Inducible epigenetic switches to drive cell fate changes

Abstract

Provided herein is a cell comprising: (a) a polynucleotide encoding a binding-triggered transcriptional switch (BTTS) comprising an intracellular domain comprising a DNA binding domain (DBD) and one or more epigenetic editors and (b) a nucleic acid comprising a binding site for the DBD, a promoter and, operably linked to the promoter, a coding sequence encoding a payload. Binding of the BTTS to an antigen results in decreased expression of the payload via epigenetic modifications. If the payload is a transcriptional repressor, then binding of the BTTS results in increased expression of a second payload that is operably linked to the first payload. These changes result in long-term epigenetic changes in expression of the payloads and can be used to steer cell fate or function changes in response to BTTS detected inputs.

Description

[0001]Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 INDUCIBLE EPIGENETIC SWITCHES TO DRIVE CELL FATE CHANGES CROSS-REFERENCING This application claims the benefit of U.S. provisional application serial no 63 / 679,519, filed on August 5, 2024, which application is incorporated herein in its entirety. STATEMENTREGARDINGFEDERALLYSPONSOREDRESEARCHThis invention was made with government support under grant nos. R01 CA253017 and R01 DA036858 awarded by the National Institutes of Health. The government has certain rights in the invention. INCORPORATION BY REFERENCE OF SEQUENCE LISTING PROVIDED AS A SEQUENCE LISTING XML FILE A Sequence Listing is provided herewith as a Sequence Listing XML, “UCSF-814WO_SEQLIST” created on August 1, 2025 and having a size of 10,717 bytes. The contents of the Sequence Listing XML are incorporated by reference herein in their entirety. INTRODUCTION A cornerstone of multicellular life is the ability of progenitor cells to stably differentiate into specialized cell types. During development, cells undergo a complex series of fate transitions, which are induced and coordinated across the cell population by specific cell-to-cell signals. Here, these are referred to as cell fate transitions because they represent a committed change that is essentially irreversible, barring unnatural or improbable events like reprogramming. Many of the signals that trigger fate transitions are transient, happening at a particular stage of development, yet the induced cell fate changes remain locked in, long after those developmental induction signals are gone. As cell engineering advances toward bottom-up construction of tissues and organs, it is becoming increasingly important to be able to direct similar types of committed cell fate transitions. Specifically, the ability to generate multiple alternative stable cell types from a common progenitor cell, and to do so in a way in which such fates are irreversible, would be critical to program more complex multi-cellular structures and Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 functions. Thus, a foundational goal for engineering cooperative multicellular systems is to develop synthetic gene circuits that can drive stable cell fate transitions. SUMMARY The technology described herein provides a way to permanently switch the fate (i.e., with a memory) of a cell based on external input signals that the cell experiences. That is to say, one can custom engineer cell fate transitions in a way that is similar to what happens in natural development, which relies upon a cell experiencing a set of transient input signals, which then results in a stable (or "permanent") change in gene expression, even after the input signals are gone. In the present system, this has been achieved by integrating a set of epigenetic chromatin regulators into a binding-triggered transcriptional switch that, when activated, releases the epigenetic chromatin regulators into the nucleus which, in turn, lead to stable DNA methylation of a target gene and resulting in stable epigenetic silencing (or activation) of a gene of interest. In some embodiments, a cell is provided that comprises: (a) a polynucleotide encoding a binding-triggered transcriptional switch (BTTS) comprising an intracellular domain comprising a DNA binding domain (DBD) and one or more epigenetic editors and (b) a recombinant nucleic acid comprising a binding site for the DBD, a promoter and, operably linked to the promoter, a coding sequence encoding a payload. Binding of the BTTS to an antigen results in decreased expression of the payload via epigenetic modifications. In some embodiments, the cell may comprise: (a) a first polynucleotide encoding a first binding-triggered transcriptional switch (BTTS), wherein the first BTTS comprises: (i) a first extracellular binding domain that recognizes a first antigen; (ii) a transmembrane domain; (iii) one or more protease cleavage domains; and (iv) a first intracellular domain comprising a first DNA binding domain (DBD) and one or more first epigenetic editors; and (b) a recombinant nucleic acid comprising a binding site for the first DBD, a first promoter and, operably linked to the promoter, a coding sequence encoding a payload. In these embodiments, binding of the first extracellular binding domain to a first antigen results in cleavage of the BTTS at the one or more protease cleavage domains to release the first intracellular domain, and the released first intracellular domain binds to the binding site of for the first DBD and decreases expression of the payload. The one or more first epigenetic editors may comprise a KRAB domain and / or a Dnmt3 subunit. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 In some embodiments, the first intracellular domain may comprise a first DNA binding domain (DBD), KRAB domain and a Dnmt3 subunit. In these embodiments, binding of the cell to the first antigen may decrease long-term, stable, decrease in expression of the payload. In some embodiments, the cell may further comprise: (c) a second polynucleotide encoding a second BTTS, wherein the second BTTS comprises: (i) a second extracellular binding domain that recognizes a second antigen; (ii) a transmembrane domain; (iii) one or more protease cleavage domains; and (iv) a second intracellular domain comprising a second DNA binding domain (DBD) and one or more second epigenetic editors. In these embodiments, the recombinant nucleic acid may further comprises a binding site for second DBD. In these embodiments, wherein binding of the second extracellular binding domain to an antigen results in cleavage of the second BTTS at the one or more protease cleavage domains to release the second intracellular domain which, in turn, wherein binds to the binding site for second DBD further decreases expression of the payload. In some of these embodiments, the first epigenetic editor may be KRAB domain and the second epigenetic editor may be Dnmt3 subunit. In these embodiments, binding of the cell to the first and second antigen may decrease long-term, stable, decrease in expression of the payload. In any embodiment, the payload cell fate regulator. In any embodiment, the payload may be a transcription factor, morphogen, therapeutic protein (cytokine, chemokine or growth factor ) or cell surface ligand / receptor for example. In some embodiments, the payload may be a transcriptional repressor. In these embodiments, the cell may further comprise a second recombinant nucleic acid comprising: a binding site for the transcriptional repressor, a second promoter and a coding sequence for a second payload that is operably linked to the second promoter. In these embodiments, a the decrease in expression of the transcriptional repressor increases expression of the second payload. A variety of methods effecting a long-term change in the expression of a payload via a transient extracellular binding event are also provided. Without wishing to be held to any specific theory, it is noted that during development cells sometimes make highly coordinated stable (irreversible) fate changes in response to extracellular stimuli / signals. In many cases these committed cell fate changes are regulated by epigenetic modifications in chromatin, which result in changes in the expression of specific Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 genes. Such gene expression changes remain, even after removal of the initiating stimulus, and represent a form of cellular memory of past inputs. The present cells provide a way to make a stable change in gene expression upon stimulation with specific input signals. This system makes use of modular receptors with extracellular ligand binding domains, fused to a binding induced protease site, and an intracellular domain that contain either the KRAB domain and / or Dnmt3 regulatory domains fused to DNA binding domains. Upon binding of the extracellular domain to a ligand (input stimulus), the receptors are cleaved, resulting in the release of these intracellular domains such that they can enter the nucleus and participate in epigenetic silencing of a target locus. In some embodiments, stable silencing occurs only when both the KRAB domain and Dnmt3 domains are recruited to a target locus simultaneously. However, in some embodiments (e.g., in embodiments that involve human neural stem cells), D3L alone is sufficient to induce silencing. This switch can be configured so that two combinatorial inputs (co-incident) are required to trigger stable silencing. Also provided is an “inverter switch” that can be used to activate (rather than silence) a target gene of interest with the ability for temporal processing. These tools provide a way to drive a cell along a specified cell fate trajectory in response to a series of user-defined input signals. These epigenetic switch circuits can be used to drive synthetic developmental pathways in regeneration and morphogenesis. For example, they can be used to drive cell therapies into specific phenotypic / functional states, only after they receive a specific signal or enter a specific tissue. There are many applications for this foundational technology in cell engineering. The present system allows one to exert control over target silencing by cell-cell contact rather than inducible promoters exogenously added molecules and, further, allows one to customize the input signals governing permanent changes in gene expression. This makes the present system suitable for, for example, multi-cell engineering protocols. These and other features may become apparent in view of the description that follows below. BRIEFDESCRIPTIONOFTHEDRAWINGSFig. 1A-1B: Cooperative epigenetic silencing by KRAB and Dnmt3 underlies stable cell fate changes. Fig. 1A: A three-state model describing a stepwise cooperation between KRAB Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 domains and Dnmt3 to drive stable transcriptional silencing. KRAB domains, and the multimeric chromatin complexes they co-recruit, can establish a repressive heterochromatin domain to silence transcription in a reversible manner. In combination with Dnmt3, which establishes DNA methylation, the silenced state is locked in as repressive DNA methylation is effectively sustained over time and cell divisions. Fig. 1B: A schematic representing an input signal that permanently switches cells to a new fate regardless of whether the input persists. This scenario corresponds to the ability of Dnmt3 to establish epigenetic memory. Fig. 2A-2C: Design of an inducible epigenetic memory switch through synNotch control of KRAB and Dnmt3L regulation. Fig. 2A: Design of a receiver cell engineered to contain a genomically integrated synthetic circuit that enables the inducible site-specific recruitment of KRAB and D3L domains to a target locus. Fig. 2B: Flow cytometry distributions of receiver cells incubated with magnetic beads as indicated. Cells were stimulated for 3 days and allowed to recover for up to 7 days by removal of beads and in the presence of DAPT. Distributions are representative of three technical replicates. Fig. 2C: Time course of the data in (2B) showing the percent of cells expressing EGFP, as determined by the threshold indicated by the dotted lines in (2B). Data are from three technical replicates. Confidence intervals are too small to be visible. Fig. 3A-3D: Fate bifurcation emerges from the synthetic memory switch. Fig. 3A: Schematic illustrating the a series of synNotch stimuli with varying duration. Fig. 3B: The percent of cells in which EGFP was silenced following the recovery period (9 days), which corresponds to fate commitment, are plotted as a function of the stimulus duration. Data shown are from three technical replicates of a single cell clone. Confidence intervals are too small to be visible. Fig. 3C: Flow cytometry distributions of receiver cells stimulated for varying amounts of time, as indicated. Distributions are representative of three technical replicates of a single cell clone. Fig. 3D: Flow cytometry distributions of receiver cells stimulated for varying amounts of time, as indicated. Distributions are representative of three technical replicates of a single cell clone. Fig. 4A-4C: Coincident KRAB and D3L stimulation is critical for effective fate switching. Fig. 4A: Time courses of the percent of cells expressing EGFP and the corresponding representative flow cytometry distributions for varying stimulation regimes. The pink shaded areas in some of the time course plots indicate the time of coincident KRAB and D3L recruitment. Data for all curves are from three technical replicates of a single cell clone. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Confidence intervals are too small to be visible. Fig. 4B: Model describing the stepwise transition of a promoter from transcriptionally active (ON) to stably silenced (OFFstable). Fig. 4C: KRAB, via H3K9me3 and the heterochromatin it establishes, primarily decreases the energy to transition to a transiently silenced state (OFFtransient). D3L, via cytosine methylation (5mC) and other downstream repressors, primarily decreases the energy to transition to the OFFstable state from the OFFtransient state, but not from the ON state. Fig. 5A-5C: Combinatorial inputs can direct specific transitions from among alternative fate choices. Fig. 5A: Design of a multipotent receiver cell type that can specifically silence either EGFP or mCherry depending on the combination of synNotch input signals. Fig. 5B: Schematic illustrating the possible transitions of the multipotent receiver cell. Cells begin in the top right double-positive quadrant and can choose among three alternative transitions based on specific input combinations. Fig. 5C: Two-dimensional flow cytometry distributions of EGFP and mCherry fluorescence. Receiver cells were co-cultured with different sender cells, as indicated in the INPUT column, for 7 days, and then allowed to recover in the presence of DAPT for 7 days. Numbers are the percentage of cells in each quadrant. Panels under the TRAJECTORY summarize the fate transitions. Distributions are representative of three technical replicates of a single cell clone. Fig. 6A-6C: Engineering a stable gene activation memory switch using an inverter circuit. Fig. 6A: Design of a receiver cell that incorporates an inverter cascade circuit architecture, whereby synNotch silencer receptors control an intermediate repressor, which in turn represses a downstream engineered target gene (mCherry). Fig. 6B: Time courses showing the percent of cells expressing mCherry (top) or CymR-EGFP (bottom) following a transient stimulus (shaded area). Data for all curves are from three technical replicates of a receiver cell line derived from a single cell clone (see Methods). Confidence intervals are too small to be visible. Fig. 6C: Time courses showing the percent of cells expressing mCherry following a transient stimulus (shaded area), and in the presence of varying amounts of the small molecule Sld1. Data for all curves are from three technical replicates. Confidence intervals are too small to be visible. Fig. 7A-7D: Epigenetic switch yields a morphological fate change that is stable over many cell divisions. Fig.7A: Design of a receiver cell that activates P-cadherin (PCAD) expression in the inverter cascade. Note that the cells also constitutively express a synthetic Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 adhesion molecule (synCAM), which has been found to disrupt baseline cell-cell adhesion. This synCAM does not participate in the epigenetic switching circuit that controls P-cadherin and is not illustrated for simplicity. Fig. 7B: Schematic highlighting the outcome of circuit activation in receiver cells upon contact with sender cells. Receiver cells undergo a morphological change as P-cadherin mediates strong homotypic adhesion. Additionally, as the sender cells also express a different cadherin, N-cadherin (NCAD), sender and activated receiver cells can self-organize into distinct spatial domains. Fig. 7C: Confocal microscopy images (maximum intensity projections) for an experiment where senders and receivers were co-cultured in an ultra-low attachment vessel. The cartoon below summarizes the morphological dynamics and highlights how the self-organized dual-spheroid structure formed early cannot be sustained over time. Images are representative of 24 technical replicates of a receiver cell derived from a single cell clone (see Methods). Fig. 7D: Same as Fig. 7C but with senders that activate both synNotch- KRAB and synNotch-D3L receptors. Fig. 8A-8F: Inducible epigenetic memory through synNotch control of KRAB and Dnmt3L regulation, related to Figs. 2A-2C. Fig. 8A: Alternative configurations of the synNotch silencer receptors with different fusion combinations of DNA binding domain and chromatin regulators. Fig. 8B: Flow cytometry distributions of receiver cells co-cultured with ligand- presenting sender cells as indicated. Cells were co-cultured for 4 days and allowed to recover for up to 7 days in the presence of the synNotch inhibitor DAPT. Distributions are representative of three technical replicates. Fig. 8C: Time course of the data in Fig. 8B showing the percent of cells expressing EGFP, as determined by the threshold indicated by the dotted lines in (8B). Data are from three technical replicates. Shaded area under the D3L curve is the 95% confidence interval (CI). Confidence intervals are too small to be visible. Fig. 8D: Time course of the percent of cells expressing EGFP when they were initially incubated with anti-myc and anti-HA beads to stably silence EGFP, followed by treatment with 5-aza or vehicle. Data are from three technical replicates. Confidence intervals are too small to be visible. Fig. 8E: Same as Fig. 8B, but with alternative co-culture experiments. Fig. 8F: Same as Fig. 8B, but with receiver cells harboring different synNotch silencer configurations. Figs 9A-9D: An epigenetic inverter circuit allows permanent transdifferentiation of mouse embryonic fibroblasts into neuronal cells. Fig. 9A: Schematics of the transdifferentiation process induced by a transient pulse of a synNotch ligand activating the epigenetic memory Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 circuit. Fig. 9B: Illustration of the epigenetic inverter circuit permanently activating the BAM transcription factors upon priming of synNotch.KRAB / D3L. Fig. 9C: Histogram showing the absolute count of TUJ1-positive neuronal cells obtained by exposing mouse C3H fibroblasts to the indicated stimuli of the epigenetic inverter BAM circuit (n=2). Fig. 9D: Representative immunofluorescence microscopy images of TUJ1-positive neuronal cells obtained by exposing mouse C3H fibroblasts to the indicated stimuli of the epigenetic inverter BAM circuit. BAM: Brn2, Ascl1 and Myt1l. Figs 10A-10F SynNotch-dependent epigenetic memory can be instructed in human IPSC-derived Neural stem / progenitor cells (NSPCs) and is inherited throughout their differentiation into mature neurons. Fig. 10A: Schematics of the experimental strategy to assess induction and inheritance of epigenetic silencing in NSPCs by transient synNotch stimulation; Fig. 10B: Illustration of the epigenetic circuit permanently silencing a phPGK.GFP reporter in response to priming of synNotch.KRAB / D3L. Fig. 10C: Graphs showing the kinetics of phPGK.GFP silencing upon transient priming of synNotch.KRAB / D3L in human IPSC-derived Neural stem / progenitor cells (NSPCs). n=3; Fig. 10D: Schematics of the experimental strategy to assess the inheritance of synNotch-based epigenetic silencing throughout the differentiation of NSPCs into mature neurons. Fig. 10E: Representative immunofluorescence microscopy images of neural stem cells cultured for seven days in neural stem cell media (Top) or in mature neural differentiation media; Fig. 10F: Histogram showing the percentage of cells expressing the phPGK.GFP reporter upon exposure to the indicated synNotch stimulations and culture media. n=3. SN.K: synNotch.KRAB; SN.D3L: synNotch.D3L. NFH: neurofilamentin H. DEFINITIONSAs used herein, the terms "treatment," "treating," “treat” and the like, refer to obtaining a desired pharmacologic and / or physiologic effect and / or a response related to the treatment. The effect can be prophylactic in terms of completely or partially preventing a disease or symptom thereof and / or can be therapeutic in terms of a partial or complete cure for a disease and / or adverse effect attributable to the disease. "Treatment," as used herein, covers any treatment of a disease in a mammal, particularly in a human, and includes: (a) preventing the disease from occurring in a subject which can be predisposed to the disease but has not yet been diagnosed as Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 having it; (b) inhibiting the disease, i.e., arresting its development; and (c) relieving the disease, i.e., causing regression of the disease. A “therapeutically effective amount” or “efficacious amount” refers to the amount of an agent (including biologic agents, such as cells), or combined amounts of two agents, that, when administered to a mammal or other subject for treating a disease, is sufficient to effect such treatment for the disease. The “therapeutically effective amount” will vary depending on the agent(s), the disease and its severity and the age, weight, etc., of the subject to be treated. The terms “individual,” “subject,” “host,” and “patient,” used interchangeably herein, refer to a mammal, including, but not limited to, murines (e.g., rats, mice), non-human primates, humans, canines, felines, ungulates (e.g., equines, bovines, ovines, porcines, caprines), lagomorphs, etc. In some cases, the individual is a human. In some cases, the individual is a non-human primate. In some cases, the individual is a rodent, e.g., a rat or a mouse. In some cases, the individual is a lagomorph, e.g., a rabbit. As used herein, the term “binding-triggered transcriptional switch” or “BTTS” refers to any polypeptide or complex of the same that is capable of transducing a specific binding event on the outside of the cell (e.g., binding of an extracellular domain of the BTTS) to a signal on the inside of the cell. Many BTTSs work by releasing the intracellular domain, which travels into the nucleus. In these embodiments, the BTTS is made up of one or more polypeptides that undergo proteolytic cleavage upon binding to the cognate ligand to release a gene expression regulator. For example, a BTTS may comprise (i) an extracellular domain comprising the antigen binding region of an antigen-specific antibody; (ii) a proteolytically cleavable sequence comprising one or more proteolytic cleavage sites; and (iii) an intracellular domain, wherein binding of the antigen binding region to the antigen induces cleavage of the sequence at the one or more proteolytic cleavage sites, thereby releasing the intracellular domain. A BTTS can be based on synNotch, A2, MESA, or force receptor, for example, although others are known or could be constructed. "Single-chain Fv" or "sFv" antibody fragments comprise the VH and VL domains of antibody, wherein these domains are present in a single polypeptide chain. In some embodiments, the Fv polypeptide further comprises a polypeptide linker between the VH and VL domains, which enables the sFv to form the desired structure for antigen binding. For a Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 review of sFv, see Pluckthun in The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds., Springer-Verlag, New York, pp. 269-315 (1994). The term "nanobody" (Nb), as used herein, refers to the smallest antigen binding fragment or single variable domain (VHH) derived from naturally occurring heavy chain antibody and is known to the person skilled in the art. They are derived from heavy chain only antibodies, seen in camelids (Hamers-Casterman et al. (1993) Nature 363:446; Desmyter et al. (2015) Curr. Opin. Struct. Biol. 32:1). In the family of "camelids" immunoglobulins devoid of light polypeptide chains are found. "Camelids" comprise old world camelids (Camelus bactrianus and Camelus dromedarius) and new world camelids (for example, Llama paccos, Llama glama, Llama guanicoe and Llama vicugna). A single variable domain heavy chain antibody is referred to herein as a nanobody or a VHHantibody. The terms “synthetic”, “chimeric” and “engineered” as used herein generally refer to artificially derived polypeptides or polypeptide encoding nucleic acids that are not naturally occurring. Synthetic polypeptides and / or nucleic acids may be assembled de novo from basic subunits including, e.g., single amino acids, single nucleotides, etc., or may be derived from pre- existing polypeptides or polynucleotides, whether naturally or artificially derived, e.g., as through recombinant methods. Chimeric and engineered polypeptides or polypeptide encoding nucleic acids will generally be constructed by the combination, joining or fusing of two or more different polypeptides or polypeptide encoding nucleic acids or polypeptide domains or polypeptide domain encoding nucleic acids. Chimeric and engineered polypeptides or polypeptide encoding nucleic acids include where two or more polypeptide or nucleic acid “parts” that are joined are derived from different proteins (or nucleic acids that encode different proteins) as well as where the joined parts include different regions of the same protein (or nucleic acid encoding a protein) but the parts are joined in a way that does not occur naturally. The term "recombinant", as used herein describes a nucleic acid molecule, e.g., a polynucleotide of genomic, cDNA, viral, semisynthetic, and / or synthetic origin, which, by virtue of its origin or manipulation, is not associated with all or a portion of the polynucleotide sequences with which it is associated in nature. The term recombinant as used with respect to a protein or polypeptide means a polypeptide produced by expression from a recombinant polynucleotide. The term recombinant as used with respect to a host cell or a virus means a host cell or virus into which a recombinant polynucleotide has been introduced. Recombinant is also Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 used herein to refer to, with reference to material (e.g., a cell, a nucleic acid, a protein, or a vector) that the material has been modified by the introduction of a heterologous material (e.g., a cell, a nucleic acid, a protein, or a vector). The term “operably linked” refers to a juxtaposition wherein the components so described are in a relationship permitting them to function in their intended manner. For instance, a promoter is operably linked to a coding sequence if the promoter affects its transcription or expression. Operably linked nucleic acid sequences may but need not necessarily be adjacent. For example, in some instances a coding sequence operably linked to a promoter may be adjacent to the promoter. In some instances, a coding sequence operably linked to a promoter may be separated by one or more intervening sequences, including coding and non-coding sequences. Also, in some instances, more than two sequences may be operably linked including but not limited to e.g., where two or more coding sequences are operably linked to a single promoter. The terms “polynucleotide” and “nucleic acid,” used interchangeably herein, refer to a polymeric form of nucleotides of any length, either ribonucleotides or deoxyribonucleotides. Thus, this term includes, but is not limited to, single-, double-, or multi-stranded DNA or RNA, genomic DNA, cDNA, DNA-RNA hybrids, or a polymer comprising purine and pyrimidine bases or other natural, chemically or biochemically modified, non-natural, or derivatized nucleotide bases. The terms “polypeptide,” “peptide,” and “protein”, used interchangeably herein, refer to a polymeric form of amino acids of any length, which can include genetically coded and non- genetically coded amino acids, chemically or biochemically modified or derivatized amino acids, and polypeptides having modified peptide backbones. The term includes fusion proteins, including, but not limited to, fusion proteins with a heterologous amino acid sequence, fusions with heterologous and homologous leader sequences, with or without N-terminal methionine residues; immunologically tagged proteins; and the like. A "vector" or "expression vector" is a replicon, such as plasmid, phage, virus, or cosmid, to which another DNA segment, i.e. an "insert", may be attached so as to bring about the replication of the attached segment in a cell. The term “heterologous”, as used herein, means a nucleotide or polypeptide sequence that is not found in the native (e.g., naturally-occurring) nucleic acid or protein, respectively. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Heterologous nucleic acids or polypeptide may be derived from a different species as the organism or cell within which the nucleic acid or polypeptide is present or is expressed. Accordingly, a heterologous nucleic acids or polypeptide is generally of unlike evolutionary origin as compared to the cell or organism in which it resides. DETAILEDDESCRIPTIONAs summarized above, the present disclosure provides a cell comprising: (a) a polynucleotide encoding a binding-triggered transcriptional switch (BTTS) comprising an intracellular domain comprising a DNA binding domain (DBD) and one or more epigenetic editors and (b) a nucleic acid comprising a binding site for the DBD, a promoter and, operably linked to the promoter, a coding sequence encoding a payload. Binding of the BTTS to an antigen results in decreased expression of the payload via epigenetic modifications. Such cells and methods of their use are described below. Cells In some embodiments, the cell may comprise: (a) a first polynucleotide encoding a first binding-triggered transcriptional switch (BTTS), wherein the first BTTS comprises: (i) a first extracellular binding domain that recognizes a first antigen; (ii) a transmembrane domain; (iii) one or more protease cleavage domains; and (iv) a first intracellular domain comprising a first DNA binding domain (DBD) and one or more first epigenetic editors; and (b) a nucleic acid comprising a binding site for the first DBD, a first promoter and, operably linked to the promoter, a coding sequence encoding a payload. In this cell, binding of the first extracellular binding domain to a first antigen results in cleavage of the BTTS at the one or more protease cleavage domains to release the first intracellular domain, and wherein the released first intracellular domain binds to the binding site of for the first DBD and decreases expression of the payload. In any embodiment, the one or more first epigenetic editors may comprise a KRAB domain and / or a Dnmt3 subunit, where the term “KRAB domain” is intended to refer to the KRAB domain of any of the family of Kruppel-associated box containing zinc finger proteins several members of which are described in (Huntley, S. et al. (2006) Genome Res. 16: 669-77). These transcription factors bind to specific sequences through their ZFP DNA binding domain, and they are believed to recruit the KRAB Associated Protein 1 (KAP1) with their conserved Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 KRAB domain (otherwise known as a Kruppel-associated box). KAP1 in turn binds a large number of effectors that promote the local formation of repressive chromatin (Iyengar, S. et al. (2011) J. Biol. Chem. 286: 26267-76). The KRAB domain, which is approximately 40 amino acids in length is conserved in this family (see, e.g., Genbank’s conserved protein domain family entry number. An ATR of the present invention may, for example, comprise a KRAB domain. Various KRAB domains are known in the family of KRAB-ZFP proteins. For example, an ATR of the present invention may comprise the KRAB domain of ZNF10, 350, 197, RBAK, ZKSCAN1, KRBOX4, ZNF274, or Kox1, for example. Exemplary sequences for KRAB domains that can be found in WO2016063264A1. The KRAB domain used recruits KAP1 and other effector proteins. The term “Dnmt3 subunit”, refers to domain of DNMT3L, DNMT3A or DNMT3B (which are subunits of the Dnmt3), several examples and orthologs of which are known. A DNMT3A subunit may comprise at least 100 amino acids (e.g., at least 100, at least 150 or at least 200 amino acids) that has at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 99% or 100% identity to the wild type human DNMT3A subunit, an isoform of which is shown below: TYGLLRRREDWPSRLQMFFANNHDQEFDPPKVYPPVPAEKRKPIRVLSLFDGIATGLLVL KDLGIQVDRYIASEVCEDSITVGMVRHQGKIMYVGDVRSVTQKHIQEWGPFDLVIGGSP CNDLSIVNPARKGLYEGTGRLFFEFYRLLHDARPKEGDDRPFFWLFENWAGVSDKRDIS RFLESNPVMIDAKEVSAAHRARYFWGNLPGMNRPLASTVNDKLELQECLEHGRIAKFS KVRTITTRSNSIKQGKDQHFPVFMNEKEDILWCTEMERVFGFPVHYTDVSN SRLARQRLLGRSWSVPVIRHLFAPLKEYFACV (SEQ ID NO:8). A DNMT3B subunit may comprise at least 100 amino acids (e.g., at least 100, at least 150 or at least 200 amino acids) that has at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 99% or 100% identity to the wild type human DNMT3B subunit, an isoform of which is shown below: CHGVLRRRKDWNVRLQAFFTSDTGLEYEAPKLYPAIPAARRRPIRVLSLFDGIATGYLVL KELGIKVGKYVASEVCEESIAVGTVKHEGNIKYVNDVRNITKKNIEEWGPFDLVIGGSPC NDLSNVNPARKGLYEGTGRLFFEFYHLLNYSRPKEGDDRPFFWMFENWAKVGDKRDIS RFLECNPVIDAIKVSAAHRARYFWGNLPGMNRPVIASKNDKLELQDCLEYNRIAKLKK Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 VQTITTKSNSIKQGKNQLFPWMNGKEDVLWCTELERIFGFPVHYTDVSNMGRGARQKL LGRSWSVPVIRHLFAPLKDYFACE (SEQ ID NO:9). A DNMT3L subunit may comprise at least 100 amino acids (e.g., at least 100, at least 150 or at least 200 amino acids) that has at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 99% or 100% identity to the wild type human DNMT3L subunit, an isoform of which is shown below: CHGVLRRRKDWNVRLQAFFTSDTGLEYEAPKLYPAIPAARRRPIRVLSLFDGIATGYMA AIPALDPEAEPSMDVILVGSSELSSSVSPGTGRDLIAYEVKANQRNIEDICICCGSLQVHT QHPLFEGGICAPCKDKFLDALFLYDDDGYQSYCSICCSGETLLICGNPDCTRCYCFECVD SLVGPGTSGKVHAMSNWVCYLCLPSSRSGLLQRRRKWRSQLKAFYDRESENPLEMFET VPVWRRQPVRVLSLFEDIKKELTSLGFLESGSDPGQLKHWDVTDTVRKDVEEWGPFDL VYGATPPLGHTCDRPPSWYLFQFHRLLQYARPKPGSPRPFFWMFVDNLVLNKEDLDVA SRFLEEPVTIPDVHGGSLQNAVRVWSNIPAIRSRHWALVSEEELSLLAQNKQSSKLAAKW PTKLVKNCFLPLREYFKYFSTELTSSL (SEQ ID NO:10). The crystal structures of these proteins in the complex are known. In some embodiments, the first intracellular domain may comprising a first DNA binding domain (DBD), a KRAB domain and a Dnmt3 subunit. The present fusion protein is believed to recruit the complex. In some embodiments, the cell may further comprise: (c) a second polynucleotide encoding a second BTTS, wherein the second BTTS comprises: (i) a second extracellular binding domain that recognizes a second antigen; (ii) a transmembrane domain; (iii) one or more protease cleavage domains; and (iv) a second intracellular domain comprising a second DNA binding domain (DBD) and one or more second epigenetic editors. In this cell, wherein the nucleic acid of (b) may further comprises a binding site for second DBD, binding of the second extracellular binding domain to an antigen results in cleavage of the second BTTS at the one or more protease cleavage domains to release the second intracellular domain, and binding of the released second intracellular domain to the binding site for second DBD further decreases expression of the payload. In the cell, the first and second DNA binding domains may be independently selected from Gal4-, LexA-, Tet-, Lac-, dCas9-, zinc-finger- and TALE-based DNA binding domains, although others are known and can be used. TALE- and CRISPR / dCas9-based transcription Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 factors are described in Lebar (Methods Mol Biol.20181772: 191-203). Examples of transcriptional activators that can be part of the fusion protein are numerous and include artificial transcription factors (ATFs) such as, e.g., Zinc-finger-based artificial transcription factors (including e.g., those described in Sera T. Adv Drug Deliv Rev. 200961(7-8):513-26; Collins et al. Curr Opin Biotechnol. 200314(4):371-8; Onori et al. BMC Mol Biol. 201314:3. In some cases, the transcriptional activator may contain a GAL4 DNA binding domain, which binds to the Gal4 responsive UAS, which has been well characterized in the art. Others are well known. In any embodiment, the first and second epigenetic editors are independently selected from a KRAB domain and a DNMT3 subunit. In some embodiments, the first epigenetic editor is a KRAB domain and the second epigenetic editor is a Dnmt3 subunit. The antigens recognized by the first and second extracellular binding domains can be different. These antigens can be independently on another cell, in the extracellular matrix or on a biocompatible material. Depending on the binding domain of the BTTSs (which may contain an antibody sequence (e.g., a scFv or nanobody), a receptor for a ligand, or a ligand for a receptor, etc. the antigen itself can be a protein that is recognized by an antibody, a ligand for the receptor or a receptor for the ligand, etc. In other words, the signaling events are not limited to antibody- antigen interactions. The nucleic acid of (b), i.e., the nucleic acid that comprises the binding site for the first DBD, etc. can be recombinant or, if the DNB is engineered to bind to a sequence that is endogenous to the genome of the cell, the nucleic acid may be endogenous to endogenous to the cell. In any embodiments, a binding site for a DBP may be knocked into an endogenous sequence, e.g., within 1 kb, within 500 nt, or within 200 nt or of an endogenous promoter (e.g., a promoter that drives the expression of a payload that is endogenous to the cell), thereby allowing one to control the expression of payloads that are endogenous to the cell. s The identity of the payload may vary depending on how the cell is going to be used. In some embodiments, the payload may be cell fate regulator, e.g., a morphogen, one or more transcription factors that induce cell differentiation, transdifferentiation or reprogramming (full or partial), or a cell surface ligand that induces differentiation of another cell. The payload may Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 also be a therapeutically relevant protein such as a cytokine, chemokine, growth factor, endogenous or artificial receptor (e.g. TCR or CAR), for example. Morphogens are signaling molecules that emanate from a restricted region of a tissue and spread away from their source to form a concentration gradient. As the fate of each cell in the field depends on the concentration of the morphogen signal, the gradient prefigures the pattern of development. Mammalian morphogens include Hh (Hedgehog), Wnt, TGF-β (Transforming growth factor β) and BMP (Bone morphogenetic protein), although others are known. In some embodiments, the morphogen may be a member of the Wnt / beta-catenin, a member of the TGF-β (Transforming growth factor β) / BMP (Bone morphogenetic protein) family, a member of the Shh (Sonic hedgehog) family, or retinoic acid (or an analog thereof). Examples of morphogens are listed in Table 1 below: 1. Wnt family (Wnt, Wnt2a, Wnt2b, Wnt3a, Wnt5a, Wnt10 Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 19. Insulin , iation or reprogramming (full or partial),have been described in a variety of publications including: (1) Oh, Yujeong, and Jiwon Jang. "Directed differentiation of pluripotent stem cells by transcription factors." Molecules and cells 42.3 (2019): 200; (2) Joung, Julia, et al. "A transcription factor atlas of directed differentiation." Cell 186.1 (2023): 209-229; (3) Smela, Merrick D. Pierson, et al. "Directed differentiation of human iPSCs to functional ovarian granulosa-like cells via transcription factor overexpression." Elife 12 (2023): e83291 ; (4) Yamamizu, Kohei, et al. "Identification of transcription factors for lineage-specific ESC differentiation." Stem cell reports 1.6 (2013): 545-559; and (5) Van Haute, Lindsey, et al. "Generation of lung epithelial- like tissue from human embryonic stem cells." Respiratory research 10 (2009): 1-13. Suitable cells (iPSCs or ESC etc.) that differentiate into the cells listed in the following table by expression of the listed transcription factor can be selected from the literature. In some cases, the following transcription factors can be combined to make a particular cell type. For example, skeletal muscle can potentially be made by expressing MYOD1 BAF60C2 and JMJD3 alone or any combination, in a suitable cell type. TRANSCRIPTION FACTOR CELL / TISSUE TYPE Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 LMX1A Dopaminergic neuron ATOH1 Dopaminergic neuron Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 MSGN1 Smooth muscle ZBTB7B Smooth muscle Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 DMRT1 Neuronal SOX9 Neuronal For example, the following cell fate regulators can be used individually or in combination in the following cells: hPSCs (ESC or iPSC): neural differentiation via combination of Brn2+Ascl1+Myt1l; NeuroD1; Ngn1; Ngn2 hPSC: cardiomyocyte differentiation: GATA4, MEF2C, Nkx2-5, Activin A, BMP4 hPSC: neural tube / floor plate: SHH hPSC: gastrula: BMP4; Nodal; WNT3A hPSC microglia: combination of MAFB, CEPBa, IRF8, PU.1, CEBPb, IRF5 Pancreatic exocrine cells, hepatocytes, or hPSCs to insulin-producing cells: PDX1, NEUROG3, MAFA hMSC (mesenchymal stem cell): myogenic differentiation (myod1); osteogenic differentiation (bmp2, runx2); chondrogenic differentiation (TGFb3, SOX9). The following transcription factors may be used in wound repair: Transcription Factor Function / Activity Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 MYC Wound repair / re-epithelialization E2F-1 Wound repair / re-epithelialization nd repair. Annu. Rev. Cell Dev. Biol. 23 (2007): 69-92. If more than one payload is expressed, they can be expressed using separate expression cassettes (all of which may be activated by the BTTS), or a single expression cassette. In the latter embodiment, the coding sequences may be separated by an IRES or a protease cleavage site may be used. In some embodiments, the cell may additionally secrete a growth factor, e.g., VEGF (Vascular endothelial growth factor), EGF (Epidermal growth factor), FGF (Fibroblast growth factor), TNF-α (Tumor necrosis factor α) or PDGF (Platelet derived growth factor), or a synthetic form thereof, if desired. In any embodiment, the payload may be a transcription factor, a morphogen, therapeutic protein or cell surface receptor. In some embodiments, the payload may be a transcriptional repressor. In these embodiments, the cell may comprise a second nucleic acid comprising: a binding site for the transcriptional repressor, a second promoter and a coding sequence for a second payload that is operably linked to the second promoter, wherein the decrease in expression of the transcriptional repressor increases expression of the second payload. In these embodiments, the second payload may be a cell fate regulator (see the examples above). The second payload may be a Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 transcription factor, morphogen, therapeutic protein or cell surface receptor. Such repressors may be adapted from the TetR, LacI, LexA, RNAi, CRISPRi systems, although several are known. In any embodiment, the cell may be in vitro. However, and ex vivo and in vivo embodiments are also envisioned. In some embodiments, the cell may be a pluripotent stem cell such as embryonic stem cells (ESCs) or induced pluripotent stem cells (iPSCs). In other embodiments, the cell may be adult stem cells (ASCs). Embryonic Stem Cells (ESC) may be derived from inner cell mass (ICM) of pre-implantation blastocyst-stage embryos. These cells can be potentially be used in allogeneic / ‘off-the-shelf’ (donor-derived) cell therapies. In some cases, such cells may be modified for immune compatibility (primarily deletion / disruption of HLA-I and HLA-II classes) (see Kim Stem Cell Reviews and Reports 17 (2021): 1053-1067. Induced Pluripotent Stem Cells (iPSCs) may be derived from at least skin, lung, heart, stomach, brain, liver, blood, kidney or muscle cells, where stemness and pluripotency can be induced by expression of Yamanaka factors (Oct 3 / 4 , Sox2, Klf4, and c-Myc) among others. These cells can generally be used for autologous (self-derived) or allogeneic (donor-derived) cell therapies. See WO 2010 / 017562A2 and Liu, Gele, et al. "Advances in pluripotent stem cells: history, mechanisms, technologies, and applications." Stem cell reviews and reports 16 (2020): 3-32. Adult stem cells include bronchioalveolar stem cells (BASCs), bulge epithelial stem cells (bESCs), corneal epithelial stem cells (CESCs), cardiac stem cells (CSCs), epidermal neural crest stem cells (eNCSCs), embryonic stem cells (ESCs), endothelial progenitor cells (EPCs), intestinal stem cells, hepatic oval cells (HOCs), hematopoetic stem cells (HSCs), keratinocyte stem cells (KSCs), mesenchymal stem cells (MSCs), neuronal stem cells (NSCs), pancreatic stem cells (PSCs), retinal stem cells (RSCs), endothelial stem cells, neural stem cells, olfactory adult stem cells, neural crest stem cells, testicular cells and skin-derived precursors (SKPs). Progenitor cells include endothelial progenitor cells, muscle stem cells, pancreatic progenitor cells, cardiac progenitor cells, ectodermal cells, mesodermal cells, and endodermal cells. A list of exemplary stem cells can be found in the table below: embryonic stem cells Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 neural progenitor cells, neural stem cells hematopoetic progenitor cells (stem cells) In other embodiments, the cell may be a therapeutic immune cell. Suitable mammalian immune cells include primary cells and immortalized cell lines. Suitable mammalian cell lines include human cell lines, non-human primate cell lines, rodent (e.g., mouse, rat) cell lines, and the like. In some instances, the cell is not an immortalized cell line, but is instead a cell (e.g., a primary cell) obtained from an individual. For example, in some cases, the cell is an immune cell, immune cell progenitor or immune stem cell obtained from an individual. As an example, Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 the cell is a lymphoid cell, e.g., a lymphocyte, or progenitor thereof, obtained from an individual. As another example, the cell is a cytotoxic cell, or progenitor thereof, obtained from an individual. In other embodiments, the cell may be a committed differentiated cell type (e.g. a fibroblast), that upon activation of the payload transdifferentiate into another committed cell type or stem cell. Such cells include, e.g., lymphoid cells, i.e., lymphocytes (T cells, B cells, natural killer (NK) cells), and myeloid-derived cells (neutrophil, eosinophil, basophil, monocyte, macrophage, dendritic cells). “T cell” includes all types of immune cells expressing CD3 including T-helper cells (CD4+ cells) and cytotoxic T-cells (CD8+ cells). A “cytotoxic cell” includes CD8+ T cells, natural-killer (NK) cells, and neutrophils, which cells are capable of mediating cytotoxicity responses. This definition includes progenitors of such cells. As would be apparent, a cell of the present disclosure may be generated by any convenient method. Nucleic acids encoding one or more components of a subject circuit may be stably or transiently introduced into the subject immune cell, including where the subject nucleic acids are present only temporarily, maintained extrachromosomally, or integrated into the host genome. Introduction of the subject nucleic acids and / or genetic modification of the subject immune cell can be carried out in vivo, in vitro, or ex vivo. In some cases, the introduction of the subject nucleic acids and / or genetic modification is carried out ex vivo. For example, a primary T lymphocyte, a stem cell, or an NK cell is obtained from an individual; and the cell obtained from the individual is modified to express components of a circuit of the present disclosure. Binding-triggered transcriptional switches (BTTSs) A BTTS is a cleavable fusion protein that contains: i. an extracellular binding domain (e.g., scFv, nanobody, ligand, etc.) that binds to antigen on another cell, in the extracellular matrix, or on a biocompatible matrix; ii. a transmembrane domain, iii. one or more binding- dependent cleavage sites that are cleaved upon binding of the extracellular binding domain to the antigen, and iv. an intracellular domain, where binding of the extracellular binding domain to the antigen induces proteolytic cleavage of the one or more binding-dependent cleavage sites to release the intracellular binding domain and whatever domains it carries. In some embodiments, the BTTS may comprise (a) an extracellular binding domain comprising a protein Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 binding domain (e.g., scFv or nanobody) that binds to a marker on a cell, (b) a force sensing region, (c) a transmembrane domain, (d) one or more force-dependent cleavage sites that are cleaved when the force sensing region is activated, and (e) an intracellular domain, where binding of the binding domain to the antigen induces proteolytic cleavage of the one or more force-dependent cleavage sites to release the intracellular binding domain. Binding of a BTTS to a marker (e.g., an antigen) on the surface of another cell or extracellular matrix (e.g., in vitro or in vivo, in target environment) activates expression decreases expression of the payload. In these embodiments, binding of the binding domain of the BTTS to the antigen induces proteolytic cleavage of the polypeptide (e.g., at one or more force-dependent cleavage sites) to release the intracellular domain. The intracellular domain then binds to a promoter and silences it, thereby reducing the expression of the payload. The general principles SynNotch circuits are described in WO 2016 / 138034, U.S. Patent No. 9,670,281, U.S. Patent No.9,834,608, Roybal et al. Cell (2016) 167(2):419-432, Roybal et al. Cell (2016) 164(4):770-9, and Morsut et al. Cell (2016) 164(4):780-91, among others. In a BTTS, the fusion protein is cleaved to release the intracellular domain when the extracellular domain of the fusion protein engages with an antigen, which may be on another cell. As such, in some cases, the fusion protein may contain a force sensing region (which is typically in the extracellular domain) and one or more force-dependent cleavage sites that are cleaved when the force sensing region is activated. The position of the force-dependent cleavage sites may vary and, in some embodiments, the fusion protein may contain at least two cleavage sites. In some cases, one of the cleavage sites may be extracellular and the other may be in the transmembrane domain or within 10 amino acids of the transmembrane domain in the intracellular domain. In any embodiment, the force sensing region and / or the one or more force- dependent cleavage sites may be from a Delta / Serrate / Lag2 (DSL) superfamily protein, as reviewed by Pintar et al (Biology Direct 20072: 1-13). For example, the force sensing region and / or the one or more force-dependent cleavage sites may be from Notch (see Morsut Cell. 2016164: 780-91), von Willebrand Factor (vWF), amyloid-beta, CD16, CD44, Delta, a cadherin, an ephrin-type receptor or ephrin ligand, a protocadherin, a filamin, a synthetic E cadherin, interleukin-1 receptor type 2 (IL1R2), major prion protein (PrP), a neuregulin or an adhesion-GPCR. Several other examples of this type of protein are known and listed in Pintar et al. Many members of this family appear to share a similar architecture: a region that unfolds and Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 opens up a protease cleavage site (e.g., EGF-like repeats; see Cordle et al Nat. Struct. Mol. Biol. 200815: 849–857), a trans-membrane segment, and a relatively short (~100–150 amino acids) intracellular domain. These sequences permit the binding-triggered release of a transcriptional activator from the membrane in their natural environment and can be readily adapted herein. In some cases, the one or more ligand-inducible proteolytic cleavage sites are selected from S1, S2, and S3 proteolytic cleavage sites. In some cases, the S1 proteolytic cleavage site is a furin-like protease cleavage site comprising the amino acid sequence Arg-X-(Arg / Lys)-Arg, where X is any amino acid. In some cases, the S2 proteolytic cleavage site is an ADAM-17-type protease cleavage site comprising an Ala-Val dipeptide sequence. In some cases, the S3 proteolytic cleavage site is a γ-secretase cleavage site comprising a Gly-Val dipeptide sequence. The S3 proteolytic cleavage site is in the transmembrane domain. In many cases, the shear force generated by binding of the extracellular domain of this fusion protein to another cell unfolds the force sensing region (which, in the case of Notch contains EGF-like repeats whereas in other proteins is made up of other sequences such as the A2 domain in vWF (see, e.g., J Thromb Haemost. 20097:2096-105, Lippok Biophys J. 2016110: 545-54, Lynch Blood. 2014123: 2585-92, Crawley, Blood. 2011118:3212-21 and Xy J Biol Chem. 2013288:6317-24) or a modified A2 domain that has, e.g., the R1597W, E1638K and I1628T substitutions. The architecture of such proteins is described in, e.g., Morsut et al, Cell. 2016164: 780-91, WO2016138034 and WO2019099689, among other places). In some cases, the fusion protein includes an S1 ligand-inducible proteolytic cleavage site. An S1 ligand-inducible proteolytic cleavage site can be located between the HD-N segment and the HD-C segment. In some cases, the S1 ligand-inducible proteolytic cleavage site is a furin-like protease cleavage site. A furin-like protease cleavage site can have the canonical sequence Arg-X-(Arg / Lys)-Arg, where X is any amino acid; the protease cleaves immediately C-terminal to the canonical sequence. For example, in some cases, an amino acid sequence comprising an S1 ligand-inducible proteolytic cleavage site can have the amino acid sequence GRRRRELDPM (SEQ ID NO:1), where cleavage occurs between the “RE” sequence. As another example, an amino acid sequence comprising an S1 ligand-inducible proteolytic cleavage site can have the amino acid sequence RQRRELDPM (SEQ ID NO:2), where cleavage occurs between the “RE” sequence. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 In some cases, the fusion protein polypeptide includes an S2 ligand-inducible proteolytic cleavage site. An S2 ligand-inducible proteolytic cleavage site can be located within the HD-C segment. In some cases, the S2 ligand-inducible proteolytic cleavage site is an ADAM-17-type protease cleavage site. An ADAM-17-type protease cleavage site can comprise an Ala-Val dipeptide sequence, where the enzyme cleaves between the Ala and the Val. For example, in some cases, an amino acid sequence comprising an S2 ligand-inducible proteolytic cleavage site can have the amino acid sequence KIEAVKSE (SEQ ID NO:3), where cleavage occurs between the “AV” sequence. As another example, an amino acid sequence comprising an S2 ligand- inducible proteolytic cleavage site can have the amino acid sequence KIEAVQSE (SEQ ID NO:4), where cleavage occurs between the “AV” sequence. In some cases, the fusion protein includes an S3 ligand-inducible proteolytic cleavage site. An S3 ligand-inducible proteolytic cleavage site can be located within the TM domain. In some cases, the S3 ligand-inducible proteolytic cleavage site is a gamma-secretase (γ-secretase) cleavage site. A γ-secretase cleavage site can comprise a Gly-Val dipeptide sequence, where the enzyme cleaves between the Gly and the Val. For example, in some cases, an S3 ligand- inducible proteolytic cleavage site has the amino acid sequence VGCGVLLS (SEQ ID NO:5), where cleavage occurs at the “GV” sequence. In some cases, an S3 ligand-inducible proteolytic cleavage site comprises the amino acid sequence GCGVLLS (SEQ ID NO:6), where cleavage occurs at the “GV” sequence. In some cases, the fusion protein polypeptide lacks an S1 ligand-inducible proteolytic cleavage site. In some cases, the BTTS lacks an S2 ligand-inducible proteolytic cleavage site. In some cases, the BTTS lacks an S3 ligand-inducible proteolytic cleavage site. In some cases, the BTTS lacks both an S1 ligand-inducible proteolytic cleavage site and an S2 ligand-inducible proteolytic cleavage site. In some cases, the BTTS includes an S3 ligand-inducible proteolytic cleavage site; and lacks both an S1 ligand-inducible proteolytic cleavage site and an S2 ligand- inducible proteolytic cleavage site. In some embodiments, the fusion protein may have an vWF A2 sequence or a variation thereof, an ADAMTS13 cleavage site (which may be described by the consensus sequence HEXXHXXGXXHD (SEQ ID NO:7) ; Crawley, Blood. 2011118:3212-21), and an S3 or γ- secretase cleavage site, although many other arrangements exist. In some embodiments, the Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 switch may contain components that are borrowed from Notch. In other embodiments, the switch may not contain components that are from Notch. For simplicity, BTTSs, including but not limited to chimeric notch receptor polypeptides, are primarily single polypeptide chains. However, BTTSs, including chimeric notch receptor polypeptides, may be divided or split across two or more separate polypeptide chains where the joining of the two or more polypeptide chains to form a functional BTTS, e.g., a chimeric notch receptor polypeptide, may be constitutive or conditionally controlled. For example, constitutive joining of two portions of a split BTTS may be achieved by inserting a constitutive heterodimerization domain between the first and second portions of the split polypeptide such that upon heterodimerization the split portions are functionally joined. Useful BTTSs that may be employed in the subject methods include, but are not limited to, modular extracellular sensor architecture (MESA) polypeptides. A MESA polypeptide comprises: a) a ligand binding domain; b) a transmembrane domain; c) a protease cleavage site; and d) a functional domain. The functional domain can be a transcription regulator (e.g., a transcription activator, a transcription repressor). In some cases, a MESA receptor comprises two polypeptide chains. In some cases, a MESA receptor comprises a single polypeptide chain. Non-limiting examples of MESA polypeptides are described in, e.g., U.S. Patent Publication No. 2014 / 0234851; the disclosure of which is incorporated herein by reference in its entirety. Useful BTTSs that may be employed in the subject methods include, but are not limited to, polypeptides employed in the TANGO assay. The subject TANGO assay employs a TANGO polypeptide that is a heterodimer in which a first polypeptide comprises a tobacco etch virus (Tev) protease and a second polypeptide comprises a Tev proteolytic cleavage site (PCS) fused to a transcription factor. When the two polypeptides are in proximity to one another, which proximity is mediated by a native protein-protein interaction, the protease cleaves the PCS to release the transcription factor. Non-limiting examples of TANGO polypeptides are described in, e.g., Barnea et al. (Proc Natl Acad Sci USA. 2008 Jan. 8; 105(1):64-9); the disclosure of which is incorporated herein by reference in its entirety. Useful BTTSs that may be employed in the subject methods include, but are not limited to von Willebrand Factor (vWF) cleavage domain-based BTTSs, such as but not limited to e.g., those containing an unmodified or modified vWF A2 domain. A subject vWF cleavage domain- based BTTS will generally include: an extracellular domain comprising a first member of a Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 binding pair; a von Willebrand Factor (vWF) cleavage domain comprising a proteolytic cleavage site; a cleavable transmembrane domain and an intracellular domain. Non-limiting examples of vWF cleavage domains and vWF cleavage domain-based BTTSs are described in Langridge & Struhl (Cell (2017) 171(6):1383-1396); the disclosure of which is incorporated herein by reference in its entirety. Useful BTTSs that may be employed in the subject methods include, but are not limited to, chimeric Notch receptor polypeptides, such as but not limited to e.g., synNotch polypeptides, non-limiting examples of which are described in PCT Pub. No. WO 2016 / 138034, U.S. Patent No. 9,670,281, U.S. Patent No.9,834,608, Roybal et al. Cell (2016) 167(2):419-432, Roybal et al. Cell (2016) 164(4):770-9, and Morsut et al. Cell (2016) 164(4):780-91; the disclosures of which are incorporated herein by reference in their entirety. The "SNIPR" switch is another example of a BTTS (Zhu et al 2022 Cell. 185: 1431-1443 and WO2021061856), although others (including the A2, MESA, Tango, ChaCha, RASER, Cal-Light switches) exist and / or can be readily designed. Expression of a BTTS in the cell may be constitutive or inducible, e.g., by binding of another BTTS to an antigen on another cell in the patient. The extracellular binding domain of the BTTS may bind to a tissue-, stromal- or disease- associated marker that is extracellular or expressed on another cell, e.g., in a target environment. In some embodiments, the markers recognized by the first and / or second BTTSs are, independently, selective for an organ, a tissue, cancer, autoimmunity, inflammation, transplant rejection, infection or degeneration, for example. Methods Also provided is a method for effecting a long-term change in the expression of a payload via a transient extracellular binding event. In the context of this method, a "transient binding event” may last up to an hour, a day or a week, whereas a “long-term” change may last at least 2x, at least 5x or at least 10x longer than the binding event (e.g., at least 2 weeks, at least a month or at least a year). In some embodiments, a long-term change may be irreversible or permanent and may last longer than two, three or four rounds of cell division, e.g., for 2 months, 6 months, 1 year, 2 year or the entire lifetime of the cell / organism. These embodiments may comprise transiently contacting a cell with the first antigen. In these embodiments, the first antigen is on a cell, in the extracellular matrix or on a Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 biocompatible material, the first intracellular domain comprises the DBD, a KRAB domain and a Dnmt3 subunit; and the contacting results in a long-term decrease in expression of the first payload. In some cell comprises the second nucleic acid and the contacting results in a long- term increase in expression of the second payload. In some embodiments, a method for effecting a long-term change in the expression of a payload via a transient extracellular binding event is provided, which method comprises transiently contacting a cell with a first and second antigens, where the cell comprises: (a) a first polynucleotide encoding a first binding-triggered transcriptional switch (BTTS), wherein the first BTTS comprises: (i) a first extracellular binding domain that recognizes a first antigen; (ii) a transmembrane domain; (iii) one or more protease cleavage domains; and (iv) a first intracellular domain comprising a first DNA binding domain (DBD) and one or more first epigenetic editors; (b) a nucleic acid comprising a binding site for the first DBD, a first promoter and, operably linked to the promoter, a coding sequence encoding a payload, wherein binding of the first extracellular binding domain to a first antigen results in cleavage of the BTTS at the one or more protease cleavage domains to release the first intracellular domain, and wherein the released first intracellular domain binds to the binding site of for the first DBD, and (c) a second polynucleotide encoding a second BTTS, wherein the second BTTS comprises: (i) a second extracellular binding domain that recognizes a second antigen; (ii) a transmembrane domain; (iii) one or more protease cleavage domains; and (iv) a second intracellular domain comprising a second DNA binding domain (DBD) and one or more second epigenetic editors; wherein the nucleic acid of (b) further comprises a binding site for second DBD, binding of the second extracellular binding domain to an antigen results in cleavage of the second BTTS at the one or more protease cleavage domains to release the second intracellular domain, and wherein the released second intracellular domain binds to the binding site for second DBD. In these embodiments, the first and second antigens are independently on a cell, extracellular matrix or a biocompatible material, the first epigenetic editor is a KRAB domain and the second epigenetic editor is a Dnmt3 subunit, and the contacting results in a long-term decrease in expression of the first payload. In some of these embodiments, the payload is a transcriptional repressor. In these embodiments, the cell may further comprise: (d) a second nucleic acid comprising: a binding site for the transcriptional repressor, a second promoter and a coding sequence for a second Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 payload that is operably linked to the second promoter, wherein the decrease in expression of the transcriptional repressor increases expression of the second payload. In these embodiments, the contacting results in a long-term increase in expression of the second payload. As described above, the first and second antigens may be different. In these embodiments, the first and second antigens may on different cells or the first antigen is on a cell and the second antigen is in the extracellular matrix or a biocompatible material, etc. If the antigens are on cells, they may be endogenous to the cells (i.e., present in wild type cells) or engineered into the cells via recombinant DNA technology. In any embodiment, the contacting may done in vivo (e.g., by administering the cell to a subject), in vitro (using cultured cells) or ex vivo (using cells that have been isolated from a subject, e.g., blood cells or stem cells). In some embodiments, transient contacting the cell with the antigen(s) may results in differentiation of the cell from one cell type to another. The present cell and method may be particularly useful for regenerative medicines, particularly those in which one wishes to to induce differentiation of a cell at a particular time and place, and the differentiation should be long term (e.g., permanent). For example, where one wishes to instruct a stem cell to become a neuron only at the right place and time, i.e., in the brain. In addition, the present system may be incorporated into a therapeutic immune cell to permanently change its state (e.g., to an effector cytotoxic state) only after it experienced a transient priming signal (e.g., a cancer- or tissue- specific marker or the like). Likewise, the system could also be used to induce fate bifurcation - where a cell population would bifurcate into two stable differentiated fates or between a stem and an effector state in response to a specific input signal. The technology may be used to make specific permanent changes in gene expression based on diverse environmental input signals. This technology may also be used to induce the expression of transcription factors or other payloads rejuvenating a committed differentiated or exhausted cell into a younger (or stem) cell. EXAMPLES The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the present invention, and are not intended to limit the scope of what the inventors regard as their invention nor are they intended to represent that the experiments below are all or the only experiments performed. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Efforts have been made to ensure accuracy with respect to numbers used (e.g. amounts, temperature, etc.) but some experimental errors and deviations should be accounted for. Unless indicated otherwise, parts are parts by weight, molecular weight is weight average molecular weight, temperature is in degrees Celsius, and pressure is at or near atmospheric. Standard abbreviations may be used, e.g., bp, base pair(s); kb, kilobase(s); pl, picoliter(s); s or sec, second(s); min, minute(s); h or hr, hour(s); aa, amino acid(s); kb, kilobase(s); bp, base pair(s); nt, nucleotide(s); i.m., intramuscular(ly); i.p., intraperitoneal(ly); s.c., subcutaneous(ly); and the like. MATERIALS AND METHODS Cell lines All cells were handled in sterile safety cabinets, and cell densities were determined using the Countess 3 automated cell counter (Thermo Fisher). Mouse L929 cells (male) were cultured in Dulbecco’s Modified Eagle’s Medium (Thermo Fisher) with 10% fetal bovine serum (UCSF) at 37 °C with 5% CO2 in humidified incubators. They were maintained at a confluence approximately between 10% and 90%. For routine passaging, cells were lifted using TrypLE Express (Thermo Fisher). However, for modified L929 cells expressing the LaG16-LFA1β2 gene, which diminishes cell adhesion, cells were simply resuspended by vigorous pipetting as trypsinization is not required. Lenti-X HEK293T cells (female) were treated similar to L929 cells. All cell lines were used for experiments before they reached 20 passages, which were counted starting from the time the cells were frozen down to make stocks. Study Design The goal of this work was to demonstrate how modular genetic circuits incorporating chromatin regulators could be deployed for the engineering of cell fate transitions and commitment. varying designs were iterated through to explore different features of the cell circuits, including DNA binding domains and synNotch receptors. Cell and molecular biology experiments were used to assess synthetic circuit performance, including flow cytometry and fluorescence microscopy. Results shown represent at least three replicates, as indicated in figure legends. This study was not blinded. Plasmid design and construction Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 SynNotch receptors harboring the chromatin regulators were constructed by iterating through the various domains. First, all synNotch variants were constitutively expressed by the human Elongation Factor 1-alpha promoter (EF-1α). All receptors contain an N-terminal CD8α signal peptide for membrane targeting and c-Myc, HA, or 3xFlag epitope tags. The epitope tags served a dual purpose for determination of surface expression and for receptor stimulation with magnetic beads. Recognition domains consisted of a the anti-CD19 scFv, anti-GFP LaG16 nanobody, or the anti-HER2 scFv clone 4D5-8. All receptors contained the mouse Notch1 minimal regulatory region (Ile1427 to Arg1752). The intracellular domains consisted of the yeast Gal4 DNA binding domain, the TetR DNA binding domain, or the LexA DNA binding domain, the latter of which also included the c-Myc nuclear localization sequence (NLS) on the N-terminal end. Lastly, the C-terminal end of the receptors consisted of the Kox1 KRAB domain, the mouse full-length Dnmt3L, or a KRAB-Dnmt3L fusion where the domains are separated by a linker. All chromatin regulator components were also separated from the DNA binding domain by a short linker. The target loci of the chromatin regulators were cloned into separate vectors. At the 5’ end, the constructs consisted of seven copies of the tetracycline responsive element, followed by eight copies of the ZFHD1 binding sites (present here only as a spacer), and lastly five copies of the Gal4 DNA binding domain target sequence. An alternative construct used for the multi-state receiver cell (Figure 5) consisted of six copies of the LexA DNA binding domain target sequence followed directly with the same five copies of the Gal4 DNA binding domain target sequence as above. All binding site arrays were followed by the constitutively active SFFV promoter, which then drove expression of various transgenes, including EGFP, mCherry, or the CymR DNA binding domain. Note at the CymR domain was directly fused to EGFP and two copies of the SV40 NLS. All targets were in turn fused at the C-terminal end to an FKBP12 destabilization domain that is regulated by ligand, Shield1. The target loci of CymR were also cloned into separate vectors with redesigned promoter regions. The 5’ end consisted of the same SFFV promoter, but without the upstream arrays of binding sites. Instead, the SFFV promoter sequence was followed immediately downstream by four copies of the CymR target sequence. This promoter construct was followed by mCherry fused to a PEST degron, or to a self-cleaving P2A peptide followed immediately by the wild-type mouse P-cadherin. These vectors also included a separate downstream Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 transcription unit consisting of the mouse phosphoglycerate kinase (PGK) promoter constitutively driving expression of TagBFP. This constitutive TagBFP served the dual purpose as first a selectable marker for FACS sorting, and second to track the localization of receiver cells in microscopy experiments. Other constitutively expressed genes were also cloned into separate vectors. SynNotch ligands were constructed by fusing the CD19 extracellular domain, non-fluorescent GFP mutant protein, or the Her2 extracellular domain to the Intercellular Adhesion Molecule 1 (ICAM1) intracellular domain (ICD). Similarly, a LaG16 anti-GFP nanobody was fused to the beta2 integrin ICD from the Lymphocyte function-associated antigen-1 (LFA-1) gene. Ligand constructs also harbored an N-terminal signal peptide for membrane targeting and either Flag or HA epitope tags for determination of surface expression via antibody staining. Ligands were constitutively expressed by the SFFV promoter. Lastly, iRFP670 constructs included the EF-1α promoter to drive constitutive expression. Mouse N-cadherin was expressed together with iRFP670 in the form of a NCAD-P2A-iRFP670 construct. All of the constructs were cloned into a pHR’SIN vector via In-Fusion (Takara Bio). Inserts were PCR amplified using either the KAPA HiFi DNA Polymerase (Roche) or the Phusion High-Fidelity DNA Polymerase (New England Biolabs) with primers containing complementary ends for In-Fusion assembly. Vectors were linearized by restriction digests. Inserts and vectors were extracted from agarose gels using the Zymoclean Gel DNA Recovery Kit (Zymo Research) and used for In-Fusion reactions. Stbl3 chemically competent E. coli were transformed with In-Fusion products and plated on agar media with carbenicillin. Plasmids were extracted using the QIAprep Spin Miniprep Kit (Qiagen) and sequence-verified by either RF Biotech or MCLab companies. Lentiviral transduction To engineer cell lines with stably integrated constructs, L929s were transduced with lentiviral particles in various combinations and sequences. To generate lentiviruses, Lenti-X HEK293T cells (Takara Bio) were plated in 6-well plates, and the next day, co-transfected the cells with the transfer plasmids plus two 2ndgeneration packaging plasmids, pMD2.G and pCMVdR8.91. The Fugene 6 HD transfection reagent (Promega) or the TransIT-VirusGEN transfection reagent (Mirus Bio) were used to transfect the cells. Viral supernatants were collected 2 days after transfection and used immediately to transduce target L929s cells. Most Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 cell lines contained multiple integrated transgenes, and these were delivered either simultaneously by infection with multiple independently generated lentiviral vectors, or as a sequence of multiple rounds of transduction. Target cells were first plated in 24-well plates, and the next day, infected with freshly prepared lentiviral supernatants as described above. Typically, cells were exposed to a range of volumes of viral supernatant, from as low as 1 µL and up to 400 µL per well in the presence of 4 μg / mL polybrene (Sigma-Aldrich), to identify an optimal viral load. Most receiver cell lines were transduced such that about 1-2 integration events per cell occurred, as estimated based on the fraction of cells deemed positive for a given transgene. However, for sender cells, high expression levels were prioritized regardless of the number of integration events. Four days after infection, expression levels of transgenes were assessed by flow cytometry. Cell line selection After at least four days following lentiviral infection, cells were lifted with TrypLE Express and transferred to a round-bottom 96-well plate. The cells were pelleted and resuspended in PBS containing 2 mM EDTA (Thermo Fisher) and 2% FBS. When selecting for cell surface expression, cells were incubated with various fluorophore-conjugated antibodies, including an anti-myc AF647 (Cell Signaling Technology), anti-HA AF647 (Cell Signaling Technology), anti-Flag AF647 (R&D Systems), and anti-Her2 PE (BioLegend). Cells were stained for 45 min at 4 °C and then washed twice. Cells expressing the various transgenes were then isolated by cell sorting using a FACSAria II (BD Biosciences). To reduce some of the variability of the polyclonal cell populations, gates for sorting were set to only select cells in the middle portions of fluorescence distributions, thereby avoiding outliers on both ends. For single cell isogenic clone selection, polyclonal receiver cell lines were first generated as described above. In a subsequent selection round, cells were again sorted by FACS and single cells were seeded into individual wells of a 96-well plate. Cells were allowed to grow for about 2 weeks and then screened for circuit performance. Single cell clones were then picked based on several criteria. First, baseline expression levels of the EGFP and mCherry target genes needed to be similar to average values of the polyclonal population. Clones were then co-cultured with the full panel of sender cells expressing all combinations of synNotch ligands (CD19, nfGFP, and Her2). Single cell clones were selected for further analysis based on the following criteria: 1) No silencing of the target genes when co-cultured with cells that do not Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 express a ligand, 2) silencing only of the intended target in response to the corresponding ligand that induces KRAB, with no off-target silencing (if both EGFP and mCherry are present), 3) no silencing in response to a ligand that should only stimulated synNotch-D3L, 4) a robust recovery of target gene expression when only synNotch-KRAB is stimulated, and 5) efficient sustained silencing in response to ligands that induce both KRAB and D3L. In the case of the inverter cell lines, with just mCherry-PEST or with mCherry-P2A-PCAD, a single cell clone was first generated as described that contains the synNotch-KRAB, synNotch-D3L, and CymR- EGFP-DD constructs. Then in subsequent lentiviral transductions, this single cell clone received the remaining CymR target constructs, thereby making the cell lines monoclonal with respect to the synNotch receptors and CymR, but polyclonal with respect to mCherry and PCAD. synNotch stimulation Receiver cells with synNotch were stimulated with either magnetic beads coated with antibodies against Myc or HA epitope tags, or by co-culture with sender cells expressing ligands recognized by the synNotch extracellular domains. For experiments involving magnetic bead stimulation, receiver cells were plated in flat-bottom 96-well plates at a density of 16,000 cells per well. During the synNotch stimulation period, which ranged from as short us 0.5 days up to 7 days, washed magnetic beads (Thermo Fisher) were added at density of 0.2 μL of bead slurry equivalent per well. Cells were passaged as normal throughout this period, and beads were re- added after each passage. To end synNotch stimulation, the gamma-secretase inhibitor DAPT (TOCRIS) was added to the wells to a final concentration to 25 μM and the beads were removed using a magnetic plate. For the experiment in which the relative timing of KRAB and D3L activity was varied, DAPT could not be used as it would inhibit both synNotch receptors. Instead, inhibition of the TetR-KRAB domain specifically was accomplished by addition of doxycycline to a final concentration of 1 μg / mL, while maintaining the anti-HA beads that stimulate synNotch-D3L. Cells were maintained in the presence of DAPT or doxycycline, as appropriate, for at least 3 passages after removal of the beads. For the experiment to test reactivation of the EGFP target gene, cells were cultured in media containing 250 nM 5-Aza-2- deoxycytidine (Sigma-Aldrich), and cells were passaged daily until essentially all cells had died from 5-aza toxicity. For the experiment to test the delay time of mCherry activation in the inverter cascade, cells were grown during both stimulation and recovery periods in the presence of varying concentrations of Shield1, as indicated. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 For experiments involving co-culture with senders and receivers, cells were mixed and plated in round-bottom 96-well plates at a total cell density of 16,000 cells per well. Sender- receiver ratios were set to 3:1 at the beginning of the co-culture, or in the cases of multiple sender types simultaneously, 3:3:1 (sender1-sender2-receiver) or 3:3:3:1 (sender1-sender2- sender3-receiver). In the latter cases, total cell density was maintained at 16,000 cells per well. During the synNotch stimulation phase, cells were passaged every day for up to 4 days, or every 1-2 days for experiments with a 7-day stimulation period. This passaging regimen was developed empirically, and the goal was to maintain 50-100% confluency to increase the chances of cell-cell contact and synNotch stimulation. SynNotch stimulation was stopped by adding DAPT as above, and cells were maintained in the presence of DAPT for the remainder of the time course. However, for the experiment involving sequential stimulation with two different types of sender cells, the first stimulation was ended by FACS sorting to isolate the receivers. Following a recovery period, receivers were then co-culture with the second set of senders as above. All experiments were performed in triplicate wells. Flow cytometry At various timepoints during synNotch stimulation or recovery, cells were lifted with TrypLE and a portion were transferred to a round-bottom 96-well plate. Cells were then washed and resuspended in Ca2+- and Mg2+-free PBS containing 2 mM EDTA and 2% FBS. Fluorescence intensities of the resuspended cells were then assessed with either the BD LSR II or BD LSR Fortessa X20 with a high-throughput sampler. In addition to forward scatter and side scatter channels, fluorescence data was also collected across four channels, including BV 421 (TagBFP), FITC (EGFP), PE CF594 (mCherry and PE-conjugated antibodies), and APC (iRFP670 and AF647-conjugated antibodies). Following acquisition, fluorescence data were first processed with FlowJo software (version 10, BD). Gates to determine the percent of cells positive or negative for EGFP or mCherry were established by untransduced (negative control) cell lines. Self-organizing spheroid assembly Sender cells, or control cells without ligands, were first lifted with TrypLE and resuspended in fresh media. Receiver cells expressed a synCAM that disrupted adhesion to tissue culture plates, and as such they did not require enzymatic dissociation. Senders and receivers were then resuspend in DMEM with 4% FBS, which was empirically determined Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 would enable cells to self-assemble and grow as spheroids for up to 2 weeks in experiments without overgrowth and cell death. Through serial dilution, 30 sender cells and 30 receiver cells were added to 96-well round-bottom ultra-low attachment plate (Corning #7007) in a final volume of 100 μL. Co-cultures were performed with at least four replicates, and in some cases, up to 24 replicates. Co-cultures were then allowed to proceed unperturbed and undergo live-cell imaging daily by confocal microscopy. Microscopy Spheroids were directly imaged in the ULA cultures plates using the Opera Phenix automated spinning-disk confocal microscope with a ×20 water-immersion objective lens (0.45 NA). Images were acquired daily for up to 8 days. Fluorescence data was acquired across four channels allowing determination of iRFP670 (senders), TagBFP (constitutively expressed in receivers), EGFP (PCAD- receivers), and mCherry (PCAD+ receivers). Cross-sections were collected every 5-10 μM to create a total Z-stack 140 μM thick. Harmony software was used to operate the microscope and for image processing. Maximum intensity projections and three- dimensional reconstructions were generated with Harmony using the entire Z-stack at each time point. Harmony software was also used to apply contrast, brightness, and gamma values for digital image visualization (applied equally across samples). Quantification and statistical analysis The numbers of replicates are indicated in the figure legends and method details. Flow cytometry raw data were first assessed using FlowJo (FlowJo LLC) to determine gates defining cells positive and negative for a given fluorescent marker. Data were then exported for further processing and visualization in Python, with the Jupyter Notebook and using Matplotlib and Seaborn libraries. For line plots, confidence intervals are displayed as shaded areas under the lines (same color). For flow cytometry distributions, only one replicate is shown but they are representative of the set. One-dimensional distributions are plotted as kernel density estimate plots with Seaborn and with a bandwidth value (bw_adjust) of 0.5. Two-dimensional distributions are also plotted as kernel density estimate plots with Seaborn, but as a series of contour lines (15) of increasing darkness and with the lowest 5% of data omitted. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 RESULTS Building inducible epigenetic memory switch through synNotch triggering of KRAB and Dnmt3L activity Synthetic circuits that can control cell fate transitions should have the following features: at least one output transcriptional unit than can exist in stable discrete “on” and “off” states, and a set of regulatory modules that can induce transitions between states in a directable and tunable manner. An ideal system can also enable cells to decide to undergo a transient or stable state change based on input identity. To construct such a cell fate transition circuit, key chromatin regulators were linked to synNotch receptors. Synthetic Notch receptors have an extracellular ligand recognition domain and a central Notch transmembrane domain that is proteolytically cleaved by gamma secretase upon ligand binding. This cleavage event releases the intracellular effector domain, allowing it to enter the nucleus. In this case, as intracellular effector domains, DNA-targeted chromatin regulators were used, such that upon activation, they would be released, enter the nucleus and bind to the appropriate target promoter. Specifically, a synNotch receptor was first constructed with an extracellular anti-CD19 single-chain variable fragment (with a Myc epitope tag) linked to an intracellular TetR DNA binding domain fused to the Kox1 KRAB domain. This myc-anti- CD19-Notch-TetR-KRAB receptor was designed, in principle, to induce KRAB mediated silencing at a target locus, upon stimulation with the extracellular ligand CD19. Next, a synNotch receptor was constructed with an extracellular anti-GFP nanobody LaG16 with an HA epitope tag linked to an intracellular Gal4 DNA binding domain fused to the non-catalytic regulatory subunit of the mouse Dnmt3 complex, Dnmt3L (D3L) (Figure 2A). The resulting HA-LaG16-Notch-Gal4-D3L receptor was designed, in principle, to recruit the Dnmt3 DNA methylation complex to the target locus, upon stimulation with the extracellular ligand, GFP (Figures 2A and 8A). Altogether, these two receptors should enable independent sensing of two different inputs that are then coupled to discrete DNA targeting by the two chromatin regulators (Figure 2A). Correspondingly, target promoters were constructed whose transcriptional state could be controlled by the two synNotch-gated chromatin regulators (Figure 2A). Binding site arrays for the TetR and Gal4 DNA binders were placed upstream of the strong constitutively active spleen focus-forming virus (SFFV) promoter, ensuring that KRAB and D3L do not directly compete Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 for DNA occupancy (Figure 2A). Altogether, this synNotch-promoter system constitutes a switching module that could control the expression of any payload gene. Enhanced green fluorescent protein (EGFP) was used as a reporter gene, whose expression would reflect the chromatin environment and transcriptional state of the SFFV promoter (Figure 2A). Lastly, the FKBP-DD degron (DD) was fused to EGFP to minimize the delay between transcriptional repression and fluorescence decay (Figure 2A). A “receiver” cell was generated using the mouse L929 fibroblast cell line, introducing the synNotch receptor and target promoter circuits via lentiviral integration. A corresponding set of “sender” L929 cells were also generated that express cell surface CD19 and / or non- fluorescent GFP mutant (nfGFP) ligands which could trigger one or both of the synNotch receptors (Figure S1B). The synNotch receptors could also be triggered by stimulating with either anti-Myc or anti-HA beads that recognize the extracellular epitope tags. Receiver cell stimulation by incubation with anti-Myc beads (Figure 2A) or by co- culture with CD19-presenting sender cells (Figure 8B) should induce KRAB domain recruitment to the target reporter promoter. Indeed, EGFP expression was substantially silenced within 3 days of stimulation, with about 80-90% of cells in the polyclonal population switching to an EGFPlowstate (Figures 2B, 2C, 8B, and 8C). However, this silencing is transient as EGFP expression is largely recovered following inhibition of synNotch by addition of the gamma- secretase inhibitor, DAPT (Figures 2B, 2C, 8B, and 8C). This is consistent with previous findings that KRAB-mediated transcriptional silencing is largely reversible and is consistent with substantial reversal of heterochromatin by activating factors recruited to the SFFV promoter. In contrast, stimulation with anti-HA beads (Figure 2A) or co-culture with nfGFP sender cells (Figure 8B), which should recruit the Dnmt3 complex to the reporter promoter, resulted in a much lower percentage of cells silencing EGFP (Figures 2B, 2C, 8B, and 8C). This is consistent with previous findings that Dnmt3-mediated transcriptional silencing is, by itself, ineffective at silencing strong, constitutively active promoters. Crucially, it is observed that concurrent stimulation of synNotch-KRAB and synNotch- D3L receptors results in the same potent EGFP silencing as KRAB alone but, importantly, additionally results in commitment to the EGFPlowstate in almost all cells, even after removal of the stimuli (Figures 2B, 2C, 8B, and 8C). Inhibition of the synNotch stimulus does not restore EGFP expression, indicating a switch to a self-sustaining silenced chromatin state and Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 epigenetic memory. This result is consistent with a model in which Dnmt3-mediated DNA methylation is greatly enhanced by KRAB mediated silencing. In other words, heterochromatin induction is required to effectively initiate silencing, but DNA methylation makes silencing stable (i.e., committed). To test this general hypothesis, the population of cells that had received dual synNotch-KRAB and synNotch-D3L stimulation were treated with the DNA methyltransferase inhibitor 5-aza-2’-deoxycytidine (5-aza). Reactivation of EGFP was observed in about 75% of cells (Figure 8D), supporting the substantial contribution of DNA methylation to the epigenetic memory observed in this system. Notably, memory could be achieved via concurrent stimulation with any input combination that stimulated both KRAB and Dnmt3 synNotch receptors, including anti-Myc and anti-HA beads (Figures 2B and 2C), co-culture with a sender cell that co-expresses both CD19 and nfGFP ligands (Figures 8B and 8C), or with simultaneous co-culture with two different sender cell types each expressing one of the ligands (Figure 8E). Interestingly, it was also observed that synNotch-gated silencing and memory establishment is most effective when KRAB and D3L are controlled by independent receptor molecules, even if they respond to a single ligand, CD19 (Figure 8F). A single synNotch receptor variant harboring a Gal4-KRAB-D3L dual-effector intracellular domain is less effective at silencing the target promoter and at establishing memory compared to the two-receptor configuration (Figure 8F). Thus, the dual receptor system not only drives maximal silencing and memory, but also maintains the flexibility of generating transient or stable switches in a single cell type depending on the input. In summary, it is shown that synNotch receptors can be used to control epigenetic regulators, allowing epigenetic memory to be induced by cell-cell contact. These synNotch circuits constitute minimal switching modules with which to engineer more complex cell fate transitions. Fate bifurcation emerges from synthetic memory switch Most memory switches are predicted to yield bistability, which can be observed as all-or- none bifurcation of individual cells into the ON or OFF states. Thus, it was examined whether these epigenetic switches yielded bistability. To avoid confounding effects of heterogeneous lentiviral transduction within a bulk cell population (potentially resulting from reporter integration into variable chromatin environments) three single cell clones derived from the polyclonal population harboring TetR-KRAB synNotch, Gal4-D3L synNotch, and the EGFP Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 response element were first isolated(Figure 2A). Circuit performance in each of the clones was first confirmed to be broadly representative of the bulk population (data not shown). Clones 1 and 3 demonstrated the most substantial functional difference between KRAB only and KRAB + D3L stimulation, where nearly 100% of cells underwent transient vs stable silencing, respectively, following a 3-day stimulation period (data not shown). These clonal cell lines were then used to test whether the epigenetic circuit could produce population bifurcation. A series of stimulation regimes were set up where receiver cells were subjected to stepwise increases in dual stimulation duration, ranging from half a day to three days (Figure 3A). A non-linear time-dependence for generating stably silenced cells was observed (Figure 3B), wherein a stimulus duration increase from 1 to 1.5 days resulted in more than two-fold increase in the proportion of EGFPlowcells. With 2 days of stimulation, >90% of the population stably switched to the silenced state (Figure 3B). Importantly, stimulus duration did not correlate with the final mean fluorescence intensity following the recovery period, but instead resulted in obvious bifurcation between silenced EGFPlowand active EGFPhighcells (Figure 3C). Intriguingly, short stimulus durations, such as the 1-day pulse, resulted in a silencing response characterized by two phases. Initially, the population uniformly represses EGFP (Figure 3D). However, over time and after the end of the stimulation period, the population resolves into distinct silenced and active subpopulations (Figure 3D), which is consistent with stochastic all-or-none fate transitions observed in natural differentiation processes. Cells in which synNotch-KRAB only was stimulated showed no bimodal population dynamics (data not shown), suggesting the dependence on D3L and epigenetic memory for this dynamic bifurcation behavior. Co-incident KRAB and D3L stimulation is critical for effective fate switching These findings demonstrate the requirement for stimulation of both synNotch-KRAB and synNotch-D3L receptors to yield an epigenetic memory response.These two-input circuits now provide an opportunity to address unanswered questions about the temporal relationship between KRAB and Dnmt3 silencing modules. For example, although co-stimulation of KRAB and D3L synNotch receptors required 2-3 days for commitment to the silenced state in essentially 100% of cells (Figure 3B), is this co-stimulation duration set by both factors equally? Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Alternatively, is KRAB silencing the rate-limiting step such that the stimulation duration of D3L can be shortened yet still be effective to completely induce memory? To gain insight into the temporal dependencies of the KRAB and D3L synNotch silencing system, the modularity of the circuit was taken advantage of to test a series of stimulation regimes where a constant stimulation duration of 3 days was maintained for both factors but the relative start time of each was varied (data not shown). Correspondingly, this set of experiments systematically alters the overlap time between KRAB and D3L stimulation and includes regimes where synNotch-D3L was stimulated entirely after synNotch-KRAB. In the isogenic clone #1 described above (data not shown), KRAB stimulation alone for 3 days results in 100% silencing but 0% memory, whereas D3L stimulation alone for 3 days results in in 0% silencing, thus precluding memory establishment (Figure 4A). Overall, as the duration of temporal coincidence between KRAB and D3L stimulation is increased, ranging from a 2-day gap to a full 3-day overlap (data not shown), the proportion of cells that stably silenced EGFP also increased (Figure 4A). Similar to the result obtained with short stimulus pulses, shorter temporal overlaps produced bifurcation of the cell population between silenced EGFPlowand active EGFPhighfates (Figure 4A). Notably, a complete fate switch did not require the full 3 days of concurrent KRAB and D3L stimulation. Indeed, even as little as half a day of overlap resulted in about 50% of the population stably silencing EGFP (Figure 4A). This suggests that D3L is highly effective at establishing memory when KRAB has completely silenced expression first, and that KRAB is the rate-limiting factor in our system for driving stable fate switching. Also of note is that the ability of KRAB to potentiate the effect of D3L in silencing diminishes relatively quickly upon its removal (Figure 4A), highlighting the strict requirement for sufficient coincident stimulation to result in fate switching. Interestingly, comparing the result of the half-day overlap (Figure 4A) to the result of the half-day total co-stimulation (Figure 3C) reveals an important principle. While the overlap time in both scenarios is the same, a half-day of total KRAB stimulation, as opposed to three days, does not result in any silencing or commitment (Figure 3C). This result suggests that the cooperation between KRAB and D3L is not merely due to the co-occupancy of the factors at the target promoter. Instead, it is likely that KRAB must first alter the chromatin environment of the promoter and establish a silenced state, which takes time, in order for D3L to then induce commitment. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Overall, these findings expand on a previous model of chromatin regulation, wherein KRAB and D3L promote the stepwise transition of a promoter first from active to transiently silenced and finally to the stably silenced state (Figure 4B). KRAB, through the establishment of H3K9me3 and HP1-dependent heterochromatin, minimizes the energy required to transition from actively transcribed to transiently silenced. In contrast, D3L, through self-sustaining DNA methylation and the recruitment of additional chromatin silencers, minimizes the energy required to transition from the transiently silenced state to the stably silenced state (Figure 3C). The notion that KRAB and D3L primarily function at distinct steps in this cascade further supports the finding that the synergy is only effective if the first transition, mediated by KRAB silencing, has been sufficiently made. These findings further elaborate on the temporal relationship of these chromatin regulators, although the precise timings of each will probably vary in response to different input types and in different chromatin contexts. Nonetheless, these findings also show how the temporal relationship between these two forms of gene silencing can be exploited in synthetic circuits for discriminating input sequences. Combinatorial inputs can direct specific transitions from among alternative fate choices It was next reasoned that a cell that not only can transition to a single comitted fate, but that can choose to transition to multiple distinct alternative fates could be engineered. Thus, the initial design was expanded (Figure 2A) by incorporating an additional epigenetically regulated target reporter promoter and an associated third synNotch input receptor (Figure 5A). The third synNotch contains an anti-Her2 scFv as the extracellular domain, allowing the receiver cell with all three synNotch receptors to respond to three different input signals, CD19, nfGFP, and Her2, presented on the surface of sender cells. Additionally, the intracellular domain of the anti-Her2 synNotch is distinct from the others in that it consists of a third orthogonal DNA binding domain, LexA, fused to the KRAB domain (Figure 5A). Correspondingly, the second target locus in the receiver cell is composed of the same constitutively active SFFV promoter, but with upstream binding sites for LexA and Gal4, but not TetR (Figure 5A). This second promoter drives the expression of mCherry and thereby allows monitoring of the expression state of each target promoter independently (Figure 5B). Distinct cell states can be determined by both the level of EGFP and mCherry expression. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Altogether, this network design represents a minimal architecture that enables independent silencing of each target construct as well as independent control of both KRAB and D3L factors. Selection of the target to be silenced should be determined by the release of TetR- KRAB and LexA-KRAB synNotch factors, which occur by CD19 and Her2 ligand stimulation, respectively (Figure 5A). Since it was observed that D3L does not substantially silence expression on its own, a single Gal4-D3L module can be shared by both target promoters to commit to the KRAB-mediated silenced state at each locus (Figure 5A). Consequently, the engineered cell should be able to occupy four distinct stable fates starting from the EGFPhighmCherryhighbaseline fate. In response to specific inputs, cells should undergo all-or-none transitions to three other alternative possible states: the EGFPlowmCherryhighfate (transition T1), the EGFPhighmCherrylowfate (transition T2), or the EGFPlowmCherrylowfate (transition T3) (Figure 5B). As before, to avoid heterogeneity from reporters integrated at different local chromatin environments, three different isogenic clones harboring the complete network were isolated and examined in parallel. The multipotent receiver cells with the different sender cell types were co- cultured for seven days to ensure sufficient stimulation, followed by recovery in the presence of the gamma secretase inhibitor DAPT. As had been observed with the initial single target receiver (Figure 2), stimulation with CD19 alone, Her2 alone, or CD19 and Her2 together, which only induce recruitment of KRAB to the two target promoters, resulted in only transient changes to alternative expression states during stimulation (Figure 5C). However, co-culture with senders presenting the same ligand sets but also with nfGFP, which recruits D3L to both targets, resulted in transitions to each of the alternative stable fates dictated by the specific KRAB synNotch variant that was stimulated (Figure 5C). CD19-nfGFP senders induced the T1 transition, Her2-nfGFP senders induced the T2 transition, and CD19-Her2-nfGFP senders induced the T3 transition, with little deviation to the other fates (Figure 5C). As a further illustration of the modularity of the system, the T3 transition could be induced by co-culture with a single sender cell type expressing all three ligands, by two different sender cell types expressing CD19 and nfGFP or Her2 and nfGFP, or by three sender cell types each expressing CD19, Her2, or nfGFP (data not shown). Thus, the specific transitions between alternative states can be selectively controlled by the combination of stimulatory inputs. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 It was also realized that the irreversible commitment to each fate would enable encoding of different stepwise transitions from the EGFPhighmCherryhighbaseline fate to the terminal EGFPlowmCherrylowfate (data not shown). For example, a receiver cell experiencing Her2 and nfGFP undergoes the T2 transition to the EGFPhighmCherrylowfate. Exposure to a second stimulus after removal of the first, specifically CD19 and nfGFP, should then result in a new transition (T5) from EGFPhighmCherrylowto the terminal EGFPlowmCherrylowfate (data not shown). Indeed, a stepwise co-culture program was performed whereby the multipotent receiver was first co-cultured with Her2-nfGFP senders, the senders were then removed by fluorescence- activated cell sorting, and then the resulting receivers were exposed to the CD19-nfGFP sender type. Cells transitioned specifically between the fates in the prescribed manner and remained in the terminal EGFPlowmCherrylowfate after synNotch inhibition (data not shown). By comparison, performing stepwise input exposure without inclusion of the nfGFP, and therefore without D3L recruitment, resulted in transient shifts between alternative states, but no evidence of additive transitions to the terminal fate (data not shown). Thus, these epigenetic modules can be used to drive a specific temporal sequence of fate transitions. Altogether, the engineered multipotent receiver cell highlights the highly flexible and precise capabilities of the synNotch-based epigenetic fate switching system. This synthetic circuit forms a foundation for complex programmable differentiation systems. Engineering a stable gene activation switch using an inverter cascade These types of irreversible epigenetic systems are able to effectively silence a target gene locus, but how can one irreversibly activate a gene locus? It was reasoned that stable activation could be accomplished by linking the modular silencing switch to an inverter module in a cascade circuit topology. In this case, the epigenetic silencing of a first gene relieves the repression of a second downstream gene, in turn leading to its activation (Figure 6A). To construct such a system, the DNA binding protein, CymR, was first placed under the control of the engineered TetO-UAS-SFFV promoter, such that KRAB and D3L recruitment silences CymR expression. Second, a downstream locus was constructed expressing mCherry (with a PEST degron) that is inhibited by CymR, specifically by placing four copies of the CymR DNA binding site (CuO) immediately downstream of a separate copy of the SFFV promoter. This configuration results in inhibition of mCherry expression when CymR is expressed under baseline conditions (Figure 6A). CymR transcriptional inhibition should be readily reversible as Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 it is unlikely to lead to chromatin remodeling or epigenetic memory. Lastly, CymR was also fused to EGFP-DD allowing the monitoring of promoter activity at each of the two stages of the silencing-inverter cascade following synNotch stimulation (Figure 6A). These constructs were integrated in different genomic locations via separate lentiviral vectors. At baseline, cells express the CymR-EGFP gene, which represses mCherry expression from the downstream promoter (Figure 6B). However, upon synNotch stimulation, CymR- EGFP is quickly silenced followed by a matched upregulation of mCherry expression (Figure 6B). When only KRAB is recruited, only transient mCherry expression is observed; inhibition of synNotch results in a loss of mCherry expression, corresponding to the reactivation of CymR- EGFP and restoration of baseline transcriptional repression (Figure 6B). In contrast, co- stimulation of both KRAB and D3L synNotch receptors resulted in stabilized mCherry activation even upon synNotch inhibition, consistent with epigenetic memory repression at the CymR-EGFP locus (Figure 6B). To further corroborate the strict inverse relationship between stable silencing of the CymR-EGFP locus and the stable activation of the mCherry output, a series of stimulations of increasing duration was performed from 1 to 3 days (data not shown). Similar to previous observations (Figure 3), shorter stimulus durations produced stable population bifurcation of the CymR-EGFP expression, and correspondingly, the mCherry output (data ). Importantly, bifurcation of CymR-EGFP was almost perfectly coupled to the bifurcation of mCherry, such that cells that silenced CymR-EGFP also completely activated mCherry and vice versa (data not shown). Although rigorous coupling between mCherry activation and CymR-EGFP silencing was observed, the gene switches do not occur simultaneously. As previously observed with transcriptional cascades, upregulation of mCherry following synNotch stimulation occurs after a delay (Figure 6B), the duration of which can be explained by the time required for CymR-EGFP protein to sufficiently diminish, which in turn is dependent on its turnover rate. It was reasoned, therefore, that the delay timing of mCherry activation can be tuned by modulating the stability of CymR-EGFP through the controllable degron, FKBP-DD (Figure 6A). Treatment of cells with increasing concentrations of the small molecule, Shield1 (Sld1) stabilizes the degron, resulting in increased baseline levels of CymR-EGFP and slower CymR- EGFP downregulation following synNotch stimulation (data not shown). Correspondingly, increasing Sld1 concentrations results in a greater delay time to activate mCherry (Figure 6C). Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Time to the first detectable increase in mCherry positive cells can be as short as 1 day and as long as 5 days following initiation of synNotch stimulation (Figure 6C). Thus, the timing of this epigenetic activation circuit can be systematically tuned. Epigenetic switching yields morphological fate change that is stable over many cell divisions It was next sought to demonstrate the utility of these epigenetic fate switching circuits by inducing functionally important irreversible changes in cell behavior. Here, the ability to irreversibly alter the adhesion properties of a cell was focused on. In multicellular development, cell adhesion plays a central role in the formation of distinct spatial domains and morphogenic structures. Cell adhesion can be manipulated by ectopic or induced expression of natural or synthetic adhesion proteins. For example, expression of cadherins can induce a population of cells to form compact aggregates due to strong homotypic interaction. Additionally, in a mixed population consisting of cells expressing differing types of cadherins, distinct cell populations segregate from one another into distinct domains, recapitulating the elementary principle of differential adhesion sorting in multicellular self-organization. This type of differential adhesion sorting is used throughout development, resulting in downstream cell lineages that move or migrate to new positions to further elaborate the body. Importantly, however, for this kind of spatial sorting to be stable over the course of development, changes in expression of the responsible adhesion molecules must be irreversible. It was first confirmed that compact aggregation and self-organization of a sender- receiver co-culture system could be induced by incorporating P-cadherin (PCAD) in the inverter activation circuit (Figure 7A and 7B). Specifically, an mCherry-P2A-PCAD fusion protein was designed that is controlled by the CymR-EGFP-DD protein (Figure 7A). Following synNotch stimulation of both KRAB and D3L, mCherry expression can be stably induced (data not shown). Importantly, when cells are co-cultured in an ultra-low attachment (ULA) culture plate, the differential adhesion properties between the sender cells that do not express a cadherin, and the activated receiver cells expressing PCAD, results in the morphogenesis of a compact core of receivers surrounded by an outer shell of sender cells (data not shown). This confirms the ability to switch between cell adhesion phenotypes by way of a synNotch-gated inverter cascade circuit, which parallels a previous study. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 It was reasoned that irreversibly induced cadherin expression would have a crucial impact in a scenario where a change in adhesion is accompanied by cell sorting, movement, or proliferation, which would cause receiver cells to physically distance themselves from the stimulating sender cells. In such a situation, contact with the sender cells, which drives the state transition, would inherently be transient. Thus, the receiver cells would only stay in their newly induced adhesion state if they irreversibly activated cadherin expression. In natural development, similar transient cell-cell communication occurs when morphogenetic events produce spatially constrained domains. To test if the epigenetic switch circuits could lead to new self-organizing behaviors by inducing irreversible changes in cell sorting, a co-culture experiment was designed involving sender cells expressing N-cadherin (NCAD) and receiver cells that do not express a cadherin at baseline, but that can induce the expression of PCAD upon receiving synNotch stimulation. First, it was confirmed that mixing of two cell types that constitutively express NCAD or PCAD results in the two populations aggregating whilst simultaneously sorting from each other into two distinct domains (data not shown). This is because the homotypic interactions (PCAD- PCAD and NCAD-NCAD) are stronger than the heterotypic interactions (NCAD-PCAD). Three alternative types of NCAD expressing sender cells were constructed. Specifically, control cells that do not express a synNotch ligand, sender cells that express the synNotch-KRAB ligand (CD19), or senders that express both the synNotch-KRAB and synNotch-D3L ligands (CD19 and nfGFP) were generated. Thus, each of these populations of NCAD sender cells invoke a different response in the receiver cells. When receiver cells were co-cultured with the control cells with no ligand, the receivers remain in an EGFPhighstate do not aggregate (data not shown). When the same receivers are mixed with the CD19+ senders that stimulate synNotch-KRAB only, EGFP is largely silenced and cells begin to induce mCherry expression and self-organize to form a PCAD-based aggregate adjacent to the sender cell aggregate (Figure 7C). Importantly, as the receiver cells continue to grow, and because the interfacial region between the sender and receiver cells is finite, the excess receiver cells lose contact with the senders and revert to the PCAD- unclustered adhesion state (green dissociated cells) (Figure 7C). Thus, when only synNotch- KRAB is induced, only the subpopulation of receiver cells that maintain contact with the sender cells are able to maintain the aggregated PCAD+ state. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Lastly, when receivers are cultured with the CD19+ nfGFP+ senders, which stimulate both KRAB and D3L and result in memory (data not shwon), it is again observed that receivers begin to induce mCherry and PCAD, resulting in a self-organizing aggregation adjacent to the sender cell mass (Figure 7D). Crucially in the presence of this sender cell type, most of the receiver cells remain in the mCherry+ PCAD+ state, even as the cells grow and gradually lose direct contact with the senders. The majority of receiver cells do not revert to the PCAD- fate (Figure 7D). Thus, because the receiver cells have irreversibly switched to the PCAD+ state, they can maintain the NCAD / PCAD dual cluster structure indefinitely, even as they move away from direct contact with the initiating sender cells that were responsible for inducing this fate change. Altogether, these results demonstrate that the epigenetic fate switching circuits can be readily adapted to control transitions between different cell adhesion states by installing a cadherin gene as a payload. In turn, it has been shown how irreversibly inducing a stable morphogenetic change, such as a change in adhesion properties, unlocks the potential for more complex trajectories of multicellular self-organization. An epigenetic inverter circuit allows permanent transdifferentiation of mouse embryonic fibroblasts into neuronal cells The epigenetic inverter circuit allows engineered cells to maintain memory of a given input and instruct a permanent transcriptional output despite even a transient pulse of the input. This would make more accessible for engineered cells to complete biological processes that physiologically require a significantly long input duration. A paradigmatic example is the transdifferentiation of embryonic fibroblasts into neurons, which requires the overexpression of transcription factors such as BAM (Brn2, Ascl1 and Myt1l) for weeks (PMID: 21617644). To assess the capacity of the synNotch-based epigene^c inverter circuit to drive neuronal transdifferen^a^on, mouse embryonic C3H fibroblasts were transduced with lentiviral vectors encoding for synNotch.TetR:KRAB, synNotch.Gal4:D3L, a TetO-Gal4bs- phPGK.CymR-GFP reporter cassette and two CymR-controlled cassettes encoding for the BAM TFs (Figs. 9A-B). Upon sorting of these transgenes-positive receiver cells, the cells were co- cultured with mitomycin-treated C3H senders cells exposing the priming ligand for either synNotch.TetR:KRAB or synNotch.Gal4:D3L or both of them. After a short pulse (three v.s. seven days) of synNotch-priming, the synNotch signaling was stopped by adding DAPT to the Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 co-culture. Neuronal transdifferentiation was assessed by measuring the number of cells acquiring expression of the neuronal marker Tuj125 days after the starting point of the co- culture. By conducting this experiment, it was demonstrated that transient (three days) priming of synNotch.KRAB+synNotch.D3L is sufficient to promote the neuronal transdifferentiation of mouse C3H embryonic fibroblasts (Figs 9C-D). Importantly, while a three-day cumate-based transcriptional activation of BAM TFs in C3H cells is not enough to induce neuronal transdifferentiation, on the other hand three-day priming of synNotch.KRAB+synNotch.D3L, but not synNotch.KRAB alone, is sufficient to significantly promote the transdifferen^a^on process and obtain neuronal cells marked by TUJ1 expression and clear morphological protrusions (Figs. 9C-D). This data represents to our knowledge the first evidence of a contact-dependent, synNotch:epigenetic memory-based induction of transdifferentiation upon just a transient input. This proof of concept paves the way to the application of similar epigenetic inverter circuits to programmable, contact-dependent differentiation (or trans-differentiation) processes in therapeutically relevant cells. Synnotch-based circuits instruct epigenetic memory in human ipscs-derived neural stem progenitor cells Given the potential of synNotch-based circuits to instruct contact-dependent epigenetic memory, their portability to a therapeutically relevant cell type such as human IPSCs-derived Neural Stem / Progenitor Cells (NSPCs) was tested, which hold promise to differentiate into multiple committed cell types (astrocytes, oligodendrocytes, neurons) and generate complex brain structures ex vivo or directly in vivo for disease modeling, drug screening and regenerative medicine applications (PMID: 26849304). As a proof of concept of the functionality of synNotch-dependent epigenetic memory in this system, hIPSC-derived NSPCs (obtained by doxy-based overexpression of Ngn2 in hIPSCs – Fattahi lab, UCSF) were transduced with lentiviral vectors encoding for synNotch.TetR:KRAB, synNotch.Gal4:D3L and a TetO-Gal4bs- phPGK.GFP reporter cassette. Upon sorting of transgenes-positive cells, either of the two synNotch receptors, or both of them, were stimulated with a short (seven days) pulse of cognate input dynabeads. Cells were then recovered from the initial stimulus by beads removal and DAPT supplementation (Figs. 10A-B). Remarkably, the transient seven-day input was enough to induce silencing of the phPGK.GFP reporter in NSPCs, which was then permanently maintained Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 until the end of the experiment (27 days). Silencing was not achieved by stimulating synNotch.KRAB alone, while was achieved by both synNotch.D3L alone and synNotch.KRAB+synNotch.D3L, the latter providing up to 2-fold increase in silencing induction efficiency compared to synNotch.D3L alone (Fig. 10C). Whether the epigenetic silencing imposed by synNotch.KRAB+synNotch.D3L in NSPCs can be inherited along their differentiation into mature neurons was assessed. To answer this question, the epigenetically silenced cells were cultured for seven more days in a media driving neuronal maturation. Treated cells lost the expression of PAX6 (stemness-associated transcription factor) and increased the expression of Neurofilamentin H-positive protrusions, indicating successful neuronal differentiation (Figs. 10D-E). However, they did not regain expression of the phPGK.GFP reporter, indicating that the epigenetic repressive state imposed at the neural stem / progenitor cell level by the synNotch-based circuit is resistant to the chromatin and transcriptional remodeling occurring during differentiation (Fig. 10F). This data represents to our knowledge the first evidence of a contact-dependent, synNotch-based epigenetic memory induction in therapeutically relevant human stem cells, inheritable upon differentiation of the stem cell into a committed cell type. This proof of concept paves the way to the application of synNotch-based epigenetic memory circuit in relevant human stem cells, such as neural stem cells, other somatic stem cells or embryonic stem cells, with multiple potential applications in the cell therapy, regenerative medicine, complex structures morphogenesis and disease modeling fields. DISCUSSIONEngineering modular input-induced epigenetic cell fate switches In development, long-term commitment to differentiated cell fates is controlled by the convergence of cell signaling with epigenetic gene regulatory systems. In this work, a series of synthetic circuits is developed that harness the epigenetic regulatory machinery to drive user- specified signal-induced cell fate transitions. These modules represent a generalizable and customizable approach to controlling synthetic cell fate decisions. Here, a cell fate change is considered to represent an irreversible change in gene expression. These synthetic epigenetic circuits also provide a means to dynamically tune fate bifurcation leading to population diversification. Furthermore, these synthetic modules can be used in combination. They can be Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 scaled up to regulate multiple genes independently and selectively, and they can also be linked in series to drive a specific trajectory of fate changes. Given that this system relies on endogenous epigenetic regulatory mechanisms, these synthetic circuits should be applicable in diverse mammalian cell types. Alternative approaches for engineering multiple stable cell fates Prior studies in synthetic biology have explored alternative approaches to achieve irreversible or stable cell fate switches. One alternative approach is the use of networks of mutually inhibitory transcription factors to create bistable and multistable systems. Another is to use recombinase triggered systems to flip selective promoter-gene relationships yielding stable inhibition or activation. Others have used orthogonal systems that mimic the writer-reader functionality of natural epigenetic systems. While each of these approaches has its advantages and highlights different principles, the approach herein of recycling the native epigenetic machinery (and the power of DNA methylation transmission as a source of memory) is highly modular, and therefore lends itself to constructing higher order systems in which fate transitions are highly targeted, tightly controlled by specific combinations and durations of signals, and can be linked together into specific fate transition cascades. Moreover, all of these diverse synthetic memory systems have the potential to be linked together to create hierarchical systems that integrate different mechanisms of cell fate determination. Cooperativity between KRAB and Dnmt3 in stable silencing The cell fate switches that have been built herein depend on the cooperation between the KRAB domain (and the KAP1 complex that it recruits) and the Dnmt3 complex to remodel chromatin and generate a self-sustaining transcriptionally inactive state.. Many studies have identified several mechanisms of interplay between heterochromatin formation and DNA methylation that can account for this cooperativity. Such potential mechanisms include direct interactions between heterochromatin protein 1 and the Dnmt3 complex, potentially enhancing its localization when KRAB / KAP1 establishes a silenced heterochromatin state. Additionally, it is well established that histone H3 lysine 4 (H3K4) trimethylation, which is strongly associated with active transcriptional activity at promoters, does not allow full activity of Dnmt3A. Thus, silencing of transcription, which is associated with H3K4 demethylation, may produce the chromatin environment allowing Dnmt3 to efficiently methylate DNA. In support of this, fusing Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Dnmt3L to the H3 N-terminal domain, which effectively provides an unmethylated H3K4 signal, potentiates DNA methylation and silencing by the Dnmt3 complex. This work examines the cooperativity of these chromatin mechanics from a temporal perspective. By systematically altering stimulation duration and relative timing of KRAB and D3L recruitment, support for a simplified model describing the transition to a silenced state as a two-step process is provided. These findings are consistent with the Dnmt3 complex participating in the second step of this process, that is, being the primary driver of establishing the stably silenced state from the transiently silenced state. This explains why D3L is mostly ineffective at establishing a silenced state on its own. In contrast, KRAB appears to primarily impact the first step, to potently overcome active transcription and in effect, prime D3L for its function by producing a permissive chromatin state for it to act. Controlling cell fate trajectories over time and space with irreversible signaling-induced epigenetic switches It is shown how inducible epigenetic silencing can be an effective foundation for engineering cell fate switches, although not all use cases will require such long-term stability. These synthetic fate switch circuits may be most valuable for cell engineering applications in which conditions, signals, and cell interactions evolve over time, as in the case of development- like processes. In this work, the change in cell adhesion properties—that is, the fate switch itself—contributes to driving the time evolution of the system wherein cells became separated from the cells that provide the initiating signal. However, as the complexity of multicellular engineering increases, it is also bound to become more dynamic, where spatial reorganization, cell proliferation, and cell mobility result in changing signals and interactions. The cell fate control modules engineered here exhibit a high level of spatial and temporal input control and can generate complex spatial and temporal cellular phenotypes. Enabling cells to flexibly decide whether to transiently change their expression or to commit to a new fate, or both simultaneously within a population allows the engineering of far more dynamic cellular systems. REFERENCES1. Perrimon, N., Pitsouli, C., and Shilo, B.-Z. (2012). Signaling Mechanisms Controlling Cell Fate and Embryonic Patterning. Cold Spring Harb Perspect Biol 4, a005975. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 2. Loh, K.M., van Amerongen, R., and Nusse, R. (2016). Generating Cellular Diversity and Spatial Form: Wnt Signaling and the Evolution of Multicellular Animals. Developmental Cell 38, 643–655. 3. Amsen, D., Helbig, C., and Backer, R.A. (2015). Notch in T Cell Differentiation: All Things Considered. Trends in Immunology 36, 802–814. 4. Lim, W.A. (2022). The emerging era of cell engineering: Harnessing the modularity of cells to program complex biological function. Science 378, 848–852. 5. Martínez-Ara, G., Stapornwongkul, K.S., and Ebisuya, M. (2022). Scaling up complexity in synthetic developmental biology. Science 378, 864–868. 6. Trentesaux, C., Yamada, T., Klein, O.D., and Lim, W.A. (2023). Harnessing synthetic biology to engineer organoids and tissues. Cell Stem Cell 30, 10–19. 7. McNamara, H.M., Ramm, B., and Toettcher, J.E. (2023). Synthetic developmental biology: New tools to deconstruct and rebuild developmental systems. Seminars in Cell & Developmental Biology 141, 33–42. 8. Teixeira, A.P., and Fussenegger, M. (2024). Synthetic Gene Circuits for Regulation of Next-Generation Cell-Based Therapeutics. Advanced Science 11, 2309088. 9. Mao, Y., Wang, S., Yu, J., and Li, W. (2024). Engineering pluripotent stem cells with synthetic biology for regenerative medicine. Medical Review. 10. Black, J.B., Perez-Pinera, P., and Gersbach, C.A. (2017). Mammalian Synthetic Biology: Engineering Biological Systems. Annual Review of Biomedical Engineering 19, 249– 277. 11. Xie, M., and Fussenegger, M. (2018). Designing cell function: assembly of synthetic gene circuits for cell biology applications. Nat Rev Mol Cell Biol 19, 507–525. 12. Ho, C., and Morsut, L. (2021). Novel synthetic biology approaches for developmental systems. Stem Cell Reports 16, 1051–1064. 13. Gardner, T.S., Cantor, C.R., and Collins, J.J. (2000). Construction of a genetic toggle switch in Escherichia coli. Nature 403, 339–342. 14. Kramer, B.P., Viretta, A.U., Baba, M.D.-E., Aubel, D., Weber, W., and Fussenegger, M. (2004). An engineered epigenetic transgene switch in mammalian cells. Nat Biotechnol 22, 867–870. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 15. Zhu, R., Del Rio-Salgado, J.M., Garcia-Ojalvo, J., and Elowitz, M.B. (2022). Synthetic multistability in mammalian cells. Science 375, eabg9765. 16. Bird, A. (2002). DNA methylation patterns and epigenetic memory. Genes & Development 16, 6–21. 17. Steffen, P.A., and Ringrose, L. (2014). What are memories made of? How Polycomb and Trithorax proteins mediate epigenetic memory. Nat Rev Mol Cell Biol 15, 340– 356. 18. Grewal, S.I.S. (2023). The molecular basis of heterochromatin assembly and epigenetic inheritance. Molecular Cell 83, 1767–1785. 19. Zofall, M., Sandhu, R., Holla, S., Wheeler, D., and Grewal, S.I.S. (2022). Histone deacetylation primes self-propagation of heterochromatin domains to promote epigenetic inheritance. Nat Struct Mol Biol 29, 898–909. 20. Allshire, R.C., and Madhani, H.D. (2018). Ten principles of heterochromatin formation and function. Nat Rev Mol Cell Biol 19, 229–244. 21. Holoch, D., Wassef, M., Lövkvist, C., Zielinski, D., Aflaki, S., Lombard, B., Héry, T., Loew, D., Howard, M., and Margueron, R. (2021). A cis-acting mechanism mediates transcriptional memory at Polycomb target genes in mammals. Nat Genet 53, 1686–1697. 22. Amabile, A., Migliara, A., Capasso, P., Biffi, M., Cittaro, D., Naldini, L., and Lombardo, A. (2016). Inheritable Silencing of Endogenous Genes by Hit-and-Run Targeted Epigenetic Editing. Cell 167, 219-232 e14. 23. Nunez, J.K., Chen, J., Pommier, G.C., Cogan, J.Z., Replogle, J.M., Adriaens, C., Ramadoss, G.N., Shi, Q., Hung, K.L., Samelson, A.J., et al. (2021). Genome-wide programmable transcriptional memory by CRISPR-based epigenome editing. Cell 184, 2503- 2519 e17. 24. Nakamura, M., Ivec, A.E., Gao, Y., and Qi, L.S. (2021). Durable CRISPR-Based Epigenetic Silencing. BioDesign Research 2021, 9815820. 25. Mlambo, T., Nitsch, S., Hildenbeutel, M., Romito, M., Müller, M., Bossen, C., Diederichs, S., Cornu, T.I., Cathomen, T., and Mussolino, C. (2018). Designer epigenome modifiers enable robust and sustained gene silencing in clinically relevant human cells. Nucleic Acids Research 46, 4456–4468. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 26. O’Geen, H., Bates, S.L., Carter, S.S., Nisson, K.A., Halmai, J., Fink, K.D., Rhie, S.K., Farnham, P.J., and Segal, D.J. (2019). Ezh2-dCas9 and KRAB-dCas9 enable engineering of epigenetic memory in a context-dependent manner. Epigenetics & Chromatin 12, 26. 27. Bannister, A.J., Zegerman, P., Partridge, J.F., Miska, E.A., Thomas, J.O., Allshire, R.C., and Kouzarides, T. (2001). Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410, 120–124. 28. Nan, X., Ng, H.-H., Johnson, C.A., Laherty, C.D., Turner, B.M., Eisenman, R.N., and Bird, A. (1998). Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature 393, 386–389. 29. Catania, S., Dumesic, P.A., Pimentel, H., Nasif, A., Stoddard, C.I., Burke, J.E., Diedrich, J.K., Cook, S., Shea, T., Geinger, E., et al. (2020). Evolutionary Persistence of DNA Methylation for Millions of Years after Ancient Loss of a De Novo Methyltransferase. Cell 180, 263-277 e20. 30. Bostick, M., Kim, J.K., Estève, P.-O., Clark, A., Pradhan, S., and Jacobsen, S.E. (2007). UHRF1 Plays a Role in Maintaining DNA Methylation in Mammalian Cells. Science 317, 1760–1764. 31. Sharif, J., Muto, M., Takebayashi, S., Suetake, I., Iwamatsu, A., Endo, T.A., Shinga, J., Mizutani-Koseki, Y., Toyoda, T., Okamura, K., et al. (2007). The SRA protein Np95 mediates epigenetic inheritance by recruiting Dnmt1 to methylated DNA. Nature 450, 908–912. 32. Groner, A.C., Tschopp, P., Challet, L., Dietrich, J.-E., Verp, S., Offner, S., Barde, I., Rodriguez, I., Hiiragi, T., and Trono, D. (2012). The Krüppel-associated Box Repressor Domain Can Induce Reversible Heterochromatization of a Mouse Locus in Vivo*. Journal of Biological Chemistry 287, 25361–25369. 33. Ayyanathan, K., Lechner, M.S., Bell, P., Maul, G.G., Schultz, D.C., Yamada, Y., Tanaka, K., Torigoe, K., and Rauscher, F.J. (2003). Regulated recruitment of HP1 to a euchromatic gene induces mitotically heritable, epigenetic gene silencing: a mammalian cell culture model of gene variegation. Genes Dev. 17, 1855–1869. 34. Urrutia, R. (2003). KRAB-containing zinc-finger repressor proteins. Genome Biol 4, 1–8. 35. Quenneville, S., Turelli, P., Bojkowska, K., Raclot, C., Offner, S., Kapopoulou, A., and Trono, D. (2012). The KRAB-ZFP / KAP1 System Contributes to the Early Embryonic Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 Establishment of Site-Specific DNA Methylation Patterns Maintained during Development. Cell Reports 2, 766–773. 36. Gilbert, L.A., Larson, M.H., Morsut, L., Liu, Z., Brar, G.A., Torres, S.E., Stern- Ginossar, N., Brandman, O., Whitehead, E.H., Doudna, J.A., et al. (2013). CRISPR-Mediated Modular RNA- of Transcription in Eukaryotes. Cell 154, 442–451. 37. Hathaway, N.A., Bell, O., Hodges, C., Miller, E.L., Neel, D.S., and Crabtree, G.R. (2012). Dynamics and Memory of Heterochromatin in Living Cells. Cell 149, 1447–1460. 38. Bintu, L., Yong, J., Antebi, Y.E., McCue, K., Kazuki, Y., Uno, N., Oshimura, M., and Elowitz, M.B. (2016). Dynamics of epigenetic regulation at the single-cell level. Science 351, 720. 39. Okano, M., Bell, D.W., Haber, D.A., and Li, E. (1999). DNA Methyltransferases Dnmt3a and Dnmt3b Are Essential for De Novo Methylation and Mammalian Development. Cell 99, 247–257. 40. Suetake, I., Shinozaki, F., Miyagawa, J., Takeshima, H., and Tajima, S. (2004). DNMT3L Stimulates the DNA Methylation Activity of Dnmt3a and Dnmt3b through a Direct Interaction*. Journal of Biological Chemistry 279, 27816–27823. 41. Bourc’his, D., Xu, G.-L., Lin, C.-S., Bollman, B., and Bestor, T.H. (2001). Dnmt3L and the Establishment of Maternal Genomic Imprints. Science 294, 2536–2539. 42. Cappelluti, M.A., Mollica Poeta, V., Valsoni, S., Quarato, P., Merlin, S., Merelli, I., and Lombardo, A. (2024). Durable and efficient gene silencing in vivo by hit-and-run epigenome editing. Nature 627, 416–423. 43. Morsut, L., Roybal, K.T., Xiong, X., Gordley, R.M., Coyle, S.M., Thomson, M., and Lim, W.A. (2016). Engineering Customized Cell Sensing and Response Behaviors Using Synthetic Notch Receptors. Cell 164, 780–791. 44. Imai, C., Mihara, K., Andreansky, M., Nicholson, I.C., Pui, C.-H., Geiger, T.L., and Campana, D. (2004). Chimeric receptors with 4-1BB signaling capacity provoke potent cytotoxicity against acute lymphoblastic leukemia. Leukemia 18, 676–684. 45. Margolin, J.F., Friedman, J.R., Meyer, W.K., Vissing, H., Thiesen, H.J., and Rauscher, F.J. (1994). Krüppel-associated boxes are potent transcriptional repression domains. Proceedings of the National Academy of Sciences 91, 4509–4513. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 46. Fridy, P.C., Li, Y., Keegan, S., Thompson, M.K., Nudelman, I., Scheid, J.F., Oeffinger, M., Nussenzweig, M.C., Fenyö, D., Chait, B.T., et al. (2014). A robust pipeline for rapid production of versatile nanobody repertoires. Nat Methods 11, 1253–1260. 47. Chédin, F., Lieber, M.R., and Hsieh, C.-L. (2002). The DNA methyltransferase- like protein DNMT3L stimulates de novo methylation by Dnmt3a. Proceedings of the National Academy of Sciences 99, 16916–16921. 48. Gowher, H., Liebert, K., Hermann, A., Xu, G., and Jeltsch, A. (2005). Mechanism of Stimulation of Catalytic Activity of Dnmt3A and Dnmt3B DNA-(cytosine-C5)- methyltransferases by Dnmt3L *. Journal of Biological Chemistry 280, 13341–13348. 49. Ooi, S.K.T., Qiu, C., Bernstein, E., Li, K., Jia, D., Yang, Z., Erdjument-Bromage, H., Tempst, P., Lin, S.-P., Allis, C.D., et al. (2007). DNMT3L connects unmethylated lysine 4 of histone H3 to de novo methylation of DNA. Nature 448, 714–717. 50. Banaszynski, L.A., Chen, L., Maynard-Smith, L.A., Ooi, A.G.L., and Wandless, T.J. (2006). A Rapid, Reversible, and Tunable Method to Regulate Protein Function in Living Cells Using Synthetic Small Molecules. Cell 126, 995–1004. 51. Martinez-Fabregas, J., Wilmes, S., Wang, L., Hafer, M., Pohler, E., Lokau, J., Garbers, C., Cozzani, A., Fyfe, P.K., Piehler, J., et al. (2019). Kinetics of cytokine receptor trafficking determine signaling and functional selectivity. eLife 8, e49314. 52. Geling, A., Steiner, H., Willem, M., Bally‐Cuif, L., and Haass, C. (2002). A γ‐ secretase inhibitor blocks Notch signaling in vivo and causes a severe neurogenic phenotype in zebrafish. EMBO reports 3, 688–694. 53. Margueron, R., and Reinberg, D. (2010). Chromatin structure and the inheritance of epigenetic information. Nat Rev Genet 11, 285–296. 54. Reinig, J., Ruge, F., Howard, M., and Ringrose, L. (2020). A theoretical model of Polycomb / Trithorax action unites stable epigenetic memory and dynamic regulation. Nat Commun 11, 4782. 55. Moris, N., Pina, C., and Arias, A.M. (2016). Transition states and cell fate decisions in epigenetic landscapes. Nat Rev Genet 17, 693–703. 56. Mojtahedi, M., Skupin, A., Zhou, J., Castaño, I.G., Leong-Quong, R.Y.Y., Chang, H., Trachana, K., Giuliani, A., and Huang, S. (2016). Cell Fate Decision as High-Dimensional Critical State Transition. PLOS Biology 14, e2000640. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 57. Stumpf, P.S., Smith, R.C.G., Lenz, M., Schuppert, A., Müller, F.-J., Babtie, A., Chan, T.E., Stumpf, M.P.H., Please, C.P., Howison, S.D., et al. (2017). Stem Cell Differentiation as a Non-Markov Stochastic Process. Cell Systems 5, 268-282.e7. 58. Lehnertz, B., Ueda, Y., Derijck, A.A.H.A., Braunschweig, U., Perez-Burgos, L., Kubicek, S., Chen, T., Li, E., Jenuwein, T., and Peters, A.H.F.M. (2003). Suv39h-Mediated Histone H3 Lysine 9 Methylation Directs DNA Methylation to Major Satellite Repeats at Pericentric Heterochromatin. Current Biology 13, 1192–1200. 59. Fuks, F., Hurd, P.J., Deplus, R., and Kouzarides, T. (2003). The DNA methyltransferases associate with HP1 and the SUV39H1 histone methyltransferase. Nucleic Acids Research 31, 2305–2312. 60. Hu, J.-L., Zhou, B.O., Zhang, R.-R., Zhang, K.-L., Zhou, J.-Q., and Xu, G.-L. (2009). The N-terminus of histone H3 is required for de novo DNA methylation in chromatin. Proceedings of the National Academy of Sciences 106, 22187–22192. 61. Otani, J., Nankumo, T., Arita, K., Inamoto, S., Ariyoshi, M., and Shirakawa, M. (2009). Structural basis for recognition of H3K4 methylation status by the DNA methyltransferase 3A ATRX–DNMT3–DNMT3L domain. EMBO reports 10, 1235–1241. 62. Guo, X., Wang, L., Li, J., Ding, Z., Xiao, J., Yin, X., He, S., Shi, P., Dong, L., Li, G., et al. (2015). Structural insight into autoinhibition and histone H3-induced activation of DNMT3A. Nature 517, 640–644. 63. Du, J., Johnson, L.M., Jacobsen, S.E., and Patel, D.J. (2015). DNA methylation pathways and their crosstalk with histone methylation. Nature Reviews Molecular Cell Biology 16, 519–532. 64. Carter, P., Presta, L., Gorman, C.M., Ridgway, J.B., Henner, D., Wong, W.L., Rowland, A.M., Kotts, C., Carver, M.E., and Shepard, H.M. (1992). Humanization of an anti- p185HER2 antibody for human cancer therapy. Proceedings of the National Academy of Sciences 89, 4285–4289. 65. Liu, X., Jiang, S., Fang, C., Yang, S., Olalere, D., Pequignot, E.C., Cogdill, A.P., Li, N., Ramones, M., Granda, B., et al. (2015). Affinity-Tuned ErbB2 or EGFR Chimeric Antigen Receptor T Cells Exhibit an Increased Therapeutic Index against Tumors in Mice. Cancer Research 75, 3596–3607. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 66. Ottoz, D.S.M., Rudolf, F., and Stelling, J. (2014). Inducible, tightly regulated and growth condition-independent transcription factor in Saccharomyces cerevisiae. Nucleic Acids Research 42, e130. 67. Mullick, A., Xu, Y., Warren, R., Koutroumanis, M., Guilbault, C., Broussau, S., Malenfant, F., Bourget, L., Lamoureux, L., Lo, R., et al. (2006). The cumate gene-switch: a system for regulated expression in mammalian cells. BMC Biotechnol 6, 43. 68. Rosenfeld, N., and Alon, U. (2003). Response Delays and the Structure of Transcription Networks. Journal of Molecular Biology 329, 645–654. 69. Haugwitz, M., Garachtchenko, T., Nourzaie, O., Gandlur, S., and Sagawa, H. (2008). Rapid, on-demand protein stabilization and destabilization using the ProteoTunerTMsystems. Nat Methods 5, iii–iv. 70. Tsai, T.Y.-C., Garner, R.M., and Megason, S.G. (2022). Adhesion-Based Self- Organization in Tissue Patterning. Annual Review of Cell and Developmental Biology 38, 349– 374. 71. Duguay, D., Foty, R.A., and Steinberg, M.S. (2003). Cadherin-mediated cell adhesion and tissue segregation: qualitative and quantitative determinants. Developmental Biology 253, 309–323. 72. Nose, A., Nagafuchi, A., and Takeichi, M. (1988). Expressed recombinant cadherins mediate cell sorting in model systems. Cell 54, 993–1001. 73. Toda, S., Blauch, L.R., Tang, S.K.Y., Morsut, L., and Lim, W.A. (2018). Programming self-organizing multicellular structures with synthetic cell-cell signaling. Science 361, 156–162. 74. Bao, M., Cornwall-Scoones, J., Sanchez-Vasquez, E., Chen, D.-Y., De Jonghe, J., Shadkhoo, S., Hollfelder, F., Thomson, M., Glover, D.M., and Zernicka-Goetz, M. (2022). Stem cell-derived synthetic embryos self-assemble by exploiting cadherin codes and cortical tension. Nat Cell Biol, 1–9. 75. Stevens, A.J., Harris, A.R., Gerdts, J., Kim, K.H., Trentesaux, C., Ramirez, J.T., McKeithan, W.L., Fattahi, F., Klein, O.D., Fletcher, D.A., et al. (2023). Programming multicellular assembly with synthetic cell adhesion molecules. Nature 614, 144–152. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 76. Cachat, E., Liu, W., Martin, K.C., Yuan, X., Yin, H., Hohenstein, P., and Davies, J.A. (2016). 2- and 3-dimensional synthetic large-scale de novo patterning by mammalian cells through phase separation. Sci Rep 6, 20664. 77. Maître, J.-L., Berthoumieux, H., Krens, S.F.G., Salbreux, G., Jülicher, F., Paluch, E., and Heisenberg, C.-P. (2012). Adhesion Functions in Cell Sorting by Mechanically Coupling the Cortices of Adhering Cells. Science 338, 253–256. 78. Brasch, J., Katsamba, P.S., Harrison, O.J., Ahlsén, G., Troyanovsky, R.B., Indra, I., Kaczynska, A., Kaeser, B., Troyanovsky, S., Honig, B., et al. (2018). Homophilic and Heterophilic Interactions of Type II Cadherins Identify Specificity Groups Underlying Cell- Adhesive Behavior. Cell Reports 23, 1840–1852. 79. Arzate-Mejía, R.G., Valle-García, D., and Recillas-Targa, F. (2011). Signaling epigenetics: Novel insights on cell signaling and epigenetic regulation. IUBMB Life 63, 881– 895. 80. Sawarkar, R., and Paro, R. (2010). Interpretation of Developmental Signaling at Chromatin: The Polycomb Perspective. Developmental Cell 19, 651–661. 81. Kiwimagi, K.A., Letendre, J.H., Weinberg, B.H., Wang, J., Chen, M., Watanabe, L., Myers, C.J., Beal, J., Wong, W.W., and Weiss, R. (2021). Quantitative characterization of recombinase-based digitizer circuits enables predictable amplification of biological signals. Commun Biol 4, 1–12. 82. Kim, T., Weinberg, B., Wong, W., and Lu, T.K. (2021). Scalable recombinase- based gene expression cascades. Nat Commun 12, 2711. 83. Wauford, N., Patel, A., Tordoff, J., Enghuus, C., Jin, A., Toppen, J., Kemp, M.L., and Weiss, R. (2023). Synthetic symmetry breaking and programmable multicellular structure formation. Cell Systems. 84. Park, M., Patel, N., Keung, A.J., and Khalil, A.S. (2019). Engineering Epigenetic Regulation Using Synthetic Read-Write Modules. Cell 176, 227-238.e20. 85. Neumann, E.N., Bertozzi, T.M., Wu, E., Serack, F., Harvey, J.W., Brauer, P.P., Pirtle, C.P., Coffey, A., Howard, M., Kamath, N., et al. (2024). Brainwide silencing of prion protein by AAV-mediated delivery of an engineered compact epigenetic editor. Science 384, ado7082. Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 86. Li, X., Chen, W., Martin, B.K., Calderon, D., Lee, C., Choi, J., Chardon, F.M., McDiarmid, T.A., Daza, R.M., Kim, H., et al. (2024). Chromatin context-dependent regulation and epigenetic manipulation of prime editing. Cell 187, 2411-2427.e25. 87. Grupp, S.A., Kalos, M., Barrett, D., Aplenc, R., Porter, D.L., Rheingold, S.R., Teachey, D.T., Chew, A., Hauck, B., Wright, J.F., et al. (2013). Chimeric Antigen Receptor– Modified T Cells for Acute Lymphoid Leukemia. New England Journal of Medicine 368, 1509– 1518. 88. Chassin, H., Müller, M., Tigges, M., Scheller, L., Lang, M., and Fussenegger, M. (2019). A modular degron library for synthetic circuits in mammalian cells. Nat Commun 10, 2013. 89. Zufferey, R., Nagy, D., Mandel, R.J., Naldini, L., and Trono, D. (1997). Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotechnol 15, 871– 875. While the present invention has been described with reference to the specific embodiments thereof, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the true spirit and scope of the invention. In addition, many modifications may be made to adapt a particular situation, material, composition of matter, process, process step or steps, to the objective, spirit and scope of the present invention. All such modifications are intended to be within the scope of the claims appended hereto.

Claims

Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 CLAIMS What is claimed is:

1. A cell comprising: (a) a first polynucleotide encoding a first binding-triggered transcriptional switch (BTTS), wherein the first BTTS comprises: (i) a first extracellular binding domain that recognizes a first antigen; (ii) a transmembrane domain (iii) one or more protease cleavage domains; and (iv) a first intracellular domain comprising a first DNA binding domain (DBD) and one or more first epigenetic editors; and (b) a nucleic acid comprising a binding site for the first DBD, a first promoter and, operably linked to the promoter, a coding sequence encoding a payload, wherein binding of the first extracellular binding domain to a first antigen results in cleavage of the BTTS at the one or more protease cleavage domains to release the first intracellular domain, and wherein the released first intracellular domain binds to the binding site for the first DBD and decreases expression of the payload.

2. The cell of claim 1, wherein the one or more first epigenetic editors comprise a KRAB domain and / or a Dnmt3 subunit.

3. The cell of claim 1 or 2, further comprising: (c) a second polynucleotide encoding a second BTTS, wherein the second BTTS comprises: (i) a second extracellular binding domain that recognizes a second antigen; (ii) a transmembrane domain (iii) one or more protease cleavage domains; and (iv) a second intracellular domain comprising a second DNA binding domain (DBD) and one or more second epigenetic editors; andAtty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 wherein the nucleic acid of (b) further comprises a binding site for second DBD, wherein binding of the second extracellular binding domain to an antigen results in cleavage of the second BTTS at the one or more protease cleavage domains to release the second intracellular domain, and wherein binding of the released second intracellular domain to the binding site for second DBD further decreases expression of the payload.

4. The cell of any prior claim, wherein the first and second DNA binding domains are independently selected from Gal4-, LexA-, Tet-, Lac-, dCas9-, zinc-finger- and TALE-based DNA binding domains.

5. The cell of any of claims 2-4, wherein the first and second epigenetic editors are independently selected from a KRAB domain, and a DNMT3 subunit.

6. The cell of claim 5, wherein the first epigenetic editor is a KRAB domain and the second epigenetic editor is a Dnmt3 subunit.

7. The cell of any prior claim, wherein the antigens recognized by the first and second extracellular binding domains are different.

8. The system of any prior claim, wherein the payload is a cell fate regulator.

9. The cell of any prior claim, wherein the payload is a transcription factor, morphogen, therapeutic protein or cell surface ligand or receptor.

10. The cell of any of claims 1-8, wherein the payload is a transcriptional repressor.

11. The cell of claim 10, wherein the cell further comprises: (d) a second nucleic acid comprising: a binding site for the transcriptional repressor, a second promoter and a coding sequence for a second payload that is operably linked to the second promoter, wherein the decrease in expression of the transcriptional repressor increasesAtty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 expression of the second payload.

12. The system of any prior claim, wherein the payload is a cell fate regulator.

13. The cell of claim 12, wherein the second payload is a transcription factor, morphogen, therapeutic protein or cell surface ligand or receptor.

14. The cell of any prior claim, wherein the cell is a stem cell.

15. The cell of any prior claim, wherein the cell is a pluripotent stem cell.

16. The cell of any prior claim, wherein the cell is an adult stem cell, differentiated committed cell, or precursor / progenitor cell.

17. The cell of any of claims 1-13, wherein the cell is a therapeutic immune cell.

18. A method for effecting a long-term change in the expression of a payload via a transient extracellular binding event, comprising: transiently contacting a cell of any prior claim with the first antigen, wherein: the first antigen is on a cell, in the extracellular matrix or on a biocompatible material; the first intracellular domain comprises the DBD, a KRAB domain and a Dnmt3 subunit; and the contacting results in a long-term decrease in expression of the first payload.

19. The method of claim 18, wherein the cell comprises the second nucleic acid and the contacting results in a long-term increase in expression of the second payload.

20. A method for effecting a long-term change in the expression of a payload via a transient extracellular binding event, comprising: transiently contacting a cell of claim 3-17 with the first and second antigens, wherein:Atty. Dkt: UCSF-814WO Client Ref.: SF-2025-025-2 the first and second antigens are independently on a cell, extracellular matrix or a biocompatible material, the first epigenetic editor is a KRAB domain and the second epigenetic editor is a Dnmt3 subunit, and the contacting results in a long-term decrease in expression of the first payload.

21. The method of claim 20, wherein the cell comprises the second nucleic acid and the contacting results in a long-term increase in expression of the second payload.

22. The method of claim 20 or 21, wherein the first and second antigens are the same or different.

23. The method of any of claims 20-22, wherein the first and second antigens are on different cells or the first antigen is on a cell and the second antigen is in the extracellular matrix or a biocompatible material.

24. The method of any of claims 20-23, wherein the contacting is done in vivo, in vitro or ex vivo.

25. The method of any of claims 20-24, wherein the transient contacting results in differentiation, transdifferentiation or full reprogramming (to a pluripotent state) or partial reprogramming (to a rejuvenated / more stemlike state) of the cell from one cell type to another.

Images