WO2022156885A1

WO2022156885A1 - Novel chemically controlled cellular switches

Info

Publication number: WO2022156885A1
Application number: PCT/EP2021/051184
Authority: WO
Inventors: Bruno Emanuel CORREIA; Sailan SHUI; Pablo GAINZA CIRAUQUI
Original assignee: Ecole Polytechnique Federale De Lausanne (Epfl)
Priority date: 2021-01-20
Filing date: 2021-01-20
Publication date: 2022-07-28

Abstract

The invention relates to activation by inhibitor release (AIR) switches and chemically disruptable heterodimers (CDH), their use in controlling cell signaling components, and their use for treatments and therapies.

Description

Novel chemically controlled cellular switches

FIELD OF THE INVENTION

SEQUENCE LISTING

The instant application contains a Sequence Listing named PAT7489PC00_ST25.txt, which has been submitted electronically in text format and is hereby incorporated by reference in its entirety.

BACKGROUND OF THE INVENTION

Synthetic biology has enabled important developments on the understanding of fundamental aspects in biology as well as in next-generation cell-based therapies¹. In synthetic biology, many strategies have been pursued to control the timing, localization, specificity and strength of transgene expression or signaling events by equipping cells with sophisticated genetic circuits governed by small-molecule controlled protein switches. Typically, protein switches function in the basis of the action of a small molecule to control the assembly or disassembly of two protein subunits². One of the most widespread switch systems is the rapamycin-controlled chemically induced dimers (CIDs), FKBP:FRB which has been used in a wide variety of applications, including to control chimeric antigen receptor T (CAR-T) cell activities as safety switches³.

However, the large majority of chemically-controlled protein switches have limitations, particularly in translational applications, due to drug toxicity/side effects^4,5 and unfavorable pharmacokinetics⁶, or concerns that non-human protein components could raise an immunogenic response^7-9. Thus, an important requirement to enhance the breadth and scope of synthetic biology applications is to expand the universe of protein-based switches and consequently of the chemical space used to control engineered cellular activities.

Beyond naturally-sourced protein switches², several methods have recently been proposed to expand the panel of available protein switches. Specifically for CIDs, Hill and colleagues used an in vitro evolution-based approach to engineer antibodies that engage with Bcl-XL only in the presence of a small-molecule drug¹⁰, and showed that these switches where active in cellular applications. Foight and colleagues used libraries of computationally designed mutants of a previously reported de novo protein scaffold to interact with a viral protease only in its drug-bound state¹¹. These were also shown to regulate cellular activities in vivo, but the crystal structures of the designs evidenced substantial differences to the predicted binding modes. Remarkable computational design work was performed by Glasgow and colleagues, where a CID was rationally designed by transplanting the binding sites of a ligand to an existing protein dimer¹². However, the precise design of key interaction residues to mediate small molecule interactions and control CIDs remain an extremely challenging computational design problem.

All these approaches focus on chemically induced dimerization systems^{2 10-12}, yet chemical disruption systems also have important applications in synthetic biology and remain much less explored^{13 14}. Therefore, we devised a strategy to design chemically-controlled switches by repurposing protein components and small molecules involved in the inhibition of proteinprotein interactions (PPI)¹⁵. Multiple PPI inhibitors have been in clinical development over the past few years and some have been approved for clinical use^{15 16}, which make them attractive molecules for synthetic biology applications in the translational domain.

Previously, we reported the design of chemically disruptable heterodimers (CDH) from known protein: peptide-motif complexes, by transferring the peptide motif from the disordered binding partner to globular proteins and using the known PPI inhibitor as a chemical disruptor¹⁷. Specifically, our CDH was based on the Bcl-XL:BIM-BH3 complex, where the BIM-BH3 interaction motif was transplanted to the Lead Design 3 (LD3) using computational design, resulting in a Bcl-XL binder with 1000-fold times higher affinity than the wildtype disordered BIM-BH3 peptide-motif. Biochemically, we showed that Bcl-XL:LD3 complex (CDH-1) was disrupted by the drug A-l 155463¹⁸, and was used as an OFF switch for CAR-T cell activity in a dose-dependent, dynamic, and reversible manner in vivo¹¹. Importantly, both protein components were sourced from human proteome, reducing the risk of inducing an immune response against the CAR-T cells¹⁹, opening exciting opportunities to bridge the gap between computational protein design and synthetic biology for translational applications.

There is a strong need to develop safe, tunable, and small molecule responsive protein switches to control biological and therapeutical activities. SUMMARY OF THE INVENTION

This object has been achieved by providing an activation by inhibitor release (AIR) switch comprising a) a first polypeptide chain comprising a first ligand-binding domain; and b) a second polypeptide chain comprising i) a second ligand binding domain, ii) a linker, said linker is fused to the second ligand binding domain and to iii) a binder; wherein the first ligand binding domain shows binding affinity to the binder, wherein the second ligand binding domain shows binding affinity to the binder and forms an auto-inhibited binding interaction with said binder in the absence of a ligand, wherein the first and the second polypeptides are two separate polypeptides, and wherein the AIR dimerizes in the presence of a ligand which disrupts and releases the auto-inhibited binding interaction and enables the interaction between the binder and the first ligand-binding domain to induce (ON) or stop (OFF) a biological activity.

A further object of the present invention is to provide a chemically disruptable heterodimer (CDH) switch comprising a) a first polypeptide chain comprising a ligand-binding domain; and b) a second polypeptide chain comprising a binder wherein the ligand-binding domain shows binding affinity to the binder, wherein the first and the second polypeptides are two separate polypeptides, and wherein the CDH dimerizes in the absence of a ligand which disrupts the heterodimeric interaction to induce (ON) or stop (OFF) a biological activity.

A further object of the present invention is to provide a chemically disruptable heterodimer (CDH) switch comprising a) a first polypeptide chain comprising a ligand-binding domain based on the mdm2 protein; and b) a second polypeptide chain comprising a binder wherein the ligand-binding domain shows binding affinity to the binder, wherein the first and the second polypeptides are two separate polypeptides, and wherein the CDH dimerizes in the absence of a ligand which disrupts the heterodimeric interaction to induce (ON) or stop (OFF) a biological activity.

A further object of the present invention is to provide a method for designing an activation by inhibitor release sensor (AIR) switch comprising a) selecting a pair “liganddigand-binding domain”; b) designing a binder to selectively interact with the ligand-binding domain of step a), wherein the ligand-binding domain is the second ligand-binding domain; c) designing or rendering the first ligand-binding domain insensitive, or less sensitive, to the ligand by modifying active site residues involved in the ligand binding; and optionally d) operably linking the second ligand binding domain to the binder, via a linker, wherein the second ligand binding domain forms an auto-inhibited binding interaction with said binder in the absence of a ligand. operably linking the second ligand binding domain to the binder, via a linker, wherein the second ligand binding domain forms an auto-inhibited binding interaction with said binder in the absence of a ligand.

A further object of the present invention is to provide an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 1, SEQ ID No. 2, SEQ ID No. 3, SEQ ID No. 4, SEQ ID No. 5, SEQ ID No. 6, SEQ ID No. 7, SEQ ID No. 8, SEQ ID No. 9 and SEQ ID No. 10.

A further object of the present invention is to provide an isolated nucleic acid encoding an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 1, SEQ ID No. 2, SEQ ID No. 3, SEQ ID No. 4, SEQ ID No. 5, SEQ ID No. 6, SEQ ID No. 7, SEQ ID No. 8, SEQ ID No. 9 and SEQ ID No. 10. A further object of the present invention is to provide an expression vector comprising an isolated nucleic acid of the invention, operably linked to a transcriptional control element.

A further object of the present invention is to provide a host cell comprising an isolated nucleic acid or an expression vector of the invention.

A further object of the present invention is to provide a host cell expressing the activation by inhibitor release (AIR) or the CDH of the invention.

A further object of the present invention is to provide a method of treating a disease in a subject, the method comprising: i) genetically modifying at least one host cell obtained from the individual with an expression vector comprising one or more nucleotide sequences encoding an activation by inhibitor release (AIR) of the invention, and where the genetic modification is carried out ex vivo; ii) introducing the genetically modified host cell into the subject; and iii) administering to the subject an effective amount of a ligand, wherein the ligand induces dimerization of the heterodimeric, conditionally active AIR, wherein said dimerization induces (ON) a biological activity of the genetically modified host cells.

A further object of the present invention is to provide a composition comprising an AIR switch, an amino acid sequence, an isolated nucleic acid, an expression vector, a host cell, and/or a CDH of anyone of the invention.

A further object of the present invention is to provide a pharmaceutical composition comprising a therapeutically effective amount of an AIR switch of anyone of claims 1 to 18, an amino acid sequence, an isolated nucleic acid, an expression vector, a host cell, and/or a CDH of the invention, and a pharmaceutically acceptable carrier or diluent.

A further object of the present invention is to provide a kit comprising a composition or a pharmaceutical composition of the invention.

A further object of the present invention is to provide a pharmaceutical composition for use in the treatment and/or prevention of a disease selected from the group comprising a cancer, an inflammatory disease, a genetic disorder, an infectious disease, or a degenerative disease.

DESCRIPTION OF THE FIGURES

Figure 1: Design of CDH3 and biophysical characterization of CDH2 and CDH3. a) Structural representation of CDH-2 composed of Bcl2 (surface) and LD3 (cartoon) with the interfacial segment colored, the hotspot residues transplanted from the BIM BH3 motif are shown in sticks, b) SPR measurements of CDH-2 binding affinity. The dissociation constant determined for the interaction of Bcl2 with LD3 is 796 pM. c) SPR drug competition assay determined the apparent IC50 of CDH-2 with Drug-2 around 67.5 nM. d) Structural representation of CDH-3 composed of the mdm2 (surface):LD6 (cartoon) complex with the interfacial segment colored, the hotspot residues transplanted from the p53 are motif shown in sticks, e) SPR measurements of CDH-3 binding affinity. The dissociation constant determined for the interaction of mdm2 with LD6 is 4.19 nM. f) SPR drug competition assay determined the apparent IC50 of CDH-3 with Drug-3 around 106 nM. g) The crystal complex structure of CDH-3 consisting of the LD6 protein (tube) and mdm2 (surface) was in a close agreement with the computational model of LD6 (tube) with complex with mdm2(not shown), interface RMSD of 0.37 A. h) Transplanted hotspot residues in LD6 in sticks aligned with p53 peptide residues in a RMSD of 0.24 A.

Figure 2: Applications of CDHs for intracellular and extracellular gene expression control. a) Schematic representation of the CDHs utilized to control transcription regulation with the GAL4/UAS system (CDH-TFs). b) CDH-TFs fold-change activity between drug treated (1 pM ) versus untreated (DMSO), showing the quantification of SEAP expression after 24 h drug treatment, c) Drug dose-dependent responses of CDH-TFs quantified by SEAP expression after 24 h drug treatment(CDH-l, circle, CDH-2, triangle, CDh-3, square), d) Schematic representation of the CDHs utilized to regulate surface signaling receptors (CDH-GEMS) in engineered cells, e) CDH-GEMS fold-change activity between drug (1 pM ) versus no drug treatment (DMSO), showing the quantification of SEAP expression after 24 h(CDH-l, circle, CDH-2, triangle, CDh-3, square). 1) Drug dose-dependent responses of CDH-TFs quantified by SEAP expression after 24 h. g) Summary of the LD3 mutants including computationally predicted decreases in affinity and the experimental measurements using SPR assays. Values shown are the predicted interaction energy of LD3 mutants (AAG in Rosetta energy units), the binding affinities with BclxL and Bcl2 (measured by SPR) and IC50S of CDH dissociation (measured by SPR drug competition assay) also for BclxL and Bcl2 with Drug-1 and Drug-2, respectively, h) Drug dose-dependent responses in engineered cells determined for the CDH- GEMS with LD3 and LD3_v3. The drug receptor utilized was BclxL (CDH-1). i) Drug dosedependent responses in engineered cells determined for the CDH-GEMS with LD3 and LD3_v3. The drug receptor utilized was Bcl2 (CDH-2). b, e, c, f, h, i) Each data point represents the mean ± s.d. of three replicates and the ICsos were calculated using four- parameter nonlinear regression ± s.d.

Figure 3: prediction of drug-insensitive receptor and development of AiR-GEMS. a) Computational design approach and architecture of the AIR-GEMS system. Starting from the CDH components (drug receptor(beige) and protein binder (green) we used a multistate design approach to search for drug receptor sequences that would retain binding to the protein binder (positive design) and become resistant to the drug (negative design). We then assembled the AIR-GEMS, were the drug triggers the expression of SEAP by activating the JAK-STAT pathway, b) Structural representation of the residues important for Drug-1 binding with BclxL. BclxL drug binding pocket (white surface) where the mutations R102E (green stick) and Fl 051 (blue stick) were performed to obtain the variant iBclxL_v3. Drug-1 is shown in sticks representation and colored in brown. Four other designable residues are highlighted on the surface, E98 in red, T109 in cyan, S145 in orange and A149 in yellow, c) Apparent IC50S of Drug-1 dissociate BclxL:LD3 and iBclxL_v3 :LD3 determined by SPR drug competition assay, d) Structural representation of the residues important for Drug-2 binding with Bcl2. Bcl2 drug binding pocket (white surface) where the mutations A100V (Red stick), D103N (Green stick) and Y201H (Orange stick) were performed to obtain the variant iBcl2_v4. Drug-2 is shown in sticks representation and colored in green. Two other designable resides are highlighted on the surface, V148 in blue, VI 56 in yellow, e) Apparent IC50S of Drug-2 dissociate Bcl2:LD3 and iBcl2_v4:LD3 determined by SPR drug competition assay, f) AIR-GEMS fold-change activity between drug (1 pM) versus no drug treatment (DMSO), showing the quantification of SEAP expression after 24 h. g) Drug dose-dependent responses in engineered cells expressing the AIR-GEMS(AIR-1, circle, Air-2, square). Each data point represents the mean ± s.d. of three replicates and the ECsos were calculated using four-parameter nonlinear regression ± s.d. Figure 4: Orthogonal chemical-specific switches enable implementation of multi-input multi-output control mode in mammalian cells. a) Scheme of multi-inputs multi-outputs (MIMO) designer cells where two drugs control two different outputs. b,c) Cells were co-transfected with AIR-1/2-GEMS and CDH-3-TF regulate SEAP and Luciferase expression, respectively. Drug concentrations ranged from 0.001 nM to 1 pM and different combinations were added depending on the protein components, Drug-1 + Drug-3 (b) and Drug-2 + Drug-3 (c). d) Scheme of multi-inputs single-output (MISO) designer cells where two drugs control one output. e,f) Quantification of SEAP activity controlled by CDH-l-TF and CDH-3-TF circuits (e) or CDH-l-GEMS and CDH-3-GEMS (f). Drug-3 concentrations ranged from 0.001 nM to 1 pM and Drug-1 treatments were performed in the presence of 1 pM Drug-3, g) Scheme of single-input multioutputs (SIMO) designer cells where one drug controls two outputs. h,i) Quantification of SEAP and Luciferase activities under Drug-1 (h) or Drug-2 (i). AIR-1/2-GEMS coupled with SEAP expression circuits were co-transfected with CDH-1/2-TF circuits which control the Luciferase production. Respective drug concentrations ranged from 0.001 nM to 1 pM. All values presented are mean ± s.d. of three replicates and curves were fitted by four-parameters nonlinear regression.

DESCRIPTION OF THE INVENTIONS

Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. The publications and applications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Nothing herein is to be construed as an admission that the present invention is not entitled to antedate such publication by virtue of prior invention. In addition, the materials, methods, and examples are illustrative only and are not intended to be limiting.

In the case of conflict, the present specification, including definitions, will control. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as is commonly understood by one of skill in art to which the subject matter herein belongs. As used herein, the following definitions are supplied in order to facilitate the understanding of the present invention. The term "comprise/comprising" is generally used in the sense of "include/including", that is to say permitting the presence of one or more features or components.

As used in the specification and claims, the singular form "a", "an" and "the" include plural references unless the context clearly dictates otherwise.

The term "amino acid" includes all of the naturally occurring amino acids as well as modified amino acids.

As used herein, "at least one" means "one or more", "two or more", "three or more", etc.

The term “about,” particularly in reference to a given quantity, is meant to encompass deviations of plus or minus ten (10) percent

“Homology” refers to the percent identity between two polynucleotide or two polypeptide moieties. Two nucleic acid, or two polypeptide sequences are “substantially homologous” to each other when the sequences exhibit at least about 50% sequence identity, preferably at least about 75% sequence identity, more preferably at least about 80% or at least about 85% sequence identity, more preferably at least about 90% sequence identity, and most preferably at least about 95%-98% sequence identity over a defined length of the molecules. As used herein, substantially homologous also refers to sequences showing complete identity to the specified sequence. Alternatively, homology can be determined by readily available computer programs or by hybridization of polynucleotides under conditions which form stable duplexes between homologous regions, followed by digestion with single stranded specific nuclease(s), and size determination of the digested fragments. DNA sequences that are substantially homologous can be identified in a Southern hybridization experiment under, for example, stringent conditions, as defined for that particular system. Defining appropriate hybridization conditions is within the skill of the art.

In general, “identity” refers to an exact nucleotide to nucleotide or amino acid to amino acid correspondence of two polynucleotides or polypeptide sequences, respectively. Percent identity can be determined by a direct comparison of the sequence information between two molecules by aligning the sequences, counting the exact number of matches between the two aligned sequences, dividing by the length of the shorter sequence, and multiplying the result by 100. In some aspects, a nucleotide or amino acid sequence of the invention, or a portion thereof, is at least 80%, namely, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to a corresponding nucleotide or amino acid sequence (SEQ ID NO identifier).

The present invention discloses a multidomain architecture referred to as activation by inhibitor release (AIR) switch to mediate assembly of cellular signaling components, which comprises, a) a first polypeptide chain comprising a first ligand-binding domain; and b) a second polypeptide chain comprising i) a second ligand binding domain, ii) a linker, said linker is fused to the second ligand binding domain and to iii) a binder; wherein the first ligand binding domain shows binding affinity to the binder, wherein the second ligand binding domain shows binding affinity to the binder and forms an auto-inhibited binding interaction with said binder in the absence of a ligand, wherein the first and the second polypeptides are two separate polypeptides, and wherein the AIR switch dimerizes in the presence of a ligand which disrupts and releases the autoinhibited binding interaction, and enables the interaction between the binder and the first ligand-binding domain to induce (ON) or stop (OFF) a biological activity.

An example of a subject AIR) switch, is represented Schematically in FIG. 3.

As used herein, the terms "peptide", "protein", "polypeptide", "polypeptide chain", "polypeptidic" and "peptidic" are used interchangeably to designate a series of amino acid residues connected to the other by peptide bonds between the alpha-amino and carboxy groups of adjacent residues.

In some aspects, the polypeptide chain, also named polypeptide scaffold, is a monomer comprising an amino acid sequence length between about 50 and 200 amino acids, preferably between about 70 and 180 amino acids, most preferably between about 80 and 160 amino acids. The polypeptide chain may comprise one or more extracellular domain and/or one or more transmembrane domain as described infra.

In one aspect, a first polypeptide chain comprises a first ligand binding domain.

Preferably, the first polypeptide chain is operably linked to the first ligand binding domain. In the same aspect, a second polypeptide chain comprises a second ligand binding domain and a linker, said linker being fused to the second ligand binding domain and to a binder. Preferably, the second polypeptide chain is operably linked to the binder which is linked to the ligand binding domain through a linker.

The present invention also encompasses one or more variants of a polypeptide, a peptide, a protein or an amino acid sequence disclosed herein. The term "variant", as used herein, means a molecule comprising substitution, deletion or addition of one or a few to a plurality of amino acids, and includes particularly conservatively substituted molecules, provided that the variant substantially retains the same function as the original sequence.

As used herein, the term "ligand" refers to any moiety that is capable of covalently or otherwise chemically binding, to a domain (i.e. the binding domain) with a specific affinity. The ligand in the context of the invention may be synthetic or natural and is selected among the non-limiting group comprising a small molecule (e.g. a drug), an antibody, an antigen binding fragment thereof, a protein-based therapeutic, and a hormone. The ligand may be an activator, an inhibitor or a disruptor.

Non-limiting examples of ligands are selected among the group comprising a Bcl-XL inhibitor, a Bcl2 inhibitor and a mdm2 inhibitor, or a combination of one or more thereof (e.g. a Bcl-XL/BCL-2 inhibitor).

In one aspect, the ligand is selected from the group comprising navitoclax, A- 1331852, A-l 155463, venetoclax, ABT-199 (GDC-0199), obatoclax mesylate (GX15-070), HA14-1, ABT-737, TW- 37, AT101, sabutoclax, gambogic acid, ARRY 520 trifluoroacetate, iMAC2, maritoclax, methylprednisolone, MIMI, ML 311, glossypol, BH3I-1, and 2- methoxy-antimycin A3.

In a preferred aspect, the ligand is selected from the group comprising A-l 155463, A- 1331852, and navitoclax (ABT-263), when the first and/or second ligand-binding domain is/are the Bcl-XL polypeptide or a derivative thereof,

In a preferred aspect, the ligand is selected from the group comprising venetoclax, obatoclax, and navitoclax (ABT-263), when the first and/or second ligand-binding domain is/are the Bcl2 polypeptide or a derivative thereof. In a preferred aspect, the ligand is selected from the group comprising NVP-CGM097, RG7388, and NVP-HDM201, when the first and/or second ligand-binding domain is/are the MDM2 polypeptide or a derivative thereof.

The term "ligand binding domain" as used herein refers to the domain of a ligand binding receptor of the present disclosure that is operably linked to a polypeptidic scaffold chain.

In one aspect of the invention, the activation by inhibitor release (AIR) of the invention, the first and/or second ligand-binding domain are independently selected from the group comprising Bcl-XL polypeptide, Bcl2 polypeptide, MDM2 polypeptide, and a derivative of one or more thereof.

Preferably, the first and/or second ligand-binding domain are independently selected from the group comprising an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 4 (iBcl-XL_v3), SEQ ID No. 5 (iBcl2_v4), SEQ ID No. 6 (BclxL), SEQ ID No. 7 (Bcl2) and SEQ ID No. 9 (MDM2).

In various aspects, a linker (also named linker region) is used to provide more flexibility and accessibility for the ligand binding domain, e.g. the second ligand binding domain. A linker may be a peptide comprising one or more Serine and/or Glycine amino acids and comprise up to 300 amino acids, preferably 8 to 100 amino acids, most preferably 10 to 50 amino acids, and more preferably 10 to 25 amino acids.

A linker region may be derived from all or part of naturally occurring molecules, such as from all or part of the extracellular region of CD8, CD4 or CD28, or from all or part of an antibody constant region.

Alternatively, the linker region may be a synthetic sequence that corresponds to a naturally occurring linker region sequence or may be an entirely synthetic linker domain sequence (such as e.g. a sequence comprising the sequence as set forth in SEQ ID No. 11 : GGGGS GGGGS GGGGS, or a functional fragment sequence thereof).

Additional non-limiting examples of linker regions which may be used in accordance to the invention include a part of human CD8 a chain, partial extracellular domain of CD28, FcyRllla receptor, IgG, IgM, IgA, IgD, IgE, an Ig hinge, a derivative or functional fragment of one or more thereof.

The AIR switch and/or CDH of the invention may further comprise one or more transmembrane domain of the present disclosure comprises a transmembrane domain. The term “transmembrane domain” means a domain of a polypeptide that includes at least one contiguous amino acid sequence that traverses a lipid bilayer when present in the corresponding endogenous polypeptide when expressed in a mammalian cell. For example, a transmembrane domain can include one, two, three, four, five, six, seven, eight, nine, or ten contiguous amino acid sequences that each traverse a lipid bilayer when present in the corresponding endogenous polypeptide when expressed in a mammalian cell. As is known in the art, a transmembrane domain can, e.g., include at least one (e.g., two, three, four, five, six, seven, eight, nine, or ten) contiguous amino acid sequence (that traverses a lipid bilayer when present in the corresponding endogenous polypeptide when expressed in a mammalian cell) that has a-helical secondary structure in the lipid bilayer. In some embodiments, a transmembrane domain can include two or more contiguous amino acid sequences (that each traverse a lipid bilayer when present in the corresponding endogenous polypeptide when expressed in a mammalian cell) that form a P-barrel secondary structure in the lipid bilayer.

The transmembrane domain may be derived from a natural polypeptide, or may be artificially designed. The transmembrane domain derived from a natural polypeptide can be obtained from any membrane-binding or transmembrane protein. For example, a transmembrane domain of a T cell receptor a or P chain, a C, chain, CD28, CD3z, CD45, CD4, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD137, ICOS, CD154, or a GITR can be used. The artificially designed transmembrane domain is a polypeptide mainly comprising hydrophobic residues such as leucine and valine. For example, a triplet of phenylalanine, tryptophan and valine can be found at each end of the synthetic transmembrane domain. Optionally, a short oligopeptide linker or a polypeptide linker, for example, a linker having a length of 2 to 10 amino acids can be arranged between the transmembrane domain and the intracellular segment as described herein. Particularly, a linker sequence having a glycine-serine continuous sequence can be used.

Additionally, or alternatively, a spacer domain can be arranged between the extracellular domain and the transmembrane domain, or between the intracellular segment and the transmembrane domain. The spacer domain means any oligopeptide or polypeptide that serves to link the transmembrane domain with the extracellular domain and/or the transmembrane domain with the intracellular segment. The spacer domain comprises up to 300 amino acids, for example about 10 to 100 amino acids, or about 25 to 50 amino acids.

The AIR switch and/or CDH of the invention may further comprise one or more intracellular signaling domain(s). The "intracellular signaling domain" means any oligopeptide or polypeptide domain known to function to transmit a signal causing activation or inhibition of a biological process in a cell, for example, activation of an immune cell such as a T cell or a NK cell.

In some aspects of the invention, the one or more intracellular signaling domain induces downstream signaling via a JAK/STAT (Janus kinase/signal transducer and activator of transcription) signaling pathway, a MAPK (mitogen-activated protein kinase) signaling pathway, a PLCG (phospholipase C gamma) signaling pathway, or a PBK/Akt (phosphati-dylinositol 3-kinase/protein kinase B) signaling pathway.

Non-limiting examples of an intracellular signaling domains are selected from the group comprising CD247/ CD3z, CD137/4-1BB, Zap70, STAT3, and STAT5.

Other examples of intracellular signaling domains include cytoplasmic regions from a TCR complex and/or a costimulatory molecule, and any variant having the same function as those sequences.

Further examples include cytoplasmic signaling domains listed in Table 2 of Sadelain et al (Sadelain, M., Brentjens, R., & Riviere, I. (2009). The promise and potential pitfalls of chimeric antigen receptors. Current opinion in immunology, 21(2), 215-223), which is incorporated herein by reference.

The AIR switch and/or CDH of the invention further comprise a binder. A "binder" means any oligopeptide or polypeptide domain known to, or designed to, interact with a ligand binding domain. The binder can further incorporate one or more motifs resulting in a stable globular binder with higher affinity than the wildtype motif. Preferably, the binder shows up to about lOOx, about 200x, about 500x, about 800x or about lOOOx times higher affinity than the wildtype motif it derives from. Preferably, the binder with higher affinity comprises an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 1 (LD3_vl), SEQ ID No. 2 (LD3_v2), SEQ ID No. 3 (LD3_v3), SEQ ID No. 8 (LD3), SEQ ID No. 10 (LD6), and a variant, derivative or a functional fragment of one of more of these sequences.

Alternatively, lower-affinity binders are also envisioned in the present invention and were selected through, for example, alanine scanning on a designed binder. Preferably, a lower-affinity binder shows up to about lOOx, about 200x, about 500x, about 800x or about lOOOx times lower affinity than the wildtype motif it derives from.

Preferably, the binder with lower affinity comprises an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 8 (LD3), SEQ ID No. 10 (LD6), and a variant, derivative or a functional fragment of one of more of these sequences.

The present invention further contemplates isolated nucleic acids. The term "nucleic acid sequence" and " nucleic acids" as used herein refers to a sequence of nucleoside or nucleotide monomers consisting of naturally occurring bases, sugars and intersugar (backbone) linkages. The term also includes modified or substituted sequences comprising non-naturally occurring monomers or portions thereof. The nucleic acid sequences of the present application may be, deoxyribonucleic acid sequences (DNA) or ribonucleic acid sequences (RNA), either double stranded or single stranded and represents the sense or antisense strand, and may include naturally occurring bases including adenine, guanine, cytosine, thymidine and uracil. The sequences may also contain modified bases. Examples of such modified bases include aza and deaza adenine, guanine, cytosine, thymidine and uracil; and xanthine and hypoxanthine.

In one aspect, the isolated nucleic acid of the invention encodes an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 1 (LD3_vl), SEQ ID No. 2 (LD3_v2), SEQ ID No. 3 (LD3_v3), SEQ ID No. 4 (iBcl-XL_v3), SEQ ID No. 5 (iBcl2_v4), SEQ ID No. 6 (Bcl-XL), SEQ ID No. 7 (Bcl2), SEQ ID No. 8 (LD3), and SEQ ID No. 9 (MDM2), and SEQ ID No. 10 (LD6), a variant, derivative or a functional fragment of one of more of these sequences.

The present invention further contemplates an expression vector comprising an isolated nucleic acid as disclosed herein, operably linked to a transcriptional control element. The terms "expression vector", “gene delivery vector” and "gene therapy vector" refer to any vector that is effective to incorporate and express one or more nucleic acid(s) of the invention, in a cell, preferably under the regulation of a promoter. A cloning or expression vector may comprise additional elements, for example, regulatory and/or post-transcriptional regulatory elements in addition to a promoter.

In addition to control sequences that govern transcription and translation, vectors and expression vectors may contain nucleic acid sequences that serve other functions as well. An expression vector may comprise additional elements, for example, the expression vector may have two replication systems, thus allowing it to be maintained in two organisms, for example in human cells for expression and in a prokaryotic host for cloning and amplification. The expression vector may have additional sequence such as 6x-histidine, c- Myc, and FLAG tags which are incorporated into the expressed polypeptides. In various embodiments, the vectors are plasmid, autonomously replicating sequences, and transposable elements.

Suitable vectors include derivatives of SV40 and known bacterial plasmids, e. g., E. coli plasmids col El, pCRl, pBR322, pMB9 and their derivatives, plasmids such as RP4; phage DNAs, e. g., the numerous derivatives of phage X, e. g., NM989, and other phage DNA, e. g., Ml 3 and filamentous single stranded phage DNA; yeast plasmids such as the 2p plasmid or derivatives thereof; vectors useful in eukaryotic cells, such as vectors useful in insect or mammalian cells; vectors derived from combinations of plasmids and phage DNAs, such as plasmids that have been modified to employ phage DNA or other expression control sequences; and the like.

Various viral vectors are used for delivering nucleic acid to cells in vitro or in vivo. Nonlimiting examples are vectors based on Herpes Viruses, Pox- viruses, Adeno-associated virus, Lenti virus, and others.

The present invention further contemplates host cells.

In one aspect the invention contemplates a host cell comprising an isolated nucleic acid or any vector or expression vector disclosed herein. Further contemplated is a host cell expressing the activation by inhibitor release (AIR) of the invention.

Preferably, the cell is a mammalian cell and is selected from the group comprising a cytotoxic cell, a T cell, a stem cell, a progenitor cell, and a cell derived from a stem cell or a progenitor cell. The T cell may be a T-helper cell, a cytotoxic T-cell, a T-regulatory cell (Treg), or a gamma-delta T cell. The cytotoxic cell may be a cytotoxic T cell or a natural killer (NK) cell. The host cell may be activated ex vivo and/or expanded ex vivo. The host cell may be an allogeneic cell. The host cell may be an autologous cell. The host cell may be isolated from a subject having a disease. In various aspects, the subject is human.

Also provided is a method for producing any of the host cells described herein. The method comprises genetically modifying the cell with any nucleic acid molecule or any vector or expression vector described herein. The genetic modification may be conducted ex vivo. The method may further comprise activation and/or expansion of the cell ex vivo.

The polypeptides disclosed herein, or nucleic acids encoding such, may be introduced into the host cells using transfection and/or transduction techniques known in the art. The nucleic acid may be integrated into the host cell DNA or may be maintained extrachromosomally. The nucleic acid may be maintained transiently or may be a stable introduction. Transfection may be accomplished by a variety of means known in the art including but not limited to calcium phosphate-DNA co-precipitation, DEAE- dextran-mediated transfection, polybrene-mediated transfection, electroporation, microinjection, liposome fusion, lipofection, protoplast fusion, retroviral infection, and biolistics. Transduction refers to the delivery of a gene(s) using a viral or retroviral vector by means of viral infection rather than by transfection. In certain embodiments, retroviral vectors are transduced by packaging the vectors into virions prior to contact with a cell. For example, a nucleic acid encoding a transmembrane polypeptide carried by a retroviral vector can be transduced into a cell through infection and pro virus integration.

In certain aspects, the nucleic acid or vector, e.g. viral vector, is transferred via ex vivo transformation. Methods for transfecting cells and tissues removed from an organism in an ex vivo setting are known to those of skill in the art. Thus, it is contemplated that cells or tissues may be removed and transfected ex vivo using the polynucleotides presented herein. In particular aspects, the transplanted cells or tissues may be placed into an organism. Thus, it is well within the knowledge of one skilled in the art to isolate antigen-presenting cells ( e.g., T-cells or NK cells) from an animal ( e.g., human), transfect the cells with the expression vector and then administer the transfected or transformed cells back to the animal.

In certain embodiments, the nucleic acid or viral vector is transferred via injection. In certain embodiments, a polynucleotide is introduced into an organelle, a cell, a tissue or an organism via electroporation. In certain embodiments, a polynucleotide is delivered into a cell using DEAE-dextran followed by polyethylene glycol. In certain embodiments, the polynucleotides encode any of the first and second transmembrane polypeptides described herein, and are inserted into a vector or vectors. The vector is a vehicle into which a polynucleotide encoding a protein may be covalently inserted so as to bring about the expression of that protein and/or the cloning of the polynucleotide. Expression vectors have the ability to incorporate and express heterologous or modified nucleic acid sequences coding for at least part of a gene product capable of being transcribed in a cell. In most cases, RNA molecules are then translated into a protein.

The cells as described herein are useful in treating and/or preventing a disease selected from the group comprising a cancer, an inflammatory disease, a genetic disorder, an infectious disease, or a degenerative disease as well as combination of one or more of these diseases, in a subject with the disorder, for example, by ex vivo therapies.

The present invention further contemplates a method for designing an activation by inhibitor release switch (AIR) switch comprising: a) selecting a pair “liganddigand-binding domain”; b) designing a binder to selectively interact with the ligand-binding domain of step a), wherein the ligand-binding domain is the second ligand-binding domain; c) designing or rendering the first ligand-binding domain insensitive, or less sensitive, to the ligand by modifying active site residues involved in the ligand binding; and optionally d) operably linking the second ligand binding domain to the binder, via a linker, wherein the second ligand binding domain forms an auto-inhibited binding interaction with said binder in the absence of a ligand. The present invention further contemplates a method of treating a disease in a subject, the method comprising: i) genetically modifying at least one host cell obtained from the individual with an expression vector comprising one or more nucleotide sequences encoding an activation by inhibitor release (AIR) of the invention, and where the genetic modification is carried out ex vivo; ii) introducing the genetically modified host cell into the subject; and iii) administering to the subject an effective amount of a ligand, wherein the ligand induces dimerization of the heterodimeric, conditionally active AIR, wherein said dimerization induces (ON) a biological activity of the genetically modified host cells.

As used herein, the term “treatment” or “treating” means any administration of a composition, pharmaceutical composition, AIR, CDH, vector, nucleic acids, compound, cell etc... of the disclosure to a subject for the purpose of: (i) inhibiting the disease, that is, arresting the development of clinical symptoms; and/or (ii) relieving the disease, that is, causing the regression of clinical symptoms.

As used herein, the term “prevention” or “preventing” means any administration of a composition, pharmaceutical composition, therapeutic agent, compound, etc... of the disclosure to a subject for the purpose of preventing the disease, that is, causing the clinical symptoms of the disease not to develop.

In the context of the present invention, the disease is selected from the group comprising a cancer, an inflammatory disease, a genetic disorder, an infectious disease, or a degenerative disease as well as combination of one or more of these diseases.

The Inventors also designed a series of chemically disruptable heterodimers with tailored affinity ranges from low picomolar to mid nanomolar and tested these their effect in regulating cellular processes. The present invention provides, in a further aspect, a chemically disruptable heterodimer (CDH) switch comprising a) a first polypeptide chain comprising a ligand-binding domain based on the mdm2 protein; and b) a second polypeptide chain comprising a binder wherein the ligand-binding domain shows binding affinity to the binder, wherein the first and the second polypeptides are two separate polypeptides, and wherein the CDH dimerizes in the absence of a ligand which disrupts the heterodimeric interaction to induce (ON) or stop (OFF) a biological activity.

The first polypeptide chain may further comprise an extracellular domain (ECI) and/or the second polypeptide chain comprises an extracellular domain (EC2).

In an aspect of the invention, the CDH further comprises c) one or more transmembrane domain and d) one or more intracellular signaling domain(s).

Preferably, the binding of a ligand to the ligand binding domain can lead to conformational change such as dissociation or association of the CDH, which modulates activity (negatively or positively) of one or more intracellular signaling domain.

Non-limiting examples of the ligand of the invention, whether an inhibitor (also referred to "a disruptor") or an activator, are selected from the group comprising a small molecule, an antibody, protein-based therapeutic, an antigen binding fragment thereof, and a hormone.

Preferably, the first ligand-binding domain comprises an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in SEQ ID No. 6, SEQ ID No. 7, and SEQ ID No. 9.

As illustrated in the examples, the inventors designed a novel CDH based on the Bcl2:LD3 interaction (CDH2) disrupted by the clinically-approved drug Venetoclax (Drug-2)²⁰'²¹. A novel CDH3 was designed based on the interaction of MDM2 protein and the p53 binding motif (disrupted by Drug-3²²). These three CDHs are applied to regulate cellular responses, including transcriptional gene expression by tailoring promoter-specific transactivating Gal4/UAS system²³ and by controlling synthetic cell surface receptors based on the GEMS platform²⁴ which modulate the JAK/STAT signaling pathway.

In an aspect of the invention, one of the first or second polypeptide chain of the CDH and/or AIR switch of the invention further comprises, or is operably linked to, or is fused to, one or more DNA binding domain. In this case, the remaining first or second polypeptide chain further comprises, or is/ operably linked to, or is fused to, one or more transcription factor. The one or more DNA binding domain is preferably selected from the group comprising Gal4, TetR, dCas9, ZFN, and TALEN, or a combination of one or more thereof.

The one or more transcription factor is selected from the group comprising VP 16, VP64, VPR and p65, or a combination of one or more thereof.

Alternatively, both polypeptide chain of the CDH and/or AIR switch of the invention further comprises, or are operably linked to, or are fused to, one or more transmembrane domain and/or one or more intracellular signaling domain.

The one or more transmembrane domain is preferably selected from the group comprising a T cell receptor a or p chain, a chain, CD28, CD3z, CD45, CD4, CD5, CD8, CD9, CD 16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD137, ICOS, CD154, or a GITR can be used.

The one or more intracellular signaling domain is preferable selected from the group comprising CD247/ CD3z, CD137/4-1BB, Zap70, STAT3, and STAT5.

In order to explore the effect of CDH affinity on cellular activity, and drug disruptor dosage, the Inventors designed a suite of CDHs with tailored affinity ranges from low picomolar to mid nanomolar and tested these their effect in regulating cellular processes. Next, the Inventors devised a novel strategy to create ON switches by repurposing CDHs. These new switches, referred to as activation by inhibitor release switches (AIRs), consist of a singlechain, fused CDH, coupled with a rationally-designed drug insensitive receptor protein (Fig. la). The exposure of the drug disruptor releases the designed binder from the fused CDH and enables dimerization with the drug insensitive receptor. Notably, as the array of the clinically-validated PPI inhibitors grows, this AIR strategy could be applied to develop ON- switchable controls for synthetic cells. Finally, and to demonstrate the applicability of the designed switches, the Inventors engineered mammalian cells equipped with multi input/output control modes which expanded the protein switches into more sophisticated controls and broaden the potential applications. Altogether, this approach provides a blueprint to design novel cellular switches for use across basic and translational synthetic biology applications. Also contemplated in the present invention are compositions comprising an AIR switch, an amino acid sequence, an isolated nucleic acid, a vector, an expression vector, a host cell, and/or a CDH of the invention as described herein.

Also contemplated in the present invention are pharmaceutical compositions comprising a therapeutically effective amount of an AIR switch, an amino acid sequence, an isolated nucleic acid, a vector, an expression vector, a host cell, and/or a CDH of the invention as described herein and a pharmaceutically acceptable carrier or diluent.

In a preferred aspect, the pharmaceutical composition of the invention is for use in the treatment and/or prevention of a disease that is selected from the group comprising a cancer, an inflammatory disease, a genetic disorder, an infectious disease, or a degenerative disease as well as combination of one or more of these diseases.

The term "therapeutically effective amount" as used herein means an amount of an AIR, CDH, vector, nucleic acids, compound, or cell high enough to significantly positively modify the symptoms and/or condition to be treated, but low enough to avoid serious side effects (at a reasonable risk/benefit ratio), within the scope of sound medical judgment.

The therapeutically effective amount of an AIR, CDH, vector, nucleic acids, compound, or cell is selected in accordance with a variety of factors including type, species, age, weight, sex and medical condition of the patient; the severity of the condition to be treated; the route of administration; the renal and hepatic function of the patient. A physician of ordinary skill in the art can readily determine and prescribe the effective amount of the drug required to prevent, counter or arrest the progress of the disease.

“Pharmaceutically acceptable carrier or diluent” means a carrier or diluent that is useful in preparing a pharmaceutical composition that is generally safe, non-toxic, and desirable, and includes carriers or diluents that are acceptable for human pharmaceutical use.

Such pharmaceutically acceptable carriers can be sterile liquids, such as water and oils, including those of petroleum, animal, vegetable or synthetic origin, such as peanut oil, soybean oil, mineral oil, sesame oil and the like. Water is a preferred carrier when the pharmaceutical composition is administered intravenously. Saline solutions and aqueous dextrose and glycerol solutions can also be employed as liquid carriers, particularly for injectable solutions. Pharmaceutically acceptable excipients include starch, glucose, lactose, sucrose, sodium stearate, glycerol monostearate, talc, sodium chloride, dried skim milk, glycerol, propylene glycol, water, ethanol and the like.

The pharmaceutical compositions may further contain one or more pharmaceutically acceptable salts such as, for example, a mineral acid salt such as a hydrochloride, a hydrobromide, a phosphate, a sulfate, etc.; and the salts of organic acids such as acetates, propionates, malonates, benzoates, etc. Additionally, auxiliary substances, such as wetting or emulsifying agents, pH buffering substances, gels or gelling materials, flavorings, colorants, microspheres, polymers, suspension agents, etc. may also be present herein. In addition, one or more other conventional pharmaceutical ingredients, such as preservatives, humectants, suspending agents, surfactants, antioxidants, anticaking agents, fillers, chelating agents, coating agents, chemical stabilizers, etc. may also be present, especially if the dosage form is a reconstitutable form. Suitable exemplary ingredients include macrocrystalline cellulose, carboxymethyf cellulose sodium, polysorbate 80, phenyletbyl alcohol, chiorobutanol, potassium sorbate, sorbic acid, sulfur dioxide, propyl gallate, the parabens, ethyl vanillin, glycerin, phenol, parachlorophenol, gelatin, albumin and a combination thereof. A thorough discussion of pharmaceutically acceptable excipients is available in REMINGTON'S PHARMACEUTICAL SCIENCES (Mack Pub. Co., N.J. 1991) which is incorporated by reference herein.

The pharmaceutical composition of the invention can further comprise at least one additional therapeutic agent or therapy selected from the group comprising radiotherapy, chemotherapy, immunotherapy and hormone therapy, or a combination of one of more thereof. In one aspect the at least one additional therapeutic agent comprises a ligand as described herein.

The invention also contemplates kits for the treatment and/or prevention of a disease of the invention. In one aspect of the invention, the kit comprises a pharmaceutical composition of the invention.

The kits of the invention may also comprise a container and a label or package insert on or associated with the container. Suitable containers include, for example, bottles, vials, syringes, etc. The containers may be formed from a variety of materials such as glass or plastic. The container holds a composition which is effective for treating the disease of disorder of the invention and may have a sterile access port (for example the container may be an intravenous solution bag or a vial having a stopper pierceable by a hypodermic injection needle). Alternatively, or additionally, the kits may further comprise a second (or third) container comprising a pharmaceutically-acceptable buffer (such as bacteriostatic water for injection (BWFI), phosphate-buffered saline, Ringer's solution and dextrose solution) and/or at least one additional therapeutic agent (such as e.g. a ligand of the invention) . It may further include other materials desirable from a commercial and user standpoint, including other buffers, diluents, filters, needles, and syringes.

The label or package insert may comprise instructions for use thereof. Instructions included may be affixed to packaging material or may be included as a package insert. While the instructions are typically written or printed materials they are not limited to such. Any medium capable of storing such instructions and communicating them to an end user is contemplated by this disclosure.

The invention also contemplates the use of a composition, pharmaceutical composition, AIR, CDH, vector, nucleic acids, compound, or cell of the invention in the manufacture of a medicament. Preferably, the medicament is for the treatment and/or prevention of a disease selected from the group comprising a cancer, an inflammatory disease, a genetic disorder, an infectious disease, or a degenerative disease as well as combination of one or more of these diseases.

The practice of the present disclosure will employ, unless otherwise indicated, conventional methods of protein chemistry, biochemistry, recombinant DNA techniques and pharmacology, within the skill of the art. Such techniques are explained fully in the literature. See, e.g., T. E. Creighton, Proteins: Structures and Molecular Properties (W.H. Freeman and Company, 1993); A. L. Lehninger, Biochemistry (Worth Publishers, Inc., current addition); Sambrook, et al., Molecular Cloning: A Laboratory Manual (2nd Edition, 1989); Methods In Enzymology (S. Colowick and N. Kaplan eds., Academic Press, Inc.); REMINGTON'S PHARMACEUTICAL SCIENCES (Mack Pub. Co., N.J. 1991); Carey and Sundberg Advanced Organic Chemistry 3. sup. rd Ed. (Plenum Press) Vols A and B (1992). The present disclosure is therefore to be considered as in all aspects illustrated and not restrictive, the scope of the invention being indicated by the appended Claims, and all changes which come within the meaning and range of equivalency are intended to be embraced therein.

EXAMPLES

Computational design of CDH3

The design of CDH-3 followed a protocol very similar to that presented in Giordano- Attianese, G. et al.¹⁷, based on a side chain grafting approach. The mdm2:p53 peptide interaction was selected as a starting point, since multiple small molecule inhibitors bind to the mdm2 receptor protein and prevent binding of p53. An 8 amino acid helical fragment (FXXXWXXL, SEQ ID No. 12 where F, W, and L are hotspot residues and X are the designable residues) was extracted from the helical binding motif of an mdm2-binding, p53- mimic, stapled peptide (PDB id: 5afg) and selected as the 'binding. A database of monomeric protein structures obtained through X-ray crystallography was then assembled, where proteins where sourced from the Protein Databank (PDB) if they met the following: (a) proteins with a global stoichiometry assigned in the PDB as monomers, (b) an amino acid sequence length between 80 and 160, (c) proteins containing helical secondary structures. The computational design protocol was executed as a script using RosettaScripts, and entailed the following steps. The MotifGraft²⁵ program in Rosetta attempted to graft the binding seed to all proteins in the database (scaffold proteins) . Proteins on which the fragment was grafted to a similar backbone fragment, with a maximum Ca root mean square deviation of 1.0 A, and where they maintained a steric compatibility (a clash score of maximum 5 Rosetta Energy Units or REU) with mdm2, were accepted. Once scaffolds were matched, residue positions surrounding the binding seed were designed using Rosetta fixed backbone design^26,27, allowing amino acid mutations to residues with a positive score in the BLOSUM62 matrix²⁸, with respect to the amino acid identity in the wildtype protein. The restriction to mutations based on the BLOSUM62 matrix was performed to prevent mutations to the original scaffold that could affect its stability, folding pathway, or solubility. The resulting designed proteins were scored using Rosetta for a predicted complex energy, AAG predicted score, a buried solvent accessible area and unsatisfied hydrogen bonds. Designs where Rosetta's AAG energy was higher than -7 REU were discarded. A visual inspection of the resulting scaffolds showed that many of them had non-globular structures, with extended conformations and poor packing of the binding seed. To remove scaffolds with non-globular structures, we computed a globularity score on proteins based on a metric created by Miller et al²⁹, which found that the mass AT of globular proteins correlates with the solvent accessible surface area A under the power law:

A_s = 6.3M^0,73

We thus computed a globularity score G =

and scaffolds with G< 0.9 were discarded from consideration. Finally, scaffolds were visually inspected, and those where a large small molecule ligand binding site was present in the original structure near the novel interface, or those where the binding seed was grafted to a terminal region of the protein, were removed from consideration. We thus selected 3 protein scaffolds, LD4, (designed from scaffold hydroquinone flavodoxin from Desulfovibrio vulgaris with PDB id: 1 AKU), LD5 (designed from scaffold with PDB id: 2IFQ) and LD6 (designed from scaffold with PDB id: 2FWE). After close inspection of LD6, residue 23 was manually mutated to Tyr, as this was the identity found in the starting stapled peptide.

Protein expression and purification

The gene sequences of all the designs were ordered from Twist Bioscience. Genes were synthesized with N-terminal His tag and cloned into pETl lb vector using Gibson assembling (New England Biolabs, E261 IS). The plasmids were transformed into BL21(DE3) E.coli (Thermofisher), and a single clone was picked to inoculate in 10ml for an overnight preculture, then transferred into 500 ml of TB media containing Ampicillin (100 pg/ml). The cultures were grown at 37 °C until OD600 reach around 0.8, then induced with 1 mM IPTG (Fishier Scientific) overnight at 20 °C. For the protein purification, the harvested pellet (thawed on ice, if needed) was resuspended in 40 ml lysis buffer (50 mM Tris, 500 mM NaCl and 5% Glycerol in pH 7.5) supplemented with 100 pg/ml PMSF (ROTH, 6376.2). The slurry were sonicated (Q Sonica) for 30 mins and clarified by centrifuging at 20,000 g for 20 mins.

For mdm2 protein, the protein was extracted from inclusion bodies after centrifugation. The collected pellet was washed twice with 50 ml lysis buffer containing 0.05% Triton-XlOO and solubilized by 20 ml lysis buffer supplemented with 8 M urea and 10 mM P-mercaptoethanol. Resuspended inclusion body was dialyzed against 1 liter of 4 M guanidine hydrochloride, pH 3.5 supplemented with 10 mM P-mercaptoethanol. Following, the protein was refolded by dropwise addition into 1 liter of 10 mM Tris-HCl, pH 7.0, containing 1 mM EDTA and 10 mM P-mercaptoethanol and slow mixing overnight at 4°C. The supernatant and the refolded solution was loaded to AKTA pure system (GE Life Science) for nickel affinity purification and followed with gel filtration. The purified proteins were concentrated, aliquoted and stored at -80 °C.

Compounds

VenetoClax (>99.9%, Chemietek CT-A199), Al 155463 (99.5%, Chemietek CT-A115) and NVP-CGM097(100% optically pure, Chemitek CT-CGM097), were directly used without further purification. Before use, VenetoClax, Al 155463, and NVP-CGM097 were each dissolved in DMSO as lOmM stocks. Stocks were aliquoted and stored at -20°C until use.

Circular dichroism spectrum

Protein samples were dissolved in the PBS saline buffer at a protein concentration around 0.2 mg/mL. The sample was loaded into 0.1 cm path-length quartz cuvette (HELLMA). The far- UV CD spectrum between 190 nm and 250 nm was recorded by J-815 spectrometer with a band width of 2.0 nm, and scanning speed was at 20 nm/min. Response time was set to 0.125 sec and spectra were averaged from 2 individual scans.

Size-exclusion chromatography coupled with multi-angle light scattering

LD6 was further characterized by Size Exclusion Chromatography coupled to Light Scattering (SEC-MALS) for solution behavior, and to study dimerization and drug-induced monomerization properties. LD6 was injected at 50-100 pM into Superdex™ 75 300/10 GL column (GE Healthcare) using an HPLC system (Ultimate 3000, Thermo Scientific) at a flow rate of 0.5 ml/min coupled in-line to a multi-angle light scattering device (miniDAWN TREOS, Wyatt). Static light-scattering signal was recorded from three different scattering angles. The scatter data were analyzed by ASTRA software (version 6.1, Wyatt). For drug- induced monomerization, 50 pM mdm2 or Bcl2 were mixed with equal molar ratio of LD6 and LD3, respectively. Then the mixtures were treated with either 100 pM DMSO or corresponding drugs then injected into the SEC-MALS.

Surface plasma resonance for assessing protein-protein binding affinity

SPR measurement was performed on a Biacore 8K (GE Life Science). Drug-receptor proteins(BclxL, Bcl2 and mdm2) were immobilized in the CM5 chip (GE Life Science) as a 1 ligand with the concentration at 5 pg/ml in pH 4.0 (mdm2) or pH 4.5 (BclxL and Bcl2) sodium acetate solutions, respectively. The counterpart of the designed protein were flowed as the analytes, and starting with 2-folds serially diluted in SPR running buffer (10 mM HEPES, 150 mM NaCl, 3 mM EDTA and 0.005% Surfactant P20) for the kinetic measurement. The equilibrium dissociation constant (KD) and kinetic parameters (kon and koff) were calculated in the mathematic model of 1 : 1 binding fit by using the Biacore 8K Evaluation software (GE Life Science).

SPR drug competition assay

Drug ICsos of disrupting the CDHs were measured on the Biacore 8K. 4 pM of designed binders, LD3 or LD6 were mixed with serial diluted drugs, varied from 0.01 nM, 0.1 nM, 1 nM, 10 nM, 100 nM, I M and 10 pM. The drug: binder mixtures then injected in the drugreceptor protein immobilized chips. IC50S were calculated using the response units at the plateau of 120 seconds.

Purification of mdm2 and LD6 for crystallization

The complex of mdm2 with LD6 was prepared by mixing each of the component at equal molar ratio. After overnight incubation, the complex was purified by size exclusion chromatography using a Superdex75 16 600 (GE Healthcare) equilibrated in 10 mM Tris pH 8, 100 mM NaCl and subsequently concentrated to ~15 mg/ml (Amicon Ultra-15, MWCO 3,000). Crystals were grown using sitting-drop vapor-diffusion method at 291K in a condition containing 1.5 M Ammonium sulfate, 0.1 M Sodium cacodylate pH 6.5. For cryo protection, crystals were briefly swished through mother liquor containing 25% glycerol.

Diffraction data were recorded with X06DA (PXIII) beamline at Paul Scherer Institute, Switzerland. Data integration was performed by with a high-resolution cut at Vo=l applied. The structure was determined by the molecular replacement using the Phenix Phaser module (REF). The searching of the initial phase was performed by using the mdm2 structure (PDB id: 5afg) and the computationally designed model LD6 as a search model. Manual model building was performed using Coot, and automated refinement using Phenix Refine.

CCK8 cell viability assay

A cell counting kit-8 (CCK-8) assay (Sigma ) was used to measure the cytotoxicity of three drugs on HEK293T cells. 10,000 cells were pre-seeded into 96 well plates with three replicates in 100 pl complete culture medium. The culture medium was changed into drug contained mediums with concentrations of 0 pM(DMSO),l pM, 5 pM and 10 pM. Post 24 hours drug incubation, 10 pl CCK-8 solution was added to each well for 2-4 hours incubation at 37 °C. The absorbance at 450 nm was determined by a multiplate reader (Tecan).

Cell transfection and drug treatment

HEK293T cells were maintained in DMEM medium with 10% FBS supplement and Pen/Strap (Thermofisher). Cells were maintained and split in every two days at the confluence around 80%. For all transfection experiments, 293T cells were pre-seeded 24 hours ahead, and the lipofectamine 3000 kit (Thermo Fisher) was used for transfection. For transcriptional regulation (CDH-TF) assay, 20 pg of reporter plasmids (S098) and 20 pg of effector plasmids (S108/S111/S113) were co-transfected per well. For CDH-GEMS or AIRGEMS assay, 1 ng SI 82, 5 ng SI 83 and 30 ng of each plasmid for pairing up the receptors were co-transfected. Drugs were added 12 hours post transfection and incubated with cells for 24 hours before the SEAP detection assay.

SEAP detection assay

SEAP activity (U/L) in cell culture supernatants was quantified by kinetic measurements at 405nM (1 measurement/minute for 30 minutes at 37 °C) of absorbance increase due to phosphatase-mediated hydrolysis of para-nitrophenyl phosphate (pNPP). 4 - 80 pL of supernatant was adjusted with water to a final volume of 80 pL, heat-inactivated (30 min at 65 °C), and mixed in a 96-well plate with 100 pL of 2 * SEAP buffer (20 mM homoarginine, 1 mM MgC12, 21% (v/v) diethanolamine, pH 9.8) and 20 pL of substrate solution containing 20 mM pNPP.

Reversibility assay (Dynamic regulation)

Cells were pre-seeded 24 hours before the transfection in 12-well plate. The ON-OFF-ON mode was grew without drug in the first 36hours and passed 1/3 of cells to the new dish with 500nM drugs supplemented for next 36hours with refreshing the drugs every 12 hours, then passed to the new dish with the removal of drugs. The OFF-ON-OFF mode was treated with 500nM drugs 12 hours post-transfection for the following 36 hours. Cells were also passed every 36 hours and cultured in the absence of drug for the hours from 36 to 72 hours, then add drugs again from 72hours. SEAP samples were taken every 12 hours from the culture supernatant. Design of weaker affinity variants.

The LD3 protein was computationally redesigned for decreased binding to BclxL with a gradient of affinity. Rosetta's alanine scanning filter was used to evaluate the change in AAG for the LD3:BclxL complex upon mutating each of the 22 residues in the interface of LD3 to Alanine. The resulting list was then sorted by the change in AAG, and three residues with positive levels of change in AAG were selected: L235 (5.0 REU), D240 (4.2 REU) and E124 (1.8 REU), where higher REU values are predicted to result in greater affinity losses. The three mutations were selected to provide a 'gradient' of affinities between LD3 and BclxL.

Drug-insensitive receptor mutations predictions

BclxL and Bcl2 were redesigned for resistance to A-l 155463 and Venetoclax, respectively, following a computational strategy similar to one used to predict drug resistance [refs]. Briefly, a set of residues in the receptor protein's (BclxL/Bcl2) binding site was selected for redesign. From this set, a number of mutations was evaluated for binding to the binder protein (LD3) (positive design) or the drug (negative design). Afterwards all mutations were ranked according to the difference in energy between the positive design and negative design. BclxL: The structure of BclxL bound to A-l 155463 (PDB id: 4QVX) was used for the negative design strategy, while the model of BclxL bound to LD3 (based on the BclxL:BIM BH3 structure with PDB id: 3FDL) was used for positive design. Six BclxL residues in the binding site of Drugl (E98, R102, F105, T109, S145 and A149) were manually selected for redesign due to their closeness to drug moeities and relative distance to LD3 in the positive design structure. Each of these residues was allowed to mutate to residues with similar size/properties, restricted to a maximum of two simultaneous mutations from wildtype: E98: {E/S}, R102: {F/R/K/D/E/H}, F105:{F/L/V/EA}; T109: {S/A/T/L/V}; S145: {S/D/E/V/A}; A149:{V/A/L/I}. The total sequence space thus consisted of 253 unique sequences. The Rosetta program was then used to redesign the positive design structure (LD3:BclxL complex) and the negative design structure (Drugl :BclxL complex) for each of the 253 sequences. A score was computed for the complex state of each sequence in each of the two states, and sequences were ranked according to ratio. Five sequences iBclxL vl (T109L, A149L), iBclxL v2 (A67V), iBclxL v3 (R102E, F23I), iBclxL v4 (R102F, T109V), and iBclxL v 5 (E98S, F105I) were selected from the top results. Bcl2: the structure of Bcl2 bound to Venetoclax (Drug 2 , PDB id: 41vt) was used for the negative design strategy, while the model of Bcl2 bound to LD3 (PDB id: 6iwb) was used for positive design. Five Bcl2 residues in binding site of Venetoclax (A100, D103, V148, V156 and Y202) were manually selected for redesign due to their closeness to drug moieties and relative distance to LD3 in the positive design structure. Each of these residues was allowed to mutate to residues with similar size/properties, restricted to a maximum of two simultaneous mutations from wildtype: A100: {A/S/T/V}, D103: {D/N/E/Q/S}, V148: {V/I/L/M/T}, V156: {V/I/L/M/T} and Y202: {Y/W/F/H/R/K/Q/E}. The total sequence space thus consisted of 251 unique sequences. The Rosetta program was then used to redesign the positive structure(PDB id: 6iwb) and the negative design structure(PDB id: 41vt) for each of the 251 sequences. A score was computed for the complex state of each sequence in each of the two states, and sequences were ranked according to the ratio. Three sequences iBcl-vl(156I_202H), iBcl2_v2(103N_202H) and iBcl2_v3(100T_103S) were selected from the top results, three mutations were enriched in the top designs which contributed to the combinatorial sequences of iBcl2_v4( 100 V_103N_202H).

Table 1 : Protein sequences of CDHs(l-3)

Table 2: Residue mutation indication ( ) of weaker binders and insensitive receptors

Results

Rational design of globular, potent, stable CDHs We previously described the design of CDH-1, composed of the BclxL:LD3 complex which dissociates in the presence of the BclxL-specific inhibitor A-l 155463 (Drug-1). However, Drug-1 is not a clinically-approved drug, which limits the potential translational application of the CDH-1. We found that LD3 also binds to Bcl2 (Fig. la), a protein from the Bcl2 family^30,31 that is closely related to BclxL, with an affinity of 0.8 nM as determined by surface plasmon resonance (SPR) (Fig. lb). Bcl2 is the target of Venetoclax³² (Drug-2), a BH3 chemical mimetic which blocks the anti-apoptotic Bcl2 protein and is clinically- approved for chronic lymphocytic leukemia (CLL) treatment²¹. We therefore assembled CDH-2, where we exchanged the BcLxL component with Bcl2 (Fig. la). Drug-2 effectively disrupts the CDH-2 heterodimer in vitro both in an SPR drug competition assay (Fig. 1c) (IC50 = 67.5 nM), as well as by the elution profiles by size-exclusion chromatography coupled with multi-angle light scattering (SEC-MALS).

To further expand the panel of CDHs and the chemical space of small-molecule disruptors, we designed a novel CDH-3 (Fig. Id), based on the interaction between p53 and mdm2 as the starting point. Over the past two decades, numerous compounds have been developed to inhibit this interaction and several candidates have been tested in clinical trials^33,34.

Particularly, the dihydroisoquinoline derivative, NVP-CGM097 (Drug-3), is currently undergoing phase 1 clinical trials^15,22,35. Leveraging the same computational design approach described for the CDH-1¹⁷, we utilized the structural information of the mdm2:p53-like stapled peptide complex (PDB ID: 5afg)³⁶ and searched for proteins that could accommodate the p53-like helical motif. We performed a structural match over approximately 11,000 putative protein scaffolds and after the design stage we selected for experimental characterization three designs which we refer to as LD4, LD5 and LD6. From these three designs, LD6 is a design based on the mus musculus thioredoxin protein³⁷, showed the best biochemical behavior being monomeric in solution and a melting temperature of 62 °C. LD6 bound to mdm2 with a dissociation constant (Kd) of 4 nM (Fig. le) as determined by SPR, showing a higher binding affinity than those reported for the wild-type p5317-29 peptide (820±60 nM) or the stapled p53 peptide which we used as the input peptide motif ( 12±3 nM)³⁶. Next, we tested if Drug-3 promoted the dissociation of the mdm2:LD6 using an SPR drug-competition assay we determined the Drug-3 with an IC50 of 106 nM for inhibiting the hetero-dimerization of mdm2:LD6 (Fig. If). These results were also confirmed using SEC- MALS where the drug separate the complex into two monomers.

To evaluate the structural accuracy of the designed CDH-3, we solved the crystal structure of the mdm2:LD6 complex at 2.9 A resolution by X-ray crystallography. The structure of the complex closely matched our computational model, with backbone RMSDs of 0.372 A over the overall structure (Fig. 1g) and 0.243 A over the p53 motif region (Fig. 1g). The conformation of key residues observed in the native interface were closely mimicked in the crystal structure of LD6 in complex with mdm2. Overall, our results show that we can robustly use computational design to create protein modules which the assembly state is controlled by the presence of small-molecules. CDHs function as OFF-switches to control intra and extra-cellular activities

To test whether the designed CDHs could be useful for distinct synthetic biology applications, we generated OFF-switches in two cellular regulation systems: transcriptional gene regulation and control of endogenous signaling pathways (Fig. 2a, d). To establish reliable drug concentration ranges for our experiments, we assessed drug toxicities and impact on the expression of reporter proteins in HEK293T cells, Drugsl-3 were well tolerated up to 10 pM.

To regulate transgene activation events by small molecule inputs we used the CDHs with a split transcription factor design (CDH-TFs) based on the well-characterized Gal4/UAS system²³ (Fig. 2a).

In brief, the CDH mediated dimerization of the Gal4 DNA binding domain and the Rel65 activation domain should triggering the transcription of reporter protein (secreted alkaline phosphatase (SEAP)). If the CDH switch is functional, the drugs should dissociate the engineered CDH-TFs and diminish SEAP expression. Upon drug treatment, all CDH-(l-3)- TFs were responsive to their respective drugs with dynamic response ranges between 3 and 10-fold (untreated vs 1 pM Drug) (Fig. 2b). The CDH-TFs showed dose-dependent responses with IC₅os of 9.6, 33.5 and 540 nM for CDH-l-TF, CDH-2-TF, and CDH-3-TF respectively (Fig. 2c). The CDH-TF responses were also small molecule specific, as each CDH-TF only responded to their respective drug. Finally, we explored the dynamic behavior of the designed components by subjecting the CDH-TFs transfected cells to intermittent drug treatments. The CDH-TFs showed a significant decrease upon drug treatment and subsequent recovery of SEAP expression upon drug withdrawal, exhibiting a reversible behavior which is central for many synthetic biology applications. Together, these results show that our extended panel of CDHs can be used as intracellular switches and may likely be adaptable to control several cellular activities in engineered cells that are dimerization/co-localization dependent.

To test modularity and performance robustness of CDHs in different molecular contexts, we used them in the generalizable extracellular molecule sensor (GEMS) receptor signaling pathway platform²⁴ and built CDH-GEMS. In these constructs the CDHs are located extracellularly and function as ectodomain switches, fused to the backbone of a erythropoietin receptor, and an intracellular interleukin 6 receptor domain which activates the JAK/STAT pathway (Fig. 2d). The CDH-GEMS trigger the activation of the JAK/STAT pathway by the dimerization of their subunits. The drugs are expected to split the CDHs inactivating the signaling pathway which will be assessed by the expression of the reporter protein SEAP. All three CDH-GEMS showed a dose-dependent response with ICsos of 1.7, 23 and 149 nM respectively for CDH(l-3)-GEMS with their cognate drugs (Fig. 2f). Their dynamic response range varied from approximately 6-fold (CDH-2) to 8-fold (CDH-1,3) (Fig. 2e). The CDHs-GEMS only function while both CDH domain be transfected and is specific to their own corresponding drugs. The CDH-GEMS were also reversible upon intermittent drug treatment. Despite that comparisons between CDH-TFs and CDH-GEMS are not straightforward due to differences in the reporter signaling pathways, we observe a general trend for the CDHs to present lower IC50s when used as extracellular modules which is consistent with overcoming drug permeability issues in cells.

Our design process optimized the CDHs for the strength of their interaction, achieving affinities as tight as 4 pM (CDH-1). However, it is unclear how tight the strength of the interaction for a specific application should be, and, indeed, many intracellular pathways effectively function with weak affinities³⁸. We hypothesized that the sensitivity of a CDH could be tuned towards lower drug demand, by weakening the interaction of the two domains in a CDH. If this assumption holds true, it should be possible to generate CDHs that function with drug concentrations below the doses required for normal drug effects, greatly reducing the risk of adverse reactions. We performed computational alanine scanning on the designed binder LD3 to generate lower-affinity binders and selected three mutants (LD3_vl-v3) which were predicted to decrease the interaction energy (AAG) to different extents (Fig. 2g). LD3 affinity to BclxL measured by SPR was approximately 4 pM and the weakest design (LD3_v3) showed a binding affinity of 22.3 nM, a decrease of more than 5,000 fold (Fig. 2g) which lowered the in vitro IC50S from 101 nM to 14.6 nM with a 7 fold shrinkage. Next, we tested whether this decrease in affinity also translated into an increased drug sensitivity comparing LD3 with LD3_v3 in CDH-GEMS engineered cells. We observed a shift of 21- fold in IC50S of the weakest binder LD3_v3 in BclxL-based CDH-GEMS application when compared to the original LD3, showing that we successfully decreased the amount of drug necessary to control the CDH (Fig. 2h). We also tested the panel of LD3 mutants in the context of the CDH-2 and observed that the affinity to Bcl2 was also reduced from 0.8 nM to 147 nM and translated into an decrease of IC50S in vitro from 71 nM to 19.5 nM. The increase of drug sensitivity comparing LD3 and the weakest binder LD3_v3 in the context of CDH-2- GEMS is 11 -fold (Fig. 2i). The tunable affinity enables the usage of lower doses of a clinically approved drug to control the designed switches in vitro and in cellular applications (Fig. 3h-i). While the overall observed trends were maintained, the magnitude of the reductions in binding affinities did not directly translated to the reductions in cellular activity assays which may reflect a number of factors associated to the complexity of performing measurements in living cells. In summary, the reduced binding affinity increased the sensitivity of the CDHs towards their small-molecule drugs. This property enlarges the panel of molecular switches that can function under reduced drug dosages which could avoid potential toxicities while remain effective at controlling designer cell therapies.

Multistate design to functionize the CDHs into ON switches

CDH components are inherently well suited to trigger OFF outputs, which are highly desirable in some settings, as we previously demonstrated by its integration in CAR-T cells¹⁷. However, switches that can induce protein colocalization are naturally better suited to obtain ON outputs. We propose that monomerization inducing components (CDHs) can be rapidly repurposed into dimerization inducing systems (CIDs), thus diversifying the protein switches available and expanding the chemical space of CID systems to control ON outputs.

To address this challenge, we developed a new CID switch architecture, dubbed activation by inhibitor release (AIR) switches (Fig. 3 a). The AIR architecture relies on three protein components: CDH drug receptor, CDH protein binder, and a rationally designed druginsensitive CDH receptor that retains the binding capability to the protein binder. These three components are assembled into two distinct polypeptide chains to form the AIR switches. In one chain the two elements of the CDH are genetically fused with a flexible linker forming an intramolecular binding interaction that will be disrupted in the presence of the drug, unveiling the binding site of the protein binder (Fig. 3 a). In the AIR's second chain, a drug-insensitive receptor has its binding site available to mediate the intermolecular interaction between the two chains only when the drug is present. By creating this multidomain architecture, with tailored components for drug sensitivity, we aim to design protein switches that heterodimerize through an allosteric drug-binding event, effectively mediating a chemically- induced proximity mechanism. Crucial to our AIR architecture is the ability to rationally design drug insensitive receptors that can retain binding to the designed binders. In computational design this is a well-defined problem of multistate optimization where the sequence space is searched to optimize simultaneously several objective functions^39,40. In our case, we performed a protocol where ligand site residues in the drug receptor close to the drug and away from the designed binder were selected and sampled for mutations that knocked-out drug binding (negative design) and maintained affinity to the protein binder (positive design)^41-43 (Fig. 4a, left) (see Methods). We first applied this protocol to the CDH-

1 composed of the BclxL:LD3 protein pair (see Methods). Six residues in BclxL (D98, R102, F105, T109, S145 and A149) were selected as designable (Fig. 3b) for the multistate design, and five putative drug insensitive BclxL mutants (iBclxL_vl-5) were selected from the computational simulations (Supplementary Table ). In an SPR-based assay iBclxL_v3 (R102E, F105I) and iBclxL_v5 (E98S, F105I) showed the greatest drug resistance and did not dissociate from LD3 in the presence of 10 pM of Drug- 1, which was sufficient to fully dissociate the wildtype BclxL (Fig. 3c). The iBclxL_v3 :LD3 retained a high affinity interaction (Kd = 3.8 nM), which was however considerably lower than BclxL:LD3 complex (Kd = 4 pM). Structural analysis of the negative design BclxL:Drug-l complex (PDB id: 4QVX) shows that Fl 051, shared by both designs, removes a Pi-stacking interaction with Drug-1 which points to this mutation as the main driver of resistance to Drug-1. To confirm the drug resistance of iBclxL_v3 in a cellular context, we tested the iBclxL_v3 paired with LD3 in the CDH-GEMS platform. In contrast to the CDH-l-GEMS, Drug-1 failed to disrupt the iBclxL_v3:LD3 complex at pM concentrations.

Next, we assembled the AIR architecture in the GEMS platform (Fig. 3a, right) to evaluate its potential as an effective ON switch. We fused the AIR components in the ectodomains of the GEMS and measured SEAP activity as a reporter for drug triggered activation. We constructed the AIR-l-GEMS using iBclxL_v3 in one chain and a genetically fused CDH-1 (LD3-GGGGSX3 linker-BclxL) in the second chain. We co-expressed these constructs in HEK293T cells, and observed that the AIR- 1 -GEMS was responsive to Drug-1, effectively turning ON the expression of the reporter gene SEAP, with a very sensitive drug response (ECso = 18.68 ± 4.65 nM) (Fig. 4g), and a 64-fold dynamic response range (untreated vs 100 nM Drug-1 treated) (Fig. 3f).

To test whether we could design a second ON-switch controlled by a different drug, we applied the same design strategy to generate a Drug-2 resistant receptor and created the AIR-

2 based on the CDH-2 components (Bcl2:LD3) which is responsive to the clinically approved drug (Drug-2). Similarly to the AIR-1, five residues located in the drug binding site (A100, D103, V148, V156, Y202) (Fig. 3d) were used for multistate design to create a druginsensitive Bcl2 variants (iBcl2). Four designs (Supplementary Table5) were screened on the AIR-GEMS platform and one showed activation upon addition of Drug-2. The iBcl2_v4 gave the strongest activation effect retained interaction with LD3 (Kd = 9.8 pM) and being resistant to Drug-2 comfirmed by SPR drug competition assay (Fig. 3e) and CDH-GEMS based cellular assay. Subsequently, the functional ON-switch design, is referred to as AIR-2- GEMS, and was built using iBcl2_v4 (A100V, D103N, Y202H) and CDH-2 components (LD3-GGGGSX3 Iinker-Bcl2) showing an ECso of 0.5 ± 0.27 nM (Fig. 3g). A comparison between the two AIR-GEMS shows that AIR-2-GEMS is approximately 36-fold more sensitive to drug treatment in terms of EC50s, but AIR-l-GEMS shows a higher magnitude of response when comparing the resting and the fully activated states (350 U/L SEAP production of AIR1-GEMS vs 160 U/L for the AIR2-GEMS at 100 nM Drug concentration) (Fig. 3f). These distinct behaviors may be due to differences in affinities of LD3 to the different drug receptors, where a lower affinity affords a more responsive receptor requiring lower drug concentrations, but the output is less sustained reaching overall lower activation levels.

These results suggest that the conversion of CDH components into the AIR architecture is an effective way to build ON switches responsive to low-doses of small-molecule drugs (including clinically approved). As the panel of clinically-tested PPI inhibitors grows in the future, our AIR approach combining the design of drug-insensitive receptor and a repurposed PPI inhibitor could be a generalizable approach to control cellular activation using protein-based ON-switches.

Multi-input multi-output control by orthogonal switches

An overarching goal in synthetic biology is to use living cells as bio-computing units⁴⁴, where orthogonal sensing components display distinct switching behaviors. The need for these complex logic devices goes beyond basic applications, as shown in the CAR T field, where there is a growing need to control multiple functions using orthogonal signals (e.g. suicide switches⁴⁵, tunable activity⁴⁶, molecule secretion⁴⁷). As a final proof-of-concept, we engineered multi-input/multi-output cells by combining several designed switches in HEK293T cells, that were used as a model system for further applications. In analogy with control systems in electrical engineering, we sought to implement three distinct input/output systems beyond single-input/single-output (SISO), such as: multi-inputs/single-output (MISO), single-input/multi-outputs (SIMO) and multi-inputs/multi-outputs (MIMO)⁴⁸. We first engineered MIMO cells to respond to two drugs with distinct outputs (Fig. 4a) by co-transfecting the CDH-3-TF with either AIR-l-GEMS (Fig. 4b) or AIR-2-GEMS (Fig. 4c). The combination of two drugs did not affect cell viability at the highest concentration of 1 pM each. The CDH-3-TF controlled luciferase reporter expression decreased in a Drug-3 dose-dependent manner and AIR-GEMS (AIR-1 and AIR-2) showed an increase of SEAP expression dependent on their cognate drugs (Fig. 4b-c). Showing an orthogonal control of two distinct cellular activities by two drug inputs in the same engineered cell population.

We then engineered MISO cells (Fig. 4d) with CDH-l-TF and CDH-3-TF dual switches (Fig. 4e) that encode OFF behaviors controlled by Drug-1 and Drug-3. As expected, the individual CDHs showed similar response curves to those observed when they were tested in isolation (Fig. 2c). Interestingly, when treated with both drugs the engineered cells showed a unique behavior where the MISO’s response curve showed three persistent response levels controlled by drug amounts (high, medium, low) (Fig. 4e), unlike the typical SISO systems which only show two persistent states (high, low) (Fig. 2c, f). Similarly, the engineered cells endowed with CDH-l-GEMS and CDH-3-GEMS dual-circuits (Fig. 4f) showed three persistent response levels. This type of output in biological systems could be utilized to produce graded levels of responses that are robust to minor oscillations in drug concentrations, which could be used to study of fundamental biological processes but also to provide persistent response levels in translational applications.

Finally, we engineered SIMO cells (Fig. 4g) using two designed switches that produce ON and OFF responses upon exposure to the same drug. We transfected cells with combinations of AIR-GEMS and CDH-TFs which controlled the expression of the reporter proteins SEAP and luciferase, respectively. Both Drug-1 and Drug-2 simultaneously activated SEAP production and suppressed the expression of Luciferase (Fig. 4h, i) in their respective switch systems, demonstrating that we successfully designed protein components which can physically dissociate and associate simultaneously under the control of the same drug. Drug-1 and Drug-2 are the only drugs working similarly with rapamycin that can control both assembly^3,49 and disassembly¹⁴ of protein components. Therefore, computational design expanded the repertoire of protein switches controlled by the same compounds that could be used to simultaneous turn ON and OFF synergistic outputs.

In summary, leveraging our computationally designed switches we created cellular biocomputing units with multiple output modalities that can be useful to deal with the inherent complexity of engineering mammalian cells and have the potential for translational applications in synthetic biology.

REFERENCES

1. Xie, M. & Fussenegger, M. Mammalian designer cells: Engineering principles and biomedical applications. Biotechnol. J. 10, 10051018-n/a (2015).

2. Stanton, B. Z., Chory, E. J. & Crabtree, G. R. Chemically induced proximity in biology and medicine. Science 359, eaao5902 (2018).

3. Wu, C.-Y., Roybal, K. T., Puchner, E. M., Onuffer, J. & Lim, W. A. Remote control of therapeutic T cells through a small molecule-gated chimeric receptor. Science 350, aab4077 (2015).

4. Barlow, A. D., Nicholson, M. L. & Herbert, T. P. Evidence for Rapamycin Toxicity in Pancreatic P-Cells and a Review of the Underlying Molecular Mechanisms. Diabetes 62, 2674-2682 (2013).

5. Sarkar, S., Ravikumar, B., Floto, R. A. & Rubinsztein, D. C. Rapamycin and mTOR- independent autophagy inducers ameliorate toxicity of polyglutamine-expanded huntingtin and related proteinopathies. Cell Death Differ. 16, 46-56 (2009).

6. Sun, J. & Sadelain, M. The quest for spatio-temporal control of CAR T cells. Cell Res. 25, 1281-1282 (2015).

7. Riddell, S. R. et al. T-cell mediated rejection of gene-modified HIV-specific cytotoxic T lymphocytes in HIV-infected patients. Nat. Med. 2, 216-223 (1996).

8. Jensen, M. C. et al. Antitransgene Rejection Responses Contribute to Attenuated Persistence of Adoptively Transferred CD20/CD19-Specific Chimeric Antigen Receptor Redirected T Cells in Humans. Biol. Blood Marrow Transplant. 16, 1245-1256 (2010).

9. Berger, C., Flowers, M. E., Warren, E. H. & Riddell, S. R. Analysis of transgene-specific immune responses that limit the in vivo persistence of adoptively transferred HSV-TK- modified donor T cells after allogeneic hematopoietic cell transplantation. Blood 107, 2294-2302 (2006).

10. Hill, Z. B., Martinko, A. J., Nguyen, D. P. & Wells, J. A. Human antibody -based chemically induced dimerizers for cell therapeutic applications. Nat. Chem. Biol. 14, 112-117 (2018).

11. Foight, G. W. et al. Multi -input chemical control of protein dimerization for programming graded cellular responses. Nat. Biotechnol. 37, 1209-1216 (2019). 12. Glasgow, A. A. et al. Computational design of a modular protein sense/response system. http://biorxiv.org/lookup/doi/10.! 101/648485 (2019) doi: 10.1101/648485.

13. Rivera, V. M. Regulation of Protein Secretion Through Controlled Aggregation in the Endoplasmic Reticulum. Science 287, 826-830 (2000).

14. Rollins, C. T. et al. A ligand-reversible dimerization system for controlling proteinprotein interactions. Proc. Natl. Acad. Sci. 97, 7096-7101 (2000).

15. Ran, X. & Gestwicki, J. E. Inhibitors of protein-protein interactions (PPIs): an analysis of scaffold choices and buried surface area. Curr. Opin. Chem. Biol. 44, 75-86 (2018).

16. Arkin, M. R., Tang, Y. & Wells, J. A. Small-Molecule Inhibitors of Protein-Protein Interactions: Progressing toward the Reality. Chem. Biol. 21, 1102-1114 (2014).

17. Giordano-Attianese, G. et al. A computationally designed chimeric antigen receptor provides a small-molecule safety switch for T-cell therapy. Nat. Biotechnol. 38, 426-432 (2020).

18. Tao, Z.-F. et al. Discovery of a Potent and Selective BCL-XL Inhibitor with in Vivo Activity. ACS Med. Chem. Lett. 5, 1088-1093 (2014).

19. Gorovits, B. & Koren, E. Immunogenicity of Chimeric Antigen Receptor T-Cell Therapeutics. BioDrugs 33, 275-284 (2019). 0. Deeks, E. D. Venetoclax: First Global Approval. Drugs 76, 979-987 (2016). 1. Roberts, A. W. et al. Targeting BCL2 with Venetoclax in Relapsed Chronic Lymphocytic Leukemia. N. Engl. J. Med. 374, 311-322 (2016). 2. Holzer, P. et al. Discovery of a Dihydroisoquinolinone Derivative (NVP-CGM097): A Highly Potent and Selective MDM2 Inhibitor Undergoing Phase 1 Clinical Trials in p53wt Tumors. J. Med. Chem. 58, 6348-6358 (2015). 3. Tang, J. C. Y. et al. A Nanobody -Based System Using Fluorescent Proteins as Scaffolds for Cell-Specific Gene Manipulation. Cell 154, 928-939 (2013). 4. Scheller, L., Strittmatter, T., Fuchs, D., Bojar, D. & Fussenegger, M. Generalized extracellular molecule sensor platform for programming cellular behavior. Nat. Chem. Biol. 14, 723-729 (2018). 5. Silva, D.-A., Correia, B. E. & Procko, E. Motif-Driven Design of Protein-Protein Interfaces, in Computational Design of Ligand Binding Proteins (ed. Stoddard, B. L.) vol. 1414 285-304 (Springer New York, 2016). 6. Fleishman, S. J. et al. RosettaScripts: A Scripting Language Interface to the Rosetta Macromolecular Modeling Suite. PLoS ONE 6, e20161 (2011). 27. Kuhlman, B. & Baker, D. Native protein sequences are close to optimal for their structures. Proc. Natl. Acad. Sci. 97, 10383-10388 (2000).

28. Henikoff, S. & Henikoff, J. G. Amino acid substitution matrices from protein blocks. Proc. Natl. Acad. Sci. 89, 10915-10919 (1992).

29. Miller, S., Janin, J., Lesk, A. M. & Chothia, C. Interior and surface of monomeric proteins. J. Mol. Biol. 196, 641-656 (1987).

30. Department of Biotechnology, Assam University, Silchar-788011, Assam, India & Choudhury, S. A comparative analysis of BCL-2 family. Bioinformation 15, 299-306 (2019).

31. Hanada, M., Aime-Sempe, C., Sato, T. & Reed, J. C. Structure-function analysis of Bcl-2 protein. Identification of conserved domains important for homodimerization with Bcl-2 and heterodimerization with Bax. J. Biol. Chem. 270, 11962-11969 (1995).

32. Souers, A. J. et al. ABT-199, a potent and selective BCL-2 inhibitor, achieves antitumor activity while sparing platelets. Nat. Med. 19, 202-208 (2013).

33. Neochoritis, C., Estrada-Ortiz, N., Khoury, K. & Dbmling, A. p53-MDM2 and MDMX Antagonists, in Annual Reports in Medicinal Chemistry vol. 49 167-187 (Elsevier, 2014).

34. Tisato, V., Voltan, R., Gonelli, A., Secchiero, P. & Zauli, G. MDM2/X inhibitors under clinical evaluation: perspectives for the management of hematological malignancies and pediatric cancer. J. Hematol. Oncol.J Hematol Oncol 10, 133 (2017).

35. Townsend, E. C. et al. The MDM2 Inhibitor NVP-CGM097 Is Highly Active in a Randomized Preclinical Trial of B-Cell Acute Lymphoblastic Leukemia Patient Derived Xenografts. Blood 126, 797-797 (2015).

36. Lau, Y. H. et al. Double Strain-Promoted Macrocyclization for the Rapid Selection of Cell-Active Stapled Peptides. Angew. Chem. Int. Ed. 54, 15410-15413 (2015).

37. Stimimann, C. U. et al. High-resolution Structures of Escherichia coli cDsbD in Different Redox States: A Combined Crystallographic, Biochemical and Computational Study. J. Mol. Biol. 358, 829-845 (2006).

38. Hein, M. Y. et al. A Human Interactome in Three Quantitative Dimensions Organized by Stoichiometries and Abundances. Cell 163, 712-723 (2015).

39. Havranek, J. J. & Harbury, P. B. Automated design of specificity in molecular recognition. Nat. Struct. Biol. 10, 45-52 (2003). 40. Loftier, P., Schmitz, S., Hupfeld, E., Sterner, R. & Merkl, R. Rosetta:MSF: a modular framework for multi-state computational protein design. PLOS Comput. Biol. 13, el005600 (2017).

41. Frey, K. M., Georgiev, I., Donald, B. R. & Anderson, A. C. Predicting resistance mutations using protein design algorithms. Proc. Natl. Acad. Sci. U. S. A. 107, 13707- 13712 (2010).

42. Reeve, S. M. et al. Protein design algorithms predict viable resistance to an experimental antifolate. Proc. Natl. Acad. Sci. U. S. A. 112, 749-754 (2015).

43. Ojewole, A. et al. OSPREY Predicts Resistance Mutations Using Positive and Negative Computational Protein Design. Methods Mol. Biol. Clifton NJ 1529, 291-306 (2017).

44. Kitada, T., DiAndreth, B., Teague, B. & Weiss, R. Programming gene and engineered- cell therapies with synthetic biology. Science 359, eaadl067 (2018).

45. Di Stasi, A. et al. Inducible Apoptosis as a Safety Switch for Adoptive Cell Therapy. N. Engl. J. Med. 365, 1673-1683 (2011).

46. Caruso, H. G. et al. Tuning Sensitivity of CAR to EGFR Density Limits Recognition of Normal Tissue While Maintaining Potent Antitumor Activity. Cancer Res. 75, 3505- 3518 (2015).

47. Pegram, H. J. et al. Tumor-targeted T cells modified to secrete IL-12 eradicate systemic tumors without need for prior conditioning. Blood 119, 4133-4141 (2012).

48. Shockley, E. M., Rouzer, C. A., Marnett, L. J., Deeds, E. J. & Lopez, C. F. Signal integration and information transfer in an allosterically regulated network. Npj Syst. Biol. Appl. 5, 23 (2019).

49. DeRose, R., Miyamoto, T. & Inoue, T. Manipulating signaling at will: chemically- inducible dimerization (CID) techniques resolve problems in cell biology. Pfliig. Arch. - Eur. J. Physiol. 465, 409-417 (2013).

Claims

1. An activation by inhibitor release (AIR) switch comprising a) a first polypeptide chain comprising a first ligand-binding domain; and b) a second polypeptide chain comprising i) a second ligand binding domain, ii) a linker, said linker is fused to the second ligand binding domain and to iii) a binder; wherein the first ligand binding domain shows binding affinity to the binder, wherein the second ligand binding domain shows binding affinity to the binder and forms an auto-inhibited binding interaction with said binder in the absence of a ligand, wherein the first and the second polypeptides are two separate polypeptides, and wherein the AIR dimerizes in the presence of a ligand which disrupts and releases the auto-inhibited binding interaction and enables the interaction between the binder and the first ligand-binding domain to induce (ON) or stop (OFF) a biological activity.

2. The AIR switch of claim 1 further comprising c) one or more transmembrane domain and/or d) one or more intracellular signaling domain(s).

3. The AIR switch of claim 1 or 2, wherein the first polypeptide chain further comprises an extracellular domain (ECI) and/or the second polypeptide chain comprises an extracellular domain (EC2).

4. The AIR switch of any one of the preceding claims, wherein the ligand induces (ON) or stops (OFF) the biological activity through a conformational reorganization of the AIR switch.

44

5. The AIR switch of any one of the preceding claims, wherein the affinity to the binder of the second ligand binding domain is equal to, or higher than, the affinity to the binder of first ligand binding domain.

6. The AIR switch of any one of the preceding claims, wherein the first ligand binding domain is insensitive, or less sensitive, to the ligand.

7. The AIR switch of any one of the preceding claims, wherein the ligand is an inhibitor or an activator selected from the group comprising a small molecule, an antibody, a proteinbased therapeutic, an antigen binding fragment thereof, and a hormone, or a combination of one or more thereof.

8. The AIR switch of any one of the preceding claims, wherein the ligand is an inhibitor selected from the group comprising a Bcl-xL inhibitor, a Bcl2 inhibitor and a mdm2 inhibitor, or a combination of one or more thereof.

9. The AIR switch of any one of the preceding claims, wherein the first and/or second ligand-binding domain is/are selected from the group comprising Bcl-xL polypeptide, Bcl2 polypeptide, MDM2 polypeptide, and a derivative or a combination of one or more thereof.

10. The AIR switch of any one of the preceding claims, wherein i) the ligand is selected from the group comprising A-l 155463, A-1331852, and navitoclax (ABT-263), when the first and/or second ligand-binding domain is/are the Bcl-xL polypeptide or a derivative thereof, ii) the ligand is selected from the group comprising venetoclax, obatoclax, and navitoclax (ABT-263), when the first and/or second ligand-binding domain is/are the Bcl2 polypeptide or a derivative thereof, or

45 iii) the ligand is selected from the group comprising NVP-CGM097, RG7388, and NVP- HDM201, when the first and/or second ligand-binding domain is/are the MDM2 polypeptide or a derivative thereof.

11. The activation by inhibitor release (AIR) of any one of the preceding claims, wherein the linker is a peptide comprising one or more Serine and/or Glycine amino acids.

12. The activation by inhibitor release (AIR) of any one of the preceding claims, wherein the binder incorporates one or more motifs to interact with the ligand binding domain of the first and/or second ligand binding domains.

13. The activation by inhibitor release (AIR) of any one of the preceding claims, wherein the binder comprises an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 1 (LD3_vl), SEQ ID No. 2 (LD3_vl), SEQ ID No. 3 (LD3_vl), SEQ ID No. 8 (LD3), SEQ ID No. 10 (LD6).

14. The activation by inhibitor release (AIR) of any one of the preceding claims, wherein the first and/or second ligand-binding domain comprises an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 4 (iBl-XL_v3), SEQ ID No. 5 (iBcl2_v4), SEQ ID No. 6 (Bcl-XL), SEQ ID No. 7 (Bcl2), SEQ ID No. 9 (MDM2).

15. The activation by inhibitor release (AIR) of any one of claims 2-14, wherein the one or more intracellular signaling domain(s) is selected from the group comprising CD247/ CD3z, CD137/4-1BB, Zap70, STAT3, STAT5, and a derivative or a combination of one or more thereof.

46

16. The activation by inhibitor release (AIR) of any one of the preceding claims, wherein

- one of the first or second polypeptide chain further comprises, or is operably linked to, or is fused to, one or more DNA binding domain, and

- the remaining first or second polypeptide chain further comprises, or is/ operably linked to, or is fused to, one or more transcription factor.

17. The activation by inhibitor release (AIR) of claim 16, wherein the one or more DNA binding domain is selected from the group comprising Gal4, TetR, dCas9, ZFN, and TALEN, or a combination of one or more thereof.

18. The activation by inhibitor release (AIR) of claim 16 or 17, wherein the one or more transcription factor is selected from the group comprising VP 16, VP64, VPR and p65, or a combination of one or more thereof.

19. A chemically disruptable heterodimer (CDH) switch comprising a) a first polypeptide chain comprising a ligand-binding domain; and b) a second polypeptide chain comprising a binder wherein the ligand-binding domain shows binding affinity to the binder, wherein the first and the second polypeptides are two separate polypeptides, and wherein the CDH dimerizes in the absence of a ligand which disrupts the heterodimeric interaction to induce (ON) or stop (OFF) a biological activity.

20. A chemically disruptable heterodimer (CDH) switch comprising a) a first polypeptide chain comprising a ligand-binding domain based on the mdm2 protein; and b) a second polypeptide chain comprising a binder wherein the ligand-binding domain shows binding affinity to the binder, wherein the first and the second polypeptides are two separate polypeptides, and wherein the CDH dimerizes in the absence of a ligand which disrupts the heterodimeric interaction to induce (ON) or stop (OFF) a biological activity.

21. The CDH of claim 19 or 20, further comprising c) one or more transmembrane domain and d) one or more intracellular signaling domain(s).

22. The CDH of any one of claims 19 to 21, wherein the first polypeptide chain further comprises an extracellular domain (ECI) and/or the second polypeptide chain comprises an extracellular domain (EC2).

23 The CDH of claim 22, wherein the extracellular domain (ECI) and/or the extracellular domain (EC2) is selected form the group comprising a part of human CD8 a chain, partial extracellular domain of CD28, FcyRllla receptor, IgG, IgM, IgA, IgD, IgE, and an Ig hinge, or a functional fragment and/or combination thereof.

24. The CDH of any one of claims 19 to 23, wherein the ligand induces or stops the biological activity through a conformational reorganization of the CDH.

25. The CDH of any one of claims 19 to 24, wherein the ligand is an inhibitor or an activator selected from the group comprising a small molecule, an antibody, protein-based therapeutic, an antigen binding fragment thereof, and a hormone.

26. The CDH of any one of claims 19 to 25, wherein the first ligand-binding domain comprises an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in SEQ ID No. 6 (Bcl-XL), SEQ ID No. 7 (Bcl2) and SEQ ID No. 9 (MDM2).

27. The CDH of any one of claims 19 to 26, wherein said one or more intracellular signaling domain(s) is selected from the group comprising CD247/ CD3z, CD137/4-1BB, Zap70, STAT3, STAT5, and a derivative or a combination of one or more thereof

28. The CDH of any one of claims 19 to 27, wherein said one or more transmembrane domain is selected from the group comprising of a transmembrane domain of a T cell receptor, a p chain, a chain, CD28, CD3z, CD45, CD4, CD5, CD8, CD9, CD 16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD137, ICOS, CD154, and a GITR, or a derivative or a combination of one or more thereof.

29. The CDH of any one of claims 19 to 28, wherein

30. The CDH of claim 29, wherein the one or more DNA binding domain is selected from the group comprising Gal4, TetR, dCas9, ZFN, and TALEN, or a combination of one or more thereof.

31. The CDH of claim 29 or 30, wherein the one or more transcription factor is selected from the group comprising VP 16, VP64, VPR and p65, or a combination of one or more thereof.

32. A method for designing an activation by inhibitor release sensor (AIR) switch comprising a) selecting a pair “ligand: ligand-binding domain”;

49 b) designing a binder to selectively interact with the ligand-binding domain of step a), wherein the ligand-binding domain is the second ligand-binding domain; c) designing or rendering the first ligand-binding domain insensitive, or less sensitive, to the ligand by modifying active site residues involved in the ligand binding; and optionally d) operably linking the second ligand binding domain to the binder, via a linker, wherein the second ligand binding domain forms an auto-inhibited binding interaction with said binder in the absence of a ligand. operably linking the second ligand binding domain to the binder, via a linker, wherein the second ligand binding domain forms an auto-inhibited binding interaction with said binder in the absence of a ligand.

33. An amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 1 (LD3_vl), SEQ ID No. 2 (LD3_v2), SEQ ID No. 3 (LD3_v3), SEQ ID No. 4 (iBcl-XL_v3), SEQ ID No. 5 (iBcl2_v4), SEQ ID No. 6 (Bcl-XL), SEQ ID No. 7 (Bcl2), SEQ ID No. 8 (LD3), SEQ ID No. 9 (MDM2), SEQ ID No. 10 (LD6).

34. An isolated nucleic acid encoding an amino acid sequence having at least 80% amino acid sequence identity to the amino acid sequence set forth in any of SEQ ID No. 1 (LD3_vl), SEQ ID No. 2 (LD3_v2), SEQ ID No. 3 (LD3_v3), SEQ ID No. 4 (iBcl-XL_v3), SEQ ID No. 5 (iBcl2_v4), SEQ ID No. 6 (Bcl-XL), SEQ ID No. 7 (Bcl2), SEQ ID No. 8 (LD3), SEQ ID No. 9 (MDM2), SEQ ID No. 10 (LD6).

35. An expression vector comprising an isolated nucleic acid of claim 34, operably linked to a transcriptional control element.

36. A host cell comprising an isolated nucleic acid of claim 34 or an expression vector of claim 35.

50

37. A host cell expressing the activation by inhibitor release (AIR) of any one of claims 1 to 18 or the CDH of any one of claims 19 to 31.

38. The host cell of claim 36 or 37, wherein the cell is a mammalian cell.

39. The host cell of any one of claims 36 to 38, wherein the cell is selected from the group comprising a cytotoxic cell, a T cell, a stem cell, a progenitor cell, and a cell derived from a stem cell or a progenitor cell.

40. A method of treating a disease in a subject, the method comprising: i) genetically modifying at least one host cell obtained from the individual with an expression vector comprising one or more nucleotide sequences encoding an activation by inhibitor release (AIR) of anyone of claims 1 to 18, and where the genetic modification is carried out ex vivo; ii) introducing the genetically modified host cell into the subject; and iii) administering to the subject an effective amount of a ligand, wherein the ligand induces dimerization of the heterodimeric, conditionally active AIR, wherein said dimerization induces (ON) a biological activity of the genetically modified host cells.

41. The method of treating of claim 40, wherein the disease is selected from the group comprising a cancer, an inflammatory disease, a genetic disorder, an infectious disease, or a degenerative disease.

51

42. The method of treating a disease of claim 40 or 41, wherein the cell is selected from the group comprising a cytotoxic cell, a T cell, a stem cell, a progenitor cell, and a cell derived from a stem cell or a progenitor cell.

43. A composition comprising an AIR switch of anyone of claims 1 to 18, an amino acid sequence of claim 33, an isolated nucleic acid of claim 34, an expression vector of claim 35, a host cell of anyone of claims 36 to 39, and/or a CDH of anyone of claims 19 to 31.

44. A pharmaceutical composition comprising a therapeutically effective amount of an AIR switch of anyone of claims 1 to 18, an amino acid sequence of claim 33, an isolated nucleic acid of claim 34, an expression vector of claim 35, a host cell of anyone of claims 36 to 39, and/or a CDH of anyone of claims 19 to 31, and a pharmaceutically acceptable carrier or diluent.

45. A kit comprising a composition of claim 43 or a pharmaceutical composition of claim

46. The pharmaceutical composition of claim 44 for use in the treatment and/or prevention of a disease selected from the group comprising a cancer, an inflammatory disease, a genetic disorder, an infectious disease, or a degenerative disease.

52