Attorney Docket No.01218-0029-00PCT Novel fusion protein specific for CD137 and CD228 I. CROSS-REFERENCE TO RELATED APPLICATIONS [0001] This application claims the benefit of priority of US Provisional Application No. 63/408,634, filed September 21, 2022; US Provisional Application No.63/413,174, filed October 4, 2022; and US Provisional Application No. 63/496,463, filed April 17, 2023; each of which is incorporated by reference herein in its entirety for any purpose. II. INTRODUCTION AND SUMMARY [0002] Cluster of differentiation 228 or CD228 (also known as melanotransferrin, MELTF, p97 and MFI2) is a glycosylphosphatidylinositol (GPI)-anchored glycoprotein that belongs to the transferrin family of iron-binding proteins and was first described as an oncofetal protein highly expressed on malignant melanoma cells (Rose et al., Proc Natl Acad Sci U S A, 1986). [0003] CD228 is expressed in a variety of cancers, including melanoma, mesothelioma, thyroid cancer, lung cancer, liver cancer, pancreatic cancer, head and neck cancer, stomach cancer, colorectal cancer, urothelial cancer, breast cancer, and cervical cancer. Melanoma, also known as malignant melanoma, is a type of cancer that develops from melanocytes, which are pigment-containing cells. Melanoma is the most dangerous type of skin cancer. In 2015, were 3.1 million people with active disease and melanoma resulted in 59,800 deaths. Surgery can be effective for early-stage melanoma but may not be a treatment option for disease that has metastasized to distant organs. Melanomas that spread often do so to the lymph nodes in the area before spreading elsewhere. Attempts to improve survival by removing lymph nodes surgically were associated with many complications but no overall survival benefit. Immunotherapy, chemotherapy, and radiation therapy have all been used, but are often not curative, particularly for late-stage melanoma. When there is distant metastasis, the cancer is generally considered incurable. The five-year survival rate of stage IV disease is 15-20%. [0004] CD137 (also known as 4-1BB and TNFRSF9) is a co-stimulatory immune receptor and a member of the tumor necrosis factor receptor (TNFR) superfamily. It is primarily expressed on activated CD4+ and CD8+ T cells, activated B cells, and natural killer (NK) cells but can also be found on resting monocytes and dendritic cells (Li and Liu, Clin Pharmacol, 2013), or endothelial cells (Snell et al., Immunol Rev, 2011). CD137 plays an important role in regulation of the immune response and thus is a target for cancer immunotherapy. CD137 ligand (CD137L) is the only known natural ligand of CD137, is constitutively expressed on several types of antigen presenting cells, such as activated B cells, monocytes, and splenic dendritic cells, and can be induced on T lymphocytes. Attorney Docket No.01218-0029-00PCT [0005] CD137L is a trimeric protein that exists as a membrane-bound form and as a soluble variant. The ability of soluble CD137L to activate CD137, e.g., on CD137-expressing lymphocytes is limited, however, and large concentrations are required to elicit an effect (Wyzgol et al., J Immunol, 2009). The natural way of activation of CD137 is via the engagement of a CD137-positive cell with a CD137L-positive cell. CD137 activation is then thought to be induced by clustering through CD137L on the opposing cell, leading to signaling via TRAF1, 2 and 3 (Yao et al., Nat Rev Drug Discov, 2013; Snell et al., Immunol Rev, 2011) and further concomitant downstream effects in the CD137-positive T cell. In the case of T cells activated by recognition of their respective cognate targets, the effects elicited by co-stimulation of CD137 are a further enhanced activation, enhanced survival and proliferation, the production of pro-inflammatory cytokines and an improved capacity to kill. [0006] The present disclosure provides, among other things, novel antibodies and antigen-binding domains thereof that bind CD228, along with novel approaches for simultaneously engaging CD137 and CD228 via one or more fusion proteins having the properties of binding specificity for CD137 and binding specificity for CD228. III. DEFINITIONS [0007] The following list defines terms, phrases, and abbreviations used throughout the instant specification. All terms listed and defined herein are intended to encompass all grammatical forms. [0008] As used herein, unless otherwise specified, “CD137” means human CD137 (huCD137). Human CD137 means a full-length protein defined by UniProt Q07011, a fragment thereof, or a variant thereof. CD137 is also known as 4-1BB, tumor necrosis factor receptor superfamily member 9 (TNFRSF9), and induced by lymphocyte activation (ILA). In some particular embodiments, CD137 of non-human species, e.g., cynomolgus CD137 or mouse CD137, is used. [0009] As used herein, unless otherwise specified, “CD228” means human CD228. Human CD228 means a full-length protein defined by UniProt P08582, a mature form thereof, an isoform thereof, a fragment thereof, or a variant thereof. Human CD228 is encoded by the MELTF gene. CD228 is also known as melanotransferrin, MELTF, p97 and MFI2, which terms may be used interchangeably herein. In some particular embodiments, CD228 of non-human species, e.g., cynomolgus CD228 or mouse CD228, is used. [0010] As used herein, “binding affinity” describes the ability of a biomolecule (e.g., a polypeptide or a protein) of the disclosure (e.g., a lipocalin mutein, an antibody, an antigen-binding domain thereof, a fusion protein, or any other peptide or protein) to bind a selected target (and Attorney Docket No.01218-0029-00PCT form a complex). Binding affinity is measured by a number of methods known to those skilled in the art including, but not limited to, fluorescence titration, enzyme-linked immunosorbent assay (ELISA)-based assays, including direct and competitive ELISA, calorimetric methods, such as isothermal titration calorimetry (ITC), and surface plasmon resonance (SPR). These methods are well-established in the art and some examples of such methods are further described herein. Binding affinity is thereby reported as a value of the dissociation constant (K _D ), half maximal effective concentration (EC ₅₀ ), or half maximal inhibitory concentration (IC ₅₀ ) measured using such methods. A lower K _D , EC ₅₀ , or IC ₅₀ value reflects better (higher) binding ability (affinity). Accordingly, the binding affinities of two biomolecules toward a selected target can be measured and compared. When comparing the binding affinities of two biomolecules toward the selected target, the term “comparable to”, “about the same,” “substantially the same” or “substantially similar” means one biomolecule has a binding affinity reported as a K _D , an EC ₅₀ , or an IC ₅₀ value that is identical or similar to that of another molecule within the experimental variability of the binding affinity measurement. Preferably, “comparable to”, “about the same,” “substantially the same” or “substantially similar” relate to a value that is within 50% deviation to a given reference value, more preferably within 20% deviation, most preferably within 10% deviation. The experimental variability of the binding affinity measurement is dependent upon the specific method used and is known to those skilled in the art. [0011] As used herein, the term “substantially” may also refer to the qualitative condition of exhibiting total or near-total extent or degree of a characteristic or property of interest. One of ordinary skill in the biological arts will understand that biological and chemical phenomena rarely, if ever, go to completion and/or proceed to completeness or achieve or avoid an absolute result. The term “substantially” is therefore used herein to capture the potential lack of completeness inherent in many biological and chemical phenomena. [0012] As used herein, the term “detect”, “detection”, “detectable”, or “detecting” is understood both on a quantitative and a qualitative level, as well as a combination thereof. It thus includes quantitative, semi-quantitative, and qualitative measurements performed on a biomolecule of the disclosure. [0013] As used herein, “detectable affinity” generally means the binding ability between a biomolecule and its target, reported by a K _D , EC ₅₀ , or IC ₅₀ value, is at most about 10 ^-5 M or lower. A binding affinity, reported by a K _D , EC ₅₀ , or IC ₅₀ value, higher than 10 ^-5 M is generally no longer measurable with common methods such as ELISA and SPR and is therefore of secondary importance. Thus, “detectable affinity” may refer to a K _D value of about 10 ^-5 M or lower as determined by ELISA or SPR, preferably SPR. Attorney Docket No.01218-0029-00PCT [0014] As used herein, “specific for”, “specific binding”, “specifically bind”, or “binding specificity” relates to the ability of a biomolecule to discriminate between the desired target (for example, CD137 and CD228) and one or more reference targets (for example, cellular receptor for neutrophil gelatinase-associated lipocalin). It is understood that such specificity is not an absolute but a relative property and can be determined, for example, by means of SPR, western blots, ELISA, fluorescence activated cell sorting (FACS), radioimmunoassay (RIA), electrochemiluminescence (ECL), immunoradiometric assay (IRMA), ImmunoHistoChemistry (IHC), and peptide scans. [0015] When used herein in the context of a biomolecule, such as an antibody, an antigen- binding domain thereof, or a fusion protein, of the present disclosure that binds to CD137 and/or CD228, the term “specific for”, “specific binding”, “specifically bind”, or “binding specificity” means that the biomolecule binds to, reacts with, or is directed against CD137 and/or CD228, as described herein, but does not substantially bind another protein. The term “another protein” includes any proteins that are not CD137 or CD228 nor proteins closely related to or being homologous to CD137 or CD228. However, CD137 or CD228 from species other than human and fragments and/or variants of CD137 or CD228 are not excluded by the term “another protein.” The term “does not substantially bind” means that a biomolecule of the present disclosure binds another protein with lower binding affinity than CD137 and/or CD228, i.e., shows a cross-reactivity of less than 30%, preferably 20%, more preferably 10%, particularly preferably less than 9, 8, 7, 6, or 5%. Whether the biomolecule specifically reacts as defined herein above can easily be tested, inter alia, by comparing the reaction of a biomolecule of the present disclosure with CD137 and/or CD228 and the reaction of said biomolecule with (an)other protein(s). [0016] As used herein, the term “lipocalin” refers to a monomeric protein of approximately 18-20 kDa in weight, having a cylindrical β-pleated sheet supersecondary structural region comprising a plurality of β-strands (preferably eight β-strands designated A to H) connected pair- wise by a plurality of (preferably four) loops at one end to thereby comprise a ligand-binding pocket and define the entrance to the ligand-binding pocket. Preferably, the loops comprising the ligand-binding pocket used in the present disclosure are loops connecting the open ends of β- strands A and B, C and D, E and F, and G and H, and are designated loops AB, CD, EF, and GH. It is well-established that the diversity of said loops in the otherwise rigid lipocalin scaffold gives rise to a variety of different binding modes among the lipocalin family members, each capable of accommodating targets of different sizes, shape, and chemical character (reviewed, e.g., in Skerra, Biochim Biophys Acta, 2000; Flower et al., Biochim Biophys Acta, 2000; Flower, Biochem J, 1996). It is understood that the lipocalin family of proteins has naturally evolved to bind a wide spectrum of ligands, sharing unusually low levels of overall sequence conservation (often with sequence identities of less than 20%) yet retaining a highly conserved overall folding Attorney Docket No.01218-0029-00PCT pattern. The correspondence between positions in various lipocalins is also well-known to one of skill in the art (see, e.g., U.S. Patent No.7,250,297). Proteins falling in the definition of “lipocalin” as used herein include, but are not limited to, human lipocalins including tear lipocalin (Tlc, Lcn1), Lipocalin-2 (Lcn2) or neutrophil gelatinase-associated lipocalin (NGAL), apolipoprotein D (ApoD), apolipoprotein M, α ₁ -acid glycoprotein 1, α ₁ -acid glycoprotein 2, α ₁ -microglobulin, complement component 8γ, retinol-binding protein (RBP), the epididymal retinoic acid-binding protein, glycodelin, odorant-binding protein IIa, odorant-binding protein IIb, lipocalin-15 (Lcn15), and prostaglandin D synthase. [0017] As used herein, unless otherwise specified, “tear lipocalin” refers to human tear lipocalin (hTlc) and further refers to mature human tear lipocalin. The term “mature” when used to characterize a protein means a protein essentially free from the signal peptide. A “mature hTlc” of the instant disclosure refers to the mature form of human tear lipocalin, which is free from the signal peptide. Mature hTlc is described by residues 19-176 of the sequence deposited with the SWISS-PROT Data Bank under Accession Number P31025, and its amino acid sequence is shown in SEQ ID NO: 1. [0018] As used herein, “Lipocalin-2” or “neutrophil gelatinase-associated lipocalin” refers to human Lipocalin-2 (hLcn2) or human neutrophil gelatinase-associated lipocalin (hNGAL) and further refers to the mature human Lipocalin-2 or mature human neutrophil gelatinase-associated lipocalin. The term “mature” when used to characterize a protein means a protein essentially free from the signal peptide. A “mature hNGAL” of the instant disclosure refers to the mature form of human neutrophil gelatinase-associated lipocalin, which is free from the signal peptide. Mature hNGAL is described by residues 21-198 of the sequence deposited with the SWISS-PROT Data Bank under Accession Number P80188, and its amino acid sequence is shown in SEQ ID NO: 2. [0019] As used herein, a “native sequence” refers to a protein or a polypeptide having a sequence that occurs in nature or having a wild-type sequence, regardless of its mode of preparation. Such native sequence protein or polypeptide can be isolated from nature or can be produced by other means, such as by recombinant or synthetic methods. [0020] The “native sequence lipocalin” refers to a lipocalin having the same amino acid sequence as the corresponding polypeptide derived from nature. Thus, a native sequence lipocalin can have the amino acid sequence of the respective naturally occurring (wild-type) lipocalin from any organism, in particular, a mammal. The term “native sequence”, when used in the context of a lipocalin, specifically encompasses naturally occurring truncated or secreted forms of the lipocalin, naturally occurring variant forms such as alternatively spliced forms and naturally occurring allelic variants of the lipocalin. The terms “native sequence lipocalin” and “wild- type lipocalin” are used interchangeably herein. Attorney Docket No.01218-0029-00PCT [0021] As used herein, a “mutein”, a “mutated” entity (whether protein or nucleic acid), or “mutant” refers to the exchange, deletion, or insertion of one or more amino acids or nucleotides, compared to the naturally occurring (wild-type) protein or nucleic acid. Said term also includes fragments of a mutein as described herein. The present disclosure explicitly encompasses lipocalin muteins (also referred to as Anticalin® proteins), as described herein, having a cylindrical β-pleated sheet supersecondary structural region comprising eight β-strands connected pair-wise by four loops at one end to thereby comprise a ligand-binding pocket and define the entrance of the ligand-binding pocket, wherein at least one amino acid located within said four loops has been mutated as compared to the native sequence lipocalin. Lipocalin muteins of the present disclosure preferably have the function of binding CD137 as described herein. [0022] As used herein, the term “fragment”, in connection with the lipocalin muteins of the disclosure, refers to proteins or polypeptides derived from full-length mature hTlc or hNGAL or lipocalin muteins that are N-terminally and/or C-terminally truncated, i.e., lacking at least one of the N-terminal and/or C-terminal amino acids. Such fragments may include at least 10 or more, such as 20 or 30 or more consecutive amino acids of the primary sequence of mature hTlc or hNGAL or the lipocalin mutein it is derived from and are usually detectable in an immunoassay of mature hTlc or hNGAL. Such a fragment may lack up to 2, up to 3, up to 4, up to 5, up to 10, up to 15, up to 20, up to 25, or up to 30 (including all numbers in between) of the N-terminal and/or C-terminal amino acids. As an illustrative example, such a fragment may lack the one, two, three, or four N-terminal (His-His-Leu-Leu) and/or one or two C-terminal amino acids (Ser-Asp) of mature hTlc. It is understood that the fragment is preferably a functional fragment of mature hTlc or hNGAL or the lipocalin mutein from which it is derived, which means that it preferably retains the binding specificity, preferably to CD137, of mature hTlc/hNGAL or lipocalin mutein it is derived from. As an illustrative example, such a functional fragment may comprise at least amino acids at positions 5-153, 5-150, 9-148, 12-140, 20-135, or 26-133 corresponding to the linear polypeptide sequence of mature hTlc. As another illustrative example, such a functional fragment may comprise at least amino acids at positions 13–157, 15-150, 18-141, 20-134, 25-134, or 28-134 corresponding to the linear polypeptide sequence of mature hNGAL. [0023] A “fragment” with respect to the corresponding target CD137 or CD228 of an antibody, an antigen-binding domain thereof, or a fusion protein of the disclosure, refers to N- terminally and/or C-terminally truncated CD137 or CD228 or protein domains of CD137 or CD228. Fragments of CD137 or fragments of CD228 as described herein retain the capability of the full- length CD137 or CD228 to be recognized and/or bound by an antibody, an antigen-binding domain thereof, or a fusion protein of the disclosure. As an illustrative example, the fragment may be an extracellular domain of CD137. For example, such an extracellular domain of CD137 may comprise amino acids of the extracellular subdomains of CD137, such as the individual or Attorney Docket No.01218-0029-00PCT combined amino acid sequences of domain 1 (residues 24-45 of UniProt Q07011), domain 2 (residues 46-86), domain 3 (87-118) and domain 4 (residues 119-159). [0024] As used herein, the term “variant” relates to derivatives of a protein or polypeptide that include mutations, for example by substitutions, deletions, insertions, and/or chemical modifications of an amino acid sequence or nucleotide sequence. In some embodiments, such mutations and/or chemical modifications do not reduce the functionality of the protein or peptide. Such substitutions may be conservative, i.e., an amino acid residue is replaced with a chemically similar amino acid residue. Examples of conservative substitutions are the replacements among the members of the following groups: 1) alanine, serine, threonine, and valine; 2) aspartic acid, glutamic acid, glutamine, asparagine, and histidine; 3) arginine, lysine, glutamine, asparagine, and histidine; 4) isoleucine, leucine, methionine, valine, alanine, phenylalanine, threonine, and proline; and 5) isoleucine, leucine, methionine, phenylalanine, tyrosine, and tryptophan. Such variants include proteins or polypeptides, wherein one or more amino acids have been substituted by their respective D-stereoisomers or by amino acids other than the naturally occurring 20 amino acids, such as, for example, ornithine, hydroxyproline, citrulline, homoserine, hydroxylysine, norvaline. Such variants also include, for instance, proteins or polypeptides in which one or more amino acid residues are added or deleted at the N- and/or C-terminus. Generally, a variant has at least about 50%, 60%, 70%, 75%, 80%, 85%, 90%, 92%, 95% or at least about 98% amino acid sequence identity with the native sequence protein or polypeptide. A variant preferably retains the biological activity, e.g., binding the same target, of the protein or polypeptide it is derived from. [0025] The term “variant”, as used herein with respect to CD137 or CD228 relates to CD137 or CD228 or a fragment thereof, respectively, that has one or more, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 40, 50, 60, 70, 80 or more amino acid substitutions, deletions and/or insertions in comparison to the native sequence of CD137 or CD228 (wild-type CD137 or CD228), such as CD137 as deposited with UniProt Q07011 or CD228 as deposited with UniProt P08582 as described herein. A CD137 variant or a CD228 variant, respectively, has preferably an amino acid identity of at least 50%, 60%, 70%, 80%, 85%, 90% or 95% with a wild-type CD137 or CD228. A CD137 variant or a CD228 variant as described herein retains the ability to bind antibodies, antigen-binding domains thereof, lipocalin muteins, or fusion proteins specific to CD137 and/or CD228 disclosed herein. [0026] The term “variant”, as used herein with respect to a lipocalin mutein, relates to a lipocalin mutein or fragment thereof of the disclosure, wherein the sequence has mutations, including substitutions, deletions, insertions, and/or chemical modifications. A variant of a lipocalin mutein as described herein retains the biological activity, e.g., binding to CD137, of the lipocalin mutein from which it is derived. Generally, a lipocalin mutein variant has at least about 50%, 60%, Attorney Docket No.01218-0029-00PCT 70%, 75%, 80%, 85%, 90%, 92%, 95%, or 98% amino acid sequence identity with the lipocalin mutein from which it is derived. [0027] The term “variant”, as used herein with respect to an antibody or an antigen- binding domain thereof relates to an antibody or an antigen-binding domain thereof of the disclosure, wherein the sequence has mutations, including substitutions, deletions, insertions, and/or chemical modifications. A variant of an antibody or an antigen-binding domain thereof as described herein retains the biological activity, e.g., binding to CD228, of the antibody or antigen- binding domain thereof from which it is derived. Generally, a variant of an antibody or an antigen- binding domain thereof has at least about 50%, 60%, 70%, 75%, 80%, 85%, 90%, 92%, 95%, or 98% amino acid sequence identity with the antibody or antigen-binding domain thereof from which it is derived. [0028] As used herein, the term “mutagenesis” refers to the introduction of mutations into a polynucleotide or amino acid sequence. Mutations are preferably introduced under experimental conditions such that the amino acid naturally occurring at a given position of the protein or polypeptide sequence can be altered, for example substituted by at least one amino acid. The term “mutagenesis” also includes the (additional) modification of the length of sequence segments by deletion or insertion of one or more amino acids. Thus, it is within the scope of the disclosure that, for example, one amino acid at a chosen sequence position is replaced by a stretch of three amino acids, leading to an addition of two amino acid residues compared to the length of the respective segment of the native protein or polypeptide amino acid sequence. Such an insertion or deletion may be introduced independently from each other in any of the sequence segments that can be subjected to mutagenesis in the disclosure. In one exemplary embodiment of the disclosure, an insertion may be introduced into an amino acid sequence segment corresponding to the loop AB of the native sequence lipocalin (cf. International Patent Publication No. WO 2005/019256, which is incorporated herein by reference in its entirety). [0029] As used herein, the term “random mutagenesis” means that no predetermined mutation (alteration of an amino acid) is present at a certain sequence position but that at least two amino acids can be incorporated with a certain probability at a predefined sequence position during mutagenesis. [0030] As used herein, the term “sequence identity” or “identity” denotes a property of sequences that measures their similarity or relationship. The term “sequence identity” or “identity” as used in the present disclosure means the percentage of pair-wise identical residues – following (homologous) alignment of a sequence of a protein or polypeptide of the disclosure with a sequence in question – with respect to the number of residues in the longer of these two sequences. Sequence identity is measured by dividing the number of identical amino acid Attorney Docket No.01218-0029-00PCT residues by the total number of residues and multiplying the product by 100. [0031] As used herein, the term “sequence homology” or “homology” has its usual meaning, and a homologous amino acid includes identical amino acids as well as amino acids which are regarded to be conservative substitutions at equivalent positions in the linear amino acid sequence of a protein or polypeptide of the disclosure (e.g., any fusion proteins or lipocalin muteins of the disclosure). [0032] A skilled artisan will recognize available computer programs, for example BLAST (Altschul et al., Nucleic Acids Res, 1997), BLAST2 (Altschul et al., J Mol Biol, 1990), and Smith- Waterman (Smith and Waterman, J Mol Biol, 1981), for determining sequence homology or sequence identity using standard parameters. The percentage of sequence homology or sequence identity can, for example, be determined herein using the program BLASTP, version 2.2.5 (November 16, 2002; (Altschul et al., Nucleic Acids Res, 1997). In some embodiments, the percentage of homology is based on the alignment of the entire protein or polypeptide sequences (matrix: BLOSUM 62; gap costs: 11.1; cutoff value set to 10 ^-3 ) including the propeptide sequences, preferably using the wild-type protein scaffold as reference in a pairwise comparison. It is calculated as the percentage of numbers of “positives” (homologous amino acids) indicated as result in the BLASTP program output divided by the total number of amino acids selected by the program for the alignment. [0033] Specifically, in order to determine whether the amino acid sequence of a lipocalin (mutein) is different from that of a reference (wild-type) lipocalin with regard to a certain position in the amino acid sequence of the reference (wild-type) lipocalin, a skilled artisan can use means and methods well-known in the art, e.g., alignments, either manually or by using computer programs such as BLAST 2.0, which stands for Basic Local Alignment Search Tool, or ClustalW, or any other suitable program which is suitable to generate sequence alignments. Accordingly, the amino acid sequence of a reference (wild-type) lipocalin can serve as “subject sequence” or “reference sequence”, while the amino acid sequence of a lipocalin mutein serves as “query sequence”. The terms “wild-type sequence”, “reference sequence” and “subject sequence” are used interchangeably herein. A preferred wild-type sequence of a lipocalin is the sequence of hTLc as shown in SEQ ID NO: 1 or hNGAL as shown in SEQ ID NO: 2. [0034] “Gaps” are spaces in an alignment that are the result of additions or deletions of amino acids. Thus, two copies of exactly the same sequence have 100% identity, but sequences that are less highly conserved, and have deletions, additions, or replacements, may have a lower degree of sequence identity. [0035] As used herein, the term “position” means the position of either an amino acid within an amino acid sequence disclosed herein or the position of a nucleotide within a nucleic Attorney Docket No.01218-0029-00PCT acid sequence disclosed herein. It is to be understood that when the term “correspond” or “corresponding” is used herein in the context of the amino acid sequence positions of one or more lipocalin muteins, a corresponding position is not only determined by the number of the preceding nucleotides or amino acids. Accordingly, the absolute position of a given amino acid in accordance with the disclosure may vary from the corresponding position due to deletion or addition of amino acids elsewhere in a (mutant or wild-type) lipocalin. Similarly, the absolute position of a given nucleotide in accordance with the present disclosure may vary from the corresponding position due to deletions or additional nucleotides elsewhere in a mutein or wild- type lipocalin 5’-untranslated region (UTR) including the promoter and/or any other regulatory sequences or gene regions (including exons and introns). [0036] A “corresponding position” in accordance with the disclosure may be the sequence position that aligns to the sequence position it corresponds to in a pairwise or multiple sequence alignment according to the present disclosure. It is preferably to be understood that for a “corresponding position” in accordance with the disclosure, the absolute positions of nucleotides or amino acids may differ from adjacent nucleotides or amino acids but said adjacent nucleotides or amino acids which may have been exchanged, deleted, or added may be comprised by the same one or more “corresponding positions”. [0037] In addition, for a corresponding position in a lipocalin mutein based on a reference sequence in accordance with the disclosure, it is preferably to be understood that the positions of nucleotides or amino acids of a lipocalin mutein can structurally correspond to the positions elsewhere in a reference lipocalin (wild-type lipocalin) or another lipocalin mutein, even if they may differ in the absolute position numbers, as appreciated by the skilled person in light of the highly-conserved overall folding pattern among lipocalins. [0038] As used interchangeably herein, the terms “conjugate”, “conjugation”, “fuse”, “fusion”, or “linked” refer to the joining together of two or more subunits, through all forms of covalent or non-covalent linkage, by means including, but not limited to, genetic fusion, chemical conjugation, coupling through a linker or a cross-linking agent, and non-covalent association. [0039] The term “fusion polypeptide” or “fusion protein” as used herein refers to a polypeptide or protein comprising two or more subunits. In some embodiments, a fusion protein as described herein comprises two or more subunits, at least one of these subunits being capable of specifically binding to CD137, and a further subunit capable of specifically binding to CD228. Within the fusion protein, these subunits may be linked by covalent or non-covalent linkage. Preferably, the fusion protein is a translational fusion between the two or more subunits. The translational fusion may be generated by genetically engineering the coding sequence for one subunit in a reading frame with the coding sequence of a further subunit. Both subunits may be Attorney Docket No.01218-0029-00PCT interspersed by a nucleotide sequence encoding a linker. However, the subunits of a fusion protein of the present disclosure may also be linked through chemical conjugation. The subunits forming the fusion protein are typically linked to each other as follows: C-terminus of one subunit to N-terminus of another subunit, or C-terminus of one subunit to C-terminus of another subunit, or N-terminus of one subunit to N-terminus of another subunit, or N-terminus of one subunit to C- terminus of another subunit. The subunits of the fusion protein can be linked in any order and may include more than one of any of the constituent subunits. If one or more of the subunits are part of a protein (complex) that consists of more than one polypeptide chain, the term “fusion protein” may also refer to the protein comprising the fused sequences and all other polypeptide chain(s) of the protein (complex). As an illustrative example, where a full-length immunoglobulin/antibody is fused to a lipocalin mutein via a heavy or light chain of the immunoglobulin/antibody, the term “fusion protein” may refer to the single polypeptide chain comprising the lipocalin mutein and the heavy or light chain of the immunoglobulin/antibody. The term “fusion protein” may also refer to the entire immunoglobulin/antibody (both light and heavy chains) and the lipocalin mutein fused to one or both of its heavy and/or light chains. [0040] As used herein, the term “subunit” of a fusion protein disclosed herein refers to a single protein or a separate polypeptide chain, which can form a stable folded structure by itself and may define a unique function of providing a binding motif towards a target. In some embodiments, a preferred subunit of the disclosure is a lipocalin mutein. In some other embodiments, a preferred subunit of the disclosure is an antibody, such as a full-length antibody, or an antigen-binding domain/fragment thereof. [0041] A “linker” that may be comprised by a fusion protein of the present disclosure joins together two or more subunits of a fusion protein as described herein. The linkage can be covalent or non-covalent. A preferred covalent linkage is via a peptide bond, such as a peptide bond between amino acids. A preferred linker is a peptide linker. Accordingly, in a preferred embodiment, said linker comprises one or more amino acids, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more amino acids. Preferred peptide linkers are described herein, including glycine-serine (GS) linkers, glycosylated GS linkers, and proline-alanine-serine polymer (PAS) linkers. In some preferred embodiments, a GS linker, such as a (G ₄ S) ₃ linker as described in SEQ ID NO: 13, is used to join together the subunits of a fusion protein. Other preferred linkers include chemical linkers. [0042] As used herein, the term “albumin” includes all mammalian albumins, such as human serum albumin or bovine serum albumin or rat serum albumin. [0043] As used herein, the term “organic molecule” or “small organic molecule” denotes an organic molecule comprising at least two carbon atoms, but preferably not more than 7 or 12 Attorney Docket No.01218-0029-00PCT rotatable carbon bonds, having a molecular weight in the range between 100 and 2,000 daltons, preferably between 100 and 1,000 daltons, and optionally including one or two metal atoms. [0044] A “sample” is defined as a biological sample taken from any subject. Biological samples include, but are not limited to, blood, serum, urine, feces, semen, or tissue, including tumor tissue. [0045] A “subject” is a vertebrate, preferably a mammal, more preferably a human. The term “mammal” is used herein to refer to any animal classified as a mammal, including, without limitation, humans, domestic and farm animals, and zoo, sports, or pet animals, such as sheep, dogs, horses, cats, cows, rats, pigs, apes such as cynomolgus monkeys, to name only a few illustrative examples. Preferably, the “mammal” used herein is human. [0046] An “effective amount” is an amount sufficient to yield beneficial or desired results. An effective amount can be administered in one or more individual administrations or doses. [0047] As used herein, “antibody” includes whole antibodies or any antigen-binding fragment (i.e., “antigen-binding portion” or “antigen-binding domain”) or single chain thereof. The terms “antibody” and “immunoglobulin” can be and are used interchangeably herein. A whole antibody refers to a glycoprotein comprising at least two heavy chains (HCs) and two light chains (LCs) inter-connected by disulfide bonds. Each heavy chain is comprised of a heavy chain variable domain/region (V _H or HCVR) and a heavy chain constant region (C _H ). The heavy chain constant region is comprised of three domains, C _H1 , C _H2 and C _H3 . Each light chain is comprised of a light chain variable domain/region (V _L or LCVR) and a light chain constant region (C _L ). The light chain constant region is comprised of one domain, C _L . The VH and VL regions can be further subdivided into regions of hypervariability, termed complementarity determining regions (CDRs), interspersed with regions that are more conserved, termed framework regions (FRs). Each V _H and V _L is composed of three CDRs and four FRs, arranged in the following order from the amino- terminus to the carboxy-terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The variable regions of the heavy and light chains contain a binding domain that interacts with an antigen (for example, CD228). The constant regions of the antibodies may optionally mediate the binding of the immunoglobulin to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component (C1q) of the classical complement system. [0048] As used herein, “antigen-binding fragment” (also referred to as “antigen-binding domain”) of an antibody refers to one or more fragments of an antibody that retain the ability to specifically bind to an antigen (e.g., CD228). It has been shown that the antigen-binding function of an antibody can be performed by fragments of a full-length antibody. Examples of binding fragments encompassed within the term “antigen-binding fragment” of an antibody include (i) a Fab fragment consisting of the V _H , V _L , C _L and C _H1 domains; (ii) a F(ab′) ₂ fragment comprising two Attorney Docket No.01218-0029-00PCT Fab fragments linked by a disulfide bridge at the hinge region; (iii) a Fab′ fragment consisting of the V _H , V _L , C _L and C _H1 domains and the region between the C _H1 and C _H2 domains; (iv) an Fd fragment consisting of the V _H and C _H1 domains; (v) a single-chain Fv fragment consisting of the V _H and V _L domains of a single arm of an antibody, (vi) a dAb fragment (Ward et al., Nature, 1989) consisting of a V _H domain; (vii) an isolated complementarity determining region (CDR) or a combination of two or more isolated CDRs which may optionally be joined by a synthetic linker; (viii) a “diabody” comprising the V _H and V _L connected in the same polypeptide chain using a short linker (see, e.g., patent documents EP 404,097; WO 93/11161; and Holliger et al., Proc Natl Acad Sci U S A, 1993); and (ix) a “domain antibody fragment” containing only the V _H or V _L , where in some instances two or more V _H regions are covalently joined. [0049] Antibodies may be polyclonal or monoclonal; xenogeneic, allogeneic, or syngeneic; or modified forms thereof (e.g., humanized, chimeric, or multispecific). Antibodies may also be fully human. [0050] As used herein, “framework” or “FR” refers to the variable domain residues other than the hypervariable region (CDR) residues. [0051] “Fragment crystallizable region” or “Fc region” refers to the C-terminal region of an immunoglobulin heavy chain, including native-sequence Fc regions and variant Fc regions. Although the boundaries of the Fc region of an immunoglobulin heavy chain might vary, the human IgG heavy-chain Fc region is usually defined to stretch from an amino acid residue at position Cys226, or from Pro230, to the carboxyl-terminus thereof numbering according to EU index of Kabat (Johnson and Wu, Nucleic Acids Res, 2000). The C-terminal lysine (residue 447 according to EU index of Kabat) of the Fc region may be removed, for example, during production or purification of the antibody, or by recombinantly engineering the nucleic acid encoding a heavy chain of the antibody. Accordingly, a composition of intact antibodies may comprise antibody populations with all K447 residues removed, antibody populations with no K447 residues removed, and antibody populations having a mixture of antibodies with and without the K447 residue. Suitable native-sequence Fc regions for use in the antibodies of the disclosure include human IgG1, IgG2 (IgG2A, IgG2B), IgG3, and IgG4. [0052] “Fc receptor” or “FcR” refers to a receptor that binds to the Fc region of an antibody. [0053] As used herein, “isolated antibody” refers to an antibody that is substantially free of its natural environment. For instance, an isolated antibody is substantially free of cellular material and other proteins from the cell or tissue source from which it is derived. An “isolated antibody” further refers to an antibody that is substantially free of other antibodies having different antigenic specificities. In the present case, an isolated antibody that binds specifically CD228 is substantially free of antibodies that specifically bind antigens other than CD228. However, an Attorney Docket No.01218-0029-00PCT isolated antibody that specifically binds CD228 may have cross-reactivity to other antigens, such as CD228 molecules from other species. [0054] As used herein, “monoclonal antibody” refers to a preparation of antibody molecules of single molecular composition. A monoclonal antibody composition displays a single binding specificity and affinity for a particular epitope. [0055] As used herein, “humanized antibody” refers to an antibody that consists of the CDRs of antibodies derived from mammals other than human, and the FR region and the constant region of a human antibody or derived from a human antibody. A humanized antibody may comprise a variable domain that has a variable region amino acid sequence which, analyzed as a whole, is closer to human than to other species as assessed using the Immunogenetics Information System (IMGT) DomainGapAlign tool, as described by Ehrenmann et al. (2010). A humanized antibody may be useful as an effective component in a therapeutic agent due to the reduced antigenicity. The term “therapeutic agent” or “therapeutically active agent”, as used herein, refers to an agent which is therapeutically useful. A therapeutic agent may be any agent for the prevention, amelioration, or treatment of a disease, a physiological condition, a symptom, or for the evaluation or diagnosis thereof. [0056] As used herein, “human antibody” includes antibodies having variable regions in which both the framework and CDR regions are derived from human germline immunoglobulin sequences. Furthermore, if the antibody contains a constant region, the constant region is also derived from human germline immunoglobulin sequences. The human antibodies of the disclosure may include amino acid residues not encoded by human germline immunoglobulin sequences (e.g., mutations introduced by random or site-specific mutagenesis in vitro or by somatic mutation in vivo). However, the term “human antibody”, as used herein, is not intended to include antibodies in which CDR sequences derived from the germline of another mammalian species, such as a mouse, have been grafted onto human framework sequences. [0057] As used herein, “antibody clone X” may also be referred to as “OMTX.” For example, “antibody clone 30” may also be referred to as “OMT30.” Antibody clone 8 may also be referred to as “EE03”; antibody clone 24 may also be referred to as “EB03”; antibody clone 30 may also be referred to as “OC04”; antibody clone 35 may also be referred to as “OB02”; and antibody clone 36 may also be referred to as “OE12.” [0058] As used herein, “AAFX” refers to a fusion protein containing antibody clone X and the lipocalin mutein having the amino acid sequence of SEQ ID NO: 40 (Lipocalin Mutein I). For example, “AAF30” refers to a fusion protein containing antibody clone 30 and Lipocalin Mutein I. The notation “HC” in the name of a fusion protein refers to the lipocalin mutein therein being conjugated to the antibody via the antibody’s heavy chain, and the notation “LC” in the name of a Attorney Docket No.01218-0029-00PCT fusion protein refers to the lipocalin mutein therein being conjugated to the antibody via the antibody’s light chain. A fusion protein name that contains a clone number followed by “HC” or “LC” refers to a fusion protein containing that antibody clone and Lipocalin Mutein I, conjugated via the antibody’s heavy chain or light chain, respectively. For example, “30HC” refers to a fusion protein containing antibody clone 30 and Lipocalin Mutein I conjugated thereto via the antibody’s heavy chain. [0059] An antibody or fusion protein may also be denoted by its heavy and light chain sequences (e.g., the fusion protein having the sequences of SEQ ID NOs: 80 and 76), one of which in the case of a fusion protein (e.g., SEQ ID NO: 80) contains the sequence of a lipocalin mutein. IV. DESCRIPTIONS OF FIGURES [0060] Figure 1 provides a schematic showing CD137 clustering by bridging CD137- positive T cells with CD228-expressing tumor cells using the fusion proteins provided herein. [0061] Figures 2A and 2B show CDR sequences of certain antibodies provided herein [0062] Figure 3 shows results of an ELISA, in which anti-CD228 antibodies bind recombinant human CD228. [0063] Figures 4A-4C show binding of anti-CD228 antibodies to CD228-expressing cells, as measured by fluorescence intensity. Fig.4D shows the EC ₅₀ value of each antibody for each cell line. [0064] Figures 5A-5E show results of biolayer inferometry (BLI) assays for binding kinetics and affinity of anti-CD228 antibodies to CD228. [0065] Figures 6A-6C show results of binding assays of anti-CD228 antibodies to CD228 and certain other transferrin family members. [0066] Figures 7A-7B and 8A-8C show results of binding assays of anti-CD228 antibodies to human, murine, and cynomolgus CD228. [0067] Figure 9 shows results of a cross-competition assay between various anti-CD228 antibodies. [0068] Figure 10 shows results of an antibody internalization assay using CD228+ tumor cell lines. [0069] Figures 11A-11I provide an overview over the design of representative fusion proteins described in this application that are bispecific for the targets CD137 and CD228. Attorney Docket No.01218-0029-00PCT Representative fusion proteins were made based on an antibody specific for CD228 and one or more lipocalin muteins specific for CD137. One or more lipocalin muteins were genetically fused to the C- and/or the N-terminus of the heavy chain and/or the light chain of a CD228-specific antibody as depicted in Figure 11A-11I. The generated fusion proteins can be bivalent to CD137 (e.g., as depicted in Figure 11A-11D) or tetravalent to CD137 (e.g., as depicted in Figure 11E- 11H), or can have even higher valency to CD137 (e.g., as depicted in Figure 11I). [0070] Figures 12A-12D show the results of ELISA experiments in which the binding to human or cynomolgus CD228 (Figures 12A-12B) or CD137 (Figures 12C-12D) of representative fusion proteins was determined as described in Example 11. CD228 or CD137 (with C-terminal His or Fc tag) was coated on a microtiter plate, and the tested agents were titrated starting with the highest concentration of 200 nM. Bound agents under study were detected via anti-human IgG Fab-HRP. The data were fit with a 4PL fit with EC ₅₀ value and the maximum signal as free parameters, and a slope that was fixed to one. The resulting EC ₅₀ values are provided in Table 4. [0071] Figure 13 illustrates the results of an ELISA experiment in which the ability of representative fusion proteins to simultaneously bind both targets, CD228 and CD137, was determined as described in Example 12. Recombinant huCD228-His was coated on a microtiter plate, followed by a titration of the fusion proteins and control anti-CD228 antibodies starting with the highest concentration of 200 nM. Subsequently, a constant concentration of biotinylated huCD137-His was added, which was detected via ExtrAvidin-Peroxidase. The data were fit with a 4PL fit with EC ₅₀ value and the maximum signal as free parameters, and a slope that was fixed to one. The resulting EC ₅₀ values are provided in Table 5. [0072] Figure 14 illustrates additional results of an ELISA experiment in which the ability of representative fusion proteins to simultaneously bind both targets, CD228 and CD137, was determined as described in Example 12. [0073] Figures 15A-15B show the results of an assessment of the target binding of fusion proteins and respective controls by flow cytometry (FACS) using CD228-positive human SH-4 cells (Figure 15A) and Flp-In-CHO cells expressing human CD137 (Figure 15B) as described in Example 13. The geometric means of the fluorescence intensity were used to calculate EC ₅₀ values using nonlinear regression (shared bottom, SLOPE = 1). EC ₅₀ values are provided in Table 6. [0074] Figures 16A-16E show the potential of representative fusion proteins to co- stimulate T cell activation in a CD228-target-dependent manner, as assessed by a CD137 bioassay as described in Example 14. NFκB-luc2/CD137 Jurkat cells were co-cultured with tumor cells expressing high (SH-4) (Figure 16A), medium (A375) (Figure 16B) or low (A549) (Figure Attorney Docket No.01218-0029-00PCT 16C) levels of CD228 as well as with CD228-negative RPMI-7951 tumor cells (Figure 16D) in the presence of various concentrations of the fusion proteins or controls. After 4 hours, luciferase assay reagent was added and luminescent signals were measured. Four-parameter logistic curve analysis was performed with GraphPad Prism® to calculate EC ₅₀ values (see table in Figure 16E). [0075] Figures 17A-17E show results of antigen recall assays in which fusion proteins co-stimulate innate and adaptive immune cytokines from peripheral blood mononuclear cells (PBMC) in response to viral peptides. [0076] Figures 18, 19, 20, and 21 show results of antigen recall assays in which fusion proteins co-stimulate T and NK cell responses in response to viral peptides. [0077] Figures 22A-22E show results of antigen recall assays in which fusion proteins co-stimulate cytotoxic effector molecules and cytokines from PBMC in response to viral peptides. [0078] Figures 23A-23D show the ability of representative fusion proteins to co-stimulate T cell activation in a CD228-target-dependent manner. Various tumor cell lines expressing different CD228 levels – SH-4 (CD228 high) (Figure 23A), Calu-1 (CD228 high) (Figure 23B), SK-MEL-24 (CD228 high) (Figure 23C), and RPMI-7951 (CD228-negative) (Figure 23D) – were seeded into anti-human CD3 coated plates. Pan T cells and various concentrations of fusion proteins or controls were added and incubated for 3 days. Levels of secreted IL-2 were used as read-out and determined by an electrochemoluminescence-based assay as described in Example 18. Similarly, Figures 24A-24B show the ability of additional representative fusion proteins to co-stimulate T cell activation in a CD228-target-dependent manner. [0079] Figures 25A-25C show production of T cell cytokines in cocultures with an anti- CD3 scFv engineered CD228-expressing tumor cell line. [0080] Figure 26 shows CD8 T cell proliferation in cocultures with an anti-CD3 scFv engineered CD228-expressing tumor cell line, and Figure 27 shows tumor cell killing in the coculture. [0081] Figures 28A-28D illustrate the assessment of storage stability of fusion proteins and controls in PBS, 50% human plasma (HPL), or 50% mouse plasma (MPL) after incubation at 37 °C for one week as described in Example 21. The stability was assessed by means of a CD137 reporter cell assay as described in Example 14. [0082] Figures 29A-29B provide the result of pharmacokinetic analyses of the bispecific fusion proteins and the parental anti-CD228 antibodies in mice as described in Example 22. Male CD-1 NUDE mice (3 mice per timepoint) were injected intravenously with fusion proteins at a dose Attorney Docket No.01218-0029-00PCT of 10 mg/kg. Drug levels were detected using a Sandwich ELISA detecting the full molecules via their targets CD228 and biotinylated CD137. The anti-CD228 antibody plasma levels were determined using an ELISA via the target CD228 and an anti-human IgG Fc antibody. Absolute concentrations are shown in Figure 29A, values normalized to C _max are shown in Figure 29B. [0083] Figure 30 shows pharmacokinetics of fusion proteins in cynomolgus monkeys. [0084] Figures 31A-31D, 32A-32C, and 33A-33B show in vivo activity of fusion proteins in humanized xenograft models. [0085] Figures 34A and 34B show CD8 T cell proliferation in cocultures with anti-CD3 scFv engineered CD228-expressing tumor cell lines. [0086] Figure 35A shows CD8 T cell count in wells with treatment added, relative to wells without treatment added. Figure 35B shows CD8 T cell mitochondrial content, represented by mean fluorescence intensity (MFI) of MitoSpy™ green FM staining. Figure 35C shows the percentages of CD8 T cell with depolarized mitochondria, as determined by staining with MitoSpy™ orange CMTMros and MitoSpy™ green FM. [0087] Figure 36A shows a schematic of creating functionally exhausted cytotoxic T cells. Figure 36B shows the percentages of divided cells (left), the levels of IFN-γ (center), and the numbers of live tumor cells (right) for T cells from different passages (P0-P4) restimulated with anti-CD3 scFv engineered CALU-1 cells. [0088] Figure 37A shows a heatmap for exhausted CD8+ T cells from Figure 36A or exhausted CD8+ T cells from a pan-cancer tumor-infiltrating lymphocytes (TILs) atlas. Figure 37B shows expression levels of TCF7 (TCF1) (left), HAVCR2 (TIM-3) (center), and TNFRSF9 (CD137) (right) in two groups of exhausted CD8+ T cells. [0089] Figure 38 shows the fold increase in carboxyfluorescein succinimidyl ester (CFSE)-low (divided) CD8 T cell counts in wells with treatment added, relative to wells without treatment added. V. DETAILED DESCRIPTION OF THE DISCLOSURE [0090] As is described herein, in one aspect, the present disclosure provides antibodies or antigen-binding domains thereof that bind CD228. Such anti-CD228 antibodies can be used by themselves as antibody therapeutics, conjugated to a therapeutic agent to generate an antibody-drug conjugate, included as part of a bispecific or multispecific antibody, or included as part of a fusion molecule. In certain embodiments, such fusion molecules include those that have a binding domain that binds a different target than CD228. For example, as described in greater Attorney Docket No.01218-0029-00PCT detail below, in some embodiments, the fusion molecule is a fusion protein that can bind both CD137 and CD228. [0091] Certain fusion proteins that target both CD137 and CD228 as provided herein were designed based upon the recognition of several factors by the present inventors. For instance, with respect to CD137, it was recognized that CD137 is expressed on multiple immune cell types and that engagement enhances innate and adaptive immunity. It was also recognized that CD137 agonism restores functionality to exhausted T cells, thereby reinvigorating pre-existing anti-tumor activity. In the case of CD228, the inventors recognized that it could be an excellent antigen target for a tumor-targeting anti-CD137 bispecific molecule for several reasons. First, CD228 expression is highly restricted to tumors, thereby minimizing the risk of toxicity. Second, the combination of two agents with a distinct mechanism of action has the potential to provide broader coverage and the opportunity to facilitate additional combination therapies. [0092] In addition, the fusion proteins that target both CD137 and CD228 as provided herein are based on the recognition that a bivalent CD137-binder, such as an antibody, may not be sufficient by itself to cluster CD137 on T cells or NK cells and lead to efficient activation, similar to the lack of activity of the trivalent soluble CD137L. A second problem observed with bivalent anti-CD137 antibodies is that active doses can in some instances elicit on-target liver toxicity, indicating a need for a more specific, tumor-targeted mode of action. [0093] Based in part upon these various insights, the present disclosure provides, among other things, novel approaches for simultaneously engaging CD137 and CD228 via one or more fusion proteins having binding specificity for CD137 and binding specificity for CD228. The binding specificity for CD228 may be provided by an antibody or antigen-binding domain thereof that binds CD228 as provided herein. Provided fusion proteins are designed to promote CD137 clustering by bridging CD137-positive T cells with CD228-expressing tumor cells located in the tumor microenvironment, as shown in exemplary Fig.1. Hence, the fusion proteins are designed to allow the localized induction of antigen-specific T cells in the tumor microenvironment, potentially reducing peripheral toxicity. This is in contrast to, for example, urelumab, a 4-1BB agonist that is active in solid tumors but that elicits on-target liver toxicity at active doses. [0094] In some aspects, the present disclosure provides antibodies or antigen-binding domains thereof that bind CD228 and fusion proteins that bind CD137 and CD228, as well as methods and useful applications therefor. The disclosure also provides methods of making CD228-binding antibodies or antigen-binding domains thereof and CD137- and CD228-binding fusion proteins described herein as well as compositions comprising such proteins. CD228- binding antibodies or antigen-binding domains thereof and CD137- and CD228-binding fusion proteins of the disclosure as well as compositions thereof may be used in methods of detecting Attorney Docket No.01218-0029-00PCT CD137 and/or CD228 in a sample, in methods of binding of CD137 and/or CD228 in a subject, or in methods of modulating immune responses in a subject. No such antibodies, antigen-binding domains thereof, or fusion proteins having these features attendant to the uses provided by present disclosure have been previously described. A. Exemplary antibodies or antigen-binding domains thereof specific for CD228 [0095] In some embodiments, an antibody or an antigen-binding domain thereof that binds CD228 (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen- binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) is provided, wherein the antibody or the antigen-binding domain thereof comprises: i) a heavy chain variable domain (VH) comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 110, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 111, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 112, and a light chain variable domain (VL) comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 116, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 117, and (f) CDR-L3 comprising the amino acid sequence of SEQ ID NO: 118; ii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 113, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 114, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 115, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 119, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 120, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 121; iii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 130, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 131, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 132, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 136, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 137, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 138; iv) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 133, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 134, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 135, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 139, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 140, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 141; Attorney Docket No.01218-0029-00PCT v) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 150, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 151, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 152, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 156, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 157, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 158; vi) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 153, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 154, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 155, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 159, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 160, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 161; vii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 170, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 171, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 172, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 176, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 177, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 178; viii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 173, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 174, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 175, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 179, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 180, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 181; ix) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 190, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 191, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 192, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 196, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 197, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 198; x) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 193, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 194, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 195, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 199, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 200, and (f) CDR- Attorney Docket No.01218-0029-00PCT L3 comprising the amino acid sequence of SEQ ID NO: 201; xi) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 210, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 211, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 212, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 216, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 217, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 218; xii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 213, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 214, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 215, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 219, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 220, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 221; xiii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 230, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 231, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 232, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 236, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 237, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 238; xiv) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 233, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 234, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 235, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 239, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 240, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 241; xv) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 250, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 251, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 252, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 256, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 257, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 258; xvi) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 253, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 254, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 255, and a VL Attorney Docket No.01218-0029-00PCT comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 259, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 260, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 261; xvii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 270, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 271, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 272, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 276, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 277, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 278; or xviii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 273, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 274, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 275, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 279, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 280, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 281. [0096] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: i) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 210, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 211, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 212, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 216, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 217, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 218; ii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 213, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 214, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 215, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 219, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 220, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 221; iii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 250, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 251, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 252, and a VL Attorney Docket No.01218-0029-00PCT comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 256, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 257, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 258; or iv) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 253, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 254, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 255, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 259, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 260, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 261. [0097] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 210, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 211, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 212, and a VL comprising (d) CDR- L1 comprising the amino acid sequence of SEQ ID NO: 216, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 217, and (f) CDR-L3 comprising the amino acid sequence of SEQ ID NO: 218; or a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 213, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 214, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 215, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 219, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 220, and (f) CDR-L3 comprising the amino acid sequence of SEQ ID NO: 221. [0098] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: i) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 122, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 124; ii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least Attorney Docket No.01218-0029-00PCT 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 142, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 144; iii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 162, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 164; iv) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 182, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 184; v) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 202, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 204; vi) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 222, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 224; vii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least Attorney Docket No.01218-0029-00PCT 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 242, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 244; viii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 262, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 264; or ix) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 282, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 284. [0099] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 222, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 224; or a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 262, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 264. [00100] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an Attorney Docket No.01218-0029-00PCT antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 222, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 224. [00101] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: i) a VH comprising the amino acid sequence of SEQ ID NO: 122, and a VL comprising the amino acid sequence of SEQ ID NO: 124; ii) a VH comprising the amino acid sequence of SEQ ID NO: 142, and a VL comprising the amino acid sequence of SEQ ID NO: 144; iii) a VH comprising the amino acid sequence of SEQ ID NO: 162, and a VL comprising the amino acid sequence of SEQ ID NO: 164; iv) a VH comprising the amino acid sequence of SEQ ID NO: 182, and a VL comprising the amino acid sequence of SEQ ID NO: 184; v) a VH comprising the amino acid sequence of SEQ ID NO: 202, and a VL comprising the amino acid sequence of SEQ ID NO: 204; vi) a VH comprising the amino acid sequence of SEQ ID NO: 222, and a VL comprising the amino acid sequence of SEQ ID NO: 224; vii) a VH comprising the amino acid sequence of SEQ ID NO: 242, and a VL comprising the amino acid sequence of SEQ ID NO: 244; viii) a VH comprising the amino acid sequence of SEQ ID NO: 262, and a VL comprising the amino acid sequence of SEQ ID NO: 264; or ix) a VH comprising the amino acid sequence of SEQ ID NO: 282, and a VL comprising the amino acid sequence of SEQ ID NO: 284. [00102] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins Attorney Docket No.01218-0029-00PCT disclosed herein) comprises: a VH comprising the amino acid sequence of SEQ ID NO: 222, and a VL comprising the amino acid sequence of SEQ ID NO: 224; or a VH comprising the amino acid sequence of SEQ ID NO: 262, and a VL comprising the amino acid sequence of SEQ ID NO: 264. [00103] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a VH comprising the amino acid sequence of SEQ ID NO: 222, and a VL comprising the amino acid sequence of SEQ ID NO: 224. [00104] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: i) a heavy chain (HC) comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 126, and a light chain (LC) comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 128; ii) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 146, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 148; iii) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 166, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 168; iv) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least Attorney Docket No.01218-0029-00PCT 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 186, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 188; v) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 206, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 208; vi) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 226, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 228; vii) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 246, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 248; viii) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 266, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 268; or ix) a HC comprising an amino acid sequence having at least 80%, at least 85%, at least Attorney Docket No.01218-0029-00PCT 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 286, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 288. [00105] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 226, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 228; or a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 266, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 268. [00106] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a HC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 226, and a LC comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 228. [00107] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: i) a HC comprising the amino acid sequence of SEQ ID NO: 126, and a LC comprising the amino acid sequence of SEQ ID NO: 128; Attorney Docket No.01218-0029-00PCT ii) a HC comprising the amino acid sequence of SEQ ID NO: 146, and a LC comprising the amino acid sequence of SEQ ID NO: 148; iii) a HC comprising the amino acid sequence of SEQ ID NO: 166, and a LC comprising the amino acid sequence of SEQ ID NO: 168; iv) a HC comprising the amino acid sequence of SEQ ID NO: 186, and a LC comprising the amino acid sequence of SEQ ID NO: 188; v) a HC comprising the amino acid sequence of SEQ ID NO: 206, and a LC comprising the amino acid sequence of SEQ ID NO: 208; vi) a HC comprising the amino acid sequence of SEQ ID NO: 226, and a LC comprising the amino acid sequence of SEQ ID NO: 228; vii) a HC comprising the amino acid sequence of SEQ ID NO: 246, and a LC comprising the amino acid sequence of SEQ ID NO: 248; viii) a HC comprising the amino acid sequence of SEQ ID NO: 266, and a LC comprising the amino acid sequence of SEQ ID NO: 268; or ix) a HC comprising the amino acid sequence of SEQ ID NO: 286, and a LC comprising the amino acid sequence of SEQ ID NO: 288. [00108] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a HC comprising the amino acid sequence of SEQ ID NO: 226, and a LC comprising the amino acid sequence of SEQ ID NO: 228; or a HC comprising the amino acid sequence of SEQ ID NO: 266, and a LC comprising the amino acid sequence of SEQ ID NO: 268. [00109] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) comprises: a HC comprising the amino acid sequence of SEQ ID NO: 226, and a LC comprising the amino acid sequence of SEQ ID NO: 228. [00110] In some embodiments, the antibody that binds CD228 provided herein (e.g., an independent antibody or an antibody that is comprised in any one of the fusion proteins disclosed herein) is a monoclonal antibody. In some embodiments, the antibody that binds CD228 provided herein is a humanized or chimeric antibody. In some embodiments, the antibody that binds CD228 provided herein is an IgG1, IgG2, IgG3, or IgG4 antibody. Attorney Docket No.01218-0029-00PCT [00111] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) binds CD228 with a K _D value of 175 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 150 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 100 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 50 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 75 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 25 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 20 nM or less. In some embodiments, the antibody or the antigen- binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 15 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 10 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 5 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 4 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 3 nM or less. In some embodiments, the antibody or the antigen- binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 2.5 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 2 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 1.5 nM or less. In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein binds CD228 with a K _D value of 1 nM or less. [00112] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) binds cynomolgus monkey CD228. In some such embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein comprises: i) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 170, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 171, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 172, and a VL Attorney Docket No.01218-0029-00PCT comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 176, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 177, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 178; ii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 173, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 174, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 175, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 179, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 180, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 181; iii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 210, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 211, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 212, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 216, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 217, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 218; iv) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 213, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 214, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 215, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 219, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 220, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 221; v) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 250, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 251, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 252, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 256, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 257, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 258; or vi) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 253, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 254, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 255, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 259, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 260, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 261. In some such embodiments, the antibody or the antigen-binding domain thereof that binds CD228 Attorney Docket No.01218-0029-00PCT provided herein comprises a set of CDRs as set forth above and: i) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 182, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 184; ii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 222, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 224; or iii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 262, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 264. In some such embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein comprises: a VH comprising the amino acid sequence of SEQ ID NO: 182, and a VL comprising the amino acid sequence of SEQ ID NO: 184; a VH comprising the amino acid sequence of SEQ ID NO: 222, and a VL comprising the amino acid sequence of SEQ ID NO: 224; or a VH comprising the amino acid sequence of SEQ ID NO: 262, and a VL comprising the amino acid sequence of SEQ ID NO: 264. [00113] In some embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen-binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) does not bind murine CD228, or binds murine CD228 with at least 100-fold reduced affinity compared to human CD228. In some embodiments, the antibody or the antigen- binding domain thereof that binds CD228 provided herein does not bind transferrin or lactotransferrin. In some such embodiments, the antibody or the antigen-binding domain thereof Attorney Docket No.01218-0029-00PCT that binds CD228 provided herein comprises: i) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 170, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 171, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 172, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 176, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 177, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 178; ii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 173, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 174, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 175, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 179, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 180, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 181; iii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 190, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 191, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 192, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 196, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 197, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 198; iv) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 193, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 194, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 195, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 199, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 200, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 201; v) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 210, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 211, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 212, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 216, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 217, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 218; vi) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 213, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 214, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 215, and a VL Attorney Docket No.01218-0029-00PCT comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 219, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 220, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 221; vii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 230, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 231, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 232, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 236, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 237, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 238; viii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 233, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 234, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 235, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 239, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 240, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 241; ix) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 250, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 251, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 252, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 256, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 257, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 258; x) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 253, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 254, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 255, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 259, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 260, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 261; xi) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: 270, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 271, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 272, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 276, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 277, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 278; or xii) a VH comprising (a) CDR-H1 comprising the amino acid sequence of SEQ ID NO: Attorney Docket No.01218-0029-00PCT 273, (b) CDR-H2 comprising the amino acid sequence of SEQ ID NO: 274, and (c) CDR-H3 comprising the amino acid sequence of SEQ ID NO: 275, and a VL comprising (d) CDR-L1 comprising the amino acid sequence of SEQ ID NO: 279, (e) CDR-L2 comprising the amino acid sequence of SEQ ID NO: 280, and (f) CDR- L3 comprising the amino acid sequence of SEQ ID NO: 281. In some such embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein comprises a set of CDRs as set forth above and: i) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 182, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 184; ii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 202, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 204; iii) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 222, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 224; iv) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 242, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 244; or Attorney Docket No.01218-0029-00PCT v) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 262, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 264; or vi) a VH comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 282, and a VL comprising an amino acid sequence having at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 284. In some such embodiments, the antibody or the antigen-binding domain thereof that binds CD228 provided herein comprises: a VH comprising the amino acid sequence of SEQ ID NO: 182, and a VL comprising the amino acid sequence of SEQ ID NO: 184; a VH comprising the amino acid sequence of SEQ ID NO: 202, and a VL comprising the amino acid sequence of SEQ ID NO: 204; a VH comprising the amino acid sequence of SEQ ID NO: 222, and a VL comprising the amino acid sequence of SEQ ID NO: 224; a VH comprising the amino acid sequence of SEQ ID NO: 242, and a VL comprising the amino acid sequence of SEQ ID NO: 244; a VH comprising the amino acid sequence of SEQ ID NO: 262, and a VL comprising the amino acid sequence of SEQ ID NO: 264; or a VH comprising the amino acid sequence of SEQ ID NO: 282, and a VL comprising the amino acid sequence of SEQ ID NO: 284. [00114] In some embodiments, an antibody or an antigen-binding domain thereof that binds CD228 (e.g., an independent antibody or antigen-binding domain thereof, or an antibody or antigen- binding domain thereof that is comprised in any one of the fusion proteins disclosed herein) is provided, wherein the antibody or the antigen-binding domain thereof competes for binding with CD228 with any one of the antibodies or antigen-binding domains thereof disclosed herein. In some such embodiments, the antibody is a monoclonal antibody. In some such embodiments, the antibody is a humanized or chimeric antibody. In some such embodiments, the antibody is an IgG1, IgG2, IgG3, or IgG4 antibody. In some such embodiments, the antibody or the antigen-binding domain thereof binds CD228 with a K _D value of 175 nM or less. In some such embodiments, the antibody or the antigen-binding domain thereof binds cynomolgus monkey CD228. In some such embodiments, the antibody or the antigen-binding domain thereof does not bind murine CD228, or binds murine CD228 with at least 100-fold reduced affinity compared to human CD228. In some Attorney Docket No.01218-0029-00PCT such embodiments, the antibody or the antigen-binding domain thereof does not bind transferrin or lactotransferrin. In some such embodiments, the antibody or antigen-binding domain thereof competes for binding with CD228 with an antibody or antigen-binding domain thereof comprising: a VH comprising the amino acid sequence of SEQ ID NO: 122, and a VL comprising the amino acid sequence of SEQ ID NO: 124; a VH comprising the amino acid sequence of SEQ ID NO: 142, and a VL comprising the amino acid sequence of SEQ ID NO: 144; a VH comprising the amino acid sequence of SEQ ID NO: 162, and a VL comprising the amino acid sequence of SEQ ID NO: 164; a VH comprising the amino acid sequence of SEQ ID NO: 182, and a VL comprising the amino acid sequence of SEQ ID NO: 184; a VH comprising the amino acid sequence of SEQ ID NO: 202, and a VL comprising the amino acid sequence of SEQ ID NO: 204; a VH comprising the amino acid sequence of SEQ ID NO: 222, and a VL comprising the amino acid sequence of SEQ ID NO: 224; a VH comprising the amino acid sequence of SEQ ID NO: 242, and a VL comprising the amino acid sequence of SEQ ID NO: 244; a VH comprising the amino acid sequence of SEQ ID NO: 262, and a VL comprising the amino acid sequence of SEQ ID NO: 264; or a VH comprising the amino acid sequence of SEQ ID NO: 282, and a VL comprising the amino acid sequence of SEQ ID NO: 284. In some such embodiments, the antibody or antigen-binding domain thereof competes for binding with CD228 with an antibody or antigen-binding domain thereof comprising: a VH comprising the amino acid sequence of SEQ ID NO: 202, and a VL comprising the amino acid sequence of SEQ ID NO: 204; a VH comprising the amino acid sequence of SEQ ID NO: 242, and a VL comprising the amino acid sequence of SEQ ID NO: 244; or a VH comprising the amino acid sequence of SEQ ID NO: 262, and a VL comprising the amino acid sequence of SEQ ID NO: 264. In some such embodiments, the antibody or antigen-binding domain thereof competes for binding with CD228 with an antibody or antigen-binding domain thereof comprising: a VH comprising the amino acid sequence of SEQ ID NO: 202, and a VL comprising the amino acid sequence of SEQ ID NO: 204; or a VH comprising the amino acid sequence of SEQ ID NO: 242, and a VL comprising the amino acid sequence of SEQ ID NO: 244. In some such embodiments, competition for binding with CD228 is measured by flow cytometry using at least one labeled antibody or antigen-binding domain thereof. [00115] In some embodiments, CDR sequences disclosed herein are defined according to the Kabat numbering scheme as described in Kabat et al. (1991),“Sequences of Proteins of Immunological Interest,” 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD. In some embodiments, CDR sequences disclosed herein are defined according to the IMGT method as described in Lefranc, M.-P., The Immunologist, 7, 132-136 (1999). [00116] Antibodies specifically binding to CD228 as included in fusion proteins of the disclosure may comprise an Fc part which allows for extending the in vivo half-life of the bispecific binding molecule of the disclosure. In some embodiments, such Fc part is preferably from human Attorney Docket No.01218-0029-00PCT origin, more preferably a human Fc part of an IgG1 or lgG4 antibody, even more preferably an engineered human Fc part of an IgG1 or lgG4 with activating or silencing effector functions. In some embodiments, silencing effector functions may be preferred over activating effector functions. In some embodiments, such an Fc part is an engineered to silence effector functions with mutation(s) at positions 234 and/or 235, numbering according to EU index of Kabat (Johnson and Wu, Nucleic Acids Res, 2000). In some embodiments, mutations in positions F234 and L235 of a provided anti-CD228 antibody may be introduced to silence effector functions. In other embodiments, mutations in positions D265 and P329 of a provided anti-CD228 antibody may be introduced to silence effector function. Numbering for both sets of these potential mutations is according to the EU index of Kabat (Shields et al., J Biol Chem, 2001). In some embodiments, the provided CD228 antibody has an engineered IgG4 backbone with the mutations S228P, F234A and L235A. [00117] Various techniques for the production of antibodies and antigen-binding domains thereof are well known in the art and described, e.g., in Altshuler et al. (2010). Thus, for example, polyclonal antibodies can be obtained from the blood of an animal following immunization with an antigen in mixture with additives and adjuvants and monoclonal antibodies can be produced by any technique which provides antibodies produced by continuous cell line cultures. Examples of such techniques are described, e.g., Harlow and Lane (1999), (1988), and include the hybridoma technique originally described by Köhler and Milstein, 1975, the trioma technique, the human B cell hybridoma technique (see e.g., Li et al., Proc Natl Acad Sci U S A, 2006; Kozbor and Roder, Immunol Today, 1983) and the EBV-hybridoma technique to produce human monoclonal antibodies (Cole et al., Cancer Res, 1984). Furthermore, recombinant antibodies may be obtained from monoclonal antibodies or can be prepared de novo using various display methods such as phage, ribosomal, mRNA, or cell display. In some embodiments, a suitable system for the expression of the recombinant (humanized) antibodies or fragments thereof may be selected from, for example, bacteria, yeast, insects, mammalian cell lines or transgenic animals or plants (see, e.g., US Patent No.6,080,560; Holliger and Hudson, Nat Biotechnol, 2005). Further, techniques described for the production of single chain antibodies (see, inter alia, US Patent No.4,946,778) can be adapted to produce single chain antibodies specific for the target of this invention. Surface plasmon resonance as employed in the BIAcore system can be used to increase the efficiency of phage antibodies. B. Exemplary fusion proteins specific for CD137 and CD228 of the disclosure. [00118] In some embodiments, a provided fusion protein contains at least two subunits in any order: (1) a first subunit that comprises a full-length antibody or an antigen-binding domain thereof specific for CD228, and (2) a second subunit that comprises a lipocalin mutein specific for CD137. In some embodiments, in the fusion protein provided herein, the antibody or the antigen- Attorney Docket No.01218-0029-00PCT binding domain thereof is any one of the antibodies or the antigen-binding domains thereof disclosed herein, in particular any one of the antibodies or the antigen-binding domains thereof disclosed in Section IV.A above. [00119] In some embodiments, a provided fusion protein also may contain at least one additional subunit, for example, a third subunit. For instance, a fusion protein may contain a third subunit specific for CD137. In some embodiments, a third subunit may be or comprise a lipocalin mutein specific for CD137. For example, two lipocalin muteins may be fused to a first antibody subunit, one at the C-terminus and one at the N-terminus of the antibody. In some embodiments, lipocalin muteins may be fused to the heavy chain or light chain of an antibody. [00120] In some embodiments, provided fusion proteins may comprise one or more additional subunits (e.g., a fourth, fifth, or sixth subunit). [00121] In some embodiments, at least one subunit may be fused at its N-terminus and/or its C-terminus to another subunit. [00122] In some embodiments, at least one subunit can be linked to another subunit via a linker. In some further embodiments, a linker is a peptide linker, for example, a glycine-serine (GS) linker, such as an unstructured GS linker or a glycosylated GS linker, or a proline-alanine- serine polymer (PAS) linker. In some embodiments, the GS linker is a (Gly ₄ Ser) ₃ linker ((G ₄ S) ₃ ) as shown in SEQ ID NO: 13. Other exemplary linkers are shown in SEQ ID NOs: 14-23. In some embodiments, a peptide linker may have from 1 to 50 amino acids, such as 1, 2, 3, 4, 5, 10, 11, 12, 13, 14, 15, 16, 1718, 19, 20, 25, 30, 35, 40, 45 or 50 amino acids. For example, when a first subunit comprises an antibody, such as a full-length antibody, a second subunit may be linked via a peptide linker between the N-terminus of the second subunit and the C-terminus of a heavy chain constant region (CH) of said antibody. In some further embodiments, a third subunit may be linked via a peptide linker between the N-terminus of the third subunit and the C-terminus of a light chain constant region (CL) of said antibody. [00123] In some embodiments, one subunit can be linked to another subunit as essentially described in Figures 11A-11I. Generally, one subunit may be fused at its N-terminus and/or its C-terminus to another subunit. For example, in some embodiments, a lipocalin mutein subunit can be fused at its N-terminus and/or its C-terminus to an antibody subunit. For further example, one lipocalin mutein can be linked, preferably via a peptide linker, to the C-terminus of the antibody heavy chain (HC), the N-terminus of the HC, the C-terminus of the antibody light chain (LC), and/or the N-terminus of the LC (Figures 11A-11D). [00124] In some embodiments, a lipocalin mutein subunit can be fused at its N-terminus and/or its C-terminus to an antibody fragment. For example, in some embodiments, a lipocalin Attorney Docket No.01218-0029-00PCT mutein may be linked, preferably via a peptide linker, to the C-terminus of a heavy chain constant region (CH) or the C-terminus of a light chain constant region (CL) of the antibody. [00125] In some embodiments, when one subunit comprises an antibody, such as a full- length antibody, the antibody and a second subunit may be linked between the N-terminus of the second subunit and the C-terminus of a heavy chain constant region (CH) of said antibody. [00126] In some embodiments, a third subunit may be linked via the N-terminus of the third subunit and the C-terminus of a light chain constant region (CL) of said antibody. [00127] In some embodiments, the fusion protein comprises an antibody, such as a full- length antibody, specific for CD228 and two lipocalin muteins specific for CD137, wherein each of the lipocalin muteins is fused at its N-terminus via a peptide linker to the C-terminus of a heavy chain of the antibody. [00128] In some embodiments, the fusion protein comprises an antibody, such as a full- length antibody, specific for CD228 and two lipocalin muteins specific for CD137, wherein each of the lipocalin muteins is fused at its N-terminus via a peptide linker to the C-terminus of a light chain of the antibody. [00129] In some embodiments, with respect to a fusion protein of the disclosure, wherein at least one subunit may be or comprise an antibody, such as a full-length antibody, the Fc function of the Fc region of the antibody to Fc receptor-positive cell may be preserved at the same time while the fusion protein is simultaneously engaging CD137 and CD228. [00130] In some embodiments, wherein at least one subunit of a provided fusion protein may be or comprise an antibody, such as a full-length antibody, the Fc function of the Fc region of the antibody to Fc receptor-positive cell may be reduced or fully suppressed by protein engineering while the fusion protein is simultaneously engaging CD137 and CD228. In some embodiments, this may be achieved, for example, by switching from the IgG1 backbone to IgG4, as IgG4 is known to display reduced Fc-gamma receptor interactions compared to IgG1. In some embodiments, to further reduce the residual binding to Fc-gamma receptors, mutations may be introduced into the IgG4 backbone such as F234A and L235A. In some embodiments, an S228P mutation may also be introduced into the IgG4 backbone to minimize the exchange of IgG4 half- antibody (Silva et al., J Biol Chem, 2015). In some embodiments, F234A and L235A mutations may be introduced for decreased ADCC and ADCP (Glaesner et al., Diabetes Metab Res Rev, 2010) and/or M428L and N434S mutations or M252Y, S254T, and T256E mutations for extended serum half-life (Dall'Acqua et al., J Biol Chem, 2006; Zalevsky et al., Nat Biotechnol, 2010). In some embodiments, an additional N297A mutation may be present in the antibody heavy chain of the fusion protein in order to remove the natural glycosylation motif. Attorney Docket No.01218-0029-00PCT [00131] In some embodiments, the Fc portion of an antibody included in a fusion protein of the disclosure may contribute to maintaining the serum levels of the fusion protein. For example, when the Fc portion binds to Fc receptors on endothelial cells and phagocytes, the fusion protein may become internalized and recycled back to the bloodstream, enhancing its half-life within the body. [00132] In one aspect, fusion proteins of the disclosure bind CD137 with high affinity. In another aspect, provided fusion proteins bind CD228 with high affinity. In some preferred embodiments, provided fusion proteins simultaneously bind CD137 and CD228. In some embodiments, the simultaneous binding to CD137 and CD228 allows provided fusion proteins to exhibit a durable anti-tumor or anti-infection response. [00133] In some embodiments, a fusion protein of the disclosure may be able to bind CD228 with a K _D value of at most about 200 nM, at most about 180 nM, at most about 150 nM, at most about 120 nM, at most about 100 nM, at most about 70 nM, or at most about 35 nM or even lower, such as about 30 nM or lower, about 25 nM or lower, or about 20 nM or lower. In some embodiments, a fusion protein of the disclosure may be able to bind CD228 with a K _D value comparable to or lower than the K _D value of the antibody specific for CD228 as included in such fusion protein or a variant of said antibody, such as an antibody having the heavy and light chains provided by SEQ ID NOs: 74 and 76 or SEQ ID NOs: 77 and 79. The K _D values of provided fusion proteins may be measured, for example, in a surface plasmon resonance (SPR) assay, such as an SPR assay as essentially described in Example 10. [00134] In some embodiments, fusion proteins of the disclosure are cross-reactive with cynomolgus CD228. In some embodiments, a fusion protein of the disclosure may be able to bind cynomolgus CD228 with a K _D value of at most about 200 nM, at most about 180 nM, at most about 150 nM, at most about 120 nM, at most about 100 nM, at most about 70 nM, or at most about 35 nM or even lower, such as about 30 nM or lower, about 25 nM or lower, or about 20 nM or lower. The K _D values of provided fusion proteins may be measured, for example, in a surface plasmon resonance (SPR) assay, such as an SPR assay as essentially described in Example 10. [00135] In some embodiments, a fusion protein of the disclosure may be able to bind CD137 with a K _D value of at most about 10 nM or even lower, such as about 7 nM or lower, about 6 nM or lower, about 5 nM or lower, or about 4 nM or lower. In some embodiments, a fusion protein of the disclosure may be able to bind CD137 with a K _D value comparable to or lower than the K _D value of the lipocalin mutein specific for CD137 that is included in a particular fusion protein, e.g., SEQ ID NO: 40, or the lipocalin mutein fused to the Fc region of an antibody, e.g., SEQ ID NO: 84. The K _D values of provided fusion proteins may be measured, for example, in an SPR Attorney Docket No.01218-0029-00PCT assay, such as an SPR assay as essentially described in Example 10. [00136] In some embodiments, a fusion protein of the disclosure may be able to bind CD228 with an EC ₅₀ value of at most about 2 nM, at most about 1.5, or at most about 1.2 nM or even lower, such as about 1 nM or lower, about 0.9 nM or lower, about 0.8 nM or lower, about 0.7 nM or lower, about 0.6 nM or lower or about 0.5 nM or lower. In some embodiments, a fusion protein of the disclosure may be able to bind CD228 with an EC ₅₀ value comparable to or lower than the K _D value of the antibody specific for CD228 as included in such fusion protein or a variant of said antibody, such as an antibody having the heavy and light chains provided by SEQ ID NOs: 74 and 76 or SEQ ID NOs: 77 and 79. The EC ₅₀ values of provided fusion proteins may be measured, for example, in an enzyme-linked immunosorbent assay (ELISA) assay, such as an ELISA assay as essentially described in Example 11. [00137] In some embodiments, fusion proteins of the disclosure are cross-reactive with cynomolgus CD228. In some embodiments, a provided fusion protein may be able to bind cynomolgus CD228 with an EC ₅₀ value of at most about 2 nM, at most about 1.5, or at most about 1.2 nM or even lower, such as about 1 nM or lower, about 0.9 nM or lower, about 0.8 nM or lower, about 0.7 nM or lower, about 0.6 nM or lower or about 0.5 nM or lower. The EC ₅₀ values of provided fusion proteins may be measured, for example, measured in an ELISA assay, such as an ELISA assay as essentially described in Example 11. [00138] In some embodiments, a fusion protein of the disclosure may be able to bind CD137 with an EC ₅₀ value of at most about 6 nM or even lower, such as about 5 nM or lower, about 4 nM or lower, about 3 nM or lower, about 2.5 nM or lower, about 2 nM or lower, about 1.5 nM or lower, about 1 nM or lower or about 0.8 nM or lower. In some embodiments, a fusion protein of the disclosure may be able to bind CD137 with an EC ₅₀ value comparable to or lower than the EC ₅₀ value of the lipocalin mutein specific for CD137 that is included in a particular fusion protein, e.g., SEQ ID NO: 40, or the lipocalin mutein fused to the Fc region of an antibody, e.g., SEQ ID NO: 84. The EC ₅₀ values of provided fusion proteins may be measured, for example, in an ELISA assay, such as an ELISA assay as essentially described in Example 11. [00139] In some embodiments, fusion proteins of the disclosure are cross-reactive with cynomolgus CD137. In some embodiments, a provided fusion protein may be able to bind cynomolgus CD137 with an EC ₅₀ value of at most about 150 nM or even lower, such as about 120 nM or lower, about 100 nM or lower, about 90 nM or lower, about 80 nM or lower, about 70 nM or lower, about 60 nM or lower, about 50 nM or lower, about 40 nM or lower, or about 30 nM or lower. The EC ₅₀ values of provided fusion proteins may be measured, for example, in an ELISA assay, such as an ELISA assay as essentially described in Example 11. [00140] In some embodiments, fusion proteins of the disclosure may be able to Attorney Docket No.01218-0029-00PCT simultaneously bind CD137 and CD228. In some embodiments, a provided fusion protein may be able to simultaneously bind CD137 and CD228, with an EC ₅₀ value of at most about 20 nM, at most about 15 nM, or at most about 10 nM or even lower, such as about 7 nM or lower, about 5 nM or lower, about 3 nM or lower, about 2.5 nM or lower, about 2 nM or lower, or about 1.5 nM or lower. The simultaneous binding may be determined, for example, in and ELISA assay, such as an ELISA assay as essentially described in Example 12. [00141] In some embodiments, a fusion protein of the disclosure may be able to bind CD137 expressed on a cell with an EC ₅₀ value of at most about 30 nM or even lower, such as about 20 nM or lower, about 15 nM or lower, about 10 nM or lower, about 9 nM or lower, about 7 nM or lower, about 5 nM or lower, about 3 nM or lower, about 2.5 nM or lower, or about 2 nM or lower. The EC ₅₀ value of a provided fusion protein may be measured, for example, in a flow cytometric analysis as essentially described in Example 13. The cell expressing CD137 may be, for example, a CHO cell transfected with human CD137 or cynomolgus CD137. [00142] In some embodiments, a fusion protein of the disclosure may be able to bind CD228 expressed on a cell with an EC ₅₀ value of at most about 20 nM or even lower, such as about 10 nM or lower, about 8 nM or lower, about 6 nM or lower, about 5 nM or lower, about 4 nM or lower, about 3 nM or lower, about 2.5 nM, about 2 nM or lower, or about 1.5 nM or lower. The EC ₅₀ value of a provided fusion protein may be measured, for example, in a flow cytometric analysis as essentially described in Example 13. The cell expressing CD228 may, for example, be a tumor cell expressing CD228, such as a SH-4 cell. [00143] In some embodiments, fusion proteins of the disclosure may be able to co- stimulate T cell proliferation and/or T cell responses. In some embodiments, provided fusion proteins lead to a comparable or stronger T cell activation as compared to that elicited by an anti- CD137 antibody, such as the reference antibody of SEQ ID NOs: 26 and 27. The stimulated T cell response or T cell activation may be measured, for example, in a CD137 bioassay as essentially described in Example 14, or in a functional T cell activation assay as essentially described in Example 18. [00144] In some embodiments, fusion proteins of the disclosure may be able to co- stimulate T cell responses in a CD228-dependent manner. In some embodiments, provided fusion proteins may lead to local induction of the IL-2 production by T cells in the vicinity of CD228- positive cells, such as CD228-transfected cells or CD228-positive tumor cells. “In the vicinity of CD228-positive cells” when used herein refers to a T cell and a CD228-positive cell are brought close to each other through a provided fusion protein which binds CD137 and CD228 simultaneously. The CD228-dependent activation of T cell by provided fusion proteins may be determined, for example, in a CD137 bioassay essentially described in Example 14, or in a Attorney Docket No.01218-0029-00PCT functional T cell activation assay as essentially described in Example 18. [00145] In some preferred embodiments, provided fusion proteins may be able to co- stimulate T cell responses in the presence of CD228-expressing tumor cells and/or in a (CD228- positive) tumor microenvironment. In some embodiments, a provided fusion protein may be able to co-stimulate T cell responses in the presence of CD228-positive tumor cells with an EC ₅₀ value of about 1.5 nM or lower, about 1.2 nM or lower, about 1 nM or lower, about 0.9 nM or lower, about 0.8 nM or lower, about 0.7 nM or lower, about 0.6 nM or lower, about 0.5 nM or lower, about 0.4 nM or lower, about 0.3 nM or lower, about 0.2 nM or lower, or about 0.15 nM or lower. The T cell activation by provided fusion proteins in the presence of CD228-expressing tumor cells and/or in a CD228-positive tumor microenvironment may be assessed, for example, in a CD137 bioassay as essentially described in Example 14. [00146] In some embodiments, provided fusion proteins are not able to co-stimulate T cell responses in the absence of CD228. In some embodiments, provided fusion proteins are not able to co-stimulate T cell responses in the absence of CD228-expressing cells. In some embodiments, a provided fusion protein may be able to discern the presence of CD228 and lead to corresponding T cell activation better than a CD137 antibody shown in SEQ ID NOs: 26 and 27. The CD228-dependent action of the fusion proteins may be determined, for example, in a CD137 bioassay essentially described in Example 14, or in a functional T cell activation assay essentially described in Example 18. [00147] In some embodiments, fusion proteins of the disclosure may be able to induce increased IL-2 secretion, such as in the presence of CD228-positive (tumor) cells. In some preferred embodiments, provided fusion proteins may be able to induce a concentration- dependent IL-2 secretion and/or demonstrate a tendency to induce enhanced IL-2 secretion at higher concentrations, preferably coating concentrations. In some embodiments, provided fusion proteins may lead to increased IL-2 secretion with a better efficiency as compared an anti-CD137 antibody, such as the reference antibody of SEQ ID NOs: 26 and 27. IL-2 secretion may be measured, for example, in a functional T cell activation assay as essentially described in Example 18. [00148] In some embodiments, provided fusion proteins are capable of stimulating CD4+ and/or CD8+ T cell and/or NK cell proliferation. In some embodiments, provided fusion proteins are capable of inducing NF-κB signaling. In some embodiments, provided fusion proteins are capable of inducing increased secretion of cytokines, such as interferon gamma, TNF-alpha, IL- 2, IL-5, IL-10, IL-13, and IP-10 (CXCL10). In some embodiments, provided fusion proteins are capable of inducing increased secretion of cytotoxic factors, such as perforin, granzyme B, and granzyme A. Attorney Docket No.01218-0029-00PCT [00149] In some embodiments, provided fusion proteins have a favorable stability and/or pharmacokinetic profile. In some embodiments, a provided fusion protein has antibody-like pharmacokinetics. In some embodiments, a provided fusion protein has a terminal half-life (e.g., an in vivo terminal half-life in mice) of about 200 hours or longer, about 250 hours or longer, or about 300 hours or longer. In some embodiments, a provided fusion protein has a longer terminal half-life than an anti-CD228 antibody having the amino acid sequences of SEQ ID NOs: 74 and 76 or SEQ ID NOs: 77 and 79, respectively. Pharmacokinetics profiles of provided fusion proteins may be analyzed as described in Example 22. In some embodiments, a favorable pharmacokinetic profile or an antibody-like pharmacokinetics may be considered to be achieved if % of c _max was above 10 % after 336 h. [00150] In some embodiments, fusion proteins of the disclosure may be able to induce increased CD8+ T cell mitochondrial content and/or reduced CD8+ T cell mitochondria depolarization. In some embodiments, provided fusion proteins may induce increased CD8+ T cell mitochondrial content and/or reduced CD8+ T cell mitochondria depolarization in the vicinity of CD228-positive cells, such as CD228-transfected cells or CD228-positive tumor cells. The induction of increased CD8+ T cell mitochondrial content by provided fusion proteins may be determined, for example, by using a polarization-independent dye such as MitoSpy™ Green FM as essentially described in Example 26. The induction of reduced CD8+ T cell mitochondria depolarization by provided fusion proteins may be determined, for example, by using a polarization-dependent dye such as MitoSpy™ Orange CMTMRos as essentially described in Example 26. [00151] In some embodiments, fusion proteins of the disclosure may be able to stimulate exhausted CD8+ T cell proliferation and/or stimulate exhausted CD8+ T cell proliferation in synergy with an anti-PD-1 or anti-PD-L1 antibody. In some embodiments, provided fusion proteins may be able to stimulate exhausted CD8+ T cell proliferation and/or stimulate exhausted CD8+ T cell proliferation in synergy with an anti-PD-1 or anti-PD-L1 antibody, in the vicinity of CD228- positive cells, such as CD228-transfected cells or CD228-positive tumor cells. In some embodiments, exhausted CD8+ T cells have reduced capacity to divide, secrete cytokines (such as IFN-γ), and/or kill tumor cells, compared to nonexhausted CD8+ T cells. In some embodiments, the fusion proteins alone may be able to stimulate exhausted CD8+ T cell proliferation. In some embodiments, the fusion proteins may be able to stimulate exhausted CD8+ T cell proliferation in synergy with an anti-PD-1 or anti-PD-L1 antibody; in such embodiments, the anti-PD-1 or anti- PD-L1 antibody alone may have no or limited ability to stimulate exhausted CD8+ T cell proliferation, but it may be able to enhance the fusion proteins’ ability to stimulate exhausted CD8+ T cell proliferation. In some embodiments, the anti-PD-1 or anti-PD-L1 antibody is nivolumab, pembrolizumab, cemiplimab, dostarlimab, atezolizumab, avelumab, or durvalumab. Attorney Docket No.01218-0029-00PCT Cell proliferation may be determined, for example, by using flow cytometry with a dye such as CFSE as essentially described in Example 28. [00152] In some embodiments, a provided fusion protein comprises an amino acid sequence shown in any one of SEQ ID NOs: 75, 76 and 78-83. [00153] In some embodiments, a provided fusion protein comprises an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or even higher sequence identity to an amino acid sequence shown in any one of SEQ ID NOs: 75, 76 and 78-83. [00154] In some embodiments, a provided fusion protein comprises amino acid sequences having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 95%, at least 97%, at least 98%, or even higher sequence identity to the amino acid sequences shown in SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, SEQ ID NOs: 78 and 83, SEQ ID NOs: 96 and 97, SEQ ID NOs: 98 and 99, SEQ ID NOs: 100 and 101, SEQ ID NOs: 102 and 103, SEQ ID NOs: 104 and 105, or SEQ ID NOs: 106 and 107. [00155] In some embodiments, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, SEQ ID NOs: 78 and 83, SEQ ID NOs: 96 and 97, SEQ ID NOs: 98 and 99, SEQ ID NOs: 100 and 101, SEQ ID NOs: 102 and 103, SEQ ID NOs: 104 and 105, or SEQ ID NOs: 106 and 107. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 80 and 76. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 82 and 79. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 75 and 81. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 78 and 83. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 96 and 97. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 98 and 99. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 100 and 101. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 102 and 103. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 104 and 105. In one embodiment, a provided fusion protein comprises the amino acid sequences shown in SEQ ID NOs: 106 and 107. C. Exemplary lipocalin muteins of the disclosure. [00156] Lipocalins are proteinaceous binding molecules that have naturally evolved to bind ligands. Lipocalins occur in many organisms, including vertebrates, insects, plants, and bacteria. Attorney Docket No.01218-0029-00PCT The members of the lipocalin protein family (Pervaiz and Brew, FASEB J, 1987) are typically small, secreted proteins and have a single polypeptide chain. They are characterized by a range of different molecular-recognition properties: their binding to various, principally hydrophobic small molecules (such as retinoids, fatty acids, cholesterols, prostaglandins, biliverdins, pheromones, tastants, and odorants), and their binding to specific cell-surface receptors and their formation of macromolecular complexes. Although they have, in the past, been classified primarily as transport proteins, it is now clear that the lipocalins fulfill a variety of physiological functions. These include roles in retinol transport, olfaction, pheromone signaling, and the synthesis of prostaglandins. Lipocalins have also been implicated in the regulation of the immune response and the mediation of cell homeostasis (reviewed, e.g., in Flower et al., Biochim Biophys Acta, 2000; Flower, Biochem J, 1996). [00157] Lipocalins share unusually low levels of overall sequence conservation, often with sequence identities of less than 20%. In strong contrast, their overall folding pattern is highly conserved. The central part of the lipocalin structure consists of a single eight-stranded anti- parallel β-sheet closed back on itself to form a continuously hydrogen-bonded β-barrel. This β- barrel forms a central cavity. One end of the barrel is sterically blocked by the N-terminal peptide segment that runs across its bottom as well as three peptide loops connecting the β-strands. The other end of the β-barrel is open to the solvent and encompasses a target-binding site, which is formed by four flexible peptide loops (AB, CD, EF, and GH). It is the diversity of the loops in the otherwise rigid lipocalin scaffold that gives rise to a variety of different binding modes each capable of accommodating targets of different size, shape, and chemical character (reviewed, e.g., in Skerra, Biochim Biophys Acta, 2000; Flower et al., Biochim Biophys Acta, 2000; Flower, Biochem J, 1996). [00158] A lipocalin mutein according to the present disclosure may be a mutein of any lipocalin. Examples of suitable lipocalins (also sometimes designated as “reference lipocalin,” “wild-type lipocalin,” “reference protein scaffolds,” or simply “scaffolds”) of which a mutein may be used include, but are not limited to, tear lipocalin (lipocalin-1, Tlc, or von Ebner’s gland protein), retinol binding protein, neutrophil lipocalin-type prostaglandin D-synthase, β-lactoglobulin, bilin- binding protein (BBP), apolipoprotein D (APOD), neutrophil gelatinase-associated lipocalin (NGAL), α2-microglobulin-related protein (A2m), 24p3/uterocalin (24p3), von Ebner’s gland protein 1 (VEGP 1), von Ebner’s gland protein 2 (VEGP 2), and Major allergen Can f 1 (ALL-1). In related embodiments, a lipocalin mutein is derived from the lipocalin group consisting of human tear lipocalin (hTlc), human neutrophil gelatinase-associated lipocalin (hNGAL), human apolipoprotein D (hAPOD) and the bilin-binding protein of Pieris brassicae. [00159] The amino acid sequence of a lipocalin mutein according to the disclosure may have a high sequence identity as compared to the reference (or wild-type) lipocalin from which it Attorney Docket No.01218-0029-00PCT is derived, for example, hTlc or hNGAL, when compared to sequence identities with another lipocalin (see also above). In this general context, the amino acid sequence of a lipocalin mutein according to the disclosure is at least substantially similar to the amino acid sequence of the corresponding reference (wild-type) lipocalin, with the proviso that there may be gaps (as defined herein) in an alignment that are the result of additions or deletions of amino acids. A respective sequence of a lipocalin mutein of the disclosure, being substantially similar to the sequences of the corresponding reference (wild-type) lipocalin, has, in some embodiments, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 82%, at least 85%, at least 87%, or at least 90% identity, including at least 95% identity to the sequence of the corresponding lipocalin. In this regard, a lipocalin mutein of the disclosure of course may contain substitutions as described herein which renders the lipocalin mutein capable of binding to CD137. [00160] Typically, a lipocalin mutein contains one or more mutated amino acid residues – relative to the amino acid sequence of the wild-type or reference lipocalin, for example, hTlc and hNGAL – in the four loops at the open end that comprise a ligand-binding pocket and define the entrance of the ligand-binding pocket (cf. above). As explained above, these regions are essential in determining the binding specificity of a lipocalin mutein for the desired target. In some embodiments, a lipocalin mutein of the disclosure may also contain mutated amino acid residues regions outside of the four loops. In some embodiments, a lipocalin mutein of the disclosure may contain one or more mutated amino acid residues in one or more of the three peptide loops (designated BC, DE, and FG) connecting the β-strands at the closed end of the lipocalin. In some embodiments, a mutein derived from tear lipocalin, NGAL or a homologue thereof, may have 1, 2, 3, 4, or more mutated amino acid residues at any sequence position in the N-terminal region and/or in the three peptide loops BC, DE, and FG arranged at the end of the β-barrel structure that is located opposite to the natural lipocalin binding pocket. In some embodiments, a mutein derived from tear lipocalin, NGAL or a homologue thereof, may have no mutated amino acid residues in peptide loop DE arranged at the end of the β-barrel structure, compared to wild-type sequence of tear lipocalin, NGAL or a homologue thereof. [00161] In some embodiments, a lipocalin mutein according to the disclosure may include one or more, such as 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or even more mutated amino acid residues in comparison to the amino acid sequence of a corresponding reference (wild-type) lipocalin, provided that such a lipocalin mutein is capable of binding to CD137. In some embodiments, a lipocalin mutein of the disclosure includes at least two, including 2, 3, 4, 5, or even more, mutated amino acid residues, where a native amino acid residue of the corresponding reference (wild-type) lipocalin is substituted by an arginine residue. [00162] Any types and numbers of mutations, including substitutions, deletions, and insertions, are envisaged as long as a provided lipocalin mutein retains its capability to bind Attorney Docket No.01218-0029-00PCT CD137, and/or it has a sequence identity of at least 60%, such as at least 65%, at least 70%, at least 75%, at least 80%, at least 85% or higher identity to the amino acid sequence of the reference (wild-type) lipocalin, for example, mature hTlc or mature hNGAL. [00163] In some embodiments, a substitution is a conservative substitution. In some embodiments, a substitution is a non-conservative substitution or one or more from the exemplary substitutions below. [00164] Specifically, in order to determine whether the amino acid sequence of a lipocalin (mutein) is different from that of a reference (wild-type) lipocalin with regard to a certain position in the amino acid sequence of the reference (wild-type) lipocalin, a skilled artisan can use means and methods well-known in the art, e.g., alignments, either manually or by using computer programs such as BLAST2.0, which stands for Basic Local Alignment Search Tool or ClustalW or any other suitable program which is suitable to generate sequence alignments. Accordingly, the amino acid sequence of a reference (wild-type) lipocalin can serve as “subject sequence” or “reference sequence”, while the amino acid sequence of a lipocalin mutein serves as “query sequence” (see also above). [00165] Conservative substitutions are generally the following substitutions, listed according to the amino acid to be mutated, each followed by one or more replacement(s) that can be taken to be conservative: Ala Ser, Thr, or Val; Arg Lys, Gln, Asn, or His; Asn Gln, Glu, Asp, or His; Asp ^ Glu, Gln, Asn, or His; Gln ^ Asn, Asp, Glu, or His; Glu ^ Asp, Asn, Gln, or His; His ^ Arg, Lys, Asn, Gln, Asp, or Glu; Ile Leu Ile, Val, Met, Ala, Phe, Pro, Tyr, or Trp; Lys ^ Arg, His, Gln, or Asn; Met ^ Thr, Leu, Tyr, Ile, Phe, Val, Ala, Pro, or Trp; Phe ^ Thr, Met, Leu, Tyr, Ile, Pro, Trp, Val, or Ala; Ser ^ Thr, Ala, or Val; Thr Ile, Met, Val, Phe, Pro, or Leu; Trp ^ Tyr, Phe, Met, Ile, or Leu; Tyr ^ Trp, Phe, Ile, Leu, or Met; Val ^ Thr, Ile, Leu, Met, Phe, Ala, Ser, or Pro. Other substitutions are also permissible and can be determined empirically or in accordance with other known conservative or non-conservative substitutions. As a further orientation, the following groups each contain amino acids that can typically be taken to define conservative substitutions for one another: i) Alanine (Ala), Serine (Ser), Threonine (Thr), Valine (Val); ii) Aspartic acid (Asp), Glutamic acid (Glu), Glutamine (Gln), Asparagine (Asn), Histidine (His); iii) Arginine (Arg), Lysine (Lys), Glutamine (Gln), Asparagine (Asn), Histidine (His); iv) Isoleucine (Ile), Leucine (Leu), Methionine (Met), Valine (Val), Alanine (Ala), Phenylalanine (Phe), Threonine (Thr), Proline (Pro); Attorney Docket No.01218-0029-00PCT v) Isoleucine (Ile), Leucine (Leu), Methionine (Met), Phenylalanine (Phe), Tyrosine (Tyr), Tryptophan (Trp). [00166] If such conservative substitutions result in a change in biological activity, then more substantial changes, such as the following, or as further described below in reference to amino acid classes, may be introduced and the products be screened for a desired characteristic. Examples of such more substantial changes are: Ala Leu or Phe; Arg Glu; Asn Ile, Val, or Trp; Asp Met; Cys Pro; Gln Phe; Glu ^ Arg; His ^ Gly; Ile Lys, Glu, or Gln; Leu ^ Lys or Ser; Lys ^ Tyr; Met Glu; Phe Glu, Gln, or Asp; Trp ^ Cys; Tyr ^ Glu or Asp; Val ^ Lys, Arg, His. [00167] In some embodiments, substantial modifications in the physical and biological properties of the lipocalin (mutein) are accomplished by selecting substitutions that differ significantly in their effect on maintaining (a) the structure of the polypeptide backbone in the area of the substitution, for example, as a sheet or helical conformation, (b) the charge or hydrophobicity of the molecule at the target site, or (c) the bulk of the side chain. [00168] Naturally occurring residues are divided into groups based on common side-chain properties: (1) hydrophobic: methionine, alanine, valine, leucine, iso-leucine; (2) neutral hydrophilic: cysteine, serine, threonine, asparagine, glutamine; (3) acidic: aspartic acid, glutamic acid; (4) basic: histidine, lysine, arginine; (5) residues that influence chain orientation: glycine, proline; and (6) aromatic: tryptophan, tyrosine, phenylalanine. In some embodiments, substitutions may entail exchanging a member of one of these classes for a member of another class. [00169] Any cysteine residue not involved in maintaining the proper conformation of the respective lipocalin also may be substituted, generally with serine, to improve the oxidative stability of the molecule and prevent aberrant crosslinking. Conversely, cysteine bond(s) may be added to the lipocalin to improve its stability. D. Exemplary CD137-specific lipocalin muteins of the disclosure. [00170] As noted above, a lipocalin is a polypeptide defined by its supersecondary structure, namely a cylindrical β-pleated sheet supersecondary structural region comprising eight β-strands connected pair-wise by four loops at one end to define thereby a binding pocket. The present disclosure is not limited to lipocalin muteins specifically disclosed herein. In this regard, the disclosure relates to a lipocalin mutein having a cylindrical β-pleated sheet supersecondary structural region comprising eight β-strands connected pair-wise by four loops at one end to define thereby a binding pocket, wherein at least one amino acid of each of at least three of said four loops has been mutated and wherein said lipocalin is effective to bind CD137 with detectable Attorney Docket No.01218-0029-00PCT affinity. [00171] In some embodiments, lipocalin muteins disclosed herein may be or comprise a mutein of mature human tear lipocalin (hTlc). A mutein of mature hTlc may be designated herein as an “hTlc mutein”. In some other embodiments, a lipocalin mutein disclosed herein is a mutein of mature human neutrophil gelatinase-associated lipocalin (hNGAL). A mutein of mature hNGAL may be designated herein as an “hNGAL mutein”. [00172] In one aspect, the present disclosure includes any number of lipocalin muteins derived from a reference (wild-type) lipocalin, preferably derived from mature hTlc or mature hNGAL, that bind CD137 with detectable affinity. In a related aspect, the disclosure includes various lipocalin muteins that are capable of activating the downstream signaling pathways of CD137 by binding to CD137. In this sense, CD137 can be regarded as a non-natural target of the reference (wild-type) lipocalin, preferably hTlc or hNGAL, where “non-natural target” refers to a substance that does not bind to the reference (wild-type) lipocalins under physiological conditions. By engineering reference (wild-type) lipocalins with one or more mutations at certain sequence positions, the present inventors have demonstrated that high affinity and high specificity for the non-natural target, CD137, is possible. In some embodiments, at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 or even more nucleotide triplet(s) encoding certain sequence positions on wild-type lipocalins, a random mutagenesis may be carried out through substitution at these positions by a subset of nucleotide triplets, with the aim of generating a lipocalin mutein which is capable of binding CD137. [00173] In some embodiments, lipocalin muteins of the disclosure may have mutated, including substituted, deleted and inserted, amino acid residue(s) at one or more sequence positions corresponding to sequence positions in the linear polypeptide sequence of a reference lipocalin, preferably hTlc or hNGAL. In some embodiments, the number of amino acid residues of a lipocalin mutein of the disclosure that are mutated in comparison with the amino acid sequence of the reference lipocalin, preferably hTlc or hNGAL, is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more such as 25, 30, 35, 40, 45 or 50, with 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or 11 being preferred and 9, 10 or 11 being even more preferred. However, it is preferred that a lipocalin mutein of the disclosure is still capable of binding CD137. [00174] In some embodiments, a lipocalin mutein of the present disclosure may lack 1, 2, 3, 4 or more amino acids at its N-terminal end and/or 1, 2 or more amino acids at its C-terminal end, in comparison to the respective reference (wild-type) lipocalin; for example, SEQ ID NOs: 32-38. In some embodiments, the present disclosure encompasses hTlc muteins as defined above, in which the first one, two, three, or four N-terminal amino acid residues of the sequence of mature hTlc (His-His-Leu-Leu; positions 1-4) and/or the last one or two C-terminal amino acid Attorney Docket No.01218-0029-00PCT residues (Ser-Asp; positions 157-158) of the linear polypeptide sequence of the mature hTlc have been deleted (e.g., SEQ ID NOs: 32-38). In some embodiments, the present disclosure encompasses hNGAL muteins as defined above, in which amino acid residues (Lys-Asp-Pro, positions 46-48) of the linear polypeptide sequence of the mature hNGAL have been deleted (SEQ ID NO: 43). Further, a lipocalin mutein of the disclosure may include the wild-type (natural) amino acid sequence of the reference (wild-type) lipocalin, preferably hTlc or hNGAL, outside the mutated amino acid sequence positions. [00175] In some embodiments, one or more mutated amino acid residues incorporated into a lipocalin mutein of the disclosure do not substantially hamper or interfere with the binding activity to the designated target and the folding of the mutein. Such mutations, including substitutions, deletions and insertions, can be accomplished at the DNA level using established standard methods (Sambrook and Russell, 2001, Molecular cloning: a laboratory manual). In some embodiments, (a) mutated amino acid residue(s) at one or more sequence positions corresponding to the linear polypeptide sequence of the reference (wild-type) lipocalin, preferably hTlc or hNGAL, is/are introduced through random mutagenesis by substituting the nucleotide triplet(s) encoding the corresponding sequence positions of the reference lipocalin with a subset of nucleotide triplets. [00176] In some embodiments, a provided lipocalin mutein that binds CD137 with detectable affinity may include at least one amino acid substitution of a native cysteine residue by another amino acid, for example, a serine residue. In some embodiments, a lipocalin mutein that binds CD137 with detectable affinity may include one or more non-native cysteine residues substituting one or more amino acids of a reference (wild-type) lipocalin, preferably hTlc or hNGAL. In some embodiments, a lipocalin mutein according to the disclosure includes at least two amino acid substitutions of a native amino acid by a cysteine residue, hereby to form one or more cysteine bridges. In some embodiments, said cysteine bridge may connect at least two loop regions. The definition of these regions is used herein in accordance with Skerra, Biochim Biophys Acta (2000), Flower (1996) and Breustedt et al. (2005). [00177] Generally, a lipocalin mutein of the disclosure may have at least about 70%, including at least about 80%, such as at least about 85%, amino acid sequence identity with the amino acid sequence of the mature hTlc (SEQ ID NO: 1) or mature hNGAL (SEQ ID NO: 2). [00178] In some aspects, the present disclosure provides CD137-binding hTlc muteins. In this regard, the disclosure provides one or more hTlc muteins that are capable of binding CD137 with an affinity measured by a K _D of about 300 nM, 200 nM, 150 nM, 100 nM, or lower. In some embodiments, provided hTlc muteins are capable of binding CD137 with an EC ₅₀ value of about 250 nM, 150 nM, 100 nM, 50 nM, 20 nM, or even lower. In some other embodiments, the CD137- Attorney Docket No.01218-0029-00PCT binding hTlc muteins may be cross-reactive with cynomolgus CD137 (cyCD137). [00179] In some embodiments, an hTlc mutein of the disclosure may interfere with the binding of CD137L to CD137. [00180] In some embodiments, provided hTlc muteins may comprise a mutated amino acid residue at one or more positions corresponding to positions 5, 26-31, 33-34, 42, 46, 52, 56, 58, 60-61, 65, 71, 85, 94, 101, 104-106, 108, 111, 114, 121, 133, 148, 150, and 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1). [00181] In some embodiments, provided hTlc muteins may comprise a mutated amino acid residue at one or more positions corresponding to positions 26-34, 55-58, 60-61, 65, 104-106, and 108 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1). [00182] In some embodiments, provided hTlc muteins may further comprise a mutated amino acid residue at one or more positions corresponding to positions 101, 111, 114 and 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1). [00183] In some embodiments, provided hTlc muteins may comprise a mutated amino acid residue at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, or even more positions corresponding to positions 5, 26-31, 33-34, 42, 46, 52, 56, 58, 60-61, 65, 71, 85, 94, 101, 104-106, 108, 111, 114, 121, 133, 148, 150 and 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1). In some preferred embodiments, the provided hTlc muteins are capable of binding CD137, in particular human CD137. [00184] In some embodiments, provided hTlc muteins may comprise a mutated amino acid residue at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, or even more positions corresponding to positions 26-34, 55-58, 60-61, 65, 104-106 and 108 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1). In some preferred embodiments, the provided hTlc muteins are capable of binding CD137, in particular human CD137. [00185] In some embodiments, a lipocalin mutein according to the disclosure may include at least one amino acid substitution of a native cysteine residue by, e.g., a serine residue. In some embodiments, an hTlc mutein according to the disclosure includes an amino acid substitution of a native cysteine residue at positions corresponding to positions 61 and/or 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO:1) by another amino acid, such as a serine residue. In this context it is noted that it has been found that removal of the structural disulfide bond (on the level of a respective naïve nucleic acid library) of wild-type hTlc that is formed by the cysteine residues 61 and 153 (cf. Breustedt et al., J Biol Chem, 2005) may provide hTlc muteins that are not only stably folded but are also able to bind a given non-natural target with high affinity. In some embodiments, the elimination of the structural disulfide bond may provide Attorney Docket No.01218-0029-00PCT the further advantage of allowing for the generation or deliberate introduction of non-natural disulfide bonds into muteins of the disclosure, thereby, increasing the stability of the muteins. However, hTlc muteins that bind CD137 and that have the disulfide bridge formed between Cys 61 and Cys 153 are also part of the present disclosure. [00186] In some particular embodiments, an hTlc mutein of the disclosure may include the amino acid substitutions Cys 61 → Ala, Phe, Lys, Arg, Thr, Asn, Gly, Gln, Asp, Asn, Leu, Tyr, Met, Ser, Pro or Trp and/or Cys 153 → Ser or Ala, at positions corresponding to positions 61 and/or 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO:1). [00187] In some embodiments, either two or all three of the cysteine codons at positions corresponding to positions 61, 101 and 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO:1) are replaced by a codon of another amino acid. Further, in some embodiments, an hTlc mutein according to the disclosure includes an amino acid substitution of a native cysteine residue at the position corresponding to position 101 of the linear polypeptide sequence of mature hTlc (SEQ ID NO:1) by a serine residue or a histidine residue. [00188] In some embodiments, a mutein according to the disclosure comprises an amino acid substitution of a native amino acid by a cysteine residue at positions corresponding to positions 28 or 105 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1). Further, in some embodiments, a mutein according to the disclosure comprises an amino acid substitution of a native arginine residue at the position corresponding to position 111 of the linear polypeptide sequence of mature hTlc (SEQ ID NO:1) by a proline residue. Further, in some embodiments, a mutein according to the disclosure comprises an amino acid substitution of a native lysine residue at the position corresponding to position 114 of the linear polypeptide sequence of mature hTlc (SEQ ID NO:1) by a tryptophan residue or a glutamic acid. [00189] In some embodiments, provided CD137-binding hTlc muteins may comprise, at one or more positions corresponding to positions 5, 26-31, 33-34, 42, 46, 52, 56, 58, 60-61, 65, 71, 85, 94, 101, 104-106, 108, 111, 114, 121, 133, 148, 150, and 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1), one or more of the following mutated amino acid residues: Ala 5 → Val or Thr; Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Thr 42 → Ser; Gly 46 → Asp; Lys 52 → Glu; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Lys 65 → Arg or Asn; Thr 71 → Ala; Val 85 → Asp; Lys 94 → Arg or Glu; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; Lys 121 → Glu; Ala 133 → Thr; Arg 148 → Ser; Ser 150 → Ile; and Cys 153 → Ser. In some embodiments, an hTlc mutein of the disclosure comprises two or more, such as 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, or more, or even all mutated amino acid residues at these sequence positions Attorney Docket No.01218-0029-00PCT of mature hTlc (SEQ ID NO: 1). [00190] In some embodiments, provided CD137-binding hTlc muteins may comprise one of the following sets of mutated amino acid residues in comparison with the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1): (a) Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; and Cys 153 → Ser; (b) Ala 5 → Thr; Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Lys 65 → Arg; Val 85 → Asp; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; Lys 121 → Glu; Ala 133 → Thr; and Cys 153 → Ser; (c) Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Lys 65 → Asn; Lys 94 → Arg; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; Lys 121 → Glu; Ala 133 → Thr; and Cys 153 → Ser; (d) Ala 5 → Val; Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Lys 65 → Arg; Lys 94 → Glu; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; Lys 121 → Glu; Ala 133 → Thr; and Cys 153 → Ser; (e) Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Thr 42 → Ser; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; Ser 150 → Ile; and Cys 153 → Ser; (f) Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Lys 52 → Glu; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Thr 71 → Ala; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; Ala 133 → Thr; Arg 148 → Ser; Ser 150 → Ile; and Cys 153 → Ser; and Attorney Docket No.01218-0029-00PCT (g) Ala 5 → Thr; Arg 26 → Glu; Glu 27 → Gly; Phe 28 → Cys; Pro 29 → Arg; Glu 30 → Pro; Met 31 → Trp; Leu 33 → Ile; Glu 34 → Phe; Gly 46 → Asp; Leu 56 → Ala; Ser 58 → Asp; Arg 60 → Pro; Cys 61 → Ala; Thr 71 → Ala; Cys 101 → Ser; Glu 104 → Val; Leu 105 → Cys; His 106 → Asp; Lys 108 → Ser; Arg 111 → Pro; Lys 114 → Trp; Ser 150 → Ile; and Cys 153 → Ser. [00191] In some embodiments, the residual region, i.e., the region differing from positions corresponding to positions 5, 26-31, 33-34, 42, 46, 52, 56, 58, 60-61, 65, 71, 85, 94, 101, 104- 106, 108, 111, 114, 121, 133, 148, 150, and 153 of the linear polypeptide sequence of mature hTlc (SEQ ID NO: 1), of an hTlc mutein of the disclosure may comprise the wild-type (natural) amino acid sequence of the linear polypeptide sequence of mature hTlc outside the mutated amino acid sequence positions. [00192] In some embodiments, an hTlc mutein of the disclosure has at least 70% sequence identity or at least 70% sequence homology to the sequence of mature hTlc (SEQ ID NO: 1). As an illustrative example, the mutein of SEQ ID NO: 32 has an amino acid sequence identity or a sequence homology of approximately 84% with the amino acid sequence of the mature hTlc. [00193] In some embodiments, an hTlc mutein of the disclosure comprises an amino acid sequence as set forth in any one of SEQ ID NOs: 32-38 or a fragment or variant thereof. [00194] In some embodiments, an hTlc mutein of the disclosure has at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or higher sequence identity to an amino acid sequence selected from SEQ ID NOs: 32-38. [00195] The present disclosure also includes structural homologues of an hTlc mutein having an amino acid sequence selected from SEQ ID NOs: 32-38, which structural homologues have an amino acid sequence homology or sequence identity of more than about 60%, preferably more than 65%, more than 70%, more than 75%, more than 80%, more than 85%, more than 90%, more than 92% and most preferably more than 95% in relation to said hTlc mutein. [00196] In some aspects, the present disclosure provides CD137-binding hNGAL muteins. In this regard, the disclosure provides one or more hNGAL muteins that are capable of binding CD137 with an affinity measured by a K _D of about 800 nM, 700 nM, 200 nM, 140 nM, 100 nM or lower, preferably about 70 nM, 50 nM, 30 nM, 10 nM, 5 nM, 2 nM, or even lower. In some embodiments, provided hNGAL muteins are capable of binding CD137 with an EC ₅₀ value of about 1000 nM, 500 nM, 100 nM, 80 nM, 50 nM, 25 nM, 18 nM, 15 nM, 10 nM, 5 nM, or lower. [00197] In some embodiments, provided CD137-binding hNGAL muteins may be cross- reactive with cynomolgus CD137. In some embodiments, provided hNGAL muteins are capable of binding cynomolgus CD137 with an affinity measured by a K _D of about 50 nM, 20 nM, 10 nM, Attorney Docket No.01218-0029-00PCT 5 nM, 2 nM, or even lower. In some embodiments, provided hNGAL muteins are capable of binding cynomolgus CD137 with an EC ₅₀ value of about 100 nM, 80 nM, 50 nM, 30 nM, or even lower. [00198] In some embodiments, an hNGAL mutein of the disclosure may interfere or compete with the binding of CD137L to CD137. In some other embodiments, an hNGAL mutein of the disclosure may be capable of binding CD137 in the presence of CD137L and/or binding a CD137/CD137L complex. [00199] In some embodiments, provided hNGAL muteins may comprise a mutated amino acid residue at one or more positions corresponding to positions 28, 36, 40-41, 49, 52, 65, 68, 70, 72-73, 77, 79, 81, 83, 87, 94, 96, 100, 103, 106, 125, 127, 132 and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2). [00200] In some embodiments, provided hNGAL muteins may comprise a mutated amino acid residue at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, or even more positions corresponding to positions 28, 36, 40-41, 49, 52, 65, 68, 70, 72-73, 77, 79, 81, 83, 87, 94, 96, 100, 103, 106, 125, 127, 132, and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2). In some preferred embodiments, the provided hNGAL muteins are capable of binding CD137, in particular human CD137. [00201] In some embodiments, provided hNGAL muteins may comprise a mutated amino acid residue at one or more positions corresponding to positions 28, 36, 40-41, 49, 52, 65, 68, 70, 72-73, 77, 79, 81, 87, 96, 100, 103, 106, 125, 127, 132 and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2) In some preferred embodiments, the provided hNGAL muteins are capable of binding CD137, in particular human CD137. [00202] In some embodiments, provided hNGAL muteins may comprise a mutated amino acid residue at one or more positions corresponding to positions 36, 87, and 96 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2) and at one or more positions corresponding to positions 28, 40-41, 49, 52, 65, 68, 70, 72-73, 77, 79, 81, 83, 94, 100, 103, 106, 125, 127, 132, and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2). [00203] In other some embodiments, provided hNGAL muteins may comprise a mutated amino acid residue at one or more positions corresponding to positions 20, 25, 28, 33, 36, 40-41, 44, 49, 52, 59, 68, 70-73, 77-82, 87, 92, 96, 98, 100, 101, 103, 122, 125, 127, 132, and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2). [00204] In other embodiments, provided hNGAL muteins may comprise a mutated amino acid residue at one or more positions corresponding to positions 36, 40, 41, 49, 52, 68, 70, 72, 73, 77, 79, 81, 96, 100, 103, 125, 127, 132, and 134 of the linear polypeptide sequence of mature Attorney Docket No.01218-0029-00PCT hNGAL (SEQ ID NO: 2) and at one or more positions corresponding to positions 20, 25, 33, 44, 59, 71, 78, 80, 82, 87, 92, 98, 101, and 122 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2). [00205] In some embodiments, a lipocalin mutein according to the disclosure may comprise at least one amino acid substitution of a native cysteine residue by, e.g., a serine residue. In some embodiments, an hNGAL mutein according to the disclosure may comprise an amino acid substitution of a native cysteine residue at positions corresponding to positions 76 and/or 175 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2) by another amino acid, such as a serine residue. In this context, it is noted that it has been found that removal of the structural disulfide bond (on the level of a respective naïve nucleic acid library) of wild-type hNGAL that is formed by the cysteine residues 76 and 175 (cf. Breustedt et al., J Biol Chem, 2005) may provide hNGAL muteins that are not only stably folded but are also able to bind a given non-natural target with high affinity. In some embodiments, the elimination of the structural disulfide bond may provide the further advantage of allowing for the generation or deliberate introduction of non-natural disulfide bonds into muteins of the disclosure, thereby, increasing the stability of the muteins. However, hNGAL muteins that bind CD137 and that have the disulfide bridge formed between Cys 76 and Cys 175 are also part of the present disclosure. [00206] In some embodiments, provided CD137-binding hNGAL muteins may comprise, at one or more positions corresponding to positions 28, 36, 40-41, 49, 52, 65, 68, 70, 72-73, 77, 79, 81, 83, 87, 94, 96, 100, 103, 106, 125, 127, 132 and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2), one or more of the following mutated amino acid residues: Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Arg or Lys; Gln 49 → Val, Ile, His, Ser or Asn; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Met, Ala or Gly; Leu 70 → Ala, Lys, Ser or Thr; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Met, Arg, Thr or Asn; Trp 79 → Ala or Asp; Arg 81 → Met, Trp or Ser; Phe 83 → Leu; Cys 87 → Ser; Leu 94 → Phe; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu and Lys 134 → Tyr. In some embodiments, an hNGAL mutein of the disclosure comprises two or more, such as 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, even more such as 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or all mutated amino acid residues at these sequence positions of mature hNGAL (SEQ ID NO: 2). [00207] In some embodiments, provided CD137-binding hNGAL muteins may comprise, at one or more positions corresponding to positions 20, 25, 28, 33, 36, 40-41, 44, 49, 52, 59, 68, 70-73, 77-82, 87, 92, 96, 98, 100, 101, 103, 122, 125, 127, 132, and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2), one or more of the following mutated amino acid residues: Gln 20 ^ Arg; Asn 25 ^ Tyr or Asp; Gln 28 → His; Val 33 ^ Ile; Leu 36 ^Met; Ala 40 ^ Asn; Ile 41 ^ Leu; Glu 44 ^ Val or Asp; Gln 49 ^ His; Tyr 52 ^Ser or Gly; Lys 59 ^ Asn; Ser 68 ^ Asp; Leu 70 ^ Met; Phe 71 ^ Leu; Arg 72 ^ Leu; Lys 73 ^ Asp; Asp 77 ^ Gln or Attorney Docket No.01218-0029-00PCT His; Tyr 78 ^ His; Trp 79 ^ Ile; Ile 80 ^ Asn; Arg 81 ^ Trp or Gln; Thr 82 ^ Pro; Cys 87 → Ser; Phe 92 ^ Leu or Ser; Asn 96 ^ Phe; Lys 98 ^ Arg; Tyr 100 ^ Asp; Pro 101 ^ Leu; Leu 103 ^ His or Pro; Phe 122 ^ Tyr; Lys 125 ^ Ser; Ser 127 ^ Ile; Tyr 132 ^ Trp; and Lys 134 ^ Gly. [00208] In some embodiments, provided CD137-binding hNGAL muteins may comprise, at one or more positions corresponding to positions 36, 40, 41, 49, 52, 68, 70, 72, 73, 77, 79, 81, 96, 100, 103, 125, 127, 132, and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2), one or more of the following mutated amino acid residues: Leu 36 ^Met; Ala 40 ^ Asn; Ile 41 ^ Leu; Gln 49 ^ His; Tyr 52 ^Ser or Gly; Ser 68 ^ Asp; Leu 70 ^ Met; Arg 72 ^ Leu; Lys 73 ^ Asp; Asp 77 Gln or His; Trp 79 Ile; Arg 81 ^ Trp or Gln; Asn 96 ^ Phe; Tyr 100 ^ Asp; Leu 103 ^ His or Pro; Lys 125 Ser; Ser 127 Ile; Tyr 132 ^ Trp; and Lys 134 ^ Gly. In some embodiments, provided CD137-binding hNGAL muteins may further comprise, at one or more positions corresponding to positions 20, 25, 33, 44, 59, 71, 78, 80, 82, 92, 98, 101, and 122 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2), one or more of the following mutated amino acid residues: Gln 20 ^ Arg; Asn 25 ^ Tyr or Asp; Val 33 ^ Ile; Glu 44 ^ Val or Asp; Lys 59 ^ Asn; Phe 71 ^ Leu; Tyr Thr Phe 92 ^ Leu or Ser; Lys 98 ^ Arg; Pro Phe 122 ^ Tyr. [00209] In some embodiments, provided CD137-binding hNGAL muteins may comprise one of the following sets of mutated amino acid residues in comparison with the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2): (a) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Lys; Gln 49 → Asn; Tyr 52 → Met; Ser 68 → Gly; Leu 70 → Thr; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Thr; Trp 79 → Ala; Arg 81 → Ser; Cys 87 → Ser; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; (b) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Arg; Gln 49 → Ile; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Met; Leu 70 → Lys; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Met; Trp 79 → Asp; Arg 81 → Trp; Cys 87 → Ser; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; (c) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Arg; Gln 49 → Asn; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Ala; Leu 70 → Ala; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Thr; Trp 79 → Asp; Arg 81 → Trp; Cys 87 → Ser; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; Attorney Docket No.01218-0029-00PCT (d) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Lys; Gln 49 → Asn; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Ala; Leu 70 → Ala; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Thr; Trp 79 → Asp; Arg 81 → Trp; Cys 87 → Ser; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; (e) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Lys; Gln 49 → Ser; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Gly; Leu 70 → Ser; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Thr; Trp 79 → Ala; Arg 81 → Met; Cys 87 → Ser; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; (f) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Lys; Gln 49 → Val; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Gly; Leu 70 → Thr; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Arg; Trp 79 → Asp; Arg 81 → Ser; Cys 87 → Ser; Leu 94 → Phe; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; (g) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Arg; Gln 49 → His; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Gly; Leu 70 → Thr; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Thr; Trp 79 → Ala; Arg 81 → Ser; Cys 87 → Ser; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; (h) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Lys; Gln 49 → Asn; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Gly; Leu 70 → Thr; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Thr; Trp 79 → Ala; Arg 81 → Ser; Phe 83 → Leu; Cys 87 → Ser; Leu 94 → Phe; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr; or (i) Gln 28 → His; Leu 36 → Gln; Ala 40 → Ile; Ile 41 → Arg; Gln 49 → Ser; Tyr 52 → Met; Asn 65 → Asp; Ser 68 → Ala; Leu 70 → Thr; Arg 72 → Asp; Lys 73 → Asp; Asp 77 → Asn; Trp 79 → Ala; Arg 81 → Ser; Cys 87 → Ser; Asn 96 → Lys; Tyr 100 → Phe; Leu 103 → His; Tyr 106 → Ser; Lys 125 → Phe; Ser 127 → Phe; Tyr 132 → Glu; and Lys 134 → Tyr. [00210] In some further embodiments, in the residual region, i.e., the region differing from positions 28, 36, 40-41, 49, 52, 65, 68, 70, 72-73, 77, 79, 81, 83, 87, 94, 96, 100, 103, 106, 125, 127, 132 and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2), an hNGAL mutein of the disclosure may include the wild-type (natural) amino acid sequence of mature hNGAL outside the mutated amino acid sequence positions. [00211] In some other embodiments, provided CD137-binding hNGAL muteins may comprise one of the following sets of mutated amino acid residues in comparison with the linear Attorney Docket No.01218-0029-00PCT polypeptide sequence of mature hNGAL (SEQ ID NO: 2): (a) Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Ser; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Arg 81 → Trp; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; (b) Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Ser; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Arg 81 → Trp; Phe 92 → Leu; Asn 96 → Phe; Lys 98 → Arg; Tyr 100 → Asp; Pro 101 → Leu; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; (c) Asn 25 → Tyr; Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Gly; Ser 68 → Asp; Leu 70 → Met; Phe 71 → Leu; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Arg 81 → Gln; Phe 92 → Ser; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; (d) Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Gly; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Tyr 78 → His; Trp 79 → Ile; Arg 81 → Trp; Phe 92 → Leu; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; (e) Asn 25 → Asp; Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Gly; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Arg 81 → Trp; Phe 92 → Leu; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; (f) Val 33 → Ile; Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Gly; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Arg 81 → Trp; Phe 92 → Leu; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; (g) Gln 20 → Arg; Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Glu 44 → Val; Gln 49 → His; Tyr 52 → Gly; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Arg 81 → Trp; Phe 92 → Leu; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Phe 122 → Tyr; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; (h) Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Ser; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Ile 80 → Asn; Arg 81 → Trp; Thr 82 → Pro; Asn 96 → Phe; Tyr 100 → Asp; Pro 101 → Leu; Leu 103 → Pro; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; Attorney Docket No.01218-0029-00PCT (i) Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Gln 49 → His; Tyr 52 → Gly; Lys 59 → Asn; Ser 68 → Asp; Leu 70 → Met; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → Gln; Trp 79 → Ile; Arg 81 → Trp; Phe 92 → Leu; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly; and (j) Leu 36 → Met; Ala 40 → Asn; Ile 41 → Leu; Glu 44 → Asp; Gln 49 → His; Tyr 52 → Ser; Ser 68 → Asp; Leu 70 → Met; Phe 71 → Leu; Arg 72 → Leu; Lys 73 → Asp; Asp 77 → His; Trp 79 → Ile; Arg 81 → Trp; Phe 92 → Leu; Asn 96 → Phe; Tyr 100 → Asp; Leu 103 → His; Lys 125 → Ser; Ser 127 → Ile; Tyr 132 → Trp; and Lys 134 → Gly. [00212] In some embodiments, in the residual region, i.e., the region differing from positions 20, 25, 28, 33, 36, 40-41, 44, 49, 52, 59, 68, 70-73, 77-82, 87, 92, 96, 98, 100, 101, 103, 122, 125, 127, 132, and 134 of the linear polypeptide sequence of mature hNGAL (SEQ ID NO: 2), of an hNGAL mutein of the disclosure may include the wild-type (natural) amino acid sequence of mature hNGAL outside the mutated amino acid sequence positions. [00213] In some embodiments, an hNGAL mutein of the disclosure has at least 70% sequence identity or at least 70% sequence homology to the sequence of mature hNGAL (SEQ ID NO: 2). As an illustrative example, the mutein of the SEQ ID NO: 40 has an amino acid sequence identity or a sequence homology of approximately 87% with the amino acid sequence of the mature hNGAL. [00214] In some embodiments, an hNGAL mutein of the disclosure comprises an amino acid sequence as set forth in any one of SEQ ID NOs: 39-57 or a fragment or variant thereof. [00215] In some embodiments, an hNGAL mutein of the disclosure has at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or higher sequence identity to an amino acid sequence selected from SEQ ID NOs: 39-57. [00216] The present disclosure also includes structural homologues of an hNGAL mutein having an amino acid sequence selected from SEQ ID NOs: 39-57, which structural homologues have an amino acid sequence homology or sequence identity of more than about 60%, preferably more than 65%, more than 70%, more than 75%, more than 80%, more than 85%, more than 90%, more than 92% and most preferably more than 95% in relation to said hNGAL mutein. [00217] In some embodiments, the present disclosure provides a lipocalin mutein that binds CD137 with an affinity measured by a K _D of about 5 nM or lower, wherein the lipocalin mutein has at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or higher sequence identity to the amino acid sequence of SEQ ID NO: 40. [00218] Suitable lipocalin muteins specific for CD137 are also described in WO Attorney Docket No.01218-0029-00PCT 2016/177762, which is incorporated herein by reference in its entirety. [00219] In some embodiments, a lipocalin mutein of the present disclosure can comprise a heterologous amino acid sequence at its N-or C-terminus, preferably C-terminus, such as a Strep II tag (SEQ ID NO: 12) or a cleavage site sequence for certain restriction enzymes, without affecting the biological activity (binding to its target, e.g., CD137) of the lipocalin mutein. [00220] In some embodiments, further modifications of a lipocalin mutein may be introduced in order to modulate certain characteristics of the mutein, such as to improve folding stability, serum stability, protein resistance or water solubility or to reduce aggregation tendency, or to introduce new characteristics to the mutein. In some embodiments, modification(s) may result in two or more (e.g., 2, 3, 4, 5, 6, 7, 8, 9, or 10) characteristics of a provided mutein being modulated. [00221] For example, it is possible to mutate one or more amino acid sequence positions of a lipocalin mutein to introduce new reactive groups, for example, for the conjugation to other compounds, such as polyethylene glycol (PEG), hydroxyethyl starch (HES), biotin, peptides or proteins, or for the formation of non-naturally occurring disulphide linkages. The conjugated compound, for example, PEG and HES, can in some cases increase the serum half-life of the corresponding lipocalin mutein. [00222] In some embodiments, a reactive group of a lipocalin mutein may occur naturally in its amino acid sequence, such as naturally occurring cysteine residues in said amino acid sequence. In some other embodiments, such reactive group may be introduced via mutagenesis. In case a reactive group is introduced via mutagenesis, one possibility is the mutation of an amino acid at the appropriate position by a cysteine residue. Exemplary possibilities of such a mutation to introduce a cysteine residue into the amino acid sequence of an hTlc mutein include the substitutions Thr 40→ Cys, Glu 73→ Cys, Arg 90→ Cys, Asp 95→ Cys, and Glu 131→ Cys of the wild-type sequence of hTlc (SEQ ID NO: 1). Exemplary possibilities of such a mutation to introduce a cysteine residue into the amino acid sequence of an hNGAL mutein include the introduction of a cysteine residue at one or more of the sequence positions that correspond to sequence positions 14, 21, 60, 84, 88, 116, 141, 145, 143, 146 or 158 of the wild-type sequence of hNGAL (SEQ ID NO: 2).The generated thiol moiety may be used to PEGylate or HESylate the mutein, for example, in order to increase the serum half-life of a respective lipocalin mutein. [00223] In some embodiments, in order to provide suitable amino acid side chains as new reactive groups for conjugating one of the above compounds to a lipocalin mutein, artificial amino acids may be introduced to the amino acid sequence of a lipocalin mutein. Generally, such artificial amino acids are designed to be more reactive and thus to facilitate the conjugation to the desired compound. Such artificial amino acids may be introduced by mutagenesis, for example, Attorney Docket No.01218-0029-00PCT using an artificial tRNA, such as para-acetyl-phenylalanine. [00224] In some embodiments, a lipocalin mutein of the disclosure is fused at its N- terminus or its C-terminus to a protein, a protein domain or a peptide, for instance, an antibody, a signal sequence and/or an affinity tag. In some other embodiments, a lipocalin mutein of the disclosure is conjugated at its N-terminus or its C-terminus to a partner, which is a protein, a protein domain or a peptide; for instance, an antibody, a signal sequence and/or an affinity tag. [00225] Affinity tags such as the Strep-tag or Strep-tag II (Schmidt et al., J Mol Biol, 1996), the c-myc-tag, the FLAG-tag, the His-tag or the HA-tag or proteins such as glutathione-S- transferase, which allow easy detection and/or purification of recombinant proteins, are examples of suitable fusion partners. Proteins with chromogenic or fluorescent properties such as the green fluorescent protein (GFP) or the yellow fluorescent protein (YFP) are suitable fusion partners for lipocalin muteins of the disclosure as well. In general, it is possible to label the lipocalin muteins of the disclosure with any appropriate chemical substance or enzyme, which directly or indirectly generates a detectable compound or signal in a chemical, physical, optical, or enzymatic reaction. For example, a fluorescent or radioactive label can be conjugated to a lipocalin mutein to generate fluorescence or x-rays as detectable signal. Alkaline phosphatase, horseradish peroxidase and β-galactosidase are examples of enzyme labels (and at the same time optical labels) which catalyze the formation of chromogenic reaction products. In general, all labels commonly used for antibodies (except those exclusively used with the sugar moiety in the Fc part of immunoglobulins) can also be used for conjugation to the lipocalin muteins of the disclosure. [00226] In some embodiments, a lipocalin mutein of the disclosure may be fused or conjugated to a moiety that extends the serum half-life of the mutein (in this regard see also International Patent Publication No. WO 2006/056464, where such strategies are described with reference to muteins of human neutrophil gelatinase-associated lipocalin (hNGAL) with binding affinity for CTLA-4). The moiety that extends the serum half-life may be a PEG molecule, a HES molecule, a fatty acid molecule, such as palmitic acid (Vajo and Duckworth, Pharmacol Rev, 2000), an Fc part of an immunoglobulin, a C _H 3 domain of an immunoglobulin, a C _H 4 domain of an immunoglobulin, an albumin binding peptide, an albumin binding protein, or a transferrin, to name only a few. [00227] In some embodiments, if PEG is used as a conjugation partner, the PEG molecule can be substituted, unsubstituted, linear, or branched. It can also be an activated polyethylene derivative. Examples of suitable compounds are PEG molecules as described in International Patent Publication No. WO 1999/64016, in U.S. Patent No. 6,177,074, or in U.S. Patent No. 6,403,564 in relation to interferon, or as described for other proteins such as PEG- modified asparaginase, PEG-adenosine deaminase (PEG-ADA) or PEG-superoxide dismutase Attorney Docket No.01218-0029-00PCT (Fuertges and Abuchowski, Journal of Controlled Release, 1990). The molecular weight of such a polymer, such as polyethylene glycol, may range from about 300 to about 70,000 daltons, including, for example, polyethylene glycol with a molecular weight of about 10,000, of about 20,000, of about 30,000 or of about 40,000 daltons. Moreover, as, e.g., described in U.S. Patent No. 6,500,930 or 6,620,413, carbohydrate oligomers and polymers such as HES can be conjugated to a mutein of the disclosure for the purpose of serum half-life extension. [00228] In some embodiments, if an Fc part of an immunoglobulin is used for the purpose of prolonging the serum half-life of the lipocalin muteins of the disclosure, the SynFusion™ technology, commercially available from Syntonix Pharmaceuticals, Inc. (MA, USA), may be used. The use of this Fc-fusion technology allows the creation of longer-acting biopharmaceuticals and may, for example, consist of two copies of the mutein linked to the Fc region of an antibody to improve pharmacokinetics, solubility, and production efficiency. [00229] Examples of albumin binding peptides that can be used to extend the serum half- life of a lipocalin mutein are, for instance, those having a Cys-Xaa ₁ -Xaa ₂ -Xaa ₃ -Xaa ₄ -Cys consensus sequence, wherein Xaa ₁ is Asp, Asn, Ser, Thr, or Trp; Xaa ₂ is Asn, Gln, His, Ile, Leu, or Lys; Xaa ₃ is Ala, Asp, Phe, Trp, or Tyr; and Xaa ₄ is Asp, Gly, Leu, Phe, Ser, or Thr as described in U.S. Patent Publication No.2003/0069395 or Dennis et al. (2002). The albumin binding protein fused or conjugated to a lipocalin mutein to extend serum half-life may be a bacterial albumin binding protein, an antibody, an antibody fragment including domain antibodies (see U.S. Patent No.6,696,245, for example), or a lipocalin mutein with binding activity for albumin. Examples of bacterial albumin binding proteins include streptococcal protein G (Konig and Skerra, J Immunol Methods, 1998). [00230] In some embodiments, if the albumin-binding protein is an antibody fragment it may be a domain antibody. Domain Antibodies (dAbs) are engineered to allow precise control over biophysical properties and in vivo half-life to create the optimal safety and efficacy product profile. Domain Antibodies are for example commercially available from Domantis Ltd. (Cambridge, UK, and MA, USA). [00231] In some embodiments, albumin itself (Osborn et al., J Pharmacol Exp Ther, 2002), or a biologically active fragment of albumin can be used as a partner of a lipocalin mutein of the disclosure to extend serum half-life. The term “albumin” includes all mammal albumins such as human serum albumin or bovine serum albumin or rat albumin. The albumin or fragment thereof can be recombinantly produced as described in U.S. Patent No.5,728,553 or European Patent Publication Nos. EP0330451 and EP0361991. Accordingly, recombinant human albumin (e.g., Recombumin® from Novozymes Delta Ltd., Nottingham, UK) can be conjugated or fused to a lipocalin mutein of the disclosure. Attorney Docket No.01218-0029-00PCT [00232] In some embodiments, if a transferrin is used as a partner to extend the serum half-life of the lipocalin muteins of the disclosure, the muteins can be genetically fused to the N- or C-terminus, or both, of non-glycosylated transferrin. Non-glycosylated transferrin has a half-life of 14-17 days, and a transferrin fusion protein will similarly have an extended half-life. The transferrin carrier also provides high bioavailability, biodistribution and circulating stability. This technology is commercially available from BioRexis (BioRexis Pharmaceutical Corporation, PA, USA). Recombinant human transferrin (DeltaFerrin™) for use as a protein stabilizer/half-life extension partner is also commercially available from Novozymes Delta Ltd. (Nottingham, UK). [00233] Yet another alternative to prolong the half-life of the lipocalin muteins of the disclosure is to fuse to the N- or C-terminus of a mutein a long, unstructured, flexible glycine-rich sequence (for example poly-glycine with about 20 to 80 consecutive glycine residues). This approach disclosed in International Patent Publication No. WO 2007/038619, for example, has also been term “rPEG” (recombinant PEG). E. Exemplary uses and applications of antibodies or antigen-binding domains thereof specific for CD228 and fusion proteins specific for CD137 and CD228. [00234] In some embodiments, fusion proteins of the disclosure may produce synergistic effects through dual-targeting of CD137 and CD228. In some embodiments, fusion proteins of the disclosure may produce a localized anti-tumor effect through dual-targeting of CD137 and CD228. Numerous possible applications for the fusion proteins of the disclosure, therefore, exist in medicine. [00235] In some embodiments, the present disclosure encompasses the use of one or more fusion proteins disclosed herein or of one or more compositions comprising such fusion proteins for simultaneously binding of CD137 and CD228. [00236] The present disclosure also involves the use of one or more fusion proteins as described for complex formation with CD137 and/or CD228. [00237] Therefore, in one aspect of the disclosure, provided fusion proteins may be used for the detection of CD137 and/or CD228. Such use may include the steps of contacting one or more of said fusion proteins, under suitable conditions, with a sample suspected of containing CD137 and/or CD228, thereby allowing formation of a complex between the fusion proteins and CD137 and/or CD228, and detecting the complex by a suitable signal. The detectable signal can be caused by a label, as explained above, or by a change of physical properties due to the binding, i.e., the complex formation, itself. One example is surface plasmon resonance, the value of which is changed during binding of binding partners one of which is immobilized on a surface such as a gold foil. Attorney Docket No.01218-0029-00PCT [00238] Fusion proteins of the disclosure may also be used for the separation of CD137 and/or CD228. Such use may include the steps of contacting one or more of said fusion proteins, under suitable conditions, with a sample supposed to contain CD137 and/or CD228, thereby allowing the formation of a complex between the fusion proteins and CD137 and/or CD228 and separating the complex from the sample. [00239] In some aspects, the present disclosure provides diagnostic and/or analytical kits comprising one or more antibodies, antigen-binding domains thereof, or fusion proteins according to the disclosure. [00240] In addition to their use in diagnostics, in yet another aspect, the disclosure contemplates pharmaceutical compositions comprising one or more antibodies, antigen-binding domains thereof, or fusion proteins of the disclosure and a pharmaceutically acceptable excipient. [00241] Furthermore, in some embodiments, provided antibodies, antigen-binding domains thereof, or fusion proteins may be used in therapy, e.g., as anti-tumor and/or anti- infection agents and/or immune modulators. In some embodiments, provided antibodies, antigen- binding domains thereof, or fusion proteins may be used in the manufacture of a medicament, e.g., a medicament for the treatment of cancer, including CD228-positive cancer. In some embodiments, antibodies, antigen-binding domains thereof, or fusion proteins of the present disclosure may be used in a method of prevention, amelioration, or treatment of human diseases, such as a cancer, including CD228-positive cancer. In some embodiments, fusion proteins of the present disclosure may be used together with an anti-PD-1 or anti-PD-L1 antibody in a method of prevention, amelioration, or treatment of a cancer, including CD228-positive cancer. In some embodiments, the anti-PD-1 or anti-PD-L1 antibody is nivolumab, pembrolizumab, cemiplimab, dostarlimab, atezolizumab, avelumab, or durvalumab. Accordingly, also provided are methods of preventing, ameliorating, or treating human diseases, such as cancer, including CD228-positive cancer, in a subject in need thereof, comprising administering to said subject a therapeutically effective amount of one or more antibodies, antigen-binding domains thereof, or fusion proteins of the disclosure or one or more compositions comprising such antibodies, antigen-binding domains thereof, or fusion proteins. Accordingly, also provided are methods of preventing, ameliorating, or treating cancer, including CD228-positive cancer, in a subject in need thereof, comprising administering to said subject a therapeutically effective amount of one or more fusion proteins of the disclosure or one or more compositions comprising such fusion protein, and further comprising administering a therapeutically effective amount of an anti-PD-1 or anti-PD-L1 antibody or one or more compositions comprising such antibody to said subject. In some embodiments, the anti-PD- 1 or anti-PD-L1 antibody is nivolumab, pembrolizumab, cemiplimab, dostarlimab, atezolizumab, avelumab, or durvalumab. In some embodiments, the cancer is CD228-positive cancer. Attorney Docket No.01218-0029-00PCT [00242] Examples of cancers that may be treated using the antibodies, antigen-binding domains thereof, or fusion proteins of the disclosure include lung cancer, e.g., non-small cell lung cancer (NSCLC), melanoma, e.g., cutaneous melanoma or intraocular malignant melanoma, pancreatic neoplasms or pancreatic cancer, mesothelioma, colorectal neoplasms or colorectal cancer (CRC), thyroid cancer, breast cancer, cholangiocarcinoma, esophageal cancer, and head and neck cancer. In some embodiments, cancer includes metastatic cancers. [00243] In some embodiments, fusion proteins of the disclosure may simultaneously target tumor cells where CD228 is expressed and activate lymphocytes of the host immune system adjacent to such tumor cells. In some embodiments, fusion proteins of the disclosure may increase targeted anti-tumor T cell activity, enhance anti-tumor immunity and, and/or have a direct inhibiting effect on tumor growth, thereby producing synergistic anti-tumor results. In some embodiments, fusion proteins of the disclosure may activate immune responses in a tumor microenvironment. In some embodiments, fusion proteins of the disclosure may reduce side effects of effector lymphocytes towards healthy cells, i.e., off-target toxicity, for example, via locally inhibiting oncogene activity and/or inducing lymphocyte activation. [00244] In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising a provided fusion protein, for inducing a localized lymphocyte response in the vicinity of CD228-positive tumor cells. Accordingly, in some embodiments, the present disclosure provides methods of inducing a localized lymphocyte response in the vicinity of CD228-positive tumor cells, comprising applying one or more fusion proteins of the disclosure or of one or more compositions comprising such fusion proteins. “Localized” means that, upon simultaneous binding T cells via CD137 and engaging CD228- positive tumor cells, T cells produce cytokines, particularly IL-2 and/or IFN gamma in the vicinity of the CD228-positive cells. Such cytokines reflect activation of T cells which may then be able to kill CD228-positive cells, either directly or indirectly by attracting other killer cells, such as cytotoxic T cells and/or NK cells. Provided fusion proteins may also be used to increase secretion of cytotoxic factors, such as perforin, granzyme B and granzyme A, by T cells in the vicinity of CD228- positive tumor cells. [00245] In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising such fusion protein, for co-stimulating T cells, and/or activating downstream signaling pathways of CD137. Preferably, a provided fusion protein co-stimulates T cells and/or activates downstream signaling pathways of CD137 when engaging CD228-positive tumor cells. Accordingly, the present disclosure provides methods of inducing T lymphocyte proliferation/activity and/or activating downstream signaling pathways of CD137, preferably when engaging tumor cells where CD228 is expressed, comprising applying Attorney Docket No.01218-0029-00PCT one or more fusion proteins of the disclosure and/or one or more compositions comprising such fusion protein(s). [00246] In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising such fusion protein, for inducing CD137 clustering and activation on T cells and directing such T cells to tumor cells where CD228 is expressed. [00247] In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising such fusion protein, for inducing increased CD8+ T cell mitochondrial content and/or reduced CD8+ T cell mitochondria depolarization in the vicinity of CD228-positive tumor cells, comprising applying one or more fusion proteins of the disclosure and/or one or more composition comprising such fusion protein to a tissue comprising a tumor. In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising such fusion protein, for inducing increased CD8+ T cell mitochondrial content, comprising applying one or more fusion proteins of the disclosure and/or one or more compositions comprising such fusion protein to a tissue comprising a tumor. In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising such fusion protein, for inducing reduced CD8+ T cell mitochondria depolarization in the vicinity of CD228-positive tumor cells, comprising applying one or more fusion proteins of the disclosure and/or one or more compositions comprising such fusion protein to a tissue comprising a tumor. [00248] In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising such fusion protein, for stimulating exhausted CD8+ T cell proliferation in the vicinity of CD228-positive tumor cells, comprising applying one or more fusion proteins of the disclosure or one or more compositions comprising such fusion protein to a tissue comprising a tumor. In some embodiments, the present disclosure encompasses the use of a fusion protein of the disclosure, or a composition comprising such fusion protein, for stimulating exhausted CD8+ T cell proliferation in the vicinity of CD228-positive tumor cells, comprising applying one or more fusion proteins of the disclosure or one or more compositions comprising such fusion protein, and applying an anti-PD-1 or anti-PD-L1 antibody or one or more compositions comprising such antibody, to a tissue comprising a tumor. In some embodiments, the anti-PD-1 or anti-PD-L1 antibody is nivolumab, pembrolizumab, cemiplimab, dostarlimab, atezolizumab, avelumab, or durvalumab. F. Production of exemplary provided antibodies or antigen-binding domains thereof specific for CD228 and fusion proteins specific for CD137 and CD228. [00249] In some embodiments, the present disclosure provides nucleic acid molecules (e.g., Attorney Docket No.01218-0029-00PCT DNA or RNA) that include nucleotide sequences encoding provided antibodies, antigen-binding domains thereof, or fusion proteins. In some embodiments, the disclosure encompasses a vector containing a provided nucleic acid molecule. In some embodiments, the disclosure encompasses a host cell containing a provided nucleic acid molecule or vector. Since the degeneracy of the genetic code permits substitutions of certain codons by other codons specifying the same amino acid, the disclosure is not limited to a specific nucleic acid molecule encoding an antibody, antigen- binding domain thereof, or fusion protein as described herein, rather, encompassing all nucleic acid molecules that include nucleotide sequences encoding a functional antibody, antigen-binding domain thereof, or fusion protein. In this regard, the present disclosure also relates to nucleotide sequences encoding provided antibodies, antigen-binding domains thereof, or fusion proteins. [00250] A nucleic acid molecule, such as DNA, is referred to as “capable of expressing a nucleic acid molecule” or “able to allow expression of a nucleotide sequence” if it includes sequence elements that contain information regarding to transcriptional and/or translational regulation, and such sequences are “operably linked” to the nucleotide sequence encoding the protein. An operable linkage is a linkage in which the regulatory sequence elements and the sequence to be expressed are connected in a way that enables gene expression. The precise nature of the regulatory regions necessary for gene expression may vary among species, but in general these regions include a promoter, which, in prokaryotes, contains both the promoter per se, i.e., DNA elements directing the initiation of transcription, as well as DNA elements which, when transcribed into RNA, will signal the initiation of translation. Such promoter regions normally include 5’ non-coding sequences involved in initiation of transcription and translation, such as the -35/-10 boxes and the Shine-Dalgarno element in prokaryotes or the TATA box, CAAT sequences, and 5’-capping elements in eukaryotes. These regions can also include enhancer or repressor elements as well as translated signal and leader sequences for targeting the native protein to a specific compartment of a host cell. [00251] In addition, 3’ non-coding sequences may contain regulatory elements involved in transcriptional termination, polyadenylation or the like. If, however, these termination sequences are not satisfactorily functional in a particular host cell, then they may be substituted with signals functional in that cell. [00252] Therefore, a nucleic acid molecule of the disclosure may be “operably linked” to one or more regulatory sequences, such as a promoter sequence, to allow expression of this nucleic acid molecule. In some embodiments, a nucleic acid molecule of the disclosure includes a promoter sequence and a transcriptional termination sequence. Suitable prokaryotic promoters are, for example, the tet promoter, the lacUV5 promoter or the T7 promoter. Examples of promoters useful for expression in eukaryotic cells are the SV40 promoter or the CMV promoter. Attorney Docket No.01218-0029-00PCT [00253] In some embodiments, a nucleic acid molecule encoding a lipocalin mutein disclosed in this application may be “operably linked” to another nucleic acid molecule encoding an antibody of the disclosure to allow expression of a fusion protein disclosed herein. [00254] In some embodiments, provided methods may include subjecting at least one nucleic acid molecule encoding mature hTlc to mutagenesis at nucleotide triplets coding for one or more positions corresponding to positions 5, 26-31, 33-34, 42, 46, 52, 56, 58, 60-61, 65, 71, 85, 94, 101, 104-106, 108, 111, 114, 121, 133, 148, 150 and 153 of the linear polypeptide sequence of hTlc (SEQ ID NO: 1), to obtain lipocalin muteins as included in provided fusion proteins. In some embodiments, provided methods may include subjecting at least one nucleic acid molecule encoding mature hNGAL to mutagenesis at nucleotide triplets coding for one or more positions corresponding to positions 28, 36, 40-41, 49, 52, 65, 68, 70, 72-73, 77, 79, 81, 83, 87, 94, 96, 100, 103, 106, 125, 127, 132 and 134 of the linear polypeptide sequence of hNGAL (SEQ ID NO: 2), to obtain lipocalin muteins as included in provided fusion proteins. In some embodiments, a provided method may include subjecting at least one nucleic acid molecule encoding mature hNGAL to mutagenesis at nucleotide triplets coding for one or more positions corresponding to positions 20, 25, 28, 33, 36, 40-41, 44, 49, 52, 59, 68, 70-73, 77-82, 87, 92, 96, 98, 100, 101, 103, 122, 125, 127, 132, and 134 of the linear polypeptide sequence of hNGAL (SEQ ID NO: 2), to obtain lipocalin muteins as included in provided fusion proteins. [00255] In addition, with respect to hTlc muteins or hNGAL muteins of the disclosure as included in the fusion proteins, in some embodiments, the naturally occurring disulfide bond between Cys 61 and Cys 153 or Cys 76 and Cys 175, respectively, may be removed. Accordingly, such muteins can be produced in a cell compartment having a reducing redox milieu, for example, in the cytoplasm of Gram-negative bacteria. [00256] With further respect to provided hTlc muteins or hNGAL muteins of the disclosure as included in the fusion proteins, the disclosure also includes nucleic acid molecules encoding such muteins which, in some embodiments, may include one or more additional mutations outside the indicated sequence positions of experimental mutagenesis. Such mutations are often tolerated or can even prove to be advantageous, for example, if they contribute to an improved folding efficiency, serum stability, thermal stability or ligand binding affinity of the lipocalin muteins and/or the fusion proteins. [00257] In some embodiments, provided nucleic acid molecules can also be part of a vector or any other kind of cloning vehicle, such as a plasmid, a phagemid, a phage, a baculovirus, a cosmid or an artificial chromosome. [00258] In some embodiments, a provided nucleic acid molecule may be included in a phagemid. As used in this context, a phagemid vector denotes a vector encoding the intergenic Attorney Docket No.01218-0029-00PCT region of a temperate phage, such as M13 or f1, or a functional part thereof fused to the cDNA of interest. For example, in some embodiments, after superinfection of bacterial host cells with such a provided phagemid vector and an appropriate helper phage (e.g., M13K07, VCS-M13 or R408) intact phage particles are produced, thereby enabling physical coupling of the encoded heterologous cDNA to its corresponding polypeptide displayed on the phage surface (Lowman, Annu Rev Biophys Biomol Struct, 1997, Rodi and Makowski, Curr Opin Biotechnol, 1999). [00259] In accordance with various embodiments, cloning vehicles can include, aside from the regulatory sequences described above and a nucleic acid sequence encoding an antibody, antigen-binding domain thereof, or fusion protein as described herein, replication and control sequences derived from a species compatible with the host cell that is used for expression as well as selection markers conferring a selectable phenotype on transformed or transfected cells. Large numbers of suitable cloning vectors are known in the art and are commercially available. [00260] The disclosure also relates, in some embodiments, to methods for the production of antibodies, antigen-binding domains thereof, or fusion proteins of the disclosure starting from a nucleic acid coding for an antibody, antigen-binding domain thereof, or fusion protein or any subunit(s) therein using genetic engineering methods. In some embodiments, a provided method can be carried out in vivo, wherein a provided antibody, antigen-binding domain thereof, or fusion protein can, for example, be produced in a bacterial or eukaryotic host organism, and then isolated from this host organism or its culture. It is also possible to produce an antibody, antigen-binding domain thereof, or fusion protein of the disclosure in vitro, for example, using an in vitro translation system. [00261] When producing an antibody, antigen-binding domain thereof, or fusion protein in vivo, a nucleic acid encoding such antibody, antigen-binding domain thereof, or fusion protein may be introduced into a suitable bacterial or eukaryotic host organism using recombinant DNA technology well known in the art. In some embodiments, a DNA molecule encoding an antibody, antigen-binding domain thereof, or fusion protein as described herein (for example, SEQ ID NOs: 91-94 and the sequence pairs of SEQ ID NOs: 91 and 88, 93 and 90, 87 and 92, or 89 and 94), and in particular a cloning vector containing the coding sequence of such an antibody, antigen- binding domain thereof, or fusion protein can be transformed into a host cell capable of expressing the gene. Transformation can be performed using standard techniques. Thus, the disclosure is also directed to host cells containing a nucleic acid molecule as disclosed herein. [00262] In some embodiments, transformed host cells may be cultured under conditions suitable for expression of the nucleotide sequence encoding an antibody, antigen-binding domain thereof, or fusion protein of the disclosure. In some embodiments, host cells can be prokaryotic, such as Escherichia coli (E. coli) or Bacillus subtilis, or eukaryotic, such as Saccharomyces Attorney Docket No.01218-0029-00PCT cerevisiae, Pichia pastoris, SF9 or High5 insect cells, immortalized mammalian cell lines (e.g., HeLa cells or CHO cells) or primary mammalian cells. [00263] In some embodiments, where a lipocalin mutein of the disclosure, including as comprised in a fusion protein disclosed herein, includes intramolecular disulfide bonds, it may be preferred to direct the nascent protein to a cell compartment having an oxidizing redox milieu using an appropriate signal sequence. Such an oxidizing environment may be provided by the periplasm of Gram-negative bacteria such as E. coli, in the extracellular milieu of Gram-positive bacteria or the lumen of the endoplasmic reticulum of eukaryotic cells and usually favors the formation of structural disulfide bonds. [00264] In some embodiments, it is also possible to produce an antibody, antigen-binding domain thereof, or fusion protein of the disclosure in the cytosol of a host cell, preferably E. coli. In this case, a provided antibody, antigen-binding domain thereof, or fusion protein can either be directly obtained in a soluble and folded state or recovered in the form of inclusion bodies, followed by renaturation in vitro. A further option is the use of specific host strains having an oxidizing intracellular milieu, which may thus allow the formation of disulfide bonds in the cytosol (Venturi et al., J Mol Biol, 2002). [00265] In some embodiments, an antibody, antigen-binding domain thereof, or fusion protein of the disclosure as described herein may be not necessarily generated or produced, in whole or in part, via use of genetic engineering. Rather, such protein can also be obtained by any of the many conventional and well-known techniques such as plain organic synthesis strategies, solid phase-assisted synthesis techniques, commercially available automated synthesizers, or by in vitro transcription and translation. It is, for example, possible that promising antibodies, antigen- binding domains thereof, or fusion proteins or lipocalin muteins included in such fusion proteins are identified using molecular modeling, synthesized in vitro, and investigated for the binding activity for the target(s) of interest. Methods for the solid phase and/or solution phase synthesis of proteins are well known in the art (see, e.g., Bruckdorfer et al., Curr Pharm Biotechnol, 2004). [00266] In some embodiments, an antibody, antigen-binding domain thereof, or fusion protein of the disclosure may be produced by in vitro transcription/translation employing well- established methods known to those skilled in the art. [00267] In some further embodiments, antibodies, antigen-binding domains thereof, or fusion proteins as described herein may also be prepared by conventional recombinant techniques alone or in combination with conventional synthetic techniques. [00268] Moreover, in some embodiments, a fusion protein according to the present disclosure may be obtained by conjugating together individual subunits, e.g., antibodies and Attorney Docket No.01218-0029-00PCT muteins as included in the fusion protein. Such conjugation can be, for example, achieved through all forms of covalent or non-covalent linkage using conventional methods. [00269] The skilled worker will appreciate methods useful to prepare antibodies, antigen- binding domains thereof, or fusion proteins contemplated by the present disclosure but whose protein or nucleic acid sequences are not explicitly disclosed herein. As an overview, such modifications of the amino acid sequence include, e.g., directed mutagenesis of single amino acid positions to simplify sub-cloning of a protein gene or its parts by incorporating cleavage sites for certain restriction enzymes. Also, these mutations can be incorporated to further improve the affinity of an antibody, antigen-binding domain thereof, or fusion protein for its target(s) (e.g., CD137 and/or CD228). Furthermore, mutations can be introduced to modulate one or more characteristics of the protein such as to improve folding stability, serum stability, protein resistance or water solubility or to reduce aggregation tendency, if necessary. [00270] Additional objects, advantages, and features of this disclosure will become apparent to those skilled in the art upon examination of the following Examples and the attached Figures, which are not intended to be limiting. Thus, it should be understood that although the present disclosure is specifically disclosed by exemplary embodiments and optional features, modification and variation of the disclosures embodied therein herein disclosed may be resorted to by those skilled in the art and that such modifications and variations are considered to be within the scope of this disclosure. VI. EXAMPLES [00271] Example 1: Generation of Anti-CD228 Antibodies [00272] A human immunoglobulin transgenic rat strain (OmniRat®; OMT, Inc.) was used to develop monoclonal antibody expressing hybridoma cells. The OmniRat® contains a chimeric human/rat IgH locus (comprising 22 human VHs, all human D and JH segments in natural configuration linked to the rat CH locus) together with fully human IgL loci (12 Vκs linked to Jκ-Cκ and 16 Vλs linked to Jλ-Cλ). See Osborn, et al. (2013) J Immunol 190(4): 1481-1490); WO 2014/093908. In response to immunization, the introduced human heavy and light chain transgenes undergo class switching and somatic mutation to generate high affinity IgG monoclonal antibodies. [00273] Transgenic rats were immunized with recombinant human CD228 protein. A subcutaneous injection included 48.2 µg of recombinant protein in Complete Freund’s adjuvant on day 1 and 48 µg of recombinant protein in Incomplete Freund’s adjuvant on days 33, 81, and 127. [00274] The presence of antibodies directed against human CD228 in sera of transgenic Attorney Docket No.01218-0029-00PCT rats was monitored by flow cytometry on days 33, 81, and 127 with RPMI-7951 cells engineered to stably express human CD228. Transgenic rats with detectable immune responses were boosted five and seven days before harvesting spleen and lymph nodes. The boost prior to organ harvest included a 90 µg intravascular injection and a 18 µg intraperitoneal injection of recombinant human CD228 protein suspended in Phosphate Buffer Solution (PBS). [00275] Splenocytes and lymphocytes demonstrating B cell lineage and specificity to fluorescently labeled recombinant human CD228 were sorted as single cells by flow activated cell sorting (FACS) into cell lysis buffer. RNA from single B cells was reverse transcribed to cDNA and amplified from primer sets to known human variable regions. Amplicons underwent DNA purification for Sanger sequencing with appropriate sequencing primers, after which heavy and light chain sequences were identified and annotated by IgBLAST. [00276] Productive heavy and light chain sequences of human variable regions identified from single B cell cloning and Sanger sequencing were selected for gene synthesis and cloned into expression vectors using traditional cloning techniques. The expression vectors encode human IgG1, human kappa, or human lambda constant regions that are in frame and downstream of the human variable region. Sanger sequencing verified the antibody expression constructs. [00277] To generate full length antibodies, heavy and light chain expression vectors were co-transfected into ExpiCHO cells according to the manufacturer’s protocols (Life Technologies). Cell culture supernatants were collected by centrifugation 9 days after transfection. Filtered culture supernatants were spiked with 10% Triton-X (final concentration 0.1%) and left overnight at 4 °C on a shake platform. Supernatants were then loaded onto a 5 mL HiTrap MabSelectSuRe column at 2.5 mL/min using an ÄKTA Avant 25 chromatography system. The column was then washed with: 5 column volumes (CV) of endo wash buffer (1X PBS, 0.1% Triton-X), 5 CV of high salt buffer (1X PBS, 0.5 M NaCl), and 7.5 CV standard 1X PBS, pH 7.4. Protein was eluted with 1 CV of 20 mM citrate, pH 3, and immediately buffer exchanged into 1X PBS, pH 7.4 using a HiPrep Sephadex G-25 Desalting column. Fractionation was determined via A280 UV watch parameters, allowing for optimal collection of purified protein. [00278] Figs.2A and 2B show the CDR sequences of the antibodies, as defined by Kabat (Fig.2A) and IMGT (Fig.2B). [00279] Example 2: Binding of Anti-CD228 Antibodies to Recombinant Human CD228 [00280] Humanized antibodies specific for human melanotransferrin (CD228) were evaluated for binding to recombinant human CD228 (R & D Systems) by ELISA. Antibodies were titrated on plates coated with 1 µg/ml recombinant human CD228, detected with an HRP-labeled Attorney Docket No.01218-0029-00PCT goat anti-human IgG secondary reagent (Thermo), and developed with TMB substrate (Thermo). OD450 values were read on a plate reader with SoftMax Pro software. OD450 values for respective experiments were transferred to GraphPad Prism 8 for plotting and analysis. The results are shown in Fig.3. [00281] Example 3: Binding of Anti-CD228 Antibodies to Cells [00282] Humanized antibodies specific for human CD228 were evaluated for relative binding to CD228-expressing cancer cell lines. Cells were incubated with a titration of Alexa-647 labeled humanized antibody clones, washed, and evaluated for fluorescence intensity by flow cytometry on an Attune NXT flow cytometer. MFI values were transferred to GraphPad Prism 8 for plotting and analysis. EC50 values were determined by non-linear regression. The numbers next to the names of the cell lines reflect estimated surface CD228 copy number as estimated using a QIFIKIT quantitative analysis kit (Agilent). [00283] Results are shown in Figs.4A-4C as raw Mean Fluorescent Intensity (MFI) values. Fig.4D shows the EC ₅₀ value of each antibody for each cell line. [00284] Example 4: Binding Profiles of Anti-CD228 Antibodies [00285] The binding kinetics and affinity of human CD228 with a C-terminal polyhistidine tag (R&D Systems) were determined by biolayer inferometry (BLI) using an Octet ^® RED384 system (Sartorius). Anti-human antibody capture AHC (GE Healthcare) biosensors were used for the analysis. Subsequently, anti-CD228 antibodies OMT8, OMT24, OMT30, OMT35, and OMT36 (IgG) at 0.5 µg/mL in HBS-EP+ buffer were captured by the anti-human IgG-Fc antibody at the chip surface for 180 seconds. After each capture step, the biosensors were washed in HBS-EB+ blank. For affinity determination, dilutions of recombinant huCD228 (100 nM, 40 nM, 16 nM, 6.4 nM, 2.6 nM, and 1.0 nM) or blanks were prepared in HBS-EP+ buffer and applied to the biosensor. The binding assay was carried out with a contact time of 300 seconds and a dissociation time of 1,200 seconds. All measurements were performed at 25 °C. Fresh AHC biosensors were used for each analysis. Data were evaluated with Satorius Octet ^® Data Analysis Software (v12.0), and results are shown in Figs.5A-5E. Single referencing was used, and the 1:1 binding model was used to fit the raw data. [00286] Example 5: Lack of Cross-reactivity of Anti-CD228 Antibodies to Transferrin and Lactotransferrin [00287] Humanized antibodies specific for human melanotransferrin (CD228) were evaluated for binding to related transferrin family members, lactotransferrin and transferrin, by ELISA. Antibodies were titrated on plates coated with 1 µg/mL of recombinant human melanotransferrin (CD228), lactotransferrin, and transferrin and detected with an HRP-labeled Attorney Docket No.01218-0029-00PCT goat anti-human IgG secondary reagent (Thermo) developed with TMB substrate (Thermo), and OD450 values were read on a plate reader with SoftMax Pro software. OD450 values for respective experiments were transferred to GraphPad Prism 8 for plotting and analysis. These results, as shown in Figs. 6A-6C, indicate the absence of cross-reactive binding of the anti- CD228 antibodies to other transferrin family members. [00288] Example 6: Cross-reactivity of Anti-CD228 Antibodies to Non-human CD228 [00289] The CD228-negative human melanoma cell line RPMI-7951 was engineered to express cynomolgus CD228 to test binding of anti-CD228 antibody clones. 50,000 RPMI-7951 cells were incubated with a titration of humanized antibody clones, washed with staining buffer to remove excess antibodies, and incubated with 250 ng/mL fluorochrome-labeled monoclonal antibodies against human IgG1 (Thermo) to detect bound antibodies. Results are shown in Fig. 7A as Mean Fluorescent Intensity (MFI) as determined by measurement on an Attune NXT flow cytometer (Thermo). [00290] Reactivity of CD228 antibody clones was tested against recombinant his-tagged cynomolgus CD228 by ELISA. Plates were coated with 1 µg/ml of recombinant his-cynomolgus CD228 in phosphate buffered saline, plates were blocked with 250 µL of Superblock Blocking Buffer (Thermo), antibodies were titrated across the plate, bound antibodies were detected with HRP-labeled goat anti-human IgG (Sigma), developed with TMB substrate (Thermo), and OD450 values were read on a plate reader with SoftmaxPro software. Results are shown in Fig.7B. [00291] Humanized antibodies specific for human CD228 were evaluated for binding to recombinant cynomolgus and murine CD228 by ELISA. Antibodies were titrated on plates coated with 1 µg/mL recombinant human, cynomolgus, and murine CD228 and detected with an HRP labeled goat anti-human IgG secondary reagent (Thermo) developed with TMB substrate (Thermo), and OD450 values were read on a plate reader with SoftMax Pro software. OD450 values for respective experiments are shown in Figs.8A-8C. [00292] These results indicate that a subset of antibody clones (OMT30, OMT35, OMT24, OMT8, OMT36, and L235) are cross-reactive to cynomolgus CD228 with minimal cross reactivity to murine CD228. [00293] Example 7: Anti-CD228 Antibody Cross-competition Assay [00294] Fluorochrome-labeled humanized monoclonal antibodies specific for human CD228 were evaluated for binding to the HT-1080 tumor cell line and the SK-MEL-5 tumor cell line in competition with unlabeled humanized monoclonal antibodies to identify potential shared epitopes. HT-1080 cells and SK-MEL-5 cells were preincubated with 1 µg/mL of unlabeled humanized antibodies in staining buffer for 30 minutes at 4 °C, washed in staining buffer, and Attorney Docket No.01218-0029-00PCT incubated with 200 ng/mL of A647-labeled versions. Bound fluorochome-labeled antibody was determined by flow cytometry on an Attune NXT flow cytometer. Results are shown in Fig.9 as the percent of Mean Fluorescent Intensity compared to preincubation with a control non-binding human IgG1 (Sigma). Clones with direct epitope or steric competition show low staining percentage values. These results indicate antibody clones 28, 32, and 35 may share a closely related epitope, and clones 8, 11, 24, 30, and 36 likely bind unique CD228 epitopes. [00295] Example 8: Internalization of antibody clones on CD228+ tumor cell lines [00296] Humanized antibodies specific for human CD228 were evaluated for relative internalization on the CD228-expressing melanoma and lung cancer cell lines, SK-MEL-5 and Calu-1, respectively. Cells were incubated with 2 µg/ml of anti-CD228 antibodies, washed three times to remove free antibody, and incubated at 37 °C in 5% CO ₂ for the times shown. At each timepoint, cells were fixed (BD Cytofix) and stained for bound human IgG1 with a fluorochome- labeled anti-human IgG1 antibody (Invitrogen). At the end of the time-course, cells were evaluated by FACS for the mean fluorescence intensity of surface bound antibody on an Attune NXT flow cytometer. Internalization results are shown in Fig.10 as the % of MFI relative to the starting MFI at timepoint 0. These results demonstrate a range of internalization rates among antibody clones, with antibody clone 35 internalizing at the fastest rate. [00297] Example 9: Expression and analysis of representative fusion proteins [00298] Representative antibody-lipocalin mutein fusion proteins were generated by fusing together CD228-specific antibodies and CD137-specific lipocalin muteins, such as the lipocalin mutein of SEQ ID NO: 40, via a linker, such as the unstructured (G ₄ S) ₃ linker of SEQ ID NO: 13, to engage CD228 and CD137 at the same time. Two different exemplary CD228-specific antibodies were used. A first CD228-specific antibody had the heavy chain provided by SEQ ID NO: 75 (or comprised a heavy chain variable domain of SEQ ID NO: 70, or comprised the heavy chain CDRs (HCDR1, HCDR2, HCDR3) of SEQ ID NOs: 58-60) and had the light chain provided by SEQ ID NO: 76 (or comprised a light chain variable domain of SEQ ID NO: 71, or comprised the light chain CDRs (LCDR1, LCDR2, LCDR3) of SEQ ID NOs: 61-63). A second CD228-specific antibody had the heavy chain provided by SEQ ID NO: 78 (or comprised a heavy chain variable domain of SEQ ID NO: 72, or comprised the heavy chain CDRs (HCDR1, HCDR2, HCDR3) of SEQ ID NOs: 64-66) and had the light chain provided by SEQ ID NO: 79 (or comprised a light chain variable domain of SEQ ID NO: 73, or comprised the light chain CDRs (LCDR1, LCDR2, LCDR3) of SEQ ID NOs: 67-69). The different formats of the fusion proteins that were generated are depicted in Figs.11A-11I, including fusion proteins that are bivalent to CD137 (e.g., as shown in Figs.11A-11D) or tetravalent to CD137 (e.g., as shown in Figs.11E-11H), or have even higher valency to CD137 (e.g., as shown in Fig.11I). The exemplary fusion proteins of SEQ ID NOs: 80 Attorney Docket No.01218-0029-00PCT and 76 (encoded by the nucleotide sequences of SEQ ID NOs: 91 and 88), SEQ ID NOs: 82 and 79 (encoded by the nucleotide sequences of SEQ ID NOs: 93 and 90), SEQ ID NOs: 75 and 81 (encoded by the nucleotide sequences of SEQ ID NOs: 87 and 92), and SEQ ID NOs: 78 and 83 (encoded by the nucleotides sequences of SEQ ID NOs: 89 and 94) were bivalent to CD137 with a CD137-specific lipocalin mutein being fused to the C-terminus of each heavy chain (Fig.11A) or each light chain (Fig.11B). [00299] The CD228-specific antibodies as well as all antibody lipocalin mutein fusion proteins described in this Example had an engineered IgG4 backbone, which contained a S228P mutation to minimize IgG4 half-antibody exchange in-vitro and in-vivo (Silva et al., J Biol Chem, 2015). Additional mutations in the IgG4 backbones may also exist in all antibodies and fusion proteins described here, including any one or more of mutations F234A, L235A, M428L, N434S, M252Y, S254T, and T256E. F234A and L235A mutations may be introduced to decrease ADCC and ADCP (Glaesner et al., Diabetes Metab Res Rev, 2010). M428L and N434S mutations or M252Y, S254T, and T256E mutations may be introduced for extended serum half-life (Dall'Acqua et al., J Biol Chem, 2006; Zalevsky et al., Nat Biotechnol, 2010). All antibodies were expressed without the carboxy-terminal lysine to avoid heterogeneity. [00300] The present disclosure also embodies asymmetrical antibody-lipocalin mutein fusion formats where, for example, one light chain of the antibody may be fused with a lipocalin mutein while the other is not. [00301] The constructs of exemplary fusion proteins were generated by gene synthesis and cloned into a mammalian expression vector. They were then transiently expressed in suspension-adapted CHO-K1 cells. The concentration of fusion proteins in the cell culture medium was measured by BLItz (ForteBio) using Protein A biosensors for kinetic assays. The titers obtained are summarized in Table 1. [00302] Table 1: Transient expression titers [00303] For each construct, one liter of the aforementioned cell culture medium was purified using Protein A chromatography (MabSelect SuRe, Cytiva), followed by size-exclusion Attorney Docket No.01218-0029-00PCT chromatography (SEC) in a buffer composed of 20 mM Histidine, 60 mM NaCl, pH 6.0. Purity, i.e., monomer content, of the final product was determined by analytical size exclusion chromatography with an Agilent AdvanceBio SEC column (300 Å, 2.7 ^m, 7.8 x 300 mm) and DPBS as running buffer at 0.8 ml/min. Yield and monomer content after the two-step purification process are summarized in Table 2. [00304] Table 2: Recovery and monomer content of transient expression [00305] Example 10: Binding of fusion proteins towards huCD228, cyCD228 or huCD137 determined by surface plasmon resonance (SPR) [00306] The binding kinetics and affinity of exemplary fusion proteins to huCD228-His, cyCD228-His or huCD137-His (recombinant human CD228, cynomolgus CD228 or human CD137 with a C-terminal polyhistidine tag, R&D Systems) were determined by surface plasmon resonance (SPR) using a Biacore 8K (GE Healthcare). [00307] The anti-human IgG Fc antibody (GE Healthcare) was immobilized on a CM5 sensor chip according to the manufacturer’s instructions. Subsequently, fusion proteins to be tested (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83) at 0.5 µg/mL in HBS-EP+ buffer were captured by the anti-human IgG-Fc antibody at the chip surface for 180 s at a flow rate of 10 µL/min. After each capture step, the needle was washed. hIgG1 versions of the two anti-CD228 antibodies included in the fusion proteins (SEQ ID NOs: 74/76 and 77/79, respectively) were also tested as controls. [00308] For affinity determination, dilutions of recombinant huCD228-His, cyCD228-His or huCD137-His (500 nM, 125 nM, 31.25 nM and 7.8 nM) or blanks were prepared in HBS-EP+ buffer and applied to the prepared chip surface. The binding assay was carried out with a contact time of 180 s, a dissociation time of 1200 s and a flow rate of 30 µL/min. All measurements were performed at 25°C. Regeneration of the chip surface was achieved with injections of 3 M MgCl ₂ for 120 s. Prior to the protein measurements, three startup cycles were performed for conditioning purposes. Data were evaluated with Biacore 8K Evaluation software (V1.1.1). Double referencing was used and the 1:1 binding model was used to fit the raw data. Attorney Docket No.01218-0029-00PCT [00309] The values determined for k _on , k _off , and the resulting equilibrium dissociation constant (K _D ) for representative fusion proteins and the control antibodies are summarized in Table 3. All bispecific fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83) bound huCD228 and cyCD228 with nanomolar affinities similar to the respective antibodies. All bispecific fusion proteins bound CD137 with affinities in the single digit nanomolar range. [00310] Table 3: Kinetic constants and affinities of fusion proteins determined by SPR [00311] Example 11: Binding of fusion proteins towards CD228 or CD137 in enzyme- linked immunosorbent assay (ELISA) [00312] An enzyme-linked immunosorbent assay (ELISA) was employed to determine the binding potency of exemplary fusion proteins to human CD228 or CD137 and cynomolgus CD228 or CD137. [00313] Recombinant huCD228-His or cyCD228-His (human or cynomolgus CD228 with a C-terminal polyhistidine tag, R&D Systems or Sino Biologics) at the concentration of 1 µg/mL in PBS was coated overnight on microtiter plates at 4 °C. After washing with PBS-0.05%T (PBS supplemented with 0.05% (v/v) Tween 20), the plates were blocked with 2% BSA (w/v) in PBS- 0.1%T (PBS supplemented with 0.1% (v/v) Tween 20) for 1 h at room temperature. After washing with 100 µL PBS-0.05%T five times, exemplary fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83) and hIgG1 versions of the two anti-CD228 antibodies included in the fusion proteins (SEQ ID NOs: 74/76 and 77/79, respectively) at different concentrations (maximum concentration: 200 nM) were added to the wells and incubated for 1 h at room temperature, followed by another wash step. Bound molecules Attorney Docket No.01218-0029-00PCT under study were detected by incubation with 1:5000 diluted anti-human IgG Fab-HRP (Jackson Laboratory) in PBS-0.1%T-2%BSA. After an additional wash step, fluorogenic HRP substrate (QuantaBlu, Thermo) was added to each well and the fluorescence intensity was detected using a fluorescence microplate reader. [00314] The same ELISA setup was also employed to determine the binding potency of fusion proteins to CD137, where huCD137-His (human CD137 with C-terminal polyhistidine tag, R&D Systems) or cyCD137-Fc (cynomolgus CD137 C-terminally fused to Fc) was instead coated on a microtiter plate. The testing agents were similarly titrated, and bound agents were detected via anti-human IgG Fab-HRP (Jackson Laboratory). [00315] The results of exemplary experiments are depicted in Figs.12A-12D, together with the fit curves resulting from a 4PL fit, where the EC ₅₀ value and the maximum signal were free parameters, and the slope was fixed to unity. The resulting EC ₅₀ values are provided in Table 4. [00316] The observed EC ₅₀ values of the fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83) towards human CD228 and cynomolgus CD228 were similar to those of the respective antibodies. The curves of the antibodies exhibited an approx. 2-fold higher plateau, potentially due to detection antibody accessibility, as such difference was not observed when an anti-human IgG Fc-HRP antibody was used for detection (data not shown). In terms of binding to human CD137, all fusion proteins showed EC ₅₀ values in the low single digit nanomolar range, and the fusion proteins also showed cross-reactivity with cynomolgus CD137. Overall, the EC ₅₀ values of the HC fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79) were slightly better than those of the corresponding LC fusion proteins (SEQ ID NOs: 75 and 81, SEQ ID NOs: 78 and 83). [00317] Table 4: ELISA data for CD228 or CD137 binding [00318] Example 12: Simultaneous binding of fusion proteins to CD228 and CD137 Attorney Docket No.01218-0029-00PCT in ELISA [00319] In order to demonstrate the simultaneous binding of exemplary fusion proteins to CD228 and CD137, a dual-binding ELISA format was used. [00320] Recombinant huCD228-His in PBS (1 µg/mL) was coated overnight on microtiter plates at 4 °C. The plates were washed five times after each incubation step with 100 µL PBS- 0.05%T. The plates were blocked with 2% BSA (w/v) in PBS-0.1%T for 1 h at room temperature and subsequently washed again. Different concentrations (maximum concentration: 200 nM) of tested fusion proteins and control anti-CD228 antibodies were added to the wells and incubated for 1 h at room temperature, followed by a wash step. Subsequently, biotinylated huCD137-His (huCD137-His-Bio, Sino Biological) was added at a constant concentration of 1 µg/mL in PBS- 0.1%T-2%BSA for 1 h. After washing, a 1:5000 dilution of ExtrAvidin-HRP (Sigma-Aldrich) in PBS-0.1%T-2%BSA was added to the wells and incubated for 1 h. After an additional wash step, fluorogenic HRP substrate (QuantaBlu, Thermo) was added to each well, and the fluorescence intensity was detected using a fluorescence microplate reader. [00321] Dual binding data of fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83) are shown in Fig.13, together with the fit curves resulting from a 4PL fit, where the EC ₅₀ value and the maximum signal were free parameters, and the slope was fixed to unity. The EC ₅₀ values are summarized in Table 5. In contrast to the anti-CD228 antibodies, all bispecific fusion proteins showed clear binding signals, demonstrating that the fusion proteins are able to engage CD228 and CD137 simultaneously. Additional dual binding data of fusion proteins (SEQ ID NOs: 76, 79, 80, 82, 96, 97, 100, 101, 102, and 103) are shown in Fig.14. [00322] Table 5: ELISA data for simultaneous target binding of both CD228 and CD137 [00323] Example 13: Flow cytometric analysis of fusion proteins binding to CD228- Attorney Docket No.01218-0029-00PCT positive cells and cells expressing human or cynomolgus CD137 [00324] Target specific binding of fusion proteins to human CD228-expressing cells and human and cynomolgus CD137-expressing cells was assessed by flow cytometry (FACS). [00325] Human melanoma SH-4 cells (ATCC, CRL7724) were used as huCD228-positive cells. CHO cells were stably transfected with human CD137 or cynomolgus CD137 using the Flp- In system (Life technologies) according to the manufacturer´s instructions. [00326] SH-4 cells were maintained in DMEM (PAN Biotech) supplemented with 10% Fetal Calf Serum (Biochrom). Transfected CHO cells were maintained in Ham´s F12 medium (Life technologies) supplemented with 10% Fetal Calf Serum (Biochrom) and 500 µg/ml Hygromycin B (Roth). Cells were cultured in cell culture flasks according to manufacturer’s instruction (37 °C, 5% CO ₂ atmosphere). [00327] For flow cytometric analysis, respective cell lines were incubated with fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83), hIgG1 versions of the two anti-CD228 antibodies included in the fusion proteins (SEQ ID NOs: 74/76 and 77/79, respectively), an IgG4 isotype control (SEQ ID NOs: 24 and 25) and a monoclonal anti-CD137 antibody (SEQ ID NOs: 26 and 27), which were detected using a fluorescently labeled anti-human IgG antibody or anti-human NGAL antibody in FACS analysis as described in the following: [00328] 5 × 10 ⁴ cells per well were incubated for 1 h in ice-cold PBS containing 5% fetal calf serum (PBS-FCS). A dilution series of the fusion proteins and control antibodies were added to the cells and incubated for 1 h on ice. Cells were washed twice with PBS and then incubated with a goat anti-hIgG Alexa647-labeled antibody (Life Technologies) or a proprietary rabbit anti- NGAL Alexa488-labeled antibody for 30 min on ice. Cells were subsequently washed and analyzed using iQue Flow cytometer (Intellicyte Screener). Mean geometric fluorescent signals were plotted and fitted with Graphpad software using nonlinear regression (shared bottom, SLOPE =1). [00329] The ability of fusion proteins to bind human CD228 as well as human CD137 expressed on cells is shown in Figs. 15A-15B. Binding affinities (EC ₅₀ s) of bispecific fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83) towards huCD228- or huCD137-expressing cells are in the (low) single digit nanomolar range, comparable to those of the respective control antibodies, and the tested fusion proteins are fully cross-reactive with cynomolgus CD137 – in contrast to the anti-CD137 antibody of SEQ ID NOs: 26 and 27 (summarized in Table 6). None of the fusion proteins bound to mock transfected cells (data not shown). Attorney Docket No.01218-0029-00PCT [00330] Table 6: Binding affinities of the fusion proteins to cells expressing human CD228, human CD137 or cynomolgus CD137 (“N/A”: not measured; “--": no value calculated) [00331] Example 14: CD228-dependent T cell co-stimulation using a CD137 reporter cell assay [00332] The potential of exemplary fusion proteins to induce activation of the CD137 signaling pathway in the presence of CD228 was assessed using a commercially available double stable transfected Jurkat cell line expressing CD137 and the luc2 gene (humanized version of firefly luciferase, wherein luc2 expression was driven by a NFκB-responsive element. In this bioassay, CD137 engagement results in CD137 intracellular signaling, leading to NFκB-mediated luminescence. [00333] Tumor cells expressing high (SH-4), medium (A375) or low (A549) levels of CD228 as well as CD228-negative RPMI-7951 tumor cells were cultured under standard conditions. One day prior to the assay, tumor cells were plated at 1.25 x 10 ⁴ cells per well and allowed to adhere overnight at 37 °C in a humidified 5% CO ₂ atmosphere. [00334] The next day, 3.75 x 10 ⁴ NF-kB-Luc2/CD137 Jurkat reporter cells were added to each well, followed by the addition of various concentrations ranging from 0.0002 nM to 10 nM of fusion proteins, a reference anti-CD137 antibody (urelumab, SEQ ID NOs: 26 and 27) or a fusion of IgG4 with the CD137-specific lipocalin mutein included in the fusion proteins (SEQ ID NO: 84). Plates were covered with a gas permeable seal and incubated at 37 °C in a humidified 5% CO ₂ atmosphere. After 4 hours, 30 µL Bio-Glo™ Reagent was added to each well and the bioluminescent signal was quantified using a luminometer (PHERAstar). Four-parameter logistic curve analysis was performed with GraphPad Prism ^® to calculate EC ₅₀ values (shared bottom, Attorney Docket No.01218-0029-00PCT fixed slope) which are summarized in the table in Fig. 16E. To demonstrate the CD228 dependency of CD137 engagement by fusion proteins, the same experiment was performed in parallel in the presence of CD228-negative cells (RPMI-7951) or in the absence of target cells. The assay was performed in triplicates. [00335] The results of a representative experiment are depicted in Figs. 16A-16D. The data demonstrate that all tested fusion proteins induced dose-dependent, CD137-mediated activation of T cells in the presence of tumor cells expressing CD228. In the case of SH-4 cells (Fig.16A) and A375 cells (Fig.16B), all fusion proteins resulted in a maximum activation that was higher than that of the reference anti-CD137 antibody. Overall, the fusion molecule of SEQ ID NOs: 80 and 76 (which is based on the CD228-specific antibody of SEQ ID NOs: 75 and 76) showed better, i.e., lower, EC ₅₀ values than the other fusion molecules. And Fig.16D shows that the activation of CD137 by fusion proteins was CD228-dependent because no activation of the NF-kB-Luc2/CD137 Jurkat cells was detected in CD228-negative cells (RPMI-7951). In contrast, the reference anti-CD137 mAb (SEQ ID NOs: 26 and 27) showed CD137-mediated T cell co- stimulation regardless of CD228 expression. [00336] Example 15: Assessment of PBMC response to viral peptides in the presence of tumor cells with high or no expression of CD228 [00337] An assay was conducted to assess the ability of the fusion proteins to co-stimulate innate and adaptive immune cytokines from PBMC in response to viral peptides in a CD228 target dependent manner. PBMC isolated from healthy donors were co-incubated with engineered CD228+ or wild type (CD228-) RPMI-7951 tumor cell lines 10:1 in RPMI with 10% FCS, and were provided viral peptides from CMV, EBV, and flu viruses (CEF peptides). Bispecific fusion proteins and controls were titrated into the assay and changes in IFN- ^, TNF- ^, IL-5, IL-12 (p70), and CXCL10 (IP-10) were measured at the end of 4-day stimulation at 37 °C in 5% CO ₂ . Supernatants were evaluated by Luminex ^® multiplex cytokine array for T cell (IFN- ^ and TNF- ^) and myeloid cell (IL-12 and CXCL10) responses. Data are shown in Figs.17A-17E as the fold change of cytokines relative to the untreated control wells. Samples were pooled from triplicate test article treatments prior to cytokine measurement. These results show that the fusion proteins (e.g., AAF30 (HC)), compared to antibodies, co-stimulate various innate and adaptive immune cytokines from PBMC in response to viral peptides in a CD228 target dependent manner and a dose dependent manner. [00338] Example 16: Assessment of PBMC cellular response to viral peptides in the presence of CD228+ tumor cells [00339] To evaluate the activity of bispecific fusion proteins in an antigen recall assay, healthy donor cryopreserved PBMC (Bloodworks Northwest) were thawed in prewarmed RPMI Attorney Docket No.01218-0029-00PCT 10% FCS, washed, and labeled with 1.5 mL of 10 nM CFSE PBS 3% FCS at room temperature. To quench the labeling reaction, cells were washed 2x with 12 mL of RPMI 10% FCS. Cells were counted and mixed 10:1 with CD228-expressing cell line RPMI-7951 ^CD228 (ATCC) (engineered to express human CD228) in RPMI-complete (10% FCS, 1X Glutamax, 1X MEM NEAA, 1X sodium pyruvate, 1X penicillin/streptomycin (Gibco™)) and distributed in non-adherent 96-well round bottom plates (Sbio), 1.5 x 10 ⁵ cells/well. CEF peptides were added to a final concentration of 100 ng/mL, and fusion protein bispecifics and controls were added at equimolar titrations in triplicate. The assay was allowed to incubate for 5 days at 37 °C 5% CO ₂ . Fig.18, Fig.19, and Fig.20 respectively show representative examples of CD8 T cell, NK cell and CD8 T cell / Treg ratio calculations from antigen recall assays in coculture with CD228-engineered RPMI-7951 cell line. Similarly, Fig.21 shows representative examples of NK cell and CD8 T cell / Treg ratio calculations from antigen recall assays. These results show that the fusion proteins (e.g., AAF30 (HC)), compared to antibodies, co-stimulate proliferation/division of CD8+ T cells and NK cells in response to viral peptides in a CD228 target dependent manner and a dose dependent manner. [00340] Example 17: Assessment of PBMC cytokine response to viral peptides in the presence of CD228+ tumor cells [00341] To evaluate the activity of bispecific fusion proteins in an antigen recall assay, healthy donor cryopreserved PBMC (Bloodworks Northwest) were thawed in prewarmed RPMI 10% FCS, washed, and labeled with 1.5 mL of 10 nM CFSE PBS 3% FCS at room temperature. To quench the labeling reaction, cells were washed 2x with 12 mL of RPMI 10% FCS. Cells were counted and mixed 10:1 with CD228-expressing cell line RPMI-7951 ^CD228 (ATCC) (engineered to express human CD228), CALU-1 cells (ATCC), or H3677 cells (Seagen) in RPMI-complete (10% FCS, 1X Glutamax, 1X MEM NEAA, 1X sodium pyruvate, 1X penicillin/streptomycin (Gibco™)) and distributed in non-adherent 96-well round bottom plates (Sbio), 1.5 x 10 ⁵ cells/well. CEF peptides were added to a final concentration of 100 ng/mL, and fusion protein bispecifics and controls were added at equimolar titrations in triplicate. The assay was allowed to incubate for 5 days at 37 °C 5% CO ₂ . At the end of the assay, plates were spun down, and supernatants were collected for cytokine evaluation. Cytokines were measured using the MILLIPLEX ^® MAP Human CD8+ T Cell Magnetic Bead Panel Premixed 17 Plex and read out on a Luminex ^® MAGPIX ^® system. Figs. 22A-22E show representative fold-change averages of cytokines across three CD228 expressing cell lines. Raw Luminex ^® data were exported and analyzed in Microsoft Excel. These data highlight consistent changes in cytotoxic effector molecules and diverse cytokines in antigen recall in the presence of bispecific fusion proteins. [00342] Example 18: Assessment of T cell activation in presence of tumor cells with high or no expression of CD228 Attorney Docket No.01218-0029-00PCT [00343] A further T cell assay was employed to assess the ability of the fusion proteins to co-stimulate T cell activation in a CD228 target-dependent manner. Fusion proteins were applied at different concentrations to anti-CD3 stimulated T cells, in the presence of tumor cell lines with different CD228 expression levels. Tested tumor cell lines included SH-4 (CD228 high), Calu-1 (CD228 high), SK-MEL-24 (CD228 high) and RPMI-7951 (CD228-negative). IL-2 secretion levels in the supernatants were used as read-out. [00344] PBMCs from healthy volunteer donors were isolated from buffy coats by centrifugation through a polysucrose density gradient (Biocoll, 1.077 g/mL, Biochrom), following Biochrom´s protocols. T lymphocytes were further purified from PBMC by magnetic cell sorting using a Pan T cell purification Kit (Miltenyi Biotec GmbH) following the manufacturer´s instructions. Purified Pan T cells were resuspended in a buffer consisting of 90% FCS and 10% DMSO, immediately frozen down and stored in liquid nitrogen until further use. [00345] For the assay, T cells were thawed and rested in culture media (RPMI 1640, Life Technologies) supplemented with 10% FCS and 1% Penicillin-Streptomycin (Life Technologies) for 16 h at 37 °C in a humidified 5% CO ₂ atmosphere. [00346] The following procedure was performed using triplicates for each experimental condition: flat-bottom tissue culture plates were pre-coated with 0.25 μg/mL anti-human CD3 antibody for 1 h at 37 °C and then washed twice with PBS. Tumor cell lines SH-4, Calu-1, SK- MEL-24 and RPMI-7951 were treated for 30 min with 30 µg/ml mitomycin C (Sigma Aldrich) in order to block proliferation. Mitomycin treated tumor cells were then washed twice with PBS and plated at 2.5 x 10 ⁴ cells per well in culture medium to allow adhesion overnight at 37 °C in a humidified 5% CO ₂ atmosphere. The target cells had before been grown under standard conditions, detached using Accutase (PAA Laboratories), and resuspended in culture media. [00347] On the next day, after washing the plates twice with PBS, 1.25 x 10 ⁴ T cells per well were added to the tumor cells. A dilution series of bispecific fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, and SEQ ID NOs: 78 and 83), reference anti-CD137 antibody (SEQ ID NOs: 26 and 27), a fusion of IgG4 with the CD137-specific lipocalin mutein included in the fusion proteins (SEQ ID NO: 84), or an isotype control (SEQ ID NOs: 24 and 25), ranging from 0.0002 nM to 10 nM, was added to corresponding wells. Plates were covered with a gas permeable seal and incubated at 37 °C in a humidified 5% CO ₂ atmosphere for 3 days. [00348] After 3 days of co-culturing, IL-2 levels in the supernatant were assessed using the human IL-2 DuoSet kit (R&D Systems) as described in the following: 384 well plates were coated for 2 h at room temperature with 1 μg/mL “Human IL-2 Capture Antibody” in PBS. Subsequently, wells were washed 5 times with 80 µl PBS supplemented with 0.05 % Tween (PBS- Attorney Docket No.01218-0029-00PCT T). After 1 h blocking in PBS-0.05%T containing 1% casein (w/w), assay supernatants and a concentration series of IL-2 standard diluted in culture medium was transferred to respective wells and incubated overnight at 4 °C. The next day, a mixture of 100 ng/mL goat anti-hIL-2-Bio detection antibody (R&D Systems) and 1 µg/mL Sulfotag-labelled streptavidin (Mesoscale Discovery) in PBS-T containing 0.5% casein were added and incubated at room temperature for 1 h. After washing, 25 μL reading buffer (Mesoscale Discovery) was added to each well, and the resulting electrochemiluminescence (ECL) signal was detected by a Mesoscale Discovery reader. Analysis and quantification were performed using Mesoscale Discovery software. [00349] Representative data are shown in Figs.23A-23D. Co-culturing of Pan T cells with SH-4 cells (CD228 high), Calu-1 cells (CD228 high) or SK-MEL-24 cells (CD228 high) in the presence of the fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, or SEQ ID NOs: 78 and 83) led to a clear, dose-dependent increase in IL-2 secretion compared to hIgG4 isotype control (Figs. 23A-23C). Overall, the increase of IL-2 secretion induced by fusion proteins was significantly higher than the one induced by the reference anti- CD137 antibody (SEQ ID NOs: 26 and 27). No increase of IL-2 secretion was observed with the fusion of IgG4 with the CD137-specific lipocalin mutein included in the fusion proteins (SEQ ID NO: 84). Additionally, co-culturing with CD228-negative RPMI-7951 cells did not increase IL-2 secretion levels with any of the fusion proteins, but with the reference anti-CD137 antibody (SEQ ID NO: 26 and 27) (Fig.23D). [00350] A similar assay was conducted using CD228+ Calu-1 cells or CD228- SK-BR-3 cells and a different but overlapping set of fusion proteins, and Figs.24A-24B show IL-2 secretion in response to stimulation by the fusion proteins and urelumab. These results show that the fusion proteins (e.g., AAF35 (HC)), compared to antibodies, co-stimulate T cell activation in a CD228 target dependent manner and in a dose dependent manner. [00351] The data indicate that the functional activity of the bispecific fusion proteins, measured by their ability to activate T cells or increase IL-2 secretion, is CD228-dependent. [00352] Example 19: Assessment of T cell cytokines produced in cocultures with anti-CD3 scFv engineered CD228-expressing tumor cell line [00353] Experiments were conducted to assess the impact of fusion proteins on cytokine and soluble 4-1BB (sCD137) production from T cells receiving direct T cell receptor stimulation from CD228-expressing tumor cells. Healthy donor PBMC were co-cultured with CD228+ CALU- 1 tumor cells (10:1, in RPMI 10% FCS) engineered to express surface anti-CD3 scFv to elicit direct T cell receptor engagement by tumor cells. A titration of fusion proteins or controls were added in triplicate in a 96-well round bottom plate. At the end of 3-day stimulation at 37 °C in 5% CO ₂ , supernatants were collected and measured for soluble cytokines IL-2, IL-13, and sCD137 Attorney Docket No.01218-0029-00PCT by Luminex ^® multiplex array. Samples were pooled from triplicate test article treatments prior to cytokine measurement. The results are shown in Figs. 25A-25C, which demonstrate that the fusion proteins augment cytokines from direct T cell tumor cell interactions. [00354] Example 20: Assessment of CD8 T cell proliferation in cocultures with an anti-CD3 scFv engineered CD228-expressing tumor cell line [00355] Experiments were conducted to assess the impact of fusion protein bispecifics on cytotoxic T cells receiving direct T cell receptor stimulation from CD228-expressing tumor cells. Healthy donor PBMCs were CFSE-labeled and co-cultured with CD228+ CALU-1 tumor cells (10:1) engineered to express surface anti-CD3 scFv to elicit T cell receptor engagement. A titration of bispecific fusion proteins or controls were added in triplicate in a 96-well round bottom plate. After 72 hours of incubation at 37 °C in 5% CO ₂ , plates were washed, stained with a live dead viability dye and antibodies specific for CD3, CD4, and CD8, and CD228. Plates were evaluated by Attune™ NxT flow cytometer for live CD8 T cell proliferation and CD228+ tumor cells. Fig. 26 shows the proportion of CD8 T cells with diluted CFSE relative to wells without treatment added. Fig. 27 shows the percent of viable tumor cells remaining relative to wells without treatment added, reflecting tumor cell killing. These results show that in the presence of CD228, the fusion proteins (e.g., AAF30 (HC)), compared to antibodies, co-stimulate CD8+ T cell proliferation and killing of tumor cells in a dose dependent manner. [00356] Example 21: Storage stability assessment of fusion proteins [00357] To assess storage stability, exemplary fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, or SEQ ID NOs: 78 and 83) were diluted to concentrations of either 1 mg/ml in PBS or 0.5 mg/ml in 50% human plasma (HPL) or 50% mouse plasma (MPL), respectively. Aliquots of these samples were then either incubated at 37 °C for one week or stored at -20°C. Subsequently, activity of the fusion proteins was determined by a CD137 reporter cell assay as described in Example 14, using CD228-expressing SH-4 cells as target cells. Fusion proteins diluted in PBS, 50% HPL or 50% MPL shortly before the assay served as reference, fusion proteins that were stored at -20 °C at all times served as untreated reference. An anti-CD137 antibody (SEQ ID NOs: 26 and 27) and an IgG4 isotype antibody (SEQ ID NOs: 24 and 25) were used as additional controls. Representative results are shown in Figs.28A-28D. [00358] The fusion molecules of SEQ ID NOs: 80 and 76 and SEQ ID NOs: 75 and 81 (which are both based on the CD228-specific antibody of SEQ ID NOs: 75 and 76) showed excellent stability under all tested conditions (Figs. 28A and 28C, respectively). The fusion molecules of SEQ ID NOs: 82 and 79 and SEQ ID NOs: 78 and 83 (which are both based on the CD228-specific antibody of SEQ ID NOs: 78 and 79) showed excellent stability in PBS and 50% Attorney Docket No.01218-0029-00PCT human plasma, but somewhat reduced stability in 50% mouse plasma, as indicated by an approx. three-fold higher EC ₅₀ value and an approx.30% lower plateau (Figs.28B and 28D, respectively). [00359] Example 22: Pharmacokinetics of fusion proteins in mice [00360] Analyses of the pharmacokinetics of representative fusion proteins (SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, or SEQ ID NOs: 78 and 83) were performed in mice. Male CD-1 NUDE mice approximately 5 weeks of age (3 mice per timepoint; Charles River Laboratories, Research Models and Services, Germany GmbH) were injected into a tail vein with a fusion protein at a dose of 10 mg/kg. The test articles were administered as a bolus using a volume of 5 mL/kg. Plasma samples from the mice were obtained at the timepoints of 5 min, 1 h, 4 h, 8 h, 24 h, 48 h, 4 d, 7 d, 14 d, 21 d, and 28 d. Sufficient whole blood – taken under isoflurane anesthesia – was collected to obtain at least 50 μL Li-Heparin plasma per animal and time. [00361] Fusion protein levels were detected using a Sandwich ELISA detecting the full bispecific construct via the targets CD228 and CD137 in a free PK assay. To this end, huCD228- His (human CD228 with a C-terminal polyhistidine tag) was dissolved in PBS (1 µg/mL) and coated overnight on microtiter plates at 4 °C. The plate was washed after each incubation step with 80 µL PBS supplemented with 0.05% (v/v) Tween 20 five times. The plates were blocked with PBS/BSA/Tween (PBS containing 2% BSA (w/v) and 0.1% (v/v) Tween 20) for 1 h at room temperature and subsequently washed. Plasma samples were diluted in PBS/BSA/Tween/standard CD-1 mouse plasma to 20% plasma concentration, added to the wells, and incubated for 1 h at room temperature. Another wash step followed. Bound agents under study were detected after 1 h incubation with a mixture of biotinylated human CD137 and Streptavidin SULFO-TAG (Mesoscale Discovery) at 1 µg/mL each diluted in PBS containing 2% BSA (w/v) and 0.1% (v/v) Tween 20. After an additional wash step, 35 µL reading buffer was added to each well and the electrochemiluminescence (ECL) signal of every well was read using a Mesoscale Discovery reader. Levels of the parental anti-CD228 antibodies were detected in an analogous manner via the target CD228 and an anti-human IgG Fc antibody (GE Healthcare). Data were transferred to Excel for analysis and quantification. A calibration curve with standard protein dilutions was prepared to back-calculate the concentration in the plasma. Non- compartmental analysis was performed with WinNonLin Software. [00362] Figs.29A-29B show plots of the plasma concentration (29A: mean concentration; 29B: normalized to C _max ) over time for the fusion proteins SEQ ID NOs: 80 and 76, SEQ ID NOs: 82 and 79, SEQ ID NOs: 75 and 81, or SEQ ID NOs: 78 and 83, plotted together with the values obtained for the parental anti-CD228 antibodies (SEQ ID NOs: 74 and 76 and SEQ ID NOs: 77 Attorney Docket No.01218-0029-00PCT and 79, respectively) as a reference. The corresponding terminal half-lives are summarized in Table 7. [00363] The data demonstrate that the fusion proteins have long, antibody-like terminal half-lives in mice. Because the assay employed to determine fusion protein plasma concentrations requires a retained binding activity both towards CD228 and CD137, the result also demonstrates that the bispecific molecules remain intact over the time course of 28 days. [00364] Table 7: Terminal half-lives in mice determined using a non-compartmental analysis [00365] Example 23: Pharmacokinetics of fusion proteins in cynomolgus monkeys [00366] Heavy chain fusion protein bispecifics 30HC and 35HC (SEQ ID NOs: 80 and 76 and SEQ ID NOs: 82 and 79, respectively) were compared for pharmacokinetic differences in cynomolgus monkeys. Animals received a single dose of fusion bispecifics at a 1 mg/kg or 6 mg/kg doses. As shown in Fig.30, bispecific fusion proteins exhibited a similar pharmacokinetic profile with slight differences after day 14, which may be due to an anti-drug antibody response. [00367] Example 24: Assessment of in vivo activity of fusion proteins in a humanized xenograft model [00368] To assess the activity of bispecific fusion proteins in vivo, humanized xenograft models using CD228+ melanoma cell lines CALU-1 (ATCC) and H3677 (SGEN, internal) were performed. Tumor cells were implanted into immune-deficient NSG mice in 25% Matrigel ^® (Corning) and monitored until tumors reached an average volume of 100 mm ³ , at which point healthy donor peripheral blood mononuclear cells (PBMC) were adoptively transferred via tail vein injection to provide a source of human T cells. Mice were subsequently dosed at 5-day intervals with equimolar amounts of bispecific fusion proteins (10 mg/kg), antibodies (non-bispecific antibodies, anti-PD-1, and/or anti-4-1BB), or controls (8 mg/kg) and monitored for tumor growth. Attorney Docket No.01218-0029-00PCT Growth curves across the Calu-1 study are shown in Fig.31A, and final tumor volumes at the end study, Day 80, are shown in Fig.31B. Growth curves across the H3677 study are shown in Fig.31C, and final tumor volumes at the end study, Day 22, are shown in Fig.31D. Statistics provided are the result of Tukey’s analysis. These results show that the fusion proteins (e.g., AAF30 (HC)), compared to antibodies, reduce CD228+ tumor cell growth in vivo. [00369] To evaluate intratumoral PD effects from bispecific-fusion-treated animals in the humanized CD228+ Calu-1 xenograft model, tumors were collected and processed into single cell suspensions for staining of tumor and immune cell populations and analysis by flow cytometry. Fig. 32A shows the CD8+ T cell to tumor cell ratio as determined by counts of CD8+ T cells divided by counts of CD228+ non-immune cells. Fig.32B shows changes in CD8+/CD4+ T cell ratio. Fig.32C shows the proportion of CD8+ T cells in active degranulation, as determined by surface expression of CD107a, at the end of the study. Dashed lines in figures indicate the CD8+/CD4+ T cell ratio (Fig.32B) or the percentage of CD107a-expressing CD8 T cells (Fig. 32C) in resting PBMC (i.e., non-tumor model) samples from the same donor. These data demonstrate enhanced expansion and activation of cytotoxic CD8+ T cells upon treatment with bispecific fusion proteins that are superior to controls. Statistics provided are the result of Tukey’s analysis. [00370] To evaluate intratumoral PD effects from bispecific-fusion-treated animals in a separate humanized CD228+ Calu-1 xenograft model, tumors were collected and processed into single cell suspensions for staining of tumor and immune cell populations and analysis by flow cytometry. Fig.33A shows the CD8+ T cell to CD4+ T cell ratio. Fig. 33B shows changes in intracellular TCF1 expression. These data demonstrate enhanced expansion and differentiation of cytotoxic CD8+ T cells into cells with more stem-like properties, indicative of greater anti-tumor potential, as identified by expression of the transcription factor TCF1 following treatment with bispecific fusion proteins targeting 4-1BB. Statistics provided are the result of Tukey’s analysis. [00371] Example 25: Assessment of CD8 T cell proliferation in cocultures with anti- CD3 scFv engineered CD228-expressing tumor cell lines [00372] As in Example 20, experiments were conducted to assess the impact of fusion protein bispecifics on cytotoxic T cells receiving direct T cell receptor stimulation from CD228- expressing tumor cells. Healthy donor PBMCs were CFSE-labeled and co-cultured with CD228+ A2058 and SK-MEL-5 melanoma cells (10:1) engineered to express surface anti-CD3 scFv to elicit T cell receptor engagement. A titration of bispecific fusion proteins or controls were added in triplicate in a 96-well round bottom plate. After 72 hours of incubation at 37 °C in 5% CO ₂ , plates were washed, stained with antibodies specific for CD3, CD4, and CD8. Plates were evaluated by Attune™ NxT flow cytometer for live CD8 T cell proliferation. Figs.34A and 34B Attorney Docket No.01218-0029-00PCT show the proportion of divided CD8 T cells relative to wells without treatment added. These results show that in the presence of CD228, the fusion proteins (e.g., AAF30 (HC)), compared to antibodies, control fusion proteins, and urelumab, co-stimulate CD8+ T cell proliferation in a dose dependent manner. [00373] Example 26: Assessment of CD8 T cell proliferation and mitochondrial function in cocultures with anti-CD3 scFv engineered CD228-expressing tumor cell lines [00374] As in Example 20, experiments were conducted to assess the impact of fusion protein bispecifics on cytotoxic T cells receiving direct T cell receptor stimulation from CD228- expressing tumor cells. Healthy donor PBMCs were co-cultured with CD228+ CALU-1 lung cancer cells (2.5:1) engineered to express surface anti-CD3 scFv to elicit T cell receptor engagement. A titration of bispecific fusion proteins or controls were added in quadruplicate in a 96-well round bottom plate. After 6 days of incubation at 37 °C in 5% CO ₂ , plates were washed and stained with antibodies specific for CD2, CD4, and CD8. Plates were evaluated by Attune™ NxT flow cytometer for live CD8 T cell expansion. Fig.35A shows CD8 T cell count in wells with treatment added, relative to wells without treatment added. For measuring CD8 T cell mitochondrial content, cells were stained with 75 nM MitoSpy™ green FM for 20 minutes at 37 °C in 5% CO ₂ . Fig.35B shows CD8 T cell mitochondrial content, represented by mean fluorescence intensity (MFI) of MitoSpy™ green FM staining. Fig.35C shows the percentages of CD8 T cell with depolarized mitochondria, as determined by staining with the polarization-dependent dye MitoSpy™ orange CMTMros (25 nM, 20 minutes at 37 °C in 5% CO ₂ ) and MitoSpy™ green FM (75 nM, 20 minutes at 37 °C in 5% CO ₂ ). These results show that in the presence of CD228, the fusion proteins (e.g., AAF30 (HC)), compared to non-bispecific CD228-binding antibodies or urelumab (20H4.9), co-stimulate CD8+ T cell proliferation in a dose-dependent manner and measurably improve the metabolic fitness of activated cytotoxic T cells. [00375] Example 27: Preparation and characterization of a T cell exhaustion model [00376] To create a model of functionally exhausted cytotoxic T cells receiving direct T cell receptor stimulation from CD228-expressing tumor cells, CALU-1 lung cancer cells were engineered to express a membrane anti-CD3 scFv fragment, and healthy donor PBMCs were serially passaged (P0-P4) on the CALU-1 cell lawns, driving progressive T cell dysfunction. Fig.36A shows a schematic of creating such functionally exhausted cytotoxic T cells. The cells recovered and banked from the serial passages were analyzed for their ability to proliferate, secrete interferon gamma (IFN-γ), and kill tumor cells, and Fig.36B shows the percentages of divided cells (left), the levels of IFN-γ (center), and the numbers of live tumor cells remaining (right) following restimulation of T cells from different passages in cocultures with anti-CD3 scFv Attorney Docket No.01218-0029-00PCT engineered CALU-1 cells. Passage 4 (P4) T cells exhibited a significant reduction in the capacity to divide, secrete cytokines, and kill tumor cells. [00377] CD8+ T cells from P4 were analyzed by single-cell RNA sequencing (RNAseq) and clustered into two states using the Uniform Manifold Approximation and Projection (UMAP) method. The two states resemble progenitor (P4 Tpex_like) and terminally exhausted (P4 Texterm_like) CD8+ T cells. Average gene expression per state was scaled for each dataset, and then shown in Fig.37A, which is a heatmap for the P4 cells or exhausted CD8+ T cells from a pan-cancer tumor-infiltrating lymphocytes (TILs) atlas, as described in Zheng et al., Pan-cancer single-cell landscape of tumor-infiltrating T cells, Science 374:6574 (2021). Additionally, Fig.37B shows expression levels of TCF7 (TCF1) (left), HAVCR2 (TIM-3) (center), and TNFRSF9 (CD137) (right) in Tpex_like and P4 Texterm_like cells. The similarity of Tpex_like and Texterm_like cells to bona fide TIL CD8+ T cells was assessed by hierarchical clustering based on marker genes of exhausted T cell states, and the results demonstrated that Tpex_like and P4 Texterm_like cells share significant overlap of gene signatures with bona fide exhausted TILs. [00378] Example 28: Assessment of functional reinvigoration of exhausted CD8 T cells in cocultures with anti-CD3 scFv engineered CD228-expressing tumor cell lines [00379] Experiments were conducted to assess the impact of fusion protein bispecifics on functionally exhausted cytotoxic T cells. Exhausted T cells from Example 27 were collected, CFSE-labeled, and again co-cultured with CD228+ anti-CD3 scFv CALU-1 tumor cells (1:1). A titration of bispecific fusion proteins, a costimulatory antibody against CD28 (clone CD28.2, BioLegend, San Diego, CA, cat. # 302901/302902), anti-PD-1 antibody nivolumab, and anti-4- 1BB agonist antibody 20H4.9 were added in triplicate in a 96-well round bottom plate. After 96 hours of incubation at 37 °C in 5% CO ₂ , plates were washed, and stained with antibodies specific for CD3, CD4, and CD8. Plates were evaluated by Attune™ NxT flow cytometer for live CD8 T cell proliferation. Fig.38 shows the fold increase in CFSE-low (divided) CD8 T cell counts in wells with treatment added, relative to wells without treatment added (dashed line = 1). These results show that agonists of 4-1BB, the fusion protein (e.g., AAF30 (HC)) and 20H4.9, reinvigorate functionally exhausted CD8+ T cell proliferation in a dose-dependent manner, while anti-PD-1 antibody nivolumab and an agonist antibody against CD28 do not increase proliferation alone. In the presence of CD228-expressing CALU-1 tumor cells, AAF30 (HC) drove increased costimulation compared to 20H4.9, and the proliferative effect of AAF30 (HC) and 20H4.9 on exhausted cytotoxic T cells was further amplified by blocking the PD-1/PD-L1 pathway with anti- PD-1 antibody nivolumab, demonstrating complementarity for these mechanisms of action and supporting clinical combination of the fusion protein bispecifics with agents that block the PD- 1/PD-L1 pathway, such as anti-PD-1 or anti-PD-L1 antibodies. Attorney Docket No.01218-0029-00PCT [00380] Embodiments illustratively described herein may suitably be practiced in the absence of any element or elements, limitation or limitations, not specifically disclosed herein. Thus, for example, the terms "comprising," "including," "containing," etc. shall be read expansively and without limitation. Additionally, the terms and expressions employed herein have been used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present embodiments have been specifically disclosed by preferred embodiments and optional features, equivalents, modifications and variations thereof may be resorted to by those skilled in the art, and that such equivalents, modifications and variations are considered to be within the scope of this invention. All patents, patent applications, textbooks and (peer-reviewed) publications described herein are hereby incorporated by reference in their entirety to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated herein by reference. Furthermore, where a definition or use of a term in a reference, which is incorporated by reference herein is inconsistent or contrary to the definition of that term provided herein, the definition of that term provided herein applies and the definition of that term in the reference does not apply. Each of the narrower species and subgeneric groupings falling within the generic disclosure also forms part of the invention. This includes the generic description of the invention with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein. In addition, where features are described in terms of Markush groups, those skilled in the art will recognize that the disclosure is also thereby described in terms of any individual member or subgroup of members of the Markush group. Further embodiments will become apparent from the following claims.

Attorney Docket No.01218-0029-00PCT VII. NON-PATENT REFERENCES 1. ROSE, T. M., PLOWMAN, G. D., TEPLOW, D. B., DREYER,  W. J., HELLSTRÖM, K. E. & BROWN, J. P.  1986.  Primary structure of the human melanoma‐asso ciated antigen p97 (melanotransferrin)  deduced from the mRNA sequence. Proc Natl Acad Sci  U S A, 83(5), 1261‐5.  2. LI, S. Y. & LIU, Y. 2013. Immunotherapy of mela noma with the immune costimulatory monoclonal  antibodies targeting CD137. Clin Pharmacol, 5, 47‐53 .  3. SNELL, L. M., LIN, G. H., MCPHERSON, A. J., MORAES,  T. J. & WATTS, T. H. 2011. T‐cell intrinsic� �effects  of GITR and 4‐1BB during viral infection and cance r immunotherapy. Immunol Rev, 244, 197‐217.  4. WYZGOL, A., MULLER, N., FICK, A., MUNKEL, S., GRIGOL EIT, G. U., PFIZENMAIER, K. & WAJANT, H.  2009. Trimer stabilization, oligomerization, and antibo dy‐mediated cell surface immobilization  improve the activity of soluble trimers of CD27L, CD 40L, 41BBL, and glucocorticoid‐induced TNF  receptor ligand. J Immunol, 183, 1851‐61.  5. YAO, S., ZHU, Y. & CHEN, L. 2013. Advances in  targeting cell surface signalling molecules for immune   modulation. Nat Rev Drug Discov, 12, 130‐46.  6. MELERO, I., BACH, N., HELLSTROM, K. E., ARUFFO, A.,� �MITTLER, R. S. & CHEN, L. 1998. Amplification  of tumor immunity by gene transfer of the co‐stimu latory 4‐1BB ligand: synergy with the CD28 co‐ stimulatory pathway. Eur J Immunol, 28, 1116‐21.  7. YANG, Y., YANG, S., YE, Z., JAFFAR, J., ZHOU, Y.,  CUTTER, E., LIEBER, A., HELLSTROM, I. &  HELLSTROM, K. E. 2007. Tumor cells expressing anti‐ CD137 scFv induce a tumor‐destructive  environment. Cancer Res, 67, 2339‐44.  8. ZHANG, H., KNUTSON, K. L., HELLSTROM, K. E., DISIS,� �M. L. & HELLSTROM, I. 2006. Antitumor  efficacy of CD137 ligation is maximized by the use  of a CD137 single‐chain Fv‐expressing whole‐cell� � tumor vaccine compared with CD137‐specific monoclonal  antibody infusion. Mol Cancer Ther, 5,  149‐55.  9. YE, Z., HELLSTROM, I., HAYDEN‐LEDBETTER, M., DAHLIN,  A., LEDBETTER, J. A. & HELLSTROM, K. E.  2002. Gene therapy for cancer using single‐chain Fv  fragments specific for 4‐1BB. Nat Med, 8, 343‐8 .  10. MARTINET, O., DIVINO, C. M., ZANG, Y., GAN, Y., MAN DELI, J., THUNG, S., PAN, P. Y. & CHEN, S. H.� � 2002. T cell activation with systemic agonistic antib ody versus local 4‐1BB ligand gene delivery  combined with interleukin‐12 eradicate liver metastas es of breast cancer. Gene Ther, 9, 786‐92.  11. YE, Q., SONG, D. G., POUSSIN, M., YAMAMOTO, T., BES T, A., LI, C., COUKOS, G. & POWELL, D. J., JR.   2014. CD137 accurately identifies and enriches for na turally occurring tumor‐reactive T cells in  tumor. Clin Cancer Res, 20, 44‐55.  12. CHACON, J. A., WU, R. C., SUKHUMALCHANDRA, P., MOLLD REM, J. J., SARNAIK, A., PILON‐THOMAS,  S., WEBER, J., HWU, P. & RADVANYI, L. 2013. Co� ��stimulation through 4‐1BB/CD137 improves the  expansion and function of CD8(+) melanoma tumor‐infi ltrating lymphocytes for adoptive T‐cell  therapy. PLoS One, 8, e60031.  13. FISHER, T. S., KAMPERSCHROER, C., OLIPHANT, T., LOVE,  V. A., LIRA, P. D., DOYONNAS, R.,  BERGQVIST, S., BAXI, S. M., ROHNER, A., SHEN, A. C. , HUANG, C., SOKOLOWSKI, S. A. & SHARP, L. L.  2012. Targeting of 4‐1BB by monoclonal antibody PF� ��05082566 enhances T‐cell function and  promotes anti‐tumor activity. Cancer Immunol Immunoth er, 61, 1721‐33.  14. SKERRA, A. 2000. Lipocalins as a scaffold. Biochim B iophys Acta, 1482, 337‐50.  15. FLOWER, D. R., NORTH, A. C. & SANSOM, C. E. 20 00. The lipocalin protein family: structural and  sequence overview. Biochim Biophys Acta, 1482, 9‐24.   16. FLOWER, D. R. 1996. The lipocalin protein family: st ructure and function. Biochem J, 318 ( Pt 1), 1‐ 14.  17. ALTSCHUL, S. F., MADDEN, T. L., SCHAFFER, A. A., ZH ANG, J., ZHANG, Z., MILLER, W. & LIPMAN, D. J.� � 1997. Gapped BLAST and PSI‐BLAST: a new generation� �of protein database search programs. Nucleic  Acids Res, 25, 3389‐402.  18. ALTSCHUL, S. F., GISH, W., MILLER, W., MYERS, E. W.  & LIPMAN, D. J. 1990. Basic local alignment  search tool. J Mol Biol, 215, 403‐10.  19. SMITH, T. F. & WATERMAN, M. S. 1981. Identificat ion of common molecular subsequences. J Mol  Biol, 147, 195‐7.  Attorney Docket No.01218-0029-00PCT WARD, E. S., GUSSOW, D., GRIFFITHS, A. D., JONES, P . T. & WINTER, G. 1989. Binding activities of a   repertoire of single immunoglobulin variable domains s ecreted from Escherichia coli. Nature, 341,  544‐6.  HOLLIGER, P., PROSPERO, T. & WINTER, G. 1993. "D iabodies": small bivalent and bispecific antibody  fragments. Proc Natl Acad Sci U S A, 90, 6444‐8.� � JOHNSON, G. & WU, T. T. 2000. Kabat database an d its applications: 30 years after the first  variability plot. Nucleic Acids Res, 28, 214‐8.  EHRENMANN, F., KAAS, Q. & LEFRANC, M. P. 2010.  IMGT/3Dstructure‐DB and  IMGT/DomainGapAlign: a database and a tool for immuno globulins or antibodies, T cell receptors,  MHC, IgSF and MhcSF. Nucleic Acids Res, 38, D301‐7 .  BULLIARD, Y., JOLICOEUR, R., ZHANG, J., DRANOFF, G.,� �WILSON, N. S. & BROGDON, J. L. 2014. OX40  engagement depletes intratumoral Tregs via activating  FcgammaRs, leading to antitumor efficacy.  Immunol Cell Biol, 92, 475‐80.  BULLIARD, Y., JOLICOEUR, R., WINDMAN, M., RUE, S. M. , ETTENBERG, S., KNEE, D. A., WILSON, N. S.,  DRANOFF, G. & BROGDON, J. L. 2013. Activating Fc  gamma receptors contribute to the antitumor  activities of immunoregulatory receptor‐targeting anti bodies. J Exp Med, 210, 1685‐93.  SILVA, J. P., VETTERLEIN, O., JOSE, J., PETERS, S.  & KIRBY, H. 2015. The S228P mutation prevents in   vivo and in vitro IgG4 Fab‐arm exchange as demonst rated using a combination of novel quantitative  immunoassays and physiological matrix preparation. J B iol Chem, 290, 5462‐9.  GLAESNER, W., VICK, A. M., MILLICAN, R., ELLIS, B.,� �TSCHANG, S. H., TIAN, Y., BOKVIST, K., BRENNER,  M., KOESTER, A., PORKSEN, N., ETGEN, G. & BUMOL,  T. 2010. Engineering and characterization of  the long‐acting glucagon‐like peptide‐1 analogue  LY2189265, an Fc fusion protein. Diabetes Metab  Res Rev, 26, 287‐96.  DALL'ACQUA, W. F., KIENER, P. A. & WU, H. 2006.  Properties of human IgG1s engineered for  enhanced binding to the neonatal Fc receptor (FcRn).� �J Biol Chem, 281, 23514‐24.  ZALEVSKY, J., CHAMBERLAIN, A. K., HORTON, H. M., KAR KI, S., LEUNG, I. W., SPROULE, T. J., LAZAR,  G. A., ROOPENIAN, D. C. & DESJARLAIS, J. R. 201 0. Enhanced antibody half‐life improves in vivo  activity. Nat Biotechnol, 28, 157‐9.  SHIELDS, R. L., NAMENUK, A. K., HONG, K., MENG, Y.� �G., RAE, J., BRIGGS, J., XIE, D., LAI, J., STADLEN ,  A., LI, B., FOX, J. A. & PRESTA, L. G. 2001.  High resolution mapping of the binding site on human  IgG1  for Fc gamma RI, Fc gamma RII, Fc gamma RIII, and� �FcRn and design of IgG1 variants with improved  binding to the Fc gamma R. J Biol Chem, 276, 6591� ��604.  ALTSHULER, E. P., SEREBRYANAYA, D. V. & KATRUKHA,  A. G. 2010. Generation of recombinant  antibodies and means for increasing their affinity. B iochemistry (Mosc), 75, 1584‐605.  HARLOW, E. & LANE, D. 1999. Using antibodies :  a laboratory manual, Cold Spring Harbor, N.Y., Cold  Spring Harbor Laboratory Press.  HARLOW, E. & LANE, D. 1988. Antibodies : a labo ratory manual, Cold Spring Harbor, NY, Cold Spring  Harbor Laboratory.  LI, J., SAI, T., BERGER, M., CHAO, Q., DAVIDSON, D. , DESHMUKH, G., DROZDOWSKI, B., EBEL, W.,  HARLEY, S., HENRY, M., JACOB, S., KLINE, B., LAZO,  E., ROTELLA, F., ROUTHIER, E., RUDOLPH, K.,  SAGE, J., SIMON, P., YAO, J., ZHOU, Y., KAVURU, M.,  BONFIELD, T., THOMASSEN, M. J., SASS, P. M.,  NICOLAIDES, N. C. & GRASSO, L. 2006. Human antib odies for immunotherapy development  generated via a human B cell hybridoma technology. P roc Natl Acad Sci U S A, 103, 3557‐62.  KOZBOR, D. & RODER, J. C. 1983. The production  of monoclonal antibodies from human  lymphocytes. Immunol Today, 4, 72‐9.  COLE, S. P., CAMPLING, B. G., LOUWMAN, I. H., KOZBO R, D. & RODER, J. C. 1984. A strategy for the� � production of human monoclonal antibodies reactive wit h lung tumor cell lines. Cancer Res, 44,  2750‐3.  HOLLIGER, P. & HUDSON, P. J. 2005. Engineered an tibody fragments and the rise of single domains.  Nat Biotechnol, 23, 1126‐36.  PERVAIZ, S. & BREW, K. 1987. Homology and struct ure‐function correlations between alpha 1‐acid  glycoprotein and serum retinol‐binding protein and i ts relatives. FASEB J, 1, 209‐14.  Attorney Docket No.01218-0029-00PCT SAMBROOK, J. & RUSSELL, D. W. 2001. Molecular cl oning : a laboratory manual, Cold Spring Harbor,  N.Y., Cold Spring Harbor Laboratory Press.  FLOWER, D. R. 2000. Beyond the superfamily: the lipo calin receptors. Biochim Biophys Acta, 1482,  327‐36.  BREUSTEDT, D. A., KORNDORFER, I. P., REDL, B. &� �SKERRA, A. 2005. The 1.8‐A crystal structure of  human tear lipocalin reveals an extended branched cav ity with capacity for multiple ligands. J Biol  Chem, 280, 484‐93.  SCHMIDT, T. G., KOEPKE, J., FRANK, R. & SKERRA,� �A. 1996. Molecular interaction between the Strep‐ tag affinity peptide and its cognate target, streptav idin. J Mol Biol, 255, 753‐66.  VAJO, Z. & DUCKWORTH, W. C. 2000. Genetically en gineered insulin analogs: diabetes in the new  millenium. Pharmacol Rev, 52, 1‐9.  FUERTGES, F. & ABUCHOWSKI, A. 1990. The clinical� �efficacy of poly(ethylene glycol)‐modified  proteins. Journal of Controlled Release, 11, 139‐148 .  DENNIS, M. S., ZHANG, M., MENG, Y. G., KADKHODAYAN,� �M., KIRCHHOFER, D., COMBS, D. &  DAMICO, L. A. 2002. Albumin binding as a general st rategy for improving the pharmacokinetics of  proteins. J Biol Chem, 277, 35035‐43.  KONIG, T. & SKERRA, A. 1998. Use of an albumin� ��binding domain for the selective immobilisation of  recombinant capture antibody fragments on ELISA plates . J Immunol Methods, 218, 73‐83.  OSBORN, B. L., OLSEN, H. S., NARDELLI, B., MURRAY,  J. H., ZHOU, J. X., GARCIA, A., MOODY, G.,  ZARITSKAYA, L. S. & SUNG, C. 2002. Pharmacokineti c and pharmacodynamic studies of a human  serum albumin‐interferon‐alpha fusion protein in cy nomolgus monkeys. J Pharmacol Exp Ther, 303,  540‐8.  LOWMAN, H. B. 1997. Bacteriophage display and discove ry of peptide leads for drug development.  Annu Rev Biophys Biomol Struct, 26, 401‐24.  RODI, D. J. & MAKOWSKI, L. 1999. Phage‐display� �technology‐‐finding a needle in a vast molecular� � haystack. Curr Opin Biotechnol, 10, 87‐93.  VENTURI, M., SEIFERT, C. & HUNTE, C. 2002. High� �level production of functional antibody Fab  fragments in an oxidizing bacterial cytoplasm. J Mol� �Biol, 315, 1‐8.  BRUCKDORFER, T., MARDER, O. & ALBERICIO, F. 2004.  From production of peptides in milligram  amounts for research to multi‐tons quantities for d rugs of the future. Curr Pharm Biotechnol, 5, 29‐ 43. 

Attorney Docket No.01218-0029-00PCT VIII. TABLE OF SEQUENCES [00381] Table of Exemplary Sequences: Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT Attorney Docket No.01218-0029-00PCT