Methods and reagents for cell-based screening of genetically encoded protein binders

Abstract

Hie disclosure provides methods for cell-based screening of genetically encoded protein binders, involving (a) providing recombinant eukaryotic cells that have a first nucleic acid encoding a membrane protein, and / or a secreted protein; and a second nucleic acid encoding a first fusion protein comprising one or more recombinant protein target-binding candidate fused to (A) a direct cellular compartment retention tag, or (B) an indirect cellular compartment retention tag; wherein the protein target-binding candidate, when expressed, is retained within a cellular compartment by means of the cellular compartment retention tag; (b) culturing the recombinant eukaryotic cell under conditions suitable to express the recombinant protein target and the protein target-binding candidate in the recombinant eukaryotic cell; and (c)detecting localization of the protein target in the recombinant eukaryotic cell in one or more of the cellular compartments.

Description

[0001]Methods and reagents for cell-based screening of genetically encoded protein binders Sequence Listing Statement A computer readable form of the Sequence Listing is filed with this application by electronic submission and is incorporated into this application by reference in its entirety. The Sequence Listing is contained in the file created on June 16, 2025 having the file name “24- 0207-WO.xml” and is 42,091 bytes in size. Background Despite advances, there is a low success rate of de design of genetically encoded protein binders against target membrane proteins, including integral membrane proteins such as G-protein couple receptors (GPCRs), ion channels, and transporters, as well as other intracellular soluble protein targets. For insoluble membrane proteins, soluble ligands can be generated by specialized membrane preparations (and mutational stabilization. However, this requires intensive effort and is available for only a limited number of targets, marking high throughput screening (HTS) of binder libraries to insoluble targets as a major unsolved challenge in drug discovery. Summary In a first aspect, the disclosure provides methods for cell-based screening of genetically encoded protein binders, comprising: (a) providing a recombinant eukaryotic cell comprising: (i) a first nucleic acid encoding a recombinant protein target, wherein the recombinant protein target comprises a membrane protein, and / or a secreted protein; and (ii) a second nucleic acid encoding a first fusion protein comprising one or more recombinant protein target-binding candidate fused, directly or via an amino acid linker, to (A) a direct cellular compartment retention tag, or (B) an indirect cellular compartment retention tag; wherein the protein target-binding candidate, when expressed, is retained within a cellular compartment by means of the cellular compartment retention tag, wherein the protein target, when expressed, is retained in the cellular compartment at a reduced level in the absence of binding to the recombinant protein target-binding candidate compared to protein target retention in the cellular compartment when bound to the recombinant protein target-binding candidate, wherein the cellular compartment comprises one or more of a plasma membrane, and a membrane-enclosed cell compartment; (b) culturing the recombinant eukaryotic cell under conditions suitable to express the recombinant protein target and the protein target-binding candidate in the recombinant eukaryotic cell; and (c) detecting localization of the protein target in the recombinant eukaryotic cell in one or more of the cellular compartments, wherein altered localization or abundance of the protein target in the one or more cellular compartment in the presence of the first fusion protein, compared to protein target localization in the one or more cellular compartment in the absence of the first fusion protein, indicates that the protein target-binding candidate binds to the protein target. In one embodiment, the first fusion protein comprises a recombinant protein target- binding candidate fused to a direct cellular compartment retention tag. In another embodiment, the direct cellular compartment retention tag comprises the amino acid sequence selected from the group consisting of SEQ ID NO:1-4 and 12. In a further embodiment, the direct cellular compartment retention tag comprises the amino acidsequence selected from the group consisting of SEQ ID NO:1-3, wherein the direct cellularcompartment retention tag is located at the C-terminus of the first fusion protein and wherein the cellular compartment is the endoplasmic reticulum. In a further embodiment, the direct cellular compartment retention tag comprises the amino acid sequence of SEQ ID NO:4 or12, wherein the direct cellular compartment retention tag is located at the N-terminus of thefirst fusion protein, and wherein the cellular compartment is the Golgi apparatus. In one embodiment, the fusion protein comprises a recombinant protein target-binding candidate fused to an indirect cellular compartment retention tag, wherein the recombinant eukaryotic cell further comprises: (iii) a third nucleic acid encoding a second fusion protein comprising a second polypeptide fused to a direct cellular compartment retention tag, wherein the second polypeptide is capable of binding to the indirect cellular compartment retention tag. In some embodiments, the second polypeptide comprises an antibody, one portion of a split fluorescent protein (i.e.: wherein the indirect cellular compartment retention tag comprises the other portion of the split fluorescent protein), any protein (naturally occurring or de novo designed) capable of binding an epitope that is present in the indirect cellular compartment retention tag), SpyCatcherTMor variants thereof, HaloTagTMor variants thereof, SNAPTM-tag or variants thereof (such as CLIPTM-tag). In some embodiments, the second polypeptide comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:6; and the indirect cellular compartment tag comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:5. In one such embodiment, the second polypeptide comprises the amino acid sequence of SEQ ID NO:6; and the indirect cellular compartment tag comprises the amino acid sequence of SEQ ID NO:5. In other embodiments, the second polypeptide comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:8, and the indirect cellular compartment tag comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:7. In some such embodiments, the second polypeptide comprises the amino acid sequence of SEQ ID NO:8, and the indirect cellular compartment tag comprises the amino acid sequence of SEQ ID NO:7. In further embodiments, the second polypeptide is fused to a direct cellular compartment retention tag that comprises the amino acid sequence selected from the group consisting of SEQ ID NO:1-4 and 12. In some such embodiments, the direct cellular compartment retention tag comprises the amino acid sequence selected from the groupconsisting of SEQ ID NO:1-3, wherein the indirect cellular compartment retention tag islocated at the C-terminus of the second fusion protein, and wherein the cellular compartment is the endoplasmic reticulum. In other embodiments, the direct cellular compartmentretention tag comprises the amino acid sequence of SEQ ID NO:4 or 12, wherein the indirectcellular compartment retention tag is located at the N-terminus of the second fusion protein, and wherein the cellular compartment is the Golgi apparatus. In one embodiment, the first nucleic acid encodes a protein target fused to an oligomerization domain and / or the second nucleic acid encodes a recombinant protein target- binding candidate fused to an oligomerization domain. In some embodiments, the recombinant protein target is not fused to a detectable domain. In other embodiments, the recombinant protein target is fused to a first detectable polypeptide. In some embodiments, the first fusion protein comprises the recombinant protein target-binding candidate fused to (i) (A) the direct cellular compartment retention tag, or (B) the indirect cellular compartment retention tag, and (ii) a second detectable polypeptide, wherein the second detectable polypeptide is optically distinguishable from the first detectable polypeptide when the recombinant protein target fused to a first detectable polypeptide. In other embodiments, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are present in one or more expression vectors, in which the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are operatively linked to a promoter. In some embodiments, the first nucleic acid and the second nucleic acid are present in one or more expression vectors. In other embodiments, the first nucleic acid is present in a first expression vector, and the second nucleic acid is present in a second expression vector. In further embodiments, the first nucleic acid and the second nucleic acid are present in the same expression vector. In a further embodiment, the third nucleic acid is present, and is present in an expression vector. In one embodiment, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, comprises an mRNA. In one such embodiment, the first nucleic acid, comprises an mRNA. In another embodiment, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are stably expressed in the recombinant eukaryotic cell. In a further embodiment, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are transiently expressed in the recombinant eukaryotic cell. In some embodiments, the protein target is a transmembrane protein or modified version thereof. In some embodiments, the transmembrane protein comprises a G-coupled protein receptor (GPCR), or modified version thereof. In other embodiments, the transmembrane protein comprises another receptor protein or modified version thereof, including but not limited to ion channels, transporters, gap junction proteins, signal transduction proteins, structural proteins, or enzymatic proteins. In some embodiments, the second nucleic acid may comprise a single second nucleic acid encoding a first fusion protein comprising a recombinant protein target-binding candidate fused, directly or via an amino acid linker, to (A) a direct cellular compartment retention tag, or (B) an indirect cellular compartment retention tag. In other embodiments, the second nucleic acid may comprise two or more second nucleic acids of the disclosure, where (for example), the recombinant protein target-binding candidate comprises two or more polypeptide components that interact, and the interacting polypeptide components are tested for binding to the protein target. The second nucleic acid may further encode a secretion signal at the N-terminus of the second fusion protein, so that the recombinant protein target binding candidate is secreted. In some embodiments, a barcode element can be added for tracking the identity of protein target and / or protein target binding candidate. In one embodiment, detecting localization of the protein target in the recombinant eukaryotic cell comprises optically distinguishing the first detectable polypeptide from the second detectable polypeptide. In some embodiments, microscopy or imaging flow cytometry may be used to resolve abundance, localization, and co-localization of target protein and binder candidate. In other embodiments, flow cytometry may be used to measure abundance (total or surface-localized) target protein and binder candidate. In other embodiments, localization of the protein target could alter the activity an enzymatic reporter, such as split antibiotic resistance enzyme, or an enzyme with compartment-dependent activity (e.g., dependent on pH or location of substrate). Independent from optical detection, cells with altered surface localization of the target (e.g., removal of surface localization due to trapping in endoplasmic reticulum) can be physically enriched via binding, e.g., by magnetically assisted cell sorting (MACS) depletion of cells that express target on the surface. The detection may comprise detection of one or more “binding phenotypes”. Exemplary such binding phenotypes include (a) increased co-localization of protein target with protein target-binding candidate, (b) a change in cellular compartment of the target or protein target-binding candidate (for example, the target can be retained in the ER by an ER- localized binder, or a protein target-binding candidate may be trafficked to the surface of the cell (cell membrane) and retained there due to interaction with the target), (c) Increased expression of target or protein target-binding candidate (for example, target retained in the ER is present at a higher overall level due to reduced degradation), (d) a change in localization of both the target and protein target-binding candidate contingent on binding (for example, formation of punctae when target is bound to oligomerized protein target-binding candidate), (e) a change in aggregation of target within a single cell compartment contingent on binding to the protein target-binding candidate, and (f) removal of target or protein target- binding candidate from a cellular compartment (for example, removal of surface-localized (cell membrane) target through interaction with an ER-localized protein target-binding candidate. The detection may further comprise quantifying the binding phenotype, including but not limited to estimate binding affinity, estimate off-rate (using surface retention or competitors), quantifying binding from an allelic series of binders with known on / off rates to epitope tag; predicting affinity or on / off rates based on expression and localization of both target and binder, across experimental conditions (e.g., oligomerization state), including via ML / supervised learning on known binders. In another embodiment, the second nucleic acid comprises a plurality of second nucleic acids, wherein the expression products of different second nucleic acids are detectably distinguishable. In one embodiment, each of the individual second nucleic acids includes a nucleic acid barcode that results in generation of a first fusion protein that is distinguishable from all of the other first fusion proteins encoded by the plurality of second nucleic acids. In a further embodiment, the first nucleic acid comprises a plurality of first nucleic acids, wherein the expression products of different first nucleic acids are different. In one such embodiment, the expression products of different first nucleic acids comprise different recombinant protein targets, and the different protein targets are detectably distinguishable. In one embodiment, each of the individual first nucleic acids includes a nucleic acid barcode that results in generation of a protein target that is distinguishable from other protein targets encoded by the plurality of first nucleic acids. In another such embodiment, the expression products of different first nucleic acids comprise one or more variants of the protein target, wherein at least one variant of the protein target is binding deficient, and wherein the different protein targets are detectably distinguishable. In one embodiment, the method comprises providing a plurality of different recombinant eukaryotic cell clones, wherein each cell clone comprises a first nucleic acid encoding a protein target that is detectably distinguishable from the first nucleic acid and / or protein target in each other cell clone. In one embodiment, each cell clone comprises a first nucleic acid including a nucleic acid barcode that results in generation of a protein target that is distinguishable from all of the other protein targets present in the other cell clones. The disclosure also provides kits, comprising (a) the first nucleic acid of any claim or embodiment disclosed herein; and (b) the second nucleic acid of any claim or embodiment disclosed herein; and optionally (c) the third nucleic acid of any claim or embodiment disclosed herein. In one embodiment of the kits, the first nucleic acid, the second nucleic acid, and the third nucleic acid (when present) are present in one or more expression vectors, operatively linked to a promoter. The disclosure also provides recombinant eukaryotic cell comprising: (a) the first nucleic acid of any claim or embodiment disclosed herein; and (b) the second nucleic acid of any claim or embodiment disclosed herein; and, optionally, (c) the third nucleic acid of any claim or embodiment disclosed herein. In one embodiment, the first nucleic acid, the second nucleic acid, and the third nucleic acid (when present) are present in one or more expression vectors, operatively linked to a promoter. The disclosure also provides libraries, comprising a plurality of recombinant eukaryotic cell clones of any embodiment disclosed herein. Description of the Figures Figure 1: Receptor trapping detected with imaging readout. (A) In the absence of binding, fluorescently tagged receptors traffic to the cell surface while the design is retained separately in the secretory pathway (left), whereas a successful binder colocalizes with the receptor in the secretory pathway (right). (B) Receptor trapping of EpCAM. (C) Correlation between in vitro binder affinities and in-cell affinities quantified by GFP-RFP pixel cross- correlation for soluble targets. Figure 2: Receptor trapping of GPCRs. KDEL-tagged binders were fused to cyclic self-oligomerizing proteins: C3-symmetry (UniProt Q7SIA8) and C5-symmetry (UniProt Q58584), or non-avid: C1 with no self-oligomerization . The negative control binder was a EpCAM-binding protein. Nanobodies targeting GFP was obtained from Fridy et al.2014 (PMID: 25362362) and tagged with RFP and KDEL. The fraction of cells with trapped target was calculated by identifying cells expressing both the binder and the target. A cell was classified as having target trapped if the correlation between GFP and RFP channel intensities exceeded 0.75. Error bars represent the 95% confidence interval. Figure 3: Binding phenotypes for GPCRs: (A) Cells containing low-affinity(kd=3.8 M) anti-GFP-nanobody did not result in trapping of GPCR 1. (B) Cells containinghigh-affinity anti-GFP nanobody (kd=0.7 nM) resulted in trapping of GPCR 1. (C) Cells containing high-affinity anti-GFP nanobody resulting in formation of GFP and RFP rich spots for GPCR 2. Figure 4: High-throughput image-based screening with receptor trapping. (A) Designed binders are synthesized on oligo arrays and cloned into a lentiviral library. (B) Low MOI transduction creates a cell library with one binder design per cell. (C) Binding is quantified by receptor trapping, and (D) in situ sequencing of a DNA barcode reveals the identity of the binder in each cell. Figure 5: Receptor trapping detected with FACS. (A) In the absence of binding, FLAG-tagged GFP-fused receptors traffic to the cell surface, where they can be labeled with fluorescently conjugated anti-FLAG antibodies (left), whereas a successful binder traps the receptor in the ER. (B) To validate the experimental concept, we sorted a population of cells expressing the FLAG-GFP-GPCR-BFP fusion protein. The cell surface was stained using an anti-FLAG antibody conjugated with a brilliant violet fluorophore. The cells also expressed target-binding protein (a KDEL and RFP-tagged anti-GFP-nanobody, kd = 0.7 nM). Cells with a high binder expression had decreased surface staining compared to cells with lower binder expression. Detailed Description and claims All references cited are herein incorporated by reference in their entirety. Within this application, unless otherwise stated, the techniques utilized may be found in any of several well-known references such as: Molecular Cloning: A Laboratory Manual (Sambrook, et al., 1989, Cold Spring Harbor Laboratory Press), Gene Expression Technology (Methods in Enzymology, Vol.185, edited by D. Goeddel, 1991. Academic Press, San Diego, CA), “Guide to Protein Purification” in Methods in Enzymology (M.P. Deutshcer, ed., (1990) Academic Press, Inc.); PCR Protocols: A Guide to Methods and Applications (Innis, et al. 1990. Academic Press, San Diego, CA), Culture of Animal Cells: A Manual of Basic Technique, 2ndEd. (R.I. Freshney.1987. Liss, Inc. New York, NY), Gene Transfer and Expression Protocols, pp.109-128, ed. E.J. Murray, The Humana Press Inc., Clifton, N.J.), RosettaCommons.org, and the Ambion 1998 Catalog (Ambion, Austin, TX). As used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. As used herein, “about” means + / - 5% of the recited value. All embodiments of any aspect of the disclosure can be used in combination, unless the context clearly dictates otherwise. Unless the context clearly requires otherwise, throughout the description and the claims, the words ‘comprise’, ‘comprising’, and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to say, in the sense of “including, but not limited to”. Words using the singular or plural number also include the plural and singular number, respectively. Additionally, the words “herein,” “above,” and “below” and words of similar import, when used in this application, shall refer to this application as a whole and not to any particular portions of the application. As used herein, the amino acid residues are abbreviated as follows: alanine (Ala; A), asparagine (Asn; N), aspartic acid (Asp; D), arginine (Arg; R), cysteine (Cys; C), glutamic acid (Glu; E), glutamine (Gln; Q), glycine (Gly; G), histidine (His; H), isoleucine (Ile; I), leucine (Leu; L), lysine (Lys; K), methionine (Met; M), phenylalanine (Phe; F), proline (Pro; P), serine (Ser; S), threonine (Thr; T), tryptophan (Trp; W), tyrosine (Tyr; Y), and valine (Val; V). Any N-terminal methionine residue in any polypeptide of the disclosure may be present or may be deleted. In a first aspect, the disclosure provides methods for cell-based screening of genetically encoded protein binders, comprising: (a) providing a recombinant eukaryotic cell comprising: (i) a first nucleic acid encoding a recombinant protein target, wherein the recombinant protein target comprises a membrane protein, and / or a secreted protein; and (ii) a second nucleic acid encoding a first fusion protein comprising one or more recombinant protein target-binding candidate fused, directly or via an amino acid linker, to (A) a direct cellular compartment retention tag, or (B) an indirect cellular compartment retention tag; wherein the protein target-binding candidate, when expressed, is retained within a cellular compartment by means of the cellular compartment retention tag, wherein the protein target, when expressed, is retained in the cellular compartment at a reduced level in the absence of binding to the recombinant protein target-binding candidate compared to protein target retention in the cellular compartment when bound to the recombinant protein target-binding candidate, wherein the cellular compartment comprises one or more of a plasma membrane, and a membrane-enclosed cell compartment; (b) culturing the recombinant eukaryotic cell under conditions suitable to express the recombinant protein target and the protein target-binding candidate in the recombinant eukaryotic cell; and (c) detecting localization of the protein target in the recombinant eukaryotic cell in one or more of the cellular compartments, wherein altered localization or abundance of the protein target in the one or more cellular compartment in the presence of the first fusion protein, compared to protein target localization in the one or more cellular compartment in the absence of the first fusion protein, indicates that the protein target-binding candidate binds to the protein target. As disclosed herein, the methods provide a high-throughput, pooled assay for protein binders against target membrane or secreted proteins, including integral membrane proteins such as GPCRs, ion channels, and transporters, as well as other intracellular soluble protein targets. The key advantages over existing methods are (1) fully cell-based, genetically encoded assay requiring no in vitro protein purification or manipulation; (2) preserves endogenous state of target protein, including amino acid sequence, expression level, co- factors, and cellular localization; and (3) ability to screen >105protein binders created by arrayed oligonucleotide synthesis. The recombinant eukaryotic cell may be any eukaryotic cell with a compartmentalizedsecretion pathway, including but not limited to vertebrate, mammalian (including but not limited to human), insect, fungal (including but not limited to yeast), and plant cells. The recombinant eukaryotic cell may be transiently or stably engineered to incorporate the recited nucleic acids, using techniques including but not limited to calcium phosphate co- precipitation, electroporation, or liposome mediated-, DEAE dextran mediated-, polycationic mediated-, or viral mediated transfection. In some embodiments, the cell is a human cell, including but not limited to cell lines commonly used for GPCR expression such as HEK, CHO, U2OS, DLD1, COS-7 and HeLa cells. As used herein, membrane proteins are proteins capable of inserting into a membrane co-translationally or after translation. As used herein, secreted proteins are proteins that are actively transported out of the cell. As used herein a membrane-enclosed cell compartment is any enclosed portion within the cytosol of a eukaryotic cell, usually surrounded by a single or double lipid layer membrane. These compartments are often, but not always, defined as membrane-bound organelles. In some embodiments, the membrane-enclosed cell compartment may be selected from endoplasmic reticulum membrane, Golgi apparatus membrane, lysosome, endosome, periplasm, or cell nucleus. As used herein, a “cellular compartment retention tag” is a peptide domain that causes a protein including it, or bound to it, to be primarily or completely retained in the cellular compartment. The retention tag may be a “direct” retention tag (the tag itself causes retention in the cellular compartment) or an “indirect” retention tag (i.e., a moiety that can be bound by a protein containing the retention tag). In a non-limiting embodiment, an indirect retention tag may comprise a detectable label that can be bound by an antibody or other protein that binds to the label, where the antibody or other protein is fused to a direct retention tag. As used herein, “altered” localization means any change relative to control, including but not limited to increased localization in the cellular compartment. In various embodiments, the altered localization or abundance results in a “binding phenotype” selected from one or more of (a) increased co-localization of protein target with protein target-binding candidate, (b) a change in cellular compartment of the target or protein target-binding candidate (for example, the target can be retained in the ER by an ER-localized binder, or a protein target-binding candidate may be trafficked to the surface of the cell (cell membrane) and retained there due to interaction with the target), (c) Increased expression of target or protein target-binding candidate (for example, target retained in the ER is present at a higher overall level due to reduced degradation), (d) a change in localization of both the target and protein target-binding candidate contingent on binding (for example, formation of punctae when target is bound to oligomerized protein target-binding candidate), (e) a change in aggregation of target with a single cell compartment contingent on binding to the protein target-binding candidate, and (f) removal of target or protein target-binding candidate from a cellular compartment (for example, removal of surface-localized (cell membrane) target through interaction with an ER-localized protein target-binding candidate. As will be understood by those of skill in the art, the “altered localization” may resultin all or a portion of bound target protein being retained in in the cellular compartment. Thealtered localization may comprise, for example a large shift in target localization relative to control, without full co-localization. In one embodiment, the first fusion protein comprises a recombinant protein target- binding candidate fused to a direct cellular compartment retention tag. The recombinant protein target-binding candidate may be present N-terminal or C-terminal to the direct cellular compartment retention tag in the first fusion protein. The fusion may be a direct fusion (no amino acid linker), or may be via an amino acid linker, which can be of any length and amino acid composition as appropriate for an intended construct. In non-limiting embodiments of all linkers described herein, a flexible GS-rich linker of between 6-100, 6-90, 6-80, 6-70, 6-60, 6-50, 6-40, 6-30, or 6-21 amino acids may be used. Non-limiting examples of direct cellular compartment retention tags are provided in Table 1, together with the cellular compartment they are retained in. Table 1. Exemplary direct cellular compartment retention tag In another embodiment, the direct cellular compartment retention tag comprises the amino acid sequence selected from the group consisting of SEQ ID NO:1-4 and 12. In a further embodiment, the direct cellular compartment retention tag comprises the amino acidsequence selected from the group consisting of SEQ ID NO:1-3, wherein the direct cellularcompartment retention tag is located at the C-terminus of the first fusion protein and wherein the cellular compartment is the endoplasmic reticulum. SEQ ID NO:1-3 are members of a family of related sequences (“KDEL tags”), which are localized to the ER through engagement with a KDEL receptor, which actively transports proteins with the C-term KDELmotif from the Golgi to the ER. KDEL tagging can increase the abundance level in the cell ofproteins fused to it, and is compatible for use with protein target-binding candidates that can form oligomers. In a further embodiment, the direct cellular compartment retention tagcomprises the amino acid sequence of SEQ ID NO:4 or 12, wherein the direct cellularcompartment retention tag is located at the N-terminus of the first fusion protein, and wherein the cellular compartment is the Golgi apparatus. In one embodiment, the fusion protein comprises a recombinant protein target-binding candidate fused to an indirect cellular compartment retention tag, wherein the recombinant eukaryotic cell further comprises: (iii) a third nucleic acid encoding a second fusion protein comprising a second polypeptide fused to a direct cellular compartment retention tag, wherein the second polypeptide is capable of binding to the indirect cellular compartment retention tag. As will be understood by those of skill in the art based on the teachings herein, the indirect cellular compartment retention tag may comprise any second polypeptide that (a) the retention tag can be fused to, and (b) is capable of binding to the indirect cellular compartment retention tag. The recombinant protein target-binding candidate may be present N-terminal or C-terminal to the indirect cellular compartment retention tag. The fusion may be a direct fusion (no amino acid linker), or may be via an amino acid linker, which can be of any length and amino acid composition as appropriate for an intended construct. The second polypeptide may be present N-terminal or C-terminal to the direct cellular compartment retention tag in the second fusion protein The fusion may be a direct fusion (no amino acid linker), or may be via an amino acid linker, which can be of any length and amino acid composition as appropriate for an intended construct. In some embodiments, the second polypeptide comprises an antibody. As disclosed herein, "antibody" refers to immunoglobulin molecules and immunologically active portions of immunoglobulin molecules, i.e., molecules that contain an antigen binding site that immunospecifically binds an epitope present in the indirect cellular compartment retention tag. As such, the term antibody encompasses not only whole antibody molecules, but also antibody fragments. Such antibody or antibody fragments thereof may include, but are not limited to monoclonal antibodies, humanized antibodies, chimeric antibodies, Fab', F(ab')2, Fab, Fv, rIgG, recombinant single chain Fv fragments (scFv), nanobodies, diabodies, triabodies, and tetrabodies. In these embodiments, the indirect cellular compartment retention tag comprises a polypeptide that the antibody or nanobody selectively binds to. As will be understood by those of skill in the art based on the present disclosure, any such antibody- antigen combination can be used. In other embodiments, the second polypeptide comprises an antibody, one portion of a split fluorescent protein (i.e.: wherein the indirect cellular compartment retention tag comprises the other portion of the split fluorescent protein), any protein (naturally occurring or de novo designed) capable of binding an epitope that is present in the indirect cellular compartment retention tag), SpyCatcherTMor variants thereof, HaloTagTMor variants thereof, SNAPTM-tag or variants thereof (such as CLIPTM-tag). Other variants are disclosed, for example, at the web site ncbi.nlm.nih.gov / pmc / articles / PMC6539128 / #:~:text=%2C14%5D.-,Table%201,- Summary%20of%20Catcher; and at web site blog.addgene.org / split-fluorescent-proteins-for- studying-protein-protein- interactions#:~:text=Highlighted%20Split%2DFluorescent%20Protein%20Tags, each incorporated by reference in their entirety. Table 2 shows non-limiting, exemplary embodiments of the indirect cellularcompartment retention tag.Table 2. Exemplary indirect cellular compartment retention tag In some embodiments, the second polypeptide comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:6; and the indirect cellular compartment tag comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:5. In one such embodiment, the second polypeptide comprises the amino acid sequence of SEQ ID NO:6; and the indirect cellular compartment tag comprises the amino acid sequence of SEQ ID NO:5. In other embodiments, the second polypeptide comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:8, and the indirect cellular compartment tag comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:7. In some such embodiments, the second polypeptide comprises the amino acid sequence of SEQ ID NO:8, and the indirect cellular compartment tag comprises the amino acid sequence of SEQ ID NO:7. In further embodiments, the second polypeptide is fused to a direct cellular compartment retention tag that comprises the amino acid sequence selected from the group consisting of SEQ ID NO:1-4 and 12. In some such embodiments, the direct cellular compartment retention tag comprises the amino acid sequence selected from the groupconsisting of SEQ ID NO:1-3, wherein the indirect cellular compartment retention tag islocated at the C-terminus of the second fusion protein, and wherein the cellular compartment is the endoplasmic reticulum. In other embodiments, the direct cellular compartmentretention tag comprises the amino acid sequence of SEQ ID NO:4 or 12, wherein the indirectcellular compartment retention tag is located at the N-terminus of the second fusion protein, and wherein the cellular compartment is the Golgi apparatus. In one embodiment, the first nucleic acid encodes a protein target fused to an oligomerization domain and / or the second nucleic acid encodes a recombinant protein target- binding candidate fused to an oligomerization domain. As used herein, the oligomerization domain is any domain that can self-assemble into a multimer (homodimer, homotrimer, homotetramer, homopentamer, homohexamer, etc.). This embodiment is useful, for example, to achieve avidity effects that greatly expand the range of affinities with which binders can be detected. The oligomerization domain may be present N-terminal or C-terminal to the protein target as encoded by the first nucleic acid. The oligomerization domain may be present in any arrangement in the first fusion protein (i.e.: N-terminal to the recombinant protein target-binding candidate and retention tag; C-terminal to the recombinant protein target-binding candidate and retention tag; or between the recombinant protein target-binding candidate and retention tag. In these embodiments, when the second fusion protein comprises a retention tag comprising the amino acid sequence selected from the group consisting of SEQ ID NO:1-3, the retention tag is located at the C-terminus of the second fusion protein. When the second fusion protein comprises a retention tag comprising the amino acid sequence of SEQ ID NO:4, the retention tag is located at the N-terminus of the second fusion protein. In all embodiments, the fusion with the oligomerization domain may be a direct fusion (no amino acid linker), or may be via an amino acid linker, which can be of any length and amino acid composition as appropriate for an intended construct. In some embodiments, the oligomerization domain comprises an amino acid sequence selected from the group consisting of SEQ ID NO:9-11, the sequences of which are shown in Table 3. Table 3. Exemplary oligomerization domains In some embodiments, the recombinant protein target is not fused to a detectable domain. For example, if the protein target is a receptor protein, detection may occur by using an antibody that binds the target regardless of whether the protein target is bound, and then detecting antibody binding using standard techniques. In other embodiments, the recombinant protein target is fused to a first detectable polypeptide. Any detectable polypeptide may be used as suitable for a given use, including but not limited to green fluorescent protein, red fluorescent protein, other fluorescent protein variants, and epitope tags including but not limited to: FLAGTM: DYKDDDDK (SEQ ID:18) FLAGTMx3: DYKDHDGDYKDHDIDYKDDDDK (SEQ ID:19) Myc: EQKLISEEDL (SEQ ID:20) HA: YPYDVPDYA (SEQ ID:21) V5: GKPIPNPLLGLDST (SEQ ID:22) His6: HHHHHH (SEQ ID:23) In some embodiments, the first fusion protein comprises the recombinant protein target-binding candidate fused to (i) (A) the direct cellular compartment retention tag, or (B) the indirect cellular compartment retention tag, and (ii) a second detectable polypeptide, wherein the second detectable polypeptide is optically distinguishable from the first detectable polypeptide when the recombinant protein target fused to a first detectable polypeptide. This embodiment is useful, for example, in, optically distinguishing the first detectable polypeptide from the second detectable polypeptide to facilitate detecting localization of the protein target in the recombinant eukaryotic cell. In other embodiments, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are present in one or more expression vectors, in which the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are operatively linked to a promoter. The promoter may be any promoter suitable for an intended use. In some embodiments, the first nucleic acid and the second nucleic acid are present in one or more expression vectors. In other embodiments, the first nucleic acid is present in a first expression vector, and the second nucleic acid is present in a second expression vector. In further embodiments, the first nucleic acid and the second nucleic acid are present in the same expression vector. In this embodiment, the first nucleic acid and the second nucleic acid may be under the control of the same promoter or may be under control of different promoters. For the case where the first and second nucleic acid are under the same promoter, the first nucleic acid and the second nucleic acid are expressed as a single nucleic acid, and located between the first nucleic acid and the second nucleic acid would be a nucleic acid sequence encoding a self-cleaving linker, including but not limited to the following: P2A: ATNFSLLKQAGDVEENPGP (SEQ ID:24) T2A: EGRGSLLTCGDVEENPGP (SEQ ID:25) In a further embodiment, the third nucleic acid is present, and is present in an expression vector. "Expression vector" includes vectors that operatively link the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, to a promoter sequence (and any other control sequence) capable of effecting expression of the encoded proteins. “Control sequences” operably linked to the nucleic acid sequences of the invention are nucleic acid sequences capable of effecting the expression of the nucleic acid molecules. The control sequences need not be contiguous with the nucleic acid sequences, so long as they function to direct the expression thereof. Thus, for example, intervening untranslated yet transcribed sequences can be present between a promoter sequence and the nucleic acid sequences and the promoter sequence can still be considered "operably linked" to the coding sequence. Other such control sequences include, but are not limited to, polyadenylation signals, termination signals, self-cleaving linkers such as 2A peptides, and ribosome binding sites. Such expression vectors include but are not limited to, plasmid and viral-based (such as lentiviral) expression vectors. The control sequence used to drive expression of the disclosed nucleic acid sequences in a mammalian system may be constitutive (driven by any of a variety of promoters, including but not limited to, CMV, SV40, RSV, actin, EF) or inducible (driven by any of a number of inducible promoters including, but not limited to, tetracycline, ecdysone, steroid-responsive). The expression vector must be replicable in a host cell either as an episome or by integration into host chromosomal DNA. In one embodiment, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, comprises an mRNA. In this embodiment, the nucleic acids are delivered as an mRNA to the cell, without being expressed from an expression vector. In one such embodiment, the first nucleic acid, comprises an mRNA. In another embodiment, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are stably expressed in the recombinant eukaryotic cell. In a further embodiment, the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are transiently expressed in the recombinant eukaryotic cell. In some embodiments, the protein target is a transmembrane protein or modified version thereof. In some embodiments, the transmembrane protein comprises a G-coupled protein receptor (GPCR), or modified version thereof. In other embodiments, the transmembrane protein comprises another receptor protein or modified version thereof, including but not limited to ion channels, transporters, gap junction proteins, signal transduction proteins, structural proteins, or enzymatic proteins. Non-limiting examples of modified versions of protein targets (including transmembrane proteins) include but are not limited to a C-term deletion, deletion of intracellular loops, N-term deletion including native signal peptide, variants known to improve / alter expression or trafficking (e.g., mutations used in structural studies), alterations to trafficking signals (signal peptide, sorting signals, etc.), a chimera of target protein with structurally related protein to improve / alter expression, trafficking, or epitope presentation; e.g., transplanting of epitope to related protein (homolog within or across species), transplanting an epitope (one or several copies) to scaffold protein, (e.g., tandem display of peptide epitope on highly expressed single-pass transmembrane domain), mutations known to alter post-translational modification, (e.g., addition / removal of glycosylation sites). The recombinant protein target binding candidate may be any polypeptide(s) of interest to be assessed for binding to the protein target. In non-limiting embodiments, the protein target binding candidate may comprise a short peptide, a single-chain antibody fragment, an antibody expressed as multiple chains, a de novo-designed protein, etc. In some embodiments, the second nucleic acid may comprise a single second nucleic acid encoding a first fusion protein comprising a recombinant protein target-binding candidate fused, directly or via an amino acid linker, to (A) a direct cellular compartment retention tag, or (B) an indirect cellular compartment retention tag. In other embodiments, the second nucleic acid may comprise two or more second nucleic acids of the disclosure, where (for example), the recombinant protein target-binding candidate comprises two or more polypeptide components that interact, and the interacting polypeptide components are tested for binding to the protein target. For example, the two or more polypeptide components may be co-receptors that associate with the target and are required for signaling or trafficking, or may be soluble co- factors, (such as components of related signaling pathways), or may be engineered co-factors, (such as a conformation-specific nanobody or engineered G protein as co-factors for GPCR target). Such protein co-factors can be expressed as separate polypeptides (i.e., the second nucleic acid may comprise two or more second nucleic acids of the disclosure), or in the same polypeptide (a single second nucleic acid encoding but components, optionally separated by a nucleic acid encoding a cleavable peptide). The second nucleic acid may further encode a secretion signal at the N-terminus of the second fusion protein, so that the recombinant protein target binding candidate is secreted. Any secretion signal can be used, including but not limited to: Mouse IgK: METDTLLLWVLLLWVPGSTGD (SEQ ID:26) Mouse IgH: MGWSCIILFLVATATGVHS (SEQ ID:27) Human serum albumin: MKWVTFISLLFSSAYS (SEQ ID:28) Human insulin: MALWMRLLPLLALLALWGPDPAAA (SEQ ID:29) VSV-G: MKCLLYLAFLFIGVNC (SEQ ID:30) Human OSM: MGVLLTQRTLLSLVLALLFPSMASM (SEQ ID:31) BM40: MRAWIFFLLCLAGRALA (SEQ ID:32) Secrecon: MWWRLWWLLLLLLLLWPMVWA (SEQ ID:33) Human IgKVIII: MDMRVPAQLLGLLLLWLRGARC (SEQ ID:34) CD33: MPLLLLLPLLWAGALA (SEQ ID:35) tPA: MDAMKRGLCCVLLLCGAVFVSPS (SEQ ID:36) Human Chymotrypsinogen: MAFLWLLSCWALLGTTFG (SEQ ID:37) Human trypsinogen-2: MNLLLILTFVAAAVA (SEQ ID:38) Human IL-2: MYRMQLLSCIALSLALVTNS (SEQ ID:39) Influenza Haemagglutinin: MKTIIALSYIFCLVLG (SEQ ID:40) Silkworm Fibroin LC: MKPIFLVLLVVTSAYA (SEQ ID:41) The cells / cell lines disclosed herein may be generated using any suitable technique. In one embodiment, nucleic acids / expression vectors comprising the recited nucleic acids can be introduced transiently into human cells, such as by plasmid transfection. In another embodiment, cell lines stably expressing the nucleic acids / expression vectors can be made, such as by lentiviral transduction or transposase-mediated integration (such as by using PiggyBac transposaseTM). Cell lines with the target nucleic acids / expression vectors can be selected for, such as by antibiotic selection through inclusion of appropriate selection markers in expression vectors. Isogenic cell clones can be made to ensure homogeneity and long-term stability of expression of the nucleic acids. In all embodiments, a barcode element can be added for tracking the identity of protein target and / or protein target binding candidate. For example, a nucleic acid barcode may be placed in the 5’ or 3’ UTR of the first and / or second nucleic acid / associated expression vector. It may be located close to the protein target and / or protein target binding candidate in the nucleic acid for ease of synthesis, sequencing, or to reduce recombination between distal elements. In one embodiment, detecting localization of the protein target in the recombinant eukaryotic cell comprises optically distinguishing the first detectable polypeptide from the second detectable polypeptide. Any suitable means may be used to detect localization of the protein target in the one or more cellular compartments. In some embodiments, microscopy or imaging flow cytometry may be used to resolve abundance, localization, and co- localization of target protein and binder candidate. In other embodiments, flow cytometry may be used to measure abundance (total or surface-localized) target protein and binder candidate. In other embodiments, localization of the protein target could alter the activity an enzymatic reporter, such as split antibiotic resistance enzyme, or an enzyme with compartment-dependent activity (e.g., dependent on pH or location of substrate). Independent from optical detection, cells with altered surface localization of the target (e.g., removal of surface localization due to trapping in endoplasmic reticulum) can be physically enriched via binding, e.g., by magnetically assisted cell sorting (MACS) depletion of cells that express target on the surface. The detection may comprise detection of one or more “binding phenotypes”. Exemplary such binding phenotypes include (a) increased co-localization of protein target with protein target-binding candidate, (b) a change in cellular compartment of the target or protein target-binding candidate (for example, the target can be retained in the ER by an ER- localized binder, or a protein target-binding candidate may be trafficked to the surface of the cell (cell membrane) and retained there due to interaction with the target), (c) Increased expression of target or protein target-binding candidate (for example, target retained in the ER is present at a higher overall level due to reduced degradation), (d) a change in localization of both the target and protein target-binding candidate contingent on binding (for example, formation of punctae when target is bound to oligomerized protein target-binding candidate), (e) a change in aggregation of target within a single cell compartment contingent on binding to the protein target-binding candidate, and (f) removal of target or protein target- binding candidate from a cellular compartment (for example, removal of surface-localized (cell membrane) target through interaction with an ER-localized protein target-binding candidate. The detection may further comprise quantifying the binding phenotype, including but not limited to estimate binding affinity, estimate off-rate (using surface retention or competitors), quantifying binding from an allelic series of binders with known on / off rates to epitope tag; predicting affinity or on / off rates based on expression and localization of both target and binder, across experimental conditions (e.g., oligomerization state), including via ML / supervised learning on known binders. In another embodiment, the second nucleic acid comprises a plurality of second nucleic acids, wherein the expression products of different second nucleic acids are detectably distinguishable. In this embodiment, a cell has a plurality (2, 3, 4, 5, 6, 7, 8, 9, 10, or more) second nucleic acids, and the method can be used to assess binding of a plurality of recombinant protein target-binding candidates for a single protein target. In one embodiment, each of the individual second nucleic acids includes a nucleic acid barcode that results in generation of a first fusion protein that is distinguishable from all of the other first fusion proteins encoded by the plurality of second nucleic acids. In a further embodiment, the first nucleic acid comprises a plurality of first nucleic acids, wherein the expression products of different first nucleic acids are different. In one such embodiment, the expression products of different first nucleic acids comprise different recombinant protein targets, and the different protein targets are detectably distinguishable. In this embodiment, the same binders can be tested for binding to a plurality (2, 3, 4, 5, or more) of protein targets. In one embodiment, each of the individual first nucleic acids includes a nucleic acid barcode that results in generation of a protein target that is distinguishable from other protein targets encoded by the plurality of first nucleic acids. In another such embodiment, the expression products of different first nucleic acids comprise one or more variants of the protein target, wherein at least one variant of the protein target is binding deficient, and wherein the different protein targets are detectably distinguishable. This embodiment is a decoy approach; co-expressing two (or more) versions of the target in the same cell, wherein at least one is a binding-deficient variant. Then binding can lead to differential localization of the two target variants, which can be interpreted on a cell-by-cell basis. This could reduce the likelihood of false positives (e.g., cells that are prone to aggregate or mis-localize the target independent of binding can be filtered out). In one embodiment, the method comprises providing a plurality of different recombinant eukaryotic cell clones, wherein each cell clone comprises a first nucleic acid encoding a protein target that is detectably distinguishable from the first nucleic acid and / or protein target in each other cell clone. In this embodiment, a plurality (2, 5, 10, 25, 50, 100, 500, 1000, 5000, or more) of cell clones are screened, and the method can be used to assess binding of one or a plurality of recombinant protein target-binding candidates for a large number of protein targets. In one embodiment, each cell clone comprises a first nucleic acid including a nucleic acid barcode that results in generation of a protein target that is distinguishable from all of the other protein targets present in the other cell clones. All of these embodiments permit high throughput library screening of variations of protein targets and / or recombinant protein target-binding candidates. In non-limiting examples, multiple protein targets may be screened that are variants biased towards conformational states or chosen to disrupt binding interface, or a mutational scan of residues important for binding, or library of different receptors. The barcodes may comprise nucleic acid barcodes (for example, DNA barcodes, including but not limited to 3’ UTR barcodes), and / or peptide barcodes. The disclosure also provides kits, comprising (a) the first nucleic acid of any claim or embodiment disclosed herein; and (b) the second nucleic acid of any claim or embodiment disclosed herein; and optionally (c) the third nucleic acid of any claim or embodiment disclosed herein. In one embodiment of the kits, the first nucleic acid, the second nucleic acid, and the third nucleic acid (when present) are present in one or more expression vectors, operatively linked to a promoter. The disclosure also provides recombinant eukaryotic cell comprising: (a) the first nucleic acid of any claim or embodiment disclosed herein; and (b) the second nucleic acid of any claim or embodiment disclosed herein; and, optionally, (c) the third nucleic acid of any claim or embodiment disclosed herein. In one embodiment, the first nucleic acid, the second nucleic acid, and the third nucleic acid (when present) are present in one or more expression vectors, operatively linked to a promoter. The recombinant eukaryotic cell may be any eukaryotic cell with a compartmentalizedsecretion pathway, including but not limited to vertebrate, mammalian (including but not limited to human), insect, fungal (including but not limited to yeast), and plant cells. The recombinant eukaryotic cell may be transiently or stably engineered to incorporate the recited nucleic acids, using techniques including but not limited to calcium phosphate co- precipitation, electroporation, or liposome mediated-, DEAE dextran mediated-, polycationic mediated-, or viral mediated transfection. In some embodiments, the cell is a human cell, including but not limited to cell lines commonly used for GPCR expression such as HEK, CHO, U2OS, DLD1, COS-7 and HeLa cells. The disclosure also provides libraries, comprising a plurality (i.e., 10, 25, 50, 100, 250, 500, 1000, 5000, 10,000, 50,000, 100,000, or more) of recombinant eukaryotic cell clones of any embodiment disclosed herein. Examples We describe a novel high-throughput, in situ cell based assay for discovery of genetically encoded protein binders against target membrane proteins (including integral membrane proteins such as GPCRs, ion channels, and transporters) as well as other intracellular soluble protein targets. The key advantages over existing methods are (1) fully human or vertebrate cell-based, genetically encoded assay requiring no in vitro protein purification or manipulation; (2) preserves endogenous state of target protein, including amino acid sequence, expression level, co-factors, and cellular localization; and (3) ability to screen >105protein binders created by arrayed oligonucleotide synthesis. Summary We describe a novel high-throughput, pooled assay for protein binders against target membrane proteins, including integral membrane proteins such as GPCRs, ion channels, and transporters, as well as other intracellular soluble protein targets. The key advantages over existing methods are (1) fully human cell-based, genetically encoded assay requiring no in vitro protein purification or manipulation; (2) preserves endogenous state of target protein, including amino acid sequence, expression level, co-factors, and cellular localization; and (3) ability to screen >105protein binders created by arrayed oligonucleotide synthesis. This method involves the use of optical screening methods to visualize intracellular protein- protein co-localization, of both the protein target and the candidate target binder protein which are co-expressed expressed in a defined higher eukaryotic (e.g. vertebrate) cell line. In one implementation a genetically encoded designed protein binder is localized within the secretory pathway by means of a genetic tag (e.g., endoplasmic reticulum (ER) retention signal), where it can interact with the extracellular face of a membrane protein target. High affinity interaction leads to “trapping” of the target in the ER, which is visualized as a change in subcellular localization. The specific gene sequence of the designed target binder can be identified by in situ optical gene sequencing on a single cell basis. Importantly, this method works on a quantitative basis to identify target binders with target affinities varying over several orders of magnitude. Background Despite advances, the relatively low success rate of de novo binder design to the challenging targets proposed here will necessitate high-throughput screens (HTS) with several key requirements: (1) library scale, able to handle 105oligo array-synthesized binders, (2) rapid readout, commensurate with ~2 week timescale of array synthesis, (3) quantitative affinity measurements, and (4) low false positive rate. Furthermore, methods are restricted by whether the target is soluble (e.g., soluble extracellular domains) or insoluble (e.g., GPCRs and ion channels in this proposal). For soluble targets, yeast surface display meets all key requirements. For insoluble membrane proteins, generation of soluble ligands currently requires intensive effort and is available for only a limited number of targets, marking HTS of binder libraries to insoluble targets as a major unsolved challenge in drug discovery. Method and proof-of-concept To improve screening of peptide binders against membrane protein targets, we developed receptor trapping, a novel, fully genetic, purification-free assay that operates directly in human cells (Fig.1A). In this assay, both the membrane protein target and the candidate binder are expressed in a human cell line. The binder is localized within the secretory pathway by means of a genetic tag (e.g., endoplasmic reticulum (ER) retention signal), where it can interact with the extracellular face of the membrane protein target. High- affinity interaction leads to “trapping” of the target in the ER, which can visualized as a change in subcellular localization (Fig 1A) or absence of target on the cell surface (Fig 3A). The expression format offers several advantages: (i) Target is close to the native expression level, can select appropriate cell line and confirm function, (ii) the binder is expressed and can be post-translationally modified in human cells, must be efficiently translated into ER, and must be soluble in the molecularly crowded environment of the secretory pathway, and (iii) binding happens in the molecularly crowded cellular environment, including a high concentration of other native membrane proteins. As a proof-of-principle, we expressed a membrane protein (EpCAM) and two computationally designed binders shown to have <50 nM affinity to the soluble extracellular domain of EpCAM in an in vitro binding assay. As negative controls, we used binders against other soluble targets that do not bind EpCAM. The anticipated localization was observed: In the absence of binding, EpCAM primarily trafficked to the cell surface, whereas the binder localized to the ER due to its C-terminal KDEL tag (Fig.1B, top). Both designed binders caused EpCAM to co-localize to the ER (Fig.1B, middle and bottom). To determine if binding and co-localization can quantitatively distinguish binders of different affinities, we trapped GFP-tagged soluble targets with binders of known affinity (Fig.1C) and found that the colocalization signal was predictive of affinity over 4 orders of magnitude. The ability to detect both weak and strong binders is crucial, as it will let us use the same system to discover weak candidate binders and perform affinity maturation. To further validate our method for GPCRs, we expressed five N-terminally GFP- fused target GPCRs (human SMO, MC4R, CXCR4, ACM1, and ADGRG1) and co- transfected with known binders and negative controls. In all cases, the fraction of cells with ER-retained target protein was significantly increased when the cells co-expressed known binders (Fig.2). For some GPCRs, we observed an additional “spot”-binding phenotype, where, in addition to broad GFP / RFP correlation through the ER (Fig 3B), binders and targets became clustered in spots in the cell (Fig 3C). Compatibility with high-throughput optical screening High-throughput receptor trapping enables library-scale experiments, including finding rare binder hits and developing them via saturating mutagenesis-based lead optimization. In order to screen libraries of 105computational designs, designs can be synthesized by massively parallel oligo pool synthesis (Fig.4A) and low-MOI lentiviral transduction used to create a cell library containing one design per cell (Fig.4B). Receptor trapping imaging (Fig.4C) followed by in situ sequencing (Fig.4D) is used to identify the design in each cell. This single-cell image-based screening method, scales to millions of cells and takes <2 months from in silico design to data (Feldman 2019, Feldman 2022). To rapidly validate receptor trapping for a target membrane protein, one approach is to tag the target with GFP and measure trapping with KDEL-tagged anti-GFP nanobodies. Since changing the target only requires changing the GFP-tagged DNA sequence, this can be readily scaled to many targets, such as the disease-relevant GPCRome. Extension of the assay to high-throughput screening of binders via commonly used lentiviral vectors requires the assay to work when binders are expressed from singly- integrated library vectors. Furthermore, many high-throughput pooled screens benefit from integration of barcodes that correspond to binder identity. Compatibility was shown for two high-throughput readouts: FACS followed by DNA sequencing, and single-cell microscopy followed by in situ sequencing. Both readouts used cell libraries made by low multiplicity of infection (MOI) lentiviral transduction of a mixture of binders and non- targeting controls, in order to ensure most cells received a single integration. To determine if affinity could be estimated directly from the binding signal, a series of binders to the same target (GFP, presented as an extracellular domain) with known affinity were compared, with and without avidity. Measuring binding signal that is consistent with the known affinity indicates that these binders can be used as a standard to estimate the affinity of binders to the same target with unknown affinity. A monotonic increase in binding signal with increasing affinity was observed across several targets, with greater sensitivity when using avidity. Together, this data indicates that binding signal can quantitatively reflect affinity over 3 orders of magnitude, from ~1 nM to ~1 M. For single-cell microscopy readout, a library of cells was fixed and imaged. Individual cells were identified using Cellpose and the localization pattern of target and binder were analyzed using Python. After fixation, cells were processed following a published in situ sequencing protocol (Feldman, 2019). Barcodes that uniquely identified the binders were read out via padlock-based in situ sequencing. The binding signal for each binder in the library was calculated by combining data for all cells mapped to the barcode for that binder. The resulting dataset showed high sensitivity, with fewer than 100 analyzed cells needed to confidently distinguish binders vs. negative controls. Compatibility with high-throughput Fluorescently Activated Cell Sorting An alternative method for finding KDEL-tagged binding hits is Fluorescently Activated Cell Sorting (FACS). With this method, one selects for cells that express GFP- fused antigen-tagged receptor (high GFP signal), which upon successful binding, stays trapped inside the cell, and is thus not labeled by fluorescently conjugated antibodies targeting the antigen-tag on the target (Fig 5A, right). Cells without working binder traffic that receptor as normal to the surface and is thus have both GFP and antibody signal (Fig 5A, left and Fig 5B). For FACS readout, the target protein was expressed with an extracellular FLAG tag, which was labeled by adding fluorescent anti-FLAG antibodies. A mixture of cells expressing binders or non-targeting controls were sorted by gating for (a) binder expression (measured via an mCherry tag), and (b) reduced ratio of target surface expression (FLAG) to total target expression (GFP tag). Sorted cells were subcultured and compared to the input population by microscopy, confirming that the sorted population was strongly enriched for binders that trapped the target protein intracellularly. Alternative compartment retention tags Early experiments found that among KDEL, Galnt2, and iDimerize retention tags, the KDEL tag provided superior binder expression and altered trafficking of the target. These and later experiments demonstrated that single-pass membrane proteins (PDL1, EpCAM) and multi-pass GPCRs (CXCR4, PAR2, ACM1, and others) could be assayed for binding using known binders (designed miniproteins, nanobodies, human antibodies, and toxins). A series of binders of varying affinity to a common domain (GFP) was used to demonstrate that the assay can measure a range of affinities. Importantly, weak binding can be detected by adding an avidity effect via fusion of a homo-oligomeric domain to the binder (several oligomers were tested), while retaining high binder expression. Adding avidity by the same means to target proteins resulted in diminished expression in some cases (PAR2). Potential modifications to targets and binders In some circumstances, it is desirable to measure binding to particular conformational states of the target protein. To achieve this and increase baseline expression, we tested both soluble miniG expression and fusion of miniG proteins to the intracellular domain of the target GPCRs (these proteins bind and stabilize the active conformation of the GPCR). We found that both soluble and fused miniG improved expression and trafficking of some targets (EP4 receptor), with fused miniG performing better. Fusion of a binder to the KDEL tag can block binding in some cases, as the KDEL tag must be fused to the C-terminus of the binder. To circumvent this, we tested indirect retention, in which the binder was fused on the N-terminus to a tag (mCherry), which was in turn bound by a nanobody fused to KDEL. For a set of control binders, both direct and indirect retention showed comparable binding signal. Agonist screening Using the receptor trapping approach to screen for GPCR agonists, we focused on the MRGPRX1, an emerging target for itch and pain. We targeted a large epitope within the orthosteric binding pocket spanning TM2 to TM7 of three active-state structures, reasoning that active-state stabilization alone would be sufficient to generate agonists. We screened a library of 13,000 designs using receptor trapping, and succeeded in mapping optical binding phenotypes for 800,000 cells to their design genotypes. Averaging optical phenotypes across cells, we ranked each design and selected 64 designs. Many of these designs were highly expressed from E. coli and subsequently screened in a calcium mobilization assay to explore their ability to stimulate intracellular signaling. Consistent with the design strategy, seven identified proteins demonstrated agonistic activity at 10 M (data not shown). This demonstrates that the receptor trapping methodology can be used for agonist identification. References 1. Liu, Q. et al. Functional GLP-1R antibodies identified from a synthetic GPCR- focused library demonstrate potent blood glucose control. mAbs 13, (2021). 2. Robertson, N. et al. Development of a novel mammalian display system for selection of antibodies against membrane proteins. J. Biol. Chem.295, 18436–18448 (2020). 3. Ren, H. et al. Function-based high-throughput screening for antibody antagonists and agonists against G protein-coupled receptors. Commun. Biol.3, 1–10 (2020). 4. Feldman, D. et al. Optical Pooled Screens in Human Cells. Cell 179, 787–799 (2019). 5. Feldman, D. et al. Pooled genetic perturbation screens with image-based phenotypes. Nat. Protoc.17, 476–512 (2022).

Claims

We claim 1. A method for cell-based screening of genetically encoded protein binders, comprising: (a) providing a recombinant eukaryotic cell comprising: (i) a first nucleic acid encoding a recombinant protein target, wherein the recombinant protein target comprises a membrane protein, and / or a secreted protein; and (ii) a second nucleic acid encoding a first fusion protein comprising one or more recombinant protein target-binding candidate fused, directly or via an amino acid linker, to (A) a direct cellular compartment retention tag, or (B) an indirect cellular compartment retention tag; wherein the protein target-binding candidate, when expressed, is retained within a cellular compartment by means of the cellular compartment retention tag, wherein the protein target, when expressed, is retained in the cellular compartment at a reduced level in the absence of binding to the recombinant protein target-binding candidate compared to protein target retention in the cellular compartment when bound to the recombinant protein target-binding candidate, wherein the cellular compartment comprises one or more of a plasma membrane, and a membrane-enclosed cell compartment; (b) culturing the recombinant eukaryotic cell under conditions suitable to express the recombinant protein target and the protein target-binding candidate in the recombinant eukaryotic cell; and (c) detecting localization of the protein target in the recombinant eukaryotic cell in one or more of the cellular compartments, wherein altered localization or abundance of the protein target in the one or more cellular compartment in the presence of the first fusion protein, compared to protein target localization in the one or more cellular compartment in the absence of the first fusion protein, indicates that the protein target-binding candidate binds to the protein target.

2. The method of claim 1, wherein the first fusion protein comprises a recombinant protein target-binding candidate fused to a direct cellular compartment retention tag.

3. The method of any one of claims 1-2, wherein the direct cellular compartment retention tag comprises the amino acid sequence selected from the group consisting of SEQ ID NO:1-4 and 12.

4. The method of any one of claims 1-3, wherein the direct cellular compartment retention tag comprises the amino acid sequence selected from the group consisting of SEQID NO:1-3, wherein the direct cellular compartment retention tag is located at the C-terminusof the first fusion protein and wherein the cellular compartment is the endoplasmic reticulum.

5. The method of any one of claims 1-3, wherein the direct cellular compartmentretention tag comprises the amino acid sequence of SEQ ID NO:4 or 12, wherein the directcellular compartment retention tag is located at the N-terminus of the first fusion protein, and wherein the cellular compartment is the Golgi apparatus.

6. The method of claim 1, wherein the fusion protein comprises a recombinant protein target-binding candidate fused to an indirect cellular compartment retention tag, wherein the recombinant eukaryotic cell further comprises: (iii) a third nucleic acid encoding a second fusion protein comprising a second polypeptide fused to a direct cellular compartment retention tag, wherein the second polypeptide is capable of binding to the indirect cellular compartment retention tag.

7. The method of claim 6, wherein the second polypeptide comprises an antibody, SpyCatcherTMor variants thereof, HaloTagTMor variants thereof, SNAP-tag or variants thereof (such as CLIP-tag).

8. The method of claim 7, wherein the second polypeptide comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:6; and the indirect cellular compartment tag comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:5, or wherein the second polypeptide comprises the amino acid sequence of SEQ ID NO:6; and the indirect cellular compartment tag comprises the amino acid sequence of SEQ ID NO:

5.

9. The method of claim 7, wherein the second polypeptide comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:8, and the indirect cellular compartment tag comprises an amino acid sequence at least 75%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100 identical to the amino acid sequence of SEQ ID NO:7, orwherein the second polypeptide comprises the amino acid sequence of SEQ ID NO:8, and the indirect cellular compartment tag comprises the amino acid sequence of SEQ ID NO:

7.

10. The method of claim 8 or 9, wherein the second polypeptide is fused to a direct cellular compartment retention tag that comprises the amino acid sequence selected from the group consisting of SEQ ID NO:1-4 and 12.

11. The method of claim 10, wherein the direct cellular compartment retention tag comprises the amino acid sequence selected from the group consisting of SEQ ID NO:1-3,wherein the indirect cellular compartment retention tag is located at the C-terminus of thesecond fusion protein, and wherein the cellular compartment is the endoplasmic reticulum.

12. The method of claim 10, wherein the direct cellular compartment retention tagcomprises the amino acid sequence of SEQ ID NO:4 or 12, wherein the indirect cellularcompartment retention tag is located at the N-terminus of the second fusion protein, and wherein the cellular compartment is the Golgi apparatus.

13. The method of any one of claims 1-12, wherein the first nucleic acid encodes a protein target fused to an oligomerization domain and / or the second nucleic acid encodes a recombinant protein target-binding candidate fused to an oligomerization domain.

14. The method of claim 13, wherein the oligomerization domain comprises an amino acid sequence selected from the group consisting of SEQ ID NO:9-11.

15. The method of any one of claims 1-14, wherein the first nucleic acid encodes a recombinant protein target fused to a first detectable polypeptide.

16. The method of any one of claims 1-15, wherein the first fusion protein comprises the recombinant protein target-binding candidate fused to: (i) (A) the direct cellular compartment retention tag, or (B) the indirect cellular compartment retention tag, and (ii) a second detectable polypeptide, wherein the second detectable polypeptide is optically distinguishable from the first detectable polypeptide when the recombinant protein target fused to a first detectable polypeptide.

17. The method of any one of claims 1-16, wherein the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are present in one or more expression vectors, in which the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are operatively linked to a promoter.

18. The method of any one of claims 1-17, wherein the first nucleic acid and the second nucleic acid are present in one or more expression vectors.

19. The method of claim 18, wherein the first nucleic acid is present in a first expression vector, and the second nucleic acid is present in a second expression vector.

20. The method of claim 18, wherein the first nucleic acid and the second nucleic acid are present in the same expression vector.

21. The method of any one of claims 18-20, wherein the third nucleic acid is present, and is present in an expression vector.

22. The method of any one of claims 1-17, wherein the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, comprises an mRNA.

23. The method of any one of claims 1-17, wherein the first nucleic acid, comprises an mRNA.

24. The method of any one of claims 1-23, wherein the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are stably expressed in the recombinant eukaryotic cell.

25. The method of any one of claims 1-23, wherein the first nucleic acid, the second nucleic acid, and / or the third nucleic acid, when present, are transiently expressed in the recombinant eukaryotic cell.

26. The method of any one of claims 16-26, wherein the detecting localization of the protein target in the recombinant eukaryotic cell comprises optically distinguishing the first detectable polypeptide from the second detectable polypeptide.

27. The method of any one of claims 1-26, wherein the second nucleic acid comprises a plurality of second nucleic acids, wherein the expression products of different second nucleic acids are detectably distinguishable.

28. The method of any one of claims 1-27, wherein the first nucleic acid comprises a plurality of first nucleic acids, wherein the expression products of different first nucleic acids are different.

29. The method of claim 28, wherein the expression products of different first nucleic acids comprise different recombinant protein targets, and the different protein targets are detectably distinguishable.

30. The method of claim 28, wherein the expression products of different first nucleic acids comprise one or more variants of the protein target, wherein at least one variant of the protein target is binding deficient, and wherein the different protein targets are detectably distinguishable.

31. The method of any one of claims 1-30, wherein the method comprises providing a plurality of different recombinant eukaryotic cell clones, wherein each cell clone comprises a first nucleic acid encoding a protein target that is detectably distinguishable from the first nucleic acid and / or protein target in each other cell clone.

32. A kit, comprising: (a) the first nucleic acid of any claim or embodiment disclosed herein; and (b) the second nucleic acid of any claim or embodiment disclosed herein.

33. The kit of claim 32, further comprising the third nucleic acid of any claim or embodiment disclosed herein.

34. The kit of claim 32 or 33, wherein the first nucleic acid, the second nucleic acid, and the third nucleic acid (when present) are present in one or more expression vectors, operatively linked to a promoter.

35. A recombinant eukaryotic cell comprising: (a) the first nucleic acid of any claim or embodiment disclosed herein; and (b) the second nucleic acid of any claim or embodiment disclosed herein.

36. The recombinant eukaryotic cell of claim 35, further comprising the third nucleic acid of any claim or embodiment disclosed herein. The recombinant eukaryotic cell of claim 35 or 36, wherein the first nucleic acid, the second nucleic acid, and the third nucleic acid (when present) are present in one or more expression vectors, operatively linked to a promoter.

38. A library, comprising a plurality of recombinant eukaryotic cell clones of any claim or embodiment disclosed herein.

Images