WO2025193981A1

WO2025193981A1 - Synthetic multivalent fusion proteins, manufacture, and uses thereof

Info

Publication number: WO2025193981A1
Application number: PCT/US2025/019826
Authority: WO
Inventors: Mireia Sola COLOM; III James Wayne BOWMAN; Ben Alexander MEINEN; Christopher David BAHL; III Frank David TEETS; Edson Norberto Carcamo NORIEGA; Maria Eugenia LLASES
Original assignee: Ai Proteins Inc
Current assignee: Ai Proteins Inc
Priority date: 2024-03-15
Filing date: 2025-03-13
Publication date: 2025-09-18
Anticipated expiration: 2026-09-15

Abstract

The present disclosure provides fusion proteins containing CD16a and LILRB4 binding domains. CD16a binding domains, LILRB4 binding domains, fusions of CD16a and LILRB4 binding domains, compositions containing such binding domains, and methods of making and using such binding domains, are provided herein, including their uses in NK cell activation and/or CD16a-mediated cytotoxicity of LILRB4-expressing tumor cells.

Description

SYNTHETIC MULTIVALENT FUSION PROTEINS, MANUFACTURE, AND USES THEREOF

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of and priority to U.S. Provisional Application No. 63/566,182, filed March 15, 2024, the entire contents of which are hereby incorporated by reference for all purposes.

SEQUENCE LISTING

[0002] The instant application contains a Sequence Listing which has been submitted electronically in XML format. The Sequence Listing XML is hereby incorporated by reference in its entirety. The XML file, created on February 20, 2025, is named AIP- 012WO_SL.xml and is 114,688 bytes in size.

FIELD

[0003] The disclosure relates generally to multivalent fusion proteins with cluster of differentiation 16a (CD 16a) and leukocyte immunoglobulin-like receptor B4 (LILRB4) binding domains, and their manufacture and use in the treatment of various disorders, including cancer.

BACKGROUND

[0004] According to the National Cancer Institute (“NCI”), in 2022 there were over 1.9 million estimated new cancer cases and over 600,000 estimated cancer deaths. See, e.g., Siegel et al. (2022) CA CANCER J. CLIN. 72: 7-33. Globally, there were an estimated 19.1 million new cases. See, e.g., Sung et al. (2021) CA CANCER J. CLIN 71: 209-249. NCI estimates that direct cancer-related medical costs in the US were $183 billion in 2015 and are projected to increase to $246 billion by 2030. See, e.g., The American Cancer Society Cancer Action Network, The Costs of Cancer 2020 Edition. Cancer results from abnormal proliferation of cells and can occur in any tissue in the body. These abnormal cells are capable of escaping the body’s natural defenses, preventing the immune system from fighting cancer cells. There is a need for immunotherapies to redirect the body’s natural defenses to target and eliminate cancer cells. One potential strategy of interest is immune cell engagers that have the ability to recognize molecules on both immune cells and cancer cells bringing the target cancer cell into proximity to trigger killing by the immune cell. [0005] Natural killer (NK) cells are innate lymphoid cells that recognize tumor cells and virally infected cells and release cytokines to generate a rapid immune response. See, e.g., Chu et al. (2022) J. TRANSL. MED. 20: 240; Paul et al. (2017) FRONT. IMMUNOL. 8: 1124; Mandal et al. (2015) HEMATOL. ONCOL. STEM CELL THER. 8: 47-55. Compared to certain other types of immune cells, NK cells are able to detect and kill virus and tumor cells without “pre-activation,” and are less likely to induce a cytokine storm, making them safer (for patients). See, e.g, Rahman et al. (2024) Histology, Natural Killer Cells, STATPEARLS at ncbi.nlm.nih.gov/books/NBK565844/; Page et al. (2024) CELL MOL. IMMUNOL. https://doi.org/10.1038/s41423-024-01145-x; Wang et al. (2020) MED. COMM. 4: e422. Moreover, there are reports that NK cells can trigger remission in certain malignancies (e.g., acute myeloid leukemia (AML)). See, e.g., Chu et al. (2022) supra.

[0006] NK cells express a number of receptor molecules (e.g., CD 16a) on their cell surfaces. The CD16 family of proteins (also known as Fey receptors, or FcyR) is a group of receptor proteins that modulates immune cell responses via the antibody-dependent cell- mediated cytotoxicity (ADCC) upon binding to the Fc region of antibodies. See, e.g, Gonzalez (2022) supra. The CD 16 protein family includes CD 16a, which is predominantly expressed on NK cells, and CD 16b, which is predominantly expressed on neutrophils. See, e.g, Coenon et al. (2022) FRONT IMMUNOL. 13:913215; Li et al. (2016) EXP MOL PATHOL. 101: 281-289; Ravetch c/ u/. (1989) J EXP MED. 170:481-497. CD 16a and CD 16b are known to have different binding affinities to Fc regions. See, e.g, Roberts et al. (2018) J. BIOL. CHEM. 293(51):19899. CD16a and CD16b share over 97% amino acid sequence identity, which has historically presented challenges for developing specific binders.

Engagement of NK cells via CD 16a has been associated with activation of immune responses, whereas Fc binding to CD16b in neutrophils can dampen ADCC and reduce or prevent immune responses. See, e.g., Treffers et al. (2019) FRONT. IMMUNOL. 9:3124. Because NK cells play an important role in antitumor and antiviral responses, as well as regulation of other immune cells, engagement of NK cells via CD16a shows promise as an immunotherapeutic candidate. See Vivier et al. (2008) NAT. IMMUNOL. 9: 503. It is believed that certain specific amino acid residues in CD 16a are believed to be responsible for CD16a’s stronger binding affinity to the Fc region than CD16b, representing an important characteristic to consider in development of CD16a-binding therapeutics. See, e.g., Roberts et al. (2018) J. BIOL. CHEM. 293(51):19899; see also Ravetch et al. (1989) supra. [0007] Despite the efforts made to date, there remains a need for immunotherapies that can specifically target CD16a relative to CD16b. See, e.g., Nikkhoi et al. (2023) FRONT. IMMUNOL. 3: 1039969. High homology between CD 16a and CD 16a poses an ongoing challenge in specifically targeting CD16a. See, e.g., Zhang et al. (2023) FRONT. IMMUNOL. 14:1207276.

[0008] Various cancer therapies have been developed in which NK cells are targeted to tumor cells that express tumor-specific antigens. Approaches include NK cell engagers that can simultaneously bind NK cells and tumor cells so that the NK cells are brought into close proximity with the tumor cells. Increased safety and efficacy of NK engagers as compared to other engager approaches (e.g, such as those targeting T-cells) is promising, but cost and manufacturability remain serious limitations. See, e.g., Chu et al. (2022) supra. In addition, specificity for CD 16a (on NK cells) and certain cancer cell antigens (e.g, on tumor cells) remains a challenge. The high homology between CD 16a and CD 16b has been a hurdle in developing molecules highly specific to CD 16a expressing NK cells (expressing CD16a) rather than, for example CD16b-expressing neutrophils. See, e.g., Nikkhoi et al. (2023) FRONT. IMMUNOL. 3: 1039969. When developing NK engagers it is important to select a tumor-associated antigen with high specificity for tumor cells to properly orient NK-mediated cytotoxicity to tumor cells rather than to other healthy cells.

[0009] Accordingly, despite advances made to date, there remains a need for improvements in cancer treatment.

SUMMARY

[0010] The disclosure is based, in part, upon the discovery of synthetic fusion proteins comprising CD 16a and LILRB4 binding domains. The synthetic CD 16a binding domain specifically and preferentially binds CD 16a (relative to CD 16b) and modulates natural killer (NK) cell activity, for example, by triggering NK cell activation. The synthetic LILRB4 binding domain specifically binds to LILRB4, including as expressed on the surface of cancer cells. Fusion proteins comprising at least one CD 16a synthetic binding domain and at least one LILRB4 synthetic binding domain facilitate apposition of NK cells to tumor cells, improving LILRB4-expressing tumor cell attack by NK cells. Provided herein, are technologies (e.g., compositions and methods of use) providing proteins containing CD 16a and LILRB4 binding domains that can engage natural killer cells through receptors such as CD 16a to attack target cells, such as cancer cells through surface- expressed antigens, such as tumor-associated antigens like LILRB4, resulting in death of the cancer cells. Compositions as provided herein have the potential for improved safety, selectivity, efficacy, potency, manufacturability, scalability, and stability compared to currently available methods and can be used to address such needs, compositions and methods.

[0011] Accordingly, the disclosure provides, among other things, fusion proteins comprising CD 16a and LILRB4 binding domains, methods of making such binding proteins, and methods of using such proteins to treat a disease or disorder by NK cell- mediated cytotoxicity of a target cell such as an LILRB4-expressing cancer cell.

[0012] In one aspect, the disclosure provides a fusion protein comprising a first binding domain that binds a first target attached by at least one linker to a second binding domain that binds to a second target, wherein the first and second binding domains are synthetic binding domains comprising synthetic binding proteins that each have an N-terminal amino acid residue and a C-terminal amino acid residue, and wherein the first target is CD 16a and the second target is LILRB4.

[0013] In certain embodiments, the first binding domain and the second binding domain are linked by a first linker. In certain embodiments, the C-terminal amino acid residue of the first binding domain is linked to the N-terminal amino acid residue of the second binding domain. Alternatively, in certain other embodiments, the C-terminal amino acid residue of the second binding domain is linked to the N-terminal amino acid residue of the first binding domain.

[0014] In certain embodiments, the fusion protein further comprises a third binding domain comprising an N-terminal amino acid residue and a C-terminal amino acid residue. In some embodiments, the N-terminal amino acid residue of the third binding domain is attached to the C-terminal amino acid residue of the second binding domain. Alternatively, in certain other embodiments, the N-terminal amino acid residue of the third binding domain is attached to the C-terminal residue of the first binding domain. In certain embodiments, the C-terminal amino acid residue of the third binding domain is attached to the N-terminal amino acid of the first binding domain. In some embodiments, the third binding domain is a synthetic binding protein that binds CD 16a.

[0015] In certain embodiments, the first linker of the fusion protein comprises an amino acid sequence selected from any of SEQ ID NOs: 64-85 and/or Table 4. [0016] In certain embodiments, the third binding domain of the fusion protein is linked to the first binding domain or the second binding domain by a second linker. In some such embodiments, second linker comprises an amino acid sequence selected from any of SEQ ID NOs: 64-85 and/or Table 4. In certain embodiments, the first linker and the second linker are different.

[0017] In certain embodiments, the amino acid sequence of the first binding domain and/or the third binding domain of the fusion protein each comprise at least one amino acid sequence selected from any of SEQ ID NOs: 20-42, 88-90, and/or as set forth in any of Tables 1A, IB, 1C, and/or ID. In some embodiments, the amino acid sequence of the first binding domain and/or the third binding domain comprises at least one amino acid sequence selected from any of SEQ ID NOs: 1-19 and/or as set forth in Table IE.

[0018] In certain embodiments, the amino acid sequence of second binding domain of the fusion protein comprises at least one amino acid sequence selected from any of SEQ ID NOs: 44-46, 86-87, Table 2A and/or Table 2B. In some embodiments, the amino acid sequence of the second binding domain comprises at least one amino acid sequence selected from any of SEQ ID NOs: 43, 47-52, Table 2C and/or Table 2D.

[0019] In certain embodiments, a fusion protein of the present disclosure comprises an amino acid sequence selected from any of SEQ ID NOs: 53-63 and/or Table 3.

[0020] In certain embodiments, the first and/or third binding domains of the fusion protein bind CD16a with a binding affinity stronger than about 1 pM to about 0.001 nM; about 1 pM to about 0.01 nM; about 1 pM to about 0.75 nM; about 1 pM to about 0.5 nM; about 1 pM to about 0.25 nM; about 1 pM to about 1 nM; about 0.75 pM to about 1 nM, about 0.5 pM to about 1 nM; about 0.25 pM to about 1 nM; about 0.10 pM to about 1 nM; about 75 nM to about 1 nM; about 50 nM to about 1 nM; about 25 nM to about 1 nM; about 10 nM to about 1 nM; and about 5 nM to about 1 nM.

[0021] In certain embodiments, the first and/or third binding domains of the fusion protein bind CD16a with a binding affinity stronger than about 1 pM, about 0.75 pM, about 0.5 pM, about 0.25 pM, about 0.1 pM, about 75 nM, about 50 nM, about 25 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4 nM, about 3 nM, about 2 nM, about 1 nM, about 0.75 nM, about 0.5 nM, about 0.25 nM, about 0. 1 nM to about 0.01 nM, and about 0.01 nM to about 0.001 nM. [0022] In certain embodiments, the second binding domain of the fusion protein binds LILRB4 with a binding affinity stronger than about 1 pM to about 0.001 nM; about 1 pM to about 0.01 nM, about 1 pM to about 0.75 nM; about 1 pM to about 0.5 nM; about 1 pM to about 0.25 nM; about 1 pM to about 1 nM; about 0.75 pM to about 1 nM, about 0.5 pM to about 1 nM; about 0.25 pM to about 1 nM; about 0.10 pM to about 1 nM; about 75 nM to about 1 nM; about 50 nM to about 1 nM; about 25 nM to about 1 nM; about 10 nM to about 1 nM; and about 5 nM to about 1 nM.

[0023] In certain embodiments, the second binding domain of the fusion protein binds LILRB4 with a binding affinity stronger than about 1 pM, about 0.75 pM, about 0.5 pM, about 0.25 pM, about 0.1 pM, about 75 nM, about 50 nM, about 25 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4 nM, about 3 nM, about 2 nM, about 1 nM, about 0.75 nM, about 0.5 nM, about 0.25 nM, about 0.1 nM to about 0.01 nM, and about 0.01 nM to about 0.001 nM.

[0024] In certain embodiments, the disclosure provides a nucleic acid encoding a fusion protein as provided herein. In some embodiments, the disclosure provides a host cell comprising a nucleic acid encoding a fusion protein provided herein.

[0025] In one aspect, the disclosure provides a pharmaceutical composition comprising a fusion protein as provided herein and a pharmaceutically acceptable carrier. In certain embodiments, the pharmaceutical composition is formulated for administration by a systemic route. In some embodiments, the systemic route is intravenous administration.

[0026] In certain embodiments, the administration of the pharmaceutical composition is before, concomitant with, or after administration of at least one other treatment. In some embodiments, the at least one other treatment is selected from a biological agent (e.g., biologies, gene therapy, peptides), a small molecule (e.g., chemotherapy, corticosteroids, antivirals, antibiotics, anti-inflammatory agents, etc.), one or more cells (e.g, immunotherapy), and/or one or more mechanical interventions (e.g, surgery, cryotherapy, radiation).

[0027] In another aspect, the disclosure provides a method of treating a subject in need of treatment comprising administering to the subject an effective amount of a fusion protein or pharmaceutical composition provided herein. [0028] In one aspect, the disclosure provides a method of treating cancer comprising administering to a subject in need of treatment an effective amount of a fusion protein or pharmaceutical composition as provided herein.

[0029] In certain embodiments, a subject in accordance with the disclosure is diagnosed as having a cancer or a population of cancerous cells. In some embodiments, the cancer or cancerous cells are of myeloid origin. In certain embodiments, the cancer or cancerous cells are diagnosed as acute myeloid leukemia (AML), myeloma (e.g, multiple myeloma), lymphoma (e g, mantle cell lymphoma), or from a solid tumor origin. In certain embodiments, the cancer or the population of cancerous cells comprises cancer cells that express LILRB4.

[0030] In yet another aspect, the disclosure provides a method of increasing tumor cell death in a population of cells comprising cancer cells and NK cells, the method comprising exposing the population of cells to a fusion protein or pharmaceutical composition as provided herein, thereby to increase cancer cell death relative to cancer cell death in the absence of the fusion protein or the pharmaceutical composition.

[0031] In one aspect, the disclosure provides a method of increasing expression of CD69, CD25, and/or CD107 on an NK cell, comprising contacting a CD16a-expressing NK cell in the presence of an LILRB4-expressing tumor cell with a fusion protein provided herein, whereupon the fusion protein binds to the NK cell and the cancer cell and results in increased expression of CD69, CD25, and/or CD107 on the NK cell relative to the expression of the CD69, CD25, and/or CD107 prior to the contact. In certain embodiments, the NK cells display increased expression of CD 107.

[0032] In one aspect, the disclosure provides a method of stimulating an increase of IFNy release from a CD16a-expressing NK cell in the presence of an LILRB4 expressing cancer cell, comprising exposing the NK cell and the cancer cell to a fusion protein provided herein so that the fusion protein binds to the NK cell and the cancer cell and stimulates the increase of IFNy release from the NK cell.

[0033] In yet another aspect, the disclosure provides a method of treating a cancer in a subject in need of treatment, comprising administering to the subject an effective amount of a fusion protein or pharmaceutical composition as provided herein, thereby to treat the cancer in the subject. [0034] In certain embodiments, a subject of the disclosure is a mammal. In some embodiments, the mammal is a human. In certain embodiments, the human is diagnosed or suspected of having cancer cells expressing LILRB4.

[0035] These and other aspects and features of the disclosure are described in the following detailed description and claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0036] FIG. 1 is a schematic representation of a CD 16a receptor in complex with an exemplary synthetic CD 16a binding domain (denoted as a ribbon diagram and labeled as a synthetic miniprotein in FIG. 1). Upon binding of an exemplary synthetic CD16a binding domain to CD 16a and crosslinking with other receptors, NK cells are activated through multiple activation pathways.

[0037] FIG. 2 is a schematic representation of an LILRB4 protein expressed on the surface of a tumor cell in complex with a synthetic LILRB4 binding domain. Binding to LILRB4 targets synthetic LILRB4 binding domains to the surface of LILRB4-expressing cells, such as LILRB4-expressing tumor cells.

[0038] FIGs. 3A-3C are schematic representations of exemplary bivalent fusion proteins. FIG. 3A is a schematic representation of an exemplary bivalent CD16a-LILRB4 fusion protein, with the CD 16a synthetic binding domain bound to a CD 16a protein on an NK cell and the synthetic LILRB4 binding domain bound to LILRB4 expressed on the surface of a tumor cell. FIG. 3B is a schematic representation of an exemplary bivalent (“D1-D2”) CD 16a- LILRB4 fusion protein (C-terminus of an exemplary synthetic CD 16a binding domain joined to N-terminus of an exemplary synthetic LILRB4 binding domain via an exemplary linker). FIG. 3C is an exemplary bivalent (“D2-D1”) LILRB4-CD16a fusion protein (C-terminus of an exemplary synthetic LILRB4 protein joined to N-terminus of an exemplary synthetic CD 16a binding domain via an exemplary linker).

[0039] FIGs. 4A-4E are schematic representations of two exemplary trivalent fusion proteins (“D1-D2-D3”). FIG. 4A is an exemplary trivalent fusion protein comprising two exemplary synthetic LILRB4 binding domains with an exemplary synthetic CD 16a binding domain linked to each of the C-terminus of the first LILRB4 binding domain and the N- terminus of the second LILRB4 binding domain. FIG. 4B is an exemplary trivalent fusion protein comprising two exemplary synthetic CD 16a binding domains with an exemplary synthetic LILRB4 binding domain linked to each of the C-terminus of the first CD 16a binding domain and the N-terminus of the second CD 16a binding domain. FIG. 4C is an exemplary trivalent fusion protein comprising two exemplary synthetic LILRB4 binding domains with the C-terminus of the first LILRB-4 binding domain linked to the N-terminus of the second LILRB4 binding domain, and the C-terminus of the second LILRB4 binding domain linked to the N-terminus of an exemplary synthetic CD 16a binding domain. FIG. 4D is an exemplary trivalent fusion protein comprising two exemplary synthetic CD 16a binding domains with the C-terminus of the first CD 16a binding domain linked to the N- terminus of the second CD 16a binding domain, and the C-terminus of the second CD 16a binding domain linked to the N-terminus of an exemplary synthetic LILRB4 binding domain. FIG. 4E is an exemplary trivalent fusion protein comprising one exemplary synthetic LILRB4 binding domain with the C-terminus of the LILRB-4 binding domain linked to the N-terminus of a first exemplary CD 16a binding domain, and the C-terminus of the first CD 16a binding domain linked to the N-terminus of an exemplary synthetic CD 16a binding domain.

[0040] FIG. 5 is a schematic of a trivalent CD16a-LILRB4 fusion protein (“D1-D2-D3”), with each of two CD 16a synthetic binding domains bound to two CD 16a proteins on an NK cell and the synthetic LILRB4 binding domain bound to LILRB4 expressed on the surface of a cancer cell.

[0041] FIGs. 6A and 6B are graphs of circular dichroism (CD) spectra, showing protein stability during heating and/or cooling. FIG. 6A is a graph of a CD spectra showing the stability profile of an exemplary synthetic monovalent CD 16a binding domain (Reference Miniprotein 1 (SEQ ID NO: 1)) over a wavelength range of 200-260 nm during a folded stage (first 25 °C data points), during heating (95 °C) and re-folding during cooling (25 °C ) (not all data points shown for clarity). FIG. 6B is a graph of a CD spectra showing the stability profile of an exemplary synthetic monovalent CD 16a binding domain (Reference Miniprotein 5 (SEQ ID NO: 5)) over a wavelength range of 200-260 nm during heating at 5°C intervals between temperatures of 25°C - 95°C (not all data points shown for clarity).

[0042] FIGs. 7A-7E show schematic representations and data related to structural and binding differences between CD 16a and CD 16b. FIG. 7A is one schematic representation showing structural differences in key binding areas between CD 16a and CD 16a, showing that exemplary synthetic CD 16a binding domains of the present disclosure bind to CD 16a, but do not bind to CD 16b. FIG. 7B are graphs depicting affinity measurement by surface plasmon resonance (SPR) of an exemplary synthetic CD 16a binding domain (Reference Miniprotein 5; SEQ ID NO: 5) for two variants of CD16a (176V (left graph of FIG. 7B) and 176F (right graph of FIG. 7B). FIG. 7C are graphs depicting affinity measurement by surface plasmon resonance (SPR) of an exemplary synthetic CD 16a binding domain (Reference Miniprotein 5; SEQ ID NO: 5) for two alleles of CD16b (NA1 (left graph of FIG. 7C) and NA2 (right graph of FIG. 7C). FIG. 7D is a graph that shows binding affinity of an exemplary synthetic CD16a miniproteins Reference Miniprotein 1 (SEQ ID NO: 1) to CD 16a. FIG. 7E are graphs that show specific, measurable binding of Reference Miniprotein 1 (SEQ ID NO: 1) to the CD16a ectodomain protein (left graph of FIG. 7E) and not to either CD 16b (right graph of FIG. 7E) or the streptavidin protein chip surface used for testing (data not shown).

[0043] FIG. 8A is a graph showing results from a binding assay using an exemplary synthetic CD16a binding domain (Reference Miniprotein 5; SEQ ID NO: 5) immobilized on an SPR chip and determining binding (y-axis) of CD 16a in the presence and absence of human serum over time (x-axis). FIG. 8B is a schematic representation of the results of FIG. 8A, showing that the binding location of the exemplary synthetic CD 16a miniprotein (at the bottom of the modeled image) does not bind to the same epitope as the Fc domain (at the top of the modeled image) and thus should not impact binding to the Fc domain of a human antibody. It is believed that the CD16a miniproteins provided herein should not compete with antibodies present in human serum through Fc regions. RU = response units.

[0044] FIG. 9 shows results from a cell binding assay using primary natural killer (NK) cells. An exemplary CD16a binding domain (Reference Miniprotein 5; SEQ ID NO: 5) was flag-tagged and cells were treated with the binding domain or with no binding domain and then stained with a detectable (e.g, fluorescent) marker that recognized the flag tag. The y-axis shows percent cell count and the x-axis shows the intensity of the fluorescent label. The peak to the left shows reference binding, when no synthetic CD 16a binding domain is present. The peak to the right shows that the synthetic CD 16a flag-tagged miniprotein bound to primary NK cells, which express CD 16a on their surfaces.

[0045] FIG. 10 is a graph of circular dichroism (CD) spectra showing the stability profile of an exemplary synthetic monovalent LILRB4 binding domain (Reference Miniprotein 47 ; SEQ ID NO: 47) over a wavelength range of 205-260 nm during heating and cooling measured at 5°C intervals between temperatures of 25°C - 95°C (not all data points shown, for clarity).

[0046] FIGS. 11A-11D are graphs depicting results from in vitro binding assays of an exemplary synthetic LILRB4 binding domain to two exemplary cell lines expressing LILRB4 (OCI-AML3 cells or MV-4-11 cells). Each graph shows cell counts (y-axis) plotted against antibody or synthetic miniprotein binding measured as mean fluorescent intensity (“MFI”; x-axis). Increased MFI as compared to reference controls indicates binding of the binding domain to cells. FIG. 11A shows unstained OCI-AML3 cells (control) and OCI-AML3 cells incubated with an anti-LILRB4 antibody, with the right-side peak indicating antibody binding to LILRB4 on OCI-AML3 cells. FIG. 11B shows OCI- AML3 cells incubated with an anti-flag tag antibody (control) and OCI-AML3 cells incubated with a flag-tagged LILRB4 binding domain followed by an anti-flag antibody. The right-side peak indicates that the synthetic binding domain has bound to LILRB4 on OCI-AML3 cells. FIG. 11C shows unstained MV-4-11 cells (control) and MV-4-11 cells incubated with an anti-LILRB4 antibody, with the right-side peak indicating antibody binding to LILRB4 on MV -4-11 cells. FIG. 11D shows MV-4-11 cells incubated with an anti-flag tag antibody (control) and MV-4-11 cells incubated with a flag-tagged LILRB4 binding domain and anti-flag tag antibody, with the right-side peak indicating synthetic binding domain has bound to LILRB4 on MV-4-11 cells.

[0047] FIG. 12 is a graph of circular dichroism (CD) spectra showing the stability profile of an exemplary synthetic bivalent CD16a-pl6-LILRB4 fusion binding domain (Reference Miniprotein 53 (SEQ ID NO: 53)) over a wavelength range of 205-260 nm during heating and measured at 5°C intervals between temperatures of 25°C - 95°C (not all data points shown for clarity).

[0048] FIGs. 13A and 13B are graphs depicting binding of an exemplary synthetic trivalent CD16a-pl6-LILRB4-pl6-CD16a fusion binding domain (Reference Miniprotein 58 (SEQ ID NO: 58)) to soluble LILRB4 (FIG. 13A) and CD 16a (FIG. 13B), measured by SPR. RU = response units.

[0049] FIGs. 14A and 14B show results from a comparison of cytotoxicity of NK cells on tumor cells (FIG. 14A) and NK cell activation(FIG. 14B) by bivalent and trivalent constructs as demonstrated by percentages of dead tumor cells (FIG. 14A) and of NK cell phenotype (FIG. 14B), each after co-culture of primary NK cells, OCI-AML3 tumor cells, and exemplary bivalent or trivalent synthetic binding domain constructs. FIG. 14A is a graph showing the percentage of dead tumor cells (y-axis; 0CI-AML3 tumor cells) after co-culture with healthy donor NK cells in presence of one of three exemplary fusion proteins: (i) a trivalent CD16a-pl6-LILRB4-pl6-CD16a binding domain (Reference Fusion Protein 58 (SEQ ID NO: 58)); (ii) a bivalent LILRB4-pl6-CD16a binding domain (Reference Fusion Protein 54 (SEQ ID NO: 54)); and (iii) a bivalent CD16a-pl6-LILRB4 (Reference Fusion Protein 53 (SEQ ID NO: 53)) binding domain at increasing concentrations (x-axis, nM). FIG. 14B is a graph showing the percentage of CD69/CD25- positive primary NK cells (y-axis) after co-culture of healthy donor NK cells with tumor cells (0CI-AML3) in presence of one of three exemplary fusion proteins: (i) a trivalent CD 16a-pl6-LILRB4-pl6-CD 16a binding domain (Reference Fusion Protein 58 (SEQ ID NO: 58)); (ii) a bivalent LILRB4-pl6-CD16a binding domain (Reference Fusion Protein 54 (SEQ ID NO: 54)); and (iii) a bivalent CD16a-pl6-LILRB4 binding domain (Reference Fusion Protein 53 (SEQ ID NO: 53)) at increasing concentrations (x-axis, nM).

[0050] FIGs. 15A and 15B show results from comparisons of three different trivalent orientations and impact on cytotoxicity (FIG. 15A) and NK activation (FIG. 15B) as measured by percentages of dead tumor cells and of NK cell phenotype after co-culture of primary NK cells, MV-4-11 tumor cells, and exemplary trivalent synthetic binding domain constructs. FIG. 15A is a graph showing the percentage of dead tumor cells (y-axis; MV- 4-11 tumor cells) after co-culture with healthy NK cells in presence of one of three exemplary trivalent fusion protein arrangements: (i) CD16a-pl6-LILRB4-pl6-CD16a (Reference Fusion Protein 58 (SEQ ID NO: 58)); (ii) LILRB4-pl6-CD16a-gl2-CD16a (Reference Fusion Protein 62 (SEQ ID NO: 62)); and (iii) CD16a-gl2-CD16a-pl6- LILRB4 (Reference Fusion Protein 61 (SEQ ID NO: 61)) at increasing concentrations (x- axis, nM). All trivalent constructs with two copies of a synthetic CD 16a binding domain and one copy of a synthetic LILRB4 binding domain are able to cause tumor cell death, but constructs with LILRB4 binding domains flanked by CD 16a binding domains such as in (i) show greater percentage of cell death than constructs with LILRB4 binding domains on a terminus of a trivalent construct such as in (ii) or (iii). FIG. 15B is a graph showing the percent of CD69/CD25 -positive primary NK cells (y-axis) after co-culture of healthy donor NK cells with tumor cells (MV-4-11) in presence of one of three exemplary trivalent fusion protein arrangements: (i) CD16a-pl6-LILRB4-pl6-CD16a (Reference Fusion Protein 58 (SEQ ID NO: 58)); (ii) LILRB4-pl6-CD16a-gl2-CD16a (Reference Fusion Protein 62 (SEQ ID NO: 62)); and (iii) CD16a-gl2-CD16a-pl6-LILRB4 (Reference Fusion Protein 61 (SEQ ID NO: 61)), each at increasing concentrations (x-axis, nM)

[0051] FIGs. 16A and 16B show results from experiments comparing different intermolecular linker lengths between binding domains to determine optimal trivalent construct orientation on cytotoxicity (FIG. 16A) and NK cell activation (FIG. 16B) as demonstrated by graphs showing the percent dead tumor cells (y-axis; 16A) and CD69/CD25 -positive primary NK cells (y-axis; 16B) after co-culture of tumor cells and primary NK cells (from a healthy donor) in the presence of one of six exemplary trivalent fusion protein arrangements each having different linker lengths compared to the others and at different concentrations (x-axis; nM): (i) CD16a-p5-LILRB4-p5-CD16a (Reference Fusion Protein 55 (SEQ ID NO: 55)); (ii) CD16a-p8-LILRB4-p8-CD16a (Reference Fusion Protein 56 (SEQ ID NO: 56)); (iii) CD16a-pl2-LILRB4-pl2-CD16a (Reference Fusion Protein 57 (SEQ ID NO: 57)); (iv)CD16a-p!6-LILRB4-pl6-CD16a (Reference Fusion Protein 58 (SEQ ID NO: 58)); (v)CD16a-p24-LILRB4-p24-CD16a (Reference Fusion Protein 59 (SEQ ID NO: 59)); (vi) CD16a-p32-LILRB4-p32-CD16a (Reference Fusion Protein 60 (SEQ ID NO: 60)).

[0052] FIGs. 17A-17C show results from experiments measuring cytotoxic activity of NK cells from an NK donor (NK donor 1) in two LILRB4-expressing AML cell lines versus a control cell line as demonstrated in graphs showing the percent of dead tumor cells (y-axis) in two LILRB4-expressing AML cell lines (FIG. 17A: OCI-AML3; and FIG. 17B: MV-4- 11 (AML)) and an LILRB4-negative cell line control (FIG. 17C: Raji (BL)) after coculture with primary NK cells from a healthy donor (NK donor 1) in presence of one of three exemplary trivalent fusion protein arrangements: (i) non-binding domain-pl 6- LILRB4-pl6-non-binding domain (LILRB4 as set forth in SEQ ID NO: 47/Reference Miniprotein 47; p!6 as set forth in SEQ ID NO: 67); (ii) CD16a-pl6-non-binding domain- p!6-CD16a binding domain (CD16a as set forth in SEQ ID NO: 5/Reference Miniprotein 5; p!6 as set forth in SEQ ID NO: 67); and (iii) CD16a-pl6-LILRB4-pl6-CD16a (as set forth in Reference Miniprotein 58 (SEQ ID NO: 58)), each at increasing concentrations (x- axis, nM). In constructs with a non-binding domain, the non-binding domain is an inert control with similar size and structure as CD16a or LILRB4 binding domains but does not bind to either target. [0053] FIGs. 18A-18C show results from experiments measuring cytotoxic activity of NK cells from an NK donor (NK donor 2) in two LILRB4-expressing AML cell lines versus a control cell line as demonstrated in graphs showing the percent of dead tumor cells (y-axis) in two LILRB-4 expressing cell lines (FIG. 18A: OCI-AML3; and FIG. 18B: MV-4-11 (AML)) and an LILRB4-negative control (FIG. 18C: Raji (BL)) after co-culture with primary NK cells from a healthy donor (NK donor 2) in presence of one of three exemplary trivalent fusion protein arrangements: (i) non-binding domain-pl 6-LILRB4-pl6-non- binding domain (LILRB4 as set forth in SEQ ID NO: 47/Reference Miniprotein 47; pl6 as set forth in SEQ ID NO: 67); (ii) CD16a-pl6-non-binding domain-pl6-CD16a (CD16a as set forth in SEQ ID NO: 5/Reference Miniprotein 5; p!6 as set forth in SEQ ID NO: 67); and (iii) CD16a-pl6-LILRB4-pl6-CD16a (as shown in Reference Fusion Protein 58, SEQ ID NO: 58), each at increasing concentrations (x-axis, nM). In constructs with a nonbinding domain, the non-binding domain is an inert control with similar size and structure as CD 16a or LILRB4 binding domains but does not bind to either target.

[0054] FIGs. 19A-19C show results from experiments measuring NK cell activation from an NK donor (NK donor 1) in two LILRB4-expressing AML cell lines versus a control cell line as demonstrated in graphs showing the percent of CD69/CD25-positive primary NK cells (y-axis) in two LILRB-4 expressing cell lines (FIG. 19A: OCI-AML3; and FIG. 19B: MV-4-11 (AML)) and an LILRB4-negative control (FIG. 19C: Raji (BL)) after co-culture with primary NK cells from a healthy donor (NK donor 1) in presence of one of three exemplary trivalent fusion protein arrangements: (i) non-binding domain-pl 6-LILRB4- p!6-non-binding domain (LILRB4 as set forth in SEQ ID NO: 47/Reference Miniprotein 47; p!6 as set forth in SEQ ID NO: 67); (ii) CD16a-pl6-non-binding domain-pl6-CD16a (CD16a as set forth in SEQ ID NO: 5/Reference Miniprotein 5; p!6 as set forth in SEQ ID NO: 67); and (iii) CD16a-pl6-LILRB4-pl6-CD16a (as set forth in Reference Fusion Protein 58 ,SEQ ID NO: 58), each at increasing concentrations (x-axis, nM). In constructs with a non-binding domain, the non-binding domain is an inert control with similar size and structure as CD 16a or LILRB4 binding domains but does not bind to either target.

[0055] FIGs. 20A-20C show results from experiments measuring NK cell activation from an NK donor (NK donor 2) in two LILRB4-expressing AML cell lines versus a control cell line as demonstrated in graphs showing the percent of CD69/CD25-positive primary NK cells (y-axis) in two LILRB-4 expressing cell lines (FIG. 20A: OCI-AML3; and FIG. 20B: MV-4-11 (AML)) and an LILRB4-negative control (FIG. 20C: Raji (BL)) after co-culture with primary NK cells from a healthy donor (NK donor 2) in presence of one of three exemplary fusion proteins: (i) a tri valent non-binding domain-pl 6-LILRB4-pl6-non- binding domain (LILRB4 as set forth in SEQ ID NO: 47/Reference Miniprotein 47; pl6 as set forth in SEQ ID NO: 67); (ii) a trivalent CD16a-pl6-non-binding domain-pl6-CD16a binding domain (CD16a as set forth in SEQ ID NO: 5/Reference Miniprotein 5; p!6 as set forth in SEQ ID NO: 67); and (iii) a trivalent CD16a-pl6-LILRB4-pl6-CD16a binding domain (as set forth in Reference Fusion Protein 58 (SEQ ID NO: 58)), each at increasing concentrations (x-axis, nM). In constructs with a non-binding domain, the non-binding domain is an inert control with similar size and structure as CD 16a or LILRB4 binding domains but does not bind to either target.

[0056] FIGs. 21A-21C are graphs showing the concentration of IFNy (y-axis; pg/mL) secreted by activated NK cells. Two LILRB-4 expressing cell lines (FIG. 21A: OCI- AML3; and FIG. 21B: MV-4-11 (AML)) and an LILRB4-negative control (FIG. 21C: Raji (BL)) were co-cultured with healthy donor primary NK cells in presence of one of three exemplary fusion proteins: (i) a trivalent non-binding domain-pl 6-LILRB4-pl6-non- binding domain(LILRB4 as set forth in SEQ ID NO: 47/Reference Miniprotein 47; p!6 as set forth in SEQ ID NO: 67); (ii) a trivalent CD16a-pl6-non-binding domain-pl6-CD16a binding domain (CD16a as set forth in SEQ ID NO: 5/Reference Miniprotein 5; p!6 as set forth in SEQ ID NO: 67); and (iii) a trivalent CD16a-pl6-LILRB4-pl6-CD16a binding domain (as set forth in Reference Fusion Protein 58 (SEQ ID NO: 58)), each at increasing concentrations (x-axis, nM). In constructs with a non-binding domain, the non-binding domain is an inert control with similar size and structure as CD 16a or LILRB4 binding domains but does not bind to either target.

[0057] FIGs. 22A-22B shows an experimental outline and graph depicting results from the experiment. FIG. 22A shows the experimental outline. In this model, tumor cells were grafted at day zero, NK cells introduced at day 3, and, in conditions (ii) and (iv) fusion protein treatment provided at days 4-8 and 10-14, with measurements taken on days 4, 7, 10, 14, and 17. FIG. 22B is a graph showing tumor cell spread/progression over two weeks (measured via bioluminescence using total flux (p/s)) in a mouse xenograft model of disseminated AML tumor (MV-4-11 cells). Animals were treated with: (i) nothing; (ii) an exemplary fusion protein (CD16a-pl6-LILRB4-pl6-CD16a; Reference Fusion Protein 58 (SEQ ID NO: 58); (iii) healthy donor-derived NK cells only; or (iv) a combination of healthy donor-derived NK cells and an exemplary fusion protein (CD16a-pl6-LILRB4- pl6-CD16a; Reference Fusion Protein 58 (SEQ ID NO: 58).

[0058] FIGs. 23A-23D are graphs showing results from in vitro serial killing cytotoxicity assays (round 1 - FIG. 23A, round 2- FIG. 23B, round 3- FIG. 23C, and round 4- FIG. 23D) measuring tumor cell proliferation (y-axis) normalized at time 0 and plotted over time (x-axis) to evaluate serial killing activity of NK cells in the presence of a trivalent fusion miniprotein (10 nM of a trivalent CD 16a-pl6-LILRB4-pl6-CD 16a binding domain, SEQ ID NO: 58) (denoted as “COC” in the legend for FIGs. 23A-23D) or two controls (tumor cells alone (denoted as “tumor only” in legend for FIGs. 23A-23D) and tumor cells with NK cells without any miniprotein (denoted as “no protein” in legend for FIGs. 23A-23D).

DETAILED DESCRIPTION

[0059] The disclosure is based, in part, upon the development of fusion proteins comprising of or containing at least one synthetic CD 16a binding domain that specifically and preferentially binds CD 16a over CD 16b and modulates NK cell activity and one or more other protein binding domains that bind to proteins expressed on non-NK target cells (e.g, tumor cells, e.g., LILRB4-expressing tumor cells), thus “engaging” NK cells to attack the target cells. The disclosure also provides newly developed synthetic LILRB4 binding domains and CD 16a binding domains. The CD 16a binding domains, LILRB4 binding domains, fusions of CD 16a and LILRB4 binding domains, and compositions containing such binding domains, and methods of making and using such binding domains, can be used for NK cell activation and/or CD16a-mediated cytotoxicity of LILRB4-expressing tumor cells.

[0060] As demonstrated herein, CD 16a expression can be leveraged to develop molecules to bring NK cells into proximity with certain other cells (e.g, cancer cells that express LILRB4). LILRB4 (also known as ILT3) is an immune checkpoint molecule expressed in myeloid antigen presenting cells (APCs), such as monocytes and dendritic cells. Elevated levels of LILRB4 correlate with poor prognosis of various cancer types (e.g., acute myeloid leukemia (AML), chronic myelomonocytic leukemia (CMML), and non-small-cell lung cancer (NSCLC)) making it a potential target for cancer immunotherapy. See Yang et al. (2022) BLOOD Set. 4: 49. Other pathologies in which LILRB4 has been implicated include B cell-related malignancies, like systemic lupus erythematosus (SLE), and allergies. See Boonpiyathad et al. (2019) ALLERGY 74: 976, and Inui et al. (2016) INT. IMM. 28: 597.

High expression of LILRB4 in APCs is thought to suppress immune response and lead to immune tolerance.

[0061] Disclosed herein are NK cell engagers that can simultaneously bind to CD16a receptors expressed on NK cells and the tumor-associated antigen LILRB4 expressed on certain cancer cells, whereupon the dual binding can trigger NK-cell induced tumor-cell specific cytotoxicity. Depending upon the circumstances, the NK cell engagers can trigger ADCC in tumor cells brought into close proximity of NK cells, where each cell is bound by a single NK engager molecule with binding domains for a surface antigen on each cell. The NK engagers can have multiple valencies, e.g., bi- and tri-specific killer cell engagers (BiKEs and TRiKEs, respectively) that bind to NK cells through CD 16a and one or two other target antigens, e.g, LILRB4 and another tumor-associated antigen. It is contemplated that the conjugates can be multivalent and contain 3, 4, 5, 6, or 7 separate miniprotein binders.

[0062] Accordingly, the present disclosure provides, among other things, synthetic CD 16a binding domains, synthetic LILRB4 binding domains, and multivalent fusions thereof, as well as methods of making such binding domains, and methods of using such proteins to treat related disorders.

I. DEFINITIONS

[0063] Unless defined otherwise, technical and scientific terms used herein have the same meaning as is commonly understood by one of skill in the art to which the claimed subject matter belongs. For example, nomenclatures utilized in connection with, and techniques of, e.g, polypeptide and polynucleotide chemistry and synthesis, molecular and cellular biology, protein biology and biochemistry, immunology, etc. as described herein are those well-known and commonly used in the art.

[0064] As used herein, the singular forms “a,” “an” and “the” include plural referents unless context clearly dictates otherwise. Thus, for example, in some embodiments, reference to, e.g, a synthetic binding domain (e.g, CD16a binding domain, LILRB4 binding domain, fusion proteins thereof, etc.) includes a single binding domain, a plurality of synthetic binding domains, etc. [0065] As used herein, the expression “and/or” in connection with two or more recited objects includes individually each of the recited objects and the various combinations of two or more of the recited objects, unless otherwise understood from the context and use.

[0066] Where the use of the term “about” is before a quantitative value, the present disclosure also includes the specific quantitative value itself, unless specifically stated otherwise. As used herein, the term “about” refers to a ±10% variation from the nominal value unless otherwise indicated or inferred. Ranges can be expressed in this disclosure as from “about” one particular value, and/or to “about” another particular value. When values are expressed as approximations by use of the antecedent “about,” it is understood that the disclosure also contemplates embodiments that specify the particular values and ranges of values without the approximations.

[0067] As used herein, the phrases “solvent accessible residue” and “solvent accessible amino acid” refer to an amino acid that, when disposed in a folded molecule (e.g, in its a tertiary conformation) and in a solvent, is characterized in that the amino acid is at least partially accessible or exposed to the solvent. Solvent accessible amino acids can be determined using a variety of approaches including, e.g, Rosetta software suite, Neighbor Count, and Neighbor vector algorithms (Durham et. al. (2009) J. MOL. MODEL. 15(9): 1093-108).

[0068] As used herein, the phrase “conservative substitution” refers to a substitution with a structurally and/or functionally similar amino acid. The following six groups each contain amino acids that are conservative substitutions for one another: 1) Serine (S) and Threonine (T); 2) Aspartic Acid (D) and Glutamic Acid (E); 3) Asparagine (N) and Glutamine (Q); 4) Arginine (R) and Lysine (K); 5) Isoleucine (I), Leucine (L), Methionine (M), and Valine (V), and 6) Phenylalanine (F), Tyrosine (Y), and Tryptophan (W). Conservative substitutions may also be defined by the BLAST (Basic Local Alignment Search Tool) algorithm, the BLOSUM substitution matrix (e.g, BLOSUM 62 matrix), or the PAM substitution: p matrix (e.g, the PAM 250 matrix). In certain embodiments, a binding domain of the disclosure comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 conservative substitutions relative to a reference amino acid sequence.

[0069] As used herein, the phrase “corresponding to” designates a position/identity of an amino acid or a nucleic acid in a polymeric molecule such as an amino acid in an amino acid sequence or a nucleic acid in a nucleic acid sequence. It is understood by the skilled artisan that such amino acids or nucleic acids in such a polymer are often designated using a canonical numbering system based on a reference related polymer, so that, for example, an amino acid in a first polymer “corresponding to” position seven in the reference amino acid, for example, need not actually be the seventh amino acid in the first polymer. Those of ordinary skill in the art are aware of methodology to identify “corresponding” amino acids or nucleic acids between two molecules (e.g., a polymer and a reference polymer), including, such as, commercially available algorithms, databases, or other information given context regarding particular polymers.

[0070] As used herein, the phrase “effective amount” refers to the amount of an active agent (e.g., a CD 16a binding domain or LILRB4 binding domain or fusion protein thereof as provided herein) sufficient to effect beneficial or desired results. An effective amount can be administered in one or more administrations, applications, or dosages and is not intended to be limited to a particular formulation or administration route.

[0071] As used herein, the term “synthetic” refers to a molecule that is (i) not naturally occurring, (ii) not present in nature, (iii) does not comprise entirely natural components, or (iv) a combination of any one of (i), (ii) and (iii). For example, a synthetic peptide does not exist naturally, is produced or otherwise modified by human intervention, such as techniques including recombinant or cell-free synthesis, and/or the peptide may comprise one or more non-naturally occurring amino acids.

[0072] As used herein, the terms/phrases “synthetic binding domain,” “synthetic miniprotein,” “binding domain” and “miniprotein” are used interchangeably, and refer to a polypeptide between about 35 to about 100 amino acids in length, e.g, from about 30 to about 90 amino acids, from about 30 to about 80 amino acids in length, from about 30 to about 70 amino acids in length, from about 35 to about 65 amino acids, from about 35 to about 60 amino acids in length, from about 40 to about 70 amino acids, from about 40 to about 65 amino acids, from about 40 to about 60 amino acids, from about 40 to about 55 amino acids in length, from about 40 to about 50 amino acids in length, or 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, or 55 amino acids in length, that are capable of binding to a given target, e.g, CD16a, e.g, LILRB4 with a desired binding affinity (e.g, stronger than 1 pM). A miniprotein or binding domain, as the terms are used herein, include one or more structural features (e.g, amino acid, primary structure or secondary structure features) and/or one or more functional features (e.g, binding properties). Binding domains may be connected to one another in fusion proteins according to various formulas such as D1-D2 or D1-D2-D3, where DI, D2, and/or D3 may be the same or different. Each binding domain is made up of a series of structural domains, e.g., alpha helix, beta sheet, and/or loop structures where R represents an alpha helix or a beta sheet (e.g., Rl, R2, R3, R4) and L represents a loop linking R1 to R2, R2 to R3, and R3 to R4 (R1-L1-R2-L2-R3-L3-R4). In the context of secondary structure, a “structural domain” can be an uninterrupted linear sequence that adopts a single type of secondary structure, for example, ten continuous amino acid residues that are all part of the same alpha helix structure or beta sheet.

[0073] As used herein, the term “linker” refers to a structure (e.g, a polypeptide linker), or a chemical crosslinker (e.g., a homobifunctional or a heterobifunctional cross linking agent) between two molecules (e.g, two synthetic CD 16a binding domains disclosed herein) or between, e.g, (i) a synthetic CD16a binding domain or a synthetic LILRB4 binding domain; and (ii) an effector, wherein each of the entities that are linked is covalently linked to one another.

[0074] As used herein, the term “loop” refers to (i) a structure (e.g, polypeptide) that connects two structural domains (e.g, a loop may be disposed between two alpha helices, between an alpha helix and a beta sheet, or between two beta sheets in a given synthetic CD16a binding domain) and/or (ii) a structure (e.g, peptide) present at the N- and/or C- terminal end of a given monovalent synthetic binding domain.

[0075] As used herein, a “non-binding domain” is a control protein and considered “inert” with respect to particular synthetic miniproteins in that it has the same size and structure (e.g, shape, e.g, folding, e.g, secondary/tertiary structure but not amino acid sequence) as a synthetic miniprotein, but does not bind to a target. For example, a non-binding domain can have the same size, shape, and folding as a synthetic CD 16a binding domain, but it does not bind to CD 16a, and, thus, can be used as a control such as in testing binding portions of multivalent constructs.

[0076] As used herein, the term “epitope” refers to a region of a protein that is specifically recognized by a binding partner, such as an antibody or another binding protein/miniprotein or binding domain. The epitope may generally span a portion of the protein. Often, proteins may have multiple such regions where binding partners can attach. Epitopes typically fall into two classes: continuous epitopes (also known as linear epitopes), which are epitopes defined by linear sequences of consecutive amino acids, and discontinuous epitopes (also known as conformational epitopes), which are epitopes defined by discontinuous amino acids that are brought together into spatial proximity when a protein is in its folded state.

[0077] As used herein, the term "paratope" refers to the specific region of a binding molecule (e.g, a binding protein, e.g., a CD16a and/or LILRB4 binding domain) that recognizes and binds an epitope of a target molecule (e.g, CD16a, e.g, LILRB4). A paratope of a given binding molecule typically comprises 5-20 amino acids that are solvent accessible and in close proximity in three-dimensional space.

[0078] As used herein, the phrase “percent identity” and “% identity” refers to the extent to which two sequences (e.g, a polypeptide) have the same amino acid or nucleotide at the same positions in an alignment. The percent identity between a polypeptide sequence and a reference sequence is defined as the percentage of amino acid residues in the polypeptide sequence that are identical to the amino acid residues in the reference sequence, after aligning the sequences and introducing gaps, if necessary, to achieve the maximum percent sequence identity. It is contemplated that a reference sequence can be an amino acid sequence corresponding to an entire CD 16a binding domain or a portion thereof. A reference sequence may be an amino acid sequence that corresponds to a particular domain or domains (e.g, an alpha helix, a loop region) or a combination of domains (e.g, a combination of a loop and an alpha helix). Alignment for purposes of determining percent sequence identity (e.g, amino acid sequence identity) can be achieved in various ways that are within the skill in the art, for instance, using publicly available computer software such as BLAST (Basic Local Alignment Search Tool), BLAST-2, ALIGN, MEGALIGN (DNASTAR), CLUSTALW, CLUSTAL OMEGA, MUSCLE, or BioPython software. For a discussion of basic issues in searching sequence databases, see Altschul et al. (1994) NATURE GENETICS 6: 119-129, which is incorporated by reference herein. Those skilled in the art can determine appropriate parameters for aligning sequences, including any algorithms needed to achieve maximal alignment over the full length of the sequences being compared.

[0079] As used herein, the term “effector” refers to a molecule or molecular entity that confers one or more particular characteristics on itself or another molecule or molecular entity with or to which it is associated. For example, an effector may include a synthetic binding protein (e.g, a miniprotein, or something other than a miniprotein that is associated with a synthetic CD 16a binding domain or a synthetic LILRB4 binding domain disclosed herein (e.g, via a covalent linkage), such as a detectable label (e.g, visualizable or otherwise measurable such as by fluorescence or radiolabel detection), small molecule, nanoparticle (e.g, a lipid nanoparticle, a polymer nanoparticle, etc.), polynucleotide (e.g, an aptamer, an siRNA, an shRNA, an oligonucleotide, etc.), a radionuclide, etc. An effector may be a synthetic binding protein (e.g, a monovalent synthetic binding protein linked to a CD16a binding domain or to an LILRB4 binding domain as disclosed herein to create a bivalent synthetic protein where one or both of the proteins causes a change, e.g, in a cellular function, e.g, in a disease state, etc.).

[0080] As used herein, the phrase “pharmaceutically acceptable” refers to those compounds, materials, compositions, and/or dosage forms which are, within the scope of sound medical judgment, suitable for use in contact with the tissues of human beings and animals without excessive toxicity, irritation, allergic response, or other problem or complication, commensurate with a reasonable benefit/risk ratio.

[0081] As used herein, the phrase “pharmaceutically acceptable carrier” as used herein refers to an agent (e.g, excipient, carrier, buffer, etc.) suitable for use in contact with the tissues of human beings and animals without excessive toxicity, irritation, allergic response, or other problem or complication, commensurate with a reasonable benefit/risk ratio. Pharmaceutically acceptable carriers include any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents, and the like, compatible with pharmaceutical administration. Standard pharmaceutical carriers may include, for example a phosphate buffered saline solution, water, emulsions (e.g, such as an oil/water or water/oil emulsions), and various types of wetting agents. The compositions also can include stabilizers and preservatives. For examples of carriers, stabilizers, and adjuvants, see e.g., Adeboye Adejare, REMINGTON: THE SCIENCE AND PRACTICE OF PHAR ACY (23^rd ed. 2020).

[0082] As used herein, the terms “subject” and “patient” refer to an organism to be treated by the methods and compositions described herein. Such organisms preferably include, but are not limited to, mammals (e.g, murines, simians, equines, bovines, porcines, canines, felines, and the like), and more preferably includes humans. [0083] As used herein, “treat”, “treating”, and “treatment” refer to the treatment of a disease, disorder, or symptom or manifestation of such in a subject, e.g., in a human. This includes: (a) preventing a disease or disorder, (b) inhibiting the disease, disorder, etc., i.e., slowing or arresting its progress or development; and (b) relieving the disease, disorder, etc., i.e., causing regression of the disease state. As used herein, “prevent”, “preventing” and “prevention” refer to causing a disease, disorder, or symptom or manifestation of such not to occur for at least a period of time in at least some subjects.

[0084] Throughout the description, where compositions are described as having, including, or comprising specific components, or where processes and methods are described as having, including, or comprising specific steps, it is contemplated that, additionally, there are compositions of the present disclosure that consist essentially of, or consist of, the recited components, and that there are processes and methods according to the present disclosure that consist essentially of, or consist of, the recited processing steps. Similarly, throughout the description, where compositions are described as consisting essentially of specific components, or where processes and methods are described as consisting essentially of specific steps, it is contemplated that, additionally, there are compositions of the present disclosure that consist of the recited components, and that there are processes and methods according to the present disclosure that consist of the recited processing steps.

[0085] Throughout the text, where an element or component is said to be included in and/or selected from a list of recited elements or components, it should be understood that the element or component can be any one of the recited elements or components, or the element or component can be selected from a group consisting of two or more of the recited elements or components.

[0086] Further, it should be understood that elements and/or features of a composition or a method described herein can be combined in a variety of ways without departing from the spirit and scope of the present disclosure, whether explicit or implicit herein. For example, where reference is made to a particular protein, that protein can be used in various embodiments of compositions of the present disclosure and/or in methods of the present disclosure, unless otherwise understood from the context. In other words, within this application, embodiments have been described and depicted in a way that enables a clear and concise application to be written and drawn, but it is intended and will be appreciated that embodiments may be variously combined or separated without parting from the present teachings and any invention provided herein. For example, it will be appreciated that all features described and depicted herein can be applicable to all aspects of any invention described and depicted herein.

[0087] The use of any and all examples, or exemplary language herein, for example, “such as” or “including,” is intended merely to illustrate better the present disclosure and does not pose a limitation on the scope of any invention unless claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of any invention disclosed herein.

[0088] It should be understood that the expression “at least one of’ includes individually each of the recited objects after the expression and the various combinations of two or more of the recited objects unless otherwise understood from the context and use.

[0089] The use of the term “include,” “includes,” “including,” “have,” “has,” “having,” “contain,” “contains,” or “containing,” including grammatical equivalents thereof, should be understood generally as open-ended and non-limiting, for example, not excluding additional unrecited elements or steps, unless otherwise specifically stated or understood from the context.

[0090] It should be understood that the order of steps or order for performing certain actions is immaterial so long as disclosed invention(s) remain operable. Moreover, two or more steps or actions may be conducted simultaneously.

[0091] As used herein, all numerical values or numerical ranges include whole integers within or encompassing such ranges and fractions of the values or the integers within or encompassing ranges unless the context clearly indicates otherwise. Thus, for example, reference to a range of 90-100%, includes 90%, 91%, 92%, 93%, 94%, 95%, 95%, 97%, etc.

II. CLUSTER OF DIFFERENTIATION 16a (CD 16a)

[0092] Provided herein are synthetic binding domains (also referred to as miniproteins), that bind the Low Affinity Immunoglobulin Gamma Fc Region Receptor III-A (FCGR3A) also referred to as cluster of differentiation 16a (CD 16a). Binding and crosslinking with another cell or receptor can induce NK cell activation and ADCC activity. CD 16a is a member of the Fc gamma receptor family that binds to IgG. CD 16a has two extracellular Ig-like domains, a transmembrane domain and a short C-terminal cytoplasmic tail. See, e.g., Coenon et al. (2022) supra. [0093] While Fc gamma receptors are broadly expressed throughout the body, CD16a is expressed on natural killer (NK) cells and macrophages and anchors to the membrane via a transmembrane domain. See, e.g., Coenon et al. (2021) supra. A highly homologous isoform, CD 16b, is expressed on neutrophils and is glycosylphosphatidylinositol (GPI) anchored to the neutrophil cell membrane. See, e.g., Coenon et al. (2022) supra, Li et al. (2016) supra,' and Ravetch et al. (1989) supra. In most individuals, CD16a is the only Fc gamma receptor expressed on NK cells and is important for antibody-dependent cell mediated cytotoxicity (ADCC). See, e.g., Patel et al. (2020) J BIO CHEM. 296: 100183.

[0094] Without being bound by theory, binding specificity of CD 16a over CD 16b appears to involve at least two key residues, 147 and 158, with G147 and Y158 in CD16a as compared to amino acids at corresponding residues in CD16b, D147 and H158 (see, e.g., FIG. 7A)

[0095] The interaction of CD 16a with IgG is thought to occur at least through the IgG upper CH2 and lower hinge region and is influenced by the glycan composition of the IgG Fc region. See, e.g., Coenon et al. (2022) supra. Additionally, CD16a can be glycosylated with high mannose and N-glycan type structures. Without being limited by theory, glycan profiles are thought to impact interactions between CD16a and binding partners. For example, glycosylation of N45 can stabilizes CD 16a and influence IgG binding. Coenon et al. (2021) supra. Fucosylation or lack thereof may also influence the signaling through IgG-CD 16a interactions. See, e.g., Gonzalez (2022) supra; Coenon et a/. (2021) supra.

[0096] CD 16a is known to lack immunoreceptor tyrosine-based activation motifs (IT AMs). Thus, CD 16a has been shown to cluster in lipid rafts and interact with other signaling molecules that contain ITAMs such as CD3zeta. Coenon et al. (2021) supra. This interaction is believed to lead to subsequent phosphorylation of kinases that trigger NK cell degranulation and calcium release into the cytosol, signaling ADCC of target cells. Id. In some contexts, ADCC by NK cells can be leveraged to target certain cells for destruction by combining CD 16a expressing cells with one or more other binding molecules, so that NK cells are engaged to induce a cytotoxic response on the target cell, such as, for example a tumor cell or a virally -infected cell. See, e.g, Gonzalez (2022) supra; Capuano (2021) supra.

[0097] As depicted schematically in FIG. 1, engagement of CD 16a with a synthetic CD 16a binding domain and an effector can activate NK cell pathways to target a target cell. When paired with another molecule to “engage” the NK cell to target another cell, e.g., a tumor cell expressing a tumor associated antigen, NK-cell mediated cytotoxicity (e.g., ADCC) of the tumor cell is induced.

[0098] Synthetic CD 16a binding domains provided herein are designed to selectively and tightly bind to CD 16a. This specificity, selectivity, and binding strength allows targeted binding to NK cells (which express CD16a) over other cell types (e.g, cells expressing CD 16b). Such binding characteristics will allow improvements in approaches targeting NK cells, including for use in targeting and treating cancer cells in subjects in need of such treatment.

III. LILRB4

[0099] Provided herein are synthetic binding domains that bind LILRB4 (also known as ILT3). LILRB4 is an immune checkpoint molecule expressed in myeloid antigen presenting cells (APCs), such as monocytes and dendritic cells. High expression of LILRB4 in APCs suppress immune response and lead to immune tolerance. Elevated levels of LILRB4 correlate with poor prognosis of various cancer types, including acute myeloid leukemia (AML), chronic myelomonocytic leukemia (CMML), and non-small-cell lung cancer (NSCLC).

[00100] LILRB4 has an extracellular segment (“ectodomain”) which consists of two immunoglobulin-like structural domains, DI and D2, each of which consists of antiparallel fragments. The D1-D2 outer domain of LILRB4 adopts a blunt interdomain angle of 107°, which is stabilized by hydrophobic interactions. In the D2 structural domain, LILRB4 appears to have two new 3(10) helix regions.

[00101] LILRB4’s intracellular segment is made up of three receptor tyrosine-based inhibitory groups (ITIMs). ITIM is commonly found on receptor molecules on the surface of some immune cells, and its primary role in signaling is to inhibit or negatively regulate immune responses. When the ITIM receptor binds to its ligand, the tyrosine in the ITIM region may be phosphorylated, triggering a series of signaling events that ultimately lead to an inhibitory response in immune cells. As depicted schematically in FIG. 2 engagement of LILRB4 with a synthetic LILRB4 binding domain can block inhibition of the immune response. IV. MULTIVALENT FUSIONS

[00102] Provided herein are fusion proteins containing multiple synthetic binding domains, capable of dual binding to the Low Affinity Immunoglobulin Gamma Fc Region Receptor III-A (FCGR3A) also referred to as cluster of differentiation 16a (CD 16a) and the tumor associated antigen LILRB4 (also known as ILT3). FIG. 3A depicts a schematic showing a fusion comprising a CD 16a binding domain linked to an LILRB4 binding domain and each binding on cells having surface-expressed CD 16a and LILRB4 proteins, respectively. FIGS. 3B and 3C show exemplary orientations of binding domains in exemplary bivalent molecules. FIG. 3B is a schematic representation of an exemplary bivalent (“D1-D2”) CD 16a- LILRB4 fusion protein with the C-terminus of an exemplary synthetic CD 16a binding domain joined to the N-terminus of an exemplary synthetic LILRB4 binding domain via an exemplary linker. FIG. 3C is a schematic representation of an exemplary bivalent (“D2-D1”) CD16a-LILRB4 fusion protein, with the CD 16a synthetic binding domain bound to a CD 16a protein on an NK cell and the synthetic LILRB4 binding domain bound to LILRB4 expressed on the surface of a tumor cell.

[00103] As depicted in FIGs. 4A-4E, a single synthetic CD 16a binding domain can be linked with linkers to two synthetic LILRB4 binding domains or two synthetic CD 16a binding domains can be linked with linkers to one LILRB4 binding domain to generate a trivalent molecule.

[00104] Immune cell engagers are molecules capable of redirecting specific immune cell types (e.g., T cells and NK cells) to act upon cells expressing tumor-associated antigens. See, e.g., Fuca et al. (2021) ESMO OPEN 6: 100046. Compared to T cells, the use of NK cells is contemplated to be advantageous because they may be less likely to trigger cytokine release syndrome or cause graft-versus-host reactions. See, e.g., Zhang et al. (2023) supra. Current NK cell engagers typically target NK cells through recombinant protein binding molecules e.g, bispecific antibodies or dual scFvs linked by a linker and are typically directed towards NK cells through recognition of the CD 16 receptor. See, e.g., Coenon et al. (2022) supra, Fuca et al. (2021) supra. The binding region for a target cell (e.g., a cancer cell, e.g, an LILRB4-expressing tumor cell) is then designed to bind a surface antigen on a cancer cell.

[00105] Various immune cell engager formats have been investigated previously. For example, bispecific killer cell engagers (BiKEs) are designed to bind to NK cells and one target antigen and trispecific killer cell engagers (TRiKEs) are designed to bind to CD 16 and target two tumor antigens. As depicted in FIG. 5 a synthetic LILRB4 binding domain when paired with synthetic CD 16a binding domains can “engage” an NK cell to target LILRB4 expressed on a cell surface (e.g., a tumor cells), thereby inducing NK-cell mediated cytotoxicity (e.g., ADCC) of the tumor cell while simultaneously blocking anti- immune response functions of the tumor cell.

[00106] Synthetic CD 16a binding domains linked to synthetic LILRB4 proteins provided herein are designed to selectively and tightly bind to CD 16a and not CD 16b to specifically target binding to NK cells and be highly selective for the LILRB4 tumor antigen reducing any off-target effects. Combining binding characteristics of multiple miniproteins will allow improvements in approaches treating cancer cells in subjects in need of such treatment.

[00107] In certain aspects, fusions of the present disclosure comprise at least two domains: a first CD 16a binding domain and a first LILRB4 binding domain. In certain embodiments, the fusion proteins disclosed herein further comprise a third domain that can be a second CD 16a binding domain (that can be the same as or different than the first CD 16a binding domain) or a second LILRB4 binding domain (that can be the same or different than the first LILRB4 binding domain). Each binding domain contains an N-terminal amino acid residue and a C-terminal amino acid residue. Linkers, such as those described herein in (e.g, see TABLE 4), may be used to link two or more domains (e.g, the first CD 16a binding domain and the first LILRB4 binding domain), generally by linking the C-terminal amino acid residue of one domain to the N-terminal amino acid residue of another domain.

A. CD 16a binding domains

[00108] CD 16a binding domains suitable for use in fusion proteins have improved manufacturability, increased specificity, and affinity as compared to existing proteins that bind to CD 16a. For example, other CD 16a antibodies and/or binding domains may suffer from off-target activities (e.g, binding to CD16b on non-NK cells). Similarly, large molecules such as biologies (e.g., monoclonal antibodies that bind to CD 16a) are expensive to produce, are challenging to produce in uniform batches of drug substance (including, especially, e.g, at commercially-scalable amounts), and can be challenging to formulate, transport, store, and administer to subjects. [00109] The synthetic CD 16a binding domains suitable for use in fusion proteins as provided herein avoid certain such disadvantages as they have strong binding affinity to CD 16a, can be engineered to have desired pharmacodynamic and pharmacokinetic properties (e.g., a desirable circulating half-life in plasma), reduced cross-reactivity (e.g, do not bind to CD 16b), are chemically and thermally stable, and are resistant to protease degradation (e.g, via L-X-R-R sites, wherein X represents any amino acid residue), deamination and lack post- translational modification (e.g, N-linked glycosylation, e.g, glycosylation through N-X-S/T, wherein X represents any amino acid residue), and are stable in different redox environments.

[00110] A CD16a binding domain provided herein can have mM-level solubility. In some embodiments, the CD16a binding domain can have a solubility greater than 0.01 mg/mL, 0.05 mg/mL, 0.1 mg/mL, 0.5 mg/mL, 1 mg/mL, 5 mg/mL, 10 mg/mL, 15 mg/mL, 20 mg/mL, 25 mg/mL, 30 mg/mL, 35 mg/mL, 40 mg/mL, 45 mg/mL, 50 mg/mL, 55 mg/mL, 60 mg/mL, 65 mg/mL, 70 mg/mL, 75 mg/mL, or 80 mg/mL in an aqueous solution.

[00111] CD16a binding domains suitable for use in fusion proteins disclosed herein are designed to specifically bind CD 16a, and preferably bind CD 16a over CD 16b. Preferably, the binding domains do not bind CD 16b. Depending upon the circumstances, the CD 16a binding domains bind CD 16a and stabilize the CD 16a ectodomain, resulting in downstream NK cell activation. CD 16a binding domains as provided herein can, in some embodiments, bind with an affinity at least at a level of a reference CD 16a binding domains.

[00112] It is contemplated that optimization of synthetic binding domains may be achieved using optimized designs, such as, for example, modifications of one or more amino acids by substitution at one or more positions with a different amino acid. Optimization, such as by amino acid modifications, allows tunability of certain characteristics such as changes to (e.g, increases in) binding affinity and/or avidity.

[00113] Synthetic CD 16a binding domains can be optimized by affinity maturation techniques. For example, affinity maturation may be used on a sequence of a binding protein to create another synthetic CD 16a binding protein with at least the same or better selectivity and/or affinity for CD16a as compared to the starting sequence. Affinity maturation can be accomplished using techniques known to those of ordinary skill in the art, including, for example, generating libraries using error prone PCR, degenerate codons, synthetic oligonucleotide pools, or a combination thereof. These libraries can then be transformed into yeast and improved variants may be isolated by methods such as magnetic, flow cytometric, and/or FACS-based approaches. Computational design/redesign strategies may also be used when affinity maturing proteins and computer programs for implementing such approaches are known in the art. Prior to affinity maturation, synthetic binding proteins may be characterized to determine functional and structural features, such as binding affinity (e.g., for CD 16a) and conformation.

[00114] CD 16a binding domains provided herein are engineered to have certain characteristics (e.g, binding affinity/avidity, binding specificity, e.g, for a target, e.g, for CD 16a). Various in silico, in vitro, and in vivo characterization assays may be used to evaluate these CD16a binding domains. For example, binding assays can be used to determine binding specificity to a target, e.g, CD 16a as compared to binding to another molecule such as, e.g., another receptor or a ligand such as CD16b. Other assays can be used to determine binding affinity of a binding protein, e.g, a CD 16a binding protein for its target, which can include for example, surface plasmon resonance (SPR), and flow cytometry.

[00115] The synthetic CD 16a binding domains of the present disclosure are designed to have certain stability characteristics. For example, a binding domain is stable in that it may retain at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 85%, 90%, 91%, 92%, 9%, 94%, 95%, 96%, 97%, 98%, 99% or substantially all of its binding affinity to CD16a upon cooling to room temperature after thermal denaturation at 95°C in a solution (e.g, phosphate buffered saline (PBS)) for at least about 5, about 10, about 15, about 20, about 25, about 30, about 35, about 40, about 45, about 50, about 55, about 60 or more minutes relative to the synthetic CD 16a binding domain prior to thermal denaturation.

[00116] In some embodiments, a synthetic CD 16a binding domain is stable in that it may retain at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 85%, 90%, 91%, 92%, 9%, 94%, 95%, 96%, 97%, 98%, 99% or substantially all of its binding affinity to CD16a after incubation at 37°C (e.g, for 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 or more hours) relative to the synthetic CD 16a binding domain prior to incubation.

[00117] Synthetic CD 16a binding domains provided herein may also display stability in resistance to chemical denaturation and/or retention of stability after exposure to chemical denaturants. For example, a synthetic CD16a binding domain may be stable in that it may retain at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 85%, 90%, 91%, 92%, 9%, 94%, 95%, 96%, 97%, 98%, 99% or substantially all of its binding affinity to CD16a in PBS following exposure to a denaturing chemical (e.g., 4M urea) at room temperature for about 15 minutes, about 30 minutes, about 45 minutes, about 1 hour, about 2 hours, about 3 hours, about 4 hours, about 5 hours, about 6 hours, about 7 hours, about 8 or more hours relative to the binding affinity of the synthetic CD 16a binding domain prior to exposure to the chemical denaturant.

[00118] Synthetic CD 16a binding domains engineered, developed, and produced herein are selected and/or specific for CD16a. That is, in some embodiments, a synthetic CD16a binding domain does not bind to a non-CD16a target (e.g, CD16b). In some embodiments, a synthetic CD 16a binding domain binds to another receptor, but binds to CD 16a with a much greater affinity. For example, the binding affinity of a synthetic CD 16a binding protein may be between 1 and 200-fold greater than that for any other binding partner (e.g, CD 16b).

[00119] Affinity of a synthetic CD16a binding domain may be modified and may vary depending on modifications made to, for example, its primary sequence. A binding affinity may be at least 1-fold greater, at least 2-fold greater, at least 3-fold greater, at least 4-fold greater, at least 5-fold greater, at least 6-fold greater, at least 7-fold greater, at least 8-fold greater, at least 9-fold greater, 10-fold greater, at least 20-fold greater, at least 30-fold greater, at least 40-fold greater, at least 50-fold greater, at least 60-fold greater, at least 70- fold greater, at least 80-fold greater, at least 90-fold greater, at least 100-fold greater, at least 125-fold greater, at least 150-fold greater, at least 175-fold greater, at least 200-fold greater than the affinity of the synthetic CD 16a binding domain for an unrelated (e.g, different) target (e.g., CD 16b).

[00120] The CD 16a binding domains disclosed herein are suitable for use in fusion proteins (e.g, such as with a binding domain that binds to LILRB4). For the example, the CD16a binding domains described herein can be bound to one or more other synthetic binding domains, which can be the same or different. The resulting proteins can be multivalent (e.g., bivalent or trivalent). For example, a first CD16a binding domain described herein can be fused to a binding domain that binds to a second binding domain that binds a different target of interest such as, e.g., a tumor associated antigen, e.g., LILRB4, etc. Such fusion proteins can comprise, between the binding domains, intermolecular linkers, such as set forth TABLE 4. Such a molecule is bivalent and can bind CD 16a on a first cell (e.g., an NK cell) as well as the second target molecule (e.g., LILRB4) on a second cell (e.g, a tumor cell). Depending on context and types of binding domains included therein, the resulting fusion proteins can bivalent or trivalent. For example, a first CD 16a binding domain provided herein can be conjugated to a second CD 16a binding domain that can be the same as or different from the first CD 16a binding domain. The resulting molecule is bivalent. Alternatively, a first CD 16a binding domain described herein can be conjugated to a second binding domain that binds a different target of interest such as, e.g., a cell-specific surface protein (e.g., a tumor- associated antigen, e.g, LILRB4 to, for example, engage the CD16a-expressing cell to a tumor cell expressing LILRB4), serum albumin e.g., to extend serum half life), etc., which may optionally be bound to a third binding domain that can be, in some embodiments, a second CD 16a binding domain or the protein to the second target. The resulting molecule is bivalent or trivalent and can bind to NK cells expressing CD 16a and another cell expressing another target (e.g., a tumor cell, e.g., an LILRB4-expressing tumor cell).

[00121] In one aspect, the disclosure provides a CD16a binding domain that comprises: (a) an amino acid sequence from 35 amino acids to 100 amino acids in length; (b) a net negative charge in phosphate buffered saline (PBS); (c) a binding affinity for CD 16a stronger than 1 pM; and (d) a stability profile such that the binding domain (i) retains at least 90% binding affinity to CD 16a upon cooling to room temperature after thermal denaturation at 95°C in PBS for at least about five minutes relative to the protein prior to thermal denaturation; (ii) retains at least 90% binding affinity to CD 16a after incubation for 16 hours at 37°C of incubation in PBS relative to the binding domain under the same conditions prior to incubating; and/or (iii) retains at least 90% binding affinity to CD16a in PBS following chemical denaturation in 4 M urea for 1 hour at room temperature relative to the binding domain prior to chemical denaturation. In another aspect, the disclosure provides a synthetic CD 16a binding domain, the binding domain comprising: (a) an amino acid sequence from 35 amino acids to 100 amino acids in length; (b) a net negative charge in PBS; (c) a binding affinity for CD 16a stronger than 1 pM; (d) at least one alpha helix; (e) at least three beta sheets; (1) at least three amino acid loops, wherein a first loop having a first amino acid sequence connects a terminal amino acid (e.g., a C-terminal amino acid) of a first beta sheet to a terminal amino acid (e.g. , a N-terminal amino acid) of a second beta sheet, and a second loop having a second amino acid sequence connects a second, terminal amino acid (e.g, an C-terminal amino acid) of the second beta sheet to a terminal amino acid (e.g., an N- terminal amino acid) of a first alpha helix sheet, and a third loop having a third amino acid sequence connects a third, terminal amino acid (e.g, an C-terminal amino acid) of the first alpha helix to a terminal amino acid (e.g, an N-terminal amino acid) of a third beta sheet; and (g) a hydrophobic core defined by at least two hydrophobic amino acids present in at least one of the alpha helices and/or at least one of the beta sheets.

[00122] In another aspect, the disclosure provides a CD 16a binding domain, the binding domain comprising (a) the first alpha helix contains at least one hydrophobic amino acid, wherein, optionally, one or more of the at least one hydrophobic amino acids is not solvent accessible; (b) the first, second, and/or third beta sheet each contains at least two or three hydrophobic amino acids, wherein, optionally, one or more of the at least two or three hydrophobic amino acids is not solvent accessible; (c) the first alpha helix contains at least one or two solvent accessible amino acids; (d) the first, second, and/or third beta sheet each contains at least one or two solvent accessible amino acids; (e) the first alpha helix contains at least one or two solvent accessible amino acids; (f) the first and/or second and/or third loop each contains at least one hydrophobic amino acid; or (g) the binding domain comprises any combination of elements selected from (a), (b), (c), (d), (e), and (f).

[00123] CD 16a binding domains of the present disclosure may include naturally-occurring or non-naturally occurring amino acids. It is understood that certain amino acids may have and/or take on different characteristics (e.g, hydrophobic, hydrophilic, neutral, etc.) depending upon the context (e.g, macro and/or micro-environment including, but not limited to surrounding amino acids, environmental conditions such as solvent type, pH, etc.).

Various terms and phrases may be used herein to describe, identify, and/or characterize amino acids. In addition, a single amino acid, at any given time, may have more than one characteristic or identity. Depending upon the context, a hydrophobic amino acid may be selected from alanine, cysteine, phenylalanine, glycine, isoleucine, leucine, methionine, proline, valine, tryptophan, tyrosine, lysine, and arginine. A hydrophilic amino acid may be an amino acid selected from cysteine, aspartic acid, glutamic acid, histidine, lysine, asparagine, glutamine, arginine, serine, threonine, tryptophan, and tyrosine. A charged amino acid may be an amino acid selected from arginine, histidine, lysine, aspartic acid (aspartate), and glutamic acid (glutamate). A positively-charged amino acid may be an amino acid selected from arginine, histidine, and lysine. A negatively-charged amino acid may be an amino acid selected from aspartic acid (aspartate) and glutamic acid (glutamate). An uncharged or neutral amino acid may be selected from alanine, cysteine, phenylalanine, glycine, histidine, isoleucine, leucine, methionine, asparagine, proline, glutamine, serine, threonine, valine, tryptophan, and tyrosine. CD 16a binding domains of present disclosure have a primary structure comprising certain key features. For example, CD16a binding domains can have amino acid sequences that include various combinations of hydrophobic and solvent accessible amino acids organized into certain domains. Primary structures (i.e., amino acid sequences) of the CD 16a binding domains will have a combination of one or more types (e.g., hydrophobic, e.g., solvent accessible) of amino acids.

[00124] In each of the foregoing aspects, the CD 16a binding domain comprises one or more of the following features: (a) free of tryptophan amino acids; (b) free of methionine amino acids; (c) free of lysine amino acids; (d) does not comprise an unpaired cysteine amino acid when cysteine amino acids are present in the binding domain; (e) free of N-linked glycosylation sites (e.g., glycosylation at N-X-S/T sites, wherein X represents any amino acid residue); (f) free of protease cleavage sites (e.g., L-X-R-R sequences, wherein X represents any amino acid residue); and (g) soluble up to at least 0.5 mM in PBS at 4 °C for one month. Preferably, the CD16a binding sites are designed to be free of glycosylation sites (e.g, free of N-linked glycosylation sites, e.g, glycosylation at N-X-S/T sites, wherein X represents any amino acid residue), for example, free of peptide sequences that are substrates for glycosylation (e.g, N-X-S/T, wherein X represents any amino acid residue), which can be a substrate for an oligosaccharyltransferase (OST) complex). Alternatively or in addition, the CD 16a binding sites are designed to be free of protease cleavage sites, for example, free of peptide sequences that are substrates for proteases (e.g, L-X-R-R (wherein X represents any amino acid residue), which can be a substrate for a Kexin/KEX2 protease). Similarly, the CD 16a binding domains may also be designed to be free of other protease cleavage sites for other proteolytic enzymes such as trypsin, chymotrypsin, elastase, subtilisin, etc. Some CD 16a binding domains may be designed to avoid cleavage by certain other enzymes, including depending upon linkers and fusion protein partners. Depending upon the circumstances, the N-terminus of the first beta sheet is preceded by one or more N-terminal amino acids and/or the C-terminus of the third beta sheet is followed by one or more C- terminal amino acids.

[00125] It is contemplated that the CD16a binding domain can comprises from 35 amino acids to 100 amino acids in length. For example, the CD 16a binding protein comprises from 35 to 95 amino acid residues in length, from 35 to 90 amino acid residues in length, from 35 to 85 amino acid residues in length, from 35 to 80 amino acid residues in length, from 35 to 75 amino acid residues in length, from 35 to 70 amino acid residues in length, from 35 to 65 amino acid residues in length, from 35 to 60 amino acid residues in length, from 35 to 55 amino acid residues in length, from 35 to 50 amino acid residues in length, from 35 to 45 amino acid residues in length, from 35 to 40 amino acid residues in length, from 40 to 95 amino acid residues in length, from 40 to 90 amino acid residues in length, from 40 to 85 amino acid residues in length, from 40 to 80 amino acid residues in length, from 40 to 75 amino acid residues in length, from 40 to 70 amino acid residues in length, from 40 to 65 amino acid residues in length, from 40 to 60 amino acid residues in length, from 40 to 55 amino acid residues in length, from 40 to 50 amino acid residues in length, from 40 to 45 amino acid residues in length, from 45 to 95 amino acid residues in length, from 45 to 90 amino acid residues in length, from 45 to 85 amino acid residues in length, from 45 to 80 amino acid residues in length, from 45 to 75 amino acid residues in length, from 45 to 70 amino acid residues in length, from 45 to 65 amino acid residues in length, from 45 to 60 amino acid residues in length, from 45 to 55 amino acid residues in length, from 45 to 50 amino acid residues in length, from 50 to 95 amino acid residues in length, from 50 to 90 amino acid residues in length, from 50 to 85 amino acid residues in length, from 50 to 80 amino acid residues in length, from 50 to 75 amino acid residues in length, from 50 to 70 amino acid residues in length, from 50 to 65 amino acid residues in length, from 50 to 60 amino acid residues in length, or from 50 to 55 amino acid residues in length. In certain embodiments, the CD16a binding domain comprises 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50 amino acids.

[00126] A CD 16a binding domain may have an amino acid sequence comprising, consisting essentially of, or consisting of an amino acid sequence having at least about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 81%, about 82%, about 83%, about 84%, about 85%, about 86%, about 87%, about 88%, about 89%, about 90%, about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98%, about 99%, about 99.5%, about 99.6%, about 99.7%, about 99.8%, about 99.9% or more or 100% identity with or to a reference sequence or component thereof, wherein the reference sequence is selected from any of SEQ ID NOs: 1-19, and portions (e.g., domains) thereof. In some embodiments, a CD 16a binding domain may have an amino acid sequence comprising an amino acid sequence having at least about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 81%, about 82%, about 83%, about 84%, about 85%, about 86%, about 87%, about 88%, about 89%, about 90%, about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98%, about 99%, about 99.5%, about 99.6%, about 99.7%, about 99.8%, about 99.9% or more or 100% identity with or to a reference sequence or component thereof selected from any of SEQ ID NOs: 20-42, 88-90, and portions thereof.

[00127] Exemplary CD 16a binding proteins (domains) useful in producing fusion proteins disclosed herein can be found in International Application No. PCT/US2025/019796, filed on March 13, 2025.

[00128] By way of a non-limiting example, a structural arrangement in a CD 16a binding domain may be depicted as N-EniLn2En3Ln4Hn5Ln6En7-C, where E is a beta sheet, H is an alpha helix, L is a loop, each of m-n? represents an integer indicating the number of amino acids in that structural domain, N and C represent N-terminal and C-terminal domains, respectively. As disclosed herein, certain CD 16a binding domains can be represented according to a formula: Rla-L2a-R2a-L2a-R3a-L3a-R4a (Formula I). In some embodiments, Ria, R2a, and R4a each corresponds to a beta sheet, and R3a corresponds to an alpha helix; LI a, L2a, and L3a are each, independently, loops. The components of Formula 1 can correspond to the aforementioned structural arrangement as follows: R3a to H_ns and each of Ria, R2a, and R4a to each of Eni, En3, and En7. The number of E amino acids does not have to be the same across sheets, for example, m, ns, and n? may be, but do not have to be, the same numbers. Similarly, m, and ne, may be, but do not have to be, the same number. For example, an exemplary formula of a synthetic CD 16a binding domain may comprise N- E8L2E9L3H11L4E6-C as depicted pictorially below denoting the amino acids in a helix domain (H) or a loop domain (L):

N-EEEEEEEELLEEEEEEEEELLLHHHHHHHHHHHLLLLEEEEEE-C.

A person skilled in the art can determine which amino acids of a given sequence constitute a loop, sheet, or a helix. See, e.g., Mirdita, et al. (2022) NAT. METHODS(19): 679-682.

[00129] Any given H domain (e.g. , H_ns) in an alpha helix that is part of a synthetic CD16a binding domain may independently contain between about 4 amino acids and about 20 amino acids in length. A helical structure (represented by H) may independently comprise 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more amino acids in length. Any given E domain (e.g, E_ns) in a beta sheet that is part of a synthetic CD 16a binding domain may independently contain between about 4 and about 16 amino acids in length. A sheet structure (represented by E) may independently comprise 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, or more amino acids in length.

[00130] Loops disposed between alpha helices and/or beta sheets may also be of the same or different lengths. Each loop may independently comprise at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20 or 25 amino acids or more in length. In some embodiments, each loop is independently between at least 2, 3, 4, 5, or 6 amino acids in length.

[00131] In a protein with more than one alpha helical structure, each alpha helix may comprise the same number of amino acids in each of its H domains or different numbers of amino acids in length in reference to the primary structure of each helical region. That is, in some synthetic CD 16a binding domains having more than one alpha helix, each helix in the binding domain is the same length. In some synthetic CD 16a binding domains having one or more alpha helices, one or more helices has a different length relative to other helical structures in the binding domain. In some embodiments, a helix has zero, one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty or more conserved amino acids (e.g. relative to other CD 16a binding domains).

[00132] In a protein with more than one beta sheet, each beta sheet may comprise the same number of amino acids in each of its E domains or different numbers of amino acids in length in reference to the primary structure of each sheet region. That is, in some synthetic CD 16a binding domains having more than one beta sheet, each beta sheet (E) in the binding domain is the same length. In some synthetic CD 16a binding domains having one or more beta sheets, one or more beta sheets has a different length relative to other beta sheets in the binding domain. In some embodiments, a beta sheet has zero, one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen or more conserved amino acids (e.g., relative to other CD 16a binding domains).

[00133] In an exemplary CD 16a binding domain disclosed herein, the binding domain comprises at least one alpha helix, at least three beta sheets, and at least three loops (a first loop, a second loop, and a third), wherein a first loop having a first amino acid sequence connects a terminal amino acid (e.g, a C-terminal amino acid) of a first beta sheet to a terminal amino acid (e.g, aN-terminal amino acid) of a second beta sheet, and a second loop having a second amino acid sequence connects a second, terminal amino acid (e.g, an C- terminal amino acid) of the second beta sheet to a terminal amino acid (e.g, an N-terminal amino acid) of a first alpha helix sheet, and a third loop having a third amino acid sequence connects a third, terminal amino acid (e.g, an C-terminal amino acid) of the first alpha helix to a terminal amino acid (e.g, an N-terminal amino acid) of a third beta sheet.

[00134] In certain embodiments, a synthetic CD 16a binding domain may have (a) a first alpha helix containing at least one hydrophobic amino acid, wherein, optionally, one or more of the at least one hydrophobic amino acids is not solvent accessible; (b) first, second, and/or third beta sheets each containing at least two or three hydrophobic amino acids, wherein, optionally, one or more of the at least two or three hydrophobic amino acids is not solvent accessible; (c) a first alpha helix containing at least one or two solvent accessible amino acids; (d) first, second, and/or third beta sheets containing at least one or two solvent accessible amino acids; (e) a first alpha helix containing at least one or two solvent accessible amino acids; (f) first and/or second and/or third loops containing at least one hydrophobic amino acid; or (g) any combination of elements selected from (a), (b), (c), (d), (e), and (f).

[00135] In some embodiments, a synthetic CD 16a binding domain may have the (a) first, second, and/or third beta sheets each containing at least two hydrophobic amino acids; (b) first, second, and third beta sheets each containing at least one solvent accessible amino acid;

(c) first, second, and third beta sheets each containing at least two hydrophobic and one solvent accessible amino acids; (d) first alpha helix containing at least four solvent accessible amino acids; and/or (e) first, second, and/or third loops containing at least one hydrophobic amino acid.

[00136] In certain embodiments, a synthetic CD16a binding domain further comprises one or more N-terminal amino acids, N-terminal to Ria, and/or one or more C-terminal amino acids, C-terminal to R4a, wherein the one or more N-terminal amino acids and/or the one or more C-terminal amino acids are part of the binding domain and not, for example, part of a loop or an intermolecular linker (e.g, between a first binding domain, e.g, DI, and a second binding domain, e.g., D2, etc.).

[00137] A synthetic CD 16a binding domain may have one of several consensus sequence structures. Consensus sequences will generally have certain “fixed” amino acid positions as well as those that can be varied, such as by changing to another amino acid. Sometimes changing amino acids at certain positions can alter the function of the synthetic binding domain by increasing or decreasing affinity for the target (/.£., CD 16a). However, all the binding domains disclosed herein, although having different primary structures have a minimal “threshold” binding affinity to CD 16a. In some embodiments, a threshold binding affinity may be stronger than about 1 pM, about 100 nM, about 10 nM, or about 1 nM.

[00138] A synthetic CD 16a binding domain disclosure may be represented according to a formula shown as one or more domains, wherein each domain optionally has one or more conserved amino acid residues and/or a particular structure (e.g., loop, e.g., helix). For example, an exemplary synthetic CD 16a binding domain comprises an amino acid sequence arranged in a primary structure of:

Rla-Lla-R2a-R2a-R3a-L3a-R4a (Formula I) where Ria, R2a, and R4a represent beta sheets, R3a represents an alpha helix, and Lla, L2a, and L3a represent loops connecting the alpha helix and/or beta sheets. The amino acid sequence of the starting “parental” protein and exemplary consensus sequences for various miniproteins developed are set forth in TABLE 1A. Bold, underlined residues represent beta sheets and correspond to Ria, R2a, and R4a, and bold, italicized residues represent helical residues and correspond to R3a, in order along a given consensus sequence. Exemplary consensus sequences for Ria, R2a, R3a, R4a, Lla, L2a, and L3a for each miniprotein are set forth in TABLE IB. Positions for each variable amino acid along the length of a consensus sequence as set forth in TABLES 1A and IB are set forth in TABLE 1C.

[00139] A synthetic CD 16a binding domain having the amino acid sequence of a first miniprotein (referred to a Reference Miniprotein 1) RTLRVTVTHPDGSVRRLTVDADDVVDTVDRLDARTPEGTVIHIEEA (SEQ ID NO: 1) was developed, characterized, and optimized (as set forth in Example 1). Derivative miniproteins are set forth in TABLE IE. Allowable residues in linear positions from N-to-C terminus in Formula I are set forth in TABLE ID, with reference to SEQ ID NO: 5 (Reference Miniprotein 5). The “allowable residues” represent those residues which may be changed from the reference residue at a given position in Reference Miniprotein 5 (SEQ ID NO: 5). For example, position 3 in TABLE ID is listed as having “M” as an allowable residue, but position 3 in SEQ ID NO: 5 as in TABLE IE is L, so the allowable residues are in addition to those set forth in SEQ ID NO: 5. To give but another example, position 42 in TABLE ID is listed as R, and Q is an allowable residue, thus, allowable residues should be considered to be those in addition to the residues present in SEQ ID NO: 5. TABLE 1A. Exemplary CD 16a Binding Domain Consensus Sequences

*Bolded and italicized text indicates amino acids present in an alpha helix (positions 24-34); and bolded, underlined text indicates amino acids present in a beta sheet (positions 2-9; 12-20; and 40-45).

TABLE IB. Exemplary CD16a Binding Domain Consensus Sequences

TABLE 1C. Exemplary Synthetic CD16a Miniprotein Variable Substitutions in CD16a

Binding Domains

TABLE ID. Allowable Residues in CD 16a Binding Domains

* Relative to linear positions and amino acids in SEQ ID NO: 5. Allowable residues are in addition to those already present in SEQ ID NO: 5.

TABLE IE. Exemplary CD 16a Binding Domains

B. LILRB4 binding domains

[00140] LILRB4 binding domains suitable for use in fusion proteins disclosed herein generally avoid certain such disadvantages as they have high binding specificity to LILRB4, can be engineered to have desired pharmacodynamic and pharmacokinetic properties (e.g, a desirable circulating half-life in plasma), reduced immunogenicity (e.g, do not elicit an immune response against them), are chemically and thermally stable, and are resistant to protease degradation (e.g, Kex L-X-R-R- (wherein X represents any amino acid residue), deamination and post-translational modification (e.g, glycosylation through N-X-S/T, wherein X represents any amino acid residue), and are stable in different redox environments.

[00141] An LILRB4 binding domain suitable for use in fusion proteins disclosed herein can have mM-level solubility. In some embodiments, an LILRB4 binding domain has a solubility greater than 0.01 mg/mL, 0.05 mg/mL, 0.1 mg/mL, 0.5 mg/mL, 1 mg/mL, 5 mg/mL or 10 mg/mL in an aqueous solution. Suitable LILRB4 binding domains generally specifically bind LILRB4. In certain embodiments, LILRB4 binding domains bind with an affinity at least at a level of a reference LILRB4 antagonist (e.g, an LILRB4 antibody, e.g, 10-202 (Immune-One)).

[00142] In certain embodiments, LILRB4 binding domains suitable for use in fusion proteins of the present disclosure comprise: (a) an amino acid sequence from 35 to 100 amino acid residues in length; (b) a net negative charge in phosphate buffered saline (PBS); (c) a binding affinity for LILRB4- stronger than 1 pM; and (d) a stability profile such that the binding domain (i) retains at least 90% binding affinity to LILRB4-upon cooling to room temperature after thermal denaturation at 95 °C in PBS for at least about five minutes relative to the binding domain prior to thermal denaturation; (ii) retains at least 90% binding affinity to LILRB4-after incubation for 16 hours at 37 °C of incubation in PBS relative to the binding domain under the same conditions prior to the incubation; and/or (iii) retains at least 90% binding affinity to LILRB4 in PBS following chemical denaturation in 4 M urea for 1 hour at room temperature relative to the binding domain prior to chemical denaturation.

[00143] In certain embodiments, LILRB4 binding domains suitable for use in fusion proteins of the present disclosure comprise: (a) an amino acid sequence from 35 to 100 amino acid residues in length; (b) a net negative charge in PBS; (c) a binding affinity for LILRB4 stronger than 1 pM; (d) at least one alpha helix; (e) at least three beta strands; and (e) at least three amino acid loops, wherein a first loop having a first amino acid sequence connects a terminal amino acid (e.g, a C-terminal amino acid) of a first beta strand to a terminal amino acid (e.g, aN-terminal amino acid) of a first alpha helix; a second loop having a second amino acid sequence connects a second terminal amino acid (e.g, an C-terminal amino acid) of the first alpha helix to a terminal amino acid (e.g, an N-terminal amino acid) of a second beta strand; and a third loop having a second amino acid sequence connects a second terminal amino acid (e.g, an C-terminal amino acid) of the second beta strand to a terminal amino acid (e.g, an N-terminal amino acid) of a third beta strand.

[00144] Exemplary LILRB4 binding proteins (domains) useful in producing fusion proteins of the present disclosure can be found in International Application No. PCT/US2025/019794, filed on March 13, 2025.

[00145] In certain embodiments, LILRB4 binding domains suitable for use in fusion proteins of the present disclosure comprise: an amino acid sequence arranged in a primary structure of

Rlb-Llb-R2b-L2b-R3b-L3b-R4b

(Formula II) wherein Rib, R2b, R3b, and R4b are regions 1, 2, 3, and 4, respectively, and Lib, L2b, and L3b are loops 1, 2 and 3, respectively. [00146] In certain embodiments, a synthetic LILRB4 binding domain further comprises one or more N-terminal amino acids, N-terminal to Rib, and/or one or more C-terminal amino acids, C-terminal to R4b, wherein the one or more N-terminal amino acids and/or the one or more C-terminal amino acids are part of the binding domain and not, for example, part of a loop or an intermolecular linker (e.g., between a first binding domain, e.g., DI, and a second binding domain, e.g., D2, etc.).

[00147] TABLE 2A provides exemplary sequences for Rib, R2b, R3b, R4b, Lib, L2b, and L3b; and TABLE 2B provides exemplary LILRB4 binding domain variable substitutions.

TABLE 2A. Exemplary LILRB4 Synthetic Binding Domain Consensus Sequences

TABLE 2B. Exemplary LILRB4 Synthetic Binding Domain Variable Substitutions

TABLE 2C. Exemplary LILRB4 Binding Domain Consensus Sequences [00148] In certain embodiments, the LILRB4-binding domain comprises an amino acid sequence of

ITVDSLLX9ASVVAYQIX18X19X20NPNVX25 VX27IX29YDEETHRYYIVTTE (SEQ ID NO: 43; see Table 2C), wherein X9 is E or V; X18 is D or Q; X19 is H, R, or S; X20 is A, D, or E; X25 is A, R or Y; X27 is E, Q, or S; and X29 is H, R, or Y.

[00149] In some embodiments, the LILRB4-binding domain comprises an amino acid sequence of any one of SEQ ID NOs: 47-52 as set forth in TABLE 2D.

TABLE 2D. Exemplary LILRB4 Synthetic Binding domains

[00150] It is contemplated that the LILRB4 binding domain can comprises from 35 amino acids to 100 amino acids in length. For example, the LILRB4-binding protein comprises from 35 to 95 amino acid residues in length, from 35 to 90 amino acid residues in length, from 35 to 85 amino acid residues in length, from 35 to 80 amino acid residues in length, from 35 to 75 amino acid residues in length, from 35 to 70 amino acid residues in length, from 35 to 65 amino acid residues in length, from 35 to 60 amino acid residues in length, from 35 to 55 amino acid residues in length, from 35 to 50 amino acid residues in length, from 35 to 45 amino acid residues in length, from 35 to 40 amino acid residues in length, from 40 to 95 amino acid residues in length, from 40 to 90 amino acid residues in length, from 40 to 85 amino acid residues in length, from 40 to 80 amino acid residues in length, from 40 to 75 amino acid residues in length, from 40 to 70 amino acid residues in length, from 40 to 65 amino acid residues in length, from 40 to 60 amino acid residues in length, from 40 to 55 amino acid residues in length, from 40 to 50 amino acid residues in length, from 40 to 45 amino acid residues in length, from 45 to 95 amino acid residues in length, from 45 to 90 amino acid residues in length, from 45 to 85 amino acid residues in length, from 45 to 80 amino acid residues in length, from 45 to 75 amino acid residues in length, from 45 to 70 amino acid residues in length, from 45 to 65 amino acid residues in length, from 45 to 60 amino acid residues in length, from 45 to 55 amino acid residues in length, from 45 to 50 amino acid residues in length, from 50 to 95 amino acid residues in length, from 50 to 90 amino acid residues in length, from 50 to 85 amino acid residues in length, from 50 to 80 amino acid residues in length, from 50 to 75 amino acid residues in length, from 50 to 70 amino acid residues in length, from 50 to 65 amino acid residues in length, from 50 to 60 amino acid residues in length, or from 50 to 55 amino acid residues in length.

In certain embodiments, the LILRB4 binding domain comprises 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50 amino acids.

C. MULTIVALENT FUSION PROTEINS

[00151] In one aspect, the disclosure provides bivalent fusion proteins comprising a CD 16a binding domain (DI) and an LILRB4 binding domain (D2).

[00152] In certain embodiments, the C-terminal amino acid residue of DI is linked (e.g., via a linker as disclosed herein) to the N-terminal amino acid residue of D2, such that the order of domains within the bivalent fusion protein, from N-terminus to C-terminus is D1-D2 (see FIG. 3A). In certain embodiments, the C-terminal amino acid residue of D2 is linked (e.g., via a linker as disclosed herein) to the N-terminal amino acid residue of DI, such that the order of domains within the bivalent fusion protein, from N-terminus to C-terminus is D2-D1 (see FIG. 3B). Sequences of exemplary bivalent fusion proteins of the disclosure are provided in TABLE 3.

[00153] In another aspect, the disclosure provides a fusion protein comprising a first binding domain that binds a first target attached by at least one linker to a second binding domain that binds a second target, wherein the first and second binding domains are synthetic binding domains comprising synthetic binding proteins that each have an N-terminal amino acid residue and a C-terminal amino acid residue, wherein the first target is CD 16a, and the second target is LILRB4. In certain embodiments, the first and second binding domains are linked by a first linker. In some such embodiments, the C-terminal amino acid residue of the first binding domain is linked to the N-terminal amino acid residue of the second binding domain. Alternatively, the C-terminal amino acid residue of the second binding domain is linked to the N-terminal amino acid residue of the first binding domain. In certain embodiments, such a fusion protein further comprises a third binding domain. In some embodiments, the third binding domain binds CD 16a.

[00154] In addition, the disclosure provides trivalent fusion proteins comprising a first CD 16a binding domain (DI), a first LILRB4 binding domain (D2), and another binding domain (D3). In certain embodiments, D3 is a second CD 16a binding domain, which can be the same or different than the first CD 16a binding domain (DI). In certain embodiments, D3 is a second LILRB4 binding domain, which can be the same or different than the first LILRB4 binding domain (D2).

[00155] Any ordering of domains DI, D2, and D3 from the N- to the C- terminus may be possible. For example, the order of domains within the trivalent fusion protein may be Dl- D2-D3 from N-terminus to C-terminus, with the C-terminal amino acid residue of DI linked (e.g., via a linker as disclosed herein) to the N-terminal amino acid residue of D2 and the C- terminal amino acid residue of D2 linked (e.g., via a linker as disclosed herein) to the N- terminal amino acid residue of D3. (see FIG. 4B, which shows an example fusion protein with D3 being a second CD16a binding domain.)

[00156] Alternatively, the order of domains within the trivalent fusion protein may be Dl- D3-D2 from N-terminus to C-terminus, with the C-terminal amino acid residue of DI linked (e.g, via a linker as disclosed herein) to the N-terminal amino acid residue of D3 and the C- terminal amino acid residue of D3 linked (e.g., via a linker as disclosed herein) to the N- terminal amino acid residue of D2. (see FIG. 4D, which shows an example fusion protein with D3 being a second CD16a binding domain.)

[00157] Alternatively, the order of domains within the trivalent fusion protein may be D2- D1-D3 from N-terminus to C-terminus, with the C-terminal amino acid residue of D2 linked (e.g, via a linker as disclosed herein) to the N-terminal amino acid residue of DI and the C- terminal amino acid residue of DI linked (e.g., via a linker as disclosed herein) to the N- terminal amino acid residue of D3. (see FIG. 4A, which shows an example fusion protein with D3 being a second LILRB4 binding domain and FIG. 4E, which shows an example fusion protein with D3 being a second CD16a binding domain.)

[00158] As a further example, the order of domains within the trivalent fusion protein may be D2-D3-D1 from N-terminus to C-terminus, with the C-terminal amino acid residue of D2 linked (e.g., via a linker as disclosed herein) to the N-terminal amino acid residue of D3 and the C-terminal amino acid residue of D3 linked (e.g., via a linker as disclosed herein) to the N-terminal amino acid residue of DI. (See FIG. 4C, which shows an example fusion protein with D3 being a second LILRB4 binding domain).

[00159] In certain embodiments, where DI and D3 comprise CD 16a binding domains, the amino acid sequence of the binding domains can be selected from at least one amino acid sequence as set forth in any of SEQ ID NOs: 20-42, 88-90 and/or Tables 1A, IB, 1C and/or ID. In some embodiments, the amino acid sequence of the first and/or third binding domain may be or comprise an amino acid sequence selected from any of SEQ ID NOs: 1-19. In certain embodiments, the first and/or third binding domains bind CD16a with a binding affinity stronger than about 1 pM, about 0.75 pM, about 0.5 pM, about 0.25 pM, about 0.1 pM, about 75 nM, about 50 nM, about 25 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4 nM, about 3 nM, about 2 nM, about 1 nM, about 0.75 nM, about 0.5 nM, about 0.25 nM, about 0.1 nM to about 0.01 nM, and about 0.01 nM to about 0.001 nM.

[00160] In certain embodiments, where D2 comprises an LILRB4 binding domain, the amino acid sequence of the binding domain can be selected from an amino acid sequence as set forth in any of any of SEQ ID NOs: 44-46, 86-87, Table 2A and/or Table 2B. In some embodiments, the amino acid sequence of the second binding domain comprises at least one amino acid sequence selected from any of SEQ ID NOs: 43, 47-52, Table 2C and/or Table 2D. In certain embodiments, the second binding domain binds LILRB4 with a binding affinity stronger than about 1 pM, about 0.75 pM, about 0.5 pM, about 0.25 pM, about 0.1 pM, about 75 nM, about 50 nM, about 25 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4 nM, about 3 nM, about 2 nM, about 1 nM, about 0.75 nM, about 0.5 nM, about 0.25 nM, about 0.1 nM to about 0.01 nM, and about 0.01 nM to about 0.001 nM.

[00161] Linkers of the provided fusion proteins may be selected from Table 4. In certain embodiments, a first linker of a fusion protein provided herein is selected from any of SEQ ID NOs: 64-85 and/or Table 4. In certain embodiments, a second linker of a fusion protein provided herein is selected from any of SEQ ID NOs: 64-85 and/or Table 4. Within a given fusion protein comprising two linkers, a first linker and a second linker may be the same or may be different from one another.

[00162] Sequences of exemplary trivalent fusion proteins of the disclosure are set forth in any of SEQ ID NOs: 53-63, as provided in TABLE 3.

[00163] In certain embodiments, the disclosure provides a nucleic acid encoding a fusion protein provided herein. In certain embodiments, the disclosure provides a host cell comprising a nucleic acid encoding a fusion protein provided herein. TABLE 3. Exemplary Synthetic CD16-LILRB4 Fusion Proteins

* Bold, italics indicate an amino acid sequence of an exemplary synthetic CD16a binding domain; regular text indicates an exemplary linker (“PX”, where X is a number of residues set forth in the construct name, e.g., P16 has 16 amino acid residues); and italic, underlined text indicates an amino acid sequence of an exemplary synthetic LILR4B binding domain. V. LINKERS

[00164] The present disclosure provides intermolecular linkers that can be used to attach CD16a and LILRB4 binding domains to one another (see, e.g., FIGS. 3A-3C, and 4A-4E). Exemplary linkers are set forth in TABLE 4.

TABLE 4. Intermolecular Linkers

VI. SYNTHESIS OF MULTIVALENT CD16-LILRB4 BINDING DOMAINS

[00165] The synthetic CD 16a and LILRB4 binder containing fusion proteins provided herein may be produced by methods known to those of ordinary skill in the art. Methods may include, for example, biological approaches, such as recombinant approaches and/or chemical approaches, such as solid phase and/or liquid phase chemical synthesis, etc., or combinations thereof.

[00166] With regard to recombinant approaches, a variety of methodologies can be implemented to produce the binding domains disclosed herein. For example, DNA molecules encoding the binding domains can be synthesized chemically and/or cloned/produced using recombinant DNA methodologies. The resulting DNA molecules encoding binding domains of interest can be ligated to other nucleotide sequences, including, for example, expression control sequences, to produce a gene expression construct (i.e., expression vector). Thereafter, the resulting expression vectors are introduced into host cells using conventional transfection or transformation techniques. Exemplary host cells include E. coli cells, Bacillus subtilis cells, Pichia Pastoris cells, Bacillus subtilis cells, Saccharomyces cerevisiae cells, Kluyveromyces lactis cells, Chinese hamster ovary (CHO) cells, human embryonic kidney 293 (HEK 293) cells, HeLa cells, baby hamster kidney (BHK) cells, monkey kidney cells (COS), and human hepatocellular carcinoma cells (e.g, Hep G2). The transformed host cells can be grown under conditions that permit the host cells to express the genes that encode the binding domains.

[00167] Specific expression and purification conditions will vary depending upon the expression system employed. For example, if a gene is expressed in E. coli, it is first cloned into an expression vector by positioning the engineered gene downstream from a suitable bacterial promoter, e.g., T7, lac, Trp or Tac, and, in some contexts, a prokaryotic signal sequence or fusion to a protein such as, e.g., Trx, MBP, SUMO, or OsmY. The expressed protein may be secreted. The expressed protein can be harvested after disruption of the cells by French press or sonication (e.g, in the presence of 4-6 M urea). Alternatively, or in addition, the binding domains can be harvested and purified or isolated from cell extracts using techniques known in the art, e.g., affinity tags such as glutathione-S-transferase (GST) or histidine tags. Protease cleavage with SUMO Protease (Ulp), thrombin, enterokinase, TEV protease, 3C protease may be used to cleave affinity tags and fusion proteins from the miniprotein binder. The protein may be further purified with reverse phase HPLC using a C- 18 column and eluted in a solvent gradient (e.g, gradient of acetonitrile). The protein may then be lyophilized to remove solvent and may be resuspended in phosphate buffered saline. Purification by reverse phase HPLC may be used to remove endotoxin from samples expressed in E. coli. [00168] If the engineered gene is expressed in eukaryotic host cells, e.g., CHO cells, it is first inserted into an expression vector containing a suitable eukaryotic promoter, a secretion signal, a poly A sequence, and a stop codon. Optionally, the vector or gene construct may contain enhancers. The vector may also optionally contain fusion domains which can be used to facilitate expression and secretion. Vectors may also optionally contain enzyme (e.g, protease) cleavage sites. The gene construct can be introduced into eukaryotic host cells using conventional transfection (e.g., for mammalian) and transformation (e.g, for yeast).

[00169] In addition, the synthetic binding domains may be produced in cell-free systems. For example, chemical synthesis such as organic chemical synthesis using liquid and/or solid phase chemical processes may be used. Such processes and tools for performing such processes, such as various automatic synthesizers, are well known to those of ordinary skill in the art and such tools are widely commercially available. More specifically, methods of chemically synthesizing polypeptides are well known in the art and include, but are not limited to, solid-phase peptide synthesis, liquid-phase peptide synthesis, and organic synthesis methods. In some synthetic approaches, an amino group of one amino acid (or amino acid derivative) is linked to a carboxyl group of another amino acid (or amino acid derivative) that has been activated by reacting it with a reagent such as dicyclohexylcarbodiimide (DCC). When the free amino group attacks the activated carboxyl group, a peptide bond is formed and dicyclohexylurea is released. In such methods, other potentially reactive groups (such as the a-amino group of the N-terminal amino acid or amino acid derivative and the carboxyl group of the C-terminal amino acid or amino acid derivative) may be blocked (“protected”) from participating in the chemical reaction. Thus, only particular active groups react such that the desired product is formed. Blocking groups useful for this purpose include, without limitation, tertbutoxy carbonyl groups (t-Boc) and benzoyloxy carbonyl groups to protect amine groups; and simple esters (such as methyl and ethyl groups) and benzyl esters to protect carboxyl groups. Blocking groups can typically be subsequently removed with a treatment that leaves peptide bonds intact (for example, treatment with dilute acid). This process of protecting reacting groups that should not react, coupling to form a peptide bond, and deprotecting reactive groups may be repeated. A peptide may be synthesized by sequentially adding amino acids to a growing peptide chain. [00170] Both liquid-phase and solid phase peptide synthesis methods can be used to make the binding domains described herein. In solid-phase peptide synthesis, the growing peptide chain is typically linked to an insoluble matrix (such as, for example, polystyrene beads) by linking the carboxyterminal amino acid to the matrix. At the end of synthesis, the peptide can be released from the matrix using a cleaving reagent that does not disrupt peptide bonds, such as hydrofluoric acid (HF). Protecting groups are also typically removed at this time. Automated, high throughput, and/or parallel peptide synthesis methods may also be used in accordance with the disclosure. For more information about peptide synthesis methods, see, e.g, Merrifield (1969) ADV. ENZYMOL. RELAT. AREAS MOL. BIOL., 32: 221-96; Fridkin et al. (1974) ANN. REV. BIOCHEM. 43(0): 419-43; Merrifield (1997) METH. ENZYMOL. 289: 3-13; Sabatino et al. (2009) CURR. OPIN. DRUG DISCOV. DEVEL., 11(6): 762-70.

[00171] Once synthesized, the binding domains can be purified using standard approaches including, for example, chromatographic (e.g., reverse phase HPLC) and affinity binding approaches. The resulting binding domains can then be characterized using a variety of chemical, biological and biophysical approaches.

VII. CHARACTERIZATION OF MULTIVALENT CD16-LILRB4 BINDING DOMAINS

A. Biophysical Characterization

[00172] The synthetic binding domains described herein may be characterized using a variety of approaches to determine, e.g, secondary and tertiary conformation, binding affinity, binding selectivity, stability (e.g, thermostability, chemical stability, propensity to degrade, etc.), solubility, etc.

[00173] For example, protein conformation may be measured via circular dichroism spectroscopy, infrared spectroscopy, NMR, X-ray crystallography, cryo-electron microscopy and AlphaFold (alphafold.ebi.ac.uk/). Binding affinity and/or selectivity may be determined using assays such as flow cytometric analyses using, e.g., yeast or mammalian cells, biolayer interferometry and/or surface plasmon resonance measurements, each of which will be able to determine different types and specificities of binding.

[00174] Binding affinity and/or avidity can be determined by measuring the equilibrium dissociation constant (KD) of a synthetic CD 16a binding domain, a synthetic LILRB4 binding domain, or a synthetic multivalent CD16a-LILRB4 fusion to a target. In some embodiments, the binding affinity (KD) of synthetic CD 16a binding domains, LILRB4 binding domains or multivalent CD16a-LILRB4 fusion binding domains are in the range of 10'⁵ M or less, or ranging down to IO'¹⁰ M or lower, (e.g, about 10'⁶,l O'⁷, 10'⁸, 10'⁹, 10'¹⁰M or less).

[00175] In some embodiments, the synthetic CD 16a binding domain, LILRB4 binding domain or multivalent CD16a-LILRB4 fusion protein comprises a binding affinity characterized by a dissociation constant ranging from about 1 pM to 1 pM. In some embodiments, the binding affinity is between about 0.001 nM to about 1 pM; about 0.01 nM to about 1 pM; about 0.1 nM to about 1 pM; about 1 nM to about 1 pM; about 1 nM to about 0.5 pM; about 1 nM to about 0.25 pM; about 1 nM to about 0.10 pM; about 1 nM to about 75 nM; about 1 nM to about 50 nM; about 1 nM to about 25 nM; about 1 nM to about 10 nM; and about 1 nM to about 5 nM. In some embodiments, the binding affinity is stronger than about 1 pM, about 0.75 pM, about 0.5 pM, about 0.25 pM, about 0.1 pM, about 75 nM, about 50 nM, about 25 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4 nM, about 3 nM, about 2 nM, about 1 nM, about 0.75 nM, about 0.5 nM, about 0.25 nM, about 0.1 nM to about 0.01 nM, and about 0.01 nM to about 0.001 nM. One of ordinary skill in the art will readily know how to calculate equilibrium dissociation constants using measured Ka (1/sec) and Kd (1/secM) of the synthetic CD16a binding domains, LILRB4 binding domains or multivalent CD16a-LILRB4 fusion binding domains.

[00176] Other analytical techniques (some of which are also used for synthesis and purification) include, without limitation, HPLC, LCMS, quantitative thin layer chromatography and others known to those of skill in the art. Stability can be measured using assays that expose binding domains to elevated temperatures (e.g, 37 °C, e.g, 95 °C, etc.) and/or chemical denaturants (e.g, urea, and guanidine hydrochloride) and then observe whether the protein refolds into its pre-exposure structure/ conformation and/or regains binding activity to a given target molecule. Degradation can be evaluated using techniques such as reverse phase HPLC or gel electrophoresis to monitor resistance of a synthetic binding domain to degradation.

B. Biochemical Characterization

[00177] The biological activity of the binding domains can be determined via in vitro and in vivo assays (see, e.g., Examples 4, 5, 6, 7, and 8). Such assays can be used to determine whether a binding domain (including, e.g., a multivalent fusion binding domain) has agonistic (or antagonistic) properties, or is able to specifically and strongly bind to a cell (e.g., a CD16a-expressing NK cell, e.g, an LILRB4-expressing tumor cell, both an NK cell and a tumor cell, etc.). For example, the suitable assays can be performed to determine whether a synthetic CD16a or LILRB4 binding domain disclosed herein can, e.g., bind to an NK cell or tumor cell, respectively such as in a way that will allow the NK cell to be engaged with the tumor cell if the two binding domains are conjugated to one another. Assays can also be conducted to measure efficacy of a multivalent fusion protein comprising CD 16a and LILRB4 to assess whether CD16a-expressing NK cells can effectively be engaged to target LILRB4-expressing tumor cells for destruction. Depending on the circumstances, suitable assays can be performed to determine whether (i) a synthetic CD 16a binding domain disclosed herein can bind to CD16a expressed on a surface of a cell e.g, partially or completely bind to and/or engage CD16a-mediated signaling or activity in NK cells and/or on target cells; (ii) a synthetic LILRB4 binding domain disclosed herein can bind to LILRB4 expressed on a surface of a cell (e.g, a tumor cell); and/or (iii) whether a synthetic multivalent fusion protein comprising at least one domain that binds to CD 16a and at least one domain that binds to LILRB4 can engage cells expressing CD16a (e.g, NK cells) to target LILRB4 expressing cells (e.g, tumor cells) for destruction.

[00178] Cellular phenotype or function, e.g, through a cell surface molecule, e.g, a receptor such as CD 16a, e.g, a surface-expressed molecule such as LILRB4, can be measured using any number of commercial assays used to characterize, e.g, cellular phenotype using surface markers, cell death of tumor cells (e.g, LILRB4-expressing tumor cells, e.g, achieved via NK-induced cytotoxicity, etc.). For example, a flow activated cell sorting (FACS)-based assay may be used to identify cell surface markers such as LILRB4 which can be present on tumor cells. FACS/flow cytometric analyses can also be used to detect surface molecules such as, e.g, CD25, CD69, and/or CD107, all of which are known NK-cell markers. CD25 and CD69 are considered early and late immune cell markers, respectively, and CD 107 is considered a marker of activated NK cells. Expression of one or more of these cell surface markers may be used to phenotype and “stage” an NK cell, e.g., to which a synthetic CD16a binding domain of the present disclosure has bound. In certain embodiments, NK cells contacted with a fusion protein comprising CD 16a and LILRB4 binding domains, in the presence of a cell expressing LILRB4, in accordance with the present disclosure, can be analyzed to determine whether the NK cells express CD25, CD69 and/or CD 107 in a greater quantity or, as a population, a greater percentage express one or more of CD25, CD69, and/or CD 107 as compared to NK cells contacted with a control fusion protein or NK cells prior to contacting and/or in the absence of a cell expressing LILRB4.

[00179] Other in vitro assays can measure, for example, cytotoxicity through evaluation of percent dead tumor cells. For example, an assay that co-cultures primary NK cells with cancer cells (e.g., a tumor cell line, e.g, a tumor cell line expressing a target such as LILRB4 and/or a control line not expressing the target) and then with fusion proteins of the present disclosure or protein controls may be performed. After incubation, flow cytometric analysis can be conducted to determine percentage of cell death. In some embodiments, cells exposed to bivalent fusion proteins (having a CD 16a and LILRB4 binding domains) will die at a greater rate than those exposed to control proteins. In some embodiments, cells exposed to trivalent fusion proteins in accordance with FIGS. 4A-4E will die at a greater rate than those exposed to bivalent or control fusion proteins.

[00180] Characterization assays may also be conducted in vivo. For example, synthetic CD 16a binding domains may be tested for selective binding by comparing a control CD 16a binding domain (e.g., a non-binding domain with the same size and shape as synthetic CD 16a binding domains, but without any identity in paratope regions) to synthetic CD 16a binding domains as provided herein (e.g, as in TABLES 1A-1E). Synthetic LILRB4 binding domains may be tested for selective binding by comparing a control LILRB4 binding domain (e.g, anon-binding domain with the same size and shape as synthetic LILRB4 binding domains, but without any identity in paratope regions) to synthetic CD 16a binding domains as provided herein (e.g., as in TABLES 2A-2D). Fusion proteins comprising CD 16a and LILRB4 binding domains may also be tested on binding assays conducted for any monovalent protein as provided herein.

[00181] Various assays may be used to evaluate efficacy of synthetic CD 16a binding domains to determine their ability to activate CD16a-mediated signaling. Certain assays can measure the presence and/or extent of CD16a signaling by its natural ligands (e.g, Fc region of IgG) and compare to synthetic CD 16a binding domains. In such an assay, a synthetic CD 16a binding domain binds to CD 16a, identifying NK cells. NK cell phenotype can be evaluated, for example, by presence of surface markers such as CD25, CD69 and/or CD 107. Various assays may also be used to evaluate efficacy of synthetic LILRB4 binding domains to determine their ability to bind to an LILRB4-expressing cell (e.g., a tumor cell). Certain assays can measure the presence and/or extent of LILRB4 binding by its natural ligands (e.g., galectin-8) and compare to LILRB4 synthetic binding domains. In such an assay, a synthetic LILRB4 binding domain binds to LILRB4, identifying a LILRB-4 expressing cell (e.g., a tumor cell).

[00182] Assays conducted with cells such as primary immune cells e.g., NK cells from healthy human subjects, NK cells from human subjects with cancer, PBMCs from healthy human subjects, PBMCs from human subjects with cancer, cancer cells such as stable tumor cell lines e.g., MV -4-11, OCI-AML-3, etc. ), cancer cells from subjects with cancer (e.g, tumor cells, etc.), may be used to evaluate and characterize synthetic CD16a and/or LILRB4 binding domains as well as multivalent synthetic CD16a-LILRB4 fusion proteins. For example, cells used in assays of the present disclosure may have a visualizable reporter that is detectable upon CD16a and/or LILRB4-binding. Cells used in assays of the present disclosure may express certain surface antigens (e.g, LILRB4, e.g., LILRB4-expressing cell lines) or not (e.g., LILRB4 negative control cell lines).

[00183] In some embodiments, CD 16a or LILRB4 miniprotein binding can be measured by contacting a population of cells expressing CD 16a or LILRB4 with a CD 16a or LILRB4 miniprotein, respectively, and measuring detection of a reporter (e.g, a fluorescent reporter, e.g., a flag-tag and/or visualizable secondary antibody, etc.).

[00184] Cell engagement can be measured by contacting a population of cells comprising CD 16a expressing cells and LILRB4 expressing cells with a synthetic multivalent CD 16a- LILRB4 miniprotein and measuring cellular viability and/or activity (e.g, NK cell activation, such as by expression of NK cell markers including CD25, CD69, and/or CD 107 and/or secretion of interferon gamma; e.g., LILRB4-expressing tumor cell death viaNK-cell induced/mediated cytotoxicity). Such a measurement may be compared to a measurement made after contacting a population of cells with a control synthetic multivalent CD 16a- LILRB4 fusion non-binding domain or other control protein and/or by contacting a population of cells comprising CD 16a expressing cells and cells that do not express LILRB4, as provided in accordance with the present disclosure. The amount of NK expression (e.g, by CD25/69 and/or 107 expression) and engagement activity (e.g., measured by cell death of LILRB-4 expressing cells) can be induced or increased contacting the population of cells (before, concomitant with, or after exposure to a ligand) with a synthetic multivalent CD 16a- LILRB4 fusion binding domain (including, e.g, as compared to binding with a fusion protein comprising a control non-binding domain in place of either CD 16a or LILRB4). [00185] Cellular signaling, e.g., through a cell surface receptor such as CD16a, can be measured using any number of commercial assays. For example, a cytological assay may be used to detect and/or quantify expression of NK cell markers (e.g., CD25, CD69, CD107), cytokine secretion (e.g., IFNv levels), and/or cell death of target cells expressing LILRB4 (e.g, tumor cell death viaNK cell-mediated cytotoxicity, e.g, LILRB4 expressing tumor cells).

VIII. PHARMACEUTICAL COMPOSITIONS

[00186] Once produced, a synthetic binding proteins and fusions as provided herein can be formulated into a pharmaceutical composition.

[00187] For therapeutic use, a CD16a-LILRB4 fusion protein as provided herein is combined with a pharmaceutically acceptable carrier. Various carriers (e.g., diluents, excipients, etc.) used in formulating and preparing pharmaceutical compositions are known and/or readily accessible to those of skill in the art. Depending upon the circumstances, a carrier can include a liquid (e.g, a sterile liquid) or a solid. A carrier may be selected from or comprise water, aqueous solvents, non-aqueous solvents, dispersion media, surfactants, antioxidants, buffers, adjuvants, tonicity agents, stabilizers, bulking agents, lyoprotectants, metal ions, chelating agents, isotonic and absorption delaying agents, and the like, that are compatible with pharmaceutical administration. The use of such media and agents for pharmaceutically active substances is known in the art. Typically a carrier is approved by United States Food and Drug Administration and meets the standards of the United States Pharmacopoeia (USP), the European Pharmacopoeia (EP), the British Pharmacopoeia, and/or other International Pharmacopoeia. Suitable formulations for use in the present disclosure are found in see e.g, Adeboye Adejare, REMINGTON: THE SCIENCE AND PRACTICE OF PHARMACY (23^rd ed. 2020). For a brief review of methods for drug delivery, see, e.g., Langer (1990) SCIENCE 249:1527-1533. The resulting pharmaceutical compositions are suitable for administration to a subject (e.g, a mammal, e.g., a human).

[00188] A pharmaceutical composition may contain formulation materials for modifying, maintaining or preserving, for example, the pH, osmolarity, viscosity, clarity, color, isotonicity, odor, sterility, stability, rate of dissolution or release, adsorption or penetration of the composition. In such embodiments, suitable formulation materials include, but are not limited to, amino acids (such as glycine, glutamine, asparagine, arginine or lysine); antimicrobials; antioxidants (such as ascorbic acid, sodium sulfite or sodium hydrogen- sulfite); buffers (such as borate, bicarbonate, Tris-HCl, citrates, phosphates or other organic acids); bulking agents (such as mannitol or glycine); chelating agents (such as ethylenediamine tetraacetic acid (EDTA)); complexing agents (such as caffeine, polyvinylpyrrolidone, beta-cyclodextrin or hydroxypropyl-beta-cyclodextrin); fillers; monosaccharides; disaccharides; and other carbohydrates (such as glucose, mannose or dextrins); proteins (such as serum albumin, gelatin or immunoglobulins); coloring, flavoring and diluting agents; emulsifying agents; hydrophilic polymers (such as polyvinylpyrrolidone); low molecular weight polypeptides; salt-forming counterions (such as sodium); preservatives (such as benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid or hydrogen peroxide); solvents (such as glycerin, propylene glycol or polyethylene glycol); sugar alcohols (such as mannitol or sorbitol); suspending agents; surfactants or wetting agents (such as pluronics, polyethylene glycol (PEG), sorbitan esters, polysorbates such as polysorbate 20, polysorbate, triton, tromethamine, lecithin, cholesterol, tyloxapal); stability enhancing agents (such as sucrose or sorbitol); tonicity enhancing agents (such as alkali metal halides, preferably sodium or potassium chloride, mannitol sorbitol); delivery vehicles; diluents; excipients and/or pharmaceutical adjuvants (see e.g, Adeboye Adejare, REMINGTON: THE SCIENCE AND PRACTICE OF PHARMACY (23^rd ed. 2020)).

[00189] In certain embodiments, a pharmaceutical composition may contain a sustained- or controlled-delivery formulation. Techniques for formulating sustained- or controlled- delivery means, such as liposome carriers, bio-erodible microparticles or porous beads and depot injections, are also known to those skilled in the art. Sustained-release preparations may include, e.g., porous polymeric microparticles or semipermeable polymer matrices in the form of shaped articles, e.g, films, or microcapsules. Sustained release matrices may include polyesters, hydrogels, polylactides, copolymers of L-glutamic acid and gamma ethyl-L- glutamate, poly (2-hydroxyethyl-inethacrylate), ethylene vinyl acetate, or poly-D(-)-3- hydroxybutyric acid. Sustained release compositions may also include liposomes that can be prepared by any of several methods known in the art.

[00190] Depending upon the circumstances, a pharmaceutical composition may contain nanoparticles, or lipid droplets, e.g., polymeric nanoparticles, liposomes, or micelles (see Anselmo et al. (2016) BIOENG. TRANSL. MED. 1: 10-29). [00191] Pharmaceutical compositions containing a fusion protein in accordance with the present disclosure can be presented in a dosage unit form and can be prepared by any suitable method. A pharmaceutical composition should be formulated to be compatible with its intended route of administration. Examples of routes of administration are intravenous (IV), intraperitoneal, intradermal, inhalation, transdermal, topical, transmucosal, intrathecal and rectal administration. In certain embodiments, the synthetic peptide is administered by subcutaneous administration.

[00192] Useful formulations can be prepared by methods known in the pharmaceutical art. For example, see e.g, Adeboye Adejare, REMINGTON: THE SCIENCE AND PRACTICE OF PHARMACY (23^rd ed. 2020). Formulation components suitable for parenteral administration include a sterile diluent such as water for injection, saline solution, fixed oils, polyethylene glycols, glycerin, propylene glycol or other synthetic solvents; antibacterial agents such as benzyl alcohol or methyl parabens; antioxidants such as ascorbic acid or sodium bisulfite; chelating agents such as EDTA; buffers such as acetates, citrates or phosphates; and agents for the adjustment of tonicity such as sodium chloride or dextrose.

[00193] For intravenous administration, suitable carriers include physiological saline, bacteriostatic water, polyethoxylated castor oil or phosphate buffered saline (PBS). The carrier should be stable under the conditions of manufacture and storage, and should be preserved against microorganisms. The carrier can be a solvent or dispersion medium containing, for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyethylene glycol), and suitable mixtures thereof.

[00194] Pharmaceutical formulations preferably are sterile. Formulations can be sterilized, for example, by methods appropriate to retain activity and stability of the CD16a-LILRB4 fusion protein included therein. Sterilization can be accomplished by any suitable method, e.g, filtration through sterile filtration membranes. Where the composition is lyophilized, filter sterilization can be conducted prior to or following lyophilization and reconstitution.

[00195] Depending upon the drug substance and formulation, the resulting dosage forms can be stable for extended periods of time, such as 1 month, 3 months, 6 months, 1 year, 2 years, 3 years, or more, when the dosage form is a liquid or solid. The formulations can be stable at room temperature or higher. It is contemplated that the dosage form is stable at ambient conditions in PBS. Alternatively the dosage form is frozen (e.g, a liquid or a lyophilizate) and stable under appropriate temperatures such as, e.g, -20°C, -80°C). [00196] Depending upon the circumstances, the dosage forms can be formulated as a unit dose, which can include, for example, about 0.01 mg, about 0.05 mg, about 0.10 mg, about 0.15 mg, about 0.20 mg, about 0.25 mg, about 0.5 mg, about 0.75 mg, about 1 mg, about 1.25 mg, about 1.5 mg, about 1.75 mg, about 2 mg, about 3 mg, about 4 mg, about 5 mg, about 6 mg, about 7 mg, about 8 mg, about 9 mg, about 10 mg, about 25 mg, about 50 mg, about 100 mg, about 250 mg, about 500 mg, about 1 g, about 1.5 g, about 2.5 g, about 5 g, or about 10 g of the drug substance.

[00197] The compositions described herein may be administered locally or systemically. It is contemplated that the compositions described herein are generally administered by parenteral administration. Preparations for parenteral administration include sterile aqueous or non-aqueous solutions, suspensions, and emulsions. In certain embodiments, the pharmaceutical composition is administered subcutaneously or may be administered intravenously, e.g., via intravenous infusion. In certain embodiments, it is contemplated that the synthetic constructs disclosed herein can be administered by systemic administration.

[00198] Generally, a therapeutically effective amount of active component, for example, a CD16a-LILRB4 fusion protein disclosed herein, is in the range of 0.01 pg/kg to 100 mg/kg, e.g, 0.1 pg/kg to 25 mg/kg, 1 pg/kg to 15 mg/kg, 10 pg/kg to 10 mg/kg, 0.1 mg/kg to 100 mg/kg, 1 mg/kg to 100 mg/kg. In certain embodiments, the effective amount is 0.01 pg/mg. In certain embodiments, the effective amount is 0.1 mg/kg. In certain embodiments, the effective amount is 1 mg/kg. In certain embodiments, the effective amount is 10 mg/kg. In certain embodiments, the effective amount is 15 mg/kg. A dose may also be a flat dose, for example, about 0.25 mg to 25 mg, and depending on context (e.g., administration route such as intravenous vs. subcutaneous vs. oral), dose can change. In certain embodiments (e.g. intravenous administration) a dose may be about 0.075 mg to about 100 mg, 0.1 mg to 100 mg, 1 mg to 100 mg, 0.1 mg to 50 mg or 1 mg to 50 mg. In certain embodiments (e.g. subcutaneous administration) a dose may be about 0.25 mg to about 2.5 mg (e.g., about 0.25 mg, about 0.5 mg, about 0.75 mg, about 1 mg, about 1.25 mg, about 1.5 mg, about 2 mg, or about 2.5 mg. In other embodiments, (e.g. oral administration) a dose may be about 5 mg to about 25 mg (e.g, about 5 mg, about 7.5 mg, about 10 mg, about 12.5 mg, about 15 mg, about 20 mg, or about 25 mg). The amount administered will depend on variables such as the type and extent of disease or indication to be treated, the overall health of the patient, the in vivo potency of the active component, the pharmaceutical formulation, and the route of administration. The initial dosage can be increased beyond the upper level in order to rapidly achieve the desired blood-level or tissue-level. Alternatively, the initial dosage can be smaller than the optimum, and the daily dosage may be progressively increased during the course of treatment. Human dosage can be optimized, e.g, in a conventional Phase I dose escalation study. Dosing frequency can vary, depending on factors such as route of administration, dosage amount, serum half-life of the synthetic peptide, and the disease, disorder, or condition being treated. Exemplary dosing frequencies are once per day, once per week and once every two weeks.

IX. METHODS OF USE AND TREATMENT

[00199] The CD 16a and LILRB4 binding domains and fusions thereof as provided herein can be used in a variety of different ways.

[00200] The fusion proteins provided herein can be used in a variety of different approaches or contexts, including, for example, therapeutic and/or diagnostic contexts. For example, monovalent CD 16a and/or LILRB4 synthetic binding proteins or the CD16a-LILRB4 fusion proteins of the present disclosure can be used, along with one ore more detectable labels (e.g, a fluorescent label), to determine if a binding domain specifically binds to CD16a as compared to CD 16b or if a binding domain specifically and tightly binds to LILRB4. In addition, such labeled miniproteins can be used to determine if binding occurs to soluble antigen and/or cell-surface bound antigen (e.g, CD16a, e.g, LILRB4). Such methods may be used, for example, in diagnostic contexts or in screens identifying proteins that specifically bind a cell type (e.g, NK cells, e.g, cancer cells).

[00201] Fusion proteins of the present disclosure can be used in a method of targeting CD 16a to a cell expressing a different cell surface antigen expressing LILRB4, such as a method which comprises contacting a cell that expresses CD 16a on its cell surface with a composition comprising a synthetic CD 16a binding protein disclosed herein, wherein the CD 16a binding protein further comprises a binding domain that binds to LILRB4, and binds to a cell-surface antigen on another cell, thereby engaging the CD16a-expressing cell to target the LILRB4-expressing cell for death (e.g, via ADCC).

[00202] For example, the fusion proteins can be used in a method of targeting CD16a and/or LILRB4, to bring cells expressing each protein into proximity with one another. The method comprises contacting a cell that expresses CD16a on its cell surface with a composition comprising a fusion protein comprising a CD 16a binding domain and an LILRB4 binding domain as provided herein. CD 16a binding domains described herein can be used to modulate CD16a-positive cell activity (e.g., NK cell activity). The method comprises contacting a cell that expresses CD 16a on its cell surface with a composition comprising the synthetic CD 16a binding domain in a fusion protein. In each method, the CD 16a binding domain or the pharmaceutical composition comprising a CD 16a binding domain further comprises an LILRB4 binding domain in a fusion protein (e.g., linked by a linker, e.g, as described in TABLE 4). In each of the foregoing methods, the fusion proteins or the pharmaceutical composition can leverage CD 16a expression to (1) bind to an NK cell and (2) engage the bound NK cell to target the LILRB4-expressing cancer cell, inducing ADCC and LILRB4-expressing cancer cell death.

[00203] Fusion proteins of the present disclosure may be used in treatment of a disease, disorder, or condition mediated by CD16a-positive cell activity. CD16a is expressed on NK cells that can target LILRB4-expressing cancer cells.

[00204] In one aspect, the disclosure provides a method of targeting a CD16a-expressing cell, the method comprising contacting a cell that expresses CD 16a on its cell surface with a composition comprising a synthetic CD 16a binding domain and a synthetic LILRB4 binding domain or pharmaceutical composition as provided herein in the presence of a tumor cell that expresses LILRB4, bringing the NK cell into the proximity of the tumor cell and inducing a cytotoxic response, resulting in death of the tumor cell.

[00205] In another aspect, the disclosure provides a method of modulating immune cell activity, the method comprising contacting an immune cell that expresses CD 16a on its cell surface with a composition comprising a synthetic CD 16a binding domain and an LILRB4 binding domain or a pharmaceutical composition as provided herein, and under conditions that permit the CD 16a binding domain to bind to the CD 16a on the immune cell, and the LILRB4 binding domain to bind to a protein on a cancer cell, wherein the activity of the immune cell is modulated in that it initiates a cytotoxic response against the cancer cell.

[00206] Without wishing to be bound by theory, it is contemplated that a fusion protein comprising a CD 16a binding domain and an LILRB4 binding domain as provided by the present disclosure promotes or increases a cell-mediated activity (e.g, ADCC) in a target cell expressing a target (e.g., LILRB4) relative to cell-mediated activity in the target cell in the absence of the fusion protein with the CD 16a binding domain. [00207] In another aspect, the disclosure provides a method of treating cancer by administering to a subject in need thereof a fusion protein comprising a synthetic CD 16a binding protein or a pharmaceutical composition as provided herein, wherein the LILRB4 targets a cancer cell, wherein the administration localizes a CD16a-expressing NK cell into proximity with a cancer cell expressing LILRB4, and promoting or increasing CD 16a- mediated cytotoxicity in the cancer cell.

[00208] In another aspect, the disclosure provides a method of treating one or more cancers in a subject in need thereof, the method comprising administering to the subject an effective amount of a fusion protein comprising a CD 16a binding domain and an LILRB4 binding domain or pharmaceutical composition as provided herein.

[00209] Subjects that can be treated include those suspected as having, having, or at risk of having a disease, disorder, or condition that would benefit from targeted cell toxicity by CD 16a expressing NK cells. The methods described herein may include a step of selecting a treatment for a subject in need thereof. The method includes (a) identifying (e.g, diagnosing) the subject with such a disease, disorder, or condition, and (b) selecting a synthetic CD 16a binding domain as described herein and selecting an LILRB4 binding domain as described herein and using a fusion protein to treat the subject.

[00210] Fusion proteins administered in an effective amount to a subject in need thereof may result in CD 16a expressing NK-cell mediated death of one or more cancer cells, which cancer cells express LILRB4.

[00211] The present disclosure provides methods of treating a subject in need thereof by administering an effective amount of the fusion protein comprising a CD 16a binding domain and an LILRB4 binding domain to the subject. The methods and compositions described herein can be used alone or in combination with other therapeutic agents and/or modalities. The phrase administered “in combination,” as used herein, is understood to mean that two (or more) different treatments are delivered to the subject during the course of the subject’s affliction with the disorder, such that the effects of the treatments on the patient overlap at a point in time. In certain embodiments, the delivery of one treatment is still occurring when the delivery of the second begins, so that there is overlap in terms of administration. This is sometimes referred to herein as “simultaneous” or “concurrent delivery.” In other embodiments, the delivery of one treatment ends before the delivery of the other treatment begins. In certain embodiments of either case, the treatment is more effective because of combined administration. For example, the second treatment is more effective, e.g., an equivalent effect is seen with less of the second treatment, or the second treatment reduces symptoms to a greater extent, than would be seen if the second treatment were administered in the absence of the first treatment, or the analogous situation is seen with the first treatment. In certain embodiments, delivery is such that the reduction in a symptom, or other parameter related to the disorder is greater than what would be observed with one treatment delivered in the absence of the other. The effect of the two treatments can be partially additive, wholly additive, or greater than additive. The delivery can be such that an effect of the first treatment delivered is still detectable when the second is delivered.

[00212] It is contemplated that therapy can be accomplished using a fusion protein as part of a combination therapy wherein the fusion protein comprising a CD 16a binding domain and an LILRB4 binding domain and, optionally, one ore more additional agents, which agent(s) may be administered prior, concomitant with, and/or after treatment with a fusion protein comprising both a CD 16a and an LILRB4 binding domain.

[00213] The combination therapy comprising the fusion protein may include one or more additional agents or treatments known to those of skill in the art and may have been previously used, be already ongoing, or added to a treatment for a subject in need thereof. In certain embodiments, the one or more treatments is selected from a biological agent (e.g., biologies, gene therapy, peptides), a small molecule (e.g., chemotherapy, corticosteroids, antivirals, antibiotics, anti-inflammatory agents, etc.), one or more cells (e.g, immunotherapy), and/or one or more mechanical interventions (e.g, surgery, cryotherapy, radiation). In some embodiments, a small molecule can include, paclitaxel and cyclophosphamide. In some embodiments, a biologic can include an immune checkpoint modulator targeting antibody such as anti-PD-1 or anti-CTLA-4 antibodies. In certain embodiments, other treatments can include, but are not limited to, cell therapy such as use of ex vivo expanded and differentiated NK cells.

[00214] In methods of the present disclosure, where administration occurs, administration can be before, during, or after administration or use of one or more other treatments. In some embodiments, one or more other treatments may be a biological agent (e.g., biologies, gene therapy, peptides), a small molecule (e.g, chemotherapy, corticosteroids, antivirals, antibiotics, anti-inflammatory agents, etc.), one or more cells (e.g, immunotherapy), and/or one or more mechanical interventions (e.g, surgery, cryotherapy, radiation). By way of non- limiting example, exemplary small molecules can include, paclitaxel and cyclophosphamide, exemplary biologies can include, for example, immune checkpoint modulator targeting antibodies such as anti-PD-1 or anti-CTLA-4 antibodies, and exemplary other treatments can include, but are not limited to, cell therapy such as use of ex vivo expanded and differentiated NK cells. In certain embodiments, the additional therapy may include a combination of therapeutics of different classes.

[00215] Exemplary diseases, disorders, or conditions that may be treated with the CD16a binding proteins disclosed herein include those such tumor and/or one or more cancer-related disorders (e.g., one or more disorders in which cells express a surface antigen, such as, e.g, a tumor associated antigen, e.g, LILRB4). Exemplary disorders include, but are not limited to myeloma (e.g, multiple myeloma), acute myeloid leukemia (AML), lymphoma (e.g, mantle cell lymphoma), solid tumors, etc.

[00216] Compositions of the present disclosure may be used to treat a subject diagnosed as having or at risk of having a one or more such conditions. In some embodiments, the subject has been diagnosed as having cancer and/or a cancerous cells.

[00217] In one aspect, the disclosure provides a method of targeting a population of LILRB4-expressing cancer cells, the method comprising contacting the population with a composition comprising a fusion protein comprising a CD 16a binding domain and an LILRB4 binding domain, wherein the fusion protein binds to LILRB4, wherein after the contacting, a greater portion of the population of cancer cells is dead as compared to contacting without the fusion protein comprising a CD 16a binding domain and an LILRB4 binding domain.

[00218] In one aspect, the disclosure provides a method of treating a subject in need thereof, the method comprising administering to the subject an effective amount of the fusion protein or pharmaceutical composition comprising two binding domains, wherein a first binding domain comprises an amino acid sequence as set forth in any of Tables 1A-1D and a second binding domain, wherein the second binding domain is set forth in any of Tables 2A-2D. In some embodiments, the fusion protein is selected from Table 3. In some such embodiments, binding domains of the fusion protein comprise one or more linkers, such as those set forth in Table 4

[00219] In methods provided herein, a subject can be diagnosed as having or at risk of having a cancer or population of cancer cells. Cancer or cancer cells may be of myeloid origin, such as, for example, in myeloma (e.g., multiple myeloma), acute myeloid leukemia (AML), lymphoma (e.g., mantle cell lymphoma), solid tumors, etc.

[00220] In one aspect, the disclosure provides method of increasing cancer cell death in a population of cells comprising cancerous cells and NK cells, the method comprising exposing the population of cells to a fusion protein as provided herein, thereby to increase cancer cell death relative to cancer cell death in the absence of the fusion protein. In some embodiments, the fusion protein comprises at least one CD 16a binding domain and an LILRB4 binding domain. In some embodiments, the fusion protein comprises two CD 16a binding domain and an LILRB4 binding domain.

[00221] In one aspect, the disclosure provides a method of increasing expression of CD69, CD25, and/or CD107 on an NK cell, the method comprising contacting a CD16a-expressing NK cell in the presence of an LILRB4-expressing tumor cell with a fusion protein as provide herein, whereupon the fusion protein binds to the NK cell and the cancer cell and results in increased expression of CD69, CD25, and/or CD 107 on the NK cell relative to the expression of the CD69, CD25, and/or CD 107 prior to the contact. In some embodiments, the fusion protein comprises at least one CD 16a binding domain and an LILRB4 binding domain. In some embodiments, the fusion protein comprises two CD 16a binding domain and an LILRB4 binding domain. In some embodiments, the NK cells display increased expression of CD107.

[00222] In another aspect, the disclosure provides a method of stimulating an increase of IFNy release from a CD16a-expressing NK cell in the presence of an LILRB4-expressing cancer cell, the method comprising exposing the NK cell and the cancer cell with a fusion protein of the present disclosure, so that the fusion protein binds to the NK cell and the cancer cell and stimulates the increase of IFNy release from the NK cell. In some embodiments, the fusion protein comprises at least one CD 16a binding domain and an LILRB4 binding domain. In some embodiments, the fusion protein comprises two CD 16a binding domains and an LILRB4 binding domain.

[00223] In another aspect, the disclosure provides method of treating a subject in need thereof, the method comprising administering to the subject an effective amount of a fusion protein of the present disclosure or a pharmaceutical composition thereof, thereby to treat the cancer in the subject. In some embodiments, the fusion protein comprises at least one CD 16a binding domain and an LILRB4 binding domain. In some embodiments, the fusion protein comprises two CD 16a binding domain and an LILRB4 binding domain. In some embodiments, the subject is a mammal. In some embodiments, the mammal is a human. In some embodiments, the human is diagnosed or suspected of having cancer cells expressing LILRB4.

[00224] A subject may be evaluated, e.g, by a healthcare provider, before, during, and/or after treatment with a composition provided herein. Depending on the outcome of the evaluation, a treatment may be continued or ceased, treatment frequency or dosage may change, or the patient may be treated with a different fusion protein comprising a CD 16a binding domain and a domain that binds to a surface antigen on another cell (e.g, a cancer cell, e.g, an LILRB4-expressing cell). Subjects may be administered a composition comprising the fusion protein for a discrete period of time according to dosage paradigms described herein, including, optionally, until the disease, disorder, or condition is treated.

X. KITS

[00225] Synthetic CD16a and LILRB4 binding domains and fusions thereof (e.g, bivalent, tri valent, etc.), as provided by the present disclosure may be included as part of a kit. A kit may comprise a container comprising or consisting essentially of a unit of a pharmaceutical composition comprising a synthetic CD 16a binding domain, an LILRB4 binding domain, and/or a CD16a-LILRB4 fusion protein (e.g, bivalent, tri valent, etc.), instructions for use, and optionally, one or more agents (e.g, a buffer or diluent, if appropriate, to dissolve the binding domain or dilute a solution containing the binding domain), and a dispenser. A kit may include a label indicating the intended use of the contents of the kit. The contents of the kit may be used for treating, monitoring and/or diagnosing a subject in need thereof.

[00226] The present disclosure is further illustrated by the following examples which should not be construed as further limiting.

EXAMPLES

EXAMPLE 1: MULTIVALENT FUSION PROTEIN DESIGN

[00227] This Example describes design of synthetic multivalent fusion proteins using synthetic fusion proteins with at least one portion that binds to a receptor on the surface of an immune cell (e.g., an NK cell) and at least one portion that binds to a tumor associated antigen (e.g., LILRB4) on the surface of a cancer cell (e.g., an AML tumor cell, etc.). These fusions comprise synthetic CD 16a binding domains (FIG. 1), synthetic LILRB4 binding domains (FIG. 2), and synthetic linkers. To engineer fusion proteins that would bind NK cells and tumor cells, CD 16a and LILRB4 sequences were genetically engineered to produce a fusion protein, with the different proteins (CD 16a and LILRB4) connected using disordered linkers of varying length and amino acid composition (see, e.g., Table 4).

[00228] As shown in schematics of FIGS. 3B and 3C, bivalent fusion proteins were synthesized to have at least a first synthetic binding domain to a first target and at least a second synthetic binding domain to a second target, connected by an intermolecular linker, linking the C-terminus of the first synthetic binding domain to the N-terminus of the second synthetic binding domain. FIG. 3A demonstrates an exemplary bivalent fusion protein binding to CD 16a on an NK cell and LILRB4 on a tumor cell.

[00229] As shown in schematics of FIGs. 4A - 4E trivalent fusion proteins were synthesized to have at least two synthetic binding domains that bind to a first target and at least a second synthetic binding domain to a second target, wherein the two synthetic binding domains that bind to a first target are each linked to the second synthetic binding domain that binds to a second target by an intermolecular linker, linking the C-terminus of the first instance of the synthetic binding domain that binds to the first target to the N-terminus of the second synthetic binding domain and the C-terminus of the second synthetic binding domain to the N-terminus of the second instance of the binding domain that binds to the first target. FIG. 4B demonstrates an exemplary trivalent fusion protein with two separate domains binding to CD 16a on an NK cell and one to LILRB4 on a tumor cell, leading to NK cell activation and tumor-cell-specific toxicity and death. FIG. 4A is an exemplary trivalent fusion protein comprising two exemplary synthetic LILRB4 binding domains with an exemplary synthetic CD 16a binding domain linked to each of the C-terminus of the first LILRB4 binding domain and the N-terminus of the second LILRB4 binding domain. FIG. 4B is an exemplary trivalent fusion protein comprising two exemplary synthetic CD 16a binding domains with an exemplary synthetic LILRB4 binding domain linked to each of the C-terminus of the first CD 16a binding domain and the N-terminus of the second CD 16a binding domain. FIG. 4D is an exemplary trivalent fusion protein comprising two exemplary synthetic CD 16a binding domains with the C-terminus of the first CD 16a binding domain linked to the N-terminus of the second CD 16a binding domain, and the C-terminus of the second CD 16a binding domain linked to the N-terminus of an exemplary synthetic LILRB4 binding domain. FIG. 4E is an exemplary trivalent fusion protein comprising two exemplary synthetic LILRB4 binding domains with the C-terminus of the first LILRB-4 binding domain linked to the N-terminus of the second LILRB4 binding domain, and the C-terminus of the second LILRB4 binding domain linked to the N-terminus of an exemplary synthetic CD 16a binding domain. FIG. 4E is an exemplary trivalent fusion protein comprising one exemplary synthetic LILRB4 binding domain with the C-terminus of the LILRB-4 binding domain linked to the N-terminus of a first exemplary CD 16a binding domain, and the C-terminus of the first CD 16a binding domain linked to the N-terminus of an exemplary synthetic CD 16a binding domain.

[00230] As a starting point, validated synthetic binding domains for CD 16a (TABLE IE) and LILRB4 (TABLE 2D) were identified as candidates for use in multivalent fusion proteins. These validated synthetic binding domains are proteins that are known to bind with a certain affinity for their targets and have a certain stability under various conditions such as high heat (see, e.g., FIGs. 6A, 6B, and 10).

EXAMPLE 2: CD 16a BINDING DOMAIN VALIDATION

[00231] From 19 unique monovalent synthetic CD16a binding domains as set forth in Table IE, two (Reference Miniprotein 1 (SEQ ID NO: 1), and Reference Miniprotein 5 (SEQ ID NO: 5) were further characterized and individually tested for CD 16a binding strength and specificity.

[00232] Binding affinity and selectivity were determined using SPR analyses. For all proteins that showed specific binding to CD 16a (as determined by measurable binding to the CD 16a ectodomain protein and not to a streptavidin protein chip surface), binding affinities were determined. In parallel with binding affinity testing, each of the proteins was individually tested for the ability to specifically bind to CD 16a (and not to CD 16b). References Miniproteins 1 and 5 (SEQ ID NOs: 1 and 5) were confirmed to bind to CD 16a by SPR (SEQ ID NO: 1: k_on 3.07 e⁴ s^M’¹; k_off 3.88 e^s’¹; KD 1.26E-05 M) and folded and thermostable as shown by circular dichroism spectroscopy from about 25°C to about 95°C (FIGs. 6A and 6B). Binding characteristics for 19 tested CD 16a binding domains are shown in TABLE 5

TABLE 5. Binding Characteristics of Exemplary Reference Miniproteins

‘interaction half-life is time after initial binding at which half of the starting pool of a plurality of CD16a binding proteins remain bound to CD 16a.

[00233] Specificity of binding to CD 16a over CD 16b was also confirmed. FIG. 7A shows one schematic representation indicating certain structural differences in key binding areas in CD 16a (which has a G147 and Y158) as compared to CD16b (which has a D147 and H158). Exemplary synthetic CD 16a binding domains bind to CD 16a with high affinity and do not bind to CD16b. FIGs. 7B and 7C are line graphs showing affinity measurement by surface plasmon resonance (SPR) of Reference Miniprotein 5 (SEQ ID NO: 5), demonstrating that CD 16a binding domains of the present disclosure are extremely selective for CD 16a as compared to CD 16b, as shown by their binding to two different variants of CD 16a, 176V and 176F (FIG. 7B) as compared to two variants of CD16b, NA1 and NA2 (FIG. 7C). Binding affinities of both Reference Miniprotein 1 (SEQ ID NO: 1) (FIG. 7D and TABLE 5) and Reference Miniprotein 2 (SEQ ID NO: 2) (TABLE 5) to CD16a were stronger than 1 pM. As seen in FIG. 7E, Reference Miniprotein 1 (SEQ ID NO: 1) showed specific binding to CD 16a as determined by measurable binding to the CD 16a ectodomain protein (left panel) and not to CD16b (right panel) or the streptavidin protein chip surface used for testing (data not shown). RU = response units.

[00234] In addition, a binding assay showed that CD 16a binding domains of the present disclosure are not be compromised in the presence of human serum albumin. That is, as shown in FIG. 8A, an exemplary CD 16a binder was immobilized on an SPR chip and binding of CD 16a was measured in the presence and absence of human serum. As illustrated in the schematic of FIG. 8B, this assay showed that binding of synthetic CD 16a proteins of the present disclosure do not compete with binding of antibodies in human serum through the Fc region of CD 16a.

[00235] In addition, as shown in FIG. 9, a cell binding assay using primary natural killer (NK) cells from healthy donors were incubated with an exemplary synthetic CD16a binding domain (Reference Miniprotein 5 SEQ ID NO: 5) that had been flag-tagged and then identified with a detectable fluorescent marker (APC). APC levels were determined and, as shown, in the absence of a synthetic CD 16a binding domain, no fluorescence was detected, whereas in presence of an exemplary CD 16a binding domain, fluorescent signal, indicating binding, was detected.

[00236] This example also describes use of single-site saturation mutagenesis (SSM) for identification and interrogation binding tolerance at each of 46 amino acid positions in an exemplary CD 16a Miniprotein.

[00237] Reference Miniprotein 5 (SEQ ID NO: 5) was selected according to its high affinity for CD 16a. To better understand the sequence requirements for binding At each of the 46 linear amino acid positions, from N-terminus to C-terminus, single amino acid substitutions were made. For example, at position 1, Arginine (R) was substituted with alanine, arginine, asparagine, aspartic acid, glutamine, glutamic acid, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, or valine. The binding affinity (KD) for CD 16a was assessed using yeast surface display. This process was continued for each of the remaining 45 amino acids in SEQ ID NO: 5, and a map of substitutions tolerances was developed as shown in TABLE ID. To be considered an allowable residue, binding of the mutated miniproteins had to occur at an affinity 120 nM or stronger. As shown in TABLE 5, there were allowable residues at positions other than 29, 32, and 41, which were immutable in that they could not be varied (to another amino acid residue) and still maintain binding at an affinity of 120 nM or stronger.

[00238] Affinity measurements (KD) were determined using titration of soluble CD 16a used to label yeast cells expressing individual substitutions of Reference Miniprotein 5 (SEQ ID NO: 5) generated as described in herein. The binding signal at 0 pM, 244 pM, 488 pM, 976 pM, 1.95 nM, 3.91 nM, 7.81 nM, 15.6 nM, 31.3 nM, 62.5 nM, 125 nM, 250 nM, 500 nM, and 1 pM of CD16a was measured. The Hill equation (Formula III),

(Formula III) where /7//) is the Hill coefficient, Ko.s is the half-saturation constant, Y is the output response, I is the input concentration. The goodness of the curve fit was assessed using the R² value of the fit equation and collected only those binding affinities with an R² greater than 0.95. In total, measurable KDs were obtained for 374 sequence variants (which variants can be envisaged according to and substitutions at least as in TABLES 1A, IB, 1C, and ID)) that are all one amino acid substitution different from the amino acid sequence of Reference Miniprotein 5 (SEQ ID NO: 5). The structure of Reference Miniprotein 5 (SEQ ID NO: 5) was predicted with AlphaFold2 and used to identify solvent accessible surface residues. These residues were then analyzed within the context of single-site-mutational data to assess mutational tolerance in terms of binding affinity to CD 16a.

[00239] Paratope residues were identified at positions corresponding to amino acid residues relative to SEQ ID NO: 5, which is a 46-mer, namely at positions 28, 29, 32, 39, 41, and 42. Certain paratope residues were found to be immutable in that they were completely intolerant to substitution without loss of function (see, e.g., TABLE 6).

[00240] The amino acid corresponding to position 28 of SEQ ID NO: 5 (with residues numbered linearly starting with 1 at the N-terminal amino acid, through 46 as the C-terminal amino acid) was 100% conserved in SEQ ID NOs: 1-19 and using SSM analysis of SEQ ID NO: 5, could only tolerate substitution to leucine (L) without materially decreasing CD16a binding potency. The amino acid corresponding to position 29 of SEQ ID NO: 5 (with residues numbered linearly starting with 1 at the N-terminal amino acid, through 46 as the C- terminal amino acid) was 100% conserved in SEQ ID NOs: 1-19 and using SSM analysis of SEQ ID NO: 5, could only tolerate substitution to glutamic acid (E) without materially decreasing CD 16a binding potency. The amino acid corresponding to position 32 of SEQ ID NO: 5 (with residues numbered linearly starting with 1 at the N-terminal amino acid, through 46 as the C-terminal amino acid) was 100% conserved in SEQ ID NOs: 1-19 and using SSM analysis of SEQ ID NO: 5 could not tolerate any substitutions without materially decreasing CD16a binding potency. The amino acid corresponding to position 39 of SEQ ID NO: 5 (with residues numbered linearly starting with 1 at the N-terminal amino acid, through 46 as the C-terminal amino acid) was either a glutamine (Q) or threonine (T) in each of SEQ ID NOs: 1-19 and using SSM analysis of SEQ ID NO: 5, which has a T at position 39, could only tolerate substitution to valine (V) without materially decreasing CD 16a binding potency. The amino acid corresponding to position 41 of SEQ ID NO: 5 (with residues numbered linearly starting with 1 at the N-terminal amino acid, through 46 as the C-terminal amino acid) was 100% conserved in SEQ ID NOs: 1-19 and using SSM analysis of SEQ ID NO: 5, could only tolerate substitution to leucine (L) without materially decreasing CD 16a binding potency. The amino acid corresponding to position 42 of SEQ ID NO: 5 (with residues numbered linearly starting with 1 at the N-terminal amino acid, through 46 as the C-terminal amino acid) was either a glutamine (Q), histidine (H), or arginine (R) in each of SEQ ID NOs: 1-19. Using SSM analysis of SEQ ID NO: 5, which is an R at position 42, the protein could only tolerate substitution to glutamine (Q) without materially decreasing CD 16a binding potency.

[00241] A paratope of a CD 16a binding protein of the present disclosure can be represented as X28X29X32X39X41X42, where X28 is V, X29 is D, X32 is D, X39 is T, X41 is I, and X42 is R. Without losing binding potency as measured by binding at 120 nM or stronger, X28 may also be L, X39 may also be V, X41 may also be L, and X42 may be Q. Neither X29 nor X32 can be substituted with other amino acids.

TABLE 6. Paratope Residues and Tolerances

* Amino acid identities at positions in SEQ ID NO: 5 Paratope was determined using SSM data with Reference Miniprotein 5 (SEQ ID NO: 5) as a starting sequence; position refers to linear, ordinal position with amino acid position 1 being the N- terminal amino acid in SEQ ID NO: 5 and amino acid 46 being the C-terminal amino acid in SEQ ID NO: 5.

[00242] Using the characterization described herein, an amino acid sequence of an exemplary monovalent CD 16a binding domain (Reference Mini protein 5; SEQ ID NO: 5) was selected for inclusion into a fusion protein with LILRB4.

EXAMPLE 3: LILRB4 BINDING DOMAIN VALIDATION

[00243] Six unique monovalent synthetic binding domains as shown in TABLE 2D, were individually tested for LILRB4 binding (see TABLE 7). .

TABLE 7. Binding Characteristics of Exemplary LILRB4 Binding Domains

[00244] Reference Miniprotein 47 (SEQ ID NO: 47) was confirmed to bind to LILRB4 by SPR (k_a 7.5 x 10⁵ s^M’¹; ka 6.38 x IO’⁰⁴ s’¹; KD 8.52 x 10’¹⁰M) and folded and thermostable as shown by circular dichroism spectroscopy from about 25°C to about 95°C (FIG. 10). [00245] In vitro binding assays were used to determine specificity of binding to LILRB4 (see FIGs. 11A-11D). FIGs. 11A and 11B show binding of an exemplary monovalent LILRB4 binding domain to OCI-AML3 cells (which express LILRB4 on their surfaces). FIG. 11A shows a comparison of unstained cells and those combined with an anti-LILRB4 antibody and FIG. 11B shows cells stained with secondary antibody only and those stained with a flag-tagged exemplary synthetic LILRB4 binding domain, indicating that the synthetic LILRB4 binding domains of the present disclosure can specifically and strongly bind to LILRB4 at least as well and, if not, better, than an anti-LILRB4 antibody. FIGs. 11C and 11D also show binding of an exemplary monovalent LILRB4 binding domain, but to MV-4- 11 cells (which express LILRB4 on their surfaces).

EXAMPLE 4: MULTIVALENT FUSION PROTEIN CHARACTERIZATION

[00246] This Example describes characterization of exemplary bivalent and trivalent fusion proteins comprising binding domains to CD 16a, LILRB4, and intermolecular linkers, developed from Reference Miniproteins characterized in Examples 2 and 3.

[00247] Several bivalent and trivalent constructs were designed and synthesized, as shown in TABLE 3. Binding characteristics of these Reference Miniproteins to human CD 16a and LILRB4 were measured and binding affinity and selectivity/specificity of each mini protein was determined using SPR (see data in TABLES 5 and 7 in Examples 2 and 3, respectively).

Stability

[00248] Thermostability of the fusion binding miniproteins was also analyzed. An exemplary trivalent fusion protein was exposed to 4M urea at temperatures of at least 50°C (FIG. 12) These proteins retained their properly folded three-dimensional structure as confirmed by circular dichroism spectroscopy. FIG. 12 (Reference Miniprotein 53, SEQ ID NO: 53) shows exemplary CD spectra of an exemplary trivalent fusion protein in 5°C intervals between 25°C - 95°C, demonstrating that trivalent fusions that bind selectively to CD 16a and that bind to LILRB4 retain their folding and binding properties (not all data points between 25°C - 95°C are shown in FIG. 12 for clarity, but all were measured). In addition, the miniproteins also retained its properly folded three-dimensional structure after heating to 95°C as confirmed by circular dichroism spectroscopy (but without urea exposure) (data not shown). Binding

[00249] An exemplary synthetic trivalent CD16a-LILRB4-CD16a fusion binding domain (Reference Miniprotein 58 (SEQ ID NO: 58)) was tested for ability to bind to cells expressing LILRB4 and CD 16a. FIGs. 13A and 13B show that the fusion protein was able to bind cells expressing LILRB4 and CD 16a.

Functionality

[00250] Cytotoxicity assays were established to detect immune cell activation and tumor cell killing, in vitro. A co-culture of tumor and primary (healthy donor-derived) NK cells was established in the presence of described multivalent proteins at different concentrations. After 16-20 hours incubation at 37°C, cells were stained with different fluorescent antibodies and phenotyped via flow cytometry. Cells were stained with fluorescent markers to distinguish tumor cells from NK cells. Dyes that distinguish dead from live cells were also used. Percentage of tumor cells was determined via flow cytometry. Live NK cells were further phenotyped using CD25 and CD69 fluorescent antibodies. Cells that expressed both markers were considered activated NK cells.

[00251] These cytotoxicity assays were performed using various bivalent and trivalent fusion constructs. The assay was initiated by co-culturing healthy donor-derived primary NK cells and indicated tumor cell lines in the presence of various exemplary synthetic fusion proteins for 16-20 hours. Exemplary bivalent and trivalent CD16a-LILRB4 fusion binding domains (Reference Miniproteins 53, 54, 58 (SEQ ID NOs: 53, 54, 58)) were tested for ability to kill tumor cells (FIG. 14A) and activate NK cells (FIG. 14B). FIG. 14A shows the percentage of dead tumor cells (y-axis; OCI-AML3 tumor cells) after co-culture with healthy donor NK cells in presence of one of three exemplary fusion proteins: (i) a trivalent CD 16a- pl6-LILRB4-pl6-CD16a binding domain; (ii) a bivalent LILRB4-pl6-CD16a binding domain; and (iii) a bivalent CD16a-pl6-LILRB4 binding domain at increasing concentrations (x-axis, nM). The results from this study indicated that two copies of CD 16a binding domain in a trivalent fusion protein resulted in more dead tumor cells than one copy of CD 16a in a bivalent fusion protein as indicated by increase in dead tumor cells in conditions treated with the trivalent fusion protein.

[00252] In addition, as shown in FIG. 14B expression of NK cell markers CD69 and CD25 increased (as shown by percentage of CD69/CD25-positive primary NK cells; y-axis) after co-culture of healthy donor NK cells with tumor cells (OCI-AML3) in presence of one of three exemplary fusion proteins: (i) a trivalent CD16a-pl6-LILRB4-pl6-CD16a fusion protein; (ii) a bivalent LILRB4-pl6-CD16a fusion protein; and (iii) a bivalent CD16a-pl6- LILRB4 fusion protein at increasing concentrations (x-axis, nM). The results from this study also indicate that two copies of CD 16a (in a trivalent fusion protein) is at least as good, if not better, than one (in a bivalent fusion protein) as indicated by percentage of cells expressing NK phenotypic markers.

[00253] Furthermore, quantitative data support the patterns seen in FIGs. 13A and 13B. The EC50 for the trivalent construct (CD16a-pl6-LILRB4-pl6-CD16a) was 35.94 pM, while the EC50 for LILRB4-pl6-CD 16a fusion protein was 203.1 pM and for CD16a-pl6-LILRB4, was 172.6 pM.

[00254] Additional trivalent construct orientations were tested for percent of dead tumor cells (FIG. 15A) and NK cell markers/activation (FIGs. 15B). As shown in FIGs. 15A and 15B, three trivalent constructs (i) CD16a-pl6-LILRB4-pl6-CD16a (SEQ ID NO: 58); (ii) LILRB4-pl6-CD16A-gl2-CD16a (SEQ ID NO: 62); and (iii) CD16a-gl2-CD16a-pl6- LILRB4 (SEQ ID NO: 61) each caused dose-dependent death in tumor cells, and while some constructs were more potent and induced higher numbers of tumor cell death at lower concentrations, all three trivalent constructs were able to kill almost all tumor cells by 0.1 nM of fusion protein. In the NK cell assay (FIG. 15B), the trivalent construct with two copies of CD 16a flanking one copy of LILRB4 showed the highest percentage of CD69/CD25 positive cells at all concentrations tested.

[00255] In addition, AML blasts were isolated from three different human subjects and treated with either the trivalent fusion, CD16a-pl6-LILRB4-pl6-CD16a (SEQ ID NO: 58), a control trivalent fusion protein, or no fusion protein at all. In two of the three subjects, at a concentration of 10 nM, the trivalent fusion protein CD16a-pl6-LILRB4-pl6-CD16a (SEQ ID NO: 58) induced surface expression of CD107, as a marker of activated NK cells, as compared to the fusion control or no fusion conditions (data not shown). It is contemplated that the sample from the subject in which CD 107 expression was not induced could be due to (i) lower overall LILRB4 expression; and/or (ii) increased number of dead AML blast cells, which could indicate greater cell sensitivity and cell death prior to NK cell engagement and activation. Based on the data, construct (i), CD16a-pl6-LILRB4-pl6-CD16a (SEQ ID NO: 58) was chosen for further characterization. EXAMPLE 5: LINKER LENGTH OF TRIVALENT FUSION BINDING DOMAINS

[00256] This example describes characterization of trivalent fusion protein function when binding domains are attached using linkers of different lengths, from five amino acids up to 32 amino acids. After CD16a-pl6-LILRB4-pl6-CD16a (SEQ ID NO: 58) was chosen for further characterization, different linker compositions and lengths were tested. The following trivalent fusion proteins were tested: (i) a trivalent CD16a-p5-LILRB4-p5-CD16a (Reference Fusion Protein 55 (SEQ ID NO: 55)); (ii) CD16a-p8-LILRB4-p8-CD16a (Reference Fusion Protein 56 (SEQ ID NO: 56)); (iii) CD16a-pl2-LILRB4-pl2-CD16a (Reference Fusion Protein 57 (SEQ ID NO: 57)); (iv)CD16a-p!6-LILRB4-pl6-CD16a (Reference Fusion Protein 58 (SEQ ID NO: 58)); (v) CD16a-p24-LILRB4-p24-CD16a (Reference Fusion Protein 59 (SEQ ID NO: 59)); and (vi) CD16a-p32-LILRB4-p32-CD16a (Reference Fusion Protein 60 (SEQ ID NO: 60)). FIGs. 16A and 16B show percent dead tumor cells (y-axis; 16A) and CD69/CD25 -positive primary NK cells (y-axis; 16B) after co-culture of tumor cells (OCI- AML-3 and primary NK cells (from healthy donors) in the presence of one of six exemplary trivalent fusion proteins each having different linker lengths compared to the others and at different concentrations (x-axis; nM). A dose-dependent increase in tumor cell death was observed, but fusion proteins with longer linkers appeared to be more potent (see FIG. 16A), and have a greater ability to activate NK cells (see FIG. 16B).

EXAMPLE 6: INDUCTION OF TUMOR CELL DEATH AND NK CELL ACTIVATION BY TRIVALENT FUSION PROTEINS

[00257] This example characterizes ability of fusion proteins to activate NK cells and kill tumor cells. CD16a-pl6-LILRB4-pl6-CD16a (SEQ ID NO: 58) was also tested against two fusion proteins in which each of the CD 16a or LILRB4 components was replaced with a nonbinding domain. Tumor cell death (FIGs. 17A-17C and 18A-18C) and NK cell phenotype were measured (FIGs. 19A-19C and 20A-20C). The non-binding domain is an inert control with similar size and structure as CD 16a or LILRB4 binding domains but does not bind to either target. FIGs. 17A-17C and 18A-18C show that tumor cells are only killed in the presence of the trivalent fusion protein with CD 16a and LILRB4 components and not in the presence of constructs that cannot engage NK cells to kill LILRB4 expressing tumor cells, either by failure to bind to CD 16a or failure to bind to LILRB4 (due to lacking CD 16a or LILRB4 binding domains that had been replaced with non-binding domains). FIGs. 17A- 17C and 18A-18C show percent of dead tumor cells (y-axis) in two LILRB-4 expressing cell lines (FIG. 17A and 19A: 0CI-AML3; and FIG. 17B, and 18B: MV-4-11 (AML)) and an LILRB4-negative control (FIG. 17C, and 19C: Raji (BL)) after co-culture with primary NK cells from a healthy donor (NK donor 1 (FIGs. 19A-19C); NK donor 2 (FIGs. 20A-20C) in presence of one of three exemplary fusion proteins: (i) atrivalent non-binding domain-pl6- LILRB4-pl6-non-binding domain; (ii) a trivalent CD16a-pl6-Scaffold-pl6-CD16a binding domain; and (iii) atrivalent CD16a-pl6-LILRB4-pl6-CD16a binding domain (SEQ ID NO: 58), where the non-binding domain is an inert control with similar size and structure as CD 16a or LILRB4 binding domains but does not bind to either target (CD 16a or LILRB4), each at increasing concentrations (x-axis, nM). The EC50 for the trivalent construct in OCI- AML3 cells was 26.65 pM (FIG. 17A) and 10.7 pM (FIG. 18A) and for MV-4-11 cells, 74.31 pM (FIG. 17B) and 18.1 pM (FIG. 18B). These results show that trivalent fusion proteins with ability to bind CD 16a on NK cells and LILRB4 on tumor cells triggers antitumor activity by (i) increasing tumor cell death. NK cell phenotype, as evidenced by CD69/CD25-positivity was also only observed with the trivalent fusion that binds CD16a on NK cells and LILRB4 on tumor cells (FIGs. 19A and 20A (OCI-AML3); 19B and 20B(MV- 4-11); and 19C and 20C (negative control). CD16a-positive NK-dependent tumor cell cytotoxicity is specific to cell lines expressing LILRB4 on their surfaces

EXAMPLE 7: CONFIRMATION OF PRO-INFLAMMATORY ACTIVITY BY TRIVALENT FUSION PROTEINS

[00258] This example describes an assay for measuring cytokine secretion by activated NK cells (interferon gamma - IFNy). IFN-y secretion indicates a pro-inflammatory response. Here, three types of cells (two LILRB4 expressing tumor cell lines and a negative control line) were treated with atrivalent fusion construct for 16-20 hours (SEQ ID NO: 58) and control miniproteins. IFNy secretion was detected in supernatants of cells using an IFNy antibody and a Luminex instrument, and were quantified using an IFNy standard curve.

[00259] As shown in FIGs. 21A-21C, an NK-cell-mediated pro-inflammatory response by secretion of IFNy is elicited in LILRB4-positive tumor cells. FIGs. 21A-21C show concentration of IFNy (y-axis; pg/mL) in two LILRB-4 expressing cell lines (FIG. 21A: OCI-AML3; and FIG. 21B: MV-4-11 (AML)) and an LILRB4-negative control (FIG. 21C: Raji (BL)) after co-culture with healthy donor primary NK cells in presence of one of three exemplary trivalent fusion proteins: (i) anon-binding domain-pl 6-LILRB4-pl6-non-binding domain; (ii) a CD16a-pl6-non-binding domain-pl6-CD16a binding domain; and (iii) a CD16a-pl6-LILRB4-pl6-CD16a binding domain (SEQ ID NO: 58), where the non-binding domain is an inert control with similar size and structure as CD16a or LILRB4 binding domains but does not bind to either target, each at increasing concentrations (x-axis, nM).

[00260] These results confirm that the trivalent CD16a-pl6-LILRB4-pl6-CD16a binding domain (SEQ ID NO: 58) is able to specifically activate NK cells.

EXAMPLE 8: IN VIVO TRIVALENT FUSION PROTEIN TREATMENT INDUCED TUMOR CYTOTOXICITY AND CONTROLLED DISSEMINATED AML TUMOR CELLS IN A MOUSE XENOGRAFT MODEL

[00261] This example shows evaluation of in vivo performance after treatment with a fusion protein and NK cells (or one or more controls) on tumor cells in a mouse xenograft model. Xenograft models were generated using NSG-Tg (IL-15) mice, which express human IL-15 in combination with the highly immunodeficient NOD-scid-gamma (NSG) mouse. MV -4-11 cells were used to generate a disseminated tumor model in these mice. As shown in the experimental outline of FIG. 22A, mice were treated with tumor cells at day zero, and at day 3, primary NK cells were introduced (isolated from healthy human donors) to an NK-only cell group and to an NK + fusion (SEQ ID NO: 58) cell group. At days 4-8 and 10-14, two groups of mice (fusion only and fusion + NK cells) were treated with the trivalent CD 16a- pl6-LILRB4-pl6-CD16a fusion (SEQ ID NO: 58). Bioluminescence was measured (total flux in p/s) at days 4, 7, 10, 14, and 17.

[00262] As seen in FIG. 22B, the CD16a-pl6-LILRB4-pl6-CD16a fusion (SEQ ID NO: 58) was able to successfully control disseminated AML tumor for at least 17 days in this mouse xenograft model, but only when NK cells and the trivalent fusion protein were present together. Neither the fusion protein alone, nor NK cells alone were sufficient to control the tumor cells, indicating that the fusion protein engages NK cells to target and kill tumor cells as designed.

EXAMPLE 9: TRIVALENT FUSION PROTEIN TREATMENT INDUCED NK SERIAL KILLING OF TUMOR CELLS

[00263] This example evaluates serial killing activity of NK cells in the presence of a trivalent fusion miniprotein. In vitro serial killing cytotoxicity assays were conducted by coculturing primary NK cells, GFP-labeled OCI-AML3 tumor cells, and a trivalent fusion miniprotein in the presence of 50 U/ml IL-2. Primary NK cells were isolated from blood samples of healthy human donors. The 0CI-AML3 cell line was purchased from ATCC and transduced with a GFP lentivirus.

[00264] Briefly, 15,000 GFP-labeled OCI-AML3 cells were plated on a 96-well plate. Subsequently, 10 nM of a trivalent CD 16a-pl6-LILRB4-pl6-CD 16a binding domain (Reference Protein 58 (SEQ ID NO: 58) denoted as “COC” in legend for FIGs. 23A-23D) and 45,000 NK cells were added. After 48 hours (one serial killing round), all tumor cells were killed by the NK cells. The NK cells were then challenged with fresh tumor cells in the presence of the trivalent CD16a-pl6-LILRB4-pl6-CD16a binding domain (SEQ ID NO: 58). Tumor cell proliferation was monitored by detecting the GFP signal over time using an Incucyte S3 imaging system.

[00265] As seen in FIGs. 23A - 23D, the trivalent fusion miniprotein enabled NK cells to sustain serial killing for at least four rounds. Tumor cell proliferation was normalized at time 0 and plotted over time. Control samples included tumor cells alone (denoted as “tumor only” in legend for FIGs. 23A-23D) and tumor cells with NK cells without any miniprotein (denoted as “no protein” in legend for FIGs. 23A-23D). The differences in proliferation rate between the samples with just tumor cells and those with tumor and NK cells correspond to the intrinsic cytotoxic activity of NK cells.

INCORPORATION BY REFERENCE

[00266] All publications and patents cited throughout the text of this specification (including all patents, patent applications, scientific publications, manufacturer's specifications, instructions, etc.), whether supra or infra, including International Application Nos. PCT/US2025/019796 and PCT/US2025/019794, filed on March 13, 2025, are hereby incorporated by reference in their entirety for all purposes. To the extent the material incorporated by reference contradicts or is inconsistent with this specification, the specification will supersede any such material.

EQUIVALENTS

[00267] The invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The foregoing embodiments are therefore to be considered in all respects illustrative rather than limiting on the invention described herein. Scope of the invention is thus indicated by the appended claims rather than by the foregoing description, and all changes that come within the meaning and range of equivalency of the claims are intended to be embraced therein.

TABLE 8. Table of Sequences

Claims

CLAIMS What is Claimed is:

1. A fusion protein comprising a first binding domain that binds a first target attached by at least one linker to a second binding domain that binds to a second target, wherein the first and second binding domains are synthetic binding domains comprising synthetic binding proteins that each have an N-terminal amino acid residue and a C-terminal amino acid residue, and wherein the first target is CD 16a and the second target is LILRB4.

2. The fusion protein of claim 1, wherein the first and second binding domains are linked by a first linker.

3. The fusion protein of claim 1 or 2, wherein the C-terminal amino acid residue of the first binding domain is linked to the N-terminal amino acid residue of the second binding domain.

4. The fusion protein of claim 1 or 2, wherein the C-terminal amino acid residue of the second binding domain is linked to the N-terminal amino acid residue of the first binding domain.

5. The fusion protein of any one of claims 1-4, further comprising a third binding domain comprising an N-terminal amino acid residue and a C-terminal amino acid residue.

6. The fusion protein of claim 5, wherein the third binding domain is a synthetic binding protein that binds CD 16a.

7. The fusion protein of claim 5 or 6, wherein the N-terminal amino acid residue of the third binding domain is attached to the C-terminal amino acid residue of the second binding domain.

8. The fusion protein of claim 5 or 6, wherein the N-terminal amino acid residue of the third binding domain is attached to the C-terminal residue of the first binding domain.

9. The fusion protein of claim 5 or 6, wherein the C-terminal amino acid residue of the third binding domain is attached to the N-terminal amino acid of the first binding domain.

10. The fusion protein of any one of claims 1-9, wherein the first linker comprises an amino acid sequence selected from any of SEQ ID NOs: 64-85 and/or Table 4.

11. The fusion protein of any one of claims 1-10, wherein the third binding domain is linked to the first binding domain or the second binding domain by a second linker.

12. The fusion protein of claim 11, wherein the second linker comprises an amino acid sequence selected from any of SEQ ID NOs: 64-85 and/or Table 4.

13. The fusion protein of any one of claims 1-12, wherein the first linker and the second linker are different.

14. The fusion protein of any one of claims 1-13, wherein the amino acid sequence of the first binding domain and/or the third binding domain comprises at least one amino acid sequence selected from any of SEQ ID NOs: 20-42, 88-90, and/or as set forth in any of Tables 1A, IB, 1C, and/or ID.

15. The fusion protein of any one of claims 1-14, wherein the amino acid sequence of the first binding domain and/or the third binding domain comprises at least one amino acid sequence selected from any of SEQ ID NOs: 1-19 and/or as set forth in Table IE.

16. The fusion protein of any one of claims 1-15, wherein the amino acid sequence of the second binding domain comprises at least one amino acid sequence selected from any of SEQ ID NOs: 44-46, 86-87, Table 2A and/or Table 2B.

17. The fusion protein of any one of claims 1-16, wherein the amino acid sequence of the second binding domain comprises at least one amino acid sequence selected from any of SEQ ID NOs: 43, 47-52, Table 2C and/or Table 2D.

18. The fusion protein of any one of claims 1-17, wherein the fusion protein comprises an amino acid sequence selected from any of SEQ ID NOs: 53-63 and/or Table 3.

19. The fusion protein of any one of claims 1-18, wherein the first and/or third binding domains bind CD 16a with a binding affinity stronger than about 1 pM to about 0.001 nM; about 1 pM to about 0.01 nM; about 1 pM to about 0.75 nM; about 1 pM to about 0.5 nM; about 1 pM to about 0.25 nM; about 1 pM to about 1 nM; about 0.75 pM to about 1 nM, about 0.5 pM to about 1 nM; about 0.25 pM to about 1 nM; about 0.10 pM to about 1 nM; about 75 nM to about 1 nM; about 50 nM to about 1 nM; about 25 nM to about 1 nM; about 10 nM to about 1 nM; and about 5 nM to about 1 nM.

20. The fusion protein of any one of claims 1-19, wherein the first and/or third binding domains bind CD 16a with a binding affinity stronger than about 1 pM, about 0.75 pM, about 0.5 pM, about 0.25 pM, about 0.1 pM, about 75 nM, about 50 nM, about 25 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4 nM, about 3 nM, about 2 nM, about 1 nM, about 0.75 nM, about 0.5 nM, about 0.25 nM, about 0.1 nM to about 0.01 nM, and about 0.01 nM to about 0.001 nM.

21. The fusion protein of any one of claims 1-20, wherein the second binding domain binds LILRB4 with a binding affinity stronger than about 1 pM to about 0.001 nM; about 1 pM to about 0.01 nM; about 1 pM to about 0.75 nM; about 1 pM to about 0.5 nM; about 1 pM to about 0.25 nM; about 1 pM to about 1 nM; about 0.75 pM to about 1 nM; about 0.5 pM to about 1 nM; about 0.25 pM to about 1 nM; about 0.10 pM to about 1 nM; about 75 nM to about 1 nM; about 50 nM to about 1 nM; about 25 nM to about 1 nM; about 10 nM to about 1 nM; and about 5 nM to about 1 nM.

22. The fusion protein of any one of claims 1-21, wherein the second binding domain binds LILRB4 with a binding affinity stronger than about 1 pM, about 0.75 pM, about 0.5 pM, about 0.25 pM, about 0.1 pM, about 75 nM, about 50 nM, about 25 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4 nM, about 3 nM, about 2 nM, about 1 nM, about 0.75 nM, about 0.5 nM, about 0.25 nM, about 0.1 nM to about 0.01 nM, and about 0.01 nM to about 0.001 nM.

23. A nucleic acid encoding the fusion protein of any one of claims 1-22.

24. A host cell comprising the nucleic acid of claim 23.

25. A pharmaceutical composition comprising the synthetic fusion protein of any one of claims 1-22 and a pharmaceutically acceptable carrier.

26. The pharmaceutical composition of claim 25, formulated for administration by a systemic route.

27. The pharmaceutical composition of claim 26, wherein the systemic route is intravenous administration.

28. The pharmaceutical composition of claim 27, wherein the administration is before, concomitant with, or after administration of at least one other treatment.

29. The pharmaceutical composition of claim 28, wherein the at least one other treatment is selected from a biological agent (e.g., biologies, gene therapy, peptides), a small molecule (e.g, chemotherapy, corticosteroids, antivirals, antibiotics, anti-inflammatory agents, etc.), one or more cells e.g., immunotherapy), and/or one or more mechanical interventions (e.g., surgery, cryotherapy, radiation).

30. A method of treating a subject in need thereof, the method comprising administering to the subject an effective amount of the fusion protein of any one of claims 1-22 or the pharmaceutical composition of any one of claims 25-29.

31. A method of treating cancer, the method comprising administering to a subject in need thereof an effective amount of the fusion protein of any one of claims 1-22 or the pharmaceutical composition of any one of claims 25-29.

32. The method of claim 30 or 31, wherein the subject is diagnosed as having a cancer or a population of cancerous cells.

33. The method of claim 32, wherein the cancer or cancerous cells are of myeloid origin.

34. The method of claim 32, wherein the cancer or cancerous cells are diagnosed as acute myeloid leukemia (AML), myeloma (e.g., multiple myeloma), lymphoma (e.g., mantle cell lymphoma), or from a solid tumor origin.

35. The method of any one of claims 32-34, wherein the cancer or the population of cancerous cells comprises cancer cells that express LILRB4.

36. A method of increasing tumor cell death in a population of cells comprising cancer cells and NK cells, the method comprising exposing the population of cells to the fusion protein of any one of claims 1-22 or the pharmaceutical composition of any one of claims 25-29, thereby to increase cancer cell death relative to cancer cell death in the absence of the fusion protein or the pharmaceutical composition.

37. A method of increasing expression of CD69, CD25, and/or CD107 on an NK cell, the method comprising contacting a CD16a-expressing NK cell in the presence of an LILRB4- expressing tumor cell with the fusion protein of any one of claims 1-22, whereupon the fusion protein binds to the NK cell and the cancer cell and results in increased expression of CD69, CD25, and/or CD107 on the NK cell relative to the expression of the CD69, CD25, and/or CD 107 prior to the contact.

38. The method of claim 37, wherein the NK cells display increased expression of CD 107.

39. A method of stimulating an increase of IFNy release from a CD16a-expressing NK cell in the presence of an LILRB4 expressing cancer cell, the method comprising exposing the NK cell and the cancer cell to the fusion protein of any one of claims 1-22 so that the fusion protein binds to the NK cell and the cancer cell and stimulates the increase of IFNv release from the NK cell.

40. A method of treating a cancer in a subject in need thereof, the method comprising administering to the subject an effective amount of the fusion protein of any one of claims 1- 22 or the pharmaceutical composition of any one of claims 25-29, thereby to treat the cancer in the subject.

41. The method of claim 40, wherein the subject is a mammal.

42. The method of claim 41, wherein the mammal is a human.

43. The method of claim 42, wherein the human is diagnosed or suspected of having cancer cells expressing LILRB4.