Having generated synthetic cytokines and synthetic
Having generated synthetic cytokines and synthetic cytokine receptors, the logical next step has been to design fully synthetic cytokine/cytokine receptor systems, modifying either cytokine/cytokine receptor binding interfaces or cytokine-unrelated combinations of substances and their binding domains, fused to truncated cytokine receptors (Figure 2). For instance, a modified cytokine/cytokine receptor was designed by generating orthogonal IL-2 and IL-2Rβ by introduction of reciprocal complementary mutations . The IL-2 variant did not bind to native IL-2Rβ, but displayed native-like binding to the orthogonal IL-2Rβ variant. Hence, this system allowed the generation of modified ipi-145 responsive to orthogonal IL-2. In this way, selective pools of cells could be activated with inert synthetic ligands without activating cells expressing the wild type receptors. Examples include the in vivo expansion of CD4+ and CD8+ T cells transgenic for orthogonal IL-2Rβ by recombinant orthogonal IL-2 without inducing IL-2 related toxicities in nontransgenic T cells . Furthermore, no CD25-mediated regulatory T cell formation has been observed in vivo using the fully synthetic IL-2 system, in contrast to global administration of WT IL-2 . Thus, this system might provide a means to introduce mutant IL-2 in the context of recent CAR T cell immunotherapy in order to increase CAR T cell selective proliferation and activity without increasing numbers of regulatory T cells or other T cells by therapeutic application of orthogonal IL-2. Like neoleukins, this strategy might increase effectiveness and reduce off-target effects (cytokine storm) for CAR T cell immunotherapy; however, this remains to be explored. Another example of a fully synthetic cytokine receptor system is based on synthetic Notch receptors (synNotch) in which the extracellular domain is replaced by a nanobody directed against GFP ; however, synNotch receptor activation is unique and not prototypical among cytokine receptors since, in contrast to dimerization-induced phosphorylation of recruited transcription factors, Notch signaling involves proteolytic cleavage of the intracellular domain which itself acts as a transcription factor and induces gene expression . Engagement of Notch receptors, with their natural ligands presented as membrane-bound surface proteins on neighboring cells, has been shown to lead to cleavage of the extracellular receptor domain, followed by ectodomain shedding by ADAM proteases and subsequent proteolysis by γ-secretases . The resulting soluble intracellular domain can subsequently act as a transcriptional activator in the nucleus . As such, synthetic Notch circuits have been established using nanobody fusion receptors on receiver cells and membrane-bound GFP ligands on transmitter cells, thus resulting in juxtacrine activation of synNotch receptors . This proof-of-principle concept has been extended to activate therapeutic CD4+ and CD8+ T cells for customized therapeutic responses, including synthetic T cell differentiation, synthetic T cell cytotoxicity, local antibody production, and secretion of innate immune mediators or immunosuppressive agents in mice . To this end, the nanobody domain has been replaced by scFv antibodies directed against CD19 and HER2 . In this study, all responses however, were, dependent on cell–cell contacts between transgenic T cells expressing synNotch and the interacting (tumor) cell expressing the synNotch ligand (GFP, CD19, or HER2). By reformatting the CAR architecture by replacing the extracellular anti-CD19 scFv for an anti-GFP–nanobody GFP and an anti-transforming growth factor (TGF)-β scFv, CAR activation via receptor dimerization was demonstrated. Therefore, dimerization of GFP was achieved by introduction of an intermolecular disulfide bond, amino acid substitution D117C, and the naturally homodimeric TGF-β. Functionally, TGF-β CAR converted TGF-β from a T cell growth suppressant to a T cell growth stimulant, based on its ability to expand CD4+ and CD8+ primary T cells . In another study, the need for cell–cell contacts has been bypassed by using soluble GFP/mCherry homo- and heteromeric fusion proteins as synthetic ligands to activate nanobody-receptor fusion proteins of the IL-6 and IL-12 cytokine receptor family . Due to the modular nature of this ligand receptor system, it is suitable to dictate exact receptor stoichiometries and combinatorial assemblies of novel receptor combinations. Although kinetic analysis of signal transduction and transcriptome analysis has revealed a high degree of functional overlap between the natural cytokine and the synthetic cytokine, subtle differences might exist . However, this may be true for all other synthetic cytokines/receptor systems as well, and remains to be further investigated. For example, all constitutively active gp130 receptors differ in their signaling output from WT gp130 by only activating signal transducer and activator of transcription (STAT)1/3, but not MAPK/ERK and PI3K–Akt signaling pathways . Moreover, the mode of extracellular EPOR activation can also influence signaling outcome and strength . These and other data have shown that signaling downstream of such receptors is likely based on the geometry of the homo- or heterodimer transiently formed upon ligand binding . Consequently, signaling downstream of immunocytokines, neoleukins, fusokines, or other synthetic cytokine ligands might differ from WT cytokine downstream signaling and has to be carefully analyzed for each ligand/receptor system . Nevertheless, fully synthetic cytokine signaling systems constitute a powerful tool that could allow the precise orchestration of cellular responses.