen-allergic patients was collected before and duringthe birch-pollen season, and its chemotactic activity was analysed usingpurified eosinophils from normal, healthy (paper I) or allergic (paper II)donors.Neutralising mouse anti-human monoclonal antibodies to IL-2, IL-5, IL-8and RANTES, and neutralising goat anti-human antibodies to leukaemiainhibitory factor (LIF) were added to the BAL fluid. In some cases,mouse and goat preimmune sera were used as a control antibody inaddition to anti-LIF.The eosinophil chemotactic activity of the individual BAL fluids wassignificantly higher during the birch-pollen season than before the season(p<0.002). Anti-IL-5 and anti-RANTES inhibited the chemotactic activityof the BAL fluid both when eosinophils from normal donors and thosefrom allergic donors were used (anti-IL-5: p<0.0001 and p<0.05,respectively; RANTES: p<0.0001 and p<0.05). In addition, anti-IL-8inhibited the migration of eosinophils from allergic donors towards theBAL fluid (p<0.05) (Fig.1).
0
20
40
60
80
100
p<0.05
p<0.05
p<0.05p<0.05
p<0.05
Fig 1. Eosinophils from allergic donors: migration towardsbroncho-alveolar lavage (BAL) fluid with and without antibody
p<0.05
NSNS
Controlantibody
Anti-IL-2Anti-RANTESAnti-IL-8Anti-IL-5BAL1/1001/1001/1001/1001/100 1/10001/10001/1000
Mig
ratio
n (%
of a
ctiv
ity in
BAL
)
30
Eosinophil chemotactic activity of recombinant IL-5, IL-8 andRANTES (paper II)RANTES was chemotactic for eosinophils both from normal and allergicdonors in a dose-dependent manner (p<0.01). Preincubation of the cellswith IL-5 (10-10-10-8 M) induced chemokinetic migration and also had anadditive effect on that of RANTES (p<0.05). Only eosinophils fromsymptomatic allergic donors responded to IL-8, and IL-5 was notsufficient to prime eosinophils from normal donors to an IL-8 response.We also tested the effect of eotaxin, and found that it was a potentchemoattractant, but only at concentrations of at least 10-7 M.In an attempt to reconstruct the activity of the BAL fluid, we mixed IL-5,IL-8 and RANTES, all in sub-optimal concentrations, together in Gey�sbuffer. The eosinophil chemotactic activity was comparable to that in theBAL fluid (on average 160% of buffer migration). Antibodies were addedto the mixture, and as in the BAL fluid, anti-IL-5 and anti-RANTESinhibited the migration of both types of eosinophils, while anti-IL-8 onlyinhibited that of eosinophils from allergic donors.Eosinophils from different individuals show varying levels of basalmigration towards buffer, probably depending on what activating factorsthey have been exposed to in the circulation. We found a negativecorrelation between the level of basal migration and the response to IL-5,indicating that in vivo-activated eosinophils had already been exposed toIL-5 in the blood. On the other hand, there was a positive correlationbetween the level of basal migration and the response to RANTES. Thisis in accordance with our finding that IL-5 and RANTES actedsynergistically. It also shows that the eosinophils had not been exposed toRANTES in the circulation, and that the production of RANTES is likelyto occur locally at the inflammatory site.
Conclusions from studies I and II:Our results suggest that although other factors may influence eosinophilmigration, IL-8 and RANTES are important for the chemotactic attraction
31
of these cells to the allergic lung. IL-5 seems to be an essential factor inthe cytokine network involved in eosinophil accumulation.
The effect of anti-IL-2 on eosinophil migration towards a mixture ofcytokines (paper III)Anti-IL-2 was one of the antibodies tested in the experiments with BALfluid (papers I and II), where it did not have any significant effects on theeosinophil chemotactic activity. In contrast, when added to the mixture ofcytokines described above (paper II), anti-IL-2 enhanced the migration ofeosinophils towards the mixture. This effect was significant foreosinophils from normal donors (p=0.04). A similar, though notsignificant, relationship was found in the allergic group.
Effects of recombinant IL-2 on eosinophil migration (paper III)The migration of eosinophils towards IL-2 at different concentrations wasslower than that towards buffer. The inhibition was significant at IL-2concentrations of 10-11 to10-8 M for eosinophils from normal donors, andat concentrations of 10-10 to10-9 M for eosinophils from allergic donors.
Preincubation of eosinophils with IL-2 (10-11 mol/L) significantly
0
10
20
30
40
50
60
70
80
90
p=0.030p=0.028
p=0.027p=0.028
EOS +IL-2
Migration towards Gey's bufferMigration towards RANTES 10-7M
EOS +IL-2 +control ab
EOS +IL-2 +anti-IL-2
EOS +IL-2 +control ab
EOS +IL-2 +anti-IL-2
EOS +IL-2
EOS EOS
Fig 2. The effect of IL-2 on eosinophil (EOS) migration towardsRANTES or buffer. Neutralisation of the effect with anti-IL-2.
Mig
ratio
n (µ
m/h
)
32
inhibited their migration towards buffer or RANTES. Preincubation ofIL-2 with anti-IL-2 completely abolished this inhibition, while the mouseIgG1 isotype control had no effect (Fig. 2).The effect of IL-2 on eosinophil migration was concentration-dependent,and the response to RANTES was significantly inhibited at IL-2concentrations of 10-16 mol/L and higher. The inhibitory effect of IL-2 onmigration towards buffer was more pronounced for eosinophils from thenormal than from the allergic donors (p=0.028). A negative correlation was found between the chemotactic response ofnormal eosinophils to RANTES (10-7M) and the inhibitory effect of IL-2on migration towards buffer (r= -0.82, p=0.021, n=7 ). Thus, cells with ahigh response to RANTES were less inhibited by IL-2.Preincubation with IL-2 also inhibited the migration towards PAF(p=0.007), IL-8 (p=0.010) and eotaxin (p=0.011). The chemotacticresponse to the optimal concentrations of PAF, IL-8 and eotaxin was not,however, completely inhibited by IL-2. Human serum albumin augmentsthe basal level of eosinophil migration, and the response varies dependingon what the eosinophils have been exposed to in vivo [101].Preincubation of eosinophils from normal individuals with IL-2 inhibitedtheir migration down to the level of random migration in the absence ofHSA. Furthermore, albumin-stimulated migration was inhibited almost tothe same degree when the cells migrated towards IL-2. Thus, IL-2inhibits the albumin-stimulated migration of the eosinophils and therebytheir response to chemoattractants.Priming, in vitro, with IL-5 prevented the inhibitory effect of IL-2 oneosinophil migration.
Effect of IL-2 on eosinophil adhesion (paper III)
The presence of IL-2, at molar concentrations of between 10-12 and 10-8,during eosinophil adhesion to E-selectin and ICAM-1 did not induce anyalterations in eosinophil adhesion. Furthermore, IL-2 did not influencethe basal adhesion of eosinophils.
33
Conclusion:The results indicate that IL-2 acts as an autocrine regulator of eosinophilmigration, and that this inhibitory effect may be down-regulated inallergy, leading to increased migration of eosinophils towardschemotactic factors.
Eosinophil chemotactic activity in intestinal perfusion fluid frompatients with ulcerative colitis (paper IV)Perfusion fluid was collected from patients with ulcerative colitis beforecommencement of treatment with a steroid enema (Pred-Clysma ) (day0), and after 7 and 28 days of treatment. Neutralising mouse anti-humanmonoclonal antibodies to IL-5, IL-8, RANTES, eotaxin, GM-CSF, MCP-3, TNF-α, and a mouse IgG1 isotype control (all from R&D systems,Europé, Ltd.) were added to the perfusion fluids before transferring themto the microchemotaxis chamber.The eosinophil chemotactic activity was higher in perfusion fluids frompatients with ulcerative colitis than in those from control patients(p=0.004) (Fig. 3).
80
100
120
140
160
180
200
220
240p<0,005
UC patients
Fig 3. Eosinophil chemotactic activity in perfusion fluid frompatients with ulcerative colitis (UC) before and after treatment withsteroid enema and in perfusion fluid from control patients.
controlpatients
Day 28Day 7Day 0
Mig
ratio
n (%
of b
uffe
r mig
ratio
n)
34
There was no significant difference in chemotactic activity betweenperfusion fluids obtained before treatment (day 0) and those obtainedafter 7 or 28 days of treatment with a steroid enema, although the activityin some of the individual perfusion fluids diminished during treatment.The chemotactic activity of the perfusion fluids obtained before treatmentwas inhibited by anti-IL-5 (p=0.005) and anti-TNF-α (p=0.017)compared to that of the corresponding IgG1-isotype control. Neutralisingantibodies were also added to pools of perfusion fluids collected duringtreatment (day 7 and day 28, respectively). None of the antibodies hadany significant effect on the activity of the perfusion fluids obtainedduring treatment.
Correlation between eosinophil markers and chemotactic activity inthe perfusion fluids (paper IV)There was a positive correlation between the eosinophil chemotacticactivity and the levels of ECP (r = 0.65, p = 0.03) and EPO (r = 0.65, p =0.03) in perfusion fluids on day 28. A similar relationship, thoughweaker, was found between the chemotactic activity and both ECP (r =0.34, p= 0.049) and EPO (r= 0.34, p = 0.05) when results were includedfrom all periods (day 0, day 7 and day 28).
Conclusion:In ulcerative colitis, eosinophils are attracted to the intestinal mucosa bychemotactic factors, of which IL-5 and TNF-α may be the mostprominent steroid-sensitive ones. The steroid-insensitive chemotacticactivities remain unidentified.
Effect of inhibitors of signal transduction on albumin-stimulatedeosinophil migration (paper V)The finding that IL-2 inhibited the HSA-stimulated eosinophil migration(paper III) raised questions about the mechanisms involved in thestimulation of eosinophils by albumin, and in the counteraction of thiseffect by IL-2.
35
Purified eosinophils diluted in Gey's solution were incubated withinhibitors of different molecules involved in signal transduction beforeaddition of HSA (1g/L) or Gey's solution. Migration assay was thenperformed. To investigate the effect of IL-2 after blocking of differentsignal transduction pathways, eosinophils were incubated with IL-2 (10-11
mol/L) after the previously described incubation with inhibitors in thepresence of albumin. The albumin-stimulated but not the basal migrationwas inhibited by Wortmannin, an inhibitor of PI3-kinase, atconcentrations of 10-8 mol/L (p=0.049) and 10-7 mol/L (p=0.012).Another, more specific inhibitor of PI3-kinase, LY-29 40 02, had a similareffect. The albumin-stimulated, but not the basal, migration was inhibitedby LY-29 40 02 at concentrations of 10-7 mol/L (p=0.046), 10-6 mol/L(p=0.022) and 10-5 mol/L (p=0.005). The PKC inhibitor RO-31-8220 hadno effect on either the albumin-stimulated or the basal migration. IL-2inhibited albumin-stimulated migration (38 ± 1.1 µm/h versus 20 ± 2.0µm/h) (p=0.005). IL-2 had no significant effect when added to the cellsafter incubation with Wortmannin or LY- 29 40 02. The inhibitory effectof IL-2 persisted after incubation with RO-31-8220.
The effects of albumin and IL-2 on forward scatter (paper V)Eosinophils were incubated with or without albumin (1g/L) or albumintogether with IL-2 (10-11 mol/L) prior to flow-cytometric analysis.Forward scatter, which reflects cell size and activation, was increased byalbumin (p=0.02). Incubation with IL-2 and albumin induced a furtherincrease in forward scatter (p=0.02). The eosinophils from two blooddonors divided into two populations differing in forward scatter afterstimulation with albumin: one population (A) with increased forwardscatter, and one population (B) with unchanged forward scatter (Fig. 4).
36
Fig 4. Distribution of eosinophils into two populationsafter stimulation with albumin
Buffer Albumin
FS=forward scatter. SS= side scatter
Dose-response relationships between albumin concentration, forwardscatter and chemokinetic response (paper V)The concentration of albumin used throughout this study (1g/L) is the onenormally used in our migration assay. However, two dose-responseexperiments with albumin were performed regarding forward scatter andchemokinesis. The chemokinetic response to albumin diminished inrelation to decreasing concentrations of albumin, and the chemokineticeffect vanished at an albumin concentration of 0.01 g/L. The forwardscatter, on the other hand, reached a peak at 0.1 g/L albumin, and thendecreased with higher and lower concentrations (Table I).
A
B
37
Table 1. Effect of different concentrations of human serum albumin(HSA) on migration and forward scatter. Two experiments withdifferent blood donors.
Experiment No1
Experiment No2
HSA conc.(g/L)
Migration(µm/h)
Forwardscatter
Migration(µm/h)
Forwardscatter
1 48 510.7 63 599.40.5 45 529.8 42 613.30.1 28 530.2 31 625.70.01 27 526.4 25 605.40.001 24 508.0 24 595.30 23 491.3 23 573.5
Effect of albumin and IL-2 on the expression of adhesion moleculeson the eosinophil cell surface (paper V)Purified eosinophils were incubated with or without albumin (1g/L) andIL-2 (10-11 mol/L) and then fixed with paraformaldehyde. Cell surfacereceptors were stained by FITC-conjugated monoclonal antibody (mAb)against CD11b, CD18, CD49d, CD49f and CD66b, and the cells wereanalysed by flow cytometry. Isotype-matched mouse mAb served asbackground control. The expression of CD49d and CD49f diminishedafter incubation with albumin (p=0.02 and p= 0.02 respectively).Albumin did not significantly affect the expression of CD11b, CD18 orCD66b. IL-2 did not significantly change the expression of any of theadhesion molecules tested.According to the distribution of eosinophils into the populationsmentioned above in two of the blood donors, the eosinophils from all thedonors were divided into two equal populations, A and B, as describedabove, and the expression of adhesion molecules was analysed separatelyin the two populations.
38
The eosinophils in population A exhibited higher expression of CD11b,CD18, CD49d and CD66b than those in population B. In contrast, theeosinophils in population A showed a lower expression of CD49f thanthose in population B.
Conclusion:The stimulation of eosinophil migration by HSA was mediated by PI3
kinase, whereas PKC did not seem to be involved. The increase inforward scatter caused both by albumin and by albumin plus IL-2indicates activation of the cells, though not directly associated withincreased migration. Decreased expression of CD49d and CD49f byalbumin may diminish the adhesiveness of the cells, which in turn mayfacilitate migration.
39
GENERAL DISCUSSION
The eosinophil granulocyte is a pro-inflammatory, tissue-damaging cell,and one of the features of allergic inflammation is extensive infiltration ofthe inflammatory site with eosinophils. The eosinophils are accompaniedby T cells, but the moderate numbers of neutrophils indicate that selectiverecruitment mechanisms are operative [25]. The presence of enhancedlevels of IL-5 in sera of allergic-asthmatic patients gives rise to selectiveeosinophil growth and differentiation, resulting in eosinophilia[102;103;103]. Selective priming of eosinophils in the circulation andspecific adhesion molecules, such as VLA-4 [104], may also result ineosinophil but not neutrophil recruitment, even if some of thechemotactic activities produced during inflammation are able to attractboth types of granulocytes. In addition, a number of chemotactic factorshave been defined as eosinophil specific, e.g. RANTES, eotaxin andMCP-3.
Recent studies on eosinophil recruitment have mainly been focused oneotaxin, but our results indicate that the recruitment of eosinophils iscarried out not by one single factor but by many different factors acting inconcert. The fact that anti-IL-5, anti-RANTES and anti-IL-8 were all ableto inhibit the chemotactic activity in BAL fluid from pollen-allergicsubjects (papers I and II) suggests that these cytokines depend on eachother for their function. Our experiments with recombinant cytokines inthe same study showed interaction between IL-5 and RANTES, and to acertain extent, IL-8. The presence of IL-5 was essential, especially forunprimed eosinophils. In vivo-primed eosinophils were not as dependenton IL-5 as the normal, unprimed eosinophils, and a combination ofRANTES and IL-8 was potently chemoattractant even in the absence ofIL-5. In study II we also found that the basal, unstimulated migration ofeosinophils, which reflects their level of in vivo activation, determinedtheir response to different cytokines. The more the eosinophils wereactivated in vivo, the less they responded to IL-5, indicating that the cells
40
activated in vivo had been exposed to IL-5 in the circulation. Thecorrelation between in vivo activation and the response to RANTES wasthe reverse, in accordance with our finding that IL-5 enhanced themigration towards RANTES. It also shows that the in vivo-activatedeosinophils had not been exposed to RANTES in the circulation, and thatRANTES probably is produced locally, at the inflammatory site. IL-8 is aCXC chemokine, which is chemotactic to neutrophils as well as to invivo-primed eosinophils [43].In our experiments, only eosinophils from symptomatic allergic subjectsresponded to IL-8, and we were unable to prime the eosinophils for an IL-8 response with IL-5 in vitro. A longer incubation time with IL-5 may benecessary to achieve priming for IL-8 [105]. Another possibility is thatother factors in addition to IL-5 are necessary for eosinophil priming.This is supported by the observation in a previous study that IL-5 alonewas not able to prime eosinophils from normal donors for a response toPAF to as great an extent as was seen in eosinophils from eosinophilicsubjects [106]. In a study on oxidative metabolism of in vivo-primedeosinophils from allergic subjects, further priming with IL-5 was found topossible. It was concluded that IL-5 alone is not responsible for the invivo priming of eosinophils [107]. IL-3 and GM-CSF may have additiveeffects to that of IL-5 in the priming, and TNF-α has been shown to primeeosinophils for PAF [108].
However, apart from the priming abilities of IL-5, the presence of thiscytokine appears essential for the response of eosinophils to lowconcentrations of chemoattractants. We found that RANTES and eotaxinwere both potent eosinophil chemoattractants at concentrations of 10-7
mol/L. However, concentrations of that magnitude are unlikely to occurin vivo [79], and the eosinophils may require IL-5 for a response to thephysiological concentrations of chemoattractants, even if these areincreased after allergen challenge.
41
Interleukin-2 and eosinophil migrationEosinophils are known to produce a variety of pro-inflammatorycytokines that contribute to the activation of T cells and epithelial cells,and that also have autocrine effects [3]. IL-2 is one of the cytokines thathave been shown to be synthesised and released by eosinophils[109;110]. We found that anti-IL-2 enhanced the migration of eosinophilstowards chemoattractants in a system where no IL-2 was added (paperIII), indicating that IL-2 was present in the cell suspension, exerting aninhibitory effect on the migration. This notion was strengthened by oursubsequent experiments with recombinant IL-2, where this cytokineinhibited both the basal migration stimulated by albumin, and thechemotaxis towards RANTES, PAF, eotaxin and IL-8. The albumin-stimulated eosinophil migration was examined more closely in study V,and will be discussed below. In the majority of our experiments weobserved that eosinophils from allergic donors were less sensitive to IL-2than normal eosinophils. We also showed that priming of eosinophilswith IL-5 in vitro made the cells completely insensitive to IL-2, whichmay explain the lower sensitivity of eosinophils from allergic donors toIL-2. Thus, IL-5 may contribute to eosinophil accumulation not only byacting as a chemokinetic and chemotactic factor, but also by eliminatingthe inhibitory effect of IL-2 on eosinophil migration. Furthermore, wefound a relationship between the chemotactic response and the sensitivityto inhibition by IL-2, indicating that the eosinophils with a highchemotactic response to RANTES were less sensitive to IL-2. Asmentioned above, exposure of eosinophils to IL-5 in the circulation leadsto an increased ability to migrate towards RANTES. In addition to thelower sensitivity to IL-2 of eosinophils from allergic patients, a lowerendogenous release of IL-2 might partly explain the less pronouncedeffect of anti-IL-2 on the migration of eosinophils from allergicindividuals. We hypothesise that IL-2 normally acts as a regulatory factoron eosinophil migration, whereas in allergy this inhibitory effect is
42
abolished by IL-5 priming, allowing increased migration of eosinophilstowards chemotactic factors.
Albumin-stimulated eosinophil migrationAlbumin is necessary for eosinophil migration towards low-molecularweight chemotactic factors, such as cytokines, in the Boyden-chamberassay [111;112]. The inhibitory effect of IL-2 on the migration ofeosinophils towards chemotactic factors may therefore depend entirely oninhibition of albumin stimulation of the cell. The mechanisms underlyingthe dependence on albumin in the Boyden chamber system are obscure. Ithas been speculated that albumin does nothing but provide theeosinophils with a favourable physico-chemical environment [113], orthat albumin diminishes the adhesiveness of the cells and hence preventsthem from sticking to the filter. In study V we showed that albumin isactively involved in eosinophil function by inducing intracellularsignalling. Albumin stimulates eosinophil chemokinesis through a PI3
kinase-dependent pathway. This stimulation may also trigger the responseof eosinophils to chemotactic factors. In our experiments with inhibitorsof PI3 kinase, the effect of IL-2 in this system could not be evaluatedbecause of the inhibitory action of both the PI3 kinase inhibitors and IL-2on eosinophil migration. A possible way to study the impact of IL-2 onthe albumin-induced signalling would be to do experiments with proteintyrosine phosphorylations. In such an assay, blocking of albumin-inducedprotein tyrosine phosphorylation by IL-2 would confirm our speculationthat IL-2 interferes with albumin signalling through PI3 kinase-dependentpathways. The only indication of this so far is the fact that IL-5 inducesPI3 kinase activity [114], and that IL-5 counteracts the inhibitory effect ofIL-2 (paper III). One of the mechanisms by which albumin stimulateseosinophil migration seems to be by down-regulation of adhesionmolecules, such as CD49d and CD49f (paper V). These molecules areable to bind to extracellular matrix proteins, and lower adhesiveness mayfacilitate eosinophil migration through the tissue. This is in agreementwith the observation that antibody blocking of CD49d increased the
43
chemokinetic migration of eosinophils stimulated by IL-5 [115].However, this mechanism of albumin stimulation is probably not thetarget for IL-2, since no changes in the expression of adhesion moleculeswere detected after incubation of the cells with IL-2. Nor did IL-2 affectthe adhesiveness of eosinophils to E-selectin or ICAM-1 (paper III).Albumin also caused an increase in forward scatter, which may beinterpreted as a sign of activation of the cells. Thus, albumin seems tohave active influence on eosinophil functions in several ways. This isimportant to keep in mind, since albumin is commonly used in eosinophilmigration assays, and has so far been regarded as an inert protein with nocapacity to activate cells. Our finding that albumin generates intracellularsignalling may give reason to question earlier results on signaltransduction and migration of eosinophils.
Eosinophil accumulation in ulcerative colitisThe mucosal inflammation in ulcerative colitis shares many features withallergic inflammation. One major difference, however, is the mucosalinfiltration by neutrophils in addition to eosinophils. We also founddifferences in the chemotactic activities between BAL fluid from allergicasthmatic subjects and perfusion fluid from patients with ulcerative colitis(paper IV). While the activity in the BAL fluid could be completelyblocked by antibodies to cytokines, an important part of the activity in theperfusion fluid remained even after addition of combinations of differentanti-cytokines. This persistent, and so far unidentified, activity alsoproved to be insensitive to in vivo steroid treatment. In contrast, thecytokine-mediated activity was inhibited by treatment of the patients withsteroid enema. As in the asthmatic lung, IL-5 was found to be animportant cytokine involved in the attraction of eosinophils to theintestinal mucosa in ulcerative colitis. RANTES, eotaxin, MCP-3 andGM-CSF seem to be of minor importance in this system, though they arenot entirely without influence, since antibodies to these cytokinesinhibited the activity of 18-23 % of the individual perfusion fluids. Inaddition, in vivo-primed eosinophils respond more readily to RANTES,
44
IL-8 [69] and eotaxin [116], and the effect of these chemokines may notbe evident until primed cells are tested. Hence, it would be of interest tostudy eosinophils from patients with active ulcerative colitis in thissystem.The expression of TNF-α is strongly enhanced in ulcerative colitis [117],and biological therapies based on antibodies against TNF-α have beenfound to improve the course of the disease [118]. In this study, anti-TNF-α inhibited the chemotactic activity in the perfusion fluid obtained beforesteroid treatment, and we conclude that IL-5 and TNF-α are togetherresponsible for the major part of the cytokine-mediated chemotacticactivity in ulcerative colitis. The importance of the eosinophilchemotactic activity in the perfusion fluid is suggested by the findings ofpositive correlations between this activity and the levels of ECP and EPOin the perfusion fluids, indicating that the chemotactic activity wasaccompanied by activated eosinophils.We therefore conclude that the eosinophil chemotactic activity in theperfusion fluids has biological relevance and plays an important role inthe recruitment of eosinophils to the intestinal mucosa. IL-5 and TNF-αwere identified as two putative components of this chemotactic activity,even though the molecular nature of other important chemoattractantsremains to be identified.
45
SUMMARY
• IL-5, RANTES and IL-8 were found to be the main chemotacticfactors in BAL fluid from pollen-allergic patients during the birch-pollen season.
• Only eosinophils from symptomatic allergic subjects responded to IL-8, and in vitro incubation with IL-5 did not prime eosinophils fromnormal subjects for an IL-8 response. IL-5 enhanced eosinophilmigration towards RANTES, and may be necessary for the responseof eosinophils to physiological concentrations of chemoattractants.Our results also indicate that different chemotactic cytokines act inconcert and may be dependent on one another for their actions.
• IL-2 inhibited albumin-stimulated eosinophil migration, and probablyacts as an autocrine regulator of eosinophil migration under normalconditions. Eosinophils from allergic subjects were less sensitive toIL-2 than those from normal subjects, and we found that in vitropriming with IL-5 prevented the inhibitory effect of IL-2 oneosinophil migration.
• In ulcerative colitis, eosinophils are attracted to the intestinal mucosaby chemotactic factors that are not present in the intestine of normalsubjects. IL-5 and TNF-α were shown to be two of the most importantsteroid-sensitive chemotactic factors in perfusion fluid from patientswith ulcerative colitis. Agents with chemotactic activity that cannot beeliminated by steroid treatment remain to be identified.
• Human serum albumin stimulates eosinophil migration through a PI3
kinase-dependent pathway. One mechanism by which albuminfacilitates eosinophil migration may be down-regulation of adhesionmolecules, such as CD49d and CD49f, on the eosinophil surface.
46
ACKNOWLEDGEMENTS
This work was carried out at the Department of Clinical Chemistry,University Hospital, Uppsala, Sweden. I wish to express my sinceregratitude to:
Per Venge, my supervisor, for being so enthusiastic and encouragingduring these years, always finding the grains of gold in my confusingresults.
Lena Douhan-Håkansson, my co-supervisor, for teaching me laboratorytechniques, for always taking the time to answer my questions, and forstimulating discussions.
Marie Carlson, for introducing me to the field of inflammatory boweldisease, for guiding me through our study on that subject and for manynice chats.
Per Sangfelt, Yesuf Taha, Magnus Thörn, Lars Lööf and YngveRaab, my co-authors, for kindness and support and for providing theclinical point of view and many good advice. Ingrid Stolt, for diggingthrough the freezers in search of my samples.
Lena Moberg, Ingunn Torsteinsdottir (before going back to Iceland),Kerstin Lindblad, Ing-Britt Persson, Agneta Trulsson, KristinaSeton, and the rest of the members of the inflammation research groupfor creating a nice atmosphere at work. Special thanks to Lena for lots ofpractical help (including calculating simple molar equations, shame onme!)
Charlotte Woshnagg, Malgorzata Karawajczyk, Jonas Byström andShengyuan Xu, my dear room-mates, for being there, chit-chatting andmaking work more fun.
47
All colleagues at the Department of Clinical Chemistry, for sharingrelaxing coffee breaks, the passion for delicious cakes, and solving ofcrosswords.
All blood-donors, for courageously holding out their arms for the needle,letting me have their eosinophils.
My parents Barbro and Lars, my sister Helena and her family, forconstant support and care.
All my friends in Uppsala, Luleå, Stockholm and other places, for manygood times.
Pablo, for leaving Argentina just to live with me in spite of being home-sick, and for always being kind and loving.Clara, for making me forgetting all about eosinophils the minute I gethome.
48
REFERENCES
1. Hirsch JG, Hirsch BI. Paul Erlich, and the discovery of theeosinophil. In: Mahmoud AAF, Austen KF, Simon AS, eds. Theeosinophil in health and disease. New York: Grune & Stratton,1980: 3-23.
2. Frigas E, Gleich GJ. Postgraduate courses: The eosinophil and thepathophysiology of asthma. J Allergy Clin Immunol 1986; 77No.4:527-37.
3. Venge P, Bergstrand H, Håkansson L. Neutrophils and eosinophils.In: Immune and Inflammatory Responses. 1994: 269-83.
4. Venge P. Eosinophil activity in bronchial asthma. Allergy Proc1994; 15:139-41.
5. Kay AB. Asthma and inflammation. J Allergy Clin Immunol 1991;87:893-910.
6. Makiyama K, Zea-Iriarte W, Tsuji X. Activation of eosinopils inthe pathophysiology of ulcerative colitis. J Gastroenterol 1995;30:64-9.
7. Raab Y, Fredens K, Gerdin B, Hällgren R. Eosinophil activation inulcerative colitis. Studies on mucosal release and localization ofeosinophil granule constituents. Digestive Diseases and Sciences1998; 43:1061-70.
8. Capron M, Leprevost C, Prin L, Tomassini M, Torpier G,MacDonald S, Capron A. Immunoglobulin-mediated activation ofeosinophils. In: Eosinophils in asthma. Academic press, 1989: 49-59.
9. Makino S, Fukuda T. Eosinophils and allergy in asthma. AllergyProc 1995; 16:13-21.
10. Gleich GJ. The eosinophil and bronchial asthma: Currentunderstanding. J Allergy Clin Immunol 1990; 85:422-36.
11. O'Byrne P. Asthma pathogenesis and allergen-induced lateresponses. J Allergy Clin Immunol 1998; 102:S85-S89.
12. Gaga M, Frew AJ, Varney VA, Kay AB. Eosinophil activation andT-lymphocyte infiltration in allergen induced late phase skin
49
reactions and classical delayed-type hypersensitivity. J Immunol1991; 147:816-22.
13. Lemanske RF. Inflammatory events in asthma: an expandingequation. J Allergy Clin Immunol 2000; 105:633-6.
14. Tomioka M, Ida S, Shindoh Y, Ishihara T, Takishima T. Mast cellsin bronchoalveolar lumen of patients with bronchial asthma. AmRev Respir Dis 1984; 129:1000-5.
15. Djukanovic R, Roche WR, Wilson JW, Beasley CRW, TwentymanOP, Howarth PH, Holgate ST. Mucosal inflammation in asthma.Am Rev Respir Dis 1990; 142:434-57.
16. Howarth PH, Bradding P, Quint D, Redington AE, Holgate ST.Cytokines and Airway Inflammation. Ann N Y Acad Sci 1994;725:69-82.
17. Broide DH, Lotz M, Cuomo AJ, Coburn DA, Federman EC,Wasserman SI. Cytokines in symptomatic asthma airways. JAllergy Clin Immunol 1992; 89:958-67.
18. Raeburn D, Webber SE. Proinflammatory potential of the airwayepithelium in bronchial asthma. Eur Respir J 1994; 7:2226-33.
19. Vignola AM, Chanez P, Campbell AM, Bousquet J, Michel FB,Godard P. Functional and phenotypic characteristics of bronchialepithelial cells obtained by brushing from asthmatic and normalsubjects. Allergy 1993; 48:32-8.
20. Pober JS, Cotran RS. The role of endothelial cells in inflammation.Transplantation 1990; 50 No.4:537-44.
21. Howard MC, Miyajima A, Coffman R. T cell-derived cytokinesand their receptors. In: Paul WE, ed. Fundamental immunology.New York: Raven Press Ltd., 1993: 763-800.
22. Del Prete G. Human Th1 and Th2 lymphocytes: their role in thepathophysiology of atopy. Allergy 1992; 47:450-5.
23. Robinson DS, Hamid Q, Ying S, Tsicopoulos A, Barkans J,Bentley AM, Corrigan C, Durham SR, Kay AB. Predominant Th2-like bronchoalveolar T-lymphocyte population in atopic asthma. NEngl J Med 1992; 326:298-304.
50
24. Kay AB. "Helper" (CD4+) T cells and eosinophils in allergy andasthma. Am Rev Respir Dis 1992; 145:22-6.
25. Venge P, Håkansson L. Current understanding of the role of theeosinophil granulocyte in asthma. Clin Exp Allergy 1991; 21:31-7.
26. De Monchy JGR, Kauffman HF, Venge P, Koeter GH, Jansen HM,Sluiter HJ, De Vries K. Bronchoalveolar eosinophilia duringallergen-induced late asthmatic reactions. Am Rev Respir Dis1985; 131:373-6.
27. MacDermott RP, Sanderson IR, Reinecker H-C. The central role ofchemokines (chemotactic cytokines) in the immunopathogenesis ofulcerative colitis and Crohn's disease. Inflammatory BowelDiseases 1998; 4:54-67.
28. D'Arienzo A, Manguso F, Astarita C, D'Armiento FP, Scarpa R,Gargano D, Scaglione G, Vicinanza G, Bennato R, Mazzacca G.Allergy and mucosal eosinophil infiltrate in ulcerative colitis.Scand J Gastroenterol 2000; 6:624-31.
29. Elliott SN, Wallace JL. Neutrophil-mediated gastrointestinalinjury. Can J Gastroenterol 1998; 12:559-68.
30. Ina K, Kusugami K, Yamaguchi T, Imada A, Hosokawa T, OhsugaM, Shinoda M, Ando T, Ito K, Yokohama Y. Mucosal interleukin-8 is involved in neutrophil migration and binding to extracellularmatrix in inflammatory bowel disease. Am J Gastroenterol 1997;92:1342-6.
31. Madara JL. Review article: Pathobiology of neutrophil interactionswith intestinal epithelia. Aliment Pharmacol Ther 1997; 11:57-63.
32. Raab Y, Gerdin B, Ahlstedt S, Hällgren R. Neutrophil mucosalinvolvement is accompanied by enhanced local production ofinterleukin-8 in ulcerative colitis. Gut 1993; 34:1203-6.
33. Bischoff SC, Meier PN, Trautwein C, Cetin Y, Maschek H, StolteM, Gebel M, Manns MP. Quantitative assessment of intestinaleosinophils and mast cells in inflammatory bowel disease.Histopathol 1996; 28:1-13.
34. Dubucquoi S, Janin A, Klein O, Desreumaux P, Quandalle P,Cortot A, Capron M, Colombel J-F. Activated eosinophils and
51
interleukin-5 expression in early recurrence of Crohn's disease. Gut1995; 67:242-6.
35. Krishnaswamy G, Liu MC, Su S-N, Kumai M, Xiao H-Q, MarshDG, Huang S-K. Analysis of Cytokine Transcripts in theBronchoalveolar Lavage Cells of Patients with Asthma. Am JRespir Cell Mol Biol 1993; 9:279-86.
36. Capron M. Dual function of eosinophils in pathogenesis andprotective immunity against parasites. Mem Inst Oswaldo Cruz1992; 87:83-9.
37. Dahl R, Venge P, Fredens K. Eosinophils. In: Barnes, ed. Asthma:Basic mechanisms and clinical management. London: AcademicPress, 1992: 111-24.
38. Silberstein DS. Eosinophil function in health and disease. Crit RevOncol Hematol 1995; 19:47-77.
39. Spry CJF. In: Eosinophils. A comprehensive review and guide tothe scientific and medical literature. Oxford, New York, Tokyo:Oxford University Press, 1988.
40. Kroegel C, Virchow J-C, Luttman W, Walker C, Warner JA.Pulmonary immune cells in health and disease: the eosinophilleukocyte (part 1). Eur Respir J 1994; 7:519-43.
41. Weller PF. Cytokine regulation of eosinophil function. ClinImmunol Immunopathol 1992; 62:55-9.
42. Collins PD, Weg VB, Faccioli LH, Watson ML, Moqbel R,Williams TJ. Eosinophil accumulation induced by humaninterleukin-8 in the guinea pig in vivo. Immunology 1993; 79:312-8.
43. Erger R, Casale TB. Interleukin-8 is a potent mediator ofeosinophil chemotaxis through endothelium and epithelium. Am JPhysiol 1995; 268:117-22.
44. Ying S, Kay AB. Chemokines in allergic asthma. In: RothenbergME, ed. Chemokines in allergic diseases. New York: MarcelDekker Inc., 2000: 383-401.
45. Curfs JHAJ, Meis JFGM, Hoogkamp-Korstanje JAA. A primer oncytokines: sources, receptors, effects and inducers. Clin MicrobiolRev 1997; 10:742-80.
52
46. Silberstein DS, David JR. The regulation of human eosinophilfunction by cytokines. Immunol Today 1987; 8:380-5.
47. Nagata M, Fukuda T, Numao T, Makino S, Dohi Y. Tumornecrosis alpha induces migration of human eosinophils. Arerugi1993; 42:549-55.
48. Tachimoto H, Bochner BS. The surface phenotype of humaneosinophils. Chem Immunol 2000; 76:45-62.
49. Hogg N, Berlin C. Structure and function of adhesion receptors inleukocyte trafficking. Immunol Today 1995; 16:327-30.
50. Garratt AN, Humphries MJ. Recent insights into ligand binding,activation and signalling by integrin adhesion receptors. Acta Anat1995; 154:34-45.
51. Goldring K, Warner JA. Cell matrix interactions in asthma. ClinExp All 1997; 27:22-7.
52. Hansel TT, De Vries IJ, Carballido JM, Braun RK, Carballido-Perrig N, Rihs S, Blaser K, Walker C. Induction and function ofeosinophil intercellular adhesion molecule-1 and HLA-DR. JImmunol 1992; 149:2130-6.
53. Koenderman L, van der Bruggen T, Schweizer RC, Warringa RAJ,Coffer P, Caldenhoven E, Lammers J-WJ, Raaijmakers JAM.Eosinophil priming by cytokines: from cellular signal to in vivomodulation. Eur Respir J 1998; 6:119-25.
54. Miyajima A, Mui AL-F, Ogorochi T, Sakamaki K. Receptors forgranulocyte-macrophage colony-stimulating factor, interleukin-3,and interleukin-5. Blood 1993; 82:1960-74.
55. Rand TH, Silberstein DS, Kornfeld H, Weller PF. Humaneosinophils express functional interleukin 2 receptors. J Clin Invest1991; 88:825-32.
56. Plumas J, Gruart V, Aldebert D, Truong MJ, Capron M, Capron A,Prin L. Human eosinophils from hypereosinophilic patientsspontaneously express the p55 but not the p75 interleukin 2receptor subunit. Eur J Immunol 1991; 21:1265-70.
57. Ishii N, Takeshita T, Kimura Y, Tada K, Kondo M, Nakamura M,Sugamura K. Expression of the IL-2 receptor gamma chain on
53
various populations in human peripheral blood. Int Immunol 1994;6:1273-7.
58. Murphy PM, Rothenberg ME. Chemokine receptors. In:Rothenberg ME, ed. Chemokines in allergic disease. New York,Basel: Marcel Dekker Inc., 2000: 53-66.
59. Petering H, Götze O, Kimmig D, Smolarski R, Kapp A, Elsner J.The biologic role of interleukin-8: functional analysis andexpression of CXCR1 and CXCR2 on human eosinophils. Blood1999; 93:694-702.
60. Elsner J, Kapp A. Actvator of human eosinophil effector functionsby CC chemokines. Allergy and Asthma Proc 1999; 20:365-9.
61. Wardlaw AJ, Moqbel R, Kay AB. Eosinophils: biology and role indisease. Adv Immunol 1995; 60:151-266.
62. Håkansson L, Carlson M, Stålenheim G, Venge P. Migratoryresponses of eosinophil and neutrophil granulocytes from patientswith asthma. J Allergy Clin Immunol 1990; 85:743-50.
63. Griffin E, Håkansson L, Formgren H, Jörgensen K, Venge P.Increased chemokinetic and chemotactic responses of eosinophilsin asthmatic patients. Allergy 1991; 46:255-65.
64. Håkansson L, Heinrich C, Rak S, Venge P. Priming of eosinophiladhesion in patients with birch pollen allergy during pollen season:effect of immunotherapy. J Allergy Clin Immunol 1997; 99:551-62.
65. Carlson M, Håkansson L, Kämpe M, Stålenheim G, Peterson C.Degranulation of eosinophils from pollen-atopic patients withasthma is increased during pollen-season. J Allergy Clin Immunol1992; 89:131-9.
66. Walker C, Virchow J-C, Bruijnzeel PLB, Blaser K. T cell subsetsand their soluble products regulate eosinophilia in allergic andnonallergic asthma. J Immunol 1991; 146:1829-35.
67. Till S, Li B, Durham S, Humbert M, Assoufi B, Huston D,Dickason R, Jeannin P, Kay AB, Corrigan C. Secretion of theeosinophil-active cytokines interleukin-5, granulocyte/macrophagecolony-stimulating factor and interleukin-3 by bronchoalveolar
54
lavage CD4+ and CD8+ T cell lines in atopics asthmatics, andatopic and non atopic controls. Eur J Immunol 1995; 25:2727-31.
68. Håkansson L, Venge P. Priming of eosinophil and neutrophilmigratory responses by interleukin-3 and interleukin-5. APMIS1994; 102:1-7.
69. Schweizer RC, Welmers BAC, Raaijmakers JAM, Zanen P,Lammers J-WJ, Koenderman L. RANTES-and Interleukin-8-Induced Responses in Normal Human Eosinophils: Effects ofPriming With Interleukin-5. Blood 1994; 83:3697-704.
70. Carlson M, Peterson C, Venge P. The influence of IL-3, IL-5 andGM-CSF on normal human eosinophil and neutrophil C3b-induceddegranulation. Allergy 1993; 48:437-42.
71. Her E, Frazer J, Austen KF, Owen WF. Eosinophil hematopoietinsantagonize the programmed cell death of eosinophils. J Clin Invest1991; 88:1982-7.
72. Carlos TM, Harlan JM. Leukocyte-endothelial adhesion molecules.Blood 1994; 84:2068-101.
73. Ebisawa M, Bochner BS, Georas SN, Schleimer RP. Eosinophiltransendothelial migration induced by cytokines. I. Role ofendothelial and eosinophil adhesion molecules in IL-1b- inducedtransendothelial migration. J Immunol 1992; 149:4021-8.
74. Bloemen PGM, Henricks PAJ, Nijkamp FP. Review: Cell adhesionmolecules and asthma. Clin Exp Allergy 1997; 27:128-41.
75. Koizumi M, King N, Lobb R, Benjamin C, Podolsky DK.Expession of vascular adhesion molecules in inflammatory boweldisease. Gastroenterology 1992; 103:840-7.
76. Bruijnzeel PLB. Eosinophil Tissue Mobilization in AllergicDisorders. Ann N Y Acad Sci 1994; 725:259-67.
77. Hoshi H, Ohno I, Honma M, Tanno Y, Yamaguchi K, Tamura G,Shirato K. IL-5, IL-8 and GM-CSF immunostaining of sputumcells in bronchial asthma and chronic bronchitis. Clin Exp Allergy1995; 25:720-8.
78. Bradding P, Roberts JA, Britten KM, Montefort S, Djukanovic R,Mueller R, Heusser CH, Howarth PH, Holgate ST. Interleukin-4, -5, and -6 and Tumor Necrosis Factor-α in normal and asthmatic
55
airways: evidence for the human mastcell as a source of thesecytokines. Am J Respir Cell Mol Biol 1994; 10:471-80.
79. Teran L, Noso N, Carroll M, Davies DE, Holgate S, Schröder J-M.Eosinophil recruitment following allergen challenge is associatedwith the release of the chemokine RANTES into asthmaticairways. J Immunol 1996; 157:1806-12.
80. Gutierrez-Ramos JC, Lloyd C, Gonzalo J-A. Eotaxin: from aneosinophilic chemokine to a major regulator of allergic reactions.Immunol Today 1999; 20:500-4.
81. Kaplan AP, Reddigan SR, Kuna P. Chemokines and the allergicresponse. Exp Dermatol 1995; 4:260-5.
82. Bischoff SC, Lorentz A, Schwengberg S, Weier G, Raab Y, MannsMP. Mast cells are an important cellular source of tumour necrosisfactor alfa in human intestinal tissue. Gut 1999; 44:643-52.
83. Stossel TP. The machinery of blood cell movements. Blood 1994;84:367-79.
84. Fay FS, Gilbert SH, Brundage RA. Calcium signalling duringchemotaxis. In: Wiley, Chichester, eds. Calcium waves, gradientsand oscillations. Ciba foundation symposium 188, 1996: 121-40.
85. Palecek SP, Horwitz AF, Lauffenburger DA. Kinetic model forintegrin-mediated adhesion release during cell migration. AnnBiomed Eng 1999; 27:219-35.
86. Southwick FS, Stossel TP. Contractile proteins in leukocytefunction. Seminars in Hematology 1983; 20:305-21.
87. Keller HU, Wilkinson PC, Abercrombie M, Beckers EL, HirschJG, Miller ME, Ramsey WS, Zigmond SH. A proposal for thedefinition of terms related to locomotion of leukocutes and othercells. Clin Exp Immunol 1977; 27:377-80.
88. Wilkinson PC. Principles of the measurement of leukocytechemotaxis using Boyden's method. In: Chemotaxis andinflammation. London: Churchill Livingston, 1974: 33-53.
89. Resnick MB, Weller PF. Mechanisms of eosinophil recruitment.Am J Respir Cell Mol Biol 1993; 8:349-55.
56
90. Liu L, Zuurbier AEM, Mul FPJ, Verhoeven AJ, Lutter R, Knol EF,Roos D. Triple role of Platelet-Activating Factor in eosinophilmigration across monolayers of lung epithelial cells: eosinophilchemoattractant and priming agent and epithelial cell activator. JImmunol 1998; 161:3064-70.
91. Hom JT, Estridge T. Antigen-Induced Recruitment of Eosinophils:Importance of CD4+ T-cells ,IL-5, and Mast Cells. ClinicalImmunology and Immunopathology 1994; 73:305-11.
92. Kameyoshi Y, Dörschner A, Mallet AI, Christophers E, SchröderJ-M. Cytokine RANTES released by thrombin-stimulated plateletsis a potent attractant for human eosinophils. J Exp Med 1992;176:587-92.
93. Sehmi R, Cromwell O, Wardlaw AJ, Moqbel R, Kay AB.Interleukin-8 is a chemo-attractant for eosinophils purified fromsubjects with a blood eosinophilia but not from normal healthysubjects. Clin Exp Allergy 1993; 23:1027-36.
94. Frieri M. Corticosteroid effects on cytokines and chemokines.Allergy and Asthma Proc 1999; 20:147-59.
95. Håkansson L, Venge P. The influence of serum on randommigration and chemotaxis of polymorphonuclear leucocytes:Methodological evaluation using sera from infection-prone patientsand normals. Scand J Immunol 1980; 11:271-82.
96. Hansel TT, De Vries IJM, Iff T, Rihs S, Wandzilak M, Betz S,Blaser K, Walker C. An improved immunomagnetic procedure forthe isolation of highly purified human blood eosinophils. JImmunol Methods 1991; 145:105-10.
97. Falk W, Goodwin RHJr, Leonard EJ. A 48-well micro chemotaxisassembly for rapid and accurate measurement of leukocytemigration. J Immunol Methods 1980; 33:239-47.
98. Håkansson L, Westerlund D, Venge P. New method for themeasurement of eosinophil migration. Journa of LeukocyteBiology 1987; 42:689-96.
99. Håkansson L, Söegaard Nielsen L, Teder P. Measurement ofneutrophil and eosinophil adhesion to E-selectin, VCAM-1, andICAM-1 by the use of transfected fibroblast cell lines. J ImmunolMethods 1994; 176:53-66.
57
100. Raab Y, Hällgren R, Knutson L, Krog M, Gerdin B. A techniquefor segmental rectal and colonic perfusion in humans. Am JGastroenterol 1992; 87:1453-9.
101. Lampinen M, Rak S, Venge P. The role of interleukin-5,interleukin-8 and RANTES in the chemotactic attraction ofeosinophils to the allergic lung. Clin Exp All 1999; 29: 314-22.
102. Corrigan CJ, Kay AB. T cells, cytokines and asthma. 2-18. 1994.Ref Type: Conference Proceeding
103. Walker C, Virchow J-C, Bruijnzeel PLB, Blaser K. T cell subsetsand their soluble products regulate eosinophilia in allergic andnonallergic asthma. J Immunol 1991; 146:1829-35.
104. Seminario MC, Bochner BS. Expression and function of beta 1integrins on human eosinophils. Mem Inst Oswaldo Cruz 1997; 92Suppl 2:157-64.
105. Heath H, Qin S, Rao P, Wu L, LaRosa G, Kassam N, Ponath PD,Mackay CR. Chemokine receptor usage by human eosinophils. JClin Invest 1997; 99:178-84.
106. Sehmi R, Wardlaw AJ, Cromwell O, Kurihara K, Waltmann P,Kay AB. Interleukin-5 selectively enhances the chemotacticresponse of eosinophils obtained from normal but not eosinophilicsubjects. Blood 1992; 79 No.11:2952-9.
107. Woschnagg C, Garcia R, Rak S, Venge P. IL-5-priming of thePMA-induced oxidative metabolism of human eosinophils fromallergic and normal subjects during a pollen season. Clin Exp All2000; In press.
108. Macari DMT, Teixeira MM, Ansari T, Jeffery PK, Hellewell PG.Priming and induction of eosinophil trafficking in guinea-pigcutaneous inflammation by tumour necrosis factor α. Br JPharmacol 1998; 125:1228-35.
109. Bossé M, Audette M, Ferland C, Pelletier G, Chu HW, DakhamaA, Lavigne S, Boulet L-P, Laviolette M. Gene expression ofinterleukin-2 in purified human peripheral blood eosinophils.Immunol 1996; 87:149-54.
110. Levi-Schaffer F, Barkans J, Newman TM, Ying S, Wakelin M,Hohenstein R, Barak V, Lacy P, Kay AB, Moqbel R. Identification
58
of interleukin-2 in human peripheral blood eosinophils. Immunol1996; 87:155-61.
111. Håkansson L, Venge P. The chemotactic response of granulocytesto the low molecular weight chemoattractants f-MLP, c5f, andLTB4 is dependent on chemokinetic factors. J Leukoc Biol 1985;38:521-30.
112. Wilkinson PC. A requirement for albumin as a carrier for lowmolecular weight leukocyte chemotactic factors. Exp Cell Res1976; 103:415-20.
113. Valerius NH. Chemotaxis, spreading and oxidative metabolism ofneutrophils: influence of albumin in vitro. Acta Pathol MicrobiolImmunol Scand 1983; 91:43-9.
114. Coffer P, Schweizer RC, Dubois GR, Maikoe T, Lammers J-WJ,Koenderman L. Analysis of signal transduction pathways in humaneosinophils activated by chemoattractants and the T-helper 2-derived cytokines interleukin-4 and interleukin-5. Blood 1998;91:2547-57.
115. Poluektova LY, Huggler GK, Patterson EB, Khan MM.Involvement of protein kinase A in histamin-mediated inhibition ofIL-2 mRNA expression in mouse splenocytes.Immunopharmacology 1990; 41:77-8.
116. Mould AW, Matthaei KI, Young IG, Foster PS. Relationshipbetween interleukin-5 and eotaxin in regulating blood and tissueeosinophilia in mice. J Clin Invest 1997; 99:1064-71.
117. Reinecker H-C, Steffen M, Witthoeft T, Pflueger F, Schreiber S,Mac Dermott RP, Raedler A. Enhanced secretion of tumournecrosis factor-alpha, IL-6 and IL-1β by isolated lamina propriamononuclear cells from patients with ulcerative colitis and Crohn'sdisease. Clin Exp Immunol 1993; 94:174-81.
118. Evans RC, Clarke L, Heath P, Stephens S, Morris AI, Rhodes JM.Treatment of ulcerative colitis with an engineered human anti-TNFalpha antibody CDP571. Aliment Pharmacol Ther 1999; 11:1031-5.
![Research Paper Sex specific inflammatory profiles of ......cells, activated by microglia and astrocytes [27, 28]. Pro-inflammatory cytokine contributions to the inflammasome associated](https://img.pdfslide.us/doc/110x75/60f96bbd0ba77f202104e786/research-paper-sex-specific-inflammatory-profiles-of-cells-activated-by.jpg)
