Seite - 25 - in Advances in Neuroimmunology

BrainSci. 2016,6, 15 4.Discussion IL-6 is implicated inthepathogenesisofautoimmunedisorders, suchasMSinhumans[20,21]. Acritical roleof IL-6 in theanimalmodelofMS,EAE,hasbeendemonstratedbyanumberof studies: systemic IL-6KOmiceareresistant toEAE[9–13], andneutralizationof IL-6withantibodies leads to areduceddisease [22],bynotwell-definedmechanisms.Moreover, studieshavedemonstratedthat the transgenicexpressionof IL-6 in theCNSbyviral systemsreducesEAE[23]andthat thesystemic administrationof IL-6also reduces the clinical symptoms inaviralmodelofEAE[24]. Thus, IL-6 canpotentiate, butalso inhibitEAE, reflecting thecomplexityof its actions. Keyquestions remain, however, includingthe identityof thekeycell types thatproduceandrespondtoIL-6andwhether the criticalactionsof IL-6areperipheralorcentral. Because theproductionof IL-6during thecourseofEAEarises fromdiversecellular sourcesboth in theperipheryandin theCNS, thespecificcontributionofeachsourceof IL-6 to thedevelopmentof thediseaseneeds tobeestablished.WehavepreviouslydemonstratedaroleofCNSIL-6 inregulating EAE, becausemice expressing IL-6 only by astrocytes (GFAP-IL6-IL-6KOmice)were capable of developingtheatypicalEAEknowntooccur inGFAP-IL6mice [13]. Furthermore, inadoptive transfer experiments,EAEis lesssevere inIL-6KOmicethaninwild-typemice,whichsuggests that IL-6 locally mediates thedisease intheCNS[10]. Thus,althoughEAEhasalwaysbeenconsideredadiseasemostly inducedperipherally, itseemsthatCNSIL-6mayalsoplayanimportantrole. Becauseastrocytic IL-6plays amajorrole inneuroinflammation[25,26]andbecauseastrocytesarethemostabundantcell intheCNS, astrocytic IL-6seemedtobeanexcellentcandidatetoexamineasthekeyregulatorofEAEintheCNS. Wehavepreviouslyshownthatcomparedtofloxedmice,Ast-IL-6KOmiceexhibitanumberof alteredbehaviorsundernormal (basal) conditions, includingchanges inactivity,anxietyandlearning; aprosurvival roleof astrocytic IL-6 is alsoapparent [17,18]. Here,wepresent results fromstudies involvinganeuropathologicalcondition,MOG35–55-inducedEAE. Incontrast toresults insystemic IL-6KOmice,astrocytic-specific IL-6deficiency isunable toprevent typical signsofEAEinduction andhasnoprominentneuropathologiceffects.However,astrocyte IL-6KOmicedidshowsignificant delaysof theonsetof theEAE,at least in femalemice,amelioratingtheclinical signs in theearlystages ofEAE. AutoreactiveT-cells can result in inflammatorydemyelinationof theCNS, andknowing that the frequencyof Tregs inMSpatients is unchanged fromcontrols [27] (although their function is impaired)couldbeaclue to thedecreaseddemyelinationseenonly inAst-IL-6KOfemales, theonly groupthatpresentedadecrease inTlymphocyte infiltration. Thus,a lowerEAEscoreatearlystages couldbeduetoan initial impairedTcell infiltrationof theCNS.Thispossibility is consistentwithour results showingthatAst-IL-6KOfemale (butnotmale)micehadadecreasednumberof infiltrates in thespinal cordand lowerscores for clinical signs. However, it is important tonote thatweonly carriedoutCresylvioletandCD3immunostainingforTcells, sowecannotruleout thatachange in Tcell subpopulations is responsible for thedifferent clinical signsobserved inAst-IL-6KOfemale mice. IL-6has amajor role inTh17 cell differentiation fromnaiveCD4+ T-cells (reviewed in [28]), particularly in theEAEmodel [14,29]. EAE-resistant IL-6KOmice showadeficiency inTh17 cell infiltrates in theCNS[30].Whenresponsiveness to IL-6 isexperimentallyeliminated inThelpercells, miceshowresistance toEAEinduction,as the IL-21pathwayis intact,butnotactive in theabsenceof IL-6signaling [31]. Th17cellsproduce IL-17 (amongothercytokines),whichenhances IL-6production byastrocytes,which, in turn, induces thedifferentiationofTh17 cells in apositive feed-back loop betweenIL-17andIL-6viaactivationofNF-κBandSTAT-3[32,33]. This loopmaybecompromised inourAst-IL-6KOanimals,whereastrocyteproductionof IL-6hasbeeneliminated.However, since Ast-IL-6KOmiceonlyshowadelay inEAE,butarenotresistant toEAE,wecanspeculate that this astrocytic loopisnotnecessary for thedevelopmentof thedisease;and/or thatneuronal, endothelial andmicroglial IL-6could insteadallowthispositive feedbackbetweenIL-17andIL-6. Inorder to test thesehypotheses, adetailedstudyof theexact lymphocyticpopulationpresent in the infiltrateswould be important,particularlywith furtherbackcrossingtoC57/Bl6mice. 25