De novo designed protein binders and modulators of the insulin receptor

Abstract

Polypeptides are disclosed having an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from SEQ ID NO: 1-28, not including any amino acid insertions at identified insertion sites, wherein the polypeptide binds to the F 1 domain of insulin receptor; or SEQ ID NO: 29-31, wherein the polypeptide binds to the LI domain of insulin receptor; fusion proteins thereof, and. methods for their use.

Description

[0001]De Novo Designed Protein Binders and Modulators of the Insulin Receptor Federal Funding Statement This invention was made with government support under Grant Nos. DK063608 and DK132361 and DK132710 and GM142937, awarded by the National Institutes of Health. The government has certain rights in the invention. 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 August 25, 2025 having the file name “24-1484-WO” and is 73,955 bytes in size. Background Insulin receptor (IR), a receptor tyrosine kinase, plays an important role in metabolism, development, growth, and proliferation. Insulin activated IR undergoes trans- autophosphorylation and phosphorylates a number of intracellular substrates, triggering two major signaling pathways – protein kinase B (AKT) pathway and the MAP Kinase (MAPK) pathway. The phosphorylation cascades control trafficking events, transcription factors, glucose and lipid metabolism, and growth in different tissues and pathophysiological conditions. Dysregulation of IR signaling causes diseases including diabetes, cancer, and aging. IR is a dimer composed of two protomers linked by disulfide bonds. The structures of IR in inactive and active states have been determined. Insulin binding at two distinct sites in the extracellular domains of IR promotes a conformational change of IR which reduces the distance between the intracellular kinase domains which leads to trans-autophosphorylation.The primary insulin binding site (site-1) is located in the L1 / -CT domains of IR, while thesecondary insulin binding site (site-2) is located on the side of the F1 domain. It has been proposed that the conformational dynamics of IR in different ligand-bound active states are critical for their biological functions. However, because of the complexity of these large complexes, and the similarity in conformation observed for ligands that exert different effects on signaling, achieving a full understanding of how receptor conformational dynamics are linked to diverse downstream signals, trafficking, and biological outcomes is a current challenge. Summary In a first aspect, the disclosure provides polypeptides comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:13, not including any amino acid insertions at identified insertion sites (i.e., any insertions are not considered when determining percent identity to the reference polypeptide), wherein the polypeptide binds to the F1 domain of insulin receptor. In various embodiment, relative to SEQ ID NO:13 are selected from substitutions listed in Options 1, 2, or 3 of Table 1. In another embodiment, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, or all 18 identified interface are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:13. In a further embodiment, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or all 12 core residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:13. In one embodiment, all identified interface residues and key residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:13. In another embodiment, the polypeptides comprise an insertion in one or more insertion sited relative to the reference sequence. In a second aspect, the disclosure provides polypeptides comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:14, not including any amino acid insertions at identified insertion sites (i.e., any insertions are not considered when determining percent identity to the reference polypeptide), wherein the polypeptide binds to the F1 domain of insulin receptor. In various embodiments, substitutions relative to SEQ ID NO:14 are selected from substitutions listed in Options 1, 2, or 3 of Table 2. In another embodiment, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or all 16 identified interface are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:14. In a further embodiment, at least 1, 2, 3, 4, 5, 6, 7, or all 8 core residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:14. In one embodiment, all identified interface residues and key residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:14. In another embodiment, the polypeptide comprises an insertion in one or more insertion sited relative to SEQ ID NO:14. In various embodiments, the disclosure provides polypeptides comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:1-12 and 15-28, wherein the polypeptide binds to the F1 domain of insulin receptor. In one embodiment, the disclosure provides polypeptides comprising an amino acid sequence at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:29-31, wherein the polypeptide binds to the L1 domain of insulin receptor. In a further embodiment, amino acid substitutions relative to the reference sequence are conservative amino acid substitutions. The disclosure also provides conjugate, comprising the polypeptide of any embodiment or combination of embodiments herein; and a therapeutic or diagnostic moiety. In one embodiment, the therapeutic or diagnostic moiety comprises insulin. In another embodiment, the disclosure provides fusion proteins comprising: (a) a first polypeptide of any embodiment or combination of embodiments herein that binds to the F1 domain of insulin receptor; and (b) a second polypeptide of any embodiment or combination of embodiments herein that binds to the binds to the L1 domain of the insulin receptor, or a polypeptide comprising an amino acid sequence at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:47 and which binds to the L1 domain of the insulin receptor; optionally wherein the first polypeptide and the second polypeptide are linked by an amino acid linker. In another embodiment, the disclosure provides fusion proteins comprising: (a) a first polypeptide of any embodiment or combination of embodiments herein that binds to the F1 domain of insulin receptor; and (b) a second polypeptide of any embodiment or combination of embodiments herein that binds to the F1 domain of insulin receptor; optionally wherein the first polypeptide and the second polypeptide are linked by an amino acid linker. In various embodiments, the fusion protein comprises an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:47-74. The disclosure also provides nucleic acids encoding the polypeptide or fusion protein of any embodiment or combination of embodiments herein, expression vectors comprising the nucleic acid operatively linked to a promoter, host cells comprising the polypeptide, fusion protein, nucleic acid, or expression vector of any embodiment or combination of embodiments herein, and pharmaceutical compositions, comprising the polypeptide, fusion protein, nucleic acid, expression vector, or host cell of any embodiment or combination of embodiments herein, and a pharmaceutically acceptable carrier. The disclosure also provides methods for using, or a use of, the polypeptide, conjugate, fusion protein, nucleic acid, expression vector, host cell, and / or the pharmaceutical composition of any embodiment or combination of embodiments herein, for any suitable purpose including but not limited to those disclosed herein, including but not limited to treating or limiting development of type 1 or type 2 diabetes, type 1 or type 2 diabetes, cancer, autoimmune disease, or thyroid eye disease. In other embodiments, the purpose includes, but is not limited to, enhancing insulin sensitivity in type 2 diabetes and regulating glucose and / or lipid metabolism in type 1 diabetes and congenital severe insulin resistance. Description of the Figures Figure 1. Insulin receptor agonist design strategy. (A) Schematic representation of the structure of the inactive, apo-IR and insulin-bound, active IR. Protomer 1 and protomer 2. Leucine-rich repeats 1 and 2 (L1, L2), Cysteine-rich (CR), Fibronectin type III-1, 2, and 3(F1, F2, and F3), C-terminal tail of -subunit ( -CT), tyrosine kinase (TK) domains. Domainsin protomers 1 and 2 are shown as L1-TK and L1’-TK’, respectively. Site-1 insulins and Site- 2 insulins. (B) Close up view of the tripartite interaction between insulin and IR L1’, -CT, and F1 domains (Site-1 insulin), and the interaction between insulin and IR F1 domain (Site- 2 insulin) (left). Target domains and designer binders (right). (C) Design strategy to regulate IR conformations and dynamics by flexibly or rigidly fusing site-1 and site-2 binders. When S1B and S2B are fused rigidly, the relative positions of the L1 and F1 domains of IR are fixed. In contrast, when they are fused with a flexible linker, the ligand-bound receptor should retain flexibility. (D) Scheme of protein binder development approach. Computationally designed protein binders will be screened using yeast display and sorted by FACS. The enriched binders will then be identified by NGS and further characterized experimentally. Representative designed IR site-2 binders are shown in the top panel. Figure 2. Characterization of IR site-2 binder. (A) Design model of S2B:IR F1 domain complex. The F1 domain is shown as transparent surface according to electrostatic potential and S2B is illustrated as cartoons with key interaction residue side chains shown as sticks. (B) Superposition of S2B:IR F1 domain complex design model and insulin:F1 complex (PDB:6PXV). S2B binds to IR F1 domain at similar areas as insulin with different topology. (C) Biolayer interferometry characterization of binding of S2B to IR F1 domain. Global kinetic fit was reported. (D) CD spectra of S2B at various temperatures. S2B slightly unfolds at 95 °C but refolded when the temperature drops. (E) IR signaling in DKO-IR-B cells treated with 10 nM wild-type (WT) or site-specific insulin mutants with or without 100 nM site-2 binder (S2B) for 10 minutes. (F) Quantification of the western blot data shown in (E). Levels of autophosphorylation were normalized to total IR levels and shown as intensities relative to that in WT insulin alone. Mean ± sem. N=3 independent experiments. Significance calculated using two-tailed student’s t-test. (G) Cryo-EM model of Insulin / S2B / IR complex. Two protomers are shown. Insulin and S2B are shown in cartoons. The cryo-EM density is shown as a transparent surface. PM, plasma membrane. Figure 3. Effects of site-1 binder and site-2 binder. Related to Fig2. (A) Yeast display of site-2 binder designs in presence of 1 nM of site-2 target protein (IR F1 domain). (B) Representative structures of identified designed IR site-2 binders are shown. (C) CD spectra of insulin at various temperatures. Insulin unfolds as temperature increases and cannot be recovered when the temperature returns to room temperature (25 °C). (D) Biolayer interferometry characterization of binding of Insulin to IR F1 domain. Global kinetic fit was reported. (E) IR signaling in DKO-IR-A cells treated with 10 nM wild-type (WT) or site- specific insulin mutants with or without 100 nM site-2 binder (S2B) for 10 minutes. (F) Quantification of the western blot data shown in (E). Levels of phosphorylation were normalized to total protein levels and shown as intensities relative to that in WT insulin alone. Mean ± sem. N=3 independent experiments. Significance calculated using two-tailed student’s t-test. (G) Biolayer interferometry characterization of binding of S2-F1-S1 to IR and IGF1R. (H) IR signaling in DKO-IR-B cells treated with the indicated ligands for 10 minutes: 0, 1, 10, 100 nM insulin, 100 nM S1B, S2B or S1B+S2B.(I) Quantification of the western blot data shown in (H). Levels of phosphorylation were normalized to total protein levels and shown as intensities relative to that in 100 nM insulin alone. Mean ± sem. N=3 independent experiments. Significance calculated using two-tailed student’s t-test. Figure 4. Cryo-EM analysis of the Insulin / S2B / IR complex. Related to Figure.2. (A) Representative size-exclusion chromatogram of mouse IR (mIR). (B) The peak fractions in (A) were visualized on SDS-PAGE by Coomassie blue staining. (C) Representative electron micrograph and 2D class averages of the insulin / S2B / IR complex. (D) Unsharpened cryo-EM map by local resolution. (E) The gold-standard Fourier Shell Correlation (FSC) curve for the cryo-EM map shown in Figure 2. (F) Flowchart of cryo-EM data processing. Figure 5. Develop de novo IR agonists. (A) Design of Site-1 and Site-2 binder fusions. To arrange L1’ domain and F1 domain into active conformation, S1B and S2B were fused with either flexible linkers or rigid interdomain connection. The N-terminus and C- terminus of S1B or S2B were highlighted with dots. (B) An example of design trajectory for building linker between the two binding domains. (C) RFdiffusionTM-generated backbones were designed using ProteinMPNNTMand predicted as monomers by AlphaFold2TM. Thesedesigns were aligned with the input structure, and those with a pLDDT score >85 and C -RMSD <1.5 were selected. The selected designs were then further predicted in complexes with the target PDB (L1+F1) and filtered again. (D) Design models and biolayer interferometry characterization of binding of S2-F1-S1, RF-405, and RF-409 to IR-ECD. Global kinetic fit was plotted. (E) CD spectra of S2-F1-S1, RF-405, and RF-409 at various temperatures. Figure 6. CryoEM structures of RF-405 / IR and S2-F1-S1 / IR complexes. (A) Cryo-EM model of RF-405 / IR complex. (B) Top view of RF-405 / IR complex structure. The design binding L1 and F1 domains in each protomer were labeled. (C) Overall structure of S2-F1-S1 / IR complex. S2-F1-S1 is shown as cartoon. (D) Top view of S2-F1-S1 / IR complex structure. (E) Close-up view of RF-405 binding at the L1 domain of one protomer. The S1B / IR L1 domain design model was overlaid. The aligned binder RMSD calculated by ChimiraXTMis 0.60 Å. (F) Close-up view of RF-405 binding at the F1 domain of one protomer. The S2B / IR F1 domain design model was overlaid and the aligned binder RMSD is 0.89 Å. (G) Close-up view of the binding of RF-405 at the L1’ and F1 domains of IR. Trp65 of RF-405 is sandwiched between the Site-1 and Site-2 binding components and highlighted. (H) Overlay of S2-F1-S1 and RF-405 aligned with the site-2 binding component. The linker of S2-F1-S1 is highlighted. Figure 7. Structures of RF-405 / IR, S2-F1-S1 / IR, and S1-F8-S2 / IR. Related to Figure 4. (A) Representative size-exclusion chromatogram of human IR (hIR). (B) The peak fractions in (A) were visualized on SDS-PAGE by Coomassie blue staining. (C) Representative size-exclusion chromatogram of RF-405 / hIR complex. (D) The peak fractions in (C) were visualized on SDS-PAGE by Coomassie blue staining. (E) Representative size- exclusion chromatogram of S2-F1-S1 / hIR complex. (F) The peak fractions in (E) were visualized on SDS-PAGE by Coomassie blue staining. (G) Cryo-EM density of RF-405 / IR and S2-F1-S1 / IR. (H) Overlay of the insulin / IR structure (PDB:6pxv) with the RF-405 / :IR complex structure. (I) Overlay of the RF-405 / IR design model with the RF-405 / IR complex structure. The aligned binder RMSD is 1.06 Å. (J) Representative 2D class averages of particles of S1-F8-S2 / IR complex. S1-F8-S2 was not able to induce stable conformation and the particles were highly heterogeneous. Figure 8. Cryo-EM analysis of the RF-405 / IR complex. Related to Figure.4. (A) Representative electron micrograph and 2D class averages of the RF-405 / IR complex. (B) Unsharpened cryo-EM map by local resolution. (C) The gold-standard Fourier Shell Correlation (FSC) curve for the cryo-EM map shown in Figure 4. (D) Flowchart of cryo-EM data processing. Figure 9. Cryo-Em analysis of the S2-F1-S1 / IR complex. Related to Figure.4. (A) Representative electron micrograph and 2D class averages of the S2-F1-S1 / IR complex. (B) Unsharpened cryo-EM map by local resolution. (C) The gold-standard Fourier Shell Correlation (FSC) curve for the cryo-EM map shown in Figure 4. (D) Flowchart of cryo-EM data processing. Figure 10. Varying binder orientation and stability results in different cell signaling outcomes. (A) Insulin triggers IR trans-autophosphorylation at multiple tyrosine residues in intracellular domains and activates two major signaling pathways. Activated IR undergoes endocytosis. The MAPK pathway promotes IR endocytosis. Key autophosphorylation residues in intracellular domains were shown. JM, juxtamembrane; TK, tyrosine kinase; CT, C-terminal domains of IR. (B) IR signaling in DKO-IR-B cells treated with the indicated ligands for 10 minutes. Mean ± sem. N = at least 3 independent experiments. Significance calculated using two-tailed student’s t-test. p values vs insulin. **p<0.01, ***p<0.001, and ****p<0.0001. (C-E) IR signaling in C2C12-IR cells by the indicated concentrations of ligands for 10 minutes. Levels of protein phosphorylation were normalized to total protein levels and shown as intensities relative to that in 100 nM insulin- treated cells. Data were fit by nonlinear regression. Mean ± sem. N= at least 3 independent experiments. (F-I) IR autophosphorylation in C2C12-IR cells by the indicated concentrations of ligands for 10 minutes. Levels of IR autophosphorylation were normalized to total IR levels and shown as intensities relative to that in 100 nM insulin-treated cells. Data were fit by nonlinear regression. Mean ± sem. N= at least 3 independent experiments. (J) Cell proliferation. C2C12-IR cells were incubated with the indicated concentrations of ligands for 24 hours and then incorporated Bromodeoxyuridine (BrdU) for 2 hours. BrdU-positive cells were analyzed by FACS. Mean ± sem. N= at least 3 independent experiments. Significance calculated using 2-way ANOVA. P value vs insulin. **p<0.01. (K) Quantification of cell surface IR at 30 minutes after incubation with 100 nM indicated ligands in primary mouse hepatocytes. Mean ± sem. N= 3 independent experiments. Significance calculated using two- tailed student’s t-test. p values vs insulin. **p<0.01. (L) Quantification of total IR at 30 minutes after incubation with 100 nM indicated ligands in primary mouse hepatocytes. Mean ± sem. N= 3 independent experiments. Figure 11. Antagonistic effects of S1-F8-S2. Related to Figure 4. (A) The sequence of linkers in flexibly-linked ligands. (B) IR signaling in DKO-IR-B cells by insulin and S1- F8-S2. Cells were treated with the indicated concentrations of S1-F8-S2 for 1 hour and then treated with 10 nM insulin for 10 minutes. (C) IR signaling in C2C12-IR cells by insulin and S1-F8-S2. Cells were treated with 100 nM S1-F8-S2 for 1 hour and then treated with the indicated concentrations of insulin for 10 minutes. (D) Quantification of the western blot data shown in (B). Levels of phosphorylation were normalized to total IR levels and shown as intensities relative to that in WT insulin alone. Mean ± SD. N=3 independent experiments. Significance calculated using 2-way ANOVA. *p<0.05, ***p<0.001, and ****p<0.0001. (E) Quantification of the western blot data shown in (C). Levels of phosphorylation were normalized to total IR levels and shown as intensities relative to that in WT insulin alone. Mean ± SD. N=4 independent experiments. Significance calculated using 2-way ANOVA. *p<0.05 and p<0.01. (F) Cell proliferation. C2C12-IR cells were incubated with the indicated concentrations of S1-F8-S2 for 24 hours in the presence or absence of 10 nM insulin. Cells were then incorporated with Bromodeoxyuridine (BrdU) for 2 hours. BrdU-positive cells were analyzed by FACS. Mean ± SD. N= 4 independent experiments. Significance calculated using 2-way ANOVA. **p<0.01 and ***p<0.001. (G) Summary of EC50. Mean ± sem. EC50 values obtained from dose-response curves in Fig.5C-I. p values were calculated by Extra sum-of-squares F Test in Prism, between insulin and ligands. p values vs insulin. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. p value vs S2-F5-S1,ap<0.05. ND, not determined. Figure 12. Designed IR agonists activate insulin resistant IR mutants (A) Close up view of binding of RF-405 at IR L1 domain (left), and insulin at IR L1 domain (right). Major interacting residues are represented as sticks. (B) Close up view of binding of RF-405 at IR F1 domain (left), and insulin at IR F1 domain (right). Major interacting residues are represented as sticks. (C) IR autophosphorylation by 10 nM insulin, S2-F1-S1, and RF-405 for 10 minutes in 293FT cells expressing WT IR or the indicated IR mutants. (D) Quantification of the western blot data shown in (C). Levels of IR autophosphorylation were normalized to total IR levels and shown as intensities relative to that in WT IR in insulin- treated cells. Mean ± SD. Significance calculated using 2-way ANOVA. N= at least 4 independent experiments. (E) IR autophosphorylation by 10 nM insulin, S2-F1-S1, and RF- 405 for 10 minutes in 293FT cells expressing WT IR or the indicated disease-causing IR mutants. (F) Quantification of the western blot data shown in (E). Levels of IR autophosphorylation were normalized to total IR levels and shown as intensities relative to that in WT IR in insulin-treated cells. Mean ± sem. Significance calculated using Two-way ANOVA. N= at least 4 independent experiments. Figure 13. Designed IR agonists tune IR signaling and reduce glucose levels. (A) Illustration of the mouse experiment with normal chow and high-fat diets.2- to 3-month-old male mice were used. (B) Insulin tolerance test in mice fed normal chow diet. Mice were injected intraperitoneally with Humulin, S2-F1-S1 or RF-409 at the indicated doses, and their blood glucose levels measured at the indicated time points after injection. Mean ± sem. Humulin, N=8; S2-F1-S1 and RF-409, N=6. Significance calculated using 2-way ANOVA. p value vs Humulin. **p<0.01, ***p<0.001, and ****p<0.0001. (C) Insulin tolerance test in mice fed a high-fat diet. Mean ± sem. Humulin, N=8; S2-F1-S1 and RF-409, N=7 mice per group. Significance calculated using 2-way ANOVA. p value vs Humulin. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. (D) Insulin tolerance test in mice fed a high-fat diet. Mean ± sem. Humulin, N=8; S2-F1-S1 and RF-409, N=7 mice per group. Significance calculated using 2-way ANOVA. p value vs Humulin with the same dose. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. (E) Overview of efficacy (Emax relative to insulin-activated IR WT) and property of IR agonists. Mean ± SD. Significance calculated using 1-way ANOVA, p value vs insulin. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. Bar was shown to represent fold changes compared to insulin functions. Values greater than insulin were displayed in the same size bar as insulin maximum values. N.D, not determined. (F) IR structural stability during activation is essential for downstream signaling, trafficking, and biological function. Rigidly-linked agonists (RF-405 and RF-409, rigid agonists) enhance the conformational stability of active IR, triggering autophosphorylation of IR at multiple tyrosine residues in three intracellular domains, balancing downstream signaling, and IR endocytosis. Flexibly-linked agonists (S2-F1-S1 and S2-F5-S1, flexible agonists) induce a similar extended T-shaped IR. However, due to the flexibility of the linker, tyrosine phosphorylation in the C-terminal domain of IR is greatly reduced, resulting in reduced activation of the MAPK pathway, inhibition of IR endocytosis, and partial agonism. Although S1-F8-S2 constructs can break the auto-inhibitory conformation of IR, their linker orientation hinders the formation of a stable active conformation, thus inhibiting IR activation. Instead, by competing with insulin for the same binding surfaces of IR, S1-Fn-S2 acts as an antagonist (flexible antagonists). Figure 14. Designed IR agonists activate IR signaling and control glucose levels in mice. Related to Figure 7. (A) IR signaling in primary mouse hepatocytes Treated with the indicated ligands for 10 minutes. (B) Quantification of the western blot data shown in (A). Levels of phosphorylation were normalized to total protein levels and shown as intensities relative to that in WT insulin. Mean ± SD. N = 3 independent experiments. Significance calculated using 2-way ANOVA. p values vs insulin. *p<0.05. (C) IR signaling in the liver and skeletal muscle of mice. Each lane contains lysate from an individual mouse. (D) Quantification of data in (C). Levels of protein phosphorylation were normalized to total protein levels and shown as intensities relative to that in insulin-treated conditions. Mean ± sem. N= 4 mice per group. PBS, N=2. Significance calculated using 1-way ANOVA. (E) Insulin tolerance test in mice fed normal chow diet. Mice were injected intraperitoneally with S2-F1-S1 or RF-409 at the indicated doses, and their blood glucose levels measured at the indicated time points after injection. Mean ± sem. N= 7 mice per group. (F) Glucose area under the curve during ITT in Fig.7B and Fig. S7E. Significance calculated using two-tailed student’s t-test. (G) Glucose area under the curve during ITT in Fig.7C. Significance calculated using two-tailed student’s t-test. (H) Glucose area under the curve during ITT in Fig.7D. Significance calculated using two-tailed student’s t-test. Detailed Description / 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.), Dang, B. et al. SNAC-tag for sequence-specific chemical protein cleavage. Nat. Methods 16, 319–322 (2019), 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, 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 all embodiments of the polypeptides disclosed herein, 1, 2, 3, 4, or 5 residues may be deleted from the N-terminus and / or the C-terminus of the polypeptide while retaining activity. 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. In a first aspect, the disclosure provides polypeptides comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:13, not including any amino acid insertions at identified insertion sites (i.e., any insertions are not considered when determining percent identity to the reference polypeptide), wherein the polypeptide binds to the F1 domain of insulin receptor. SKLEEIEELLKELSKTNPLAKDILWVIEVRTEDGHDPKSELVFIRQYLKTLNTPEAREILKI VAP (S2B1; SEQ ID NO:13) In this aspect the polypeptides bind to the F1 domain of insulin receptor and can be used, for example, as a component of a fusion protein functioning as insulin receptor agonists, thereby enhancing insulin sensitivity in type 2 diabetes and regulating glucose and lipid metabolism in type 1 diabetes and congenital severe insulin resistance. In some embodiments, the polypeptide comprises an amino acid sequence at least 75% identical to the amino acid sequence of SEQ ID NO:13. In other embodiments, the polypeptide comprises an amino acid sequence at least 90% identical to the amino acid sequence of SEQ ID NO:13. In further embodiments, substitutions relative to SEQ ID NO:13 are selected from substitutions listed in Options 1, 2, or 3 of Table 1. Site saturation mutagenesis studies were conducted SEQ ID NO:13 (S2B1), and Table 1 provides a list of best substitutions (Option 1; increased or maintained activity), and tolerable substitutions (Options 2 or 3: retained activity, though at a lower level than SEQ ID NO:13), that identified mutations that can be incorporated into SEQ ID NO:13. In some embodiments, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, or all 18 identified interface are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:13. Interface residues (i.e., at the binding interface between the polypeptide and the F1 domain of insulin receptor) relative to SEQ ID NO:13 are provided in column 5 of Table 1. As used throughout the application, conservative amino acid substitutions involve replacing a residue by a residue having similar physiochemical characteristics, e.g., substituting one aliphatic residue for another (such as Ile, Val, Leu, or Ala for one another), or substitution of one polar residue for another (such as between Lys and Arg; Glu and Asp; or Gln and Asn). Other such conservative substitutions, e.g., substitutions of entire regions having similar hydrophobicity characteristics, are known. Amino acids can be grouped according to similarities in the properties of their side chains (in A. L. Lehninger, in Biochemistry, second ed., pp.73-75, Worth Publishers, New York (1975)): (1) non-polar: Ala (A), Val (V), Leu (L), Ile (I), Pro (P), Phe (F), Trp (W), Met (M); (2) uncharged polar: Gly (G), Ser (S), Thr (T), Cys (C), Tyr (Y), Asn (N), Gln (Q); (3) acidic: Asp (D), Glu (E); (4) basic: Lys (K), Arg (R), His (H). Alternatively, naturally occurring residues can be divided into groups based on common side-chain properties: (1) hydrophobic: Norleucine, Met, Ala, Val, Leu, Ile; (2) neutral hydrophilic: Cys, Ser, Thr, Asn, Gln; (3) acidic: Asp, Glu; (4) basic: His, Lys, Arg; (5) residues that influence chain orientation: In other embodiments, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or all 12 core residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:13. Residues that are present at the polypeptide core are shown in column 6 of Table 1. These residues help maintain structure of the polypeptide. In some embodiments, all identified interface residues and key residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:13. In a further embodiment, the polypeptides may comprise an insertion in one or more insertion sites of the polypeptide. The insertion may be any one or more amino acid, and may comprise a functional domain as described herein, or one or more amino acids for additional spacing or for any other purpose. In one embodiment, an insertion in the loop regions is 1-3, 1-2, 1, 2, or 3 amino acids in length. Table 1, column 7 show the position of potential insertion sites relative to SEQ ID NO:13. Table 1. Annotation of SEQ ID NO:13 (S2B1) The disclosure also provides polypeptides comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:14, not including any amino acid insertions at identified insertion sites (i.e., any insertions are not considered when determining percent identity to the reference polypeptide), wherein the polypeptide binds to the F1 domain of insulin receptor. YRVTLHSDDEELLKDLEWAARVLGIEVRHNDKTKTVTFHSDDKSELEFLESLARMNGVEVRT EG (S2B2; SEQ ID NO:14) In this aspect the polypeptides bind to the F1 domain of insulin receptor and can be used, for example, as a component of a fusion protein functioning as insulin receptor agonists, thereby enhancing insulin sensitivity in type 2 diabetes and regulating glucose and lipid metabolism in type 1 diabetes and congenital severe insulin resistance. In some embodiments, the polypeptide comprises an amino acid sequence at least 75% identical to the amino acid sequence of SEQ ID NO:14. In other embodiments, the polypeptide comprises an amino acid sequence at least 90% identical to the amino acid sequence of SEQ ID NO:14. In further embodiments, substitutions relative to SEQ ID NO:14 are selected from substitutions listed in Options 1, 2, or 3 of Table 2. Site saturation mutagenesis studies were conducted SEQ ID NO:14 (S2B2), and Table 2 provides a list of best substitutions (Option 1; increased or maintained activity), and tolerable substitutions (Options 2 or 3: retained activity, though at a lower level than SEQ ID NO:14), that identified mutations that can be incorporated into SEQ ID NO:14. In some embodiments, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or all 16, identified interface are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:14. Interface residues (i.e., at the binding interface between the polypeptide and the F1 domain of insulin receptor) relative to SEQ ID NO:14 are provided in column 5 of Table 2. In other embodiments, at least 1, 2, 3, 4, 5, 6, 7, or all 8 core residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:14. Residues that are present at the polypeptide core are shown in column 6 of Table 2. These residues help maintain structure of the polypeptide. In other embodiments, all identified interface residues and key residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:14. In a further embodiment, the polypeptides may comprise an insertion in one or more insertion sites of the polypeptide. The insertion may be any one or more amino acid, and may comprise a functional domain as described herein, or one or more amino acids for additional spacing or for any other purpose. In one embodiment, an insertion in the loop regions is 1-3, 1-2, 1, 2, or 3 amino acids in length. Table 2, column 7 show the position of potential insertion sites relative to SEQ ID NO:14. Table 2. Annotation of SEQ ID NO:14 (S2B2) The disclosure further provides polypeptides comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:1-12 and 15-28, wherein the polypeptide binds to the F1 domain of insulin receptor. In this aspect the polypeptides (sequence shown in Table 3) bind to the F1 domain of insulin receptor and can be used, for example, as a component of a fusion protein that serves as an insulin receptor agonist, permitting its use, for example, to treat diabetes and severe insulin resistance syndromes. In one embodiment, the polypeptides comprise an amino acid sequence at least 75% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:1-12 and 15-28. In another embodiment, the polypeptides comprise an amino acid sequence at least 90% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:1-12 and 15-28. Table 3. F1-domain Binder Sequences In another aspect, the disclosure provides polypeptides comprising an amino acid sequence at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:29- 31, wherein the polypeptide binds to the L1 domain of insulin receptor. In this aspect the polypeptides bind to the L1 domain of insulin receptor and can be used, for example, as a component of a fusion protein that serves as an insulin receptor agonist, permitting its use, for example, to treat diabetes and severe insulin resistance syndromes. In some embodiments, the polypeptides comprise an amino acid sequence at least 85% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:29-31. In other embodiments, the polypeptides comprise an amino acid sequence at least 90% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:29-31. In further embodiments, amino acid substitutions relative to the reference sequence are conservative amino acid substitutions. The disclosure also provides conjugate, comprising the polypeptide of any embodiment or combination of embodiments herein; and a therapeutic or diagnostic moiety. The polypeptides of the disclosure bind to the insulin receptor (IR), and thus the conjugates of this embodiment can be used, for example to deliver therapeutics or diagnostics via the IR. In non-limiting embodiments, the therapeutic moiety may include insulin (and the conjugate used to treat, for example, type 1 or 2 diabetes and cancer. In these embodiments, the conjugates may be used for insulin replacement in subjects with type 1 or type 2 diabetes. Such methods permit use of reduced amounts of insulin, as the polypeptides of the disclosure have therapeutic potentials to improve insulin sensitivity in type 1 and type 2 diabetes patients. A combination therapy of insulin with the conjugates or fusion proteins of the disclosure lowers insulin requirements in patients with diabetes to reduce side effects of hyperinsulinemia. As the polypeptides and fusion proteins bind and activate insulin-binding deficient and disease-causing IR mutants, they provide the basis for therapeutic interventions for these rare but devastating conditions that currently lead to early morbidity. Given the potential cancer inducing properties of insulin, the conjugates are also beneficial for patients with both diabetes and cancer The therapeutic or diagnostic moiety may be conjugated to the polypeptide via any suitable technique. If the therapeutic or diagnostic moiety is a polypeptide, the conjugate may comprise a fusion protein. Alternatively, the therapeutic or diagnostic moiety may be conjugated via any chemical conjugation process as appropriate in light of the specific therapeutic or diagnostic moiety. In one embodiment, the therapeutic or diagnostic moiety comprises insulin. In another embodiment, the disclosure provides fusion proteins comprising: (a) a first polypeptide of any embodiment or combination of embodiments herein that binds to the F1 domain of insulin receptor; and (b) a second polypeptide of any embodiment or combination of embodiments herein that binds to the binds to the L1 domain of the insulin receptor, or a polypeptide comprising an amino acid sequence at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:75 and which binds to the L1 domain of the insulin receptor; optionally wherein the first polypeptide and the second polypeptide are linked by an amino acid linker. PEVKKEAFKAFMLLMDALFLAEDPNIRKTIEELIERLEKADENDDEEKLKEIIKKAKTIWKQ VL (SEQ ID NO:75) As disclosed in the examples, fusion proteins comprising S2B and embodiments permit binding to the L1 and F1’ domains, and breaks the autoinhibitory conformation of IR, induces conformational rearrangement, and stabilize an active state of IR. In another embodiment, the disclosure provides fusion proteins comprising: (a) a first polypeptide of any embodiment or combination of embodiments herein that binds to the F1 domain of insulin receptor; and (b) a second polypeptide of any embodiment or combination of embodiments herein that binds to the F1 domain of insulin receptor; optionally wherein the first polypeptide and the second polypeptide are linked by an amino acid linker. A fusion of two S2 binders also can activate insulin receptor signaling. When treating C2C12 cells overexpressing hIR with fusion proteins according to claim 32 at 1 μM, activation of receptor, AKT and ERK was observed by western blot. Similar to insulin, S2-S2 fusion increased IR autophosphorylation and AKT phosphorylation in C2C12 cells expressing human IR. As compared to insulin, S2-S2 fusion increased ERK phosphorylation by 80%. In some embodiments of the conjugates and fusion proteins, the amino acid linker is present. In some embodiments, the amino acid linker is between 0-40 amino acids in length. In further embodiments, the fusion protein may comprise an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:47-74. The amino acid sequences of the exemplary fusion proteins are provided in Tables 4-6. In Tables 4 and 6, the bold font residues are amino acid linkers between the first and second polypeptides. In Table 5, the underlined residues are amino acid linkers between the first and second polypeptides. Table 4. Exemplary fusion proteins Table 5. Rigid Linkers Fusion with linker highlighted. S2B component – linker – S1B component Table 6. All Flexible Fusions In another aspect the disclosure provides nucleic acids encoding the polypeptide or fusion protein of any embodiment or combination of embodiments of the disclosure. The nucleic acid sequence may comprise single stranded or double stranded RNA or DNA in genomic or cDNA form, or DNA-RNA hybrids, each of which may include chemically or biochemically modified, non-natural, or derivatized nucleotide bases. Such nucleic acid sequences may comprise additional sequences useful for promoting expression and / or purification of the encoded peptide or chimeric molecular construct, including but not limited to polyA sequences, modified Kozak sequences, and sequences encoding epitope tags, export signals, and secretory signals, nuclear localization signals, and plasma membrane localization signals. It will be apparent to those of skill in the art, based on the teachings herein, what nucleic acid sequences will encode the polypeptide or fusion protein of the disclosure. In a further aspect, the disclosure provides expression vectors comprising the nucleic acid of any aspect of the disclosure operatively linked to a suitable control sequence, such as a promoter. “Expression vector” includes vectors that operatively link a nucleic acid coding region or gene to any control sequences capable of effecting expression of the gene product. “Control sequences” operably linked to the nucleic acid sequences of the disclosure are nucleic acid sequences such as a promoter) 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, and ribosome binding sites. Such expression vectors can be of any type, including but not limited plasmid and viral-based 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 the host organisms either as an episome or by integration into host chromosomal DNA. In various embodiments, the expression vector may comprise a plasmid, viral-based vector, or any other suitable expression vector. In another aspect, the disclosure provides host cells that comprise the polypeptide, fusion protein, nucleic acid or expression vector (i.e.: episomal or chromosomally integrated) disclosed herein, wherein the host cells can be either prokaryotic or eukaryotic. The cells can be transiently or stably engineered to incorporate the expression vector of the disclosure, using techniques including but not limited to bacterial transformations, calcium phosphate co- precipitation, electroporation, or liposome mediated-, DEAE dextran mediated-, polycationic mediated-, or viral mediated transfection. In another aspect, the disclosure provides pharmaceutical compositions, comprising the polypeptide, the recombinant nucleic acid, the expression vector, or the recombinant host cell of any of any embodiment or combination of embodiments, and a pharmaceutically acceptable carrier. The pharmaceutical compositions of the disclosure can be used, for example, in the methods of the disclosure described herein. The pharmaceutical composition may further comprise (a) a lyoprotectant; (b) a surfactant; (c) a bulking agent; (d) a tonicity adjusting agent; (e) a stabilizer; (f) a preservative and / or (g) a buffer. In some embodiments, the buffer in the pharmaceutical composition is a Tris buffer, a histidine buffer, a phosphate buffer, a citrate buffer or an acetate buffer. The pharmaceutical composition may also include a lyoprotectant, e.g. sucrose, sorbitol or trehalose. In certain embodiments, the pharmaceutical composition includes a preservative e.g. benzalkonium chloride, benzethonium, chlorohexidine, phenol, m-cresol, benzyl alcohol, methylparaben, propylparaben, chlorobutanol, o-cresol, p-cresol, chlorocresol, phenylmercuric nitrate, thimerosal, benzoic acid, and various mixtures thereof. In other embodiments, the pharmaceutical composition includes a bulking agent, like glycine. In yet other embodiments, the pharmaceutical composition includes a surfactant e.g., polysorbate-20, polysorbate-40, polysorbate- 60, polysorbate-65, polysorbate-80 polysorbate-85, poloxamer-188, sorbitan monolaurate, sorbitan monopalmitate, sorbitan monostearate, sorbitan monooleate, sorbitan trilaurate, sorbitan tristearate, sorbitan trioleaste, or a combination thereof. The pharmaceutical composition may also include a tonicity adjusting agent, e.g., a compound that renders the formulation substantially isotonic or isoosmotic with human blood. Exemplary tonicity adjusting agents include sucrose, sorbitol, glycine, methionine, mannitol, dextrose, inositol, sodium chloride, arginine and arginine hydrochloride. In other embodiments, the pharmaceutical composition additionally includes a stabilizer, e.g., a molecule which, when combined with a protein of interest substantially prevents or reduces chemical and / or physical instability of the protein of interest in lyophilized or liquid form. Exemplary stabilizers include sucrose, sorbitol, glycine, inositol, sodium chloride, methionine, arginine, and arginine hydrochloride. The polypeptide, conjugate, fusion protein, nucleic acid, expression vector, or cell of any embodiment or combination of embodiments herein may be the sole active agent in the pharmaceutical composition, or the composition may further comprise one or more other active agents suitable for an intended use. In a further aspect, the disclosure provides methods for using, or a use of the polypeptide, the recombinant nucleic acid, the expression vector, the recombinant host cell, and / or the pharmaceutical composition of any embodiment or combination of embodiments of the disclosure, for any suitable purpose including but not limited to those disclosed herein. In various embodiments, the purpose includes, but is not limited to, treating or limiting development of type 1 or type 2 diabetes, cancer, autoimmune disease, or thyroid eye disease. In other embodiments, the purpose includes, but is not limited to, enhancing insulin sensitivity in type 2 diabetes and regulating glucose and / or lipid metabolism in type 1 diabetes and congenital severe insulin resistance. As used herein, "treat" or "treating" means accomplishing one or more of the following: (a) reducing the severity of the disorder; (b) limiting or preventing development of symptoms characteristic of the disorder(s) being treated; (c) inhibiting worsening of symptoms characteristic of the disorder(s) being treated; (d) limiting or preventing recurrence of the disorder(s) in patients that have previously had the disorder(s); and (e) limiting or preventing recurrence of symptoms in patients that were previously symptomatic for the disorder(s). Examples Binding of insulin to the insulin receptor (IR) induces conformational changes in the extracellular portion of the receptor that lead to activation of the intracellular kinase domain and the AKT and MAPK pathways, and downstream modulation of glucose metabolism and cell proliferation. We reasoned that designed agonists that induce different conformational changes in the receptor might induce different downstream responses, which could be useful both therapeutically and to shed light on how extracellular conformation is coupled to intracellular signaling. We used de novo protein design to first generate binders to individual IR extracellular domains, and then to fuse these together in different orientations and with different conformational flexibility. We describe a series of synthetic agonists that signal through the IR that differ from insulin and from each other in the induction of receptor autophosphorylation, MAPK activation, intracellular trafficking, and cell proliferation. We identify designs that are more potent than insulin causing much longer lasting reductions in glucose levels, and that retain signaling activity on disease-causing receptor mutants that do not respond to insulin. These results inform our understanding of how changes in receptor conformation and dynamics are transmitted to downstream signaling, and our synthetic agonists have considerable therapeutic potential for diabetes and severe insulin resistance syndromes. There are only a small number of natural ligands for any given receptor, and the conformational dynamics they produce in the target receptor are not easily modulated; in contrast protein design can in principle generate a wide range of ligands that bind to multiple domains in receptor extracellular regions and tune their relative orientations and dynamics. To probe the relationship between extracellular conformation of the IR and intracellular signaling, we set out to design proteins that modulate the conformational dynamics of the extracellular domains of IR and to determine the effect of these on the extent of induced autophosphorylation and downstream signaling (Fig.1C). Such designed IR agonists also have therapeutic potential: although recombinant insulin and its analogs have been used to treat type 1 and type 2 diabetes for nearly a century, these treatments are not without limitations, including complications in manufacturing processes and storage, which could potentially be reduced for hyperstable, easy-to-manufacture designed proteins16. Results We hypothesized that synthetic molecules that engage two insulin binding sites in the extracellular domains of IR and induce conformational changes to reduce the distance between intracellular kinase domains could activate IR signaling. To generate new IR agonists that bring together the different domains in the extracellular region in different orientations, we used a two-step approach: we first sought to design binders for the L1 domain and F1 domain of IR (Fig.1B) and second to fuse the individual binders together to induce conformational rearrangements in the receptor (Fig.1C). Such synthetic binders could break the autoinhibitory conformation of IR by displacing the -CT motif from the L1 domain of IR, and stabilize the active conformation by simultaneously binding to two protomers. We reasoned that by varying the rigidity of the linker between the two binding domains, we could explore the effects of the conformational stability of the active state of IR on downstream signaling, intracellular trafficking, and its functions (Fig.1C). Previously, a miniprotein binder was developed to bind the L1 domain of IR using a RosettaTMbased design pipeline17(we refer to this below as S1B). Therefore, we started by designing IR site- 2 binders. De novo design of IR site-2 binders We set out to design site-2 binders that bind to the outer side of the F1 domain in the inactive state of IR (Fig.1B,D). We used the RosettaTMRIF dock method17followed by RosettaTMFast Design to design binders to the F1 domain structure from a high-resolution cryo-EM model (PDB:6PXV)8,17, and filtered the designs based on RosettaTMmetrics (‘ddg','sasa','contact_molecular_surface',’contact patch’)17and DeepAccNetTM(‘Plddt’)18. We constructed a library of 11,452 designs that passed the filters and screened them using yeast display (Fig.1D). Seventeen designs bound the F1 domain on the yeast surface at 1 nM (Fig. 3A,B). After optimization through site-saturation mutagenesis (SSM) and combination mutagenesis, we selected several designs and determined their binding affinity for the F1 domain of IR to be sub-nanomolar to nanomolar using biolayer interferometry (BLI). These binders are either helical bundles or ferredoxin scaffolds (Fig.2A, Fig.3B) and are modeled to bind to insulin binding sites on the F1 domain. One binder, referred to as S2B, which binds to the F1 domain of IR with a KD of 1.9 nM, significantly higher than insulin (KD of 21 M, Fig.3D), was selected for further study (Fig.2C). S2B is stable at 95 °C (Fig.2D), whereas insulin irreversibly unfolds at high temperatures (Fig.3C). We investigated the effect of the S2B design on IR signaling. There are two IR isoforms, a short isoform, IR-A and a long isoform, IR-B19. We treated DKO-IR-B cells (IR and IGF1R double knockout preadipocytes expressing human IR-B) with insulin wild-type (WT), insulin ValA3E (a site-1 binding defective mutant), and insulin LeuA13R (a site-2 binding defective mutant) in the presence and absence of S2B. Consistent with previous work, insulin ValA3E did not activate IR, while insulin LeuA13R partly activated IR and downstream signaling in DKO-IR-B cells (Fig.2E,F). Cotreatment with insulin site-1 and site-2 binding defective mutants can activate IR10. Similarly, when the designed S2B was co- treated with the insulin site-2 binding defective mutant, we observed increased IR autophosphorylation (pY1150,1151 IR) and downstream signaling compared to insulin site-2 binding defective mutant or S2B alone (Fig.2E,F). Similar results were obtained in cells expressing IR-A, a short isoform of IR (Fig.3E,F). Thus, while inactive on its own, S2B can synergize with site-1 only binding insulin to activate both isoforms of IR. To determine the mechanism of receptor activation by S2B, we determined a 6 Å cryo-EM structure of IR bound with S2B together with insulin (Fig.2G, 4). In the structure, two insulin and two designed S2B molecules bind to IR site-1 and site-2, respectively, and promote the symmetric T-shaped IR conformation. Thus, the functionally mimicry of site-2 binding insulin by S2B arises from similar effects on IR conformation. Design of Site-1 and Site-2 binder fusions We next sought to link S2B with S1B (Fig.5A,B) such that binding to the L1 and F1’ domains would break the autoinhibitory conformation, induce conformational rearrangement, and stabilize an active state of IR (Fig.1C). We overlaid the design models of S2B / F1 and S1B / L1 on the active conformation of IR (PDB: 8DTL) by superimposing the F1 and L1 domains20. We then used RFdiffusionTMto build a rigid connector between the two binders21(Fig.5B,C). The first helix of L1B (Up to residue 3) and the last helix of S2B (Residue 54 onwards) were rebuilt by RFdiffusionTMto link the L1- and S2-binding interfaces. Surface residues near the diffused regions were masked and redesigned using ProteinMPNNTMalong with the newly built regions.32 sequences were generated for each backbone. Designs passing AF222and RosettaTMmetrics17(pae interaction, plddt, ddg, contact surface) were selected for experimental characterization. To determine how conformational stability affects IR signaling, we also generated more flexible fusions by linking the two domains with GS linkers in both orientations; in the overlaid composite structure the distance between the S1B- C and the S2B-N terminus (S1-Fn-S2) is ~18 Å and between the S2B-C terminus and S1B N- terminus (S2-Fn-S1) is ~9 Å; we therefore used slightly longer linkers (Fn; n, linker lengths = 3 to 11 residues) in the S1-Fn-S2 than the S2-Fn-S1 constructs (Fn; n = 1 to 8 residues) (Fig. 11A). The binding affinities of the two-domain binding constructs were estimated by BLI against the biotinylated IR extracellular domain (IR-ECD). S2-F1-S1 bound to IR-ECD with a KDof 0.37 nM, while RF-405 and RF-409 bound to IR-ECD with a KDof 1.3 nM and 8.1 nM, respectively (Fig.5D). The fusion constructs are quite specific as they did not bind IGF1R, a homologous receptor tyrosine kinase (RTK) that can be activated by insulin23,2425,26(Fig.3G). Circular dichroism (CD) temperature melting studies showed that both the flexibly-linked and rigidly-linked binders are hyper-thermostable (Fig.5E). Cryo-EM complex structure determination To determine whether the synthetic fusion constructs induce the intended conformational changes, we determined cryo-EM structures of the RF-405 / IR and S2-F1- S1 / IR complexes at resolutions 4 Å and 8 Å, respectively (Fig.4A-D, S3-5). The cryo-EM structure of RF-405 / IR complex exhibits an extended T-shaped architecture. A strong density between the two protomers at the top part of the IR was observed and unequivocally assigned to RF-405. The design models of the S1B and S2B of the RF-405 fit well into the cryo-EM density as rigid bodies without further refinement; consistent with the models, the S1B binds the L1 domain of IR (site-1), while the S2B contacts a side surface of the F1 domain of the adjacent IR protomer (site-2) (Fig.6E-G, Fig.7I). S1B and S2B of RF-405 are linked through a continuous -helix as predicted in the design model (Fig.6A,B). Both site-1 and site-2 interfaces involve both hydrophobic and electrostatic interactions, and Trp65 of RF-405 is sandwiched between the S1B and S2B components, enhancing the rigidity of the design (Fig. 6G). The -CT motif is displaced from the L1 domain upon the binding of the S1B of RF- 405. RF-405 crosslink the two IR protomers by simultaneously contacting the site-1 and site- 2, thereby stabilizing the extended T-shaped active conformation (Fig.6A,B). This conformation is similar to that induced by S59720, suggesting that the designed binders and S597 induce similar conformational changes for receptor activation. The cryo-EM structure of the S2-F1-S1 in complex with IR was resolved at a lower resolution than the RF-405 / IR complex: the flexibility of S2-F1-S1 likely translates to increased flexibility of the complex relative to the rigid RF-405 (Fig.6C,D, Fig.7G). The structure of the S2-F1-S1 / IR complex has a similar extended T-shape to that of RF-405 / IR complex (Fig.6C,D). S1B and S2B of S2-F1-S1 are linked by a short flexible loop (Fig. 6D,H); superimposition of the bound S2-F1-S1 and RF-405 revealed that, due to the rigid linkage, the distance between the S1B and S2B domains of RF-405 is shorter than those of S2-F1-S1 (Fig.6H). The compact conformation of RF-405 allows its S1B component to simultaneously contact the L1 domain of one protomer and a loop in the top region of the F1 domain of the other protomer, further increasing the stability of the active conformation of IR. We next attempted to determine the cryo-EM structures of the S1-F8-S2 / IR complexes (Fig.7J). Unlike S2B-S1B fusion constructs, 2D class averaging of S1-F8-S2 / IR revealed a high degree of conformational heterogeneity. Even though individual domains could be identified in a subset of 2D class averages, no high-resolution features were apparent, such as clearly identifiable secondary structural elements. These data suggest that the S1-F8-S2 / IR complex samples a much greater range of conformations than the S2-F1-S1 and RF-405 complexes. Characterization of synthetic agonists We next tested the effects of the fusion constructs on IR activation and downstream signaling. S1-Fn-S2 increased levels of IR autophosphorylation (pY1150,1151) to about 20% of those in insulin-treated cells but did not increase pAKT or pERK levels (Fig.10B). In the presence of insulin, S1-F8-S2 functioned as an antagonist, inhibiting insulin-dependent IR activation (Fig.11A-E) and cell proliferation (Fig.11F). These results are consistent with our structural observations (Fig.7J) and indicate that S1-F8-S2 may disrupt the autoinhibitory conformation of IR, but does not stabilize IR in an active conformation. In contrast to S1B-S2B fusion, we found that the flexibly-linked S2-Fn-S1 constructs functioned as partial (biased) agonists while the rigidly-linked RF-405 and RF-409 were full insulin mimicking, balanced agonists. The rigidly-linked constructs RF-405 and RF-409 elicited pY1150,1151 IR and pAKT and pERK levels similar to insulin, while S2-Fn-S1 constructs increased IR autophosphorylation (pY1150,1151 IR) and pAKT levels, but pERK levels were only 40% of those in insulin-treated cells (Fig.10B; increasing the linker length reduces IR autophosphorylation). The increased dynamics of S2-Fn-S1 compared to the rigid RF-401 and RF-409 appears to compromise balanced signaling. We analyzed levels of pY1150,1151 IR, pAKT, and pERK over a wide range of ligand concentrations in C2C12-IR cells (Fig.10C-E, 11G). RF-405 and RF-409 potently activated both pAKT and pERK levels similar to insulin, while the flexibly-linked S2-F1-S1 and S2-F5-S1 primarily activated pAKT (Fig.10C-E, 11G), again indicating partial agonism. S2-F5-F1, which has a longer flexible linker, was significantly less effective at increasing pERK levels compared to S2-F1-S1 (Fig.10E, 11G). A similar signaling pattern was observed with the single chain peptide S59720,27,28. As S2-Fn-S1 is a partial agonist of IR signaling while RF-405 and RF-409 appear to be full balanced agonists, suggesting that altered dynamics (rather than structural changes per se) might lead to different IR autophosphorylation patterns. When insulin activates the IR, several tyrosine residues in the juxtamembrane (e.g. Y960), kinase (e.g. Y1146, Y1150, Y1151), and C-terminal domains (e.g. Y1316, Y1322) undergo trans-autophosphorylation (Fig.10A). These phosphorylation sites are crucial for recruiting downstream substrates1,29. To determine if the designed agonists differ in their efficiency at activating phosphorylation in these three domains, we measured the IR phosphorylation levels at these three intracellular sites. RF-405 increased autophosphorylation sites at all sites more significantly than insulin (Fig.10F-I, 11G). While S2-F1-S1 increased pY1150,1151 levels more than insulin (Fig. 10C, 11G), phosphorylation at Tyr960 and Tyr1146 was lower (Fig.10F,G, 11G), and in the C-terminal domain pY1316 and pY1322 phosphorylation was only 50% of insulin (Fig. 10H,I, 11G), indicating partial agonism. Thus conformational dynamics modulate the extent of IR phosphorylation during IR activation, which in turn modulates IR signaling. The MAPK pathway regulates cell growth and proliferation30. We next investigated the differences in the extent of MAPK pathway (pERK levels) activation of our synthetic agonists on cell proliferation. We compared the ability of the designed agonists and insulin to induce proliferation of C2C12-IR cells (Fig.10J). RF-405 and RF-409 induced cell proliferation at levels comparable to insulin. At 1 nM, cells treated with RF-405 and RF-409 showed a two-fold increase in proliferation compared to insulin. This enhanced proliferation may reflect the higher potency of RF-405 and RF-409 in inducing IR autophosphorylation. S2-F1-S1, which stimulated lower pERK levels, still induced cell proliferation, but S2-F5-S1 which is more defective in the MAPK pathway activation did not. Insulin-activated IR undergoes endocytosis which can terminate and redistribute IR signaling3132. The MAPK pathway plays a crucial role in regulating IR endocytosis33(Fig. 10A). To determine the effects of our synthetic agonists on IR endocytosis, we incubated S2- F1-S1, RF-409, or insulin with primary hepatocytes and analyzed cell surface IR and total IR levels. RF-409 and insulin, but not S2-F1-S1, significantly reduced cell surface IR levels (Fig.10K; none of the three molecules reduced total IR levels, Fig.10L). These data suggest that RF-409, like insulin, induces IR endocytosis, but S2-F1-S1 does not, supporting a role of the MAPK activation in promoting IR endocytosis. Taken together, these data suggest that by modulating the relative orientation of the L1 and F1 domains of the IR during activation, different signaling and trafficking outcomes can be obtained. An equimolar mixture of unlinked S1B and S2B, did not induce IR phosphorylation or pERK or pAKT activation (Fig.3H,I), indicating that engaging the two domains independently has no effect on IR activation. Combining them with rigid constructs to match and stabilize the active state of the IR results in insulin-like signaling and induction of endocytosis, whereas allowing a wide range of conformations with flexible linkers results in partial agonism (reduced pERK signaling) and less endocytosis. Designed IR agonists activate IR mutants that are resistant to insulins. Our fusion constructs and insulin make different sets of contacts with the IR: site-1 insulin interacts with the L1 domain and -CT motif, while the S1B binds solely to the L1 domain (Fig.12A). To test the importance of each binding site in activating IR, we first introduced mutations on the L1 domain of IR (F64A and F96A) that remove most of the interactions between L1 and insulin but only a subset of the interactions with the designed agonists (Fig.12A; the designs have a much larger interaction surface area with L1 than insulin– 1481.8 vs 793.1 Å2buried surface area respectively). As expected, insulin could not activate IR F64A and F96A (Fig.12C,D), but the mutants could still be activated by our designed agonists. To evaluate the role of site-2 interface in activating IR, we introduced K484E and L552A mutations (Fig.12B-D). Insulin and S2-F1-S1 could not activate the IR K484E / L552A mutants, supporting the importance of site2 interface on IR activation. Insulin induces a large conformational change and generates intra- and inter-domain contacts that stabilize the active IR state8. Arg345 in the L2 domain and Glu697 in the -CT forms a salt bridge in the insulin-induced compact T-shaped IR and the R345A mutant is insulin resistant. We found that unlike insulin, S2-F1-S1 and RF-405, were able to fully activate the IR R345A mutants (Fig.12C,D). Therefore, the activation of IR by designed agonists is less dependent on this Arg-Glu salt bridge formation and the binding energy of the designed agonists is sufficient to overcome this loss of internal stabilization of the active state. Mutations in the insulin binding sites of IR cause rare but severe insulin resistance syndromes such as Donohue syndrome and Rabson-Mendenhall syndrome34–36. Patients produce insulin normally, but are not able to properly regulate glucose metabolism. We introduced disease-causing mutations in the L1 (R14W37and N15K38), F1 (D496N39, D496K) and -CT (D707A)40domains of IR at sites that do not contribute to the interaction with the designed IR agonists (Fig.12E,F). As expected, insulin could not activate the R14W, N15K, D496K, and D707A IR mutants. In contrast, both S2-F1-S1 and RF-405 could fully activate these disease-causing mutants (Fig 12E,F). These results highlight the differences in activation mechanism of our designed agonists, and such designed agonists could be beneficial for patients with insulin-binding deficient IR mutants. Designed agonists mimic insulin functions in vivo We set out to investigate the function of our designed agonists in vivo. The IR structure and sequence are highly conserved between humans and mice. We first tested if the designed agonists can activate IR in mouse cells (Fig.14A,B). We isolated primary mouse hepatocytes and compared insulin- and designed agonist-induced IR signaling. In primary hepatocytes, RF-409 increased IR autophosphorylation (pY1150,1151 IR) and pAKT similarly to insulin, while S2-F1-S1 induced pY1150,1151 IR and pAKT less potently than RF-409. Due to the high basal levels of pERK in primary hepatocytes, despite insulin stimulation, we could not observe significant increase in pERK. We next compared IR signaling induced by S2-F1-S1 and RF-409 in metabolic tissues including liver and skeletal muscle (Fig.14C,D). In both tissues, S2-F1-S1 was less effective in stimulating IR autophosphorylation, but it could increase levels of pAKT to similar levels as RF-409. As in primary hepatocytes, mouse liver exhibited high basal levels of pERK, and we did not observe significant difference between all tested molecules. In skeletal muscle, RF-409 significantly increased pERK levels, while S2-F1-S1 did not, confirming partial agonism in vivo. To determine the metabolic effects of our designed agonists, we conducted an insulin tolerance test (ITT) in mice fed with a normal diet (Fig.13A,B; Fig.14E,F). S2-F1-S1 reduced glucose levels in mice as effectively as insulin, while RF-409 was even more effective than insulin. The designed agonists had longer-lasting effects on glucose levels compared to insulin: with insulin, glucose rapidly dropped (within 30 minutes) and began to rise again shortly thereafter, while with half the dose of RF-409, glucose levels decreased to the same level and then remained low throughout the experiment. To determine the physiological effects of designed agonists under diabetic conditions, we conducted ITT in diet-induced obese mice, which have higher basal blood glucose levels. Similar to healthy mice, RF-409 and S2-F1-S1 slowly reduced glucose levels but exhibited prolonged glucose lowering effects compared to insulin. RF-409 was more effective at lowering glucose levels than S2-F1-S1 (Fig.13C; Fig.14G). Strikingly, after a single injection of RF-409, mice maintained low glucose levels for 6 h, whereas mice treated with insulin rose glucose levels within 2 h (Fig.13D; Fig.14H). Overall, our designed agonists effectively lower glucose levels in mice, with RF-409 proving to be more potent than S2-F1- S1 (Fig.13B,C; Fig.14F). Our series of designed agonists provide insight into the mechanism of signaling through the IR, and how receptor autophosphorylation is tied to downstream outcomes. RF- 405, S2-F1-S1 and S2-F5-S1 all bind to the IR with high affinity (Fig.5, 13E) but generate different signal transduction processes. The rigid RF-405 agonist induces a highly ordered active conformation of IR (Fig.6) which leads to efficient IR autophosphorylation in the juxtamembrane (pY960), tyrosine kinase (pY1150,py1151) and C-terminal (pY1316 and pY1322) domains. This results in strong activation of the PI3K-AKT (pAKT), and MAPK (pERK) pathways (Fig.10C-I, 13E, Fig.11G). The intermediate flexibility agonist S2-F1-S1 induces a more dynamic conformation of the receptor that is less able to autophosphorylate, particularly the C-terminal domain (Y1316 and Y1322), which reduces activation of the MAPK pathway, endocytosis. The C-terminal tyrosine phosphorylation sites have been identified as docking sites for SHP241. SHP2 is an upstream regulator of MAPK pathway and controls IR signaling by dephosphorylating IR and IR substrates and by promoting receptor endocytosis31,33,42–45. This may explain why S2-F1-S1 is less effective in activating the MAPK pathway and IR endocytosis. The high flexibility S2-F5-S1 showed less potency, particularly in MAPK pathway activation (Fig 10B, 10E, 13E). In addition, designs such as S1-F8-S2 does not favor an active conformation and hence does not signal, but does disrupt the autoinhibitory state of IR and is functionally antagonist, competing out insulin. Discussion Understanding how physiological signals are transmitted to downstream pathways, particularly those involving conformational dynamics, has been challenging. Here, we demonstrate that computationally designed agonists with controlled extents of internal flexibility induce states of the target receptor with matching dynamics. We describe rigid and flexible agonists that bring together the receptor site-1 and site-2 containing domains and drive the receptor into a T-shaped active state, but with different dynamics as indicated by the cryo-EM structure resolution. Comparison of the effects of the rigid and flexible agonist shows that the extent of ordering directly impacts autophosphorylation of IR during activation, MAPK versus AKT signaling, and trafficking, indicating that the conformational stability of IR during activation is a key factor in controlling its biological activity (Fig.13F). The designed agonists are also highly stable and easy to produce recombinantly, making them ideal candidates for large-scale production and storage, and to our knowledge are the most potent completely synthetic IR agonists available described to date. The designed molecules can be used for treating IR-related diseases. Given the potential cancer inducing properties of insulin46–49, the designs disclosed herein may be beneficial for patients with both diabetes and cancer. Second, our designed agonists exhibit high binding affinity and selectivity for IR, and may be used where off-target activation of IGF1R is problematic such as cancer, autoimmune disease, and thyroid eye disease50–52. Third, as our designed agonists bind and activate insulin-binding deficient and disease- causing IR mutants, they may be used for therapeutic interventions for these rare but devastating conditions that currently lead to early morbidity. Table 7. Cryo-EM data collection and structure refinement statistics. STAR Methods Computational design of the IR site 2 binders The computational design method was using the method previously described17. In brief, IR structure 6PXV was downloaded from Protein Data Bank and relaxed by RosettaTMguided by experimental design-guided relaxation. The F1 domain (residue 467-590) was extracted as the targeting domain and the insulin binding side was selected as the targeting interface. For each residue of the selected interface, Rotamer Interaction Field (RIF) was generated. Later, mini-protein scaffold set, composed of 3 helical, 4 helical and ferredoxin scaffolds, were used to search for global shape complementarity with PatchDockTM. The docked scaffolds were then sequenced sequence-optimized using RosettaTMFastDesign and evaluated by DeepAccNetTM 18pLDDT and RosettaTMMetrics including ddG, contact patch, and contact molecular surface17. A total of 11280 oligos encoding designed site 2 binders passed the filters. Computational design of the linked binders Flexibly-linked IR binders were generated by linking S1B and S2B with Gly-Ser (GS) linkers with various lengths. Non Interface residues 1, 13, 22, 25, 27, 29, 30, 35, 36, 39, 43, 44, 45, 48, 51, 52, 53, 54, 55, 58, 59, 60, 62, 64 were redesigned by MPNNTMto increase the solubility of S2-F1-S1. To generate Rigidly-fused IR binders, we aligned the design models of S1B / L1 and S2B / F1 to the S2-F1-S1 / IR complex as the starting point. The first helix of S1B (Up to residue 3) and the last helix of S2B (Residue 54 onwards) were rebuilt by RFdiffusionTMto link the L1- and F1-binding interfaces. Surface residues near the diffused regions were masked and redesigned using ProteinMPNNTMalong with the newly built regions.32 sequences were generated for each backbone. Then, the designs were predicted by AlphaFold2TM(AF2), relaxed and scored by RosettaTM. The top 28 designs with the highest AF2 pLDDT and lowest RMSD to design were selected for experimental characterization. Sequence optimization was further performed to improve solubility of the Rigidly-linked IR binders. Yeast surface display screening for IR site 2 binders with FACS The yeast surface display screening was performed using the protocol as previously described14,16. Briefly, DNAs encoding the minbinder sequences were transformed into EBY- 100 yeast strains. The yeast cells were grown in CTUG medium and induced in SGCAA medium. After washing with FACS-buffer (PBS, Fisher Scientific, supplemented with 1% w / v bovine serum albumin, SigmaAldrich), the cells were incubated with 1uM biotinylated IR F1 domain together with streptavidin–phycoerythrin (SAPE, ThermoFisher, 1:100) and anti-c-Myc fluorescein isothiocyanate (FITC, Miltenyi Biotech, 6.8:100) for 60 min. After washing twice with FACS buffer, the yeast cells were then resuspended in the buffer and screened via FACS. Only cells with PE and FITC double-positive signals were sorted for next-round screening. After another round of enrichment, the cells were titrated with biotinylated IR F1 domain at 100 nM, 10 nM and 1 nM for 60 min, washed, and further stained with both streptavidin–phycoerythrin (SAPE, ThermoFisher) and anti-c-Myc fluorescein isothiocyanate (FITC, Miltenyi Biotech) at 1:100 ratio for 30 min. After washing twice with FACS buffer, the yeast cells at different concentrations were sorted individually via FACS and regrown for 2 days. Next, the cells from each subpool were lysed and their sequences were determined by next-generation sequencing. Protein binder expression and purification Synthetic genes encoding designed proteins were purchased from Genscript or Integrated DNA Technologies (IDT) in the pET29b expression vector or as eBlocksTM(IDT) and cloned into customized expression vectors53using Golden Gate cloning. A His6x tag was included either at the N-terminus or the C-terminus as part of the expression vector. Proteins were expressed using autoinducing TBII media (Mpbio) supplemented with 50x5052 and 20 mM MgSO4in BL21 DE3 E.coli cells. Proteins were expressed under antibiotic selection at 25 °C overnight after initial growth for 6-8 h at 37 °C. Cells were harvested by centrifugation at 4000x g and resuspended in lysis buffer (20 mM Tris, 300 mM NaCl, 5 mM imidazole, pH 8.0) containing protease inhibitors (Thermo Scientific) and Bovine pancreas DNaseI (Sigma- Aldrich) before lysis by sonication. Proteins were purified by Immobilized Metal Affinity Chromatography (IMAC). Cleared lysates were incubated with 0.1-0.5 mL nickel NTA beads (Qiagen) for 20-40 minutes before washing beads with 5-10 column volumes of lysis buffer, 5-10 column volumes of wash buffer (20 mM Tris, 300 mM NaCl, 30 mM imidazole, pH 8.0). Proteins were eluted with 1-4 mL of elution buffer (20 mM Tris, 300 mM NaCl, 300 mM imidazole, pH 8.0). All protein preparations were as a final step polished using size exclusion chromatography (SEC) on Superdex 75 Increase 10 / 300GL columns (Cytiva) using PBS buffer (Fisher Scientific). SDS-PAGE and LC / MS were used to verify peak fractions. Proteins were concentrated to concentrations between 0.5-10 mg / mL and stored at room temperature or flash frozen in liquid nitrogen for storage at -80. Thawing of flash-frozen aliquots was done at room temperature. All purification steps from IMAC were performed at ambient room temperature. Biolayer interferometry (BLI) The BLI experiments were performed on an OctetRED96TMBLI system (ForteBio) at room temperature in HBS-EP buffer (Cytiva Life Sciences) supplemented with 0.2 % w / v bovine serum albumin (BSA, SigmaAldrich). Prior to measurements, streptavidin-coated biosensors were first equilibrated for at least 10 min in the assay buffer. Biotinylated target proteins (IR F1 domain, IR extracellular domain (ECD) (ACROBIOsystems INR-H82E6), IGF1R ECD (ACROBIOsystems IGR-H82E3)) were immobilized onto the biosensors by dipping them into a solution with 100 nM protein until the loading signal reaches 0.5 - 1 nm. Association and dissociation of the analytes were monitored by dipping biosensors in the solutions containing analytes at various concentrations for 300s followed by dipping the biosensors in fresh buffer for 300s. Experiments were performed at 25 °C while rotating at 1000 rpm. Global kinetic or steady-state fits were performed on buffer-subtracted data using the manufacturer’s software (Data Analysis 12.1) assuming a 1:1 binding model. Circular Dichroism Spectroscopy CD spectra were recorded in a 1 mm path length cuvette at a protein concentration between 0.3-0.5 mg / mL on a J-1500 instrument (Jasco). For temperature melts, data were recorded at 222 nm between 25 and 95 °C every 2 °C, and wavelength scans between 190 and 260 nm at 10 °C intervals starting from 25 °C. Experiments were performed in 10 mM sodium phosphate buffer (pH 7.4), 50 mM NaCl. The high tension (HT) voltage was monitored according to the manufacturer's recommendation to ensure optimal signal-to-noise ratio for the wavelengths of interest. Protein expression and purification for cryo-EM For structural studies, the short isoform of human insulin receptor (hIR) or mouse insulin receptor (mIR, sharing 94% sequence homology with hIR) were cloned into pEZT- BM expression vectors as described previously8–10. To improve expression and protein behavior, seven mutations (Y960F, S962A, D1120N, R1333A, I1334A, L1335A, L1337A: amino acid numbering of short isoform of hIR without signal peptide) were introduced to hIR, and two mutations (Y962F and D1122N, amino acid numbering of short isoform of mIRwithout signal peptide) were introduced to mIR. The Human Rhinovirus 3 C recognition site(3C), affinity purification tag Tsi3 (T6SS secreted immunity protein three from Pseudomonas aeruginosa) and His8tag were fused to the C-terminus of both proteins. The expression and purification of hIR and mIR were performed following previously described protocols with minor modifications8–10. Briefly, the plasmids were transformed to Escherichia coli strain DH10Bac to produce bacmid DNA. Recombinant baculovirus was generated by transfecting Sf9 cells with bacmid DNA using Cellfectin reagent (Gibco). hIR or mIR proteins were expressed in FreeStyleTM293-F cells by infecting the cells with thevirus at 1:10 (virus: cell, v / v) ratio. Six hours after infection, 8 mM sodium butyrate wasadded to boost protein expression. Cells were cultured in a shaking incubator supplementedwith 8% CO2 for 48-60 h at 30 °C before harvesting.The cells were resuspended in lysis buffer containing 40 mM Tris-HCl pH 7.5,400 mM NaCl (Buffer A) with Protease Inhibitor Cocktail (Roche) and lysed by using aFrench Press cell disruptor. The membrane fraction was obtained by ultracentrifugation of thecell lysate for 1 h at 100,000 g at 4 °C. To extract the protein from the membrane fraction,Dodecyl maltoside (DDM, Anatrace) was added to a final concentration of 1% (m / v) with stirring overnight. The supernatant containing the solubilized protein was obtained byultracentrifugation for 1 h at 100,000 g at 4 °C. The supernatant was added with 2 mM CaCl2and Tse3 protein-conjugated Sepharose resin (GE Healthcare) and incubated at 4 °C for 1 hbefore being loaded onto a column by gravity flow. The resin was subsequently washed with40 column volumes (CV) of buffer containing 40 mM Tris-HCl pH 7.5, 400 mM NaCl, 2 mMCaCl2, 5% glycerol (v / v), 0.05% DDM (m / v) (Buffer B) and eluted by HRV-3C proteasecleavage at 4 °C overnight. The protein was then concentrated using a 100 kDa cutoffconcentrator (Millipore), loaded onto a SuperoseTM6 increase 10 / 300 GL size-exclusioncolumn (Cytiva), and eluted with buffer containing 20 mM HEPES pH 7.4, 150 mM NaCland 0.03% DDM (Buffer C). The dimer fractions of mIR or hIR proteins were identified by SDS-PAGE and pooled. To make Insulin / S2B / mIR complex for cryo-EM analyses, commercial insulin (I2643, recombinantly expressed in yeast) and site-2 binder (S2B) were added to mIR at a molar ratio of 4:4:1 (insulin: S2B: mIR). To make RF-405 / hIR, S2-F1-S1 / hIR, and S1-F8-S2 / hIR complexes, RF-405, S2-F1-S1, S1-F8-S2 binders were added to hIR at a molar ratio of 4:1(binder: hIR). After incubation for 30 min, the protein mixtures were concentrated to 6-8 mg ml-1 using 100 kDa cutoff concentrators (Millipore) and subject to cryo-EM gridpreparation immediately. All purification and following steps were performed at 4 °C or onice. For analytical size-exclusion chromatography of S2-F1-S1 / hIR and RF-405 / hIR complexes, S2-F1-S1 or RF-405 were added to hIR at a molar ratio of 4:1 (binder: hIR).After incubation for 30 min, the protein mixtures were then loaded onto a SuperoseTM 6increase 3.2 / 300 analytical size-exclusion column (Cytiva) and eluted with Buffer C. Cryo-EM data collection and image processing EM data acquisition, image processing, and model building, and refinement were performed following previous protocols with some modifications8–10,20. The samples of IR in complex with insulin / S2B, RF-405, or S2-F1-S1 were applied to glow-discharged QuantifoilTMR1.2 / 1.3300-mesh gold holey carbon grids (QuantifoilTM, Micro Tools GmbH, Germany). Grids were blotted under 100% humidity at 4 °C and plung-frozen in liquid ethane using a Mark IV VitrobotTM(Thermo Fisher Scientific). Micrographs were collected in the counting mode on either GlaciosTMor Titan KriosTMmicroscopes (Thermo Fisher Scientific) with either Falcon4TM(Thermo Fisher Scientific) or K3 SummitTMdirect electron detectors (Gatan). The nominal magnification and pixel size of each data set are summarized in Table 7. Motion-correction and dose-weighting of the micrographs were carried out using the Motioncor2TMprogram (version 1.2)54. GCTF 1.06 was used for CTF correction55. Template- based particle picking was carried out using the autopick tool in RELIONTM4.055,56. Particles were cleaned up with multiple rounds of 2D and 3D classification in RELIONTM. Good particles were selected and subjected to 3D refinement with C2 symmetry. The exact procedures are summarized in supplementary figures. The initial mode for 3D classification and refinement was generated using the SGD method in RELIONTM. The refined maps were further improved by using Bayesian polishing and CTF refinement at the final stage. The Fourier Shell Correlation (FSC) 0.143 criterion was used for estimating the resolution of the maps. Local resolution was calculated in RELIONTM. Model building and refinement To build the atomic models of the IR structures with different binders bound, the published model of each domain of human IR (PDB ID: 6PXV) and the predicted models of S2B, RF-405 or S2-F1-S1 using AlphaFold2TM 57were docked into the cryo-EM maps as rigid-body. The models were adjusted manually in Coot 0.9858. The models were refined using the real-space refinement module in PhenixTM1.1859. Model quality was checked using MolprobityTMas a part of the PhenixTMvalidation tool set60. Model statistics are summarized in Table S1. Structural figures were rendered in ChimeraXTM1.761,62. Mouse strains and Husbandry Animal work described in this manuscript has been approved and conducted under the oversight of the Columbia University Institutional Animal Care and Use Committee. Mice (C57BL / 6J, Jackson Laboratory, #000664) were fed a standard rodent chow (Lab diet, #5053) or high-fat diet (HFD) (D12492; Research Diets). All animals were maintained in a specific antigen-free barrier facility (temperature, 20-26 °C; humidity, 30-70%) with 12 h light / dark cycles (6 a.m. on and 6 p.m. off). Two to three-month-old male mice were used in this study. Cell cultures, Transfection, and Viral Infection IR and IGF1R double knockout brown preadipocytes expressing only human IR-B (DKO-IR-B) or mouse IR-A (DKO-IR-A) were kindly provided by Dr. Ronald Kahn63. DKO-IR-B, DKO-IR-A, 293FT (Invitrogen, #R70007), and C2C12 (ATCC, CRL-1722) were cultured in high-glucose DMEM supplemented with 10% (v / v) FBS, 2 mM L-glutamine, and 1% penicillin / streptomycin and maintained in monolayer culture at 37 °C and 5% CO2 incubator. Plasmid transfections into 293FT cells were performed with LipofectamineTM2000 (Invitrogen). To generate C2C12 cells expressing human IR-A, 293FT cells were transfected with pBabe-IR-A-GFP, pCMV-gag / pol, and pCMV-VSV-G. Virus were collected at 2- and 3-days after transfection and concentrated with homemade virus concentrator. C2C12 cells were infected with concentrated virus and polybrene (4 ug / ml). Cells were selected with 2 mg / ml of puromycin at 3 days after infection and sorted using FACS sorter (Sony Ma900). To generate IGF1R knockout C2C12 cells expressing human IR-A (C2C12-IR), lentiviruses were packaged in 293FT cells by transfecting the cells with lentiCRISPRTMvector (Addgene # 52962, gRNA: CACCGCTATGGTGGAGAGGTAACAG (SEQ ID NO: 76)), psPAX2 (Addgene #12260), and pMD2.G (Addgene #12259). Viruses were collected at 2- and 3-days after transfection and concentrated. C2C12 cells expressing IR-A were infected with concentrated virus and polybrene (4 mg / ml). Cells were selected with blasticidin (10 mg / ml) for 3 weeks. Following passage of an aliquot of each cell line for three to four weeks, a fresh batch of cells was thawed and propagated. There were no signs of mycoplasma contamination. IR signaling assay The IR signaling assay was performed as described earlier with some modification9,10,20,64. For IR mutants assay, 293FT cells were transfected with Myc-tagged IR mutants or WT. One day later, the cells were serum starved for 14-16 h. Serum-starved cells were treated with insulin (I2526, Sigma) or designed binders for 10 min. For binder validation, DKO-IR-A, DKO-IR-B, or C2C12-IR cells were used. Two days after seeding, the cells were serum starved for 6 h. Serum-starved cells were treated with insulin or designed binders for 10 min. To analyze antagonistic effects of binders, cells were serum starved for 4 h, treated with the indicated concentrations of binders for 1 h, and then treated for 10 min with insulin at the indicated concentrations. After treatment, cells were incubated with cell lysis buffer B [50 mM Hepes pH 7.4,150 mM NaCl, 10% (v / v) Glycerol, 1% (v / v) Triton X-100, 1 mM EDTA, 10 mM sodiumfluoride, 2 mM sodium orthovanadate, 10 mM sodium pyrophosphate, 0.5 mM dithiothreitol(DTT), 2 mM phenylmethylsulfonyl fluoride (PMSF)] supplemented with cOmpleteTMProtease Inhibitor Cocktail (Roche), PhosSTOPTM(Roche), and 25 U / ml turbo nuclease (Accelagen) on ice for 1 h. After centrifugation at 18,213 g at 4°C for 20 min, cell lysate samples were made with SDS-PAGE protein loading buffer. Cell lysates were analyzed by SDS-PAGE and Western blotting. Anti-IR-pY1150 / 1151 (1:2000, 19H7, Cell signaling; labeled as pY IR (or pY IGF1R), #3024), anti-IR-pY1146 (1:1000, D6D5L, Cell signaling, #80732S), anti-IR-pY960 (1:1000, Invitrogen, #44-800G), anti-IR-Y1316 (1:1000, Invitrogen, #44-807G), anti-IR-Y1322 (1:1000, Invitrogen, #44-809G), anti-Myc (1:2000; 9E10, Roche; labeled as IR, #11667149001), anti-IR (1:500; CT3, Santa Cruz, sc-57342), anti-AKT (WB, 1:2000; 40D4, #2920), anti-pS473 AKT (WB, 1:2000; D9E, #4060), anti- ERK1 / 2 (WB, 1:2000; L34F12, #4696), and anti-pERK1 / 2 (WB, 1:2000; 197G2, #4377) were used as primary antibodies. For quantitative Western blots, anti-rabbit immunoglobulin G (IgG) (H+L) (DylightTM800 conjugates, #5151) and anti-mouse IgG (H+L) (Dylight 680 conjugates, #5470) (Cell signaling) were used as secondary antibodies. The membranes were scanned with the OdysseyTMInfrared Imaging System (LI-COR, Lincoln, NE). IR signaling analysis in vivo IR signaling in vivo analysis was performed as described earlier with some modifications10,20,65,66.2-3-months-old male mice were fasted overnight. Following anesthesia, mice were injected with 6 nmol Humulin (Eli Lilly) or 9 nmol designed agonists per mouse via inferior vena cava. Livers and skeletal muscle were removed at 5 min and 10 min after injection, respectively. Tissues were homogenized in cell lysis buffer B supplemented with cOmpleteTMProtease Inhibitor Cocktail (Roche), PhosSTOPTM(Sigma), and 25 U / ml turbo nuclease (Accelagen), homogenized with FisherbrandTM Bead Mill homogenizer, and then incubated on ice for 1hr. After centrifuge at 20,817 g at 4ºC for 30 min, the concentrations of cell lysate were measured using Micro BCA Protein Assay Kit (Thermo Scientific). The lysates were then analyzed by quantitative western blotting (Li- COR, Lincoln, NE). Primary mouse hepatocytes isolation Mouse primary hepatocytes were isolated from 2-3-month-old male mice with a standard two-step collagenase perfusion procedure as described earlier20,66. Isolated hepatocytes were resuspended with attached medium [Williams’ Medium E supplemented with 5% (v / v) FBS, 10 nM insulin, 10 nM dexamethasone, and 1% penicillin / streptomycin] and plated on collagen (Sigma, C3867)-coated dishes. After 4 h, the medium was changed to serum free low-glucose DMEM supplemented with 1% penicillin / streptomycin. After 14-16 h, the cells were treated with insulin to analyze IR signaling and IR endocytosis. Cell surface biotinylation and streptavidin pulldown Cell surface labeling was performed as previously described with some modifications67. Primary mouse hepatocytes were treated with 100 nM ligands for 30 min. Cells were washed with in cold PBS, pH 8.0 (Corning, 21-030-CM) and incubated in 0.5 mg / ml EZ-linkTMSulfo-NHS-LC-Biotin (Thermo Fisher, 21335) dissolved in PBS, pH 8.0 on ice for 10 min. The labeling reaction was quenched in 50 mM glycine in two sequential 10 min incubations on ice. Cells were incubated with RIPA buffer (Thermo Fisher, 89901) supplemented with Halt Protease & Phosphatase inhibitor cocktail (Thermo Fisher, 78442) on ice for 1 h. After centrifugation at 18,213 g at 4°C for 20 min, the supernatant was taken for streptavidin pulldown. Lysates were incubated with streptavidin magnetic beads (Thermo Fisher, 88817) overnight and washed 3 times with RIPA buffer. Beads were eluted with SDS sample buffers and samples were analyzed by SDS-PAGE and Western blotting. Anti-IR (1:500; CT3, Santa Cruz, sc-57342), anti-beta catenin (1:1000, D10A8, Cell signaling, #8480), anti-Actin (1:1000, C4, Santa Cruz, #sc-47778). Insulin tolerance test Mice were fasted for 2h (healthy mice) or 5 h (diet-induced obese mice) and their blood glucose levels (T=0) were measured with tail bleeding (Contour Next). Mice were then injected intraperitoneally with PBS, Humulin or designed agonists. Their blood glucose levels at the indicated time points after injection were measured with tail bleeding. Cell proliferation assay C2C12-IR cells were seeded in 0.1 million cells per 35 mm cell culture dish. One day later, cells were serum starved for 24 h and then treated for 24 h with indicated concentrations of binders in the presence or absence of insulin. One day later, cells were pulsed with 10 M bromodeoxyuridine (BrdU) for 2 h, then fixed with 70% cold ethanol. The fixed cells were washed with FACS blocking solution (0.02% Triton X-100 and 1% BSA in PBS), denatured with 3N HCl for 30 min, neutralized with phosphate / citrate buffer for 10 min, and then washed three times with the FACS blocking solution.10uL of FITC-anti-BrdU antibody (BD, # 556028) was added to each sample and incubated at RT for 2.5 h. cells were washed with the FACS blocking solution and stained with propidium iodide (BD, #550825). Cells were analyzed using BD FACSCantoTMII from the Flow Cytometry Core of the Columbia Center for Translational Immunology (CCTI) and Herbert Irving Comprehensive Cancer Center (HICCC). Data was processed with FlowJoTM. Data statistical analysis PrismTM10 was used for the generation of graphs and for statistical analyses. Results are presented as mean ± s.d. or mean ± s.e.m. Two-tailed unpaired t tests were used for pairwise significance analysis. For dose-response curve, the relative intensity signals were fitted to a four-parameter sigmoidal concentration-response curve, from which the pEC50 values (negative logarithmic values of half-maximum effective concentration (EC50) values) were used to calculate the mean and s.e.m. Log-transformed EC50 values were analyzed by Extra sum-of-squares F Test in Prism, between insulin and designed agonists. Two-way ANOVA followed by Tukey’s multiple comparisons tests were used in case of more than two factors’ comparisons. No power analysis for sample sizes were performed. Randomization and blinding methods were not used, and data were analyzed after the completion of all data collection in each experiment. References 1. Haeusler, R. A., McGraw, T. E. & Accili, D. Biochemical and cellular properties of insulin receptor signalling. Nat. Rev. Mol. Cell Biol.19, 31–44 (2018). 2. Petersen, M. C. & Shulman, G. I. Mechanisms of Insulin Action and Insulin Resistance. Physiol. Rev.98, 2133–2223 (2018). 3. White, M. F. & Kahn, C. R. Insulin action at a molecular level - 100 years of progress. Mol. Metab.52, 101304 (2021). 4. Kasuga, M., Karlsson, F. A. & Kahn, C. R. Insulin stimulates the phosphorylation of the 95,000-dalton subunit of its own receptor. Science 215, 185–187 (1982). 5. Kasuga, M., Zick, Y., Blithe, D. L., Crettaz, M. & Kahn, C. R. Insulin stimulates tyrosine phosphorylation of the insulin receptor in a cell-free system. Nature 298, 667– 669 (1982). 6. White, M. F., Maron, R. & Kahn, C. R. Insulin rapidly stimulates tyrosine phosphorylation of a Mr-185,000 protein in intact cells. Nature 318, 183–186 (1985). 7. Sun, X. J. et al. Structure of the insulin receptor substrate IRS-1 defines a unique signal transduction protein. Nature 352, 73–77 (1991). 8. Uchikawa, E., Choi, E., Shang, G., Yu, H. & Bai, X.-C. Activation mechanism of the insulin receptor revealed by cryo-EM structure of the fully liganded receptor–ligand complex. Elife 8, e48630 (2019). 9. An, W. et al. Activation of the insulin receptor by insulin-like growth factor 2. Nat. Commun.15, 2609 (2024). 10. Li, J. et al. Synergistic activation of the insulin receptor via two distinct sites. Nat. Struct. Mol. Biol.29, 357–368 (2022). 11. Choi, E. & Bai, X.-C. The Activation Mechanism of the Insulin Receptor: A Structural Perspective. Annu. Rev. Biochem.92, 247–272 (2023). 12. McKern, N. M. et al. Structure of the insulin receptor ectodomain reveals a folded-over conformation. Nature 443, 218–221 (2006). 13. Croll, T. I. et al. Higher-resolution structure of the human insulin receptor ectodomain: Multi-modal inclusion of the insert domain. Structure 24, 469–476 (2016). 14. Scapin, G. et al. Structure of the insulin receptor–insulin complex by single-particle cryo-EM analysis. Nature 556, 122–125 (2018). 15. Gutmann, T. et al. Cryo-EM structure of the complete and ligand-saturated insulin receptor ectodomain. J. Cell Biol.219, jcb.201907210 (2020). 16. Sims, E. K., Carr, A. L. J., Oram, R. A., DiMeglio, L. A. & Evans-Molina, C.100 years of insulin: celebrating the past, present and future of diabetes therapy. Nat. Med.27, 1154–1164 (2021). 17. Cao, L. et al. Design of protein-binding proteins from the target structure alone. Nature 605, 551–560 (2022). 18. Hiranuma, N. et al. Improved protein structure refinement guided by deep learning based accuracy estimation. Nat. Commun.12, 1340 (2021). 19. Seino, S. & Bell, G. I. Alternative splicing of human insulin receptor messenger RNA. Biochem. Biophys. Res. Commun.159, 312–316 (1989). 20. Park, J. et al. Activation of the insulin receptor by an insulin mimetic peptide. Nat. Commun.13, 5594 (2022). 21. Watson, J. L. et al. De novo design of protein structure and function with RFdiffusion. Nature 620, 1089–1100 (2023). 22. Bennett, N. R. et al. Improving de novo protein binder design with deep learning. Nat. Commun.14, 2625 (2023). 23. Li, J., Choi, E., Yu, H. & Bai, X.-C. Structural basis of the activation of type 1 insulin- like growth factor receptor. Nat. Commun.10, 4567 (2019). 24. Kavran, J. M. et al. How IGF-1 activates its receptor. Elife 3, (2014). 25. Kjeldsen, T. et al. The ligand specificities of the insulin receptor and the insulin-like growth factor I receptor reside in different regions of a common binding site. Proc. Natl. Acad. Sci. U. S. A.88, 4404–4408 (1991). 26. Schumacher, R., Mosthaf, L., Schlessinger, J., Brandenburg, D. & Ullrich, A. Insulin and insulin-like growth factor-1 binding specificity is determined by distinct regions of their cognate receptors. J. Biol. Chem.266, 19288–19295 (1991). 27. Jensen, M., Hansen, B., De Meyts, P., Schäffer, L. & Ursø, B. Activation of the insulin receptor by insulin and a synthetic peptide leads to divergent metabolic and mitogenic signaling and responses. J. Biol. Chem.282, 35179–35186 (2007). 28. Kanter, J. E. et al. A novel strategy to prevent advanced atherosclerosis and lower blood glucose in a mouse model of metabolic syndrome. Diabetes 67, 946–959 (2018). 29. Liao, Z. et al. Comprehensive insulin receptor phosphorylation dynamics profiled by mass spectrometry. FEBS J.289, 2657–2671 (2022). 30. Gehart, H., Kumpf, S., Ittner, A. & Ricci, R. MAPK signalling in cellular metabolism: stress or wellness? EMBO Rep.11, 834–840 (2010). 31. Hall, C., Yu, H. & Choi, E. Insulin receptor endocytosis in the pathophysiology of insulin resistance. Exp. Mol. Med.52, 911–920 (2020). 32. Chen, Huang, L., Qi, X. & Chen, C. Insulin Receptor Trafficking: Consequences for Insulin Sensitivity and Diabetes. Int. J. Mol. Sci.20, (2019). 33. Choi, E. et al. Mitotic regulators and the SHP2-MAPK pathway promote IR endocytosis and feedback regulation of insulin signaling. Nat. Commun.10, 1473 (2019). 34. Semple, R. K., Savage, D. B., Cochran, E. K., Gorden, P. & O’Rahilly, S. Genetic syndromes of severe insulin resistance. Endocr. Rev.32, 498–514 (2011). 35. Melvin, A., O’Rahilly, S. & Savage, D. B. Genetic syndromes of severe insulin resistance. Curr. Opin. Genet. Dev.50, 60–67 (2018). 36. Musso, C. et al. Clinical course of genetic diseases of the insulin receptor (type A and Rabson-Mendenhall syndromes): a 30-year prospective. Medicine 83, 209–222 (2004). 37. Grasso, V. et al. Six cases with severe insulin resistance (SIR) associated with mutations of insulin receptor: Is a Bartter-like syndrome a feature of congenital SIR? Acta Diabetol.50, 951–957 (2013). 38. Kadowaki, T. et al. Five mutant alleles of the insulin receptor gene in patients with genetic forms of insulin resistance. J. Clin. Invest.86, 254–264 (1990). 39. Bamborschke, D. et al. Ultra-rapid emergency genomic diagnosis of Donahue syndrome in a preterm infant within 17 hours. Am. J. Med. Genet. A 185, 90–96 (2021). 40. Hart, L. M. et al. An insulin receptor mutant (Asp707 --> Ala), involved in leprechaunism, is processed and transported to the cell surface but unable to bind insulin. J. Biol. Chem.271, 18719–18724 (1996). 41. Rocchi, S. Interaction of SH2-containing protein tyrosine phosphatase 2 with the insulin receptor and the insulin-like growth factor-I receptor: studies of the domains involved using the yeast two-hybrid system. Endocrinology 137, 4944–4952 (1996). 42. Myers, M. G., Jr et al. The COOH-terminal tyrosine phosphorylation sites on IRS-1 bind SHP-2 and negatively regulate insulin signaling. J. Biol. Chem.273, 26908–26914 (1998). 43. Chen, Y.-N. P. et al. Allosteric inhibition of SHP2 phosphatase inhibits cancers driven by receptor tyrosine kinases. Nature 535, 148–152 (2016). 44. Hof, P., Pluskey, S., Dhe-Paganon, S., Eck, M. J. & Shoelson, S. E. Crystal structure of the tyrosine phosphatase SHP-2. Cell 92, 441–450 (1998). 45. Nichols, R. J. et al. RAS nucleotide cycling underlies the SHP2 phosphatase dependence of mutant BRAF-, NF1- and RAS-driven cancers. Nat. Cell Biol.20, 1064–1073 (2018). 46. Ling, S. et al. Inequalities in cancer mortality trends in people with type 2 diabetes: 20 year population-based study in England. Diabetologia 66, 657–673 (2023). 47. Pearson-Stuttard, J. et al. Trends in predominant causes of death in individuals with and without diabetes in England from 2001 to 2018: an epidemiological analysis of linked primary care records. Lancet Diabetes Endocrinol 9, 165–173 (2021). 48. Song, M. Cancer overtakes vascular disease as leading cause of excess death associated with diabetes. Lancet Diabetes Endocrinol 9, 131–133 (2021). 49. Leitner, B. P., Siebel, S., Akingbesote, N. D., Zhang, X. & Perry, R. J. Insulin and cancer: a tangled web. Biochem. J 479, 583–607 (2022). 50. Smith, T. J. & Janssen, J. A. M. J. L. Insulin-like Growth Factor-I Receptor and Thyroid-Associated Ophthalmopathy. Endocr. Rev.40, 236–267 (2019). 51. Tsui, S. et al. Evidence for an association between thyroid-stimulating hormone and insulin-like growth factor 1 receptors: a tale of two antigens implicated in Graves’ disease. J. Immunol.181, 4397–4405 (2008). 52. Buck, E. et al. Compensatory insulin receptor (IR) activation on inhibition of insulin- like growth factor-1 receptor (IGF-1R): rationale for cotargeting IGF-1R and IR in cancer. Mol. Cancer Ther.9, 2652–2664 (2010). 53. Wicky, B. I. M. et al. Hallucinating symmetric protein assemblies. Science 378, 56–61 (2022). 54. Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017). 55. Zhang, K. Gctf: Real-time CTF determination and correction. J. Struct. Biol.193, 1–12 (2016). 56. Kimanius, D., Dong, L., Sharov, G., Nakane, T. & Scheres, S. H. W. New tools for automated cryo-EM single-particle analysis in RELION-4.0. Biochem. J 478, 4169– 4185 (2021). 57. Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). 58. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D Biol. Crystallogr.66, 486–501 (2010). 59. Liebschner, D. et al. Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr D Struct Biol 75, 861–877 (2019). 60. Chen, V. B. et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. D Biol. Crystallogr.66, 12–21 (2010). 61. Pettersen, E. F. et al. UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci.30, 70–82 (2021). 62. Pettersen, E. F. et al. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem.25, 1605–1612 (2004). 63. Nagao, H. et al. Distinct signaling by insulin and IGF-1 receptors and their extra- and intracellular domains. Proc. Natl. Acad. Sci. U. S. A.118, (2021). 64. Li, J., Wu, J., Hall, C., Bai, X.-C. & Choi, E. Molecular basis for the role of disulfide- linked CTs in the activation of insulin-like growth factor 1 receptor and insulin receptor. Elife 11, (2022). 65. Park, J. et al. MAD2-Dependent Insulin Receptor Endocytosis Regulates Metabolic Homeostasis. Diabetes 72, 1781–1794 (2023). 66. Choi, E., Zhang, X., Xing, C. & Yu, H. Mitotic Checkpoint Regulators Control Insulin Signaling and Metabolic Homeostasis. Cell 166, 567–581 (2016). 67. Ramachandran, K. V. & Margolis, S. S. A mammalian nervous-system-specific plasma membrane proteasome complex that modulates neuronal function. Nat. Struct. Mol. Biol. 24, 419–430 (2017).

Claims

We claim 1. A polypeptide comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:13, not including any amino acid insertions at identified insertion sites (i.e., any insertions are not considered when determining percent identity to the reference polypeptide), wherein the polypeptide binds to the F1 domain of insulin receptor.

2. The polypeptide of claim 1, comprising an amino acid sequence at least 75% identical to the amino acid sequence of SEQ ID NO:

13.

3. The polypeptide of claim 1, comprising an amino acid sequence at least 90% identical to the amino acid sequence of SEQ ID NO:

13.

4. The polypeptide of any one of claims 1-3, wherein substitutions relative to SEQ ID NO:13 are selected from substitutions listed in Options 1, 2, or 3 of Table 1.

5. The polypeptide of any one of claims 1-3, wherein substitutions relative to SEQ ID NO:13 are selected from substitutions listed in Options 1 or 2 of Table 1.

6. The polypeptide of any one of claims 1-3, wherein substitutions relative to SEQ ID NO:13 are selected from substitutions listed in Option 1 of Table 1. The polypeptide of any one of claims 1-6, wherein at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, or all 18 identified interface are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:

13.

8. The polypeptide of any one of claims 1-7, wherein at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or all 12 core residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:

13.

9. The polypeptide of any one of claims 1-8, wherein all identified interface residues and key residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:13.

10. The polypeptide of any one of claims 1-9, comprising an insertion in one or more insertion site relative to the reference sequence.

11. A polypeptide comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:14, not including any amino acid insertions at identified insertion sites (i.e., any insertions are not considered when determining percent identity to the reference polypeptide), wherein the polypeptide binds to the F1 domain of insulin receptor.

12. The polypeptide of claim 11, comprising an amino acid sequence at least 75% identical to the amino acid sequence of SEQ ID NO:

14.

13. The polypeptide of claim 11, comprising an amino acid sequence at least 90% identical to the amino acid sequence of SEQ ID NO:

14.

14. The polypeptide of any one of claims 11-13, wherein substitutions relative to SEQ ID NO:14 are selected from substitutions listed in Options 1, 2, or 3 of Table 2.

15. The polypeptide of any one of claims 11-13, wherein substitutions relative to SEQ ID NO:14 are selected from substitutions listed in Options 1 or 2 of Table 2.

16. The polypeptide of any one of claims 11-13, wherein substitutions relative to SEQ ID NO:14 are selected from substitutions listed in Option 1 of Table 2.

17. The polypeptide of any one of claims 11-16, wherein at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or all 16 identified interface are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:

14.

18. The polypeptide of any one of claims 11-17, wherein at least 1, 2, 3, 4, 5, 6, 7, or all 8 core residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:14.

19. The polypeptide of any one of claims 11-18, wherein all identified interface residues and key residues are identical (not substituted), or conservatively substituted, relative to SEQ ID NO:

14.

20. The polypeptide of any one of claims 11-19, comprising an insertion in one or more insertion site relative to SEQ ID NO:

14.

21. A polypeptide comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:1-12 and 15-28, wherein the polypeptide binds to the F1 domain of insulin receptor.

22. The polypeptide of claim 21, comprising an amino acid sequence at least 75% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:1-12 and 15-28.

23. The polypeptide of claim 21, comprising an amino acid sequence at least 90% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:1-12 and 15-28.

24. A polypeptide comprising an amino acid sequence at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:29-31, wherein the polypeptide binds to the L1 domain of insulin receptor.

25. The polypeptide of claim 24, comprising an amino acid sequence at least 85% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:29- 31.

26. The polypeptide of claim 24, comprising an amino acid sequence at least 90% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:29-27. The polypeptide of any one of claims 21-26, wherein amino acid substitutions relative to the reference sequence are conservative amino acid substitutions.

28. A conjugate, comprising: (a) the polypeptide of any one of claims 1-27; and (b) a therapeutic or diagnostic moiety.

29. The conjugate of claim 28, wherein the therapeutic or diagnostic moiety comprises insulin.

30. A fusion protein comprising: (a) a first polypeptide according to any one of claims 1-23; and (b) a second polypeptide according to any one of claims 24-27, or a polypeptide comprising an amino acid sequence at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO:75 and which binds to the L1 domain of the insulin receptor; optionally wherein the first polypeptide and the second polypeptide are linked by an amino acid linker.

31. The fusion protein of claim 30, wherein the first polypeptide is N-terminal to the second polypeptide.

32. A fusion protein comprising: (a) a first polypeptide according to any one of claims 1-23; and (b) a second polypeptide according to any one of claims 1-23; optionally wherein the first polypeptide and the second polypeptide are linked by an amino acid linker.

33. The fusion protein of any one of claims 30-32, wherein the amino acid linker is present.

34. The fusion protein of any one of claims 30-33, wherein the amino acid linker is between 0-40 amino acids in length.

35. The fusion protein of any one of claims 30-34, comprising an amino acid sequence at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence selected from the group consisting of SEQ ID NO:47-74.

36. A nucleic acid encoding the polypeptide or fusion protein of any preceding claim.

37. An expression vector comprising the nucleic acid of claim 36 operatively linked to a promoter.

38. A host cell comprising the polypeptide, fusion protein, nucleic acid, or expression vector of any preceding claim.

39. A pharmaceutical composition, comprising the polypeptide, fusion protein, nucleic acid, expression vector, or host cell of any of the preceding claims, and a pharmaceutically acceptable carrier.

40. A method for using, or a use of the polypeptide, conjugate, fusion protein, nucleic acid, expression vector, host cell, and / or the pharmaceutical composition of any of the preceding claims, for any suitable purpose including but not limited to those disclosed herein.

41. The method or use of claim 40, wherein the purpose includes, but is not limited to, treating or limiting development of type 1 or type 2 diabetes, cancer, autoimmune disease, or thyroid eye disease.

42. The method or use of claim 40, wherein the purpose includes, but is not limited to, enhancing insulin sensitivity in type 2 diabetes and regulating glucose and / or lipid metabolism in type 1 diabetes and congenital severe insulin resistance.

Images