Advertisement

A strategy for the selection of monovalent antibodies that span protein dimer interfaces

  • Jamie B. Spangler
    Correspondence
    Recipient of a Leukemia and Lymphoma Society Career Development Program fellowship. To whom correspondence may be addressed. E-mail: [email protected]
    Footnotes
    Affiliations
    Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, California 94305

    Department of Molecular and Cellular Physiology, Stanford University School of Medicine, Stanford, California 94305

    Department of Structural Biology, Stanford University School of Medicine, Stanford, California 94305

    Department of Biomedical Engineering, Johns Hopkins University, Baltimore, Maryland 21218

    Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, Maryland 21218
    Search for articles by this author
  • Ignacio Moraga
    Footnotes
    Affiliations
    Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, California 94305

    Department of Molecular and Cellular Physiology, Stanford University School of Medicine, Stanford, California 94305

    Department of Structural Biology, Stanford University School of Medicine, Stanford, California 94305
    Search for articles by this author
  • Kevin M. Jude
    Affiliations
    Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, California 94305

    Department of Molecular and Cellular Physiology, Stanford University School of Medicine, Stanford, California 94305

    Department of Structural Biology, Stanford University School of Medicine, Stanford, California 94305
    Search for articles by this author
  • Christina S. Savvides
    Affiliations
    Department of Biology, Stanford University School of Medicine, Stanford, California 94305
    Search for articles by this author
  • K. Christopher Garcia
    Correspondence
    An investigator of the Howard Hughes Medical Institute. To whom correspondence may be addressed. E-mail: [email protected]
    Affiliations
    Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, California 94305

    Department of Molecular and Cellular Physiology, Stanford University School of Medicine, Stanford, California 94305

    Department of Structural Biology, Stanford University School of Medicine, Stanford, California 94305
    Search for articles by this author
  • Author Footnotes
    1 Both authors contributed equally to this work.
Open AccessPublished:August 06, 2019DOI:https://doi.org/10.1074/jbc.RA119.009213
      Ligand-induced dimerization is the predominant mechanism through which secreted proteins activate cell surface receptors to transmit essential biological signals. Cytokines are a large class of soluble proteins that dimerize transmembrane receptors into precise signaling topologies, but there is a need for alternative, engineerable ligand scaffolds that specifically recognize and stabilize these protein interactions. Recombinant antibodies can potentially serve as robust and versatile platforms for cytokine complex stabilization, and their specificity allows for tunable modulation of dimerization equilibrium. Here, we devised an evolutionary strategy to isolate monovalent antibody fragments that bridge together two different receptor subunits in a cytokine–receptor complex, precisely as the receptors are disposed in their natural signaling orientations. To do this, we screened a naive antibody library against a stabilized ligand–receptor ternary complex that acted as a “molecular cast” of the natural receptor dimer conformation. Our selections elicited “stapler” single-chain variable fragments (scFvs) of antibodies that specifically engage the interleukin-4 receptor heterodimer. The 3.1 Å resolution crystal structure of one such stapler revealed that, as intended, this scFv recognizes a composite epitope between the two receptors as they are positioned in the complex. Extending our approach, we evolved a stapler scFv that specifically binds to and stabilizes the interface between the interleukin-2 cytokine and one of its receptor subunits, leading to a 15-fold enhancement in interaction affinity. This demonstration that scFvs can be selected to recognize epitopes that span protein interfaces presents new opportunities to engineer structurally defined antibodies for a broad range of research and therapeutic applications.

      Introduction

      Ligand-mediated receptor dimerization is the most prevalent signaling mechanism used by secreted proteins to activate their cognate cell surface receptors. In particular, transmembrane receptors of the cytokine and receptor tyrosine kinase families signal when oriented into specific receptor dimer geometries (
      • Ullrich A.
      • Schlessinger J.
      Signal transduction by receptors with tyrosine kinase activity.
      ,
      • Klemm J.D.
      • Schreiber S.L.
      • Crabtree G.R.
      Dimerization as a regulatory mechanism in signal transduction.
      • Wang X.
      • Lupardus P.
      • Laporte S.L.
      • Garcia K.C.
      Structural biology of shared cytokine receptors.
      ). Cytokines constitute a class of soluble ligands that act through dimeric membrane-embedded receptors to elicit a wide range of biological activities, particularly those pertinent to immune regulation (
      • Wang X.
      • Lupardus P.
      • Laporte S.L.
      • Garcia K.C.
      Structural biology of shared cytokine receptors.
      ,
      • Bazan J.F.
      A novel family of growth factor receptors: a common binding domain in the growth hormone, prolactin, erythropoietin and IL-6 receptors, and the p75 IL-2 receptor β-chain.
      • Rochman Y.
      • Spolski R.
      • Leonard W.J.
      New insights into the regulation of T cells by γ(c) family cytokines.
      ). Cytokines bind to receptor extracellular domains (ECDs)
      The abbreviations used are: ECD
      extracellular domain
      JAK
      Janus kinase
      STAT
      signal transducer and activator of transcription
      scFv
      single-chain variable fragment
      IL
      interleukin
      γc
      γ-chain
      MACS
      magnetic-activated cell sorting
      HC
      heavy-chain
      LC
      light-chain
      PDB
      Protein Data Bank
      RMSD
      root mean square deviation
      VH
      variable heavy
      VL
      variable light
      CDR
      complementarity-determining region
      TReg
      regulatory T
      NK
      natural killer
      SPR
      surface plasmon resonance
      SA
      streptavidin.
      and either reorient quiescent dimers or enforce dimerization of monomeric subunits (Fig. 1A, left) (
      • Cunningham B.C.
      • Ultsch M.
      • De Vos A.M.
      • Mulkerrin M.G.
      • Clauser K.R.
      • Wells J.A.
      Dimerization of the extracellular domain of the human growth hormone receptor by a single hormone molecule.
      • Watowich S.S.
      • Hilton D.J.
      • Lodish H.F.
      Activation and inhibition of erythropoietin receptor function: role of receptor dimerization.
      ,
      • Ihle J.N.
      • Witthuhn B.A.
      • Quelle F.W.
      • Yamamoto K.
      • Silvennoinen O.
      Signaling through the hematopoietic cytokine receptors.
      ,
      • Syed R.S.
      • Reid S.W.
      • Li C.
      • Cheetham J.C.
      • Aoki K.H.
      • Liu B.
      • Zhan H.
      • Osslund T.D.
      • Chirino A.J.
      • Zhang J.
      • Finer-Moore J.
      • Elliott S.
      • Sitney K.
      • Katz B.A.
      • Matthews D.J.
      • Wendoloski J.J.
      • Egrie J.
      • Stroud R.M.
      Efficiency of signalling through cytokine receptors depends critically on receptor orientation.
      ,
      • Livnah O.
      • Stura E.A.
      • Middleton S.A.
      • Johnson D.L.
      • Jolliffe L.K.
      • Wilson I.A.
      Crystallographic evidence for preformed dimers of erythropoietin receptor before ligand activation.
      • Constantinescu S.N.
      • Keren T.
      • Socolovsky M.
      • Nam H.
      • Henis Y.I.
      • Lodish H.F.
      Ligand-independent oligomerization of cell-surface erythropoietin receptor is mediated by the transmembrane domain.
      ). Extracellular dimerization results in activation of Janus kinases (JAKs) constitutively associated with the receptor intracellular domains, which recruit and activate signal transducer and activator of transcription (STAT) proteins to effect transcriptional programs that determine cell fate (
      • Delgoffe G.M.
      • Murray P.J.
      • Vignali D.A.A.
      Interpreting mixed signals: the cell's cytokine conundrum.
      • Levy D.E.
      • Darnell J.E.
      Stats: transcriptional control and biological impact.
      ,
      • Murray P.J.
      The JAK-STAT signaling pathway: input and output integration.
      • O'Shea J.J.
      • Plenge R.
      JAK and STAT signaling molecules in immunoregulation and immune-mediated disease.
      ).
      Figure thumbnail gr1
      Figure 1Novel evolutionary selection strategy isolates antibody-based “staplers” that dimerize cytokine receptor chains. A, schematic contrasting receptor heterodimerization and activation induced by cytokine, antibody, or “stapler” scFv binding, respectively. B, layout and selection scheme for yeast display-based evolution of cytokine–receptor complex staplers. A yeast-displayed scFv library is evolved through iterative rounds of selection, each of which comprises negative clearance steps against the individual complex components and positive selection for binders to the fully assembled ternary complex. C, flow cytometry histograms depicting the binding of yeast-displayed IL-4 stapler scFv to the indicated cytokine–receptor complexes and individual components thereof. TC, ternary complex. D, kinetic SPR binding profiles for the interactions between the immobilized IL-4 stapler scFv and soluble Super-4/IL-4Rα/γc TC components versus the fully assembled complex. All proteins were flowed at a concentration of 60 μm. E, surface plasmon resonance kinetic binding traces depicting the interaction between immobilized IL-4 stapler scFv and dilutions of the Super-4/Il-4Rα/γc ternary complex. The equilibrium dissociation constant (KD) value was calculated by fitting the equilibrium titration curve to a logistic model via nonlinear regression. The top curve represents a concentration of 20 μm, and subsequent curves represent 3-fold serial dilutions.
      There has been a great deal of interest in harnessing the agonistic potential of JAK/STAT cytokines as immunotherapeutics, but so far success has been very limited for several practical reasons (
      • Overwijk W.W.
      • Schluns K.S.
      Functions of γC cytokines in immune homeostasis: current and potential clinical applications.
      ,
      • Pellegrini M.
      • Mak T.W.
      • Ohashi P.S.
      Fighting cancers from within: augmenting tumor immunity with cytokine therapy.
      ). First, the short in vivo half-life (typically <5 min) of cytokines mandates frequent injection or continuous infusion. Second, cytokines are pleiotropic, often activating a wide range of cell types expressing shared receptors, which hinders efficacy and can lead to systemic toxicity. Finally, these ligands are difficult to re-engineer or modify without concerns about immunogenicity (
      • Donohue J.H.
      • Rosenberg S.A.
      The fate of interleukin-2 after in vivo administration.
      • Andrews A.-L.
      • Holloway J.W.
      • Holgate S.T.
      • Davies D.E.
      IL-4 receptor α is an important modulator of IL-4 and IL-13 receptor binding: implications for the development of therapeutic targets.
      ,
      • Borden E.C.
      • Sen G.C.
      • Uze G.
      • Silverman R.H.
      • Ransohoff R.M.
      • Foster G.R.
      • Stark G.R.
      Interferons at age 50: past, current and future impact on biomedicine.
      ,
      • Boyman O.
      • Sprent J.
      The role of interleukin-2 during homeostasis and activation of the immune system.
      • Spangler J.B.
      • Moraga I.
      • Mendoza J.L.
      • Garcia K.C.
      Insights into cytokine–receptor interactions from cytokine engineering.
      ). Thus, there exists a need for new modulators of protein dimerization that are based on protein scaffolds with both improved druglike properties and the capacity to serve as engineering substrates.
      Monoclonal antibodies present stable, engineerable scaffolds that benefit from extended in vivo half-life due to interactions with neonatal Fc receptors (
      • Finkelman F.D.
      • Madden K.B.
      • Morris S.C.
      • Holmes J.M.
      • Boiani N.
      • Katona I.M.
      • Maliszewski C.R.
      Anti-cytokine antibodies as carrier proteins: prolongation of in vivo effects of exogenous cytokines by injection of cytokine-anti-cytokine antibody complexes.
      ,
      • Roopenian D.C.
      • Christianson G.J.
      • Sproule T.J.
      • Brown A.C.
      • Akilesh S.
      • Jung N.
      • Petkova S.
      • Avanessian L.
      • Choi E.Y.
      • Shaffer D.J.
      • Eden P.A.
      • Anderson C.L.
      The MHC class I-like IgG receptor controls perinatal IgG transport, IgG homeostasis, and fate of IgG-Fc-coupled drugs.
      ), and they can act as bivalent dimerization modulators for cytokine receptors. Previous work has demonstrated that certain cytokine receptor-targeted bivalent antibodies can activate signaling in the absence of cytokine (Fig. 1A, center) (
      • Harwerth I.M.
      • Wels W.
      • Marte B.M.
      • Hynes N.E.
      Monoclonal antibodies against the extracellular domain of the erbB-2 receptor function as partial ligand agonists.
      • Müller-Newen G.
      • Küster A.
      • Wijdenes J.
      • Schaper F.
      • Heinrich P.C.
      Studies on the interleukin-6-type cytokine signal transducer gp130 reveal a novel mechanism of receptor activation by monoclonal antibodies.
      ,
      • Kai M.
      • Motoki K.
      • Yoshida H.
      • Emuta C.
      • Chisaka Y.
      • Tsuruhata K.
      • Endo C.
      • Muto M.
      • Shimabe M.
      • Nishiyama U.
      • Hagiwara T.
      • Matsumoto A.
      • Miyazaki H.
      • Kataoka S.
      Switching constant domains enhances agonist activities of antibodies to a thrombopoietin receptor.
      ,
      • Frederickson S.
      • Renshaw M.W.
      • Lin B.
      • Smith L.M.
      • Calveley P.
      • Springhorn J.P.
      • Johnson K.
      • Wang Y.
      • Su X.
      • Shen Y.
      • Bowdish K.S.
      A rationally designed agonist antibody fragment that functionally mimics thrombopoietin.
      ,
      • Nakano K.
      • Kojima T.
      • Kasutani K.
      • Senoh C.
      • Natori O.
      • Ishii S.
      • Tsunoda H.
      • Hattori K.
      Effective screening method of agonistic diabodies based on autocrine growth.
      ,
      • Zhang H.
      • Wilson I.A.
      • Lerner R.A.
      Selection of antibodies that regulate phenotype from intracellular combinatorial antibody libraries.
      • Moraga I.
      • Wernig G.
      • Wilmes S.
      • Gryshkova V.
      • Richter C.P.
      • Hong W.-J.
      • Sinha R.
      • Guo F.
      • Fabionar H.
      • Wehrman T.S.
      • Krutzik P.
      • Demharter S.
      • Plo I.
      • Weissman I.L.
      • Minary P.
      • et al.
      Tuning cytokine receptor signaling by re-orienting dimer geometry with surrogate ligands.
      ). However, these molecules enforce unnatural receptor dimer geometries, which can impact the nature of signal activation (Fig. 1A, center) (
      • Moraga I.
      • Wernig G.
      • Wilmes S.
      • Gryshkova V.
      • Richter C.P.
      • Hong W.-J.
      • Sinha R.
      • Guo F.
      • Fabionar H.
      • Wehrman T.S.
      • Krutzik P.
      • Demharter S.
      • Plo I.
      • Weissman I.L.
      • Minary P.
      • et al.
      Tuning cytokine receptor signaling by re-orienting dimer geometry with surrogate ligands.
      ). Antibodies that potentiate cytokine action by enhancing cytokine–receptor binding have also been reported (
      • Kuruganti S.
      • Miersch S.
      • Deshpande A.
      • Speir J.A.
      • Harris B.D.
      • Schriewer J.M.
      • Buller R.M.L.
      • Sidhu S.S.
      • Walter M.R.
      Cytokine activation by antibody fragments targeted to cytokine-receptor signaling complexes.
      ,
      • Adams R.
      • Burnley R.J.
      • Valenzano C.R.
      • Qureshi O.
      • Doyle C.
      • Lumb S.
      • Del Carmen Lopez M.
      • Griffin R.
      • McMillan D.
      • Taylor R.D.
      • Meier C.
      • Mori P.
      • Griffin L.M.
      • Wernery U.
      • Kinne J.
      • et al.
      Discovery of a junctional epitope antibody that stabilizes IL-6 and gp80 protein:protein interaction and modulates its downstream signaling.
      ). Here, we sought to develop an evolutionary strategy for the discovery of antibody-based constructs that simultaneously engage two interacting subunits within dimerized receptors, as they are disposed in their native signaling conformations induced by endogenous cytokines (Fig. 1A, right).
      Antibody Fab fragments or single-chain variable fragments (scFvs) are widely known to recognize “composite” ligand surfaces, such as Fabs that bind across the interfaces between different subunits of a multimeric antigen. Yet it is not clear whether such composite interface-recognizing Fabs can be evolved to specifically recognize and bridge protein interfaces. We wished to isolate monovalent antibodies (which we term “staplers”) that bind to the composite dimer surface formed by the “stem region” of the receptors when complexed with their natural cytokine ligands (i.e. the “product complex”) (Fig. 1B). We used a soluble cytokine–receptor ECD complex to act as a “molecular cast” of the receptor dimer signaling conformation to isolate monovalent antibody fragments specific for the productive signaling complex, but not for the individual complex components. We hypothesized that the evolved antibodies would be specific for the “locked” receptor dimer stem region, binding to the receptor subunits in the geometry induced by the natural cytokine (Fig. 1A).
      We focused our stapler engineering efforts on the interleukin-4 (IL-4) cytokine, a member of the family of common γ-chain (γc) cytokines, which are critical for lymphocyte homeostasis and have been used to treat diseases such as cancer, chronic infection, and autoimmune disorders (
      • Rochman Y.
      • Spolski R.
      • Leonard W.J.
      New insights into the regulation of T cells by γ(c) family cytokines.
      ,
      • Overwijk W.W.
      • Schluns K.S.
      Functions of γC cytokines in immune homeostasis: current and potential clinical applications.
      ). IL-4 signals through either a type I heterodimeric receptor comprised of IL-4 receptor-α (IL-4Rα) and the γc subunit or a type II heterodimeric receptor comprised of IL-4Rα and IL-13Rα1 (
      • Wang X.
      • Lupardus P.
      • Laporte S.L.
      • Garcia K.C.
      Structural biology of shared cytokine receptors.
      ,
      • LaPorte S.L.
      • Juo Z.S.
      • Vaclavikova J.
      • Colf L.A.
      • Qi X.
      • Heller N.M.
      • Keegan A.D.
      • Garcia K.C.
      Molecular and structural basis of cytokine receptor pleiotropy in the interleukin-4/13 system.
      ). The crystal structure of the type I IL-4 cytokine–receptor complex extracellular domains has been determined, revealing that the membrane-proximal domains of the receptor ECDs form a “stem region” of intimate receptor–receptor contact (Fig. S1) (
      • LaPorte S.L.
      • Juo Z.S.
      • Vaclavikova J.
      • Colf L.A.
      • Qi X.
      • Heller N.M.
      • Keegan A.D.
      • Garcia K.C.
      Molecular and structural basis of cytokine receptor pleiotropy in the interleukin-4/13 system.
      ). We targeted these stem regions to isolate monovalent antibody fragments that bridge the type I IL-4 receptor dimer by simultaneously binding to the membrane-proximal domains of IL-4Rα and γc. Crystallographic analysis showed that our evolved stapler indeed binds an interface spanning the two receptor subunits. We then extended our selection strategy to evolve scFvs that stabilize the interface between another γc cytokine, interleukin-2 (IL-2), and one of its receptor subunits. We showed that these scFvs enhance IL-2–receptor interaction and potentiate cytokine signaling in the context of an activity-impaired mutant. With this proof-of-concept demonstration for evolution of customized scFvs that recognize composite surfaces, our strategy could, in principle, enable discovery of stapler antibodies that modulate the activity of cytokines and other classes of natural ligands for a variety of therapeutic indications.

      Results

      Devising an evolutionary scheme for selection of dimerizing stapler scFvs

      To screen for IL-4 staplers, we selected a diverse (109 clones) yeast-displayed nonimmune human scFv library (
      • Feldhaus M.J.
      • Siegel R.W.
      • Opresko L.K.
      • Coleman J.R.
      • Feldhaus J.M.W.
      • Yeung Y.A.
      • Cochran J.R.
      • Heinzelman P.
      • Colby D.
      • Swers J.
      • Graff C.
      • Wiley H.S.
      • Wittrup K.D.
      Flow-cytometric isolation of human antibodies from a nonimmune Saccharomyces cerevisiae surface display library.
      ) against the type I IL-4 cytokine–receptor ternary complex (i.e. cytokine bound to two receptor subunits) through iterative rounds of magnetic-activated cell sorting (MACS) (Fig. 1B). There were two critical considerations for the selection of stapler scFvs: 1) the stability of the cytokine–receptor complex selection reagent that served as the “molecular cast” of the signaling dimer and 2) the use of a multistep selection scheme to positively select against these stabilized complexes and negatively select against the individual subunits to enrich for scFvs that specifically bind to receptor dimers.
      To ensure stability of the target complex throughout the IL-4 stapler selections, we substituted the cytokine with an engineered variant of IL-4 (denoted “Super-4”) that has 3700-fold higher affinity for γc (
      • Junttila I.S.
      • Creusot R.J.
      • Moraga I.
      • Bates D.L.
      • Wong M.T.
      • Alonso M.N.
      • Suhoski M.M.
      • Lupardus P.
      • Meier-Schellersheim M.
      • Engleman E.G.
      • Utz P.J.
      • Fathman C.G.
      • Paul W.E.
      • Garcia K.C.
      Redirecting cell-type specific cytokine responses with engineered interleukin-4 superkines.
      ) and used the stabilized Super-4/IL-4Rα/γc ternary complex as our positive selection reagent. For each round of selection, we carried out two steps of negative selection against 1) γc and 2) IL-4/IL-4Rα binary complex preceding a positive selection step against the Super-4/IL-4Rα/γc target complex (Fig. S2). This sequence was iterated over six rounds of selection, decreasing only the concentration of the target complex to increase stringency.

      Staplers selectively and specifically bind the cytokine–receptor ternary signaling complex

      Following selections, the library converged on three scFv clones denoted A2 (henceforth referred to as “IL-4 stapler”), A8, and A11 (Fig. S3). The yeast-displayed IL-4 stapler, A8, and A11 scFvs exhibited robust binding to the Super-4 ternary complex but not to any of the individual components or binary complex combinations thereof (Fig. 1C and Fig. S4). These scFvs were also specific for the Super-4/IL-4Rα/γc ternary complex and showed no reactivity with the IL-13/IL-4Rα/IL-13Rα1 or IL-2/IL-2Rβ/γc ternary complexes, indicating that receptor chain engagement occurred only in the context of the assembled IL-4Rα/γc heterodimer. On-yeast binding studies also revealed that IL-4 stapler bound weakly to the IL-4Rα chain and the Super-4/IL-4Rα binary complex, but binding was substantially enhanced in the presence of the full Super-4/IL-4Rα/γc ternary complex (Fig. 1C). Importantly, our engineered stapler scFvs did not bind to the type II IL-4 receptor complex composed of IL-4Rα/IL-13Rα1 and are thus highly specific for the IL-4Rα/γc type I IL-4 receptor heterodimer. This represents a major difference from endogenous IL-4, which cross-reacts with both type I and type II receptors and induces pleiotropic effects that have limited the utility of cytokine as a therapeutic to boost the type 2 helper T cell immune response (
      • Junttila I.S.
      • Creusot R.J.
      • Moraga I.
      • Bates D.L.
      • Wong M.T.
      • Alonso M.N.
      • Suhoski M.M.
      • Lupardus P.
      • Meier-Schellersheim M.
      • Engleman E.G.
      • Utz P.J.
      • Fathman C.G.
      • Paul W.E.
      • Garcia K.C.
      Redirecting cell-type specific cytokine responses with engineered interleukin-4 superkines.
      ).
      The binding and specificity of recombinantly expressed IL-4 stapler scFv to the Super-4/IL-4Rα/γc target complex was evaluated via surface plasmon resonance. Consistent with on-yeast studies, IL-4 stapler scFv showed no binding to either Super-4 or γc and bound the IL-4Rα subunit with weak affinity, but it interacted strongly with the full ternary complex (Fig. 1D). The equilibrium binding constant (KD) for the stapler/ternary complex interaction was found to be 880 nm, whereas IL-4 stapler/IL-4Rα interaction was >10-fold weaker (Fig. 1E and Fig. S5), confirming the selectivity of IL-4 stapler for the active signaling complex. The A8 and A11 scFvs also bound the Super-4 ternary complex, but with affinities in the micromolar range (Fig. S4). By contrast, the IL-4 cytokine binds the IL-4Rα subunit alone with ∼150 pm affinity (
      • Hage T.
      • Sebald W.
      • Reinemer P.
      Crystal structure of the interleukin-4/receptor α chain complex reveals a mosaic binding interface.
      ).

      Crystal structure of the ternary complex–bound IL-4 stapler reveals shared epitope between the IL-4Rα and γc subunits

      To obtain structural evidence that the IL-4 stapler scFv recognizes a composite epitope formed by the conjunction of two receptor subunits (Fig. 1A), we reformatted the stapler scFv as a Fab fragment and determined the crystal structure of the Super-4/IL-4Rα/γc ternary complex bound to the stapler Fab to 3.1 Å resolution (Fig. 2A, Fig. S6, and Table 1). As selections and binding studies were performed in the presence of cytokine, we expected that the stapler would engage the membrane-proximal “stem regions” of the receptor subunits, distal from the site of cytokine engagement. The Fab heavy-chain (HC)/light-chain (LC) heterodimer axis coincides almost perfectly with the IL-4Rα/γc dimer interface. We also found close agreement between the orientation of the IL-4Rα and γc subunits in the presence of the stapler compared with the natural ternary complex. The receptor chains in the stapler-bound Super-4/IL-4Rα/γc ternary complex align closely with their conformations in the native IL-4 type I ternary complex (PDB code 3BPL, RMSD = 0.506 Å for IL-4Rα and RMSD = 0.643 Å for γc). Furthermore, fixing the IL-4Rα1 subunit in both structures, the γc subunit is rotated by just 5.3° in the presence of the stapler. By comparison, a diabody that homodimerizes the erythropoietin receptor but elicits only weak agonist activity induces a 151° relative rotation in the receptor subunits (
      • Moraga I.
      • Wernig G.
      • Wilmes S.
      • Gryshkova V.
      • Richter C.P.
      • Hong W.-J.
      • Sinha R.
      • Guo F.
      • Fabionar H.
      • Wehrman T.S.
      • Krutzik P.
      • Demharter S.
      • Plo I.
      • Weissman I.L.
      • Minary P.
      • et al.
      Tuning cytokine receptor signaling by re-orienting dimer geometry with surrogate ligands.
      ). The IL-4 stapler Fab interface with the receptor subunits is partitioned between the variable domains; whereas the IL-4Rα interface is dominated by contacts with the variable heavy (VH) chain, the γc interface principally comprises contacts with the variable light (VL) chain (Fig. 2B and Fig. S7). For both the VH and VL domains, all three complementarity-determining regions (CDRs) are implicated in receptor interactions. The more extensive Fab interface with IL-4Rα compared with γc rationalizes the weak affinity observed between IL-4 stapler and IL-4Rα alone (Fig. 2B and Fig. S7).
      Figure thumbnail gr2
      Figure 2Stapler recognizes a composite epitope between two receptor subunits to bridge the dimer interface. A, orthosteric views of the crystallographic structure of the IL-4 stapler Fab fragment bound to the Super-4/IL-4Rα/γc ternary complex. B, enlarged views of the IL-4 stapler interfaces with the IL-4Rα and γc subunits. At the left, IL-4 stapler-binding sites on IL-4Rα and γc are depicted in gray, with interacting residues in the CDRs of the VH (magenta) and VL (olive) domains of the IL-4 stapler Fab shown as sticks. At the right, the CDRs on VH (magenta) and VL (olive) of the IL-4 stapler Fab are shaded.
      Table 1Crystallographic statistics for Super-4/IL-4Rα/γc/IL-4 stapler Fab complex
      ParametersValues
      Data collection
       Wavelength0.9999
       Resolution range63.1–3.10 (3.22–3.10)
      Values in parentheses are for the highest-resolution shell.
       Space groupF 41 3 2
       Unit cell328.1, 328.1, 328.1, 90, 90, 90
       Total reflections493,043 (45,665)
       Unique reflections27,836 (1462)
       Completeness (%)95.0 (53.3)
       Redundancy17.7 (16.8)
       Mean I/σ(I)14.9 (0.92)
      R-merge0.275 (4.18)
       CC1/20.997 (0.343)
      Refinement statistics
       Reflections used in refinement26,537 (1462)
       Reflections used for Rfree1327 (74)
      Rwork0.199 (0.316)
      Rfree0.256 (0.4118)
       Number of non-hydrogen atoms7209
        Macromolecules7091
        Ligands98
        Solvent20
       Protein residues922
       RMSD (bonds)0.003
       RMSD (angles)0.57
       Ramachandran favored (outliers) (%)95.0 (0)
       Average B-factor77.1
        Macromolecules76.6
        Ligands112
        Solvent77.1
       Number of TLS groups22
      a Values in parentheses are for the highest-resolution shell.

      Isolating stapler scFvs that stabilize cytokine–receptor interactions

      Given our success in engineering scFvs that recognize epitopes that span multiple cytokine receptor subunits, we attempted to extend the stapling concept to select for interface-bridging scFvs that stabilize interactions between cytokines and their cognate receptor subunits (Fig. 3A). In particular, we aimed to evolve a single-chain species that potentiates the interaction between the interleukin-2 (IL-2) cytokine and one of its receptor subunits. IL-2 signals through either a high-affinity (KD ≈ 10 pm) heterotrimeric receptor comprised of the IL-2Rα, IL-2Rβ, and γc chains or an intermediate-affinity (KD ≈ 1 nm) heterodimeric receptor consisting of only the IL-2Rβ and γc chains (Fig. S8) (
      • Liao W.
      • Lin J.-X.
      • Leonard W.J.
      Interleukin-2 at the crossroads of effector responses, tolerance, and immunotherapy.
      ,
      • Wang X.
      • Rickert M.
      • Garcia K.C.
      Structure of the quaternary complex of interleukin-2 with its α, β, and γc receptors.
      • Stauber D.J.
      • Debler E.W.
      • Horton P.A.
      • Smith K.A.
      • Wilson I.A.
      Crystal structure of the IL-2 signaling complex: paradigm for a heterotrimeric cytokine receptor.
      ). The IL-2Rα chain is not implicated in signaling but rather calibrates IL-2 sensitivity through its differential expression levels on various cell types. IL-2Rα is highly expressed on regulatory T (TReg) cells but virtually absent from naive CD8+ T and natural killer (NK) cells; hence, TReg cells are 100-fold more responsive to IL-2 than these immune effector cell types (
      • Liao W.
      • Lin J.-X.
      • Leonard W.J.
      Interleukin-2 at the crossroads of effector responses, tolerance, and immunotherapy.
      ,
      • Malek T.R.
      The biology of interleukin-2.
      ). A variant of IL-2 with 200-fold enhanced affinity for the IL-2Rβ chain (denoted “Super-2”) was previously engineered to sensitize IL-2Rα–deficient cells to IL-2. Super-2 promoted expansion of cytotoxic T cells to effectively inhibit tumor growth in mice (
      • Levin A.M.
      • Bates D.L.
      • Ring A.M.
      • Krieg C.
      • Lin J.T.
      • Su L.
      • Moraga I.
      • Raeber M.E.
      • Bowman G.R.
      • Novick P.
      • Pande V.S.
      • Fathman C.G.
      • Boyman O.
      • Garcia K.C.
      Exploiting a natural conformational switch to engineer an interleukin-2 “superkine.”.
      ). We hypothesized that an scFv stapler that stabilized the IL-2/IL-2Rβ interaction would recapitulate the behavior of Super-2. Moreover, therapeutic use of an exogenous antibody offers several advantages over administration of a cytokine mutant, such as improved serum half-life (
      • Finkelman F.D.
      • Madden K.B.
      • Morris S.C.
      • Holmes J.M.
      • Boiani N.
      • Katona I.M.
      • Maliszewski C.R.
      Anti-cytokine antibodies as carrier proteins: prolongation of in vivo effects of exogenous cytokines by injection of cytokine-anti-cytokine antibody complexes.
      ,
      • Roopenian D.C.
      • Christianson G.J.
      • Sproule T.J.
      • Brown A.C.
      • Akilesh S.
      • Jung N.
      • Petkova S.
      • Avanessian L.
      • Choi E.Y.
      • Shaffer D.J.
      • Eden P.A.
      • Anderson C.L.
      The MHC class I-like IgG receptor controls perinatal IgG transport, IgG homeostasis, and fate of IgG-Fc-coupled drugs.
      ), controllable dose-dependent activation of IL-2–responsive cells, and the ability to selectively modulate rather than compete with endogenous IL-2.
      Figure thumbnail gr3
      Figure 3Evolved staplers specifically engage cytokine–receptor binary complex with engineerable affinity. A, schematic contrasting sequential assembly of a cytokine–receptor ternary complex in the absence (top) or presence (bottom) of a binary complex–stabilizing stapler scFv. B, yeast display platform diagram and selection strategy for evolution of cytokine–receptor binary complex stapler scFvs. C, binding of yeast-displayed IL-2B (left) or amIL-2B (right) stapler scFv to cytokine–receptor complexes and individual components thereof. Note the selective recognition of the IL-2/IL-2Rβ binary complex and compatibility with γc binding. TC, ternary complex. D, kinetic surface plasmon resonance binding profiles for the IL-2B stapler:IL-2 BC (left) and amIL-2B stapler:IL-2 BC (right) interactions. As shown in the traces, the affinity-matured stapler variant improves affinity 14-fold through deceleration of the dissociation rate. KD values were calculated by fitting equilibrium titration curves to a logistic model via nonlinear regression.
      Recently, Kuruganti et al. (
      • Kuruganti S.
      • Miersch S.
      • Deshpande A.
      • Speir J.A.
      • Harris B.D.
      • Schriewer J.M.
      • Buller R.M.L.
      • Sidhu S.S.
      • Walter M.R.
      Cytokine activation by antibody fragments targeted to cytokine-receptor signaling complexes.
      ) selected antibody fragments that stabilize the IFN/interferon receptor 2 (IFNAR2) binary complex to potentiate IFN signaling. Following a selection scheme similar to those used for IL-4 stapler discovery (Fig. 3B and Fig. S9), we evolved an scFv that uniquely recognizes the IL-2/IL-2Rβ binary complex but not either individual component thereof. We performed iterative rounds of selection against the aforementioned scFv library (
      • Feldhaus M.J.
      • Siegel R.W.
      • Opresko L.K.
      • Coleman J.R.
      • Feldhaus J.M.W.
      • Yeung Y.A.
      • Cochran J.R.
      • Heinzelman P.
      • Colby D.
      • Swers J.
      • Graff C.
      • Wiley H.S.
      • Wittrup K.D.
      Flow-cytometric isolation of human antibodies from a nonimmune Saccharomyces cerevisiae surface display library.
      ), clearing independently against 1) Super-2 and 2) IL-2Rβ prior to positively selecting for binders to the Super-2/IL-2Rβ complex, again capitalizing on the enhanced affinity of a mutant cytokine (Super-2) to stabilize the target complex (Fig. 3B). Selections converged on a single scFv clone (denoted the IL-2 binary complex stapler, or “IL-2B stapler”) that was then affinity-matured through selection of a site-specific mutagenic library of the VH and VL CDR3 regions (Fig. S10).
      Both the parent IL-2B stapler and the affinity-matured variant thereof (denoted the “amIL-2B stapler”) bound the IL-2/IL-2Rβ binary complex, but not the cytokine or receptor independently (Fig. 3C). IL-2B staplers also bound the IL-2/IL-2Rβ/γc ternary complex, indicating that they recognize an epitope that is noncompetitive with γc engagement of the IL-2/IL-2Rβ binary complex, which is critical for signaling activity. The IL-2B stapler bound the stabilized Super-2/IL-2Rβ binary complex with similar affinity (KD = 210 nm) to the IL-2/IL-2Rβ interaction (
      • Levin A.M.
      • Bates D.L.
      • Ring A.M.
      • Krieg C.
      • Lin J.T.
      • Su L.
      • Moraga I.
      • Raeber M.E.
      • Bowman G.R.
      • Novick P.
      • Pande V.S.
      • Fathman C.G.
      • Boyman O.
      • Garcia K.C.
      Exploiting a natural conformational switch to engineer an interleukin-2 “superkine.”.
      ), and the amIL-2BC stapler exhibited a 14-fold higher affinity for the binary complex (KD = 15 nm; Fig. 3D and Fig. S11A). Both IL-2B stapler scFvs enhanced the IL-2/IL-2Rβ interaction ∼15-fold, although this improvement was less dramatic than the 70-fold enhancement observed with the Super-2 cytokine mutant (Fig. 4A). Due to this more modest affinity improvement and the high potency of WT IL-2 signaling, IL-2B staplers did not potentiate cytokine activity on IL-2–responsive cells, as measured by phosphorylation of the downstream transcription factor signal transducer and activator of transcription 5 (STAT5, Fig. S11B). We speculated that the effect of the staplers would be more pronounced in the context of an impaired IL-2 variant reported by Shanafelt et al. (
      • Shanafelt A.B.
      • Lin Y.
      • Shanafelt M.C.
      • Forte C.P.
      • Dubois-Stringfellow N.
      • Carter C.
      • Gibbons J.A.
      • Cheng S.L.
      • Delaria K.A.
      • Fleischer R.
      • Greve J.M.
      • Gundel R.
      • Harris K.
      • Kelly R.
      • Koh B.
      • et al.
      A T-cell-selective interleukin 2 mutein exhibits potent antitumor activity and is well tolerated in vivo.
      ), which exhibits weakened activity due to its reduced affinity for IL-2Rβ. This variant (which we denote “iIL-2”) has been shown to selectively bias cytokine activity toward particular immune cell subsets based on their IL-2 receptor expression profiles (
      • Shanafelt A.B.
      • Lin Y.
      • Shanafelt M.C.
      • Forte C.P.
      • Dubois-Stringfellow N.
      • Carter C.
      • Gibbons J.A.
      • Cheng S.L.
      • Delaria K.A.
      • Fleischer R.
      • Greve J.M.
      • Gundel R.
      • Harris K.
      • Kelly R.
      • Koh B.
      • et al.
      A T-cell-selective interleukin 2 mutein exhibits potent antitumor activity and is well tolerated in vivo.
      ). IL-2B stapler and amIL-2B stapler enhanced iIL-2–induced STAT5 activity 9- and 23-fold, respectively (Fig. 4B), establishing the potential utility of binary complex stapler scFvs to selectively potentiate immune cells for such applications as T cell activation in cancer immunotherapy (
      • Liao W.
      • Lin J.-X.
      • Leonard W.J.
      Interleukin-2 at the crossroads of effector responses, tolerance, and immunotherapy.
      ,
      • Rosenberg S.A.
      Raising the bar: the curative potential of human cancer immunotherapy.
      ). Taken together, our biophysical, structural, and functional characterization of staplers demonstrates that through stringent selection against stabilized cytokine–receptor complexes, we have engineered antibody fragments that specifically recognize and modulate the stability of protein interfaces.
      Figure thumbnail gr4
      Figure 4Binary complex staplers strengthen cytokine–receptor interactions to potentiate signaling of an affinity-impaired cytokine variant. A, equilibrium surface plasmon resonance titrations of the IL-2:IL-2Rβ binary complex interaction without scFv present (filled blue circles) and with IL-2BC stapler scFv (gold circles) or amIL-2B stapler scFv (cyan circles) compared with the Super-2:IL-2Rβ interaction (unfilled blue circles). KD values were calculated by fitting equilibrium titration curves to a logistic model via nonlinear regression. IL-2B staplers increase the IL-2:IL-2Rβ interaction affinity ∼15-fold. B, YT-1 cell STAT5 phosphorylation responses to impaired affinity IL-2 (iIL-2) without scFv present (filled blue circles) and with saturating levels of IL-2B stapler (gold circles) or amIL-2B stapler (cyan circles) present compared with WT IL-2 (unfilled blue circles). EC50 values are indicated. Error bars, S.D.

      Discussion

      The ability to isolate scFvs that use a single binding site to bridge bimolecular interactions from a naive antibody fragment library presents an exciting opportunity in antibody engineering. A naive scFv library of diversity 109, which was derived from pooled spleen and lymph nodes of 58 human donors (
      • Feldhaus M.J.
      • Siegel R.W.
      • Opresko L.K.
      • Coleman J.R.
      • Feldhaus J.M.W.
      • Yeung Y.A.
      • Cochran J.R.
      • Heinzelman P.
      • Colby D.
      • Swers J.
      • Graff C.
      • Wiley H.S.
      • Wittrup K.D.
      Flow-cytometric isolation of human antibodies from a nonimmune Saccharomyces cerevisiae surface display library.
      ), yielded three unique IL-4 ternary complex stapler clones and one unique IL-2 binary complex stapler clone, and it will be interesting to explore the frequency of stapler antibodies in both natural and synthetic libraries going forward. Moreover, our IL-2 cytokine–receptor stapler demonstrates the unexpected ability to stabilize a protein interface. Previous work has shown that catalytic antibodies can be elicited by raising antibodies that target the transition state of reactions (
      • Pollack S.J.
      • Jacobs J.W.
      • Schultz P.G.
      Selective chemical catalysis by an antibody.
      ,
      • Tramontano A.
      • Janda K.D.
      • Lerner R.A.
      Catalytic antibodies.
      ). Our biophysical and functional data suggest that staplers instead act through stabilization of the receptor heterodimer to bias the equilibrium toward the dimeric over the monomeric receptor state.
      Whereas our study is limited to a proof of concept and in vitro studies, a potential advantage of the stapler platform is that it allows for modulation of the heterodimerization properties of membrane-bound receptors through recognition of specific receptor complexes. Regulating heterodimer distribution on cells has value from both a research discovery and a therapeutic perspective (
      • Junttila I.S.
      • Creusot R.J.
      • Moraga I.
      • Bates D.L.
      • Wong M.T.
      • Alonso M.N.
      • Suhoski M.M.
      • Lupardus P.
      • Meier-Schellersheim M.
      • Engleman E.G.
      • Utz P.J.
      • Fathman C.G.
      • Paul W.E.
      • Garcia K.C.
      Redirecting cell-type specific cytokine responses with engineered interleukin-4 superkines.
      ). Altered dimerization patterns could have profound functional implications, particularly for cytokine systems such as IL-4, which signal through formation of either a type I or type II receptor complex to evoke distinct immune behavior (
      • Wang X.
      • Lupardus P.
      • Laporte S.L.
      • Garcia K.C.
      Structural biology of shared cytokine receptors.
      ,
      • LaPorte S.L.
      • Juo Z.S.
      • Vaclavikova J.
      • Colf L.A.
      • Qi X.
      • Heller N.M.
      • Keegan A.D.
      • Garcia K.C.
      Molecular and structural basis of cytokine receptor pleiotropy in the interleukin-4/13 system.
      ). In addition, the specificity of staplers makes a compelling case for their use in targeted activation of transplanted cells. Previous efforts have been employed to engineer natural cytokines (
      • Spangler J.B.
      • Moraga I.
      • Mendoza J.L.
      • Garcia K.C.
      Insights into cytokine–receptor interactions from cytokine engineering.
      ,
      • Junttila I.S.
      • Creusot R.J.
      • Moraga I.
      • Bates D.L.
      • Wong M.T.
      • Alonso M.N.
      • Suhoski M.M.
      • Lupardus P.
      • Meier-Schellersheim M.
      • Engleman E.G.
      • Utz P.J.
      • Fathman C.G.
      • Paul W.E.
      • Garcia K.C.
      Redirecting cell-type specific cytokine responses with engineered interleukin-4 superkines.
      ,
      • Levin A.M.
      • Bates D.L.
      • Ring A.M.
      • Krieg C.
      • Lin J.T.
      • Su L.
      • Moraga I.
      • Raeber M.E.
      • Bowman G.R.
      • Novick P.
      • Pande V.S.
      • Fathman C.G.
      • Boyman O.
      • Garcia K.C.
      Exploiting a natural conformational switch to engineer an interleukin-2 “superkine.”.
      ,
      • Levin D.
      • Schneider W.M.
      • Hoffmann H.-H.
      • Yarden G.
      • Busetto A.G.
      • Manor O.
      • Sharma N.
      • Rice C.M.
      • Schreiber G.
      Multifaceted activities of type I interferon are revealed by a receptor antagonist.
      ), to design synthetic cytokine fusions (
      • Moraga I.
      • Spangler J.B.
      • Mendoza J.L.
      • Gakovic M.
      • Wehrman T.S.
      • Krutzik P.
      • Garcia K.C.
      Synthekines are surrogate cytokine and growth factor agonists that compel signaling through non-natural receptor dimers.
      ), or to computationally engineer de novo cytokine mimetics (
      • Silva D.-A.
      • Yu S.
      • Ulge U.Y.
      • Spangler J.B.
      • Jude K.M.
      • Labão-Almeida C.
      • Ali L.R.
      • Quijano-Rubio A.
      • Ruterbusch M.
      • Leung I.
      • Biary T.
      • Crowley S.J.
      • Marcos E.
      • Walkey C.D.
      • Weitzner B.D.
      • et al.
      De novo design of potent and selective mimics of IL-2 and IL-15.
      ) to redirect cytokine activities to specific cell subsets. The stapler strategy allows for an additional layer of control through administration of a molecule that is noncompetitive with natural cytokine activity.
      The engineering opportunities available to the scFv scaffold enable possibilities for molecular therapeutic design. Stapler scFvs can be conjugated to scFvs that engage cell surface markers to form bispecific antibodies that selectively modulate dimerization behavior on particular immune cell subsets for targeted disease therapy. Efforts to achieve cell subset-specific signaling by directly fusing cytokines to surface marker–directed antibodies have been unsuccessful, as the high potency of cytokines nullifies the effect of targeting antibodies, resulting in nonspecific activation of many immune cell subtypes (
      • Shusterman S.
      • London W.B.
      • Gillies S.D.
      • Hank J.A.
      • Voss S.D.
      • Seeger R.C.
      • Reynolds C.P.
      • Kimball J.
      • Albertini M.R.
      • Wagner B.
      • Gan J.
      • Eickhoff J.
      • DeSantes K.B.
      • Cohn S.L.
      • Hecht T.
      • et al.
      Antitumor activity of Hu14.18-IL2 in patients with relapsed/refractory neuroblastoma: a Children's Oncology Group (COG) phase II study.
      ,
      • Tzeng A.
      • Kwan B.H.
      • Opel C.F.
      • Navaratna T.
      • Wittrup K.D.
      Antigen specificity can be irrelevant to immunocytokine efficacy and biodistribution.
      • Yu A.L.
      • Gilman A.L.
      • Ozkaynak M.F.
      • London W.B.
      • Kreissman S.G.
      • Chen H.X.
      • Smith M.
      • Anderson B.
      • Villablanca J.G.
      • Matthay K.K.
      • Shimada H.
      • Grupp S.A.
      • Seeger R.
      • Reynolds C.P.
      • Buxton A.
      • et al.
      Anti-GD2 antibody with GM-CSF, interleukin-2, and isotretinoin for neuroblastoma.
      ). In contrast, the affinities of the stapler scFv and cell surface marker–targeted scFvs can be easily evolved through manipulation of the CDR loop regions (as was done in our binary stapler affinity maturation) to finely control their relative potencies and thereby optimize functional selectivity. Furthermore, discovery and structural characterization of the IL-4 stapler paves the way for engineering staplers that modulate other γc-engaging cytokines (including IL-7, IL-9, and IL-21) (
      • Overwijk W.W.
      • Schluns K.S.
      Functions of γC cytokines in immune homeostasis: current and potential clinical applications.
      ). In summary, we have established a new evolutionary approach to isolate a novel class of stapler scFvs. These engineered molecules offer insight into the biophysical requirements for dimerization and signal activation, and they provide a structural framework for the development of targeted immunotherapeutics.

      Experimental procedures

      Protein expression and purification

      Human IL-2 and variants thereof (amino acids 1–133), the IL-2Rβ ectodomain (amino acids 1–214), γc (amino acids 34–232), IL-2Rα (amino acids 1–217), IL-15 (amino acids 1–114), and IL-15Rα (amino acids 1–67) were expressed and purified using a baculovirus expression system as described (
      • Wang X.
      • Rickert M.
      • Garcia K.C.
      Structure of the quaternary complex of interleukin-2 with its α, β, and γc receptors.
      ). Super-2 refers to the previously reported IL-2 variant denoted clone H9 (
      • Levin A.M.
      • Bates D.L.
      • Ring A.M.
      • Krieg C.
      • Lin J.T.
      • Su L.
      • Moraga I.
      • Raeber M.E.
      • Bowman G.R.
      • Novick P.
      • Pande V.S.
      • Fathman C.G.
      • Boyman O.
      • Garcia K.C.
      Exploiting a natural conformational switch to engineer an interleukin-2 “superkine.”.
      ). Additionally, human IL-4 and variants thereof (amino acids 1–129), IL-13 (amino acids 1–112), the IL-4Rα ectodomain (amino acids 1–202), and the IL-13Rα1 ectodomain (amino acids 1–310) were expressed and purified using a baculovirus expression system (
      • LaPorte S.L.
      • Juo Z.S.
      • Vaclavikova J.
      • Colf L.A.
      • Qi X.
      • Heller N.M.
      • Keegan A.D.
      • Garcia K.C.
      Molecular and structural basis of cytokine receptor pleiotropy in the interleukin-4/13 system.
      ). The sequence for the Super-4 variant of IL-4 was provided previously (
      • Junttila I.S.
      • Creusot R.J.
      • Moraga I.
      • Bates D.L.
      • Wong M.T.
      • Alonso M.N.
      • Suhoski M.M.
      • Lupardus P.
      • Meier-Schellersheim M.
      • Engleman E.G.
      • Utz P.J.
      • Fathman C.G.
      • Paul W.E.
      • Garcia K.C.
      Redirecting cell-type specific cytokine responses with engineered interleukin-4 superkines.
      ). scFv clones isolated from selections were expressed and purified in the baculovirus system via transfer of their variable regions into the pAcGP67A vector (BD Biosciences) with an N-terminal gp67 signal peptide and a C-terminal hexahistidine tag. scFvs were expressed with the VH and VL chains separated by a 15-amino acid (Gly4Ser)3 linker. To prevent isomerization, scFvs were eluted in 10 mm HEPES, pH 7.3, 200 mm imidazole in the presence of 15% glycerol and 500 mm NaCl and then exchanged into standard HEPES-buffered saline (10 mm HEPES, pH 7.3, 150 mm NaCl). The VH and VL chains of stapler scFvs were delineated using the International Immunogenetics Information System (IMGT) V-QUEST tool (
      • Brochet X.
      • Lefranc M.-P.
      • Giudicelli V.
      IMGT/V-QUEST: the highly customized and integrated system for IG and TR standardized V-J and V-D-J sequence analysis.
      ,
      • Giudicelli V.
      • Brochet X.
      • Lefranc M.-P.
      IMGT/V-QUEST: IMGT standardized analysis of the immunoglobulin (IG) and T cell receptor (TR) nucleotide sequences.
      ). For IL-4 stapler Fab expression, the identified VH followed by the human IgG1 constant heavy 1 (CH1) domain and hinge region and the identified VL domain followed by the human κ constant light (CL) domain were separately cloned into the pAcGP67A vector containing 3C protease-cleavable basic and acidic leucine zippers, respectively, for high-fidelity Fab chain pairing (
      • Bankovich A.J.
      • Raunser S.
      • Juo Z.S.
      • Walz T.
      • Davis M.M.
      • Garcia K.C.
      Structural insight into pre-B cell receptor function.
      ,
      • Chang H.C.
      • Bao Z.
      • Yao Y.
      • Tse A.G.
      • Goyarts E.C.
      • Madsen M.
      • Kawasaki E.
      • Brauer P.P.
      • Sacchettini J.C.
      • Nathenson S.G.
      A general method for facilitating heterodimeric pairing between two proteins: application to expression of alpha and beta T-cell receptor extracellular segments.
      ). The resulting Fab HC and LC constructs were transfected independently, and their corresponding viruses were co-titrated to determine optimal infection ratios for equivalent expression of the two chains. The secreted IL-4 stapler Fab was treated with 3C protease to remove the leucine zippers prior to size-exclusion chromatographic purification.
      For biotinylated protein expression, Super-4, IL-4Rα, IL-13Rα1, or γc with a C-terminal biotin acceptor peptide, LNDIFEAQKIEWHE, were co-expressed with the BirA ligase enzyme in the presence of excess biotin (100 mm), or the purified proteins were biotinylated in vitro using the soluble BirA ligase enzyme in 0.5 mm Bicine, pH 8.3, 100 mm ATP, 100 mm magnesium acetate, and 500 mm biotin (Sigma). For surface plasmon resonance (SPR) studies, purified stapler scFvs were biotinylated using the EZ-Link Sulfo-NHS-LC-biotinylation kit (Pierce) according to the manufacturer’s protocol. All proteins contained C-terminal hexahistidine tags and were isolated by nickel chromatography. They were then further purified to >98% homogeneity by size-exclusion chromatography on a Superdex 200 column (GE Healthcare), equilibrated in 10 mm HEPES (pH 7.3) and 150 mm NaCl.

      Tissue culture

      YT-1 human NK cells (
      • Yodoi J.
      • Teshigawara K.
      • Nikaido T.
      • Fukui K.
      • Noma T.
      • Honjo T.
      • Takigawa M.
      • Sasaki M.
      • Minato N.
      • Tsudo M.
      TCGF (IL 2)-receptor inducing factor(s). I. Regulation of IL 2 receptor on a natural killer-like cell line (YT cells).
      ) were cultured in RPMI complete medium (RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mm l-glutamine, minimum nonessential amino acids, sodium pyruvate, 25 mm HEPES, and penicillin-streptomycin (Gibco)). The Ramos human B lymphocyte cell line (
      • Klein G.
      • Giovanella B.
      • Westman A.
      • Stehlin J.S.
      • Mumford D.
      An EBV-genome-negative cell line established from an American Burkitt lymphoma; receptor characteristics. EBV infectibility and permanent conversion into EBV-positive sublines by in vitro infection.
      ) was grown in RPMI containing 10% FBS, l-glutamine (2 mm), and penicillin/streptomycin (Gibco). Prior to stimulation, Ramos cells were cultured overnight in modified growth medium containing only 2% FBS (“starved”).

      Yeast library selection of stapler scFvs

      General yeast display methodologies were modified from protocols described previously (
      • Levin A.M.
      • Bates D.L.
      • Ring A.M.
      • Krieg C.
      • Lin J.T.
      • Su L.
      • Moraga I.
      • Raeber M.E.
      • Bowman G.R.
      • Novick P.
      • Pande V.S.
      • Fathman C.G.
      • Boyman O.
      • Garcia K.C.
      Exploiting a natural conformational switch to engineer an interleukin-2 “superkine.”.
      ,
      • Boder E.T.
      • Wittrup K.D.
      Yeast surface display for screening combinatorial polypeptide libraries.
      ,
      • Chao G.
      • Lau W.L.
      • Hackel B.J.
      • Sazinsky S.L.
      • Lippow S.M.
      • Wittrup K.D.
      Isolating and engineering human antibodies using yeast surface display.
      ). The previously described human nonimmune scFv yeast library (
      • Feldhaus M.J.
      • Siegel R.W.
      • Opresko L.K.
      • Coleman J.R.
      • Feldhaus J.M.W.
      • Yeung Y.A.
      • Cochran J.R.
      • Heinzelman P.
      • Colby D.
      • Swers J.
      • Graff C.
      • Wiley H.S.
      • Wittrup K.D.
      Flow-cytometric isolation of human antibodies from a nonimmune Saccharomyces cerevisiae surface display library.
      ) was kindly provided by K. Dane Wittrup and was used to isolate the IL-4 and IL-2 stapler scFvs. For all selection schemes, the first round of sorting was performed with 1 × 1010 cells from the naive yeast library to achieve 10-fold coverage of the number of transformants, and subsequent rounds used 1 × 108 yeast cells. Tetramers were formed by incubating a 4:1 ratio of biotinylated protein to Alexa 647–conjugated streptavidin (SA-647) for 15 min on ice. Magnetic bead selections were performed using LS MACS separation columns (Miltenyi Biotec) according to the manufacturer’s protocol. The naive scFv library and each iterative panel of selected clones were grown fresh overnight at 30 °C in SDCAA liquid medium (pH 4.5), followed by induction in SGCAA liquid medium (pH 4.5) for 2 days at 20 °C.

      IL-4 ternary complex stapler scFv selections

      In the first round of selection for the IL-4 ternary complex staplers, yeast-displayed scFvs were incubated with biotinylated γc bound to magnetic beads coated with SA (400 nm γc in 250 μl of SA-coated beads) (Miltenyi Biotec) in the presence of 1 μm nonbiotinylated IL-4Rα and Super-4, an IL-4 mutant that exhibits 3700-fold higher affinity for γc than WT IL-4 (
      • Junttila I.S.
      • Creusot R.J.
      • Moraga I.
      • Bates D.L.
      • Wong M.T.
      • Alonso M.N.
      • Suhoski M.M.
      • Lupardus P.
      • Meier-Schellersheim M.
      • Engleman E.G.
      • Utz P.J.
      • Fathman C.G.
      • Paul W.E.
      • Garcia K.C.
      Redirecting cell-type specific cytokine responses with engineered interleukin-4 superkines.
      ), in PBE solution (PBS, pH 7.2, 1% BSA, and 1 mm EDTA) for 2 h at 4 °C. In each subsequent round of selection, two steps of negative selection and one step of positive selection were carried out in sequence. In a first step, yeast were stained with biotinylated γc that had been tetramerized via incubation with SA-647 (500 nm) for 1 h in PBE at 4 °C and negatively selected by incubating the yeast with 50 μl of anti-Alexa 647 microbeads (Miltenyi Biotec) in PBE for 15 min at 4 °C. In a second step, the flow-through yeast from the first step were incubated with SA-647–tetramerized biotinylated IL-4Rα (500 nm) and 1 μm of nonbiotinylated Super-4 for 1 h in PBE at 4 °C. The yeast were then negatively selected via incubation with 50 μl of anti-Alexa 647 microbeads (Miltenyi Biotec) in PBE for 15 min at 4 °C. In a final step, the flow-through yeast from the second step were incubated with SA-647–tetramerized biotinylated γc (500 nm) in the presence of 1 μm nonbiotinylated IL-4Rα and Super-4 for 1 h in PBE at 4 °C and positively selected by incubation with 50 μl of anti-Alexa 647 microbeads (Miltenyi Biotec) in PBE for 15 min at 4 °C. This workflow was repeated over six rounds of selection in which only the concentration of the positive selection reagents was decreased to isolate scFvs specifically binding the IL-4 ternary complex with high affinity. Three scFv clones, denoted A2 (the “IL-4 stapler”), A8, and A11, were isolated, and all exhibited specific binding for the IL-4 ternary complex and no binding for the individual components of that complex (i.e. Super-4, IL-4Rα, or γc). The complementarity-determining region 3 (CDR3) domain sequences of isolated IL-4 staplers are provided in Fig. S3.

      IL-2 binary complex stapler scFv selections and affinity maturation

      In the first round of selection for an IL-2 binary complex stapler, the naive library of yeast-displayed scFvs was stained with biotinylated IL-2Rβ bound to SA-coated magnetic beads (400 nm IL-2Rβ in 250 μl of SA-coated beads) (Miltenyi Biotec) in the presence of 1 μm nonbiotinylated Super-2, a variant of IL-2 with 200-fold greater affinity for the IL-2Rβ receptor subunit (
      • Levin A.M.
      • Bates D.L.
      • Ring A.M.
      • Krieg C.
      • Lin J.T.
      • Su L.
      • Moraga I.
      • Raeber M.E.
      • Bowman G.R.
      • Novick P.
      • Pande V.S.
      • Fathman C.G.
      • Boyman O.
      • Garcia K.C.
      Exploiting a natural conformational switch to engineer an interleukin-2 “superkine.”.
      ), for 2 h at 4 °C. In each subsequent round of selection, two negative selection steps and one positive selection step were conducted sequentially. In a first step, yeast were stained with SA-647–tetramerized biotinylated Super-2 (500 nm) for 1 h in PBE at 4 °C. Negative selections were performed by incubating the yeast with 50 μl of anti-Alexa 647 microbeads (Miltenyi Biotec) in PBE for 15 min at 4 °C. In a second step, the flow-through yeast from the first step were incubated with SA-647–tetramerized biotinylated IL-2Rβ (500 nm) for 1 h in PBE at 4 °C. Negative selections were again performed by incubating the yeast with 50 μl of anti-Alexa 647 microbeads (Miltenyi Biotec) in PBE for 15 min at 4 °C. In a final step, the flow-through yeast from the second step were incubated with SA-647–tetramerized biotinylated IL-2Rβ (500 nm) and 1 μm nonbiotinylated Super-2 for 1 h at 4 °C. Positive selections were conducted by incubating the yeast with 50 μl of anti-Alexa 647 microbeads (Miltenyi Biotec) in PBE for 15 min at 4 °C. Iterations of this workflow were repeated over six rounds of selection in which only the concentrations of the positive selection reagents were decreased to isolate scFv clones specifically binding with high affinity to the IL-2 binary complex. Selections converged on a single clone, the IL-2 binary complex stapler (“IL-2B stapler”).
      The site-directed mutagenic library used for affinity maturation of IL-2B stapler was constructed via assembly PCR of 16 primers spanning the VH and VL chains of the scFv. The following degenerate codons were used: Ala-104, RYK; Val-105, RYK; Ser-106, DMY; Gly-107, RBY; Thr-108, DMY; Phe-109, HWY; Tyr-236, WWY; Asp-237, VAK; Asp-238, VAK; Asn-239, VAK; Tyr-240, WWY. The assembly PCR was carried out using Pfu Ultra DNA polymerase (Stratagene), and the product was further amplified using primers containing sequence homology to pCT302 for recombination inside the yeast. Insert DNA was combined with linearized vector backbone and electroporated into EBY100 yeast as described previously. Electroporation yielded a library of 1 × 108 transformants. Selections were performed as detailed for the IL-2 binary complex stapler, but with 10-fold reduced concentrations of positive selection reagents to heighten stringency. Selections converged on a single clone, the affinity-matured IL-2 binary complex stapler (amIL-2B stapler). The CDR3 domain sequences of the IL-2 binary complex stapler (IL-2B stapler) and the affinity-matured IL-2 binary stapler (amIL-2B stapler) are provided in Fig. S9.

      Yeast surface stapler scFv-binding studies

      Yeast displaying stapler scFvs (2 × 105/well) were plated in a 96-well plate and incubated in PBE buffer containing a 1 μm concentration of the indicated biotinylated IL-4 or IL-2 complex components for 2 h at room temperature. Cells were then washed and incubated for 20 min at 4 °C with 50 nm SA-647 diluted in PBE. After a final wash, cells were analyzed for binding using an Accuri C6 flow cytometer (BD Biosciences). Representative plots from one of three independent experiments are presented. Cytometry data were analyzed using FlowJo software (Tree Star, Inc.).

      Surface plasmon resonance binding measurements

      Binding interactions were characterized via SPR studies using Biacore SA sensor chips (GE Healthcare) on a Biacore T100 instrument at 25 °C. All recombinantly produced proteins were purified via size-exclusion chromatography on an FPLC instrument and filtered through a 0.22-μm centrifugal column prior to SPR analysis. Biotinylated scFv stapler clones were immobilized to the chip surface at low density (RUmax < 200 response units), and measurements were performed at a flow rate of 30 ml/min to minimize mass transport contributions and prevent rebinding of the analyte. Dilutions of the analytes (IL-4 cytokine–receptor complex components) were exposed to the surface channels for 60 s, and dissociation was measured for 180–300 s. For the case of protein complexes, complex components were incubated for 30 min at room temperature prior to SPR analysis. An irrelevant scFv was used in the reference channel to subtract nonspecific binding from the measurements. Experiments were performed in HBS-P+ buffer (GE Healthcare) supplemented with 0.2% BSA. Data were visualized and processed using the Biacore T100 evaluation software (GE Healthcare). Equilibrium titration curves were fitted to a logistic model via nonlinear regression, and equilibrium dissociation constant (KD) values were determined using GraphPad Prism analysis software assuming 1:1 binding interactions.

      Crystallization and data collection

      For crystallographic analysis, the Super-4, IL-4Rα, and γc viruses were co-expressed in insect cells in the presence of 5 μm kifunensine, added to culture at the time of infection. The resulting kifunensine-sensitized Super-4 ternary complex was purified via nickel chromatography and then treated overnight at 4 °C with 1:100 (w/w) endoglycosidases F1 and H to remove N-linked glycans and with 1:100 (w/w) carboxypeptidases A and B (Sigma) to remove C-terminal hexahistidine tags. The complex was then purified by size-exclusion chromatography. Separately, the IL-4 stapler Fab was secreted in insect cells via co-infection of heavy- and light-chain viruses and purified via nickel chromatography. The assembled Fab was then treated overnight at 4 °C with 1:1000 (w/w) 3C protease to remove acidic and basic leucine zippers and hexahistidine tags and purified by size-exclusion chromatography. Stoichiometrically equivalent amounts of the purified Super-4/IL-4Rα/γc ternary complex and the IL-4 stapler Fab were co-incubated for 4 h at 4 °C to allow for complex formation, and the resulting Fab-bound cytokine–receptor complex was purified by co-elution over a Superdex-200 size-exclusion chromatography column and concentrated to >10 mg/ml. Super-4/IL-4Rα/γc/IL-4 stapler Fab complex crystals were grown in sitting drops at 22 °C in 22.5% PurePEG (Microlytic), 0.1 m sodium citrate, pH 5.5, and 0.6 m ammonium nitrate. Crystals were cryoprotected with the addition of 30% ethylene glycol (Sigma) and flash-cooled in liquid nitrogen. A 3.1 Å data set was collected at beamline 8.2.1 at the Advanced Light Source. Diffraction data were processed using HKL2000. Crystallographic data collection and refinement statistics are reported in Table 1.

      Structure determination and refinement

      The Super-4/IL-4Rα/γc/IL-4 stapler Fab complex structure was solved by molecular replacement using PHASER (
      • McCoy A.J.
      • Grosse-Kunstleve R.W.
      • Adams P.D.
      • Winn M.D.
      • Storoni L.C.
      • Read R.J.
      Phaser crystallographic software.
      ). The IL-4 ternary complex structure reported previously (PDB code 3BPL) (
      • LaPorte S.L.
      • Juo Z.S.
      • Vaclavikova J.
      • Colf L.A.
      • Qi X.
      • Heller N.M.
      • Keegan A.D.
      • Garcia K.C.
      Molecular and structural basis of cytokine receptor pleiotropy in the interleukin-4/13 system.
      ) was used to model Super-4, IL-4Rα, and γc, and models of the IL-4 stapler Fab heavy and light chains were obtained from the previously solved structure of a human Fab (PDB code 3SOB) (
      • Bourhis E.
      • Wang W.
      • Tam C.
      • Hwang J.
      • Zhang Y.
      • Spittler D.
      • Huang O.W.
      • Gong Y.
      • Estevez A.
      • Zilberleyb I.
      • Rouge L.
      • Chiu C.
      • Wu Y.
      • Costa M.
      • Hannoush R.N.
      • Franke Y.
      • Cochran A.G.
      Wnt antagonists bind through a short peptide to the first β-propeller domain of LRP5/6.
      ). Iterative model rebuilding and refinement were performed using the Phenix software (
      • Adams P.D.
      • Grosse-Kunstleve R.W.
      • Hung L.W.
      • Ioerger T.R.
      • McCoy A.J.
      • Moriarty N.W.
      • Read R.J.
      • Sacchettini J.C.
      • Sauter N.K.
      • Terwilliger T.C.
      PHENIX: building new software for automated crystallographic structure determination.
      ) and COOT (Crystallographic Object-Oriented Toolkit) (
      • Emsley P.
      • Cowtan K.
      Coot: model-building tools for molecular graphics.
      ). For initial refinement, rigid body, coordinate, and real-space refinement were used with individual atomic displacement parameter refinement. Translation, libration, and screw-rotation refinement was added in later iterations. Ramachandran and rotamer analysis were performed using MolProbity (
      • Davis I.W.
      • Leaver-Fay A.
      • Chen V.B.
      • Block J.N.
      • Kapral G.J.
      • Wang X.
      • Murray L.W.
      • Arendall 3rd, W.B.
      • Snoeyink J.
      • Richardson J.S.
      • Richardson D.C.
      MolProbity: all-atom contacts and structure validation for proteins and nucleic acids.
      ). Atomic interactions were identified using the PISA (protein interfaces, surfaces, and assemblies) service at the European Bioinformatics Institute (
      • Krissinel E.
      • Henrick K.
      Inference of macromolecular assemblies from crystalline state.
      ). Structural figures were created using PyMOL (The PyMOL Molecular Graphics System, version 2.0 Schrödinger, LLC). Crystallographic software was installed and configured by SBGrid (
      • Morin A.
      • Eisenbraun B.
      • Key J.
      • Sanschagrin P.C.
      • Timony M.A.
      • Ottaviano M.
      • Sliz P.
      Collaboration gets the most out of software.
      ).

      YT-1 signal response experiments and phospho-flow cytometric analysis

      Approximately 3 × 105 YT-1 cells were plated in each well of a 96-well plate, washed with FACS buffer, and resuspended in FACS buffer containing serial dilutions of stapling scFvs or cytokines (used as positive controls for signal activation). Cells were stimulated for 20 min at 37 °C and immediately fixed by the addition of formaldehyde to 1.5% followed by incubation for 10 min at room temperature. Cells were then permeabilized with 100% ice-cold methanol for 30 min at 4 °C. The fixed and permeabilized cells were washed twice with FACS buffer and incubated with Alexa 488–conjugated anti-STAT5 pTyr-694 antibody (BD Biosciences) diluted in FACS buffer for 2 h at room temperature. Cells were then washed twice in FACS buffer, and mean fluorescence intensity was quantified on an Accuri C6 flow cytometer. Dose-response curves were fitted to a logistic model, and half-maximal effective concentrations (EC50 values) were computed in the GraphPad Prism data analysis software after subtraction of the mean fluorescence intensity of unstimulated cells and normalization to the maximum signal intensity induced by cytokine stimulation.

      Author contributions

      J. B. S. and I. M. evolved, biophysically characterized, and functionally validated stapler antibody fragments; J. B. S., I. M., and K. M. J. performed crystallographic studies of the stapler/cytokine receptor complexes and determined and refined these structures; J. B. S., I. M., K. M. J., and C. S. S. designed and prepared the cytokine proteins and recombinant antibody fragments used in this work; J. B. S., I. M., K. M. J., and K. C. G. designed experiments; J. B. S. and I. M. prepared the figures for this manuscript; J. B. S., I. M., and K. C. G. wrote the paper; and K.C.G. conceived of, designed, and supervised the research.

      Acknowledgments

      We thank members of the Garcia laboratory for helpful advice and discussions and A. Velasco, D. Waghray, and S. Fischer for technical assistance. The Advanced Light Source is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the United States Department of Energy under Contract DE-AC02-05CH11231.

      References

        • Ullrich A.
        • Schlessinger J.
        Signal transduction by receptors with tyrosine kinase activity.
        Cell. 1990; 61 (2158859): 203-212
        • Klemm J.D.
        • Schreiber S.L.
        • Crabtree G.R.
        Dimerization as a regulatory mechanism in signal transduction.
        Annu. Rev. Immunol. 1998; 16 (9597142): 569-592
        • Wang X.
        • Lupardus P.
        • Laporte S.L.
        • Garcia K.C.
        Structural biology of shared cytokine receptors.
        Annu. Rev. Immunol. 2009; 27 (18817510): 29-60
        • Bazan J.F.
        A novel family of growth factor receptors: a common binding domain in the growth hormone, prolactin, erythropoietin and IL-6 receptors, and the p75 IL-2 receptor β-chain.
        Biochem. Biophys. Res. Commun. 1989; 164 (2554900): 788-795
        • Rochman Y.
        • Spolski R.
        • Leonard W.J.
        New insights into the regulation of T cells by γ(c) family cytokines.
        Nat. Rev. Immunol. 2009; 9 (19543225): 480-490
        • Cunningham B.C.
        • Ultsch M.
        • De Vos A.M.
        • Mulkerrin M.G.
        • Clauser K.R.
        • Wells J.A.
        Dimerization of the extracellular domain of the human growth hormone receptor by a single hormone molecule.
        Science. 1991; 254 (1948064): 821-825
        • Watowich S.S.
        • Hilton D.J.
        • Lodish H.F.
        Activation and inhibition of erythropoietin receptor function: role of receptor dimerization.
        Mol. Cell Biol. 1994; 14 (8196600): 3535-3549
        • Ihle J.N.
        • Witthuhn B.A.
        • Quelle F.W.
        • Yamamoto K.
        • Silvennoinen O.
        Signaling through the hematopoietic cytokine receptors.
        Annu. Rev. Immunol. 1995; 13 (7612228): 369-398
        • Syed R.S.
        • Reid S.W.
        • Li C.
        • Cheetham J.C.
        • Aoki K.H.
        • Liu B.
        • Zhan H.
        • Osslund T.D.
        • Chirino A.J.
        • Zhang J.
        • Finer-Moore J.
        • Elliott S.
        • Sitney K.
        • Katz B.A.
        • Matthews D.J.
        • Wendoloski J.J.
        • Egrie J.
        • Stroud R.M.
        Efficiency of signalling through cytokine receptors depends critically on receptor orientation.
        Nature. 1998; 395 (9774108): 511-516
        • Livnah O.
        • Stura E.A.
        • Middleton S.A.
        • Johnson D.L.
        • Jolliffe L.K.
        • Wilson I.A.
        Crystallographic evidence for preformed dimers of erythropoietin receptor before ligand activation.
        Science. 1999; 283 (9974392): 987-990
        • Constantinescu S.N.
        • Keren T.
        • Socolovsky M.
        • Nam H.
        • Henis Y.I.
        • Lodish H.F.
        Ligand-independent oligomerization of cell-surface erythropoietin receptor is mediated by the transmembrane domain.
        Proc. Natl. Acad. Sci. U.S.A. 2001; 98 (11296286): 4379-4384
        • Delgoffe G.M.
        • Murray P.J.
        • Vignali D.A.A.
        Interpreting mixed signals: the cell's cytokine conundrum.
        Curr. Opin. Immunol. 2011; 23 (21852079): 632-638
        • Levy D.E.
        • Darnell J.E.
        Stats: transcriptional control and biological impact.
        Nat. Rev. Mol. Cell Biol. 2002; 3 (12209125): 651-662
        • Murray P.J.
        The JAK-STAT signaling pathway: input and output integration.
        J. Immunol. 2007; 178 (17312100): 2623-2629
        • O'Shea J.J.
        • Plenge R.
        JAK and STAT signaling molecules in immunoregulation and immune-mediated disease.
        Immunity. 2012; 36 (22520847): 542-550
        • Overwijk W.W.
        • Schluns K.S.
        Functions of γC cytokines in immune homeostasis: current and potential clinical applications.
        Clin. Immunol. 2009; 132 (19428306): 153-165
        • Pellegrini M.
        • Mak T.W.
        • Ohashi P.S.
        Fighting cancers from within: augmenting tumor immunity with cytokine therapy.
        Trends Pharmacol. Sci. 2010; 31 (20561689): 356-363
        • Donohue J.H.
        • Rosenberg S.A.
        The fate of interleukin-2 after in vivo administration.
        J. Immunol. 1983; 130 (6601147): 2203-2208
        • Andrews A.-L.
        • Holloway J.W.
        • Holgate S.T.
        • Davies D.E.
        IL-4 receptor α is an important modulator of IL-4 and IL-13 receptor binding: implications for the development of therapeutic targets.
        J. Immunol. 2006; 176 (16751391): 7456-7461
        • Borden E.C.
        • Sen G.C.
        • Uze G.
        • Silverman R.H.
        • Ransohoff R.M.
        • Foster G.R.
        • Stark G.R.
        Interferons at age 50: past, current and future impact on biomedicine.
        Nat. Rev. Drug Discov. 2007; 6 (18049472): 975-990
        • Boyman O.
        • Sprent J.
        The role of interleukin-2 during homeostasis and activation of the immune system.
        Nat. Rev. Immunol. 2012; 12 (22343569): 180-190
        • Spangler J.B.
        • Moraga I.
        • Mendoza J.L.
        • Garcia K.C.
        Insights into cytokine–receptor interactions from cytokine engineering.
        Annu. Rev. Immunol. 2015; 33 (25493332): 139-167
        • Finkelman F.D.
        • Madden K.B.
        • Morris S.C.
        • Holmes J.M.
        • Boiani N.
        • Katona I.M.
        • Maliszewski C.R.
        Anti-cytokine antibodies as carrier proteins: prolongation of in vivo effects of exogenous cytokines by injection of cytokine-anti-cytokine antibody complexes.
        J. Immunol. 1993; 151 (8393043): 1235-1244
        • Roopenian D.C.
        • Christianson G.J.
        • Sproule T.J.
        • Brown A.C.
        • Akilesh S.
        • Jung N.
        • Petkova S.
        • Avanessian L.
        • Choi E.Y.
        • Shaffer D.J.
        • Eden P.A.
        • Anderson C.L.
        The MHC class I-like IgG receptor controls perinatal IgG transport, IgG homeostasis, and fate of IgG-Fc-coupled drugs.
        J. Immunol. 2003; 170 (12646614): 3528-3533
        • Harwerth I.M.
        • Wels W.
        • Marte B.M.
        • Hynes N.E.
        Monoclonal antibodies against the extracellular domain of the erbB-2 receptor function as partial ligand agonists.
        J. Biol. Chem. 1992; 267 (1353079): 15160-15167
        • Müller-Newen G.
        • Küster A.
        • Wijdenes J.
        • Schaper F.
        • Heinrich P.C.
        Studies on the interleukin-6-type cytokine signal transducer gp130 reveal a novel mechanism of receptor activation by monoclonal antibodies.
        J. Biol. Chem. 2000; 275 (10671483): 4579-4586
        • Kai M.
        • Motoki K.
        • Yoshida H.
        • Emuta C.
        • Chisaka Y.
        • Tsuruhata K.
        • Endo C.
        • Muto M.
        • Shimabe M.
        • Nishiyama U.
        • Hagiwara T.
        • Matsumoto A.
        • Miyazaki H.
        • Kataoka S.
        Switching constant domains enhances agonist activities of antibodies to a thrombopoietin receptor.
        Nat. Biotechnol. 2008; 26 (18157117): 209-211
        • Frederickson S.
        • Renshaw M.W.
        • Lin B.
        • Smith L.M.
        • Calveley P.
        • Springhorn J.P.
        • Johnson K.
        • Wang Y.
        • Su X.
        • Shen Y.
        • Bowdish K.S.
        A rationally designed agonist antibody fragment that functionally mimics thrombopoietin.
        Proc. Natl. Acad. Sci. U.S.A. 2006; 103 (16973749): 14307-14312
        • Nakano K.
        • Kojima T.
        • Kasutani K.
        • Senoh C.
        • Natori O.
        • Ishii S.
        • Tsunoda H.
        • Hattori K.
        Effective screening method of agonistic diabodies based on autocrine growth.
        J. Immunol. Methods. 2009; 347 (19508873): 31-35
        • Zhang H.
        • Wilson I.A.
        • Lerner R.A.
        Selection of antibodies that regulate phenotype from intracellular combinatorial antibody libraries.
        Proc. Natl. Acad. Sci. U.S.A. 2012; 109 (23019357): 15728-15733
        • Moraga I.
        • Wernig G.
        • Wilmes S.
        • Gryshkova V.
        • Richter C.P.
        • Hong W.-J.
        • Sinha R.
        • Guo F.
        • Fabionar H.
        • Wehrman T.S.
        • Krutzik P.
        • Demharter S.
        • Plo I.
        • Weissman I.L.
        • Minary P.
        • et al.
        Tuning cytokine receptor signaling by re-orienting dimer geometry with surrogate ligands.
        Cell. 2015; 160 (25728669): 1196-1208
        • Kuruganti S.
        • Miersch S.
        • Deshpande A.
        • Speir J.A.
        • Harris B.D.
        • Schriewer J.M.
        • Buller R.M.L.
        • Sidhu S.S.
        • Walter M.R.
        Cytokine activation by antibody fragments targeted to cytokine-receptor signaling complexes.
        J. Biol. Chem. 2016; 291 (26546677): 447-461
        • Adams R.
        • Burnley R.J.
        • Valenzano C.R.
        • Qureshi O.
        • Doyle C.
        • Lumb S.
        • Del Carmen Lopez M.
        • Griffin R.
        • McMillan D.
        • Taylor R.D.
        • Meier C.
        • Mori P.
        • Griffin L.M.
        • Wernery U.
        • Kinne J.
        • et al.
        Discovery of a junctional epitope antibody that stabilizes IL-6 and gp80 protein:protein interaction and modulates its downstream signaling.
        Sci. Rep. 2017; 7 (28134246): 37716
        • LaPorte S.L.
        • Juo Z.S.
        • Vaclavikova J.
        • Colf L.A.
        • Qi X.
        • Heller N.M.
        • Keegan A.D.
        • Garcia K.C.
        Molecular and structural basis of cytokine receptor pleiotropy in the interleukin-4/13 system.
        Cell. 2008; 132 (18243101): 259-272
        • Feldhaus M.J.
        • Siegel R.W.
        • Opresko L.K.
        • Coleman J.R.
        • Feldhaus J.M.W.
        • Yeung Y.A.
        • Cochran J.R.
        • Heinzelman P.
        • Colby D.
        • Swers J.
        • Graff C.
        • Wiley H.S.
        • Wittrup K.D.
        Flow-cytometric isolation of human antibodies from a nonimmune Saccharomyces cerevisiae surface display library.
        Nat. Biotechnol. 2003; 21 (12536217): 163-170
        • Junttila I.S.
        • Creusot R.J.
        • Moraga I.
        • Bates D.L.
        • Wong M.T.
        • Alonso M.N.
        • Suhoski M.M.
        • Lupardus P.
        • Meier-Schellersheim M.
        • Engleman E.G.
        • Utz P.J.
        • Fathman C.G.
        • Paul W.E.
        • Garcia K.C.
        Redirecting cell-type specific cytokine responses with engineered interleukin-4 superkines.
        Nat. Chem. Biol. 2012; 8 (23103943): 990-998
        • Hage T.
        • Sebald W.
        • Reinemer P.
        Crystal structure of the interleukin-4/receptor α chain complex reveals a mosaic binding interface.
        Cell. 1999; 97 (10219247): 271-281
        • Liao W.
        • Lin J.-X.
        • Leonard W.J.
        Interleukin-2 at the crossroads of effector responses, tolerance, and immunotherapy.
        Immunity. 2013; 38 (23352221): 13-25
        • Wang X.
        • Rickert M.
        • Garcia K.C.
        Structure of the quaternary complex of interleukin-2 with its α, β, and γc receptors.
        Science. 2005; 310 (16293754): 1159-1163
        • Stauber D.J.
        • Debler E.W.
        • Horton P.A.
        • Smith K.A.
        • Wilson I.A.
        Crystal structure of the IL-2 signaling complex: paradigm for a heterotrimeric cytokine receptor.
        Proc. Natl. Acad. Sci. U.S.A. 2006; 103 (16477002): 2788-2793
        • Malek T.R.
        The biology of interleukin-2.
        Annu. Rev. Immunol. 2008; 26 (18062768): 453-479
        • Levin A.M.
        • Bates D.L.
        • Ring A.M.
        • Krieg C.
        • Lin J.T.
        • Su L.
        • Moraga I.
        • Raeber M.E.
        • Bowman G.R.
        • Novick P.
        • Pande V.S.
        • Fathman C.G.
        • Boyman O.
        • Garcia K.C.
        Exploiting a natural conformational switch to engineer an interleukin-2 “superkine.”.
        Nature. 2012; 484 (22446627): 529-533
        • Shanafelt A.B.
        • Lin Y.
        • Shanafelt M.C.
        • Forte C.P.
        • Dubois-Stringfellow N.
        • Carter C.
        • Gibbons J.A.
        • Cheng S.L.
        • Delaria K.A.
        • Fleischer R.
        • Greve J.M.
        • Gundel R.
        • Harris K.
        • Kelly R.
        • Koh B.
        • et al.
        A T-cell-selective interleukin 2 mutein exhibits potent antitumor activity and is well tolerated in vivo.
        Nat. Biotechnol. 2000; 18 (11062441): 1197-1202
        • Rosenberg S.A.
        Raising the bar: the curative potential of human cancer immunotherapy.
        Sci. Transl. Med. 2012; 4 (22461638): 127ps8
        • Pollack S.J.
        • Jacobs J.W.
        • Schultz P.G.
        Selective chemical catalysis by an antibody.
        Science. 1986; 234 (3787262): 1570-1573
        • Tramontano A.
        • Janda K.D.
        • Lerner R.A.
        Catalytic antibodies.
        Science. 1986; 234 (3787261): 1566-1570
        • Levin D.
        • Schneider W.M.
        • Hoffmann H.-H.
        • Yarden G.
        • Busetto A.G.
        • Manor O.
        • Sharma N.
        • Rice C.M.
        • Schreiber G.
        Multifaceted activities of type I interferon are revealed by a receptor antagonist.
        Sci. Signal. 2014; 7 (24866020): ra50
        • Moraga I.
        • Spangler J.B.
        • Mendoza J.L.
        • Gakovic M.
        • Wehrman T.S.
        • Krutzik P.
        • Garcia K.C.
        Synthekines are surrogate cytokine and growth factor agonists that compel signaling through non-natural receptor dimers.
        eLife. 2017; 6 (28498099): e22882
        • Silva D.-A.
        • Yu S.
        • Ulge U.Y.
        • Spangler J.B.
        • Jude K.M.
        • Labão-Almeida C.
        • Ali L.R.
        • Quijano-Rubio A.
        • Ruterbusch M.
        • Leung I.
        • Biary T.
        • Crowley S.J.
        • Marcos E.
        • Walkey C.D.
        • Weitzner B.D.
        • et al.
        De novo design of potent and selective mimics of IL-2 and IL-15.
        Nature. 2019; 565 (30626941): 186-191
        • Shusterman S.
        • London W.B.
        • Gillies S.D.
        • Hank J.A.
        • Voss S.D.
        • Seeger R.C.
        • Reynolds C.P.
        • Kimball J.
        • Albertini M.R.
        • Wagner B.
        • Gan J.
        • Eickhoff J.
        • DeSantes K.B.
        • Cohn S.L.
        • Hecht T.
        • et al.
        Antitumor activity of Hu14.18-IL2 in patients with relapsed/refractory neuroblastoma: a Children's Oncology Group (COG) phase II study.
        J. Clin. Oncol. 2010; 28 (20921469): 4969-4975
        • Tzeng A.
        • Kwan B.H.
        • Opel C.F.
        • Navaratna T.
        • Wittrup K.D.
        Antigen specificity can be irrelevant to immunocytokine efficacy and biodistribution.
        Proc. Natl. Acad. Sci. U.S.A. 2015; 112 (25733854): 3320-3325
        • Yu A.L.
        • Gilman A.L.
        • Ozkaynak M.F.
        • London W.B.
        • Kreissman S.G.
        • Chen H.X.
        • Smith M.
        • Anderson B.
        • Villablanca J.G.
        • Matthay K.K.
        • Shimada H.
        • Grupp S.A.
        • Seeger R.
        • Reynolds C.P.
        • Buxton A.
        • et al.
        Anti-GD2 antibody with GM-CSF, interleukin-2, and isotretinoin for neuroblastoma.
        N. Engl. J. Med. 2010; 363 (20879881): 1324-1334
        • Brochet X.
        • Lefranc M.-P.
        • Giudicelli V.
        IMGT/V-QUEST: the highly customized and integrated system for IG and TR standardized V-J and V-D-J sequence analysis.
        Nucleic Acids Res. 2008; 36 (18503082): W503-W508
        • Giudicelli V.
        • Brochet X.
        • Lefranc M.-P.
        IMGT/V-QUEST: IMGT standardized analysis of the immunoglobulin (IG) and T cell receptor (TR) nucleotide sequences.
        Cold Spring Harb. Protoc. 2011; 2011 (21632778): 695-715
        • Bankovich A.J.
        • Raunser S.
        • Juo Z.S.
        • Walz T.
        • Davis M.M.
        • Garcia K.C.
        Structural insight into pre-B cell receptor function.
        Science. 2007; 316 (17431183): 291-294
        • Chang H.C.
        • Bao Z.
        • Yao Y.
        • Tse A.G.
        • Goyarts E.C.
        • Madsen M.
        • Kawasaki E.
        • Brauer P.P.
        • Sacchettini J.C.
        • Nathenson S.G.
        A general method for facilitating heterodimeric pairing between two proteins: application to expression of alpha and beta T-cell receptor extracellular segments.
        Proc. Natl. Acad. Sci. U.S.A. 1994; 91 (7972074): 11408-11412
        • Yodoi J.
        • Teshigawara K.
        • Nikaido T.
        • Fukui K.
        • Noma T.
        • Honjo T.
        • Takigawa M.
        • Sasaki M.
        • Minato N.
        • Tsudo M.
        TCGF (IL 2)-receptor inducing factor(s). I. Regulation of IL 2 receptor on a natural killer-like cell line (YT cells).
        J. Immunol. 1985; 134 (2578514): 1623-1630
        • Klein G.
        • Giovanella B.
        • Westman A.
        • Stehlin J.S.
        • Mumford D.
        An EBV-genome-negative cell line established from an American Burkitt lymphoma; receptor characteristics. EBV infectibility and permanent conversion into EBV-positive sublines by in vitro infection.
        Intervirology. 1975; 5 (181343): 319-334
        • Boder E.T.
        • Wittrup K.D.
        Yeast surface display for screening combinatorial polypeptide libraries.
        Nat. Biotechnol. 1997; 15 (9181578): 553-557
        • Chao G.
        • Lau W.L.
        • Hackel B.J.
        • Sazinsky S.L.
        • Lippow S.M.
        • Wittrup K.D.
        Isolating and engineering human antibodies using yeast surface display.
        Nat. Protoc. 2006; 1 (17406305): 755-768
        • McCoy A.J.
        • Grosse-Kunstleve R.W.
        • Adams P.D.
        • Winn M.D.
        • Storoni L.C.
        • Read R.J.
        Phaser crystallographic software.
        J. Appl. Crystallogr. 2007; 40 (19461840): 658-674
        • Bourhis E.
        • Wang W.
        • Tam C.
        • Hwang J.
        • Zhang Y.
        • Spittler D.
        • Huang O.W.
        • Gong Y.
        • Estevez A.
        • Zilberleyb I.
        • Rouge L.
        • Chiu C.
        • Wu Y.
        • Costa M.
        • Hannoush R.N.
        • Franke Y.
        • Cochran A.G.
        Wnt antagonists bind through a short peptide to the first β-propeller domain of LRP5/6.
        Structure. 2011; 19 (21944579): 1433-1442
        • Adams P.D.
        • Grosse-Kunstleve R.W.
        • Hung L.W.
        • Ioerger T.R.
        • McCoy A.J.
        • Moriarty N.W.
        • Read R.J.
        • Sacchettini J.C.
        • Sauter N.K.
        • Terwilliger T.C.
        PHENIX: building new software for automated crystallographic structure determination.
        Acta Crystallogr. D Biol. Crystallogr. 2002; 58 (12393927): 1948-1954
        • Emsley P.
        • Cowtan K.
        Coot: model-building tools for molecular graphics.
        Acta Crystallogr. D Biol. Crystallogr. 2004; 60 (15572765): 2126-2132
        • Davis I.W.
        • Leaver-Fay A.
        • Chen V.B.
        • Block J.N.
        • Kapral G.J.
        • Wang X.
        • Murray L.W.
        • Arendall 3rd, W.B.
        • Snoeyink J.
        • Richardson J.S.
        • Richardson D.C.
        MolProbity: all-atom contacts and structure validation for proteins and nucleic acids.
        Nucleic Acids Res. 2007; 35 (17452350): W375-W383
        • Krissinel E.
        • Henrick K.
        Inference of macromolecular assemblies from crystalline state.
        J. Mol. Biol. 2007; 372 (17681537): 774-797
        • Morin A.
        • Eisenbraun B.
        • Key J.
        • Sanschagrin P.C.
        • Timony M.A.
        • Ottaviano M.
        • Sliz P.
        Collaboration gets the most out of software.
        eLife. 2013; 2 (24040512): e01456