ASSOCIATE EDITORS K. Frank Austen Harvard Medical School, Boston, Massachusetts, USA Tasuku Honjo KyotoUniversity, Kyoto,Japan Fritz Melchers University ofBasel, Basel, Switzerland Hidde Ploegh Massachusetts Institute of Technology, Massachusetts, USA Kenneth M. Murphy Washington University, St. Louis, Missouri, USA AcademicPressisanimprintofElsevier 525BStreet,Suite1800,SanDiego,CA92101-4495,USA 225WymanStreet,Waltham,MA02451,USA TheBoulevard,LangfordLane,Kidlington,Oxford,OX51GB,UK 32JamestownRoad,London,NW17BY,UK Radarweg29,POBox211,1000AEAmsterdam,TheNetherlands Firstedition2014 Copyright©2014ElsevierInc.Allrightsreserved Nopartofthispublicationmaybereproduced,storedinaretrievalsystemortransmittedin anyformorbyanymeanselectronic,mechanical,photocopying,recordingorotherwise withoutthepriorwrittenpermissionofthepublisher PermissionsmaybesoughtdirectlyfromElsevier’sScience&TechnologyRights DepartmentinOxford,UK:phone(+44)(0)1865843830;fax(+44)(0)1865 853333; email:permissions@elsevier.com.Alternativelyyoucansubmityourrequestonlineby visitingtheElsevierwebsiteathttp://elsevier.com/locate/permissions,andselecting ObtainingpermissiontouseElseviermaterial Notice Noresponsibilityisassumedbythepublisherforanyinjuryand/ordamagetopersonsor propertyasamatterofproductsliability,negligenceorotherwise,orfromanyuseor operationofanymethods,products,instructionsorideascontainedinthematerialherein. Becauseofrapidadvancesinthemedicalsciences,inparticular,independentverificationof diagnosesanddrugdosagesshouldbemade ISBN:978-0-12-800100-4 ISSN:0065-2776 ForinformationonallAcademicPresspublications visitourwebsiteatstore.elsevier.com PrintedandboundinUSA 14 15 16 17 11 10 9 8 7 6 5 4 3 2 1 CONTRIBUTORS BenjamineArellano DepartmentofNeurology&Neurotherapeutics,UniversityofTexasSouthwesternMedical CenteratDallas,Dallas,Texas,USA K.ChristopherGarcia HowardHughesMedicalInstitute;DepartmentofMolecularandCellularPhysiology; DepartmentofStructuralBiology,andPrograminImmunology,StanfordUniversitySchool ofMedicine,Stanford,California,USA SvenJarius DivisionofMolecularNeuroimmunology,DepartmentofNeurology,Universityof Heidelberg,Heidelberg,Germany KusumamJoseph DepartmentofBiochemistryandMolecularBiology,MedicalUniversityofSouthCarolina, Charleston,SouthCarolina,USA AllenP.Kaplan DepartmentofMedicine,MedicalUniversityofSouthCarolina,Charleston,South Carolina,USA MichaelLevy DepartmentofNeurology,JohnsHopkinsUniversity,Baltimore,Maryland,USA LaurenceMacia DepartmentofImmunology,MonashUniversity,Clayton,Victoria,Australia CharlesR.Mackay DepartmentofImmunology,MonashUniversity,Clayton,Victoria,Australia CraigMcKenzie DepartmentofImmunology,MonashUniversity,Clayton,Victoria,Australia JuanL.Mendoza HowardHughesMedicalInstitute;DepartmentofMolecularandCellularPhysiology; DepartmentofStructuralBiology,andPrograminImmunology,StanfordUniversitySchool ofMedicine,Stanford,California,USA IgnacioMoraga HowardHughesMedicalInstitute;DepartmentofMolecularandCellularPhysiology; DepartmentofStructuralBiology,andPrograminImmunology,StanfordUniversitySchool ofMedicine,Stanford,California,USA LuigiD.Notarangelo DivisionofImmunologyandTheMantonCenterforOrphanDiseaseResearch,Boston Children’sHospital,HarvardStemCellInstitute,HarvardMedicalSchool,Boston, Massachusetts,USA MariaPotamitis DepartmentofImmunology,MonashUniversity,Clayton,Victoria,Australia AdvancesinImmunology,Volume121 #2014ElsevierInc. ix ISSN0065-2776 Allrightsreserved. http://dx.doi.org/10.1016/B978-0-12-800100-4.09988-8 x Contributors SaranyaSasidharan DepartmentofNeurology,JohnsHopkinsUniversity,Baltimore,Maryland,USA RoyJ.Soberman DepartmentofMedicine,DivisionofNephrology,MassachusettsGeneralHospital,Boston, Massachusetts,USA JamieSpangler HowardHughesMedicalInstitute;DepartmentofMolecularandCellularPhysiology; DepartmentofStructuralBiology,andPrograminImmunology,StanfordUniversitySchool ofMedicine,Stanford,California,USA OlafStuve DepartmentofNeurology&Neurotherapeutics,UniversityofTexasSouthwesternMedical CenteratDallas;NeurologySection,VANorthTexasHealthCareSystem,MedicalService, Dallas,Texas,USA;DepartmentofNeurology,KlinikumrechtsderIsar,Technische Universita¨tMu¨nchen,Mu¨nchen,andDepartmentofNeurology,HeinrichHeineUniversity Du¨sseldorf,Du¨sseldorf,Germany JianTan DepartmentofImmunology,MonashUniversity,Clayton,Victoria,Australia AlisonN.Thorburn DepartmentofImmunology,MonashUniversity,Clayton,Victoria,Australia ChristineA.Vaine DepartmentofMedicine,DivisionofNephrology,MassachusettsGeneralHospital,Boston, Massachusetts,USA MartinWeber DepartmentofNeuropathology,UniversityMedicalCenter,GeorgAugustUniversity, Go¨ttingen,Germany,andDepartmentofNeurology,UniversityMedicalCenter,Georg AugustUniversity,Go¨ttingen,Germany BrigitteWildemann DivisionofMolecularNeuroimmunology,DepartmentofNeurology,Universityof Heidelberg,Heidelberg,Germany CHAPTER ONE Multifarious Determinants of Cytokine Receptor Signaling Specificity Ignacio Moraga*,†,{,}, Jamie Spangler*,†,{,}, Juan L. Mendoza*,†,{,}, K. Christopher Garcia*,†,{,},1 * HowardHughesMedicalInstitute,StanfordUniversitySchoolofMedicine,Stanford,California,USA †DepartmentofMolecularandCellularPhysiology,StanfordUniversitySchoolofMedicine,Stanford, California,USA {DepartmentofStructuralBiology,StanfordUniversitySchoolofMedicine,Stanford,California,USA }PrograminImmunology,StanfordUniversitySchoolofMedicine,Stanford,California,USA 1Correspondingauthor:e-mailaddress:[email protected] Contents 1. Introduction 2 2. Ligand–ReceptorComplexFormation:GeometryandAffinity 3 2.1 Surfacereceptorcomplexgeometryandsignalactivation 4 2.2 Ligand–receptoraffinity:Stabilityoftheternarycomplex 8 2.3 Ligand–receptoraffinity:Kineticsofcomplexformation 12 3. ReceptorDynamicsandEndosomalTrafficking 14 4. IntracellularProteinLevelsandSignalingActivation 19 4.1 Jaksandreceptorexpressionandaffinity 20 4.2 CrosstalkbetweenSTATs 20 5. TuningCytokineSignalingviaProteinEngineering 23 5.1 Modulationoftrafficking:IL-2andEGF 24 5.2 Modulationofligandaffinityandactivities:IFNandIL-3 25 5.3 Increasingligandspecificity:IL-2andIL-4 26 6. Perspectives 28 Acknowledgments 29 References 29 Abstract Cytokinesplaycrucialrolesinregulatingimmunehomeostasis.Twoimportantcharac- teristicsofmostcytokinesarepleiotropy,definedastheabilityofonecytokinetoexhibit diversefunctionalities,andredundancy,definedastheabilityofmultiplecytokinesto exertoverlappingactivities.Identifyingthedeterminantsforuniquecellularresponses tocytokinesinthefaceofsharedreceptorusage,pleiotropy,andredundancywillbe essentialinordertoharnessthepotentialofcytokinesastherapeutics.Here,wediscuss thebiophysical(ligand–receptorgeometryandaffinity)andcellular(receptortrafficking AdvancesinImmunology,Volume121 #2014ElsevierInc. 1 ISSN0065-2776 Allrightsreserved. http://dx.doi.org/10.1016/B978-0-12-800100-4.00001-5 2 IgnacioMoragaetal. andintracellularabundanceofsignalingmolecules)parametersthatcontributetothe specificity of cytokine bioactivities. Whereas the role of extracellular ternary complex geometry in cytokine-induced signaling is still not completely elucidated, cytokine- receptor affinity is known to impact signaling through modulation of the stability andkineticsofternarycomplexformation.Receptortraffickingalsoplaysanimportant andlikelyunderappreciatedroleinthediversificationofcytokinebioactivitiesbutithas been challenging to experimentally probe trafficking effects. We also review recent effortsto quantify levels ofintracellular signaling components, assecond messenger abundance can affect cytokine-induced bioactivities both quantitatively and qualita- tively. We conclude by discussing the application of protein engineering to develop therapeutically relevant cytokines with reduced pleiotropy and redirected biological functionalities. 1. INTRODUCTION Four-helicalcytokinesaresecretedproteinsthatregulatemostfacetsof immune function and numerous other aspects of mammalian physiology (Bazan,1989,1990).Cytokinesexerttheirbiologicalactivitiesbyinducing cell surface receptor dimerization in either homo- or hetero-oligomeric assemblies (Stroud & Wells, 2004; Wang, Lupardus, Laporte, & Garcia, 2009;Watowich,Hilton,&Lodish,1994).Inthecanonicalcytokinesignal- ing pathway, assembly of the cytokine-receptor complex activates tyrosine kinases of the Janus Kinase (Jak) and Tyk2 family, which are constitutively bound to receptors (Ihle, Witthuhn, Quelle, Yamamoto, & Silvennoinen, 1995).Jaks,inturn,phosphorylateandactivateSignalTransducerandActi- vatorofTranscription(STAT)transcriptionfactors(Levy&Darnell,2002; Schindler,Hoey,&McKnight,1996)tomodulategeneexpressionand,ulti- mately,determinecellfate(Murray,2007;O’Shea&Plenge,2012).Inaddi- tiontotheirsignalingthroughtheJak/STATpathway,somecytokinescan alsoactivatetheAktandErkpathways(Platanias,2005),aswellasothersig- nalingnetworks(Gough,Levy,Johnstone,&Clarke,2008;Heinrichetal., 2003; Malek, 2008; Schindler, Levy, & Decker, 2007; van Boxel-Dezaire, Rani, & Stark, 2006). Ithasbeenwellestablishedthatcytokinesexhibittwofeatures:(1)plei- otropy, the capacity of one cytokine to elicit a multitude of diverse func- tional responses; and (2) redundancy, the ability of multiple different cytokines to effect overlapping activities (Ozaki & Leonard, 2002). The properties of pleiotropy and redundancy emanate from the degenerate nature of cytokine complexes. A single cytokine may engage more than one receptor complex to activate distinct sets of Jaks and STATs, leading SpecificityofCytokineSignaling 3 todiversefunctionaleffects(Zurawski,Vega,Huyghe,&Zurawski,1993). Receptor subunits may also be shared between several cytokines, and the limitednumberofJak(four)andSTAT(seven)proteinsresultsinredundant activation of Jak/STAT combinations by distinct cytokine complexes (Pestka, Krause, Sarkar, et al., 2004; Pestka, Krause, & Walter, 2004; Vignali&Kuchroo,2012).Itis,however,remarkablethatdespiteusingsuch aseeminglyconstrainedsetofsignalingproteinsinafinitenumberofcom- binations,cytokinesarestillabletopromoteabroadrangeofactivitiesandto regulate a highly complex immune system (Delgoffe, Murray, & Vignali, 2011). Our understanding of the multifarious mechanisms through which cytokinesaffectsuchadiverserangeofbiologicalactivitiesremainsincom- plete,andwehaveyettoilluminatethedetailsofhowfunctionalspecificity is achieved given the rampant redundancy and pleiotropy of cytokines. Whatwedoknowisthatthereisnoclearcorrelationbetweentheparticular signaling molecule that is activated and the bioactivity that results. For instance,althoughthecytokinesInterleukin(IL)-6andIL-10bothactivate STAT3, their roles in immune regulation are diametrically opposed, with IL-10 eliciting an anti-inflammatory and IL-6 eliciting a proinflammatory response (Hunter & Kastelein, 2008; Mosser & Zhang, 2008; Murray, 2007).Anotherexampleofdivergentfunctionaloutcomesbeingtransmitted throughasharedsignalingmoleculeisfoundinthetypeIinterferon(IFN) system in which more than 16 subtypes engage an identical receptor com- plex yet elicit distinct biological activities (Borden et al., 2007; Piehler, Thomas, Christopher Garcia, & Schreiber, 2012). Gaining insight into themolecularmechanismsthatunderliethisobservedfunctionalspecificity will greatly advance our understanding of cytokine biology and immune regulation. In this review, we summarize findings that demonstrate the importance of biophysical (ligand–receptor binding kinetics and complex stability) and cellular (receptor trafficking and abundance or localization of signaling molecules) parameters in the determination and diversification ofcytokineactivities.Wealsodiscussrecentexamplesofhowtheseparam- eterscanbeusedtomodulatecytokineactivitiesthroughimplementationof biomolecular engineering techniques. 2. LIGAND–RECEPTOR COMPLEX FORMATION: GEOMETRY AND AFFINITY Signal activation is initiated when cytokines engage the extracellular domains of their cognate receptors. Many receptor subunits are shared by multiplecytokinesbutpairedwithdifferentreceptorchainstoformunique 4 IgnacioMoragaetal. signalingcomplexes(Liao,Lin,&Leonard,2011;Pestka,Krause,&Walter, 2004; Wang et al., 2009). For example, IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21sharethecommongchain(g )(Rochman,Spolski,&Leonard,2009); c IL-6, IL-11, leukemia-inducible factor, oncostatin M (OSM), and ciliary neurotrophic factor (CNTF) share the gp130 receptor (Boulanger, Bankovich, Kortemme, Baker, & Garcia, 2003; Silver & Hunter, 2010); and granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-5, and IL-3 share the common b chain (b ) (Bagley, Woodcock, Hercus, c Shannon,&Lopez,1995;Tavernieretal.,1991;Wangetal.,2009).Insome instances, identical receptor complexes are shared by multiple cytokines which induce distinct biological activities. This is the case for type I IFNs, as well as for the two cytokines IL-4 and IL-13, which signal through heterodimeric complexes composed of IL-4Ra, g , and IL-13Ra1 c (Chomarat & Banchereau, 1998; Junttila et al., 2008; Uze, Schreiber, Piehler,&Pellegrini, 2007).Giventheir sharedusageofreceptorsubunits, ithasbeenproposedthatparameterssuchasthegeometryandaffinityofthe ligand–receptorcomplexmayaccountforfunctionaldiversityofthesecyto- kines (LaPorte et al., 2008). The role of the inter-subunit geometry of the ligand–receptor complex has been extensively studied but there is still no consensus on whether or not modulation of receptor–ligand architectures can influence downstream signal activation (Ballinger & Wells, 1998). On the other hand, the importance of the ligand–receptor binding affinity in quantitativeandpossiblyqualitativeaspectsofdownstreamsignalingevents has been well established (Pang, Qin, & Zhou, 2011; Piehler et al., 2012). Affinityimpactsthekineticsofcomplexformationaswellasthestabilityof thecytokine-receptorcomplex,therebytuningboththestrengthanddura- tion of signal activation. 2.1. Surface receptor complex geometry and signal activation As more cytokine-receptor ECD complexes have been solved in recent years, it has become apparent that these complexes exhibit a diverse range of molecular architectures, yet all of these structural orientations result in activation of the Jak/STAT pathway (Boulanger, Chow, Brevnova, & Garcia, 2003; de Vos, Ultsch, & Kossiakoff, 1992; Hansen et al., 2008; Jones, Logsdon, & Walter, 2008; LaPorte et al., 2008; Livnah et al., 1996; Livnah et al., 1999; Logsdon, Deshpande, Harris, Rajashankar, & Walter, 2012; McElroy, Dohm, & Walsh, 2009; Rickert, Wang, Boulanger, Goriatcheva, & Garcia, 2005; Ring et al., 2012; Syed et al., 1998; Thomas SpecificityofCytokineSignaling 5 etal.,2011,2012;Walteretal.,1995;Wang,Rickert,&Garcia,2005).The toleranceofsignalactivationforvariablebindingtopologiesraisesthequestion of whether mere dimerization of receptors is sufficient for the activation of downstream signaling (i.e., dimerization serves as a binary “on/off” switch) orwhetherparticularreceptorgeometriesserveinstructiverolesindetermin- ingthedegreeandnatureofreceptoractivation. Several early studies suggested that the structural constraints on dimer- ization are quite relaxed, since grafting of the extracellular domain (ECDs)ofonereceptorontotheintracellulardomain(ICD)ofanunrelated receptorresultedinfunctionalsignalactivationofthechimericmolecule.As oneofmanysuchexamples,replacingtheECDoftheerythropoietin(Epo) receptor (EpoR) with those of epidermal growth factor receptor (EGFR), insulin-likegrowthfactorreceptor(IGFR),orc-Kitproducesanactivecon- struct that induces STAT5 phosphorylation and Ba/F3 proliferation in a ligand-dependent manner (Ohashi, Maruyama, Liu, & Yoshimura, 1994). Similarly, fusion of the gp130 ICD to EpoR ECD leads to Epo-mediated activation of STAT1 and STAT3, consistent with activation of the full- length gp130 by its native ligands (Kawahara et al., 2006; Schaeffer et al., 2001). Other examples of receptor ectodomain swapping have been reported for the type I IFN (Pattyn et al., 1999) and IL-4/IL-13 (Fujiwara, Hanissian, Tsytsykova, & Geha, 1997; Heller et al., 2012) sys- tems,andinallcases,thechimericreceptorselicitsomeextentofactivation of the expected Jak/STAT signaling molecules. These results suggest that cytokinesignalactivationisquitepermissivewithrespecttocomplexdimer geometry.However,theconfoundingfactorinthesesignalingstudiesisthat theywereperformedincelllinesexpressinghighlevelsofchimericreceptors and conducted with elevated doses of ligand, thereby limiting their physi- ologicalsignificance.Invivo,theconsequencesofcytokineactionaretightly regulatedbytherelativeabundanceofvariouscelltypeswhosesensitivities to stimulation are governed by receptor expression levels. Thus, while invitrosignalingofachimericreceptormayappeartobequalitativelysimilar to that of wild-type (wt) receptor in terms of STAT activation patterns (whichisnotsurprisinggiventhattheICDsusedinthechimericreceptors are identical to those of the wild-type receptors), these studies offer an incomplete perspective of cytokine behavior by failing to account for the finelytunedbalanceofreceptorexpressionlevelsmaintainedbythediverse celltypesthatexistsinvivo.Inordertopaintaclearerpictureoftheactivityof thesechimericreceptors,quantitativestudiesmustbeundertakeninwhich theEC andE valuesofinducedSTATactivationaredirectlycompared 50 max 6 IgnacioMoragaetal. to those of wt receptors for a range of cell types. In addition, sophisticated invivomodelsmustbedevelopedtoexaminethebroaderfunctionalconse- quences of chimeric receptor signaling. Inapparentcontrasttothearchitecturalpromiscuityobservedinchime- ric receptor studies, other experimental findings have presented evidence that geometry of the surface complex plays a crucial role in determining thenatureandextentofthetransmittedsignal.Itappearsthatcertainorien- tationsoftheEpoandthrombopoietin(Tpo)complexesaremorelikelyto activate signaling than others. For these studies, the authors substituted the EpoRECDwiththecoiled-coildimerizationdomainoftheyeasttranscrip- tion factor Put3 (Swaminathan, Flynn, Reece, & Marmorstein, 1997), inducing constitutive dimerization of the receptor and thus activating downstreamsignalinginaligand-independentmanner.Therotationalori- entation of the EpoR ICD was then systematically modulated by inserting alanineresiduesintothejuxtamembraneregion,whichwaspresumedtobe helicalsuchthateachresidueinsertionwouldbepredictedtorotatethehelix byapproximatelyone-thirdofaturn.Somealanineinsertionsarenotcom- patible with signal activation, other insertions lead to wt-like signal activa- tion,andyetanothergroupofinsertionspreferentiallyactivatedoneofthe two downstream signaling pathways (Constantinescu, Huang, Nam, & Lodish,2001;Staerketal.,2011).Thesefindingssuggestthattherearepre- ferredrotationalorientationsforsignaling,butintheabsenceofdirectstruc- tural information these conclusions remain speculative. Nevertheless, the results of these EpoR rotational studies resonate with those obtained from studies investigating the functionality of disulfide-mediated EpoR dimers. Cysteine pairs were introduced in the transmembrane (TM) region of the receptor,forcingconstitutivedimerizationintheabsenceofligand.Among the six disulfide bridges evaluated, only three cysteine–cysteine linkages were capable of promoting robust ligand-independent signal activation, suggesting that specific orientations of the TM domain are required to achieve a productively signaling Epo complex (Constantinescu, Keren, et al., 2001; Kubatzky et al., 2001; Lu, Gross, & Lodish, 2006; Watowich etal., 1992).Moreover,itwasreportedthattheintroductionof asparagine mutations to induce changes in the interhelical packing of the EpoR TM segments resulted in complete activation of STAT5, but did not induce the activation of Erk, suggesting that ligand-induced signal diversification couldalsoberegulatedbythegeometryandpackingoftheTMdimerinter- action (Becker et al., 2008). Here again, in the absence of direct structural information,theprecisemolecularconsequencesofthesemutationsremain