Novel TNFR2-binding molecule

Abstract

The present disclosure provides a binding molecule that specifically binds to TNFR2 but not to TNFR1.Preferably, the binding molecule is an FcγR-independent agonist of TNFR2, and more preferably has at least two valency with respect to the binding of TNFR2.The binding molecule has a variety of uses, including diagnostics and therapeutics.The ability of the binding molecule to modulate immune response means that they are particularly suitable for the treatment of autoimmune diseases or inflammatory conditions.

Description

[Technical field]

[0001] The present invention relates to a binding molecule specific for TNFR2 (also known as tumor necrosis factor receptor 2, TNFRSF1B and CD120B). The binding molecule typically binds to TNFR2 and can induce TNFR2 signaling independent of FcγR (Fc gamma receptor) binding. The binding molecule can be used to target cells expressing TNFR2. The present invention also relates to the use of the binding molecule in methods of treatment and diagnosis, particularly in methods of treating disorders such as inflammatory disorders and autoimmune diseases. [Background technology]

[0002] TNFα is a cytokine that plays a wide range of roles in inflammation, differentiation, and tissue homeostasis, and exists as both soluble and transmembrane forms. It is an important inflammatory mediator that is targeted by TNF-blocking reagents in the clinic to suppress unwanted immune responses in inflammatory and autoimmune conditions. TNFR1 and TNFR2 are distinct receptors for tumor necrosis factor-alpha (TNFα), and signaling through each receptor induces distinct effects. Signaling through TNFR1 induces strong and diverse pro-inflammatory effects, whereas signaling through TNFR2 has a wide range of effects ranging from immune stimulation to suppression of immune responses to tissue regeneration. The two receptors also differ from each other with respect to the cells in which they are expressed. TNFR1 is ubiquitously expressed, whereas TNFR2 is found in a more restricted population of cells, particularly a subset of immune cells, endothelial cells, and neurons. Immune cells that express TNFR2 include immune suppressive cell types, such as regulatory T cells (Treg), regulatory B cells (Breg), and MDSCs (myeloid-derived suppressor cells), as well as other cell types, such as CD8+ T cells.

[0003] TNFR1 and TNFR2 have an extracellular domain containing four characteristic cysteine-rich domains (CRDs), a transmembrane segment, and an intracellular domain. Unlike TNFR1, TNFR2 does not contain a cytoplasmic death domain, but may still play a role in apoptosis by controlling cell death regulators such as TRAF2 and cIAP. TNFR2 molecules can associate with each other to form TNFR2 dimeric and trimeric complexes, which can bind TNFα with high affinity, thereby resulting in the formation of oligomers of trimeric TNF-TNFR2 complexes capable of intracellular signaling. Both TNFR1 and TNFR2 can be expressed on the surface of the same cell type, but the two do not form complexes with each other. TNFR2 can bind both soluble and membrane-bound TNFα. Binding of membrane TNFα can effectively activate TNFR2, but with soluble TNFα, oligomerization of the ligand-bound TNFR2 trimers into hexamers or higher order complexes is inappropriate, thus failing to induce effective TNFR2 activation. Like TNFα, lymphotoxin-alpha (LTα) is also thought to act as a ligand for TNFR2.

[0004] When TNFR2 is activated by membrane TNFα, the intracellular domain of the receptor is thought to recruit pre-existing cytoplasmic TRAF2 trimers, which leads to the recruitment of E3 ligases cIAP1 and cIAP2 to the TNFR2 complex, which results in classical NFκB signaling. TRAF2 recruitment is also associated with activating non-classical NFκB signaling pathways. Binding of membrane TNFα to TNFR2 is also thought to trigger signaling through the PI3K / Akt pathway, which maintains survival, enhances proliferation, and promotes other functions such as cell adhesion and migration. TNFR2 knockout mice are more susceptible to inflammatory diseases, but they have Treg cells and are less responsive. Naturally occurring polymorphisms of TNFR2 also exist and are associated with a wide variety of autoimmune diseases.

[0005] The expression and function of TNFR2 on Treg cells makes it a desirable target for modulating the immune system, especially since TNFR2 is found on a more restricted subset of cells than TNFR1.The ability to stimulate Treg cells provides a way to control undesirable immune responses in a wide range of conditions where the ability to suppress immune responses can be useful, such as allergy, autoimmunity, transplantation, graft-versus-host disease (GVHD) and infectious diseases.Similar to Treg cells, other cell types that express TNFR2 can be targeted to produce desired functions.Therefore, there is currently a need for agents that target and activate TNFR2. Summary of the Invention

[0006] The present invention provides a binding molecule that can bind to the extracellular domain of TNFR2. In a preferred embodiment, the binding molecule of the present invention can bind to TNFR2 and induce signal transduction through the receptor, thus acting as an agonist. In a further particularly preferred embodiment, the binding molecule can act as an agonist of TNFR2 independent of FcγR binding. In a particularly preferred embodiment, the binding molecule of the present invention is an antibody or an antibody fragment. In a further particularly preferred embodiment of the present invention, the binding molecule is selective in that it specifically binds to TNFR2 but does not bind to TNFR1.

[0007] Binding molecules can have a variety of uses, from diagnostic applications to therapeutic uses.Because binding molecules have the ability to modulate the immune system, they are particularly useful for treating autoimmune and inflammatory conditions.Binding molecules may be used to downregulate the immune system through their effect on Treg cells, and are therefore particularly useful in treating such conditions, including graft-versus-host disease (GvHD).

[0008] Thus, in one embodiment, the present invention provides a binding molecule that specifically binds to TNFR2 but not to TNFR1, which is an FcγR-independent agonist of TNFR2 and has at least two valencies with respect to the binding of TNFR2. In a particularly preferred embodiment, the binding molecule has at least four valencies with respect to TNFR2. In a further particularly preferred embodiment, the binding molecule has four valencies with respect to TNFR2. In a particularly preferred embodiment, the binding molecule has at least six valencies with respect to TNFR2. In a further particularly preferred embodiment, the binding molecule has six valencies with respect to TNFR2.

[0009] In another particularly preferred embodiment, the binding molecule of the invention comprises at least one binding domain of TNFR2 which is a VHH binding domain specific for TNFR2. In another particularly preferred embodiment, all binding domains specific for TNFR2 are VHH binding domains specific for TNFR2. In a further particularly preferred embodiment, the binding molecule is a hexavalent antibody comprising six VHH binding domains specific for TNFR2.

[0010] The present invention also provides the following: A binding molecule of the invention for use as a medicament. A binding molecule of the invention for use in a method for treating or preventing an autoimmune or inflammatory disorder. A method of treating or preventing an autoimmune or inflammatory disorder comprising administering a binding molecule of the invention to a subject suffering from or at risk of the disorder. Use of a binding molecule of the invention for the manufacture of a medicament for use in a method of treating or preventing an autoimmune or inflammatory disorder. A method of stimulating cell proliferation comprising contacting a target cell expressing TNFR2 with a binding molecule of the invention. A pharmaceutical composition comprising a binding molecule of the invention and a pharma- ceutically acceptable carrier. A method of detecting TNFR2 comprising contacting a test sample with a binding molecule of the invention and detecting binding of the binding molecule to TNFR2, if present. A diagnostic method comprising administering a binding molecule of the invention to a subject and detecting binding of the binding molecule to TNFR2, if present. A binding molecule of the invention for use in a method of diagnosis, the method comprising administering a binding molecule of the invention to a subject and detecting binding of the binding molecule to TNFR2.

[0011] Sequence Listing This application contains a Sequence Listing in electronic form which forms part of the filed application. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 shows the nomenclature and scheme of the domain structure of exemplary anti-TNFR2 antibody variants with N-terminally organized TNFR2 binding sites (Fab or scFv domains). CI=clonal identifier of parent anti-TNFR2 antibody, e.g., mouse anti-human TNFR2 antibodies C4, C19, C9, C40, etc.; N297A refers to the corresponding mutation in the heavy chain of human IgG1 that reduces FcγR interactions; N-RGY refers to the N297A mutation plus a mutation that promotes IgG1 hexamer formation. [Diagram 2] Figure 1 shows Western blot analysis of variants of anti-TNFR2 antibody C4 with an N-terminal unidirectional TNFR2 binding site. Flag-tagged variants (200 ng) of the indicated C4 constructs were analyzed by Western blotting with anti-human IgG and anti-Flag antibodies. [Diagram 3]Figure 1 shows TNFR2-mediated IL8 production in HT1080-Bcl2-TNFR2 cells after treatment with C4 variants with unidirectionally organized N-terminal TNFR2 binding sites. HT1080-Bcl2-TNFR2 cells stably expressing human TNFR2 were stimulated with supernatants containing the indicated concentrations of various C4 constructs. The following day, supernatants were harvested and analyzed for their IL8 content by ELISA. The data shown are from parallel experiments and are shown in three diagrams for better resolution. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. [Figure 4] Figure 1 shows TNFR2-mediated cell death in Kym-1 cells after treatment with C4 variants with unidirectional N-terminal TNFR2 binding sites. Kym-1 cells were stimulated with supernatants containing various C4 constructs at the indicated concentrations. The next day, cell viability was determined using crystal violet staining. The highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) (oligo TNF80) served as a positive control for TNFR2 engagement. [Diagram 5] TNFR2-mediated p100 processing in Kym-1 cells after treatment with C4 variants with unidirectional N-terminal TNFR2 binding sites. Kym-1 cells were stimulated overnight with supernatants containing various C4 variants at the indicated concentrations. Treatment was performed in the presence of 20 μM ZVAD to prevent TNFR2-induced cell death. A nonameric TNFR2-specific mutant of mouse TNF (oligo TNF80), which potently activates human and mouse TNFR2, was used as a positive control. [Figure 6]Figure 1 shows TNFR2-mediated IL8 production in HT1080-Bcl2-TNFR2 cells after treatment with C19, C27, and C40 variants with unidirectional N-terminal TNFR2 binding sites. HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the various constructs at the indicated concentrations. The next day, supernatants were harvested and analyzed for their IL8 content by ELISA. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. Note: even constructs derived from the low affinity antibody C27 showed significant agonist activity (lower panel). [Figure 7] FIG. 1 shows the nomenclature and scheme of domain structure of anti-TNFR2 antibody variants with N- and C-terminally organized TNFR2 binding sites (Fab or scFv domains). CI = clone identifier, e.g., C4, C19, etc.; N297A refers to the corresponding mutation in the heavy chain of human IgG1 that reduces FcγR binding. [Figure 8] Figure 1 shows Western blot analysis of C4 variants with N- and C-terminal fused TNFR2 binding sites. Flag-tagged variants of the indicated C4 constructs (200 ng) were analyzed by Western blotting with anti-human IgG and anti-Flag antibodies. Note that the heavy chain of C4-IgG1(N297A)-HC:scFvC4 does not contain a Flag tag. [Figure 9]Figure 1 shows TNFR2-mediated IL8 production in HT1080-Bcl2-TNFR2 cells after treatment with C4 constructs with N- and C-terminal TNFR2 binding sites. HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the various C4 constructs at the indicated concentrations. The next day, supernatants were harvested and analyzed for their IL8 content by ELISA. The data shown are from parallel experiments and are shown in three diagrams for better resolution. The maximum response induced by the positive control (oligo TNF80; highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R)) is shown by the dotted line. The corresponding constructs with only the N-terminal binding site were analyzed in parallel for comparison. [Figure 10] Figure 1 shows TNFR2-mediated cell death in Kym-1 cells after treatment with C4 constructs with N- and C-terminally organized TNFR2 binding sites. Kym-1 cells were stimulated with supernatants from HEK293 transfected cells containing various C4 constructs at the approximate concentrations indicated. The following day, cell viability was determined using an MTT assay. Highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) was used as a positive control. [Figure 11] Figure 1 shows the binding characteristics of TNFR2-specific antibodies. (A) Equilibrium binding of the indicated anti-TNFR2 antibody-GpL fusion proteins to cells expressing human TNFR2 at 37°C. (B) HeLa-TNFR2 cells were pretreated with the indicated amounts of anti-TNFR2 antibodies or left untreated. 5ng / ml GpL-TNF was added and binding was measured after 1 hour of incubation. Non-specific binding values ​​were obtained by analyzing the binding of the various constructs to non-transfected HeLa cells. [Figure 12] Figure 3. αTNFR2 antibody competition. HeLa-TNFR2 cells were pretreated with 3-10 μg / ml of the indicated anti-TNFR2 antibodies or left untreated. 50 ng / ml of various αTNFR2-IgG1(N297A)-LC:GpL variants were added and binding measured after 1 hour. [Figure 13]Figure 2 shows that tetravalent mAb-bound TNFR2 molecules induce much stronger IL8 induction than TNFR2 molecules bound by conventional bivalent mAb variants. (A, B) Paired TNFR2-negative HeLa and HeLa-TNFR2 cells were stimulated for 8 h with the indicated concentrations of bivalent and tetravalent GpL fusion proteins of TNFR2-specific mAb C4 (see inset of scheme). Supernatants of HeLa-TNFR2 cells were then analyzed for TNFR2-induced IL8 production (B) and repeatedly washed cells (HeLa and HeLa-TNFR2) were analyzed for cell-associated GpL activity. Specific binding to TNFR2 shown in (A) was calculated by subtracting nonspecific binding values ​​obtained from HeLa cells from total binding values ​​derived from HeLa-TNFR2 cells. Note that luciferase activity was normalized according to the ratio of the number of GpL reporter domains to the number of TNFR2-binding domains in the two mAb variants (1 vs. 0.5). (C) IL8 production by the mAb variants was plotted directly as a function of their specific binding. The latter was converted to "occupied receptors / cell" with the help of the number of cells in the assay and the measured specific activity (RLU / molecule) of the GpL construct. The maximum binding of the two constructs from A is indicated by the dotted vertical line. The dotted line shows the linear regression of IL8 production as a function of receptor occupancy. [Figure 14] Figure 1 shows the binding of TNFR2-specific antibodies to human, mouse, and cynomolgus TNFR2. HEK293 cells were transiently transfected with human, mouse, and cynomolgus TNFR2, or empty vector. Equilibrium binding of the indicated anti-TNFR2 antibody-GpL fusion proteins was then determined at 37°C. Binding to cells transfected with empty vector was determined to obtain nonspecific binding values. [Figure 15] FIG. 1 shows the domain structure of TNFR2 and a representation of the extracellular TNFR2 domains recognized by anti-TNFR2 antibodies C4, C19, C27, and C40, as well as a scheme of the tetravalent CI-IgG1(N297A)-HC:scFvCI variant derived therefrom. CI, anti-TNFR2 clone identifier; CRD, cysteine-rich domain. [Figure 16] Figure 1 shows Western blot analysis of monospecific and bispecific TNFR2-specific CI-IgG-HC:scFvCI variants. Flag-tagged variants (200 ng) of the indicated antibody constructs were analyzed by Western blotting with anti-human IgG (lower panel) and anti-Flag antibodies (upper panel). Conventional IgG variants were included for comparison. Note that i) the heavy chains of C4-IgG1(N297A), C4-IgG1(N297A)-HC:scFvC4, and C4-IgG1(N297A)-HC:scFvC19 do not contain a Flag tag; ii) the conventional C19 variant used was the mouse IgG1(D265A) isoform. [Figure 17] Figure 1 shows TNFR2-mediated IL8 production in HT1080-Bcl2-TNFR2 cells after treatment with mono (upper panel) and biparatopic (lower panel) TNFR2-specific CI-IgG-HC:scFvCI variants. Responses induced by 200ng / ml of highly agonistic oligomerizing TNFR2-specific TNF variants are shown (dotted line). Note that in the upper panel, the C27 antibody with lower affinity still conferred some agonistic activity in the tetrameric form. [Figure 18] Figure 2 shows TNFR2-mediated cell death in Kym-1 cells after treatment with monospecific (upper panel) and bispecific (lower panel) TNFR2-specific CI-IgG-HC:scFvCI variants. Kym-1 cells were stimulated with supernatants containing the various constructs at the approximate concentrations indicated. The following day, cell viability was determined using crystal violet staining. Conventional antibodies (upper panel) were analyzed in parallel for comparison. [Figure 19] Figure 1 shows TNFR2-mediated p100 processing in Kym-1 cells after treatment with monospecific (A) and bispecific (B) TNFR2-specific CI-IgG-HC:scFvCI variants. Cells were stimulated overnight with supernatants containing the various constructs at the approximate concentrations indicated. Treatment was performed in the presence of 20 μM ZVAD to prevent TNFR2-induced cell death. [Figure 20] Purification and functional analysis of C4-IgG1(N297A)-HC:scFvC4, C4-IgG1(N297A)-LC:scFvC4, and C4-IgG1(N297A)-LC:scFvC4-HC:scFvC4. (A) Silver stained SDS-PAGE gel. (B) Gel filtration analysis. (C) IL8 induction in HT1080-Bcl2-TNFR2 cells. (D) Cell death induction in Kym-1 cells. (E) Induction of p100 processing in Kym-1 cells. ZVAD was added to prevent cell death. [Figure 21] Figure 1. Ex vivo assessment of Treg and Tcon cells after treatment with tetravalent TNFR2 agonists. PBMCs were collected from donors, cultured, harvested, and then treated with tetravalent antibodies (monospecific or bispecific; Fab-HC-scFv format). Untreated negative control cells were also assessed, as was a positive control with Star2 (a highly potent oligomerized TNFR2-selective TNF variant (Chopra et al., 2016) also called TNC-sc(mu)TNF(143N / 145R)TNFR2 agonist (oligo TNF80). The top panel shows the percentage fold change in Tregs and the bottom panel shows the percentage fold change in Tcon cells. [Figure 22] Figure 1 shows dose-dependent increase in Tregs in syngeneic FoxP3-luciferase transgenic mice by human-mouse cross-reactive tetravalent TNFR2 agonist C19-IgG1(N297A)-HC:scFvC19. Transgenic FoxP3-Luci mice were administered 8 doses of TNFR2 agonist (N=1) and then assessed daily by bioluminescence imaging. After 4 days, mice were sacrificed and cells from the spleen were analyzed. Panel (A) shows the timeline of the experiment. Panel (B) shows the results of FACS analysis of splenocytes. [Diagram 23]Figure 1 shows the dose-dependent increase of Tregs in syngeneic FoxP3-luciferase transgenic mice by the human-mouse cross-reactive tetravalent TNFR2 agonist C19-IgG1(N297A)-HC:scFvC19. Transgenic FoxP3-luciferase mice were administered different doses of C19-IgG1(N297A)-HC;scFvC19 and then assessed daily by bioluminescence imaging. After 4 days, mice were euthanized and cells from the spleen were analyzed by flow cytometry. [Figure 24] Figure 1 shows the production and analysis of TNFR2-specific TNF constructs. Panel (A): Domain structure of TNF-based molecules. TNF80 refers to a TNFR2-specific mutant of soluble TNF. Panel (B): Mouse soluble TNF80-based constructs were purified by anti-Flag affinity purification with the help of the Flag tag contained in all constructs and analyzed by SDS-PAGE and silver staining. Panel (C): Gel filtration analysis of purified proteins shown in "B". [Diagram 25] Figure 25 shows the agonist activity of different formats of soluble murine TNF80. Panel (A): Kym-1 cells were stimulated with various muTNF80 variants at the indicated concentrations. The following day, cell viability was determined using crystal violet staining. Panel (B): HT1080-Bcl2-TNFR2 cells were stimulated with various TNF80 variants at the indicated concentrations shown in Figure 24. The following day, supernatants were harvested and analyzed for their IL8 content by ELISA. Data shown are from four independent experiments. [Figure 26] Figure 1 shows in vivo Treg expansion by Fc(DANA)-TNC-muTNF80. Mice carrying luciferase under the control of the Foxp3 promoter were injected iv once with the indicated amounts of Fc(DANA)-TNC-muTNF80 and analyzed daily by bioluminescence imaging (BLI) (top panel). BLI photos on days 0 and 4 are shown in the middle panel. The bottom panel shows flow cytometry analysis-based quantification of Treg frequency on day 4. [Figure 27]FIG. 1 demonstrates that the binding molecules of the invention demonstrate selective binding to TNFR2 but not to TNFR1. HEK293 cells were transiently transfected with (i) a TNFR2 expression plasmid; (ii) an expression plasmid encoding the extracellular domain of TNFR1 with a GPI anchor domain; or (iii) an empty vector (ev). The following day, the indicated GpL-tagged binding molecules were added to the cells at the indicated concentrations. After 1 h of incubation at 37° C. and washing twice with PBS, the remaining cell-associated GpL activity was determined. GpL-TNF was used as a positive control that binds to both TN receptor types. [Figure 28] FIG. 1 shows that IgG1(N297A)-HC:scFv, IgG1(N297A)-HC:TNC, and IgG1(N-RGY) variants of 005-B08 from WO 2020 / 089474 are strong TNFR2 agonists as measured using IL-8 release. The maximum response induced by C4-IgG1(N297A)-HC:C4 as a positive control is shown by the dotted line. [Figure 29] Figure 1 shows that variation in linker length between heavy and light chains in scFvs and scFv fused to heavy or light chains of C4-IgG1(Durv)-scFv:C4 variants does not affect TNFR2 agonism activity. (A) Domain structure of the constructs. "G4S" refers to the linker sequence GGGGS and "Durv" indicates the constant region sequence of the antibody durvalumab. (B) Gel filtration of purified C4-IgG1(Durv)-HC:scFvC4(G4S)4 and C4-IgG1(Durv)-LC:scFvC4(G4S)4. (C) IL-8 release assay. The maximum response induced by the highly active oligomeric TNFR2 specific TNF mutant TNF(143N / 145R) positive control is indicated by the dotted line. [Diagram 30]Figure 2: Sequence optimization does not affect the activity of C4-IgG1(Durv)-HC / LC:scFvC4 variants. (A) Gel filtration analysis of purified C4-IgG1(Durv)-HC / LC:scFvC4 variants. (B) IL8 release assay as a measure of TNFR2 activation. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. [Diagram 31] Figure 2: TNFR2 binding of TNFR2 specific sdAbs. Equilibrium binding of the indicated TNFR2 specific VHH-Fc-GpLVHH constructs to cells expressing human TNFR2 at 37°C. Specific binding was obtained by subtracting non-specific binding values ​​(Hek293 cells) from transfectants expressing human TNFR2 (total binding). [Diagram 32] Figure 1 shows binding of TNFR2-specific hexavalent 3xVHH-Fc-GpL constructs to human and mouse TNFR2. Equilibrium binding of the indicated TNFR2-specific 3xVHH-Fc-GpL VHH constructs to cells expressing human TNFR2 at 37°C. The GpL reporter domain is cloned C-terminal to the Fc domain. Open circles = non-specific binding; open squares = total binding; closed circles = specific binding. [Diagram 33] Figure 1 shows that 3xVHH-Fc(DANA) variants of VHH:C18, VHH:C74, VHH:C188, and VHH:C238 block TNFα binding to TNFR2. HT1080-Bcl2-TNFR2 cells were pretreated with the indicated amounts of TNFR2-specific VHH constructs, or the CD40-specific construct 3xVHH:CD40(V12t)-Fc(DANA) as a negative control, or left untreated. 5ng / ml GpL-TNF was added and binding was measured after 1 hour incubation at 37°C. [Diagram 34]Figure 1: TNFR2-specific VHHs do not bind to TNFR1. HEk293 cells were transiently transfected with expression vectors encoding TNFR1 and TNFR2 and then analyzed for binding of the indicated proteins. Binding to empty vector-transfected cells served as a negative control for non-specific binding. TNF-GpL served as a positive control for successful TNFR1 binding. Filled circles = total binding; x = non-specific binding; open circles = specific binding. [Diagram 35] Figure 1: VHH:C18, VHH:C188, and VHH:C238 bind to CRD4 of TNFR2, whereas VHH:C74 binds to CRD3. (A) Black 96-well plates were coated with protein G and then loaded with 3xVHH-Fc (DANA) versions of the indicated TNFR2-specific nanobodies. After removing unbound proteins, the immobilized antibody constructs were analyzed for binding of GpL-tagged deletion mutants (ed) of TNFR2, shown in the Western blot in the right panel. Finally, the binding of these GpL-fusion proteins was quantified. (B) Clones C188 and C4 compete with TNF for binding to human TNFR2. HT1080-Bcl2-TNFR2 cells were pretreated with the indicated amounts of TNFR2-specific VHH constructs or the CD20-specific mAb Rituximab as a negative control, or left untreated. 5 ng / ml GpL-TNF was added and binding was measured after 1 hour incubation at 37°C. [Diagram 36] Figure 1 shows TNFR2-mediated IL8 release after treatment with VHH-based constructs with four N-terminal only TNFR2 binding sites. HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the indicated constructs and the resulting supernatants were analyzed for IL-8 release the next day. Maximal responses induced by the highly active oligomeric TNFR2-specific TNF mutants TNF(143N / 145R) (upper panel) or C4-IgG1(N297A)-HC:C4 (middle and lower panels) served as positive controls and are indicated by dotted lines. [Figure 37]Figure 2: TNFR2-mediated IL8 release and Kym-1 cell death after treatment with VHH-based constructs with six N-terminal only TNFR2 binding sites. (A) HT1080-Bcl2-TNFR2 cells were stimulated with the indicated constructs and IL-8 release was measured in the conditioned medium. The maximal response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. (B) Kym-1 cells were stimulated with the various hexavalent variants and the following day cell viability was determined using the MTT assay. The maximal killing induced by 200ng / ml of the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is again shown by the dotted line. [Figure 38] Side-by-side comparison of agonist activity of VHH:C188-Fc, 2xVHH:C188-Fc-GpL and 3xVHH:C188-Fc. (A, B) Purified proteins were analyzed by SDS-PAGE (A) and gel filtration (B). (C) Measurement of IL-8 release as an indicator of TNFR2 activation. The maximal response induced by the highly active oligomeric TNFR2 specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. (D) Cell viability assay after stimulation of Kym-1 cells with various VHH:C188 variants. The maximal killing induced by C4-IgG1(N297A)-HC:scFvC4 is shown by the dotted line. [Figure 39] Side-by-side comparison of IL8 release and Kym-1 cell death induction activity of VHH-Fc-GpL, 2xVHH-Fc-GpL, and 3xVHH-Fc-GpL variants of C18 and C74. (A) IL-8 release assay. Maximal response induced by highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown. [Diagram 40]Figure 2: Linker length between VHH domains of 3xVHH:C188-IgG1(Durv) variants does not affect TNFR2 agonism. Linker length between VHH domains of 3xVHH:C188-IgG1(Durv) variants does not affect TNFR2 agonism. (A) Domain structure of constructs. (B) SDS-PAGE analysis of purified constructs. (C) Gel filtration analysis. (D) HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the indicated constructs. The next day, supernatants were analyzed by ELISA for their IL8 content. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown by the dotted line. The parent molecule 3xVHH:C188-Fc(DANA) was included for comparison. [Diagram 41] Figure 2 shows that constructs of different domain architectures with 4, 6 or 9 VHH:C188 domains induce strong TNFR2 agonism. (A) Domain architecture of constructs. TNC refers to the trimerization domain of tenascin-C. (B) IL-8 release assay. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown by the dotted line. [Diagram 42] Protein G enhances the agonist activity of sequence-optimized variants of anti-TNFR2 mAb C4. (A) Western blot analysis with anti-human IgG1 antibody of cell culture supernatants of HEK293 cells transiently transfected with expression plasmids encoding the light and heavy chains of the indicated antibody variants. (B) HT1080-TNFR2 cells were challenged with the indicated concentrations of C4 variants in the presence and absence of 500 ng / ml protein G (PG). The next day, cell supernatants were analyzed by ELISA for IL8 production. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown by the dotted line. The enhanced activity in the presence of protein G indicates the presence of potential antagonism that can be achieved by generating constructs with the appropriate domain structure. [Diagram 43]Figure 2 shows TNFR2 agonism of sequence-optimized variants of 3xVHH:C188-Fc(Durv). (A) Western blot analysis with anti-human IgG1 antibody of cell culture supernatants of HEK293 cells transiently transfected with expression plasmids encoding the indicated 3xVHH:C188-Fc(Durv) variants. (B) HT1080-TNFR2 cells were challenged with the indicated concentrations of 3xVHH:C188-Fc(Durv) variants and parental 3xVHH:C188-Fc(DANA) molecules. The next day, cell supernatants were analyzed by ELISA for IL8 production. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is indicated by the dotted line. [Diagram 44] Figure 2 shows that TNFR2 agonists increase Foxp3+ regulatory T cells (Tregs) in human peripheral blood mononuclear cells (PBMCs). Fold change in Treg frequency in human peripheral blood mononuclear cells (PBMCs) 4 days after stimulation with the indicated TNFR2 agonists compared to untreated controls from the same donors. Samples were measured by flow cytometry and Tregs were defined as CD3+CD4+FoxP3+ T cells. The number of individual donors is indicated as n. Significance was tested by one-group t-test with a hypothesis value of 1 and is indicated by *P=0.05, **P=0.01, ***P=0.001, ****P<0.0001. [Diagram 45] Figure 1 shows activation of Tregs by C4-IgG1(N297A)Hc-scFvC4 antibody. Human PBMCs were stimulated with 100ng / ml C4-IgG1(N297A)Hc-scFvC4 for 4 days and expression of markers associated with Treg activity was measured by flow cytometry. Fold change expression compared to untreated controls of the same donor is shown. Tregs were defined as CD3+CD4+FoxP3+ T cells. Mean values ​​± SD and each dot represents an individual donor. [Figure 46]Figure 1. Activation of Tregs by 3xVHHC188(G4S)1-Fc(Durv) antibody. Human PBMCs were stimulated with the indicated concentrations of 3xVHHC188(G4S)1-Fc(Durv) for 4 days and expression of markers associated with Treg activity was measured by flow cytometry. Fold change in expression compared to untreated controls of the same donors is shown. Tregs were defined as CD3+CD4+FoxP3+ T cells. Mean values ​​± SD and each dot represents an individual donor. [Figure 47] Figure 2: TNFR2 agonists expand human TNFR2-expressing CD4 T cells in vitro. Enriched CD4 T cells from human TNFR2 knock-in mice were stimulated with different concentrations of (A) C4-C4 (C4-IgG1(N297A)-HC:scFvC4) or (B) 3xVHH:C188-Fc (DANA), and each agonist increased Treg frequency after 4 days of culture. Data represented fold change in the increase of CD25highFoxP3highCD4 T cells compared to untreated (isotype control) samples. Data were analyzed by one-way ANOVA. *p≦0.05, **p≦0.005. [Figure 48] Figure 1 shows serum retention of Fc(DANA)-TNC-muTNF80. IL-8 concentrations were measured by ELISA in HT1080-Bcl2-TNFR2 cell culture supernatants stimulated with different dilutions of serum (isolated from peripheral blood at given time points) from wild-type mice injected with 75 μg Fc(DANA)-TNC-muTNF80 or isotype control antibody at 0 h. The time points in the diagram refer to the delta t post-injection at which blood samples were collected. The x-axis gives the dilution of these samples in an in vitro assay for the determination of TNFR2 agonist activity. [Figure 49]TNFR2 agonists expand Tregs in vivo and upregulate activation markers: (A) Treg frequency of viable cells in spleens isolated on day 4 after stimulation with 75 μg Fc(DANA)-TNC-muTNF80 in wild-type mice. (B) Fold change of selected extracellular markers defined as mean fluorescence intensity of Tregs normalized to isotype control antibody samples. [Figure 50] TNFR2 agonists increase splenic Tregs in vivo. (A) Fold change in Treg frequency of viable cells in spleens from huTNFR2-expressing mice isolated on day 4 after stimulation with 300, 1000 μg 3xVHH:C188-Fc(DANA) or 1000 μg isotype control antibody. Fold change was calculated relative to isotype control. Comparison of mean values ​​between treatment groups and isotype control was performed via unpaired t-test. *p≦0.05, ns=not significant. (B) Fold change in Treg frequency (Foxp3+) of CD4+CD3+ T cells in spleens from huTNFR2-expressing mice isolated on day 4 after stimulation with 40 μg, 100 μg, or 250 μg C4-IgG1(N297A)-HC:scFv:C4. Fold change was calculated relative to untreated control. Comparison of mean values ​​between treatment groups and isotype controls was performed via unpaired t-test. *p≦0.05, ns=not significant. [Figure 51] TNFR2 agonists increase survival after allogeneic hematopoietic cell transplantation (allo-HCT)-induced acute graft-versus-host disease (GvHD). (A) Survival frequency after allo-HCT-induced experimental acute GvHD (FVB / N H-2q→C57BL / 6 H-2b, MHC major histocompatibility antigen mismatch model) in mice. Statistical significance was evaluated by log-rank test between C19-IgG1(N297A)-HC;scFvC19 and isotype control antibody. *p≦0.05. (B) Pathology score of intestinal samples isolated on day 6 after allo-HCT based on immune cell infiltrates and intestinal crypt injury. (C) Ex vivo analysis on day 6 of relative bioluminescence signal from allogeneic donor T cells in the indicated organs normalized to the signal from mice treated with isotype control antibody. [Figure 52] Figure 1: C4-IgG1(N297A)-HC:scFvC4 expands Tregs from African green monkey (vervet monkey) peripheral blood mononuclear cells (PBMCs). Fold and total changes in Treg frequency in vervet monkey PBMCs 5 days after stimulation with the indicated C4-IgG1(N297A)-HC:scFvC4 compared to untreated controls from the same donors. Samples were measured by flow cytometry and Tregs were defined as CD3+ CD4+ FoxP3+ T cells. Expansion was tested in n=1 vervet monkey samples. [Diagram 53] Figure 1 shows the amino acid sequence of the 3xVHH-Fc(durv) (hexavalent) format based on the C188 parent and sequence-optimized variant 14. The figure thus illustrates a preferred hexavalent binding molecule polypeptide. The CDRs shown, the entire set of three CDRs, the variable region, and the entire polypeptide represent preferred embodiments, as well as any variants thereof discussed further herein. [Figure 54] Figure 1 shows the amino acid sequence of the C4-IgG1(Durv)-HC:scFvC4(G4S)4 (tetravalent) format based on the C4 parent. The figure thus illustrates a preferred tetravalent binding molecule polypeptide. The depicted CDRs, set of three light chain CDRs, set of three heavy chain CDRs, entire set of six CDRs, variable regions, variable region pairs, and entire polypeptide represent preferred embodiments, as well as any variants thereof discussed further herein. [Figure 55] FIG. 1 shows the amino acid sequences of tetravalent C4var7-IgG1(Durv)-HC:scFvC4var7(G4S)4 and C4var16-IgG1(Durv)-LC:scFvC4var16(G4S)4 based on sequence-optimized variant 7 and variant 16, respectively. The figures thus illustrate preferred tetravalent binding molecule polypeptides. The depicted CDRs, set of three light chain CDRs, set of three heavy chain CDRs, entire set of six CDRs, variable regions, variable region pairs, and entire polypeptides represent preferred embodiments, as well as any variants thereof discussed further herein. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] TNFR2 and TNFR2 target cells The present invention provides binding molecules that bind to TNFR2, in particular to the extracellular domain of TNFR2. TNFR2 comprises a characteristic extracellular domain, a transmembrane segment, and an intracellular domain. The extracellular region contains four repeated cysteine-rich domains (CRDs). The TNFR2 to which the binding molecules of the present invention bind is typically a mammalian TNFR2. In a particularly preferred embodiment, the TNFR2 to which the binding molecules of the present invention bind is human TNFR2. The amino acid sequence of human TNFR2 is provided as SEQ ID NO: 1. The human TNFR2 amino acid sequence provided in SEQ ID NO: 1 is subdivided into a signal peptide comprising amino acids 1-22; an extracellular domain from amino acids 23-257; and a transmembrane region from amino acids 258-287. The binding molecules of the present invention typically bind to the extracellular domain of TNFR2.

[0014] In one embodiment, the binding molecule of the present invention can bind to human TNFR2, but can also bind to TNFR2 of different species, such as mouse TNFR2 and / or monkey TNFR2. In one embodiment, the binding molecule binds to human, mouse, and monkey TNFR2. In one embodiment, the binding molecule can bind to human and cynomolgus monkey TNFR2. In one embodiment, the binding molecule can bind to human and African green monkey TNFR2. In another embodiment, the binding molecule can bind to human, cynomolgus monkey, and African green monkey TNFR2. In another embodiment, the binding molecule can bind to human and mouse TNFR2. In another embodiment, the binding molecule can bind to human, mouse, and African green monkey TNFR2. In another embodiment, the binding molecule can bind to human, mouse, and cynomolgus monkey TNFR2. In another embodiment, the binding molecule can bind to human, mouse, cynomolgus monkey, and African green monkey TNFR2. In another embodiment, the binding molecule can bind to TNFR2 of at least those species.

[0015] The binding molecule of the present invention is typically used to bind to target cells, particularly target cells that express TNFR2. The target cell can be any cell type that expresses TNFR2. In one embodiment, the target cell is an immune cell, an endothelial cell, a mesenchymal stem cell, or a neuron. In a particularly preferred embodiment, the target cell is an immune cell. In a particularly preferred embodiment, the target cell is a Treg cell. In one embodiment, the immune cell that expresses TNFR2 is selected from regulatory T cells (Treg), regulatory B cells (Breg), and MDSCs (myeloid-derived suppressor cells). In one embodiment, the target cell can be a Treg cell, which is a Foxp3+CD25+ regulatory T cell.

[0016] binding molecule The present invention provides a binding molecule that binds to TNFR2. In a preferred embodiment, the binding molecule of the present invention binds to TNFR2 and stimulates signal transduction through TNFR2. In a preferred embodiment, the binding molecule is an agonist antibody or antibody fusion protein, or an antibody fragment, or an antibody fragment fusion protein, against TNFR2, thereby activating the receptor. In a preferred embodiment, the binding molecule is an antibody that specifically binds to TNFR2, preferably an antibody that specifically binds to TNFR2 and acts as an agonist thereof even without FcgR binding. In a preferred embodiment, the binding molecule of the present invention binds to TNFR2 to a greater extent than TNFR1. For example, the binding strength for TNFR2 can be at least 10, 50, 100, 500, 1000, or more than TNFR1. In one embodiment, the binding strength for TNFR2 can be at least 10,000, or at least 100,000 times stronger than TNFR1. In one embodiment, the binding strength for TNFR2 can be at least 10,000, or at least 100,000 times stronger than TNFR1. In one embodiment, the binding molecule does not bind to TNFR1 at all. In a preferred embodiment, the binding molecule, in particular the binding molecule of an antibody or antibody fragment or a fusion protein thereof of the invention, has a KD of binding of 10 -6 M, 10 -7 M, 10 -8 M, 10 -9 M, or 10 -10A binding molecule is said to specifically bind TNFR2 if it has a value lower than M. In one embodiment, the KD value is measured using cells expressing TNFR2, such as HeLa transfectants expressing TNFR2. In a preferred embodiment, the KD value is determined using the approach described in Kums J, Nelke J, Ruth B, Schafer V, Siegmund D, Wajant H.MAbs. 2017 Apr;9(3):506-520. Thus, in one embodiment, a fusion of the binding molecule with Gaussia princeps luciferase as a reporter domain is used to measure the KD value of the binding molecule as described in Kums et al., supra. In one embodiment, a cell line expressing TNFR2, such as HeLa or HEK293 cells stably expressing TNFR2 on its surface, may be used for such measurements. Typically, the binding molecules of the invention bind to TNFR2 to induce TNFR2 signaling. In one embodiment, the binding molecule, particularly an antibody, is agonistic for TNFR2 but not for TNFR1. In one embodiment, a binding molecule of the invention binds to TNFR2 but not to any other TNFR superfamily members.

[0017] In a particularly preferred embodiment, the binding molecule of the present invention is an antibody. The term "antibody" as used herein refers to any functional antibody or antibody fusion protein or antibody fragment that can typically specifically bind to an antigen of interest, as generally outlined in Chapter 7 of Paul, WE (Ed.): Fundamental Immunology 2nd Ed. Raven Press, Ltd., New York 1989, or as set forth in the "Periodic Table of Antibodies" (https: / / absoluteantibody.com / periodic-table-of-antibodies / ), which is incorporated herein by reference. Reference to an antibody herein includes antibody fragments, unless otherwise stated. The antibody of the present invention typically specifically binds to TNFR2. Reference to an antibody includes all of the various antibody formats discussed herein, including antibodies that do not have the naturally occurring four-chain structure of antibodies found in humans, for example, in a preferred embodiment, the antibody of the present invention may be or include a single-chain antibody. The antibody of the present invention may also include non-antibody sequences, such as ligand-based molecules that bind to TNFR2. In one embodiment, an antibody of the invention may comprise a polypeptide(s) that is a fusion of an antibody sequence with a non-antibody sequence, for example a fusion of a TNC peptide as discussed herein with a TNF-alpha polypeptide, particularly a TNFR2 specific mutant such as TNF80 (or TNF(143N / 145R)). In a particularly preferred embodiment, the binding molecule is an antibody that does not comprise any non-antibody binding site specific for TNFR2.

[0018] Any of the antibodies described herein may further comprise one or more ligand-based molecules, particularly ligand-based molecules specific for TNFR2. As an example, the present application describes various hexavalent antibodies that may increase the valency of the molecule to more than six by including one or more ligand-based binding molecules of TNFR2, such as any of those described herein. In one embodiment, any of the antibodies described herein may further comprise at least one ligand-based binding molecule of TNFR2 that is part of, linked to, or conjugated to a polypeptide. In one embodiment, a hexavalent binding molecule that comprises a VHH domain may be further modified to include at least one ligand-based binding molecule that is part of, linked to, or conjugated to a polypeptide. In one embodiment, the binding molecule comprises two polypeptides, each of which comprises three consecutive VHH domains, a heavy chain constant region lacking a CH1 region, together with a ligand-based TNFR2 binding molecule that is part of, linked to, or conjugated to a polypeptide. In one embodiment, the binding molecule of the present invention is modified such that each heavy chain has a ligand-based TNFR2 binding molecule that is part of, linked to, or conjugated to a polypeptide. In one embodiment, the light chain of the binding molecule can be so modified. In a preferred embodiment, the binding molecule can have a ligand-based molecular ligand based on TNFR2 that is part of, linked to, or conjugated to the C-terminal half of the polypeptide. The present invention also provides various tetravalent binding molecules, in which the binding molecule will have a ligand-based molecule rather than an scFv or other antibody-based TNFR2 binding site at the "south" or "C-terminal" portion. Thus, for example, various C4 and C4 variant antigen binding sites are described herein, and the present invention provides binding molecules in which the "north" antigen binding site is a C4 or C4 variant antigen binding site and the "south" antigen binding site is a ligand-based TNFR2 antigen binding molecule.

[0019] The antibody of the present invention can also include antibody part or fragment.For example, the antibody of the present invention can just include the antibody Fc region that is modified so that it does not bind to Fc receptor and the ligand-based binding molecule linked to the Fc region.In one embodiment, the ligand-based molecule present in the antibody of the present invention is the TNF-based ligand of TNFR2, such as the TNF promoter or its mutant.In one embodiment, the antibody is a single-chain type of TNF trimer.

[0020] The "valency" of a binding molecule, as used herein, refers to the number of binding sites that a binding molecule has, and the valency of TNFR2 refers to the number of binding sites that a binding molecule has on TNFR2. The "specificity" of a binding site refers to what it binds to, for example, it can mean which target molecule a binding molecule binds to, but also where on the target it binds. With respect to an antibody, the specificity of an antibody can also mean which target the antibody binds to, and can also be used to mean where on the antigen it binds to, with respect to the epitope bound on the target molecule. Binding molecules can have different specificities in the sense that they bind to epitopes on different target molecules (antigens). However, binding molecules can have different specificities in the sense that they bind to different epitopes on the same target molecule (antigen). For example, a bispecific binding molecule can be bispecific in the sense that it can bind to two different target molecules, or it can be bispecific in the sense that it can bind to two different epitopes of the same target molecule, particularly TNFR2.

[0021] In one embodiment, all of the binding sites of the binding molecule of the invention bind to the same target. In one embodiment, if the binding molecule of the invention is an antibody, all of the antigen binding sites of the antibody will bind to the same epitope of TNFR2. In another embodiment, the binding molecule of the invention will comprise at least two different target binding sites, each of which binds to a different target site on TNFR2, e.g., each of which recognizes a different epitope of TNFR2. In one embodiment, the binding molecule of the invention is an antibody that comprises at least two antigen binding sites that bind to different epitopes of TNFR2. In one embodiment, at least one of the binding sites of the molecule binds to a target other than TNFR2. For example, in one embodiment, the binding molecule of the invention may comprise binding sites with different specificities that recognize targets of different cell types. For example, the different specificities can be used as homing entities to link targets to cells that express TNFR2 with different cell types. In one embodiment, the binding molecule of the invention may comprise at least one binding site that binds to a target that alters the serum half-life of the binding molecule, e.g., in one embodiment, the binding molecule may bind to serum albumin. In one embodiment, only one of the binding sites of the molecule targets something other than TNFR2, hi one embodiment, there may be two such sites.

[0022] The binding molecule of the present invention has at least two valencies with respect to TNFR2. In one embodiment, the binding molecule of the present invention has two, three, four, five, six, seven, eight, nine, ten, eleven, or twelve valencies with respect to TNFR2, or at least these valencies. In a preferred embodiment, the binding molecule of the present invention has two valencies. In another preferred embodiment, the binding molecule of the present invention has three valencies. In one embodiment, the binding molecule of the present invention is trispecific. In another preferred embodiment, the binding molecule of the present invention has at least four valencies. Thus, in a preferred embodiment, the binding molecule of the present invention is four valencies. In a preferred embodiment, the binding molecule of the present invention has at least six valencies. In a preferred embodiment, the binding molecule of the present invention has six valencies. In a preferred embodiment, the binding molecule of the present invention has twelve valencies. In one embodiment, the binding molecule of the present invention is multispecific, for example, the binding molecule of the present invention has at least two valencies, preferably at least four valencies. A particularly preferred embodiment is a four-valent antibody. A further particularly preferred embodiment is a six-valent antibody. A preferred hexavalent antibody is an antibody whose TNFR2 binding site is a single domain antibody (sdAb) specific for TNFR2. A particularly preferred hexavalent antibody is an antibody whose TNFR2 binding site is a VHH binding domain specific for TNFR2. SdAbs and VHHs are discussed further below.

[0023] Along with the "valency" of the binding molecules of the present invention, a further point is the orientation / localization of the binding site of the binding molecule. This can be best considered using the example of the antibodies shown in Figures 1 and 7. Figure 1 represents an antibody with (N-terminal) unidirectional antigen binding sites, which can simply be considered to be antibodies in the "North" direction, for example, in one embodiment, such a binding molecule can be considered to have antigen binding sites that bind to TNFR2 at what would be the N-terminal portion of the heavy and light chains of a conventional four-chain antibody, which can be located at the N-terminal portion of the polypeptide of the binding molecule. The orientation of the binding site can also be defined by referring to binding sites in the same direction as "parallel" binding sites, for example, if all of the binding sites are at the N-terminal portion of the antibody chain or all are at the C-terminal portion of the antibody chain, the binding sites can be considered to be parallel. Such binding sites can also be considered to be unidirectional. Figure 7 of the present application represents an antibody with dual N-terminal and C-terminal antigen binding sites, which can be considered to be in the "North" + "South" direction or in the "reverse" direction. In one embodiment, such a binding site may be considered to have antigen binding sites in more than one plane or more than one orientation. In one embodiment, the construct of the invention is one in which the antigen binding site is present only in the "North" orientation, and therefore the antigen binding site is N-terminal. In one embodiment, the binding molecule of the invention is a North construct in which the variable domains of HC and LC are exchanged with scFv, for example (scFv-IgG(N297A) and scFv-IgG(N297A)-HC:TNC). In another embodiment, the binding molecule of the invention may have binding sites in both the "North" and "South" orientation, thereby effectively having antigen binding sites at both the N-terminal and C-terminal positions.

[0024] In one embodiment, the "north" or "N-terminus" facing the antigen binding site may be considered to be in the same direction or plane as the binding site of a conventional monomeric IgG molecule, in particular what is normally considered to be the N-terminal portion of the light and heavy chain variable regions of such a conventional IgG molecule. In one embodiment, the "south" or "C-terminus" facing the antigen binding site may be considered to be in the direction or plane in which a conventional monomeric IgG molecule does not have an antigen binding site. In one embodiment, the orientation is opposite to that of the Fab subdomains (VH+VL) in a conventional monomeric IgG molecule, thus having a TNFR2 binding domain C-terminal to the Fc domain towards the adjacent cell of an antigen expressing cell. Such a binding site may be considered to be opposite to the "normal" orientation of the Fab domain in an antibody. The hexameric IgG molecule depicted in FIG. 1 has antigen binding sites at the top and bottom of the hexamer, but it is not considered to have "north" and "south" or "reverse" oriented antigen binding sites, in the sense that it does not have an antigen binding site at what would be, for example, the C-terminal portion of a conventional monomeric IgG molecule, since all of the antigen binding sites are present in the same format as a conventional monomeric IgG molecule. In a preferred embodiment, the antibody of the invention comprises an antigen binding site at the N-terminus of the polypeptide that constitutes the antibody. In one embodiment, the binding molecule of the invention will be an antibody, in the sense that it has antigen binding sites at both the N-terminus and the C-terminus of the antibody polypeptide chain. In a particularly preferred embodiment, the binding molecule is a tetravalent antibody with two N-terminal ("north") and two C-terminal ("south") binding sites for TNFR2. In a preferred embodiment, the antibody is an IgG with an scFv at the C-terminus of each light chain. In a further preferred embodiment, the antibody is an IgG with an scFv at the C-terminus of each heavy chain. In a further preferred embodiment, the antibody is an IgG with an scFv at the C-terminus of each heavy and light chain, and such an antibody is hexavalent.

[0025] In a preferred embodiment, the binding molecule used is a multimer of individual subunits. For example, in a preferred embodiment, the binding molecule of the invention is a dimer. In one embodiment of the invention, the dimer is in the form of two scFv-Fab molecules. In a preferred embodiment, the binding molecule of the invention is a trimer. In another preferred embodiment, the binding molecule of the invention is a hexamer. In one embodiment, the binding molecule used can be an antibody. In one embodiment, the binding molecule used can be an antibody that is an antibody fusion protein hexamer. In a preferred embodiment, the heavy chain of the individual IgG molecules that make up the hexamer can contain an RGY peptide motif, particularly at its C-terminus. In a particularly preferred embodiment, the antibody of the invention contains a mutation of the RGY motif in the heavy chain, particularly in the Fc tail, which is E345R / E430G / S440Y. The RGY motif is described, for example, by de Jong et al (2016) PLOS Biology | DOI: 10.1371 / journal.pbio.1002344, which is incorporated by reference in its entirety, as well as in particular with respect to the RGF motif and its use in forming hexamers.

[0026] Typically, when the binding molecule of the present invention comprises an assembly of multimers or smaller units, it is still considered to be a "binding molecule" or "antibody" of the present invention, even if it may comprise, for example, several subunits that may be considered to be antibodies individually, such as IgG and scFv. Thus, for example, the "antibody" of the present invention may comprise an antibody fragment or fragments as part of the whole "antibody" of the present invention. As discussed below, it may include other entities such as the ligand of TNFR2, for example TNF-alpha, particularly TNF80 and other TNFR2-specific TNF mutants. Any specific format described herein may be modified to further include a ligand of TNF2.

[0027] Antibody hexamers represent one embodiment of the present invention. In one embodiment, the antibody of the present invention comprises at least a heavy chain modification at E345, in particular E345R. In another embodiment, the antibody of the present invention may comprise at least a heavy chain modification at E430, in particular E430G. In another embodiment, the antibody of the present invention may comprise at least a heavy chain modification at S440, in particular S440Y. In another embodiment, the antibody of the present invention may comprise at least a heavy chain modification at E345 and E430, in particular E345R and E430G. In a preferred embodiment, the antibody of the present invention may comprise at least a heavy chain modification at E345, E439, and S440, in particular E345R, E439G, and S440Y. In a preferred embodiment, the antibody of the present invention comprises such modifications and comprises a hexamer of IgG units. In a further preferred embodiment, the antibody of the present invention undergoes on-target hexamer formation when it binds to TNFR2.

[0028] In a preferred embodiment, the binding molecules of the present invention may include a sequence that results in trimerization. Examples of such sequences include the trimerization domain of collagen, leucine zipper, T4 fibritin, and trimerization domain of TRAF family members. The binding molecules of the present invention may, in some embodiments, include such trimerization sequences. In one embodiment, the binding molecules of the present invention may include or be trimers due to the presence of such domains. In a preferred embodiment, the trimerization sequence may be a sequence derived from tenascin-C (TNC). However, in any of the embodiments discussed herein that refer to TNC, a general trimerization sequence may be used, for example any specific sequence referred to herein that is not simply TNC, but in a preferred embodiment, TNC is used. In a preferred embodiment, the TNC sequence used is derived from chicken TNC. In another preferred embodiment, the TNC sequence used is derived from mammalian TNC. In a further preferred embodiment, the TNC is human TNC. In one embodiment, the binding molecules of the present invention do not include TNC.

[0029] In another preferred embodiment, the tenascin-C (TNC) trimerization domain may be used to assemble individual antibody or other subunits into multimers. Thus, the binding molecules, particularly the antibodies of the present invention, may comprise a TNC trimerization domain peptide, such as the peptide of SEQ ID NO: 2. The antibodies of the present invention may comprise variant TNC peptides, such as peptides that have 5, 4, 3, 2, or 1 amino acid sequence changes compared to SEQ ID NO: 2, but can still lead to the formation of multimers. In a preferred embodiment, the C-terminus of the heavy chain of the antibody comprises or is linked to a TNC peptide sequence. In one embodiment, such an approach is used to lead to trimers of individual TNFR2 binding moieties, such as antibody subunits, antibodies, and / or ligands. Any of the binding molecules described herein, unless otherwise stated, may comprise a TNC peptide as a way to allow the association of individual units, particularly as a way to form multimers.

[0030] In one embodiment, the binding molecule of the present invention is an FcγR-independent agonist of TNFR2.For example, in the example where the binding molecule is an antibody, the antibody will lack Fc region and therefore will not bind to the FcγR of that region.Alternatively, in another embodiment, the antibody will have an Fc region but will be modified so that it cannot bind to FcγR.Various Fc modifications are discussed herein and may be used to achieve modification.

[0031] In one embodiment, the binding molecule of the present invention is or comprises an antibody selected from IgG, IgM, IgA, IgE, and IgD antibodies. The binding molecule of the present invention may comprise an Fc modification discussed herein that renders it Fc gamma receptor independent. In a particularly preferred embodiment, the binding molecule of the present invention is or comprises an IgG antibody. In a particularly preferred embodiment, the binding molecule of the present invention is or comprises an IgG1 antibody. The antibody may be or comprise an IgG2, IgG3, or IgG4 antibody. In a particularly preferred embodiment, the antibody of the present invention may comprise a single domain antibody, such as a VHH single domain antibody. In one embodiment, the antibody of the present invention comprises at least two such dAbs, particularly VHHs.

[0032] In a preferred embodiment, the binding molecule provided specifically binds to TNFR2 but not to TNFR1. In a further preferred embodiment, the binding molecule provided is an FcγR-independent agonist of TNFR2. The binding molecule has at least bivalency with respect to binding to TNFR2.

[0033] In a preferred embodiment, the binding site of the binding molecule of the present invention can be selected from an antibody antigen-binding site and a ligand that binds to TNFR2. In a particularly preferred embodiment, the binding molecule of the present invention is an antibody or comprises at least a portion of an antibody. As defined herein, an antibody includes various formats that include an antigen-binding site, but not naturally occurring antibody formats. An antibody does not necessarily have to be a whole antibody, for example, an antibody of the present invention can be or comprise an antibody fragment, such as a Fab2 antibody fragment. An antibody of the present invention can also comprise an antibody fragment, for example, in a preferred embodiment, an antibody of the present invention can comprise an antigen-binding site provided by an scFv fragment or Fab as part of the antibody. An antibody of the present invention can be or comprise a single chain antibody.

[0034] Reference to an antibody includes antibody molecules that contain, are linked to, or are conjugated to a moiety that is not itself derived from an antibody-based sequence. Thus, for example, reference herein to an antibody includes an antibody that is linked or conjugated to a TNFR2 ligand or a modified version of a TNFR2 ligand. Examples of TNFR2 ligands that can be used or modified to be used include TNFα and LTα, and variants thereof. In one embodiment, the binding molecules, particularly antibodies, of the invention can include a sequence that results in multimerization. One example of such a sequence is that the antibodies of the invention can include a heavy chain with a sequence that results in multimerization, for example by including an RGY mutation that results in multimerization. In a preferred embodiment, this is an Fc tailpiece that includes an RGY modification that results in hexamerization. In another particularly preferred embodiment, the antibodies of the invention can include a tenascin-C trimerization domain peptide sequence (TNC) at the C-terminus of the heavy chain. Thus, in one embodiment, the binding molecules, particularly antibodies, of the invention can include a peptide sequence from tenascin-C provided as SEQ ID NO: 2, or a variant TNC peptide that can still result in multimerization. In another embodiment, the antibody may comprise a human TNC sequence that results in trimerization. In another embodiment, the antibody may comprise a trimerization domain, such as any of the specific domains mentioned herein.

[0035] In one embodiment, all of the binding sites of the TNFR2 binding molecule of the present invention can be antibody antigen binding sites. In another embodiment, the binding molecule of the present invention comprises at least one antibody antigen binding site of TNFR2 and at least one ligand or ligand derivative that binds to TNFR2, in other words, the binding molecule of the present invention can comprise a mixture of such binding sites. In one embodiment, the binding molecule of the present invention comprises an antibody that further comprises a ligand-based binding site of TNFR2. In a preferred embodiment, the binding molecule of the present invention, particularly the antibody of the present invention, does not comprise a ligand-based TNFR2 binding site. In embodiments in which the binding molecule is or comprises an antibody, the antigen binding site generally comprises six CDRs, i.e., three light chain variable region CDRs (LCDR1, LCDR2, and LCDR3), and three heavy chain CDRs (HCDR1, HCDR2, and HCDR3). However, as further discussed below, the binding molecule of the present invention can comprise a VHH binding domain of TNFR2, in particular, and other single domain binders. Such a VHH domain comprises three CDRs, i.e., CDR1, CDR2, and CDR3. The present invention provides a binding molecule comprising a binding site with any of the specific sets of six or three CDRs described herein in the antigen binding site of TNFR2. In another embodiment, the binding molecule may comprise any pair of VH and VL regions described herein. In a further embodiment, the binding molecule may comprise a binding site comprising or consisting of any of the VHHs described herein. Such specific sequences are described in the tables of this application. Such CDR sets are also described herein by reference to SEQ ID NO. Such CDR sets, VL and VH pairs, or VHH domains may be used in any of the structures described herein, including the structures shown in the figures and used in the examples. Variant sequences of any of the specific sequences described herein may also be used.

[0036] In a preferred embodiment, the antigen-binding site of the antibody and the ligand-based binding site are in different orientations, so that the two can be in opposite orientations, e.g., the antigen-binding site can be in the "north" orientation and the ligand-based binding site can be in the "south" or "reverse" orientation. In another embodiment, a binding molecule can have an antibody antigen-binding site in both the "north" and "south" orientations, but the ligand-based binding site can have a binding site in only one orientation, e.g., the "south" orientation. "North" and "south" refer to whether the antigen-binding site is at the N-terminal or C-terminal portion of a polypeptide. Thus, for example, in a conventional antibody, the antigen-binding site provided by the variable region of the antibody is at the N-terminal portion of the heavy and light chain polypeptides. In contrast, some of the binding molecules provided represent antibodies in which the antigen-binding site is at or towards the C-terminus of the polypeptide, e.g., scFvs are at or linked to the C-terminal portions of the light and heavy chain polypeptides. Some binding molecules include both N-terminal and C-terminal antigen-binding sites.

[0037] In one embodiment of the present invention, an antibody is used that has only a "North"-directed binding site. Figure 1 of the present application shows preferred TNFR2 binding molecules, as well as illustrative examples of Fabs that have a valency and are not binding molecules of the invention by themselves.

[0038] In the formats shown in the figures, particularly in Figures 1, 7, 24, 29, 40, and 41, when referring to parent TNFR2-specific IgG molecules, the structures are indicated with "CI" or "clone identifier", with the clone identifier indicating which clone the CDRs of the IgG antigen-binding site are derived from in the examples. When such structures are referred to herein, but no clone identifier is used, the structures are not limited to a reference to a specific clone, and therefore are not limited to CDRs derived from a particular clone from the examples, and the antigen-binding site may be derived from any suitable TNFR2 antibody. Thus, for example, reference to scFv-C4-IgG1(N297A) includes the clone identifier (CI) of clone 4, whereas reference to scFv-IgG1(N297A) indicates a binding molecule in such a format, but with an antigen-binding site that is not limited to a sequence derived from a particular clone. For any of the formats described herein in relation to a particular clone, one of the other antigen-binding sites provided herein may be used instead, and formats that are not limited to a particular clone are similarly provided. For any binding molecule that comprises an antigen-binding site that comprises heavy and light chain variable regions, the C4 and C4 variant CDRs, CDR sets, variable regions, and variable region pairs (see Table 10) may be used in preferred embodiments.

[0039] In a preferred embodiment, the binding molecule of the invention is selected from: i. A Fab2 antibody, both specific for TNFR2. ii. IgG antibodies that contain an Fc region that cannot activate Fc receptors, in particular cannot bind FcγR. Both valencies of the molecule are specific for TNFR2 and can be considered to be "north" directed. In one embodiment, the Fc region of the antibody contains a modification that means it does not bind to Fc receptors, such as any of the specific modifications described herein. In one embodiment, the antibody is an IgG1 antibody. In a preferred embodiment, the antibody has the format IgG1 (N297A) shown in Figure 1, but is not limited to a specific clone. iii. A tetravalent antibody molecule comprising four antigen binding sites specific for TNFR2 that are all "north" directed, the tetravalent antibody molecule having the basic structure of an IgG molecule modified such that each light and heavy chain has an antigen binding site linked to it at its N-terminus. In one embodiment, each antigen binding site comprises a heavy and light chain variable region (scFv domain) pair linked to a light or heavy chain of the basic four-chain IgG structure. In one embodiment, the antibody comprises the heavy and light chain constant regions of the basic IgG molecule with each N-terminus of the antibody chain having an scFv linked to it. The antibody typically comprises an Fc region that has been modified so that it does not bind to FcγR. In one embodiment, the antibody comprises any of the modifications discussed herein that prevent or reduce FcγR binding. In one embodiment, the antibody has the format scFv-IgG1(N297A) shown in FIG. 1, but is not limited to sequences originating from a particular clone. iv. A hexavalent antibody that is a multimer of three IgG antibody subunits, in which the Fc region of each individual IgG subunit is unable to bind to FcγR. All six antigen binding sites may be considered to be "north" directed. In one embodiment, the Fc regions are unable to bind to receptors of the Fc gamma-receptor family because they contain a modification such as any of the modifications disclosed herein. In another embodiment, the Fc region contains the N297A modification. Any suitable method may be used to multimerize the individual IgG subunits. In a particularly preferred embodiment, the subunits are associated through the use of tenascin-C peptide (TNC), particularly through the presence of a peptide derived from TNC, such as the peptide sequence of SEQ ID NO:2. For example, in one embodiment, the heavy chain of the IgG subunit contains or is linked to a tenascin peptide, and association with the tenascin peptide results in the assembly of the multimer. In a preferred embodiment, the peptide has the sequence of SEQ ID NO:2, or a variant sequence thereof that is still able to form multimers. In one embodiment, the antibody has the format IgG1(N297A)-HC:TNC shown in FIG. 1, but is not limited to a particular clonal sequence. v. A 12-valent antibody representing a multimer of three subunits. In one embodiment, the subunits comprising an IgG structure may be considered as having the variable regions each exchanged with a scFv specific for TNFR2, with all of the antigen binding sites facing "north", e.g., scFv-IgG1(297A)-HC:TNC. The antibody is unable to bind to Fc receptors. In one embodiment, the Fc region is modified so that it is unable to bind, and therefore the antibody does not bind to FcγR, e.g., any modification discussed herein is used. In a preferred embodiment, the N297A modification is used. The three subunits may be brought together using any suitable method, in one embodiment, the TNC trimerization domain is used. In a preferred embodiment, the C-terminus of each heavy chain contains or is linked to a TNC trimerization domain, resulting in multimerization of several Fab2s or IgGs, e.g., as in SEQ ID NO:2. In one embodiment, the antibody has the format scFv:IgG1(N297A)-HC:TNC shown in FIG. 1, but is not limited to sequences from a particular clone. vi. Hexameric IgG antibodies. Typically, all of the antigen binding sites of the antibody are specific for TNFR2 such that the antibody has 12 valencies. The Fc region of each individual IgG subunit has a modification that prevents binding to FcγR. Any of the modifications discussed herein may be used, but in a preferred embodiment, the modification is N297A. Any suitable method of achieving hexamer formation, either on-target or off-target, may be used. In a particularly preferred embodiment, each heavy chain of the individual IgG subunit contains an RGY mutation, allowing hexamer formation. In one embodiment, the antibody contains an RGY mutation plus an N297A mutation (IgG1(N-RGY) shown in Figure 1), such that both mutations prevent Fc receptor binding and the mutations result in hexamer formation. Thus, in one embodiment, the binding molecule has the format IgG1(N-RGY) shown in Figure 1, but is not limited to sequences from a particular clone.

[0040] In a particularly preferred embodiment, the binding molecule is one of the molecules described in (iii) to (vi) above. In another preferred embodiment, the binding molecule is a tetravalent binding molecule, such as a molecule described in (iv). In a further preferred embodiment, the binding molecule is a molecule having 12 valencies, such as selected from the valencies described in (v) and (vi) above.

[0041] In one embodiment of the present invention, antibodies are used that have both N- and C-terminally located binding sites on an antibody scaffold with parallel-oriented chains (Fc, IgG, Fc-TNC, TNC-Fc, IgG-TNC). Thus, antibodies with "reverse" oriented binding sites are used. Figure 7 of the present application shows illustrative examples of particularly preferred TNFR2 binding molecules with binding sites that can be considered to be in the "reverse" orientation. In a preferred embodiment, the binding molecules of the present invention are selected from the group described below: i. An antibody comprising a Fab region linked to an scFv, where the orientation of the scFv is opposite to the binding site of the Fab of TNFR2, such that the scFv domain is linked to the C-terminus of the LC or cleaved HC. In a preferred embodiment, the scFv is linked to the C-terminus of the light chain of the Fab. Thus, in one embodiment, the antibody of the invention is of the format Fab-LC:scFv shown in Figure 7, but is not limited to sequences from a particular clone. ii. An antibody that is an IgG molecule with an scFv attached to the C-terminus of each heavy chain and an Fc region that has been modified so that it does not bind to Fc receptors, particularly FcγR. The Fc region may include any of the modifications discussed herein that eliminate Fc binding, particularly the N297A heavy chain modification. In one embodiment, the antibody is of the format IgG1(N297A)-HC:scFv shown in Figure 7, but is not limited to sequences from a particular clone. iii. An antibody that is an IgG molecule with an scFv attached to the C-terminus of each light chain and an Fc region that has been modified so that it does not bind to Fc receptors, particularly FcγR. The Fc region may include any of the modifications discussed herein that eliminate Fc binding, particularly the N297A heavy chain modification. In one embodiment, the antibody is of the format IgG1(N297A)-LC:scFv shown in Figure 7, but is not limited to sequences from a particular clone. iv. An antibody that is an IgG molecule with an scFv attached to the C-terminus of each light and heavy chain, and an Fc region that has been modified so that it does not bind to Fc receptors, particularly FcγR. The Fc region may include any of the modifications discussed herein that eliminate Fc binding, particularly the N297A heavy chain modification. In one embodiment, the antibody is of the format IgG1(N297A)-LC:scFv-HC:scFv shown in Figure 7, but is not limited to sequences from a particular clone. v. An antibody that is an IgG molecule in which the heavy and light chain variable regions have each been exchanged with a TNFR2-specific scFv, and the heavy chains of the IgG molecule have each been modified to have a TNFR2-specific scFv linked to them. Thus, such an antibody has hexavalency with respect to TNFR2. The Fc region of the antibody has been modified so that it does not bind to Fc receptors, particularly FcγR. The Fc region may include any of the modifications discussed herein that eliminate Fc binding, particularly the N297A heavy chain modification. In one embodiment, the antibody is of the format LC:scFv-HC:scFv-IgG1(N297A)-HC:scFv shown in Figure 7, but is not limited to sequences from a particular clone. vi. An antibody that is an IgG molecule in which the heavy and light chain variable regions have each been exchanged into a scFv, and the light chains of the IgG molecule have each been modified to have an scFv linked to them. Thus, such an antibody has hexavalency with respect to TNFR2. The Fc region of the antibody has been modified so that it does not bind to Fc receptors, particularly FcγR. The Fc region may include any of the modifications discussed herein that eliminate Fc binding, particularly the N297A heavy chain modification. In one embodiment, the antibody is of the format LC:scFv-HC:scFv-IgG1(N297A)-LC:scFv shown in Figure 7, but is not limited to sequences from a particular clone. vii. An antibody that is an IgG molecule in which the heavy and light chain variable regions have each been exchanged into a scFv, and the light and heavy chains of the IgG molecule have each been modified to have an scFv linked to them. Thus, such an antibody has 8 valency with respect to TNFR2. The Fc region of the antibody has been modified so that it does not bind to Fc receptors, particularly the FcγR receptor. The Fc region may include any of the modifications discussed herein that eliminate Fc binding, particularly the N297A heavy chain modification. In one embodiment, the antibody is of the format LC:scFv-HC:scFv-IgG1(N297A)-LC:scFv-HC:scFv shown in Figure 7, but is not limited to sequences from a particular clone. viii. A multimer comprising three individual IgG antibody molecules comprising a tenascin-C (TNC) trimerization peptide at the C-terminus of the heavy chain, where each tenascin peptide is linked to an scFv at its C-terminus. The antibody molecules have 12 valencies with respect to TNFR2. The Fc region of the antibody is modified so that it does not bind to Fc receptors, particularly FcγR. The Fc region may comprise any of the modifications discussed herein that eliminate Fc binding, particularly the N297A heavy chain modification. In one embodiment, the antibody is of the format IgG1(N297A)-HC:TNC-scFv shown in Figure 7, but is not limited to sequences from a particular clone.

[0042] As discussed above, the formats discussed above shown in Figure 1 are provided and are not limited to sequences originating from a particular clone. In one embodiment, specific formats are also provided that do not have sequences from the specific clones provided in the examples or variant sequences discussed herein. In one embodiment for a format with an Fc region, the heavy chain comprises a heavy chain modification discussed herein, such as a DANA modification discussed herein, and does not comprise a simple N297A modification.

[0043] In a particularly preferred embodiment, a set of six CDRs or VL and VH sequences from one of the C4 clones or specific variants of the C4 clones provided herein are used in one of the formats shown in Figures 1 and 7. They may also be used in the format depicted in Figure 29. In an embodiment, any of the sets of six CDRs listed in Table 5 or pairs of variable regions listed in Table 6 may be used in any of the formats described herein, particularly those formats depicted in Figures 1 and 7. Variant sets of CDRs or pairs of variable regions may also be used as further discussed herein. Particularly preferred CDRs, CDR sets, and variable regions used are those described for C4 and C4 variants described herein.

[0044] In one particularly preferred embodiment, the binding molecule of the invention is an antibody selected from one of the following formats, some of the advantages of the formats are mentioned in brackets: · Fab1-scFv (high tissue penetration, short-lived); · IgG-HC:scFv (easy generation of bispecific variants that can be tested against a variety of antibodies); · scFv-IgG (easy generation of bispecific variants); · IgG-TNC (potentially useful in identifying pathways); and · scFv-IgG-TNC (Highly active, easy generation of bispecific variants).

[0045] In a particularly preferred embodiment, where IgG is referred to above, IgG1 is used. Alternatively, in another embodiment, where IgG1 is referred to in a particular format, the same format is similarly provided where the IgG can be any IgG subtype.

[0046] In one embodiment, a binding molecule of the invention comprises: (a) comprising TNFR2 binding sites at both the N-terminus and C-terminus of one or more heavy chains of the antibody; and / or (b) comprising TNFR2-binding sites at both the N-terminus and C-terminus of one or more light chains of the antibody; It is an antibody.

[0047] In one preferred embodiment, a binding molecule comprises an scFv at the C-terminal portion of a heavy chain of the binding molecule, which scFv binds to TNFR2. In another embodiment, a binding molecule of the invention comprises an scFv at the C-terminal portion of a light chain of the molecule.

[0048] In a particularly preferred embodiment, the binding molecule of the invention is an antibody that does comprise an Fc region carrying, linked to or conjugated to a ligand specific for TNFR2 at the C-terminal portion of the heavy chain, hi one embodiment, each heavy chain has three such ligands present at the C-terminal portion of the heavy chain.

[0049] In a particularly preferred embodiment, the binding molecule of the invention is an antibody or Fc region with C-terminal TNF trimerization followed by a TNFR2 specific TNF mutant, particularly TNF 80. In one embodiment, the polypeptide may comprise a single polypeptide chain comprising three TNF-alpha sequences capable of trimerization, such as three TNF80 units.

[0050] In a preferred embodiment, a binding molecule of the invention is bivalent with spatially mobile C- and N-terminally located antigen binding sites.

[0051] In one embodiment, the antibody of the present invention is a bivalent antibody comprising a Fab specific for TNFR2, and the heavy or light chain of the Fab portion of the antibody comprises a scFv specific for TNFR2. One advantage of such small molecules is that they may exhibit high tissue penetration. In a preferred embodiment, the antibody may comprise a scFv linked to the light chain of the Fab, for example, the two being part of the same polypeptide. In a preferred embodiment, the antibody is of the format Fab-LC:scFv shown in FIG. 7, but is not limited to sequences from the clones provided in the examples of the present invention, and may be any suitable sequence from an antibody for TNFR2. The Fab and scFv domains may, for example, in these embodiments, recognize the same or different TNFR2 epitopes.

[0052] In a preferred embodiment, the binding molecule of the invention, particularly the antibody, is tetravalent. In one embodiment, the binding molecule of the invention is in the format Fab-HC-scFv, the antibody comprises an Fc region modification that eliminates or reduces Fc receptor binding, and the antibody comprises an scFv at the C-terminus of each heavy chain. In another preferred embodiment, the antibody is in the format Fab-HC-scFv, and the antibody is bispecific in the sense that it has two different specificities for TNFR2. For example, in a preferred embodiment, the scFvs present bind to different epitopes than the antigen binding site of IgG. In an alternative embodiment, all of the TNFR2 binding sites recognize the same epitope of TNFR2. In a further preferred embodiment, the antibody of the invention is in the format scFv:IgG, the light and heavy chains of the antibody are modified to include scFvs at the N-terminal portion, and the Fc region of the antibody is modified so that it does not bind to Fc receptors. Any modification discussed herein that reduces or eliminates Fc binding may be used. Such antibodies may be bispecific in the sense that the scFv expressed on the heavy chain polypeptide may bind to one epitope of TNFR2 and the scFv expressed on the light chain has another specificity for TNFR2. In alternative embodiments, the scFv regions may all bind to the same epitope of TNFR2.

[0053] In one embodiment, the binding molecule of the invention is hexavalent. In one embodiment, the antibody is of the format IgG-LC:scFv-HC:scFv, also with an Fc region modified so that it does not bind to Fc receptors, e.g., with any of the Fc modifications discussed herein that eliminate or reduce binding to Fcγ receptors. In a preferred embodiment, the IgG is an IgG1. In another preferred embodiment, the antibody is hexavalent, which comprises an antibody Fc region, each of the heavy chains being linked at its C-terminal portion to a trimer of TNF-α or a variant thereof.

[0054] In another preferred embodiment, the binding molecule of the present invention is 12-valent. In a preferred embodiment, the 12-valent binding molecule is a multimer of three individual antibody subunits in which the heavy and light chain variable regions of the antibody are exchanged into a scFv molecule. In a preferred embodiment, the three antibody subunits are associated because the heavy chain of each antibody contains a tenascin-C peptide at the C-terminus that contains the trimerization domain of TNC. The binding molecule does not bind to Fc receptors, and in a preferred embodiment, contains an Fc region with any of the modifications described herein. In another embodiment, the binding molecule of the present invention can be a hexabody that contains six IgG molecules. For example, the binding molecule can be a hexabody of six IgG antibodies, with the Fc region containing the RGY hexabody mutation, as well as modifications that mean that they do not bind to Fc receptors, such as any of the specific modifications described herein. In a preferred embodiment, the IgG is an IgG1.

[0055] In a preferred embodiment, the binding molecule of the invention is tetravalent or dodecavalent with respect to TNFR2. * )-HC-TNC and scFv-IgG( * )(in the formula, *indicates an Fc region modification that eliminates or reduces Fc receptor binding). In a particularly preferred embodiment, the IgG portion of the binding molecule is an IgG1. HC-TNC indicates that the heavy chain of the IgG comprises a tenascin peptide, such as the peptide of SEQ ID NO: 2 or a variant thereof. In another embodiment, the binding molecule of the invention is an antibody selected from the format scFv-IgG ( * )-HC-TNC, scFv-IgG( * ), and IgG ( * )-HC:RGY. * For the )-HC:RGY format, again the preferred IgG portion of the molecule is an IgG1 antibody.

[0056] In a particularly preferred embodiment, the binding molecule of the present invention can cause the formation of hexamers of TNFR2 molecules on the surface of a cell, or at least hexamers. In one embodiment, the binding molecule of the present invention can bind to TNFR2 receptors in cis, binding to multiple TNFR2 receptors on the surface of the same cell. In one embodiment, the binding molecule can at least cause the formation of hexamers of TNFR2 on the surface of the same cell. In another embodiment, the binding molecule of the present invention can bind to TNFR2 molecules in trans, i.e., the binding molecule binds to at least one TNFR2 molecule on the surface of one cell and at least one TNFR2 molecule on the surface of another cell. In one embodiment, the binding molecule can bind to TNFR2 in cis and trans. For example, the binding molecule can have the ability to bind in cis and trans, and in another embodiment, can actually bind in cis and trans. In one embodiment, the binding molecule of the present invention causes the formation of hexamers of TNFR2 on both cells when trans binding occurs. In one embodiment, the orientation of the binding site can affect whether the antibody binds to TNFR2 in a cis and / or trans manner. In one embodiment, the binding molecule of the present invention is a molecule with at least two TNFR2 binding sites with opposite orientations relative to each other, and preferably, the binding molecule has at least two such TNFR2 binding sites that are antibody antigen binding sites.In a preferred embodiment, the binding molecule of the present invention is a molecule with at least two TNFR2 binding sites that can bind to TNFR2 molecules on different cells, allowing trans-binding, and preferably, the binding molecule has at least two such TNFR2 binding sites that are antibody antigen binding sites.In another preferred embodiment, the binding molecule of the present invention has at least 4 valency and can bind to multiple different molecules of TNFR2 on the surface of the same cell, and preferably, the binding molecule has at least 6 valency.

[0057] In another embodiment, the binding molecule of the present invention can be a diabody.Diabodies are typically small bivalent antigen-binding antibody moieties, and there are a variety of diabody formats, any of which can be used, for example, a diabody can comprise a heavy chain variable domain connected to a light chain variable domain on the same polypeptide chain, which is connected by a peptide linker that is too short to allow pairing between the two domains on the same chain.This leads to pairing with the complementary domain of another chain and the assembly of a dimeric molecule with two antigen-binding sites of TNFR2.

[0058] A wide variety of antibody formats exist and may be used as or as a part of the antibody of the present invention, for example, Fab, Fv, scFv, (ScFv)2, (ScFv2), bispecific (scFv2), triabody, tetrabody, bispecific sc(Fv)2, tandem di-scFv, tandem tri-scFv, and triabody all represent possible antibody formats of the present invention. As discussed above, the binding molecules of the present invention have at least two spatially mobile potential cell-cell connected TNFR2 binding sites, or at least six "parallel" or "unidirectional" oriented cis-acting binding sites. In embodiments where the formats mentioned herein have a valency, they may be used as a part of the antibody of the present invention in such formats, rather than the whole antibody.

[0059] In one embodiment, the binding molecule of the present invention may contain a variant binding site of TNFR2 rather than one of the specific binding sites of TNFR2 described herein. In a preferred embodiment, the variant binding molecule of TNFR2 is defined by its ability to cross-block the binding of one of the specific binding molecules of TNFR2 described herein to TNFR2. The ability to cross-block can be evaluated using a binding assay that measures the binding of the binding molecule to TNFR2 and the ability of additional binding molecules to cross-block such binding to TNFR2. In one embodiment, the ability to cross-block is tested using a monovalent molecule that simply contains one copy of the binding site of TNFR2. For example, the ability of an scFv that contains a binding site to cross-block an scFv that differs from one of the specific binding molecules described herein only by having a binding site of TNFR2 is tested. In another embodiment, the ability of actual binding molecules to cross-block each other is tested without converting them into scFv molecules. In a further embodiment, the binding molecule of the present invention can block or reduce the binding of TNFα to TNFR2. In another embodiment, a binding molecule of the invention is capable of reducing or blocking the binding of lymphotoxin-alpha to TNFR2.

[0060] In one embodiment, the affinity of the binding domain for TNFR2 in a binding molecule, particularly an antibody, of the invention is about 400 nM or less, 200 nM or less, e.g., about 100 nM, 50 nM, 20 nM, 10 nM, 1 nM, 500 pM, 250 pM, 200 pM, 100 pM, or less. DIn one embodiment, the binding affinity is 50 pM or less. For example, in one embodiment, the individual TNFR2 binding sites in the binding molecules of the invention exhibit such affinity for TNFR2, rather than the overall affinity for the binding molecule. In one embodiment, such affinity is the overall avidity of the antibody for TNFR2. In one embodiment, the affinity of the individual antigen binding sites of the binding molecules of the invention can be less than 1 μM, less than 750 nM, less than 500 nM, less than 250 nM, less than 200 nM, less than 150 nM, less than 100 nM, less than 75 nM, less than 50 nM, less than 10 nM, less than 1 nM, less than 0.1 nM, less than 10 pM, less than 1 pM, or less than 0.1 pM. In some embodiments, the Kd is about 0.1 pM to about 1 μM.

[0061] In one embodiment, the antibody of the present invention is a monoclonal antibody. In one embodiment, the antibody of the present invention is an IgG1, IgG2, or IgG4 antibody, or comprises a heavy chain constant region of that class of antibody. In a particularly preferred embodiment, the antibody of the present invention is an IgG1 isotype antibody, or comprises such an antibody. In a preferred embodiment, the constant region used is a human IgG constant region. In a particularly preferred embodiment, a human IgG1 constant region is used. This may have any of the modifications described herein. The term "antibody" encompasses antibodies from any suitable species of origin, including, but not limited to, chicken, shark, and mammals, such as mice, hamsters, rabbits, goats, cows, llamas, non-human primates, and humans. In a preferred embodiment, the antibody of the present invention is a humanized antibody. In another preferred embodiment, the binding molecules of the present invention, particularly antibodies, are fully human. Non-limiting examples of methods for generating humanized antibodies are known in the art, for example from Riechmann et al. (Nature. 1988 Mar 24; 332(6162):323-7) or Jones et al. (Nature. 1986 May 29-Jun 4; 321(6069):522-5). In a further preferred embodiment, the antibody of the invention is a human antibody. As discussed herein, the binding molecules, particularly the antibodies, of the invention may contain non-human sequences, such as tenascin-c peptide, but in a preferred embodiment, all of the antibody-based sequences of the antibody are human, despite the presence of tenascin peptide sequences. The term "antibody" encompasses monomeric antibodies (e.g., IgD, IgE, IgG) or oligomeric antibodies (e.g., IgA or IgM). The term "antibody" also encompasses, but is not limited to, isolated antibodies and modified antibodies, such as genetically engineered antibodies, such as chimeric or humanized antibodies.

[0062] In a preferred embodiment, the binding molecule of the present invention has only one specificity for TNFR2. Thus, in a preferred embodiment, all of the antigen-binding sites of the antibody of the present invention that are specific for TNFR2 are specific for the same epitope of TNFR2. In another embodiment, the binding molecule of the present invention comprises more than one specificity for TNFR2. For example, this may be because the binding molecule comprises both antibody antigen-binding sites as well as a ligand-based TNFR2 binding site. In a preferred embodiment, the binding molecule of the present invention is an antibody that comprises an antigen-binding site that recognizes more than one epitope of TNFR2. For example, in a preferred embodiment, the binding molecule of the present invention, particularly an antibody, is bispecific for TNFR2. In a preferred embodiment, in the binding molecule of the present invention, one specificity of the binding molecule for TNFR2 is provided by a pair of light and heavy chain variable regions, while the other is provided by an scFv fragment that is present as part of the entire binding molecule. In another embodiment, all of the TNFR2 binding sites of the binding molecule, particularly an antibody, are provided by scFv with the binding molecule formed from two different polypeptides, each of which comprises an scFv with a different specificity.

[0063] In one embodiment, the binding molecules of the invention may comprise an Fc region or an antibody molecule with irrelevant specificity, and the binding ability to TNFR2 is provided by other parts of the binding molecule. For example, in one embodiment, the binding molecules of the invention comprise an Fc region at the C-terminal part of the heavy chain that is a TNF-alpha molecule, particularly a mutant TNF-alpha molecule selective for TNFR2, such as a TNF80 mutant. In one embodiment, the binding molecules of the invention may comprise an immunoglobulin with irrelevant specificity, and the C-terminal part of the heavy chain may be a TNF-alpha molecule, particularly a mutant TNF-alpha molecule, such as TNF80. In one embodiment, the Fc or heavy chain of the irrelevant antibody may comprise TNF-alpha, particularly a TNF80 mutant, as part of the same polypeptide. In one embodiment, they may comprise a single chain polypeptide comprising three TNF-alpha polypeptides, particularly three TNF80 polypeptides. In another embodiment, the heavy chain may comprise a TNC peptide or a variant thereof, prior to the TNF-alpha part of the polypeptide. In a further preferred embodiment, a binding molecule of the invention may comprise an Fc region that does not have a variable region, but rather has two heavy chain portions of the Fc region as part of a single polypeptide.

[0064] An example of a preferred binding molecule that comprises a TNF ligand is Fc(DANA)-TNC-TNF80, as depicted in FIG. 24, where DANA represents the heavy chain of the Fc region that comprises D265A and N297A mutations, and TNC represents the tenascin trimerization domain peptide sequence. Thus, a preferred binding molecule of the present invention comprises three Fc regions, each of which has a DANA modification of D265A and N297A. In a further preferred embodiment, each of these heavy chain polypeptides comprises a TNC peptide. In a further preferred embodiment, each heavy chain polypeptide of each Fc region comprises a TNF-alpha sequence, particularly a TNF80 sequence, following the TNC peptide sequence.

[0065] A further preferred binding molecule of the invention is Fc(DANA)-TNF80, in which each heavy chain polypeptide comprises three Fc regions including the D265A and N297A DANA modifications and TNF-alpha, particularly TNF80 sequences. The invention also provides as binding molecules the Fc-TNC-TNF80 and Fc-TNF80 format molecules shown in FIG.

[0066] In another embodiment, a binding molecule is provided that specifically binds to TNFR2 but not to TNFR1, the binding molecule being an FcγR-independent agonist of TNFR2, having at least bivalency with respect to binding to TNFR2, and comprising a mutant TNF-alpha that binds to TNFR2 but not to TNFR1, and a polypeptide comprising a tenascin peptide sequence that oligomerizes the polypeptide. In one embodiment, such a binding molecule does not comprise an antibody sequence. In one embodiment, the molecule comprises polypeptides each comprising a mutant TNF-alpha sequence together with a tenascin peptide, such as any tenascin polypeptide disclosed herein. In a preferred embodiment, the tenascin peptide sequence results in oligomerization of such polypeptides. In one embodiment, the mutant TNF-alpha has a 143N / 145R modification, resulting in selective binding to TNFR2 compared to TNFR1. Such a TNF mutant may also be referred to as TNF80.

[0067] Binding molecules with more than one specificity for TNFR2 In a preferred embodiment, all antigen-binding sites of the binding molecule specific for TNFR2 have the same specificity for TNFR2.In particular, they may all have the same amino acid sequence and therefore bind to the same epitope.In one embodiment, all antigen-binding sites are specific for the same epitope of TNFR2.In another embodiment, all TNFR2 antigen-binding sites can cross-block the binding of other TNFR2 antigen-binding sites to TNFR2.

[0068] However, in a further preferred embodiment, the binding molecule may have antigen binding sites that bind to TNFR2, but at least two of those antigen binding sites have different specificities for TNFR2. In one embodiment, the binding molecule comprises at least two antigen binding sites of TNFR2 that bind to different epitopes of the same TNFR2 molecule. In another embodiment, the at least two antigen binding sites do not cross-block each other's binding to TNFR2. In one embodiment, the binding molecule is biparatopic with respect to TNFR2 in the sense that it comprises antigen binding sites that recognize at least two different epitopes of TNFR2. In another embodiment, at least 2, 3, 4, 5, or 6 different specificities for TNFR2 are conferred by the antigen binding sites. In one embodiment, all of the antigen binding sites of the binding molecule specific for TNFR2 have different specificities for TNFR2.

[0069] In one embodiment, the binding molecule is a hexavalent binding molecule comprising two polypeptides, each of which comprises at least two different antigen binding sites with different specificity for TNFR2.In one embodiment, each polypeptide comprises three such specificities.In one embodiment, the two polypeptides comprise different specificities on the same polypeptide, but the two polypeptides are identical.In another embodiment, the two polypeptides are not identical, and each polypeptide comprises three different specificities for TNFR2, so that there are six different specificities for TNFR2 in total.

[0070] In another embodiment, where the antibody comprises "north" and "south" directed antigen binding sites, the "north" and "south" directed antigen binding sites confer different specificities for TNFR2. For example, in one embodiment, any tetravalent antigen binding molecule described herein may comprise two antigen binding sites with one specificity for TNFR2 and two antigen binding sites with another specificity. In one embodiment, the two "south" directed scFvs may have different specificities for TNFR2 than the "north" directed antigen binding sites.

[0071] sdB, SdAb and VHH binding domains In a particularly preferred embodiment, the binding molecules of the present invention comprise single domain binders (sdB). Single domain binders include, for example, non-Ig engineered protein scaffolds, such as darpins, affibodies, adnectins, anticalins proteins, or peptides and others. Thus, whenever reference is made to sdB, sdAb, HCAb, and VHH, it may be possible to use darpins, affibodies, adnectins, anticalins, or peptides capable of binding to TNFR2, and the term sdB encompasses such binding entities as used.

[0072] Examples of single binding domain binders (sdBs) include in particular single domain antibodies (sdAbs), such as heavy chain only antibodies (HCAbs), in particular VHH domain antibodies. The use of VHH antigen binding domains is a particularly preferred embodiment. Single domain antibodies (sdAbs) are antibody fragments consisting of a single monomeric variable antibody domain. Like whole antibodies, sdAbs can selectively bind to a specific antigen. sdAbs can be antibody fragments that can be engineered from a single monomeric variable domain of either camelid heavy chain antibodies (VHH) or cartilaginous fish IgNARs (VNAR), or developed from camelized human antibodies. Any such sdAbs may be used. Particularly preferred sdAbs are VHH domains. In one embodiment, the binding molecules of the invention may comprise at least two sdAb domains. In one embodiment, the binding molecules may comprise at least two sdAb domains on the same polypeptide. In a preferred embodiment, an antibody of the invention may be a single domain antibody or a molecule comprising a single domain antibody, where the total valency of the binding molecule is at least 2. For example, an antibody of the invention may comprise two such single domain antibodies linked together as part of a whole antibody.

[0073] In another preferred embodiment, the binding molecule of the present invention may comprise a VHH binding domain as sdAb. SdAbs from organisms such as camelids, sharks, and other cartilaginous fish that produce heavy chain-only antibodies may be used. These single domain variable fragments of heavy chain-only antibodies are called VHH or nanobodies or sdAbs. VHHs retain the immunoglobulin fold shared by antibodies using three hypervariable loops, CDR1, CDR2, and CDR3, to bind to their targets. VHH fragments (e.g., NANOBODY®) are recombinant, antigen-specific, single domain, variable fragments derived from camelid heavy chain antibodies.

[0074] The binding molecule of the invention is preferably at least bivalent with respect to TNFR2, and in one embodiment where the binding molecule comprises a VHH fragment, it comprises at least two such VHH fragments. The individual VHH fragments may be linked together, for example through the use of a linker, and thus in one embodiment may be expressed as a single polypeptide comprising altogether at least two VHH fragments, each capable of specifically binding to TNFR2. The VHH fragments do not have an Fc region, and in one embodiment where a VHH fragment is present as part of the binding molecule of the invention, the binding molecule lacks an Fc region and for that reason does not bind to Fc receptors. In one embodiment, the binding molecule of the invention comprises or at least comprises three, four, five, six or more VHH fragments specific for TNFR2. In another embodiment, the binding molecule of the invention comprises a single VHH domain antibody. In a particularly preferred embodiment, the antibody comprises at least four sdAbs. In a particularly preferred embodiment, the antibody comprises four sdAbs. In a particularly preferred embodiment, the antibody comprises at least six sdAbs. In a particularly preferred embodiment, the antibody comprises six sdAbs. In a further preferred embodiment, the sdAb is a VHH. In one embodiment, the binding molecule comprises the number of sdBr.

[0075] In one embodiment, the binding molecule of the present invention comprises an Fc region but no CH1 region and can bind to TNFR2 without the involvement of a light chain. In a preferred embodiment, the binding molecule of the present invention comprises one or more VHHs, comprises an Fc region, and can bind to TNFR2 without the involvement of a light chain. In another embodiment, the binding molecule of the present invention comprises one or more VHHs and comprises a TNC trimerization domain or an Fc region, or both, as depicted in the portion of the structure shown in Figure 41.

[0076] Antibody constant region and Fc region functions In a preferred embodiment, the binding molecule of the invention does not bind to Fc receptors, and in particular does not bind to FcγR receptors. In a preferred embodiment, the binding molecule of the invention is an antibody and does not bind to Fc receptors, either because it does not contain an Fc region or, alternatively, because it has an Fc region that has been modified so that it does not bind to Fc receptors. Fc domains, as used herein, are generally represented by -(CH 2 CH 3 ) 2 where CH2 is a heavy chain CH2 domain, CH3 is a heavy chain CH3 domain, and there are two CHs, one from each heavy chain. 2 CH 3 In one embodiment, the antibody of the invention comprises the -CH 2 CH 3 In one embodiment, the antibody of the present invention does not include a CH fragment. 2 In one embodiment, the antibody of the invention does not contain a CH domain. 3 Does not include the domain.

[0077] In one embodiment, the binding molecules of the invention bind to an Fc gamma receptor, but to a substantially reduced extent compared to the binding of the same antibody comprising an unmodified Fc region to the FcgR (e.g., at least 70%, at least 80%, at least 90%, at least 95%, at least 98%, at least 99%, or 100% reduction in binding to the FcγR compared to the binding of the same antibody comprising an unmodified Fc region measured to the FcγR). In a particularly preferred embodiment, the binding molecule has no detectable binding to the FcγR at all. Binding, including the presence or absence of binding, can be determined using a variety of techniques known in the art, including but not limited to equilibrium methods (e.g., enzyme-linked immunosorbent assay (ELISA); KinExA, Rathanaswami et al. Analytical Biochemistry, Vol. 373:52-60, 2008; or radioimmunoassay (RIA)), or surface plasmon resonance assays or other mechanisms of kinetic-based assays (e.g., BIACORE™ analysis or Octet™ analysis (forteBIO)), and other methods such as indirect binding assays, competitive binding assays, fluorescence resonance energy transfer (FRET), gel electrophoresis and chromatography (e.g., gel filtration).

[0078] In one embodiment, if an Fc region is present, the Fc region used is mutated, in particular comprising the mutations described herein. In one embodiment, the mutations are to eliminate binding to Fc receptors, in particular FcγR. In a preferred embodiment, the antibody of the present invention is mutated so as not to bind to Fc receptors.

[0079] In one embodiment, the antibody of the present invention may comprise an Fc region, but does not comprise the variable region normally found in naturally occurring four-chain antibodies. In a preferred embodiment, such an Fc region may comprise other sequences as part of the polypeptide comprising the heavy chain of the Fc region, in particular a TNC peptide and / or TNF-alpha (in particular TNF80), and preferably both, in particular in the C-terminal portion of the polypeptide.

[0080] In one embodiment, the antibody of the invention may comprise aglycosyl IgG, for example to provide reduced Fc function, particularly a near Fc-null phenotype. In one embodiment, the antibody of the invention has a modification at N297, and particularly N297A. In one embodiment, the antibody of the invention has a modification at F243 and / or F244, particularly a modification meaning that the antibody comprises aglycosyl IgG. In one embodiment, the antibody of the invention may comprise a F243A and / or a F244A heavy chain modification. In another embodiment, one or more of F241, F243, V262, and V264 may be modified, particularly to amino acids that affect glycosylation. In one embodiment, the antibody of the invention may have the following modifications: F241A, F243A, V262E, and V264E. Such modifications are discussed in Yu et al (2013) 135(26): 9723-9732, which is incorporated by reference in its entirety, particularly with reference to the modifications discussed therein. Such modifications provide a way to modulate, for example, Fc receptor binding. Modifications that affect the glycosylation of the antibody may be present. Additionally, the antibody of the present invention may be produced in a cell type that affects glycosylation as a further approach to glycoengineering. In one embodiment, the fucosylation, sialylation, galactosylation, and / or mannosylation of the antibody of the present invention may be altered either by sequence modification and / or via the type of cell used to produce the antibody.

[0081] In one embodiment, an antibody of the invention has an alteration at positions 297 and / or 299. For example, in one embodiment, an antibody of the invention comprises in its heavy chain an N297A alteration, preferably N297Q, or a mutation of Ser or Thr to other residues at position 299. In one embodiment, an antibody of the invention has both of those alterations.

[0082] In one embodiment, the binding molecule of the present invention, particularly the antibody, comprises two different heavy chains, and the heavy chains comprise modifications that allow the different heavy chains to preferentially associate with each other, compared to the heavy chains that associate with the same heavy chain. Such an approach can be used particularly when the antibody has more than one specificity with respect to TNFR2, for example when the antibody binds to two different epitopes of TNFR2. In one embodiment, the two different heavy chains comprise knob-into-hole mutations. In certain aspects, the knob-into-hole mutation is a T366W mutation in one heavy chain constant region, and a T366S, L368A, and Y407V mutation in the other domain. In certain aspects, the modification comprises a charge-pair mutation. In certain aspects, the charge-pair mutation is a T366K mutation in one heavy chain constant region, and a corresponding L351D mutation in the other domain. In an alternative embodiment, rather than having a modification that results in preferential pairing of different heavy chains, the heavy chains contain a modification that means that heterodimers comprising two heavy chains can be preferentially purified from homodimers that comprise only one type of heavy chain. For example, the modification may alter the affinity for Protein A, such that one heavy chain can still bind to Protein A, but the modified heavy chain cannot, meaning that a heterodimer of two different heavy chains can be purified based on its affinity for Protein A.

[0083] In other embodiments, the heavy and light chains may contain modifications that alter whether or not disulfide bridges form between them. In various embodiments, the modifications include mutations that generate engineered disulfide bridges between the light and heavy chains. As described herein, an "engineered disulfide bridge" is a mutation that provides non-endogenous cysteine ​​amino acids in two or more polypeptides such that a non-natural disulfide bond forms when two or more domains associate. Engineered disulfide bridges are described in detail by Merchant et al. (Nature Biotech (1998) 16:677-681), which is incorporated herein by reference in its entirety. In certain embodiments, the mutation that generates the engineered disulfide bridge is a K392C mutation in one of the first or second CH3 domains and a D399C in the other CH3 domain. In a preferred embodiment, the mutation that generates the engineered disulfide bridge is a S354C mutation in the first or second CH3 domain and a Y349C in the other CH3 domain. In another preferred embodiment, the mutation that generates the engineered disulfide bridge is a 447C mutation in both the first and second CH3 domains provided by an extension of the C-terminus of the CH3 domain incorporating the KSC tripeptide sequence.

[0084] In one embodiment, the binding molecule of the present invention may contain a modification that alters the serum half-life of the binding molecule. Thus, in another embodiment, the antibody of the present invention has an Fc region modification(s) that alters the half-life of the antibody. Such modifications as well as modifications that alter Fc function may be present. In a preferred embodiment, the antibody of the present invention has a modification(s) that alters the serum half-life of the antibody. In a particularly preferred embodiment, the antibody of the present invention has a modification(s) that alters the serum half-life of the antibody compared to an antibody that lacks such modifications. In one embodiment, the modifications result in an increase in serum half-life. In another embodiment, they result in a decrease in serum half-life. In another preferred embodiment, the antibody of the present invention contains one or more modifications that collectively silence the Fc region and decrease the serum half-life of the antibody compared to an antibody that lacks such modifications.

[0085] In a particularly preferred embodiment, the antibody has a constant region with little to no effector function, such as an antibody derived from the FDA approved antibody durvalumab with Fc modifications L234F / L235E / P331S. The Fc modifications in durvalumab serve to remove Fc function, and therefore the use of light and heavy chain constant regions or durvalumab is a particularly effective way to provide a constant region with a desirable lack of Fc function. Durvalumab has obtained clinical approval, which further illustrates why the use of its constant region represents a particularly preferred embodiment. Thus, in a preferred embodiment, the binding molecule comprises the Fc region of durvalumab.

[0086] In a preferred embodiment, when the binding molecule comprises a VHH binding domain, the polypeptide comprising the VHH domain further comprises CH2 and CH2 domains of durvalumab. In a further preferred embodiment, the binding molecule comprises a heavy chain and a light chain, and the antibody comprises the light and heavy chain constant regions of durvalumab. In any embodiment, when the binding molecule comprises a constant region sequence derived from durvalumab, they may include any of the constant region modifications discussed herein. Durvalumab has a human IgG1 backbone. Thus, in a further preferred embodiment, when the binding molecule comprises a constant region, it may comprise a human IgG region. In a preferred embodiment, the binding molecule may comprise a human IgG1 region. Such regions may be modified to remove Fc function. In one embodiment, the constant region may be modified to delete the CH1 region, particularly when the antigen binding domain is sdBr.

[0087] The heavy and light chain constant region sequences of durvalumab are provided as SEQ ID NOs: 135 and 136, respectively. In a preferred embodiment, the binding molecules of the invention comprise such light and heavy chain variable sequences or variants of such sequences. In a preferred embodiment, the variant sequence or sequences have at least 90% sequence identity with the relevant specific sequence. In another embodiment, the variant has at least 95% sequence identity. In one embodiment, the binding molecules of the invention comprise the heavy and light chain constant region sequences of SEQ ID NOs: 135 and 136, but with one or more constant region sequence modifications discussed herein. In a preferred embodiment, such heavy and light chain constant regions are used in the tetravalent binding molecules provided herein.

[0088] A variant of the heavy chain constant region of the durvalumab sequence with a deletion of the CH1 region is provided as SEQ ID NO: 316. In a preferred embodiment, such a constant region is used when the binding molecule comprises an sdAb-based antigen binding site, preferably a VHH antigen binding site. Such a sequence, however, one or more of the other constant region modifications discussed herein may be used. In an embodiment, the binding molecule of the invention may comprise SEQ ID NO: 316 or a variant thereof having at least 90% sequence identity. In another embodiment, the variant may have at least 95% sequence identity. The variant in such an embodiment still has a deletion of CH1. Such a heavy chain constant region with a CH1 deletion is preferably used in an embodiment in which at least one of the antigen binding sites is a VHH antigen binding domain specific for TNFR2.

[0089] Exemplary Preferred Highly Valent TNFR2 Binding Molecules In a preferred embodiment, the binding molecule of the present invention has at least 4 valency with respect to TNFR2. In one embodiment, the valency of the binding molecule is 4-9 with respect to TNFR2. Without wishing to be bound by a particular theory, it is believed that a higher density of TNFR2 binding sites in a limited space may help to promote TNFR2 agonist activity. Surprisingly, the inventors have found that it is possible to convert known TNFR2 antagonist antibodies into TNFR2 agonist molecules through reformatting the antibodies into the formats described herein, particularly with 4 or higher valency.

[0090] In a particularly preferred embodiment, the binding molecules provided are tetravalent with respect to TNFR2. Examples of tetravalent formats are provided in the examples of the present application. The invention provides any of those formats comprising one of the specific binding domains disclosed herein. Examples of preferred formats include: scFv-IgG (see FIG. 1); IgG-HC:scFv (see FIG. 7); IgG-LC:scFv (see FIG. 7). IgG-HC:scFv represents a particularly preferred format. IgG-LC:scFv also represents a particularly preferred embodiment. In one embodiment, the scFv group is linked to the heavy or light chain in such a format by a linker. Thus, in a preferred embodiment, the scFv is linked using a G4S linker. In another embodiment, the antibody will comprise a linker separating at least two of the binding domains. An example of a preferred tetravalent format is illustrated in FIG. 29. The construct shown in FIG. 29 comprises a C4 binding domain, which represents a preferred binding domain. However, also provided is the format depicted in Figure 29, in which the VH and VL regions of each binding site are any of the variant C4 sequences described herein. The format depicted in Figure 29 can also be used when the binding site is a binding molecule that includes any of the sets of six CDRs described herein.

[0091] A further particularly preferred binding molecule format of the invention is a hexavalent binding molecule, particularly a binding molecule having six specific TNFR2 binding sites. The inventors have unexpectedly found that binding molecules in a hexameric format exhibit less variation in TNFR2 agonist activity than when different TNFR2 antigen binding domains are used to generate otherwise identical hexavalent molecules.

[0092] Exemplary hexavalent formats include any format used in the examples or figures of this application.In a preferred embodiment, the hexavalent antibody comprises at least two binding sites specific for TNFR2 on the same polypeptide.In a particularly preferred embodiment, the hexavalent antibody comprises two polypeptides, each polypeptide having three binding sites for TNFR2, particularly three VHHs, particularly the specific VHHs described herein.

[0093] In one embodiment, the VHH antigen binding sites in the hexavalent binding molecule are all the same.In another embodiment, the binding molecule can have at least two antigen binding sites with different specificities for TNFR2, for example, binding to different epitopes.In another embodiment, the antigen binding sites of the binding molecule specific for TNFR2 can all have different specificities for TNFR2, i.e., bind to different epitopes.In another embodiment, the antigen binding sites can be different in the sense that they do not cross-block each other when binding to TNFR2.

[0094] Examples of preferred hexavalent binding molecule formats include the formats described in the examples and figures of this application. In the examples where the figures or examples describe an antibody format with a specific TNFR2 binding domain, the invention also provides antibodies of the same format, but with one of the other TNFR2 binding domains described herein. In a particularly preferred embodiment, the TNFR2 binding domain is a VHH domain. In a preferred embodiment, the VHH domain used is one of the specific VHH domains or variants thereof described herein. In a preferred embodiment, the binding molecule is in one of the hexavalent formats depicted in Figure 40. Further preferred formats and specific sequences are provided in Figures 54 and 55. A particularly preferred format comprises two polypeptides, each of which has three VHH domains. In a further particularly preferred format, the binding molecule is in the format 3xVHH-Fc. For example, a particularly preferred format is 3xVHH-Fc. A further preferred format is 3xVHH-Fc(DANA). A particularly preferred VHH domain used in the binding molecules of the invention is C188. More preferred VHH domain is one of the specific variants of C188 described herein.Thus, 3xVHH:C188-Fc is a particularly preferred binding molecule.More preferred binding molecule is 3xVHH:C188-Fc(DANA).Alternatively, C188 in any of these formats can be substituted for one of the other VHH domains described herein.

[0095] In a further particularly preferred embodiment, when the hexavalent binding molecule comprises an Fc region, it comprises the Fc region of the antibody durvalumab. In an alternative embodiment, the binding molecule comprises the Fc region of durvalumab, but any Fc region described herein.

[0096] Generation and screening of binding molecules In one embodiment, the binding molecules, particularly antibodies, of the invention are mutated to provide improved affinity for the target antigen or antigens, particularly TNFR2. Such variants have been produced by mutating the CDRs (Yang et al., J. Mol. Biol., 254, 392-403, 1995), chain shuffling (Marks et al., Bio / Technology, 10, 779-783, 1992), using mutant strains of E. coli (Low et al J. Mol. Biol., 250, 359-368, 1996), DNA shuffling (Patten et al Curr. Opin. Biotechnol., 8, 724-733, 1997), phage display (Thompson et al., J. Mol. Biol., 256, 77-88, 1996), and sex-specific PCR (Crameri et al Nature, 391, 288-291, 1998). Vaughan et al. (supra) discuss these affinity maturation methods. Binding domains for use in the present invention may be generated by any suitable method known in the art, for example, CDRs may be obtained from non-human antibodies, including commercially available antibodies, and grafted onto a human framework, or alternatively chimeric antibodies may be prepared with non-human variable regions and human constant regions, etc. Since the antibodies of the present invention may not be in a naturally occurring format, such screening may be used to identify antibody antigen-binding sites that have the desired properties for binding TNFR2, which will then be reformatted so that they are in one of the formats described herein.

[0097] A person skilled in the art may use any suitable method known in the art to generate antibodies for use in the antibodies of the present invention. Antigen polypeptides for use in generating antibodies, e.g., for use in immunizing a host or for use in panning, e.g., phage display, may be prepared from genetically engineered host cells comprising an expression system by processes well known in the art, or they may be recovered from natural biological sources. In this application, the term "polypeptide" includes peptides, polypeptides, and proteins, which are used interchangeably unless otherwise specified. An antigen polypeptide may in some instances be part of a larger protein, such as a fusion protein fused to an affinity tag or the like. In one embodiment, a host may be immunized with a cell transfected with TNFR2, e.g., a cell expressing TNFR2 on its surface.

[0098] If immunization of an animal is required, antibodies generated against the antigenic polypeptide may be obtained by administering the polypeptide to an animal, preferably a non-human animal, using well-known and conventional protocols, see, for example, Handbook of Experimental Immunology, DM Weir (ed.), Vol 4, Blackwell Scientific Publishers, Oxford, England, 1986). Many warm-blooded animals may be immunized, such as rabbits, mice, rats, sheep, cattle, camels, or pigs. However, mice, rabbits, pigs, and rats are generally most preferred. Monoclonal antibodies can be prepared by any method known in the art, such as the hybridoma technique (Kohler & Milstein, 1975, Nature, 256:495-497), the trioma technique, the human B cell hybridoma technique (Kozbor et al 1983, Immunology Today, 4:72), and the EBV hybridoma technique (Cole et al Monoclonal Antibodies and Cancer Therapy, pp77-96, Alan R Liss, Inc., 1985). Antibodies can also be produced using single lymphocyte antibody techniques, for example, by cloning and expressing immunoglobulin variable region cDNA generated from a single lymphocyte selected for production of specific antibodies by the methods described in Babcook, J. et al 1996, Proc. Natl. Acad. Sci. USA 93(15):7843-7848l; WO 92 / 02551; WO 2004 / 051268, and WO 2004 / 106377.Antibodies for use in the present invention can also be generated using various phage display methods known in the art, see, for example, Brinkman et al. (in J. Immunol. Methods, 1995, 182: 41-50), Ames et al. (J. Immunol. Methods, 1995, 184:177-186), Kettleborough et al. (Eur. J. Immunol. 1994, 24:952-958), Persic et al. (Gene, 1997 187 9-18), Burton et al. (Advances in Immunology, 1994, 57:191-280), and WO 90 / 02809; WO 91 / 10737; WO 92 / 01047; WO 92 / 18619; WO 93 / 11236; WO 95 / 15982; WO 95 / 20401; and U.S. Pat. Nos. 5,698,426; 5,223,409; and 5,403,484. Nos. 5,580,717; 5,427,908; 5,750,753; 5,821,047; 5,571,698; 5,427,908; 5,516,637; 5,780,225; 5,658,727; 5,733,743; 5,969,108, and WO 20011 / 30305.

[0099] In one example, an antibody of the invention is fully human, in particular one or more of the variable domains are fully human. A fully human molecule is one in which the variable and constant regions (if present) of both the heavy and light chains are all of human origin or are substantially identical to sequences of human origin, although not necessarily from the same antibody. Examples of fully human antibodies may include, for example, antibodies produced by the phage display methods described above, as well as antibodies produced in mice in which the mouse immunoglobulin variable region and optionally the constant region genes have been replaced with their human counterparts, e.g., the general terms are described in EP 0546073, U.S. Pat. No. 5,545,806, U.S. Pat. No. 5,569,825, U.S. Pat. No. 5,625,126, U.S. Pat. No. 5,633,425, U.S. Pat. No. 5,661,016, U.S. Pat. No. 5,770,429, EP 0438474, and EP 0463151, which are incorporated by reference.

[0100] In one example, the antigen-binding site of the antibody according to the invention, particularly the variable region, is humanized. Humanized (including CDR-grafted antibody) as used herein refers to a molecule having one or more complementarity determining regions (CDRs) of a non-human species and framework regions of a human immunoglobulin molecule (see, for example, U.S. Pat. No. 5,585,089; WO 91 / 09967, which are incorporated herein by reference). It is recognized that this may only be necessary to transfer the specificity determining residues of a CDR, and not the entire CDR (see, for example, Kashmiri et al., 2005, Methods, 36, 25-34). Although a preferred embodiment, the entire CDR or multiple CDRs are grafted. A humanized antibody may optionally further comprise one or more framework residues from the non-human species from which the CDR is derived. As used herein, the term "humanized antibody molecule" refers to an antibody molecule containing one or more CDRs (including one or more modified CDRs, if desired) from a donor antibody (e.g., a murine monoclonal antibody) whose heavy and / or light chains have been grafted onto the heavy and / or light chain variable region framework of an acceptor antibody (e.g., a human antibody). For review, see Vaughan et al, Nature Biotechnology, 16, 535-539, 1998. In one embodiment, only one or more specificity-determining residues from any one of the CDRs described herein above are transferred to the human antibody framework, rather than the entire CDR being transferred (see, e.g., Kashmiri et al., 2005, Methods, 36, 25-34). In one embodiment, only the specificity-determining residues from one or more of the CDRs described herein above are transferred to the human antibody framework. In another embodiment, only the specificity-determining residues from each of the CDRs described herein above are transferred to the human antibody framework.

[0101] When grafting CDRs or specificity determining residues, any suitable acceptor variable region framework sequence may be used for the class / type of donor antibody from which the CDRs are derived, including mouse, primate, and human framework regions. Suitably, the humanized antibody according to the present invention has a variable domain comprising a human acceptor framework region and one or more of the CDRs provided herein. Examples of human frameworks that can be used in the present invention are KOL, NEWM, REI, EU, TUR, TEI, LAY, and POM (Kabat et al., supra). For example, KOL and NEWM can be used for the heavy chain, REI can be used for the light chain, and EU, LAY, and POM can be used for both the heavy and light chains. Alternatively, human germline sequences may be used; these are available at http: / / www2.mrc-lmb.cam.ac.uk / vbase / list2.php.

[0102] In the humanized antibody molecule of the present invention, the acceptor heavy and light chains do not necessarily have to be derived from the same antibody, and may, if desired, comprise composite chains with framework regions derived from different chains. The framework regions do not necessarily have to have the exact same sequence as the acceptor antibody. For example, unusual residues may be replaced with residues that occur more frequently for that class or type of acceptor chain. Alternatively, selected residues in the acceptor framework regions may be altered so that they correspond to residues found at the same positions in the donor antibody (see Reichmann et al 1998, Nature, 332, 323-324). Such alterations should be the minimum necessary to restore the affinity of the donor antibody. Protocols for selecting residues in the acceptor framework regions that may need to be altered are described in WO 91 / 09967. Derivatives of the framework may have 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more amino acids replaced with alternative amino acids, e.g., donor residues. Donor residues are those residues from the donor antibody, i.e. the antibody from which the CDRs were originally derived, specifically residues at corresponding positions from the donor sequence are taken. Donor residues may be replaced by suitable residues (acceptor residues) from the human acceptor framework.

[0103] Residues in antibody variable domains are conventionally numbered according to a system devised by Kabat et al., 1987, in Sequences of Proteins of Immunological Interest, US Department of Health and Human Services, NIH, USA (hereinafter "Kabat et al. (supra)"). This numbering system is used herein unless otherwise indicated. The Kabat residue designations do not necessarily correspond directly to the linear numbering of amino acid residues. The actual linear amino acid sequence may contain fewer or additional amino acids than the strict Kabat numbering, but this corresponds to the shortening or insertion of structural components other than the framework or complementarity determining regions (CDRs) of the basic variable domain structure. The exact Kabat numbering of residues can be determined for a given antibody by alignment of the residues of homology of the antibody's sequence with the "standard" Kabat numbering sequence. The CDRs of the heavy chain variable domain are located at residues 31-35 (CDR-H1), residues 50-65 (CDR-H2), and residues 95-102 (CDR-H3) according to the Kabat numbering system. However, according to Chothia (Chothia, C. and Lesk, AMJ Mol. Biol., 196, 901-917(1987)), the loop equivalent to CDR-H1 extends from residue 26 to residue 32. Thus, unless otherwise indicated, "CDR-H1" as used herein is intended to refer to residues 26-35 as described by a combination of the Kabat numbering system and the topological loop definition of Chothia. The CDRs of the light chain variable domain are located at residues 24-34 (CDR-L1), residues 50-56 (CDR-L2), and residues 89-97 (CDR-L3) according to the Kabat numbering system.

[0104] In one embodiment, the invention extends to the antibody sequences disclosed herein, hi another embodiment, the invention extends to humanized versions of such antibodies.

[0105] One skilled in the art can test the variant or humanized sequences of the CDRs in any suitable assay, such as those described herein, to ensure that activity is maintained.

[0106] In one embodiment, variant binding molecules can be identified by identifying such binding molecules that can cross-block the specific binding molecules described herein. Cross-blocking antibodies can be identified using any suitable method in the art, for example, by using competitive ELISA or BIAcore assays, where the binding of the cross-blocking antibody to the antigen (TNFR2) prevents the binding of the antibody of the present invention, or vice versa. Such cross-blocking assays may use cells expressing TNFR2 as targets. In one embodiment, flow cytometry is used to evaluate binding to cells expressing TNFR2.

[0107] Variant binding molecules may be used that still retain the desired properties of the binding molecules of the present invention.Therefore, antibodies that have a certain degree of sequence identity to binding molecules and in particular to the specific molecules described herein are also provided.Sequence identity may be over the entire length of the polypeptide, or may relate to a specific portion, such as just over the variable region or just over the CDR sequence. The degree of identity and similarity can be readily calculated (Computational Molecular Biology, Lesk, AM, ed., Oxford University Press, New York, 1988; Biocomputing. Informatics and Genome Projects, Smith, DW, ed., Academic Press, New York, 1993; Computer Analysis of Sequence Data, Part 1, Griffin, AM, and Griffin, HG, eds., Humana Press, New Jersey, 1994; Sequence Analysis in Molecular Biology, von Heinje, G., Academic Press, 1987, Sequence Analysis Primer, Gribskov, M. and Devereux, J., eds., M Stockton Press, New York, 1991, using the BLAST™ software available from NCBI (Altschul, SF et al., 1990, J. Mol. Biol. 215:403-410; Gish, W. & States, DJ 1993, Nature Genet. 3:266-272. Madden, TL et al., 1996, Meth. Enzymol. 266:131-141;Altschul, SF et al., 1997, Nucleic Acids Res. 25:3389-3402; Zhang, J. & Madden, TL 1997, Genome Res. 7:649-656).The present invention also extends to the polypeptide sequences disclosed herein and sequences that are at least 80% similar or identical thereto, such as sequences that have 85% or more, such as 90% or more, particularly 95%, 96%, 97%, 98%, or 99% or more similarity or identity. In one embodiment, the variant has at least 80% sequence identity. In another embodiment, the variant has at least 90% sequence identity. In another embodiment, the variant has at least 95% sequence identity. In one embodiment, the sequence may have at least 99% sequence identity to at least one of the specific sequences provided herein. Whenever a variant is referred to herein, the variant may have such a degree of sequence identity / "identity" as used herein indicates that at any particular position in the aligned sequences, the amino acid residue is identical between the sequences. "Similarity" as used herein indicates that at any particular position in the aligned sequences, the amino acid residue is of a similar type of residue between the sequences. For example, leucine may be substituted for isoleucine or valine. Other amino acids that can often be substituted for one another include, but are not limited to, the following: - Phenylalanine, tyrosine, and tryptophan (amino acids with aromatic side chains); - lysine, arginine, and histidine (amino acids with basic side chains); - Aspartic acid and glutamic acid (amino acids with acidic side chains); - asparagine and glutamine (amino acids with amide side chains); and - Cysteine ​​and methionine (amino acids with sulfur-containing side chains).

[0108] In one embodiment, the variant may have 1-10, e.g. 1, 2, 3, 4, 5, or up to those values ​​of amino acid sequence changes, or at least those values, or up to those values ​​of changes, as long as the variant is still capable of specifically binding to TNFR2, preferably acting as its agonist in the format previously described. In another embodiment, the antibody of the invention may have at least 5, 6, 7, 8, 9, 10, 11, or 12 amino acid sequence changes compared to one CDR of the specific antibody described herein, e.g., that number of sequence changes in the six CDRs that make up the antigen binding site. In one embodiment, when the binding molecule comprises a VHH domain, the VHH domain may comprise a set of three CDRs having the above number of amino acid changes compared to the specific set of three CDRs for the VHH provided herein. The antibody of the invention may have that number of sequence changes in the CDRs compared to the specific antibody described herein. It may have up to that number of sequence changes. It may have at least that number of amino acid sequence changes. Such variant antibody molecules typically retain the ability to specifically bind to TNFR2. They may also retain other functions or more as described herein.

[0109] In a preferred embodiment, when variant pairs of variable regions are used, they have at least 90% sequence identity with the specific pairs described herein. In another embodiment, they have at least 90% sequence identity. In another embodiment, they have at least 95% sequence identity. In one embodiment, they have 10 or fewer amino acid sequence changes. In another embodiment, they have 5 or fewer amino acid sequence changes compared to the specific sequence. In one embodiment, they have 1, 2, or 3 sequence changes compared to the specific sequence. The variants retain the binding ability of TNFR2. For example, the variants may have at least 50% of the binding activity of the specific sequence. In another embodiment, the variants have at least 75% of the binding activity. In one embodiment, the variants have at least the same binding activity. In one embodiment, the sequence changes are only in the framework regions. In another embodiment, the sequence changes are conservative amino acid changes.

[0110] In the embodiment of identifying the CDR set, a variant set may be used. For example, a set with a total of up to 10 amino acid changes. For example, a variant with a maximum of 5 amino acid sequence changes. In one embodiment, there may be 1, 2, or 3 amino acid changes. The amino acid sequence changes may be all conservative in one embodiment. The variant CDR still allows the antigen binding site to bind to TNFR2. For example, the variant may have at least 50% of the binding activity of the specific sequence. In another embodiment, they have at least 75% of the binding activity. In one embodiment, they have at least the same binding activity.

[0111] In one embodiment, the variant sequence has the ability to block binding of a binding site that contains a specific sequence. In one embodiment, the variant sequence competes for binding. The examples of this application include exemplary assays that can be used to measure binding and thus blocking / competition.

[0112] It is recognized that this aspect of the invention also extends to variants of the specific binding molecules, particularly humanized and modified forms, including antibodies in which amino acids have been isomerized, deamidated, glycosylated, or mutated in the CDRs to remove cysteine ​​residues as described herein above.

[0113] In preferred embodiments, sequence identity is referred to, it is related to the full length of the two sequences being compared.For example, the sequence identity level specified can be over the length of CDR.It can be over the length of variable region.It can be over the full length of sequence, for example over the full length of the shorter of the two sequences.

[0114] Whenever a specific sequence is indicated, variants may be used, such as those defined above or below. Variant CDRs may be used, which may be a variant CDR set. Variant variable regions may be used, which may be pairs of such variable regions. Variant hinge regions may be used. Variant constant regions may be used. In one embodiment, the variants may be variants of the entire polypeptide. In one embodiment, such variants have at least 90% sequence identity compared to the specific sequence. In one embodiment, they have at least 95% sequence identity. The variant antigen binding sites still retain TNFR2 binding activity. The variants may retain any of the functions described herein. The variant binding molecules typically retain TNFR2 agonist activity. The variants may retain any of the functions described herein. Preferred variants have 1-10 amino acid changes. Further variants have 1-5 amino acid changes. Preferred variants have 1, 2, or 3 amino acid changes. Other variants have only one amino acid change compared to the specific sequence. In one embodiment, the variant may be a biosimilar of the binding molecules described herein.

[0115] Further Exemplary Antibodies and CDR Sequences The examples of this application include various specific antibody molecules, including molecules with TNF ligands as part of their structure, to help illustrate the invention and its advantages. The invention is not limited to those sequences. In one embodiment, the binding molecules, particularly antibodies, of the invention can include specific sequences from those exemplary antibodies or their variants. Variant antibodies are also provided, including variants as defined herein, particularly those that specifically bind to TNFR2 but not to TNFR1 in an FcγR-independent manner.

[0116] Table 3 of the present invention describes various pairs of expression constructs used to express specific antibodies. The table also describes the polypeptides expressed by the vector pairs. In one embodiment, the present invention provides a binding molecule comprising one of the matched combinations of polypeptides described in Table 3. The original antibody clone used to generate the non-VHH-based antibodies in the examples is a murine antibody. In one embodiment, the present invention provides a binding molecule comprising one of the matched combinations of polypeptides described in Table 3, where the mouse CDR sequences are retained but the remainder of the antibody is humanized. In another embodiment, the binding molecule of the present invention comprises one of the matched combinations of polypeptides described in the table, except that the entire mouse sequence is replaced with the sequence of a human antibody capable of specifically binding to TNFR2. In one embodiment, a binding molecule is used that cross-blocks the binding molecule resulting from the matched combination of polypeptides described in Table 3. In another embodiment, a binding molecule is used that cross-reacts with the binding molecule resulting from the matched combination of polypeptides described in Table 3. In one embodiment, the binding molecule of the present invention, particularly an antibody, comprises an antigen-binding site comprising a set of six CDRs from those matched combinations of heavy and light chains described in Table 3 of the present application. In another embodiment, a binding molecule, particularly an antibody, of the invention comprises one pair of light and heavy chain variable region pairs from the pairs set forth in Table 3.

[0117] Table 4 of the present application lists the SEQ ID NOs of various polypeptides expressed by the expression vectors used in the examples of the present application. In one embodiment, the binding molecule of the present invention comprises at least one of those polypeptides. In another embodiment, the binding molecule of the present invention comprises at least one of those polypeptides, which is humanized and retains the mouse CDR sequences, but replaces the remainder of the mouse sequences with human sequences. In a preferred embodiment, the binding molecule may comprise two of those polypeptides, or a humanized version, in which all but the mouse CDRs are replaced with the corresponding sequences. In one embodiment, the binding molecule of the present invention, particularly the antibody, comprises an antigen-binding site comprising a set of six CDRs from those of the corresponding combination of heavy and light chain SEQ ID NOs listed in Table 4 of the present application. In another embodiment, the binding molecule of the present invention, particularly the antibody, comprises one pair of light and heavy chain variable region pairs from the variable regions in the SEQ ID NOs listed in Table 4.

[0118] Table 5 of the present application lists CDR sequences from specific antibodies used as a starting point for the antibody formats described in the present application. The three CDRs from the light chain are grouped together, as are the three CDRs from the heavy chain, and the light and heavy chains are grouped together. In one embodiment, the binding molecule of the present invention comprises a set of three light chain CDRs as listed in Table 5. In another embodiment, it comprises a set of three heavy chain CDRs as listed in Table 6, and thus comprises both light and heavy chain CDR sequences from one of the clones or functional variants of such sequences. In one embodiment, the binding molecule of the present invention comprises CDRs for more than one specificity, and thus the binding molecule may comprise, for example, two polypeptides that collectively comprise six CDRs from one of the antibodies listed in Table 5, but the binding molecule may comprise another additional set of six CDRs of an antibody, for example present as an scFv that is part of the binding molecule. In one embodiment, the binding molecule of the present invention is triparatopic with respect to TNFR2, and has three different specificities for TNFR2. In one embodiment, at least one of the specificities is conferred by a set of six CDRs from Table 5 and variants of those CDRs. In another embodiment, at least two of the specificities are conferred by a set of six CDRs from the table, or variants thereof. In a further embodiment, all three of the specificities are similar. In another embodiment, the binding molecules of the invention may be tetraparatopic, e.g., at least one, two, or three specificities are conferred by a set of six CDRs listed in Table 5, or variants thereof. In one embodiment, all four specificities may be conferred by a set of six CDRs from Table 5, or variants thereof.

[0119] In a preferred embodiment, the binding molecule of the invention comprises an antigen-binding site comprising a set of six CDRs from the C40 clone. In another preferred embodiment, the binding molecule comprises an antigen-binding site comprising a set of six CDRs from the C19 clone. In a further preferred embodiment, the antibody has both an antigen-binding site comprising a set of six CDRs from the C40 clone and an additional antigen-binding site also comprising a set of six CDRs from the C19 clone. In a preferred embodiment, the CDRs are derived from the C4 clone. In one embodiment, all antigen-binding sites of the antibody of the invention comprise the same set of six CDRs. In another embodiment, different antigen-binding sites of the antibody comprise different sets of six CDRs. In one embodiment, instead of the six CDRs from C40, C19, or C4, a variant set of six CDRs may be used, such as any variant discussed herein.

[0120] In one embodiment, any antibody format discussed herein may be provided in which the six CDRs of the antigen-binding site are the CDRs of C40. In another embodiment, they are the CDRs of C19. In another embodiment, they are the CDRs of C4. In another embodiment, the CDRs are variants of C40, C19, or C4 that still retain the ability to bind to TNFR2.

[0121] The present invention also provides variants of the above binding molecules and specific binding molecules used in the examples. For example, in one embodiment, the variant may have 1, 2, 3, 4, 5, 6, or more amino acid sequence changes in the CDR sequence of the antigen binding site compared to the six CDRs listed in Table 5 as a set and still act as an agonist of TNFR2. In another embodiment, the variant may have six or more such amino acid changes, for example, the variant may have 6 to 15 such changes, for example, 6, 7, 8, 9, 10, or more such changes. In one embodiment, the antibody of the present invention may have at least the number of the specified amino acid sequence changes. In another embodiment, the antibody of the present invention may have up to the recited number. In a further embodiment, the antibody of the present invention may have the recited number of amino acid sequence changes. In one embodiment, the sequence changes may be conservative amino acid sequence changes. In another embodiment, the sequence changes may include non-conservative sequence changes. A variety of specific modifications are discussed herein that may be present in such variant molecules. For any embodiment described herein where a particular antibody format is mentioned, in a further embodiment there may be provided or used a binding molecule which is an antibody of that format from the examples of the present application but which has been modified to replace all non-murine CDR sequences with human CDR sequences.

[0122] Preferred C4 and C4 variant antigen-binding sites In a particularly preferred embodiment, the binding molecule of the present invention comprises an antigen-binding site comprising the sequence of the C4 clone or a variant sequence thereof. For example, the binding molecule may comprise an antigen-binding site comprising the light and heavy chain variable regions from the C4 clone. The binding molecule may comprise six CDRs of the C4 clone. Variant C4 variable regions and CDR sequences may also be used with the present application, which describes preferred examples thereof.

[0123] The light and heavy chain variable regions of the C4 clone are set forth in SEQ ID NOs: 295 and 296, respectively, and may be used in the antigen-binding site. The light chain variable region CDR1, CDR2, and CDR3 sequences of the C4 clone are set forth in SEQ ID NOs: 69 / 70 / 71, respectively, and the three CDRs may be used in the antigen-binding site. The heavy chain variable region CDR1, CDR2, and CDR3 sequences of the C4 clone are provided as SEQ ID NOs: 102 / 103 / 104, respectively, and the three CDRs may be used in the antigen-binding site. In a preferred embodiment, the antigen-binding site comprises all six CDRs from the CDRs, thus comprising the light chain CDRs of SEQ ID NOs: 69 / 70 / 71 and the heavy chain CDRs of SEQ ID NOs: 102 / 103 / 104. Variant C4 sequences may also be used. In one embodiment, a variant set of six of those CDRs may be used, which has up to 10 amino acid sequence changes compared to the sequence of the specific CDR. In one embodiment, the variant set has up to 5 amino acid sequence changes. In one embodiment, the set of variants may have 1, 2, or 3 amino acid sequence changes compared to the sequences of a specific set of CDR6s.

[0124] Particularly preferred C4 variant light chain variable regions may be provided and used as SEQ ID NOs: 262-265. CDRs from those light chain variable regions may be provided and used as SEQ ID NOs: 266-277. The variant CDR set may have up to 10 amino acid sequence changes compared to the sequence of the specific light chain variable CDR. For example, the variant set may have up to 5 amino acid sequence changes compared to the specific light chain variable CDR. Particularly preferred heavy chain variable regions may be set forth and used as SEQ ID NOs: 278-281. CDRs from those heavy chain variable regions may be set forth and used as SEQ ID NOs: 282-293. Variants with up to 10 amino acid sequence changes with respect to the heavy chain CDRs may be used. In one embodiment, a variant set with up to 5 amino acid sequence changes compared to the sequence of the heavy chain CDR set may be used.

[0125] In a preferred embodiment, the binding molecule of the invention comprises an antigen-binding site comprising a light chain variable region selected from one of SEQ ID NOs: 295, 262-265, or any variant thereof, and a heavy chain variable region selected from one of SEQ ID NOs: 306, 278-281, or any variant thereof. In a preferred embodiment, the binding molecule comprises an antigen-binding site comprising a light chain variable region comprising CDR1, CDR2, and CDR3 of SEQ ID NOs: 69, 70, and 71, or a variant thereof, and a heavy chain variable region having CDR1, CDR2, and CDR3 of SEQ ID NOs: 102, 103, and 104, or a variant thereof. In a further preferred embodiment, the binding molecule comprises an antigen-binding site comprising a set of three light chain CDRs for CDR1 / CDR2 / CDR3 selected from SEQ ID NOs: 266 / 267 / 268; 269 / 270 / 271; 272 / 273 / 274; 275 / 276 / 278, or a variant thereof. In a further preferred embodiment, the binding molecule comprises an antigen-binding site comprising a set of three heavy chain CDRs for CDR1 / CDR2 / CDR3 selected from SEQ ID NO: 282 / 283 / 284;285 / 286 / 287;288 / 289 / 290;291 / 292 / 293, or variants thereof. In one embodiment, the binding site comprises both such light and heavy chain CDRs. For example, in one embodiment, the binding molecule comprises the light chain, heavy chain, or both CDRs from one of the specific C4 variants provided herein.

[0126] C4 and C4 variant-based binding sites can be used in any of the antibody formats discussed herein, including antigen binding sites that include light and heavy chain variable regions. In a preferred embodiment, such antigen binding sites are used in tetravalent binding molecules of the present invention. In a particularly preferred embodiment, a set of six CDRs or a VL and VH sequence pair from a C4 clone or one of the specific variants of the C4 clone provided herein is used in one of the formats shown in the figures, particularly in Figures 1 and 7. They can also be used in the format shown in Figure 29A. Variants of specific sequences can also be used.

[0127] In a particularly preferred embodiment of the invention, the binding molecule of the invention comprises a polypeptide having the sequence of SEQ ID NO: 259 and a polypeptide having the sequence of SEQ ID NO: 261. In one embodiment, the binding molecule consists of two of each such polypeptide. In a further preferred embodiment, the binding molecule comprises two such polypeptides, but the light chain variable region of SEQ ID NO: 261 is exchanged for a variable region of one of SEQ ID NOs: 262-265 or a variant thereof. In a further preferred embodiment, the binding molecule comprises two such polypeptides, but the heavy chain variable region of SEQ ID NO: 259 is exchanged for a variable region of one of SEQ ID NOs: 278-281 or a variant thereof. In a preferred embodiment, both the heavy and light chain variable regions are so exchanged.

[0128] In a further preferred embodiment, the binding molecule comprises a polypeptide having a sequence as set forth in Figure 54 or 55. In one embodiment, the binding molecule comprises (a) a polypeptide having a sequence set forth in SEQ ID NO: 332, or a variant sequence having at least 90% sequence identity; and (b) a polypeptide having the sequence of SEQ ID NO: 333 or a variant sequence having at least 90% sequence identity. Includes.

[0129] In one embodiment, the binding molecule comprises two polypeptides as defined in (a) and two polypeptides as defined in (b).

[0130] In one embodiment, the binding molecule is a binding molecule that comprises the light and heavy chain variable regions present in SEQ ID NOs: 332 and 333, or variant sequences thereof having at least 90% sequence identity, and is capable of acting as a TNFR2 agonist.

[0131] In one embodiment, the binding molecule is (c) a polypeptide having the sequence of SEQ ID NO: 334 or a variant sequence having at least 90% sequence identity; and (d) a polypeptide having the sequence of SEQ ID NO: 335 or a variant sequence having at least 90% sequence identity. Includes.

[0132] In one embodiment, the binding molecule comprises two polypeptides as defined in (a) and two polypeptides as defined in (b).

[0133] In one embodiment, the binding molecule is a binding molecule that comprises the light and heavy chain variable regions present in SEQ ID NOs: 334 and 335, or variant sequences thereof having at least 10% sequence identity, and is capable of acting as a TNFR2 agonist.

[0134] In one embodiment, the binding molecule is (e) a polypeptide having the sequence of SEQ ID NO: 336 or a variant sequence having at least 90% sequence identity; and (f) a polypeptide having the sequence of SEQ ID NO: 337 or a variant sequence having at least 90% sequence identity. Includes.

[0135] In one embodiment, the binding molecule comprises two polypeptides as defined in (a) and two polypeptides as defined in (b).

[0136] In one embodiment, the binding molecule is a binding molecule that comprises the light and heavy chain variable regions present in SEQ ID NOs: 336 and 337, or variant sequences thereof having at least 10% sequence identity, and is capable of acting as a TNFR2 agonist.

[0137] In a preferred embodiment, all of the antigen binding sites specific for TNFR2 in a binding molecule of the invention are one of the C4 or C4 variant based binding sites discussed herein.

[0138] Table 10 lists examples of preferred light and heavy chain pairings for C4 variant-binding antigen-binding sites. In a preferred embodiment, the pairs of sequences shown in Table 10 can be used as antigen-binding sites. In another embodiment, a pair of sequences with at least 90% sequence identity with one of the pairs of sequences can be used. A variant with 95% sequence identity can be used.

[0139] The C4 and specific C4 variants described herein may be used in any of the binding formats described herein, as well as their CDR variants, light chain CDR sets, heavy chain CDR sets, CDR6 sets, light chain variable regions, heavy chain variable regions, and variable region pairs.Specific pairs of C4 variants that can be used are described in Table 10, and the variant sequences of those specific pairs can also be used.Figures 54 and 55 show particularly preferred C4 variant-based polypeptides that can be used, particularly heavy and light chains.

[0140] Preferred VHH sequences In a particularly preferred embodiment, the TNFR2 binding domain of the binding molecule of the invention is a VHH domain. Thus, in a preferred embodiment, all of the TNFR2 binding domains in the binding molecule are VHH domains. The examples of the present application describe the generation of 14 VHH regions specific for TNFR2, the sequences of which are provided in SEQ ID NOs: 140-153. Each of those specific VH regions represents a preferred VHH domain for use in the binding molecules of the invention, including any of the binding molecule formats described herein. Variant sequences of such specific VHH domains may be used, so long as the variants are still capable of binding to TNFR2 and the binding molecule acts as an agonist of TNFR2.

[0141] Particularly preferred VHH domains used in binding molecules are C188 VHH binding domains of SEQ ID NO: 140 or variants thereof. The three CDRs of C188 are provided as SEQ ID NO: 154 / 155 / 156. In a preferred embodiment, the binding molecule comprises an antigen binding site that comprises these three CDRs. In a preferred embodiment, the antigen binding site specific for TNFR2 comprises all three CDRs. In addition, variant CDRs may be used, as long as the binding site can still bind to TNFR2.

[0142] As described in the examples of the present application, specific variant sequences of C188 have been generated as C188-VHH_1 to C188-VHH_15, the sequences of which are set forth in SEQ ID NOs: 198 to 212, respectively. Thus, in one embodiment, the TNFR2 binding domain of the binding molecule of the present invention has one VHH domain of SEQ ID NOs: 198 to 212. Preferably, all TNFR2 binding domains have the same sequence. In a particularly preferred embodiment, the variant has one sequence of SEQ ID NOs: 198 to 211. Again, preferably, all binding domains have the same sequence. In one embodiment, a variant VHH binding domain having at least 90% sequence identity with one of SEQ ID NOs: 198 to 211 may be used. In another embodiment, the sequence identity level is at least 95%. Such variants can still act as TNFR2 agonists.

[0143] In a further preferred embodiment, the binding molecule comprises an antigen binding site which is one of the specific C188 VHH variants provided as SEQ ID NOs: 198-211 or a variant of such sequence. In one embodiment, the TNFR2 specific binding sites are all such VHH variant binding domains. In one embodiment, the binding molecule comprises a set of three CDRs from one of those C188 variant VHH binding domains. For example, in one embodiment, the binding molecule comprises an antigen binding domain comprising a set of three CDRs selected from: 213 / 214 / 215; 216 / 217 / 218; 219 / 220 / 221; 222 / 223 / 224; 225 / 226 / 227; 227 / 229 / 230; 231 / 232 / 233; 234 / 235 / 236; 237 / 238 / 239; 240 / 241 / 242; 243 / 244 / 245; 246 / 247 / 248; 249 / 250 / 251; 252 / 253 / 254; 255 / 256 / 257; or a variant set thereof. In a preferred embodiment, the binding molecule comprises an antigen binding domain comprising a set of three CDRs selected from: 213 / 214 / 215; 216 / 217 / 218; 219 / 220 / 221; 222 / 223 / 224; 225 / 226 / 227; 227 / 229 / 230; 231 / 232 / 233; 234 / 235 / 236; 237 / 238 / 239; 240 / 241 / 242; 243 / 244 / 245; 246 / 247 / 248; 249 / 250 / 251; 252 / 253 / 254; or a variant set thereof, wherein the variant has up to 10 amino acid sequence changes overall, yet the VHH is still capable of binding to TNFR2. In one embodiment, the variant has up to five amino acid sequence changes and is still able to bind to TNFR2.

[0144] In a particularly preferred embodiment, the binding molecule using such a VHH domain is tetravalent or hexavalent binding for such a VHH, for example one of the formats described herein. The specific VHH regions of SEQ ID NOs: 140-153 and 198-212 may be used in any binding molecule format described herein, including those shown in the figures and examples. In one embodiment, the specific VHH regions of SEQ ID NOs: 140, 142-145, 149, 151, and 153 may be used in any binding molecule format described herein, including those shown in the figures and examples. In a particularly preferred embodiment, the binding molecule may comprise an antigen binding site comprising a VHH having a variable region selected from one of those of SEQ ID NOs: 198-211, or a variant thereof having at least 90% sequence identity that is still capable of binding to TNFR2. In one embodiment, one of the binding molecule formats in the examples or figures may be modified to use such an antigen binding site.

[0145] Figures 40 and 41 illustrate particularly preferred formats of VHH antibodies. The hexavalent format shown is particularly preferred. Any of the VHH domains described herein or a set of three CDRs from those VHH domains may be used in VHH-based binding molecules, particularly the C188 or specific C188 variant binding molecules described herein. A particularly preferred format is the 3xVHH-Fc format, optionally with a linker between the VHH binding domains. A particularly preferred binding molecule comprises two polypeptides of SEQ ID NO: 197. A further particularly preferred binding molecule comprises two polypeptides of a sequence corresponding to SEQ ID NO: 197, except that the VHH domain is one of SEQ ID NOs: 198-212. A further particularly preferred binding molecule comprises two polypeptides of a sequence corresponding to SEQ ID NO: 197, except that the VHH domains are different and comprise a set of three CDRs from one of SEQ ID NOs: 213-257.

[0146] In one embodiment, the binding molecule comprises a polypeptide having the amino acid sequence of SEQ ID NO: 331 or a variant thereof having at least 90% sequence identity. In another embodiment, the binding molecule comprises three such polypeptides. In one embodiment, the binding molecule comprises a polypeptide that is a variant of the amino acid sequence of SEQ ID NO: 331 with a different hinge region, the variant having at least 90% amino acid sequence identity. In another embodiment, it comprises three such polypeptides. In another embodiment, the binding molecule comprises an antigen-binding site comprising a variable region present in SEQ ID NO: 331 or a variant having at least 90% sequence identity that can still act as an agonist. In one embodiment, the binding molecule comprises an antigen-binding site comprising three CDRs present in the amino acid sequence of SEQ ID NO: 331.

[0147] In one embodiment, the binding molecule comprises a variant of a specific VHH region discussed herein, such as one of the variants described above. For example, the variant may be a variant having at least 80% amino acid sequence identity with one of the specific sequences. The variant may have at least 90% sequence identity. The variant may have at least 95% sequence identity. The variant may have at least 98% sequence identity. The variant may have at least 99% sequence identity. In one embodiment, the variant has 1-10 amino acid sequence changes. In one embodiment, the variant may have 1-5 amino acid sequence changes. The variant may have 1, 2, 3, 4, or 5 amino acid sequence changes compared to the specific sequence. The variant may have 1, 2, or 3 amino acid sequence changes compared to the specific sequence. Possible variants of a given sequence are discussed elsewhere herein, and any such type of variant may be used. The variant preferably retains at least the same TNFR2 binding activity. For example, the variant may have at least the same agonist activity. In one embodiment, the variant may have at least 50% of the activity of the original molecule. In one embodiment, the variant may have at least 75%, preferably at least 80%, more preferably at least 90% of the activity of the original molecule. One preferred binding molecule comprises the sequence of SEQ ID NO: 197, although variants of that sequence may be used, such as any type of variant mentioned herein. The VHH domains of SEQ ID NOs: 198-212 are also preferred, and variants thereof may be used, including any type of variant discussed herein.

[0148] In one embodiment, the binding molecule can comprise both sdBr binding domain and other non-sdBr binding domain.In one embodiment, the binding molecule can comprise both VHH binding domains in combination with non-VHH antigen binding domains, for example, C188 and C188 variant antigen binding domain.In one embodiment, the binding molecule can comprise both C188, C188 variant binding domain, and C4 or C4 variant antigen binding domain binding domain.

[0149] The C188 and specific C188 variants described herein may be used in any of the binding formats described herein, as well as their CDRs, sets of CDRs, or variable regions. Figure 53 shows particularly preferred polypeptides that may be used, as well as polypeptides that may be variants of their specific sequences.

[0150] Further preferred binding molecules Particularly preferred binding molecule formats are tetravalent. One particular preferred VH domain for use in binding molecules, particularly tetravalent binding molecules, is the C4 VH domain. Particularly preferred formats are IgG1-HC:scFv or IgG1-HC:scFv. In a preferred format, the VH and VL domains in such formats are domains of C4. One particularly preferred format is C4-IgG1(Durv)-HC:scFvC4(G4S)4. In a particularly preferred embodiment, such binding molecules comprise two polypeptides, the light chain sequence of SEQ ID NO:261 and the heavy chain sequence of SEQ ID NO:259. In one embodiment, binding molecules are provided that comprise such polypeptides, except that the VH and / or VL regions are one of the specific such regions described herein. As described in the Examples, variants of the C4 VL sequence having the sequences set forth in SEQ ID NOs: 262-265 may be generated and any of these VL sequences may be replaced with sequences in the C4-IgG1(Durv)-HC:scFvC4(G4S)4 molecule. As described in the Examples, variants of the C4 VH sequence having the sequences set forth in SEQ ID NOs: 278-281 may be generated and any of these VH sequences may be replaced with sequences in the C4-IgG1(Durv)-HC:scFvC4(G4S)4 molecule. In one embodiment, both the VL and VH region sequences are sequences selected from SEQ ID NOs: 262-265 for the light chain and SEQ ID NOs: 278-281 for the heavy chain. In a preferred embodiment, the VL and VH sequences used are the sequences of SEQ ID NOs: 295 and 296, respectively.

[0151] Assay In one embodiment, functional assays may be used to determine whether a binding molecule of the present invention has a particular property or properties, such as any of the properties mentioned herein. Thus, functional assays may be used to evaluate a binding molecule of the present invention, particularly an antibody. In one embodiment, one or more of the assays described in the examples of this application may be used to evaluate a particular binding molecule and evaluate whether it has a desired property or properties. In a particularly preferred embodiment, one of the assays described in Example 1 may be used.

[0152] The binding molecule of the present invention can bind to TNFR2. The ability of the binding molecule of the present invention or the candidate binding molecule to bind to TNFR2 can be evaluated in a variety of ways. For example, in one embodiment, the ability of the binding molecule to bind to TNFR2 is evaluated by using TNFR2 protein, for example by using a technique such as surface plasmon resonance using TNFR2 or a part thereof bound to a chip. In a particularly preferred embodiment, the ability of the binding molecule to bind to TNFR2 is evaluated using cells expressing TNFR2 on their surface. In one embodiment, the candidate molecule is labeled and then screened for its ability to bind to cells expressing TNFR2 using a technique such as ELISA or flow cytometry. In another embodiment, the candidate molecule may be incubated with cells expressing TNFR2, and then the bound candidate molecule may be detected using a secondary agent, for example a labeled antibody specific for the species of the candidate molecule. In one embodiment, the binding molecules of the invention are labeled, for example, using luciferase tagging (e.g., Gaussia princeps luciferase (GpL)) variants of the binding molecules, particularly antibodies or fusion proteins, as described, for example, in Kums et al., MAbs. 2017 Apr;9(3):506-520. Such tagged binding molecules may also be used in competitive binding assays.

[0153] In one embodiment, the binding molecule of the present invention can (a) oligomerize TNFR2 on the surface of cells expressing TNFR2, preferably forming higher-order clusters of TNFR2 hexamers or TNFR2 trimers on the cell surface; (b) induce TNFR2 signaling; and / or (c) stimulate the proliferation of leukocytes, preferably T cells, more preferably Treg cells.In one embodiment, the binding molecule of the present invention can bind to the existing higher-order complexes of TNFR2 receptors on the cell surface, particularly to the trimers of TNFR2 on the cell surface.For example, in one embodiment, the binding molecule of the present invention can cause the formation of superclusters of TNFR2 receptors on the same cell, or bridging two cells, or both.

[0154] In one embodiment, the candidate molecule is measured for its ability to cause oligomerization of individual TNFR2 molecules on the surface of cells expressing TNFR2. In one embodiment, the binding molecule of the present invention can cause such oligomerization. For example, in one embodiment, the binding molecule of the present invention can cause the formation of hexamers or other oligomers of TNFR2 trimers. In one embodiment, such oligomerization can be detected using high-resolution microscopy. In one embodiment, the oligomerization of TNFR2 can be detected using techniques such as PALM, dSTQRM, chemical cross-linking, or FRET.

[0155] In a preferred embodiment, the ability of the binding molecules of the present invention to activate TNFR2 is evaluated by measuring the downstream effects of TNFR2 activation, rather than directly measuring the binding itself. For example, in one embodiment, whether the binding molecule induces IL-8 production may be used as a way to establish whether a given binding molecule can induce TNFR2 signaling. NFκB activation in response to TNFR2 signaling typically results in the production of IL-8, which may be used as a way to evaluate whether a given binding molecule can activate TNFR2 signaling in a desired embodiment, and whether the binding molecule of the present invention can activate such signaling. In one embodiment, a molecule known to activate TNFR2 signaling, such as the binding molecule of the present invention, may be used as a positive control. An example of a preferred cell line that can be used to evaluate the ability of a binding molecule to bind to TNFR2 and stimulate IL-8 production is HT1080-Bcl2-TNFR2. In one embodiment, the functional assay involves contacting the binding molecule to be evaluated with cells expressing TNFR2, e.g., HT1080-Bcl2-TNFR2 cells, incubating the two overnight, and then analyzing the supernatant samples for the presence of IL-8. In one embodiment, an ELISA, e.g., the BD OptEIA™ IL8 ELISA kit, is used for such evaluation. In one embodiment, such methods may also include a positive control known to stimulate TNFR2 signaling. In one embodiment, such an assay may be used to screen one or more binding molecules for their ability to bind and activate IL-8 production, e.g., to screen a library of potential binding molecules to identify binding molecules that activate IL-8 production. In one embodiment, such an assay may be used to check whether variants of the binding molecules of the invention are still active or are more active than the specific binding molecules of the present application.

[0156] Another assay that can be used to evaluate the binding molecules of the present invention is a viability assay, where TNFR2 can induce cell death in some cell lines / cell types by inhibition of survival proteins (TRAF2, cIAP1, cIAP2) and simultaneous upregulation of endogenously produced TNF-induced TNFR1. In a preferred embodiment, Kym-1 cells (Schneider et al., 1999) are used for such evaluation. For example, cells expressing TNFR2, particularly Kym-1 cells seeded in plates or wells, may be incubated with the test binding molecule, followed by crystal violet staining, and then measuring the OD at 595 nm. A positive control known to induce cell killing, e.g. "cell death", may be used, e.g., a control mixture (one example is 200ng / ml TNF, 200ng / ml TRAIL, 200ng / ml CD95L, 25μg / ml CHX, 1% (weight / volume) sodium azide). The values ​​observed may be normalized relative to untreated cells and cells treated with the "death mixture."Again, such methods may be used to screen for binding molecules with desired properties.

[0157] In a preferred embodiment, a binding molecule of the invention induces IL-8 production, e.g. as assessed in the assays described above, and is also capable of inducing cell death, e.g. as assessed in the assays described above. In another preferred embodiment, a binding molecule of the invention induces IL-8 production but does not significantly induce cell death, e.g. as measured using the assays in the Examples of this application.

[0158] In a preferred embodiment, the binding molecule of the present invention can bind to TNFR2 and initiate signal transduction through activation of non-classical NFκB pathway. In a particularly preferred embodiment, the binding molecule of the present invention can activate non-classical NFκB pathway, which is indicated by processing of p100 to p52. In one embodiment, cells expressing TNFR2 are contacted with the binding molecule to be evaluated, and then the cells are harvested and cell lysates are evaluated to measure p100 and p52 levels. In one embodiment, Western blotting is performed on cell lysates to determine whether p100 is processed to p52 using antibodies against p100 and p52. Any suitable cells may be used, but in a preferred embodiment, Kym-1 cells are used for evaluation. Positive controls, such as binding molecules known to be active, may be used. Negative controls, such as cells incubated in the absence of binding molecules and evaluated in the same manner, may also be performed. In a preferred embodiment, the binding molecules of the invention are capable of activating the canonical NFκB pathway, as measured, for example, by IL-8 production, as well as the non-canonical NFκB pathway, for example, by assessing whether p100 is processed to p52. In another preferred embodiment, the binding molecules are also capable of activating cell killing, as assessed, for example, using the cell viability assays described above.

[0159] Typically, the binding molecule of the present invention does not bind to FcγR. In a preferred embodiment, the candidate binding molecule is evaluated not only for its ability to bind and activate TNFR2, but also for its ability not to bind and activate FcγR. In one embodiment, the binding molecule of the present invention is evaluated for its ability to bind to Fc receptors, particularly FcγR. The lack of binding to Fc receptors may be evaluated to determine, for example, whether CDC or ADCC activity is exhibited, preferably not by the binding molecule of the present invention.

[0160] In another embodiment, the ability of the binding molecules of the invention to stimulate cell activation and / or expansion, e.g. to stimulate specific immune cells, is assessed, as the binding molecules of the invention are typically capable of effecting activation and / or expansion of cells such as T cells. In one embodiment, the ability of the binding molecules of the invention to stimulate Treg cells from PBMCs is assessed. In one embodiment, the ability of the binding molecules to expand Tregs is assessed by: - isolating the PBMCs and then culturing the PBMCs (e.g., at high density, e.g., 10 million cells / ml); - Seeding PBMCs after harvesting the cells (e.g. at a lower density of 1 million cells / ml); - incubating the cells with the candidate binding molecule for 4 days; and - Performing an analysis to determine cell number The evaluation is carried out by a method including:

[0161] In a preferred embodiment, a negative control is performed in which cells are cultured without contacting with the candidate binding molecule. The cells may be evaluated using flow cytometry to stain specifically for CD4+CD25+FoxP3+ cells. In a preferred embodiment, the number of FoxP3+CD25+ cells within the CD3+CD4+ cell population is measured. The cells may also be stained with antibodies specific for CD3 and / or CD8. In a preferred embodiment, the binding molecules of the present invention result in a higher number of CD4+CD25+FoxP3+ cells compared to incubation in the absence of the binding molecule. In another embodiment, the candidate binding molecules may also be compared to the specific binding molecules of the present invention to evaluate whether, for example, a variant binding molecule can increase Tregs to the same or greater extent than the specific binding molecules of the present invention.

[0162] In another preferred embodiment, FoxP3-Luci mice are used to examine Treg cell expansion, which express luciferase under the control of mouse FoxP3 promoter and act as a marker for Treg cells. For example, such mice can be injected with a candidate TNFR2 agonist, and then Treg cells are imaged using bioluminescence imaging. A positive control with a known TNFR2 agonist can also be performed, as well as a negative control. In one embodiment, if a variant or candidate binding molecule produces equivalent or higher Treg levels compared to a known TNFR2 agonist described herein, as assessed by bioluminescence imaging, in a preferred embodiment, the variant or candidate binding molecule is classified as a binding molecule of the present invention. Such evaluation can also be combined with ex vivo evaluation, for example by later sacrificing animals, isolating cells, and then examining Treg numbers.

[0163] In another preferred embodiment, to test human specific molecules, transgenic mice expressing human TNFR2 or especially the extracellular domain of human TNFR2, obtained after random insertion of human TNFR2 cDNA into mouse genome or more specifically into mouse TNFR2 locus, are used to examine Treg levels, especially expansion. Such transgenic mice can be crossed with FoxP3-Luci transgenic mice for in vivo imaging of Treg expansion. At the time of sacrifice, individual tissues can also be processed through imaging for changes in Treg levels compared to negative control animals. Treg expansion and / or Treg / Teff ratio can also be quantified using flow cytometry originating from spleen cells, white blood cells in blood or other tissues. Alternatively, immune-deficient mice, such as NSG mice, can be injected with human PBMC or human Treg, and Treg expansion can be determined through flow cytometry.

[0164] The efficacy of a particular binding molecule may be evaluated in an in vivo system, such as an animal model. For example, various graft-versus-host disease (GvHD) models may be used, and the binding molecule may be administered to such animal models, and then compared with a control animal of the same GvHD animal model, but not administered with the binding molecule. In one embodiment, the binding molecule of the present invention causes or reduces GvHD in an animal model. Other animal models, such as models of conditions such as psoriasis, inflammatory bowel disease (IBD), lupus, multiple sclerosis, type I diabetes, neurological diseases, and atherosclerosis, may be used in the same manner. In vivo testing may also be used to determine whether a given binding molecule can increase Treg in vivo.

[0165] Conjugates, fusion proteins, and effector molecules The binding molecule of the present invention, particularly the antibody, may be conjugated to an effector molecule. Thus, if desired, the binding molecule for use in the present invention, and particularly the antibody, may be conjugated to one or more effector molecules (multiple). It is recognized that the effector molecule may comprise a single effector molecule, or two or more molecules linked to form a single moiety that can bind to the binding molecule of the present invention, particularly the antibody. When it is desired to obtain a binding molecule according to the present invention, particularly the antibody, linked to an effector molecule, this can be prepared by standard chemical or recombinant DNA techniques, where the binding molecule, particularly the antibody, is linked to the effector molecule directly or via a coupling agent. The techniques of conjugating such effector molecules to antibodies are well known in the art (see Hellstrom et al., Controlled Drug Delivery, 2nd Ed., Robinson et al., eds., 1987, pp. 623-53; Thorpe et al., 1982, Immunol. Rev., 62:119-58 and Dubowchik et al., 1999, Pharmacology and Therapeutics, 83, 67-123).Specific chemical techniques include, for example, those described in WO 93 / 06231, WO 92 / 22583, WO 89 / 00195, WO 89 / 01476 and WO 03 / 031581. Alternatively, where the effector molecule is a protein or polypeptide, linkage may be achieved using recombinant DNA techniques as described in WO 86 / 01533 and EP 0 392 745. In one embodiment, the binding molecules of the invention, in particular antibodies, may comprise an effector molecule.The term effector molecule, as used herein, includes, for example, drugs, toxins, biologically active proteins, such as enzymes, antibodies, or antibody fragments, synthetic or naturally occurring polymers, nucleic acids and fragments thereof, such as DNA, RNA, and fragments thereof, radionuclides, in particular radioactive iodides, radioisotopes, chelated metals, nanoparticles, and reporter groups, such as fluorescent compounds or compounds that can be detected by NMR or ESR spectroscopy.

[0166] Examples of effector molecules may include cytotoxins or cytotoxic agents, including any agent that is harmful to (e.g., kills) cells. Examples include combrestatins, dolastatins, epothilones, staurosporines, maytansinoids, spongiostatins, rhizoxins, halichondrins, roridin, hemiasterlin, taxol, cytochalasin B, gramicidin D, ethidium bromide, emetine, mitomycin, etoposide, teniposide, vincristine, vinblastine, colchicine, doxorubicin, daunorubicin, dihydroxyanthracin dione, mitoxantrone, mithramycin, actinomycin D, 1-dehydrotestosterone, glucocorticoids, procaine, tetracaine, lidocaine, propranolol, and puromycin, as well as analogs or homologs thereof. Effector molecules also include, but are not limited to, antimetabolites (e.g., methotrexate, 6-mercaptopurine, 6-thioguanine, cytarabine, 5-fluorouracil, decarbazine), alkylating agents (e.g., mechlorethamine, thiotepa, chlorambucil, melphalan, carmustine (BSNU), and lomustine (CCNU), cyclophosphamide, busulfan, dibromomannitol, streptozotocin, mitomycin C, and cis-dichlorodiamineplatinum(II) (DDP) cisplatin), anthracyclines (e.g., daunorubicin (formerly daunomycin) and doxorubicin), antibiotics (e.g., dactinomycin (formerly actinomycin), bleomycin, mithramycin, anthramycin (AMC), calicheamicin, or duocarmycin), and mitotic inhibitors (e.g., vincristine and vinblastine).

[0167] Other effector molecules include chelated radionuclides, e.g. 111 In and 90 Y.,Lu 177 , Bismuth 213 ,californium 252 ,iridium 192 , and tungsten 188 / rhenium188 or drugs, such as, but not limited to, alkylphosphocholines, topoisomerase I inhibitors, taxoids, and suramin. Other effector molecules include proteins, peptides, and enzymes. Enzymes of interest include, but are not limited to, proteases, hydrolases, lyases, isomerases, and transferases. Proteins, polypeptides, and peptides of interest include, but are not limited to, immunoglobulins, toxins such as abrin, ricin A, Pseudomonas endotoxin, or diphtheria toxin, proteins such as insulin, tumor necrosis factor, alpha-interferon, beta-interferon, nerve growth factor, platelet-derived growth factor, or tissue plasminogen activator, thrombotic agents, or anti-angiogenic agents such as angiostatin or endostatin, or biological response modifiers such as lymphokines, interleukin-1 (IL-1), interleukin-2 (IL-2), granulocyte-macrophage colony-stimulating factor (GM-CSF), granulocyte-colony stimulating factor (G-CSF), nerve growth factor (NGF), or other growth factors and immunoglobulins.

[0168] Other effector molecules can include detectable substances, for example, useful in diagnosis.Exemplary detectable substances include various enzymes, prosthetic groups, fluorescent materials, luminescent materials, bioluminescent materials, radioactive nuclides, positron-emitting metals (for use in positron emission tomography), and non-radioactive paramagnetic metal ions.In general, see U.S. Patent No. 4,741,900 for metal ions that can be conjugated to antibodies for use as diagnostic agents. Suitable enzymes include horseradish peroxidase, alkaline phosphatase, beta-galactosidase, or acetylcholinesterase; suitable prosthetic groups include streptavidin, avidin, and biotin; suitable fluorescent materials include umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine, fluorescein, dansyl chloride, and phycoerythrin; suitable luminescent materials include luminol; suitable bioluminescent materials include luciferase, luciferin, and aequorin; and suitable radionuclides include, 125 I, 131 I, 111 In, and 99 Examples include Tc.

[0169] In another embodiment, the effector molecule can increase or decrease the half-life of the binding molecule, particularly the antibody, in vivo, and / or reduce immunogenicity, and / or enhance delivery across epithelial barriers to the immune system. Examples of suitable effector molecules of this type include polymers, albumin, albumin-binding proteins, or albumin-binding compounds, such as those described in WO 05 / 117984. When the effector molecule is a polymer, it can generally be a synthetic or naturally occurring polymer, such as an optionally substituted linear or branched polyalkylene, polyalkenylene, or polyoxyalkylene polymer, or a branched or unbranched polysaccharide, such as a homo- or heteropolysaccharide. Specific optional substituents that may be present in the synthetic polymers mentioned above include one or more hydroxy, methyl, or methoxy groups. Particular examples of synthetic polymers include optionally substituted linear or branched poly(ethylene glycol), poly(propylene glycol), poly(vinyl alcohol), or derivatives thereof, in particular optionally substituted poly(ethylene glycol), such as methoxypoly(ethylene glycol), or derivatives thereof.

[0170] The binding molecule of the present invention, particularly the antibody, may be conjugated to a molecule that modulates or changes serum half-life.The binding molecule of the present invention, particularly the antibody, may be conjugated to albumin, for example, to modulate serum half-life.In another embodiment, the binding molecule of the present invention, particularly the antibody, may comprise a peptide linker that is an albumin-binding peptide.Examples of albumin-binding peptides are included in WO2015 / 197772 and WO2007 / 106120, which are incorporated by reference in their entirety.

[0171] In another embodiment, the binding molecule of the present invention, particularly the antibody, is not conjugated to an effector molecule. In one embodiment, the binding molecule of the present invention, particularly the antibody, is not conjugated to a toxin. In another embodiment, the binding molecule of the present invention, particularly the antibody, is not conjugated to a radioisotope. In another embodiment, the binding molecule of the present invention, particularly the antibody, is not conjugated to an imaging agent.

[0172] Along with using conjugation to link the antibody molecules of the invention to effector molecules, linking different portions of the whole antibody, or, for example, linking a ligand-based TNFR2 binding site to the antibody portion of the molecule of the invention, can be used.

[0173] In one embodiment, the binding molecule of the present invention may be a fusion protein. "Fusion protein" as referred to herein is not limited to a specific type of fusion protein, as long as the portions of the fusion protein are fused by covalent bonds. For example, the portions of the fusion protein may be fused by expression in one or more single polypeptide chain(s), by one or more disulfide linkages, by chemical conjugation (preferably by chemical conjugation using click chemistry), and / or by any other covalent linkage known in the art as suitable linkage of proteins. Preferably, the portions of the fusion protein are fused by expression in one or more single polypeptide chain(s) and / or by one or more disulfide linkages. For example, the binding molecule, particularly an antibody, may comprise a polypeptide comprising a heavy or light chain constant region fused to a polypeptide sequence of an scFv. In another embodiment, the polypeptide may comprise a tenascin peptide sequence, particularly a heavy chain constant region sequence fused to a tenascin peptide sequence at the C-terminus of the polypeptide. In other embodiments, fusion can be by one or more disulfide linkages, by chemical conjugation (preferably by chemical conjugation using click chemistry), and / or by any other linkage known in the art as a suitable linkage for proteins.

[0174] In one embodiment, the binding molecule of the present invention comprises a moiety that affects the serum half-life of the molecule. For example, the binding molecule of the present invention may comprise an Fc tail, serum albumin, and / or a moiety that is a conjugate of serum albumin, and PEG. In a preferred embodiment, the binding molecule of the present invention comprises, is conjugated to, or has a binding site of serum albumin. In one embodiment, the binding molecule is conjugated to serum albumin or a variant thereof to change the serum half-life of the binding molecule. In another embodiment, the binding molecule is an antibody that comprises an antigen-binding site of serum albumin. In another embodiment, the binding molecule of the present invention, particularly an antibody, may comprise a peptide sequence that binds to serum albumin.

[0175] In one embodiment, the binding molecules of the invention may include a linker or multiple linkers. Linkers may be used to link parts of the binding molecule together. In one embodiment, the binding moieties are separated by a linker. In a preferred embodiment, where the binding molecule includes at least one linker, the binding molecule includes a glycine / serine linker, such as GS or G4S (GGGGS), or multiple copies of one or both such linkers. In one embodiment, a single G4S linker may be used. In one embodiment, the linker may include multiple G4S units, for example, each linker having 1-7 repeats of that sequence. In one embodiment, 1-5 G4S linker units are used consecutively. Any suitable linker may be used in the present invention, and alternative embodiments are provided in which the linker is not limited to the G4S linker sequence, but may be any suitable linker, for example, in those embodiments using a specific G4S linker. In one embodiment where three VHH domains are present contiguously in the same polypeptide, the VHH domains may be separated using linkers on either side of the central VHH, for example to separate the central VHH from the other VHH domains.

[0176] In one embodiment, the binding molecule of the present invention may comprise a hinge region. In the case of a (tetravalent) IgG format comprising a VH and a VL, such a hinge region typically connects the CH1 domain to the CH2 domain. In the case of a VHH, such a hinge region typically connects the most C-terminal sdAB / VHH to the Fc portion. In one embodiment, the hinge may be an IgG1 hinge region, or may be 2, or 3, or 4 hinge regions. It may be such a human hinge region. In one embodiment, the hinge region is the hinge region of the human IgG1 consensus sequence. In one embodiment, the hinge region is the hinge region of durvalumab. In one embodiment, the hinge region may be a variant of the durvalumab hinge region.

[0177] Many of the polypeptides provided herein include signal or leader sequences. Where applicable, particularly preferred polypeptides are those in which the signal or leader sequence has been removed.

[0178] Pharmaceutical Compositions In one embodiment, the invention provides a pharmaceutical composition comprising (a) a binding molecule of the invention, particularly an antibody; and (b) a pharma- ceutically acceptable carrier, diluent, and / or excipient. In one embodiment, the pharmaceutical composition comprises a binding molecule of the invention that is an antibody. In one embodiment, the pharmaceutical composition of the invention comprises a binding molecule of the invention and a carrier, stabilizer, excipient, diluent, solubilizer, surfactant, emulsifier, preservative, and / or adjuvant. In one embodiment, the pharmaceutical composition of the invention is in solid or liquid form. In one embodiment, the pharmaceutical composition may be in the form of a powder, tablet, liquid, or aerosol. In one embodiment, the pharmaceutical composition of the invention is provided in frozen form.

[0179] The composition of the present invention is usually provided as a sterile pharmaceutical composition.The pharmaceutical composition of the present invention may further comprise a pharmaceutically acceptable auxiliary.In another embodiment, such auxiliary is not present in the pharmaceutical composition of the present invention.The present invention also provides a process for preparing pharmaceutical or drug composition, comprising adding and mixing the binding molecule of the present invention, particularly the antibody, with one or more pharmaceutically acceptable excipients, diluents or carriers.

[0180] The pharma- ceutically acceptable carrier in the therapeutic composition may further contain liquids such as water, saline, glycerol, and ethanol. Such carriers may be used, for example, so that the pharmaceutical composition is formulated as a tablet, pill, dragee, capsule, liquid, gel, syrup, slurry, and suspension for ingestion by the patient. The term "pharmaceutically acceptable excipient" as used herein typically refers to a pharma- ceutically acceptable formulation carrier, solution, or additive for enhancing the desired characteristics of the composition of the present invention. Excipients are well known in the art and include buffers (e.g., citrate buffer, phosphate buffer, acetate buffer, and bicarbonate buffer), amino acids, urea, alcohol, ascorbic acid, phospholipids, proteins (e.g., serum albumin), EDTA, sodium chloride, liposomes, mannitol, sorbitol, and glycerol. The solution or suspension may be encapsulated in liposomes or biodegradable microspheres. Suitable carriers may be large, slowly metabolized macromolecules such as proteins, polypeptides, liposomes, polysaccharides, polylactic acids, polyglycolic acids, polymeric amino acids, amino acid copolymers, and inactive virus particles. Pharmaceutically acceptable salts, such as inorganic acid salts, such as hydrochlorides, hydrobromides, phosphates, and sulfates, or salts of organic acids, such as acetates, propionates, malonates, and benzoates, may be used.

[0181] In certain embodiments, pharmaceutical compositions may contain formulation materials for the purpose of modifying, maintaining, or preserving certain characteristics of the composition, such as pH, osmolality, viscosity, color, isotonicity, odor, sterility, stability, dissociation or release rate, absorption or penetration. A thorough discussion of pharmaceutically acceptable carriers is available in Remington's Pharmaceutical Sciences (Mack Publishing Company, NJ 1991). Additional pharmaceutical compositions include formulations that include the binding molecules of the present invention in sustained or controlled delivery formulations. Techniques for formulating a variety of sustained or controlled delivery means are known to those skilled in the art. The binding molecules of the present invention may also be entrapped in microcapsules, prepared, for example, by coacervation techniques or interfacial polymerization, in colloidal drug delivery systems, or in macroemulsions. Such techniques are also disclosed in Remington's Pharmaceutical Sciences.

[0182] A subject is typically administered a therapeutically effective amount of the pharmaceutical composition, and thus the binding molecule of the present invention. The term "therapeutically effective amount" typically refers to the amount of therapeutic agent required to treat, ameliorate, or prevent the targeted disease or condition, or to exhibit a detectable therapeutic or preventative effect. The exact therapeutically effective amount for a human subject depends on the severity of the disease state, the subject's general health, the subject's age, weight, and sex, diet, time and number of administrations, drug combination(s), reaction sensitivity, and tolerance / response to treatment. This amount can be determined by routine experimentation and is within the physician's judgment. Generally, a therapeutically effective amount is 0.01 mg / kg to 50 mg / kg, such as 0.1 mg / kg to 20 mg / kg / day. Alternatively, the dose can be 1 to 500 mg / day, such as 10 to 100, 200, 300, or 400 mg / day. In one embodiment, the amount of a given dose is at least sufficient to effect a specific function.

[0183] In one embodiment, the binding molecule of the present invention can be administered in combination with another treatment for the condition being treated, in particular a TNF-alpha blocker. In one embodiment, the second drug with which the binding molecule of the present invention is used is an anti-inflammatory agent. For example, the binding molecule of the present invention can be provided simultaneously, sequentially, or separately with such additional agent. In another embodiment, the binding molecule of the present invention can be provided in the same pharmaceutical composition as the second therapeutic agent. Examples of drugs that can be provided in the same composition or used with the binding molecule of the present invention include TNF-alpha blockers, rapamycin, ustekinumab, or agents used as standard treatment in autoimmune or inflammatory diseases. Examples of TNF-alpha blockers that can be used with the binding molecule of the present invention include, but are not limited to, infliximab, adalimumab, etanercept, golimumab, and certolizumab. In one embodiment, the second drug is an antibody specific for TNF-alpha. In one embodiment, the second drug administered can be a steroid.

[0184] In a preferred embodiment, the therapeutic agent of the present invention, when in a pharmaceutical preparation, may be present in a unit dosage form. A suitable dose may be calculated for a patient according to the patient's body weight, but for example a suitable dose may be in the range of 0.01-20 mg / kg, such as 0.1-20 mg / kg, such as 1-20 mg / kg, such as 10-20 mg / kg, or such as 1-15 mg / kg, such as 10-15 mg / kg. To effectively treat a condition, a suitable dose for use in humans in the present invention may be in the range of 0.01-1000 mg, such as 0.1-1000 mg, such as 0.1-500 mg, such as 500 mg, such as 0.1-100 mg, or 0.1-80 mg, or 0.1-60 mg, or 0.1-40 mg, or such as 1-100 mg, or 1-50 mg of the dual targeting protein of the present invention, which may be administered parenterally, for example subcutaneously, intravenously, or intramuscularly. Such doses are repeated, if necessary, at appropriate intervals selected as appropriate by the physician. The binding molecules of the present invention may be, for example, lyophilized for storage and reconstituted in a suitable carrier prior to use. Lyophilization and reconstitution techniques can be used.

[0185] The binding molecules and pharmaceutical compositions of the present invention can be administered by any number of routes, including, but not limited to, oral, intravenous, intramuscular, intraarterial, intramedullary, intrathecal, intracerebroventricular, transdermal, transdermal (see, for example, WO 98 / 20734), subcutaneous, intraperitoneal, intranasal, intestinal, topical application, sublingual, intravaginal, or rectal routes. Hyposprays can also be used to administer the pharmaceutical compositions of the present invention. Direct delivery of the composition can generally be achieved by subcutaneous, intraperitoneal, intravenous, or intramuscular injection, or by delivery to the interstitial space of a tissue. In a preferred embodiment, administration is via intravenous administration, e.g., in a preferred embodiment, administration is via intravenous injection. In another preferred embodiment, administration is via subcutaneous administration, e.g., via subcutaneous injection. The compositions can also be administered to a specific tissue of interest. In some embodiments, the binding molecules of the present invention are administered via site-specific or targeted local delivery techniques. Examples of site-specific or targeted local delivery techniques include various implantable depot sources of binding molecules, or local delivery catheters, such as infusion catheters, indwelling catheters, or needle catheters, synthetic grafts, adventitial wraps, shunts and stents, or other implantable devices, site-specific carriers, direct injection, or direct application.

[0186] The dosage treatment may be a single dose schedule or a multiple dose schedule. If the product is for injection or infusion, it may take the form of a suspension, solution, or emulsion in an oily or aqueous medium and may contain formulatory agents such as suspending agents, preservatives, stabilizing agents, and / or dispersing agents. Alternatively, the binding molecule, particularly the antibody, may be in a dry form for reconstitution with a suitable sterile liquid before use. In one embodiment, the pharmaceutical composition comprising the binding molecule of the present invention is provided in a lyophilized form. If the composition is administered by a route that uses the digestive tract, the composition must contain an agent that protects the binding protein from degradation and releases the binding molecule once it is absorbed from the digestive tract. In another embodiment, the sprayable formulation according to the present invention may be provided as a single dose unit (e.g., a sealed plastic container or vial), for example, filled in a foil envelope. Each vial contains a unit dose in a volume, for example, 2 ml, of solvent / solution buffer.

[0187] The pharmaceutical composition of the present invention can be provided as a container that provides a means for administering to a subject.In one embodiment, the pharmaceutical composition of the present invention can be provided as a pre-filled syringe.Accordingly, the present invention provides such a loaded syringe.Similarly, the present invention also provides an autoinjector loaded with the pharmaceutical composition of the present invention.

[0188] In one embodiment, the formulation is provided as a formulation for topical administration, including inhalation.Suitable inhalation preparations include inhalable powders, metered-dose aerosols containing propellant gas, or inhalation solutions without propellant gas.The inhalable powders according to the invention containing active substances may consist of the above-mentioned active substances alone or a mixture of the above-mentioned active substances with physiologically acceptable excipients.These inhalable powders may contain monosaccharides (e.g. glucose or arabinose), disaccharides (e.g. lactose, saccharose, maltose), oligosaccharides and polysaccharides (e.g. dextran), polyalcohols (e.g. sorbitol, mannitol, xylitol), salts (e.g. sodium chloride, calcium carbonate), or mixtures of these with each other.Monosaccharides and disaccharides are preferably used, and lactose or glucose are used in particular, but not exclusively, in the form of their hydrates.

[0189] Particles for deposition in the lungs should have a particle size of less than 10 microns, for example 1-9 microns, for example 1-5 μm. The particle size of the active ingredient (e.g., antibody or fragment) is particularly important. Propellant gases that can be used to prepare inhalable aerosols are known in the art, suitable propellant gases are selected from hydrocarbons, such as n-propane, n-butane, or isobutane, and halohydrocarbons, such as chlorinated and / or fluorinated derivatives of methane, ethane, propane, butane, cyclopropane, or cyclobutane. The above propellant gases may be used alone or in combination. Particularly suitable propellant gases are halogenated alkane derivatives selected from TG11, TG12, TG134a, and TG227. Among the above halogenated hydrocarbons, TG134a (1,1,1,2-tetrafluoroethane) and TG227 (1,1,1,2,3,3,3-heptafluoropropane), and mixtures thereof, are particularly suitable. The propellant gas-containing inhalable aerosols may also contain other ingredients, such as cosolvents, stabilizers, surfactants (surfactants), antioxidants, lubricants, and means for adjusting the pH. All these ingredients are known in the art. The propellant gas-containing inhalable aerosols according to the invention may contain up to 5% by weight of active substance. The aerosols according to the invention contain, for example, 0.002 to 5% by weight, 0.01 to 3% by weight, 0.015 to 2% by weight, 0.1 to 2% by weight, 0.5 to 2% by weight, or 0.5 to 1% by weight of active ingredient.

[0190] Alternatively, local administration to the lungs can also be by administration of a liquid solution or suspension formulation using a device such as a nebulizer, e.g., a nebulizer connected to a compressor (e.g., a Pari LC-Jet Plus® nebulizer connected to a Pari Master® compressor manufactured by Pari Respiratory Equipment, Inc., Richmond, Va.).

[0191] The nebulizer formulation according to the invention may be provided as a single dose unit (e.g., sealed plastic container or vial) packed in a foil envelope. Each vial contains a unit dose in a volume of solvent / solution buffer, e.g., 2 mL. The invention also provides a syringe loaded with a composition comprising a binding molecule of the invention, particularly an antibody. In one embodiment, a pre-filled syringe is provided loaded with a unit dose of a binding molecule of the invention, particularly an antibody. In another embodiment, an autoinjector is provided loaded with a binding molecule of the invention, particularly an antibody. In a further embodiment, an IV bag is provided loaded with a binding molecule of the invention, particularly an antibody. Similarly, the binding molecule of the invention, particularly an antibody, is provided in lyophilized form in a vial or in a similar container.

[0192] It is also envisioned that the binding molecules of the invention may be administered by using gene therapy. To accomplish this, DNA sequences encoding the required polypeptides, e.g., the heavy and light chains of an antibody molecule, are introduced into the patient under the control of appropriate DNA constructs such that the antibody chains are expressed from the DNA sequences and assemble in situ.

[0193] After being formulated, the composition of the present invention can be directly administered to a subject.The subject to be treated can be an animal.However, in one or more embodiments, the composition is adapted for administration to humans.In a particularly preferred embodiment, the subject is a human.

[0194] In one embodiment, when a binding molecule of the invention is administered together with another agent, both may be administered simultaneously, sequentially or separately, hi one embodiment, both may be administered in the same pharmaceutical composition.

[0195] In one embodiment, the pharmaceutical composition of the present invention is administered prior to, simultaneously with, and / or after transplantation. For example, the pharmaceutical composition may be administered simultaneously with cell, tissue, or organ transplantation, such as any of those mentioned herein. In another embodiment, the material to be transplanted into the subject will be treated ex vivo before transplantation. In one embodiment, the binding molecules of the present invention may be used to prevent or reduce rejection of the transplanted material. In one embodiment, the binding molecules of the present invention may be used to prevent, treat, or ameliorate immune responses against the transplanted material. In a particularly preferred embodiment, the binding molecules of the present invention are used for purposes related to graft-versus-host disease.

[0196] The present invention also extends to a kit comprising the binding molecule of the present invention.In one embodiment, a kit is provided that comprises any of the binding molecules of the present invention, optionally together with instructions for administration.In yet another embodiment, the kit further comprises one or more reagents for carrying out one or more functional assays.In another embodiment, a kit is provided that contains a single-chamber or multi-chamber pre-filled syringe.The present invention also provides a kit for single-dose administration units.

[0197] Pathological, medical, and diagnostic uses Also provided is the use of the binding molecule of the present invention for use as a medicine.In another embodiment, the binding molecule of the present invention is provided for use in a method of treatment of the human or animal body.It should be noted that in the various therapeutic uses described herein that refer to the binding molecule, unless otherwise stated, pharmaceutical compositions comprising the binding molecule may also be used, and vice versa.The binding molecule of the present invention can also be used in diagnosis, including both in vivo diagnosis and also in vitro diagnosis, such as such diagnosis carried out on a sample from a subject.

[0198] As further discussed below, the binding molecules of the present invention may be used to treat conditions. As used herein, the term "treat" or "treatment" refers to both therapeutic treatment and prophylactic or preventative measures, where the purpose is to prevent or slow (attenuate) undesirable physiological changes or disorders. Beneficial or desirable clinical results include, but are not limited to, alleviation of symptoms, diminishment of the extent of disease, stabilization of disease state (i.e., not worsening), delay or slowing of disease progression, improvement or alleviation of disease state, and remission (either partial or complete), whether detectable or undetectable. "Treatment" may also mean prolonging survival compared to expected survival if not receiving treatment. Subjects in need of treatment include those already with a condition or disorder, as well as those prone to having a condition or disorder, or those in whom a condition or disorder is to be prevented.

[0199] In a particularly preferred embodiment, the binding molecules of the present invention may be used to modulate the immune system. For example, the binding molecules of the present invention may be used to stimulate cells of the immune system, for example to activate certain cells of the immune system. In one embodiment, the cells may be stimulated to proliferate. In a preferred embodiment, the binding molecules of the present invention are used to activate cells that express TNFR2 on their surface. For example, the cells may be white blood cells, in particular T cells. In a particularly preferred embodiment, the binding molecules of the present invention are used to activate Treg cells. For example, the binding molecules of the present invention may be used to stimulate Treg cells, which then suppress, reduce or prevent an immune response.

[0200] The ability of the present invention to modulate the immune system means that it represents a particularly good way to target, for example, autoimmune or inflammatory disorders. Thus, the present invention provides a binding molecule or pharmaceutical composition of the present invention for use in a method for treating or preventing an autoimmune or inflammatory disorder. The present invention provides a binding molecule or pharmaceutical composition for use in such a method, comprising: (a) the disorder is graft-versus-host disease (GvHD), preferably for use in a method in which the binding molecule is administered prior to, simultaneously with, or following cell, tissue, or organ transplantation; or (b) the disorder involves impaired function or unwanted proliferation of leukocytes, preferably T cells, more preferably Teff cells, and such disorders may exhibit an imbalance of Tregs compared to Teff cells, e.g., due to an increase in the number or activity of Teffs that is not balanced in terms of Treg numbers and / or immunosuppressive properties, and in some embodiments the invention promotes Tregs, in particular shifting the balance of effector cells against or towards normal cells; (c) the disorder is selected from inflammatory bowel disease (e.g., ulcerative colitis, Crohn's disease, and celiac disease), multiple sclerosis, myasthenia gravis, skin autoimmune diseases such as pemphigus vulgaris or bullous pemphigoid, and type I diabetes, and atherosclerosis; A binding molecule or pharmaceutical composition is provided.

[0201] The present invention can be used to treat graft-versus-host disease (GVHD). In one embodiment, the present invention is used to promote Treg activity before cell, tissue, or organ transplantation. For example, in one embodiment, the present invention is used to promote Treg activity before cell transplantation, particularly before stem cell transplantation, preferably before hematopoietic stem cell transplantation. In another embodiment, rather than stimulating Treg in the recipient before transplantation, the present invention is used to increase Treg in cells, tissues, or organs transplanted into the host. In a further embodiment, they are used as part of the treatment of non-malignant hematopoietic diseases.

[0202] The invention may be used to reduce, prevent, or treat immune responses against a transplant, for example against a transplanted cell, tissue, or organ. Thus, the invention may be used to reduce, prevent, or treat graft-versus-host disease (GvHD). In one embodiment, the invention may be used in methods in which the transplanted material is a cell, such as a cell population. In one embodiment, the transplanted material is or includes hematopoietic stem cells (HSCs). In another embodiment, the transplanted material may be an organ or tissue, such as a heart, lung, kidney, cornea, or other organ transplant. In another embodiment, the transplanted material may be a graft, such as a skin graft. In one embodiment, the invention provides a method comprising administering a binding molecule of the invention to treat, prevent, or ameliorate an undesired immune response against a transplanted cell, tissue, or organ. In one embodiment, the method may further comprise actually performing the transplant. In another embodiment, the binding molecule of the invention is administered to the subject before, during, and / or after the transplant. In a further embodiment, rather than administering the binding molecule to the subject, the method comprises treating the transplanted material with the binding molecule ex vivo before it is transplanted.

[0203] In one embodiment, rather than treating, preventing, or ameliorating the disease itself, the present invention is used to help ensure that the treatment of the disease, i.e., transplanted cells, tissues, or organs, is effective by preventing or reducing the severity of GvHD. Thus, the present invention can be used in a variety of embodiments in which the disease is treated by transplanting cells, tissues, or organs. In a preferred embodiment, the condition can be a condition that is treated via stem cell transplantation, e.g., hematopoietic stem cell (HSC) transplantation. In some embodiments, the subject has or is otherwise afflicted with a metabolic storage disease to be treated by transplantation. The subject may be afflicted with or otherwise afflicted with a metabolic disorder selected from the group consisting of glycogen storage diseases, mucopolysaccharidoses, Gaucher disease, Hurler disease, sphingolipidoses, metachromatic leukodystrophy, or any other disease or disorder that may benefit from the treatments and therapies disclosed herein, including, but not limited to, severe combined immunodeficiency, Wiskott-Aldrich syndrome, hyperimmunoglobulin M (IgM) syndrome, Chediak-Higashi disease, hereditary lymphohistiocytosis, osteopetrosis, osteogenesis imperfecta, storage diseases, thalassemia major, sickle cell disease, systemic sclerosis, systemic lupus erythematosus, multiple sclerosis, juvenile rheumatoid arthritis, and other conditions that may be treated by administration of hematopoietic stem cell transplantation therapy, including, but not limited to, those described in "Bone Marrow Transplantation for Non-Malignant Disease," ASH Education Book, 1 :319-338(2000). In one embodiment, when the invention relates to transplantation, the transfer can be of allogeneic cells, tissues, or organs. In one embodiment, the transferred cells can be cells expressing chimeric antigen receptors (CARs). In some embodiments, the subject is in need of chimeric antigen receptor T cell (CART) therapy. For example, such therapy can form part of the method of the invention.In another preferred embodiment, the present invention provides a method of promoting colonization of a cell population, tissue, or organ in a subject by treating, reducing, or preventing an immune response against said population, tissue, or organ.

[0204] The ability of the binding molecules of the present invention to modulate the immune system also makes them a particularly valuable approach for targeting autoimmune diseases. Thus, in another embodiment, the subject to be treated has an autoimmune disorder. In a particularly preferred embodiment, the autoimmune disorder is multiple sclerosis. In a further particularly preferred embodiment, the subject has ulcerative colitis. In another particularly preferred embodiment, the condition is scleroderma. In one embodiment, the condition to be treated is lupus. Further examples of autoimmune diseases include scleroderma, Crohn's disease, type 1 diabetes, or another autoimmune pathology described herein. In one embodiment, the autoimmune disease to be treated is selected from ulcerative colitis, Crohn's disease, celiac disease, inflammatory bowel disease, multiple sclerosis, lupus, Graves' disease, and type 1 diabetes. In one embodiment, the subject has type 1 diabetes and is treated. In one embodiment, the condition is atherosclerosis.

[0205] In a preferred embodiment, the condition to be treated is a condition involving undesirable inflammation. In a preferred embodiment, the condition is arthritis. For example, the present invention can be used to treat rheumatoid arthritis or osteoarthritis. Non-limiting types of examples of diseases that can be treated include rheumatoid arthritis, polyarticular juvenile idiopathic arthritis, psoriatic arthritis, and juvenile arthritis. In another preferred embodiment, the condition to be treated is selected from multiple sclerosis, ankylosing spondylitis, Crohn's disease, and ulcerative colitis.

[0206] In one embodiment, the ability of the present invention to stimulate Treg cells is used as a method of treating allergies, hi another embodiment, the ability to stimulate Treg cells can be used as a method of treating asthma.

[0207] Detection and Diagnosis The binding molecule of the present invention can be used to detect TNFR2.For example, the present invention provides a method comprising contacting the binding molecule of the present invention with a test sample and detecting any binding of the binding molecule.The binding molecule of the present invention can be labeled or linked to an enzyme that allows the detection of the binding molecule and thus the detection molecule binds.In one embodiment, such a detection method can be, for example, an ELISA assay or flow cytometry as a method to detect whether cells in a test sample express TNFR2 on their surface.The binding molecule of the present invention can be used for in vitro detection as well as for in vivo detection of TNFR2.

[0208] In one embodiment, the invention provides an in vivo method of detecting TNFR2, comprising administering a labeled binding molecule of the invention, and then detecting the location of the antibody in the subject's body. In another embodiment, the binding molecule of the invention may be used to diagnose a condition, for example to identify a reduction in cells expressing TNFR2, and in particular to identify a reduction in a particular subset of cells expressing TNFR2. In a preferred embodiment, the invention provides a method of stratifying a patient, comprising subclassifying the patient based on the level of TNFR2, for example based on the level of a particular cell type expressing TNFR2.

[0209] The invention also provides a kit for detecting TNFR2 comprising a binding molecule of the invention and, optionally, instructions for using the binding molecule in a method for detecting TNFR2.

[0210] In one embodiment, the invention provides a binding molecule of the invention for determining whether or not to administer a drug to a subject based on detection of, e.g., TNFR2, e.g., TNFR2 levels, e.g., the number of a particular cell type expressing TNFR2, or its location, as a companion diagnostic.

[0211] The binding molecule of the present invention can be used for imaging. In one embodiment, the binding molecule can be used for in vivo imaging. Thus, also provided is an imaging method, which comprises administering the labeled binding molecule of the present invention to a subject and then detecting the label. In one embodiment, imaging can also allow the detection of cells expressing TNFR2 in vivo. In one embodiment, such a method can be used to identify the location or composition of such cells. In another embodiment, the binding molecule can be used for in vitro imaging method.

[0212] As discussed herein, the present invention may be used to treat a subject. "Subject" or "individual" or "animal" or "patient" or "mammal" refers to any subject, particularly a mammalian subject, whose diagnosis, prognosis, or treatment is desired. Mammalian subjects include humans, livestock animals, farm animals, and zoo animals, sport animals, or pet animals, such as dogs, cats, guinea pigs, rabbits, rats, mice, horses, cattle, cows, etc. In a particularly preferred embodiment, the subject is a human.

[0213] The binding molecules of the present invention can also be used to stimulate T cell proliferation, particularly Treg cell proliferation.In one embodiment, such a method is carried out in vitro or ex vivo, for example by contacting Treg cells with the binding molecules of the present invention.In another embodiment, the present invention can further comprise, for example after expanding the Treg population or after contacting with the binding molecules, transferring the cells into a subject, whereby the proliferation occurs in vivo after the cells are transferred into a subject.

[0214] Further embodiments The invention also provides methods of treating or preventing an autoimmune or inflammatory disorder of the invention in a subject suffering from or at risk for the disorder.

[0215] The invention also provides such a method for treating or preventing graft-versus-host disease (GvHD), comprising administering a binding molecule to a subject and transplanting a cell, tissue, or organ into the subject, wherein the binding molecule is administered before, during, or after transplantation.

[0216] The present invention further provides a method as follows: (a) the disorder involves a reduction in the number, dysfunction or unwanted proliferation of white blood cells, preferably T cells, more preferably Treg cells; or (b) the disorder is one that would benefit from increased number and / or function of Tregs; (c) the disorder is selected from inflammatory bowel disease (e.g., ulcerative colitis, Crohn's disease, and celiac disease), atherosclerosis, lupus, multiple sclerosis, type I diabetes, myasthenia gravis, pemphigus vulgaris, and bullous pemphigoid.

[0217] The invention also provides the use of a binding molecule of the invention for the manufacture of a medicament for use in a method of treating or preventing an autoimmune or inflammatory disorder.

[0218] The present invention further provides such uses, wherein the medicament is for use in a method of treating: (a) the disorder involves impaired function or unwanted proliferation of leukocytes, preferably T cells, more preferably Treg cells; or (b) the disorder is selected from inflammatory bowel disease (e.g., ulcerative colitis, Crohn's disease, or celiac disease), lupus, multiple sclerosis, type I diabetes, atherosclerosis, myasthenia gravis, pemphigus vulgaris, and bullous pemphigoid.

[0219] The invention also provides a diagnostic method comprising administering a binding molecule of the invention to a subject and detecting binding of the binding molecule to TNFR2, preferably wherein the binding molecule is labeled and binding of the binding molecule to TNFR2 is detected via the label.

[0220] Numbered embodiments The following represent further preferred embodiments of the present invention.

[0221] [1]. A binding molecule that specifically binds to TNFR2 but not to TNFR1, which is an FcγR-independent agonist of TNFR2 and has at least bivalency with respect to binding to TNFR2.

[0222] [2].(a) oligomerizing TNFR2 on the surface of cells expressing TNFR2, preferably clustering six or more TNFR2 molecules on the cell surface; (b) induces TNFR2 signaling; and / or (c) stimulating the proliferation of leukocytes, preferably T cells, more preferably Treg cells; The binding molecule according to [1].

[0223] [3].(a) comprising a moiety that modifies serum half-life, preferably a moiety selected from the group consisting of an Fc tail, serum albumin, a moiety that is a conjugate of serum albumin, and PEG; (a) is bivalent with respect to TNFR2 binding; (b) is trivalent with respect to TNFR2 binding; (c) is tetravalent with respect to TNFR2 binding; or (d) having at least tetravalency with respect to TNFR2 binding; The binding molecule according to [1] or [2].

[0224] [4].(a) It is an antibody; (b) an antibody that contains a ligand for TNFR2; (c) comprising at least a portion of an antibody constant region, preferably an Fc region; or (d) a ligand for TNFR2, preferably comprising a ligand derived from TNFα or lymphotoxin-alpha; Preferably, (a) an antibody that contains TNFR2 binding sites at both the N-terminus and C-terminus of one or more heavy chains of the antibody; or (b) an antibody that contains TNFR2 binding sites at both the N-terminus and C-terminus of one or more light chains of the antibody; The binding molecule according to any one of [1] to [3].

[0225] [5].(a) the antibody is a bivalent antibody; (b) the antibody is a trivalent antibody; (c) the antibody is a tetravalent antibody; (d) the antibody is a hexavalent antibody; (e) the antibody is a heavy chain-only antibody; (f) the antibody is a single domain antibody, e.g., a single domain Ab (sdAb), VHH, or nanobody; (g) the antibody is a multimer of at least two antibodies; (h) the antibody comprises or is linked to a multimerizing moiety, preferably a tenascin-c peptide (TNC), and preferably the antibody is a multimer of at least two antibodies associated through the tenascin-c peptide sequence; (i) the antibody is linked to one or more molecules derived from a ligand of TNFR2, preferably TNF-α or lymphotoxin alpha; A binding molecule according to [4].

[0226] [6].(a) the antibody lacks an Fc region; or (b) the antibody comprises an altered Fc region that is incapable of binding to an Fc receptor; A binding molecule according to claim [4] or [5].

[0227] [7].The binding molecule is - IgG with Fc modifications that eliminate Fc receptor binding; - scFv:IgG with Fc modifications that eliminate Fc receptor binding in IgG; - IgG-HC:TNC, where the IgG has a modification that eliminates Fc receptor binding; - scFv:IgG-HC:TNC, in which the IgG has a modification that eliminates Fc receptor binding; and - Hexameric IgG an antibody in an antibody format selected from: - HC represents the heavy chain of IgG; - LC stands for IgG light chain; - scFv: IgG denotes an scFv that is linked to or forms part of an IgG molecule; and - HC:TNC denotes a heavy chain (HC) comprising or linked to a tenascin-c trimerization domain peptide (TNC); A binding molecule according to any one of the preceding claims.

[0228] [8].The binding molecule is - Fab-LC:scFv; - IgG-HC:scFv, in which the IgG has a modification that eliminates Fc receptor binding, preferably N297A; - IgG-LC:scFv, in which the IgG has a modification that eliminates Fc receptor binding, preferably N297A; - IgG-LC:scFv-HC:scFv; - LC:scFv-HC:scFv-IgG-HC:scFv; - LC:scFv-HC:scFv-IgG-LC:scFv; - LC:scFv-HC:scFv-IgG-LC:scFv-HC:scFv; and - IgG-HC-TNC-scFv; - sdAb / VHH an antibody in an antibody format selected from the group consisting of - HC represents the heavy chain of IgG; - LC stands for IgG light chain; - LC: scFv denotes an scFv linked to or forming part of a light chain; - HC: scFv denotes an scFv linked to or forming part of a heavy chain; - HC:TNC denotes a heavy chain (HC) comprising or linked to tenascin (TNC); - TNC-scFv denotes tenascin linked or bound to an scFv; - sdAb denotes a single domain antibody; and - VHH represents the variable heavy chain region of a camelid antibody, Preferably, the binding molecule is an antibody, the antibody comprising: - Fab-scFv; - IgG-HC:scFv; - LC:scFv-HC:scFv-IgG1; and - LC:scFv-HC:scFv-IgG1-TNC The antibody format is selected from The binding molecule according to any one of [1] to [7].

[0229] [9].A binding molecule that specifically binds to TNFR2 but not to TNFR1, the binding molecule being an FcγR-independent agonist of TNFR2 and having at least bivalency with respect to binding to TNFR2, and comprising a mutant TNF-alpha that binds to TNFR2 but not to TNFR1, and a polypeptide comprising an Fc region, or a tenascin peptide sequence that oligomerizes the polypeptide, or both; Preferably, (a) does not contain antibody sequences; (b) mutant TNF-alpha that binds to TNFR2 but not TNFR1, and a polypeptide that includes a tenascin peptide; (c) the TNF-alpha mutant is TNF80; and / or (d) having at least trivalency with respect to TNFR2, preferably hexavalency with respect to TNFR2; A binding molecule, a binding molecule.

[0230]

[10] . The binding molecule according to any one of [1] to

[10] for use as a pharmaceutical.

[0231]

[11] . A binding molecule according to any one of [1] to [9] for use in a method for treating or preventing an autoimmune or inflammatory disorder, preferably comprising: (a) for use in a method in which the disorder is graft-versus-host disease (GvHD) and preferably the binding molecule is administered prior to, simultaneously with, or following transplantation of a cell, tissue, or organ; or (b) the disorder involves impaired function or unwanted proliferation of leukocytes, preferably T cells, more preferably Treg cells; or (c) The binding molecule, wherein the disorder is selected from inflammatory bowel disease (e.g., ulcerative colitis, Crohn's disease, or celiac disease), atherosclerosis, lupus, multiple sclerosis, type I diabetes, myasthenia gravis, pemphigus vulgaris, and bullous pemphigoid.

[0232]

[12] . A method for stimulating cell proliferation, comprising contacting a target cell expressing TNFR2 with a binding molecule according to any one of [1] to [9].

[0233]

[13] A pharmaceutical composition comprising the binding molecule according to any one of [1] to [9] and a pharma- ceutically acceptable carrier.

[0234]

[14] A method for detecting TNFR2, comprising contacting a test sample with the binding molecule according to any one of [1] to [9], and detecting binding of the binding molecule to TNFR2, Preferably, the method, wherein the binding molecule is labeled and binding of the binding molecule to TNFR2 is detected via the label.

[0235]

[15] . A method for treating a TNFR2-associated inflammatory disease comprising administering to a subject a binding molecule according to any one of claims 1 to 15 and detecting binding of the binding molecule to TNFR2, Preferably, the binding molecule is labeled and binding of the binding molecule to TNFR2 is detected via the label. The binding molecule according to any one of [1] to [9] for use in a diagnostic method.

[0236] All documents mentioned herein are incorporated by reference. References to the singular herein using terms such as "a" and "an" also include the plural unless otherwise stated. When something is referred to as "comprising" herein, in another embodiment, the invention may "consist essentially of" what is described. In another embodiment, the invention may "consist of" what is described. Unless otherwise defined, all scientific and technical terms used herein have the meaning commonly understood by those skilled in the art to which the invention belongs. EXAMPLES

[0237] [Example 1] material and method Introduction The following provides a summary of the materials and methods used when performing the experiments described in the subsequent experiments.

[0238] Cell lines and cell culture conditions HEK293T cells, HeLa cells, HeLa-TNFR2 (Weiss et al., 1997), HT1080-TNFR2 (Gerspach et al., 2006, Cell Death Differ., 13(2): 273-284), and Kym-1 cells were cultured in RPMI1640 medium (Sigma-Aldrich, Irvine, United Kingdom) supplemented with 10% FCS (GIBCO, EU Approved, South America). Cells were cultured under standard conditions (5% CO 2 The cells were cultured at 37°C for 1 h.

[0239] Antibody cloning, production, and purification Heavy and light chain-encoding cDNA fragments of in-house generated TNFR2-specific antibodies (Medler et al., 2019) were generated, sequenced, and then cloned into the pCR3 expression vector (Invitrogen) using standard PCR and cloning techniques. Antibody chain variants (mutants, fusion proteins) were generated using synthetic DNA fragments encoding the amino acid sequences of interest and standard cloning techniques. Expression plasmids for TNFR2-specific antibodies typically contain a FLAG-tag.

[0240] Antibodies and antibody variants were produced in HEK293T cells using the transfection reagent PEI (polyethyleneimine; Polyscience Inc., Warrington, USA) as described, for example, in Medler and Wajant (2021). To quantify the amount of protein secreted into the supernatant, SDS-PAGE and anti-FLAG Western blotting with in-house produced FLAG-tagged standard protein (anti-FLAG M2 antibody, Sigma-Aldrich, St. Louis, USA) were performed.

[0241] The supernatant containing the produced TNFR2-specific antibodies was purified by anti-FLAG agarose affinity chromatography. After the FLAG-tagged antibodies bound to anti-FLAG M2 affinity gel (Sigma-Aldrich, St. Louis, USA), the antibodies were eluted using an excess of FLAG-peptide (100 μg / ml) (Sigma-Aldrich, St. Louis, USA) and fractions were collected. Purification was performed in a TBS buffer system. To exchange TBS against PBS, the protein was then dialyzed overnight at +4°C.

[0242] Non-FLAG tagged antibodies were purified via Protein A chromatography followed by size exclusion chromatography. Purity and concentration of purified proteins / antibodies were determined by SDS-PAGE and silver staining in comparison to Amersham's "Low Molecular Weight Calibration Kit for SDS Electrophoresis" (GE Healthcare UK Limited, Little Chalfont, UK), which contains defined amounts of proteins of known molecular weight. The silver staining kit was purchased from Thermo Scientific (Rockford, USA).

[0243] Antibody Format Figure 1 shows antibody formats in which the TNFR2 binding site is unidirectional N-terminal, showing the nomenclature and domain structure of each. The abbreviations used in the names given for the various antibody formats in the figures are: CI = clone identifier, e.g., C4, C19, etc.; N297A refers to the corresponding mutation in the heavy chain of human IgG1 that reduces FcγR interaction; N-RGY refers to the N297A mutation plus a mutation that promotes IgG1 hexamer formation; and TNC refers to the tenascin-C trimerization domain peptide that is present at the C-terminus of the heavy chain and acts to trimerize three individual antibodies together. The antibody formats and their respective N-terminal antigen binding numbers are summarized in the table below.

[0244] A table that summarizes the information

[0245] [Table 1]

[0246] Figure 7 shows antibody formats in which the TNFR2 binding sites are organized at both the N-terminus and C-terminus, showing the nomenclature and domain structure for each. The same abbreviations are used as in Figure 1 for clone identifiers, the presence of the N297A modification, and the presence of the tenascin trimerization peptide.

[0247] [Table 2]

[0248] Table 3 below summarizes the binding molecules referred to in Figures 1-26 and discussed in the Examples that follow, including the two expression vectors used to express each antibody polypeptide.

[0249] [Table 3-1]

[0250] [Table 3-2]

[0251] [Table 3-3]

[0252] Table 4 below shows the SEQ ID NOs for the light and heavy chain polypeptides encoded by the expression constructs referred to in Table 3 above.

[0253] [Table 4-1]

[0254] [Table 4-2]

[0255] [Table 4-3]

[0256] Table 5 below provides the CDR sequences of specific antibody clones that were used as the starting point for antigen binding sites for the various antibody formats used in this example other than the VHH-based antibodies whose light and heavy chain variable regions and CDR sequences are discussed in subsequent examples. The shaded portion is included simply to help indicate light and heavy chain CDRs from the same clone.

[0257] [Table 5-1]

[0258] [Table 5-2]

[0259] [Table 6]

[0260] Table 7 provides the sequence numbers of further related proteins.

[0261] [Table 7]

[0262] [Table 8]

[0263] [Table 9]

[0264] [Table 10]

[0265] Gel filtration Purified antibody fusion proteins (50–200 μg) were analyzed for their native weight and potential protein aggregation by gel filtration using an UltiMate 3000 HPLC system (Thermo Fisher) and aqueous SEC-1 column performance check standards (#AL0-3042, Phenomenex, Torrance, CA, USA) on MabPac SEC-1 columns (Thermo Fisher) or Gravitrap or MAbSelect columns.

[0266] Cell binding assay Affinity (K) of TNFR2-targeting antibodies to human, cynomolgus monkey, and mouse TNFR2 D The binding activity (values) of the TNFR2-encoding plasmids was assessed by a cell equilibrium binding assay. HEK293T cells were therefore transiently transfected with human, cynomolgus or mouse TNFR2-encoding expression plasmids or with an empty vector control (EV) using the PEI method. Aliquots of HEK293T transfectants were treated with increasing concentrations of the indicated GpL (Gaussia princeps luciferase)-tagged anti-TNFR2 antibodies and incubated for 1 h in standard cell culture conditions before being washed twice with ice-cold PBS. Cells were finally resuspended in 50 μl of RPMI 1640 medium supplemented with 0.5% FCS and transferred to black 96-well plates. Cell-associated luciferase activity (RLU; relative light units) was detected by a LUmo luminometer (anthos Mikrosysteme GmbH, Friesoythe, Germany) immediately after addition of the substrate 1.5 μM coelenterazine (Carl Roth, Karlsruhe, Germany).

[0267] The affinity of TNFR2-specific antibodies and antibody variants was also evaluated in HeLa cells and HeLa cells stably expressing human TNFR2. HeLa and HeLa-TNFR2 cells were seeded in 24-well plates. The next day, cells were challenged with the indicated concentrations of GpL-labeled TNFR2 antibodies for 1 h at 37°C. To remove unbound antibodies, cells were washed 10 times with ice-cold PBS. Cells were then scraped, collected in 50 μl of medium (RPMI, 0.5% FCS) and transferred to black 96-well plates for measurement as described above. The specific binding of the different TNFR2-specific antibodies / antibody variants was calculated by subtracting the nonspecific binding (values ​​obtained from HeLa cells) from the total binding (values ​​obtained from HeLa-TNFR2). The obtained values ​​were fitted using the nonlinear regression analysis option in GraphPad Prism 5.

[0268] Analysis of IL8 induction Cytokine induction is a well-established readout for NFκB activation in response to TNFR2 signaling. We analyzed the intensity of NFκB activation by measuring the amount of IL8 after TNFR2 stimulation with the help of a commercial IL8 ELISA kit (IL8 ELISA Set, BD OptEIA™, San Diego, USA).

[0269] HT1080-Bcl2-TNFR2 cells, which responded to TNFR2 stimulation with strong IL8 induction, were seeded in 96-well plates (20.000 cells / well). The next day, cells were stimulated with the TNFR2-targeting reagent of interest and the highly potent TNFR2-selective TNF variant oligomerizing TNF80 (=TNC-sc(mu)TNF(143N / 145R)) as a positive control (Chopra et al., 2015). After overnight incubation, 50 μl of the supernatant was analyzed with the aid of the BD OptEIA™ IL8 ELISA kit as described in the manufacturer's manual.

[0270] Viability assay TNFR2 can induce cell death in some cell lines / cell types by inhibiting survival proteins (TRAF2, cIAP1, cIAP2) and concomitantly upregulating TNF-induced TNFR1. This type of response has been demonstrated, among others, in Kym-1 cells (Schneider et al., 1999). To test whether TNFR2-targeting constructs can induce this TNFR2 response, Kym-1 cells were seeded in 96-well plates (20.000 cells / well) and stimulated the next day with the TNFR2-targeting reagent of interest. To define complete cell killing, an aliquot of cells was also treated with a "cell death" control mixture ((200ng / ml TNF, 200ng / ml TRAIL, 200ng / ml CD95L, 25μg / ml CHX, 1% (wt / vol) sodium azide)). The remaining plastic-adherent cells were finally stained with 70μl of crystal violet solution. After 20 min, the plate was washed with dHO. 2 After dissolution in methanol, the OD at 595 nm was measured by a PHOmo photometer (anthos Mikrosysteme GmbH, Friesoythe, Germany) to quantify the viability. Values ​​were normalized according to untreated cells (100% viability) and cells treated with a cytotoxic agent mixture (0% viability).

[0271] Activation of the non-canonical NFκB pathway Processing of p100 to p52 is a major hallmark of non-canonical NFκB pathway activity. To assess the ability of anti-TNFR2 antibodies to induce p100 processing, 1 × 10 Kym-1 cells were cultured in vitro. 61 were seeded in each well of a 6-well tissue culture plate. The next day, cells were stimulated with the TNFR2 targeting reagent of interest. After overnight incubation in standard cell culture conditions, total cell lysates were prepared (one wash with PBS, cells scraped in 1 ml PBS, centrifuged (2 min, 12000 rpm), cell sediments were lysed in 4x Laemmli sample buffer, sonicated for 25 s at maximum amplitude (UP100H Ultrasonic Processor, Hielscher, Germany), heated at 95°C for 5 min, and debris removed by centrifugation (2 min, 12000 rpm)). To assess p100 processing, Western blotting with standard techniques and standard equipment was performed using anti-p100 / p52 antibodies from Sigma Aldrich (Darmstadt, Germany).

[0272] Materials and Methods References

[0273] [Table 11]

[0274] [Example 2] Western blot analysis of variants of the anti-TNFR2 antibody C4 with parallel organized (unidirectional) N-terminal TNFR2 binding sites Flag-tagged variants of the anti-TNFR2 antibody C4 with a unidirectional (parallel organized) N-terminal TNFR2 binding site were expressed in HEK293T cells as described in Example 1. The antibody formats evaluated were those shown in Figure 1, the specific antibody used had the heavy and light chain variable regions of the clone 4 TNFR2 antibody, and the "clone identifier" shown in the figure is "C4". The cells were transfected with plasmids encoding each chain (HC, LC) of the various C4 constructs. The antibody variant-containing supernatants were analyzed by WB together with various amounts of antibody standards of known concentration, and the concentration of the antibody variants in the supernatants was concluded from the standard antibody signal. Volumes of the supernatants containing 200 ng of each antibody variant were again subjected to Western blotting with anti-human IgG and anti-Flag antibodies. The results obtained are shown in Figure 2, and all of the antibody formats were successfully expressed and found in the supernatants collected from the cells.

[0275] [Example 3] TNFR2-Mediated IL8 Production Following Treatment with a C4 Variant Bearing a Unidirectional (Parallel Organized) N-Terminal TNFR2 Binding Site The ability of antibody variants with unidirectional (parallel organized) N-terminal TNFR2 binding sites was assessed using antibody variants containing heavy and light chain variable regions from the original clone 4 antibody specific for TNFR2. Again, the structure of the variant is that shown in Figure 1, but the "clone identifier" is C4, indicating the origin of the antigen binding site. A positive control was performed using a highly active oligomeric form of the TNFR2-specific TNF mutant TNF(143N / 145R).

[0276] To carry out the evaluation, HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the various concentrations of C4 constructs shown in Figure 3. The next day, the supernatants were harvested and analyzed by ELISA for their IL8 content. The results obtained are shown in Figure 3. The maximum response induced was obtained with the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R), included as a positive control, the results of which are shown by the dotted line in the graph.

[0277] As can be seen from Figure 3, the most active antibody variants in the assay were those with multiple unidirectionally (parallel) organized N-terminal TNFR2 binding sites. The lowest levels of IL-8 production were observed for the C4-Fab, C4-Fab2, and C4-IgG1(N297A) variants, which have two or fewer TNFR2 binding sites.

[0278] [Example 4] TNFR2-Mediated Cell Death in Kym-1 Cells Following Treatment with a C4 Variant Bearing Unidirectional (Parallel Organized) N-Terminal TNFR2 Binding Sites Next, the ability of antibody formats with an N-terminal TNFR2 binding site to induce cell death in Kym-1 cells was assessed, again using the clone 4 light and heavy chain variable regions that confer specificity for TNFR2. The results obtained are shown in Figure 4, and the abbreviations used are again those from Figure 1, except that the use of "C4" denotes the origin of the antigen binding site.

[0279] Kym-1 cells were stimulated with supernatants containing various concentrations of C4 constructs as shown in Figure 4. The next day, cell viability was determined using crystal violet staining. The results obtained are shown in Figure 4. The highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) was again a positive control for TNFR2 engagement. The most effective antibody variants in inducing apoptosis were the scFv:C4-IgG1(N297A) and scFv:C4-IgG1-(N297A)-HC:TNC variants, the results of which are shown in the middle panel of Figure 4. The results of binding molecules based on the binding sites from clones C19, C27, and C40 are shown in Figure 6.

[0280] [Example 5] TNFR2-mediated p100 processing in Kym-1 cells following treatment with a C4 variant bearing unidirectionally (parallel) organized N-terminal TNFR2 binding sites The ability of C4 variants with unidirectionally (parallel) organized N-terminal TNFR2 binding sites to induce the processing of p100 to p52, thus indicating the induction of signaling through TNFR2, was evaluated. Kym-1 cells were stimulated overnight with supernatants containing the various concentrations of C4 variants shown in Figure 5. The structures of the variants used are shown in Figure 1, where "C4" stands for "clonal identifier". Treatments were performed in the presence of 20 μM ZVAD to prevent TNFR2-induced cell death. A nonameric TNFR2-specific mutant of mouse TNF (oligo TNF80) was used as a positive control. Western blotting was performed using antibodies specific for the indicated proteins, p100, p52, and β-actin as a control. The results obtained are shown in Figure 5.

[0281] [Example 6] Western blot analysis of C4 variants with N-terminal and C-terminal organized TNFR2 binding sites The experiments described in Example 2 for antibody variants with unidirectional (parallel organized) N-terminal TNFR2 binding sites were also performed for the C4 antibody variants depicted in Figure 7, which have both N- and C-terminal TNFR2 binding sites. The variants contained FLAG tags in the heavy and light chains, except for the heavy chain of the C4-IgG1(N297A)-HC:scFvC4 variant, which does not contain a Flag tag.

[0282] Each C4 construct was transfected into cells, and antibody variants were purified from the cell supernatant. 200ng of each antibody variant was used for Western blotting. The supernatants containing Flag-tagged variants were analyzed by Western blotting with anti-human IgG and anti-Flag antibodies. The results are shown in Figure 8, and all antibody formats were successfully expressed and found in the supernatants collected from cells.

[0283] [Example 7] TNFR2-mediated IL8 production in HT1080-Bcl2-TNFR2 cells following treatment with a C4 construct bearing N- and C-terminally organized TNFR2 binding sites An equivalent experiment to that described in Example 3 was also performed on a C4 variant with both N- and C-terminally organized TNFR2 binding sites as shown in Figure 7. To provide a comparison, multiple antibody variants with only an N-terminal TNFR2 binding site were also tested alongside variants with both N- and C-terminal TNFR2 binding sites.

[0284] HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing different concentrations of C4 constructs as shown in Figure 9. The next day, the supernatants were harvested and analyzed for their IL8 content by ELISA. The results obtained are shown in Figure 9. The data shown are from parallel experiments, but are shown in three diagrams for better resolution. The maximum response induced by the positive control (highly active oligomeric form of the TNFR2-specific TNF mutant TNF(143N / 145R)) is shown by the dotted line.

[0285] As shown in Figure 9, antibody variants with both N- and C-terminal TNFR2 binding sites tended to produce the most positive results, but valency also had an impact. For example, the results in the middle panel show that such an antibody format induces higher levels of IL8 production than the C4-IgG1(N297A) variant, which has only an N-terminal TNFR2 binding site.

[0286] [Example 8] TNFR2-mediated cell death in Kym-1 cells following treatment with a C4 construct bearing N- and C-terminally organized TNFR2 binding sites An experiment equivalent to that carried out in Example 4 was also carried out for C4 antibody variants, in which the antibody has both N-terminal and C-terminal TNFR2 binding sites. Kym-1 cells were stimulated with supernatants containing the various concentrations of C4 constructs shown in Figure 10. The following day, cell viability was determined using an MTT assay. The highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) was used as a positive control. The results obtained are shown in Figure 10.

[0287] [Example 9] Binding characteristics of TNFR2-specific antibodies to HeLa cells expressing human TNFR2. The binding characteristics of several TNFR2-specific antibodies from various clones were evaluated using Hela cells expressing TNFR2. The equilibrium binding of anti-TNFR2 antibody GpL fusion protein to cells expressing TNFR2 was evaluated at 37°C, and the results are shown in Figure 11A. Specific, non-specific, and total binding levels are shown for each antibody. The first graph shows the results of the GpL-Flag-TNFwt positive control for binding to TNFR2.

[0288] Further experiments were performed using GpL-tagged TNF rather than GpL-tagged antibody. HeLa-TNFR2 cells were pretreated with the indicated amounts of anti-TNFR2 antibody or left untreated. 5ng / ml GpL-TNF was added and binding was measured after 1 hour incubation at 37°C. Binding to non-transfected HeLa cells was then analyzed to obtain non-specific binding values. The results obtained are shown in Figure 11B.

[0289] [Example 10] αTNFR2 antibody competition HeLa-TNFR2 cells were pretreated with 3-10 μg of anti-TNFR2 antibodies or left untreated as shown in Figure 12. 50 ng / ml of various αTNFR2-IgG1(N297A)-LC:GpL variants were added and binding was measured after 1 hour as discussed in Materials and Methods of Example 1. The results obtained are shown in Figure 12.

[0290] [Example 11] Tetravalent mAb-occupied TNFR2 molecules induce significantly stronger IL-8 induction than TNFR2 molecules occupied by conventional bivalent mAb variants Paired TNFR2-negative Hela cells and HeLa-TNFR2 cells were stimulated for 8 hours with the concentrations shown in (A) and (B) of the GpL fusion protein of the bivalent and tetravalent TNFR2-specific mAb C4 (see the inset of the scheme). The supernatant of HeLa-TNFR2 was then analyzed for TNFR2-induced IL8 production, the results of which are shown in (B). Repeatedly washed cells (HeLa, and HeLa-TNFR2) were analyzed for cell-associated GpL activity. Specific binding to TNFR2 was calculated by subtracting the nonspecific binding value obtained from HeLa cells from the total binding value derived from HeLa-TNFR2 cells, the results of which are shown in (A). Note that luciferase activity was normalized according to the ratio of the number of GpL reporter domains to the number of TNFR2 binding domains in the two mAb variants (1 vs. 0.5).

[0291] IL8 production by the mAb variants was plotted directly as a function of their specific binding. The latter was converted to "occupied receptors / cell" with the help of the number of cells in the assay and the measured specific activity (RLU / molecule) of the GpL construct. The results are shown in (C). The maximum binding of the two constructs obtained from A is indicated by the dashed vertical line. The dotted line shows the linear regression of IL8 production as a function of receptor occupancy.

[0292] [Example 12] Binding characteristics of TNFR2-specific antibodies to HEK293T cells expressing human TNFR2 The binding properties of anti-TNFR2 antibodies from different clones were evaluated. HEK293 cells were transiently transfected with human, mouse, and cynomolgus TNFR2 or empty vector. Equilibrium binding of the indicated anti-TNFR2 antibody GpL fusion proteins was then determined at 37°C. Binding to empty vector transfected cells was determined to obtain non-specific binding values. The results obtained are shown in Figure 14. The antibody that specifically bound to all three human, mouse, and cynomolgus TNFR2 was the C19-IgG1-LC-GpL antibody. The C4-IgG1-LC-GpL antibody bound to human TNFR2 but not to monkey or mouse TNFR2. The C40-IgG1-LC-GpL antibody bound to human and monkey TNFR2 but not to mouse TNFR2.

[0293] FIG. 15 shows the domain structure of TNFR2 and indicates the portion of extracellular TNFR2 recognized by anti-TNFR2 antibodies C4, C19, C27, and C40, and the tetravalent CI-IgG1(N297A)-HC:scFvCI variant derived therefrom.

[0294] In addition, the recognition of the extracellular portion of TNFR2 by the anti-TNFR2 VHH VHH:C18, VHH:C74, VHH:C188 and VHH:C238 clones and constructs derived therefrom is depicted in a schematic diagram based on FIG.

[0295] [Example 13] Western blot analysis of monospecific and bispecific TNFR2-specific CI-IgG-HC:scFvCI variants Several different antibody variants were expressed and analyzed by Western blotting in a manner equivalent to the approach used in Examples 2 and 6. The 11 variants analyzed had TNFR2 binding sites from different clones, different heavy chain modifications, many of which also had a C-terminal TNFR2 binding site, and the variants tested are shown in Figure 16. The variants were expressed and analyzed in an equivalent manner as described in Examples 2 and 6.

[0296] The top panel of Figure 16 shows the results of blotting with anti-FLAG antibody against the FLAG tag with which the antibody was labeled. Note that i) the heavy chains of C4-IgG1(N297A), C4-IgG1(N297A)-HC:scFvC4, and C4-IgG1(N297A)-HC:scFvC19 do not contain a Flag tag; and ii) the conventional C19 variant used was the mouse IgG1(D265A) isoform. The bottom panel of Figure 16 shows the results of a Western blot with anti-hIgG.

[0297] [Example 14] TNFR2-mediated IL8 production in HT1080-Bcl2-TNFR2 cells The analysis of the ability of anti-TNFR2 antibodies to stimulate IL8 production in HT1080-Bcl2-TNFR2 cells was carried out in a manner equivalent to Examples 3 and 7. Several monospecific antibody variants were evaluated, as well as several bispecific antibodies with two different specificities for TNFR2. To carry out the evaluation, HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the various concentrations of constructs shown in Figure 17. The next day, the supernatants were harvested and analyzed by ELISA for their IL8 content. The results obtained are shown in Figure 17, where the upper panel shows the results of the monospecific antibodies and the lower panel shows the results of the bispecific antibodies.

[0298] The top panel shows that in each instance, the antibody variant with both N- and C-terminal TNFR2 binding sites stimulated higher levels of TNF production than the corresponding antibody variant with only the N-terminal TNFR2 binding site. The results for the C40-IgG1(N297A)-HC:scFvC40 binding molecule were particularly positive in the results shown in the top panel.

[0299] [Example 15] TNFR2-mediated cell death in Kym-1 cells following treatment with monospecific and bispecific TNFR2-specific CI-IgG-HC:scFvCI variants The ability of monospecific and bispecific antibodies to induce cell death was evaluated in a manner equivalent to the approach used in Examples 4 and 8. Kym-1 cells were stimulated with supernatants containing different concentrations of C4 constructs as shown in Figure 18. The following day, cell viability was determined using crystal violet staining. The results obtained are shown in Figure 18. The results of the monospecific antibodies are shown in the upper panel and the results of the biparatopic antibodies are shown in the lower panel.

[0300] [Example 16] TNFR2-mediated p100 processing in Kym-1 cells following treatment with monospecific and bispecific TNFR2-specific CI-IgG-HC:scFvCI variants The ability of monospecific and bispecific antibodies to effect signaling through TNFR2 was evaluated by examining p100 processing in a manner equivalent to Examples 5 and 9. Kym-1 cells were stimulated overnight with supernatants containing the various concentrations of C4 variants shown in Figure 19. Treatments were performed in the presence of 20 μM ZVAD to prevent TNFR2-induced cell death. A nonameric TNFR2-specific mutant of mouse TNF (oligo.TNF80) was used as a positive control. Western blotting was performed using antibodies specific for the indicated proteins, p100, p52, and β-actin as a control. The results obtained are shown in Figure 19.

[0301] [Example 17] Purification and functional analysis of C4-IgG1(N297A)-HC:scFvC4, C4-IgG1(N297A)-LC:scFvC4, and C4-IgG1(N297A)-LC:scFvC4-HC:scFvC4 The constructs shown in Figure 20 were evaluated using the methods discussed in Example 1 and used in the previous examples. The results obtained are shown in Figure 20. (A) Silver stained SDS-PAGE gel. (B) Gel filtration analysis. (C) IL8 induction in HT1080-Bcl2-TNFR2 cells. (D) Cell death induction in Kym-1 cells. (E) Induction of p100 processing in Kym-1 cells. ZVAD was added to prevent cell death.

[0302] [Example 18] Activity Table 7 below summarizes the activity of the various antibody formats used in the previous examples, produced as described in the Materials and Methods section of Example 1. The antibody formats evaluated were those with antigen binding sites derived from the C4 clone, and "Bs" in the table indicates the number of "North" and "South" antigen binding sites of the antibody. The table summarizes the activity observed in the IL-8 assay (Activity I - classical NFkB pathway > IL8) or Activity II (TRAF2 depletion > non-classical NFkB pathway (= p100 processing) and Kym-1 killing)).

[0303] [Table 12]

[0304] [Example 19] Ex vivo assessment of Treg expansion The ability of specific TNFR2 binding molecules to expand regulatory T cells was evaluated. The tetravalent antibodies in Fab-HC-scFv format had an N297A amino acid mutation in the heavy chain of the Fc region to eliminate Fc function, and a base Fab of IgG1 Fab. PBMCs were isolated from four donors and cells from each donor were evaluated individually. PBMCs were isolated from 10 cells / mL. 7The cells were cultured at a density of 10 cells / ml for 2 days. Then, PBMCs were harvested and plated in multi-well plates with 10 cells. 6 Cells were seeded at a density of 1000 / ml, 500,000 cells / well. The cells were subdivided into (a) untreated control cells; (b) cells treated with 100 ng / ml STAR2 agonist molecule; (c) cells treated with a tetravalent antibody with only one specificity (1000-0.3 ng / ml); and (d) cells treated with a tetravalent antibody with two specificities for TNFR2 (1000-0.3 ng / ml). After 4 days of culture, the viability of the cells was evaluated. The cells were stained with antibodies specific for CD3, CD8, CD4, CD25, and FoxP3. The cells were then evaluated via flow cytometry. Figure 21 shows the results obtained in the upper panel for Foxp3+CD25+ cells among CD3+CD4+ cells to measure Treg cells and their expansion. The lower panel of Figure 21 shows Foxp3- cells among CD3+CD4+ cells to measure Tcon cells. All figures show the percentage change compared to the cell number before the 4-day incubation.

[0305] [Example 20] In vivo Treg expansion The ability of the provided TNFR2 agonist to expand Treg cells was also evaluated using syngeneic FoxP3-Luci mice. FoxP3-Luci mice are transgenic mice that express a fusion protein of FoxP3 and luciferase. This means that it is possible to effectively image the location of Treg cells in vivo. Mice were administered TNFR2 agonists and bioluminescence imaging was performed daily. After 4 days, the mice were sacrificed and ex vivo analysis was performed on cells isolated from the mouse spleens to evaluate Foxp3, CD3, and CD4 using flow cytometry. The timeline of the experiment is summarized in the top panel (A) of Figure 22. The results of the ex vivo analysis are shown in the bottom panel (B) of Figure 22. The experiment was repeated on multiple mice for treatment groups as shown in Figure 23.

[0306] [Example 21] Generation and analysis of TNFR2-specific TNF constructs TNF-alpha based molecules were generated as shown in part (A) of Figure 24. All molecules included the TNF80 molecule, which represents a mutant form of TNF-alpha that has lost TNFR1 binding and is thus specific for TNFR2. The structures shown from left to right in part (A) are: - Fc-TNC-TNF80 - contains three Fc regions in which the C-terminal region of each polypeptide contains a tenascin peptide (TNC) followed by a single TNF80 sequence (TNF80 corresponds to the TNFR2 specific mutant TNF (143N / 145R)). The tenascin peptide forms a trimer, which in turn means that there are two trimers of mutant TNF-alpha. - Fc-TNF80 - the C-terminal region of each polypeptide contains three Fc regions containing a single TNF80 sequence, resulting in two trimers of mutant TNF-alpha being present in the molecule. - Fc(DANA)-TNC-TNF80 - corresponds to Fc-TNC-TNF80 with the heavy chain modifications D265A and N297A (DANA). - TNF80 - corresponds to the basic TNF80 mutant in its trimeric form. - TNC-scTNF80 - corresponds to a single chain molecule comprising three TNF80 mutant molecules, the chain further comprising a TNC peptide resulting in trimerization of the three single chain polypeptides.

[0307] Each of the above mouse soluble TNF80-based constructs was used to express the molecule. The constructs were then purified by anti-Flag affinity purification utilizing the Flag tag contained in all constructs. They were then analyzed by SDS-PAGE and silver staining, with the results shown in FIG. 24, part (B). Gel filtration analysis of the purified proteins shown in FIG. 24, panel (B) was then performed, with the results shown in FIG. 24, panel (C). The gel filtration results show that the expressed binding molecules do not aggregate and represent homogenous protein species. The methodology used in this example was as described in the Materials and Methods of Example 1 above.

[0308] [Example 21] Agonistic activity of different formats of soluble murine TNF80. The ability of the constructs discussed in Example 20 to act as agonists of TNFR2 was then assessed using Kym-1 cells to determine the ability of the binding molecules to induce cell death, and HT1080-TNFR2 cells to determine the ability of the binding molecules to induce IL-8 release. The methodology described above in Materials and Methods of Example 1 was used. The results obtained are shown in Figure 25.

[0309] The top panel of Figure 25 shows the results of Kym-1 cells expressing TNFR2 stimulated with various muTNF80 variants at the indicated concentrations, then assessed the following day to determine the ability of the constructs to induce cell death using crystal violet staining. As can be seen, the TNC-sc(mu)TNF80 binding molecule gave the greatest reduction in cell viability, followed by the Fc(DANA)-TNC-TNF80 binding molecule.

[0310] The bottom panel of Figure 25 shows the results for HT1080-Bcl2-TNFR2 cells. HT1080-Bcl2-TNFR2 cells were stimulated with various TNF80 variants at the indicated concentrations shown in Figure 1. The following day, supernatants were harvested and analyzed by ELISA for their IL8 content. The data shown are from four independent experiments. As can be seen, basal murine TNF80 gave the lowest level of IL8 induction, whereas the other binding molecules stimulated higher levels of IL-8 release.

[0311] [Example 22] In vivo Treg expansion by Fc(DANA)-TNC-muTNF80 To demonstrate the ability of the binding molecule containing TNF80 to act as an agonist of TNFR2, an in vivo experiment was then carried out in mice using Fc(DANA)-TNC-muTNF80 binding molecule. The results obtained are shown in Figure 26. Mice carrying luciferase under the control of the Foxp3 promoter were injected once iv with the indicated amount of Fc(DANA)-TNC-muTNF80 and analyzed daily by bioluminescence imaging (BLI), as well as evaluated for Treg frequency on day 4 using FACS analysis. The results of the BLI analysis are shown in the top panel of Figure 26. Photographs of the BLI on day 0 and day 4 are shown in the middle panel of Figure 26. The results of the FACS analysis carried out on day 4 are shown in the bottom panel of Figure 26.

[0312] [Example 22] Demonstration of specific binding to TNFR2 Experiments were carried out to demonstrate the ability of the provided binding molecules to bind TNFR2 but not TNFR1. HEK293 cells were transiently transfected with (i) a TNFR2 expression plasmid; (ii) an expression plasmid encoding the extracellular domain of TNFR1 with a GPI anchor domain; or (iii) an empty vector (ev). The following day, various binding molecules bearing a GpL tag were evaluated for their ability to bind TNFR2 but not significantly bind TNFR1. The results obtained are shown in Figure 27. The proteins shown in Figure 27 were added to HEK293 cells at the indicated concentrations. After 1 hour of incubation at 37°C and washing twice with PBS, the remaining cell-associated GpL activity was determined. GpL-TNF was used as a positive control that binds to both TNF receptor types.

[0313] [Example 23] Reformatting of the known 005-B08 antibody To illustrate that "reformatting" known anti-TNFR2 antibodies to put them into the formats discussed herein results in useful agonists for TNFR2, even if the known antibody is an anti-TNFR2 antagonist antibody, the known TNFR2 antagonist was reformatted into some of the tetravalent formats provided. The 005-B08 antibody disclosed in WO2020 / 089474 was reformatted into some of the formats described herein. Thus, antibodies were generated in the formats IgG1(N297A)-HC:scFv, IgG1(N297A)-HC:TNC, and IgG1(N-RGY) in which all antigen binding sites originate from the 005-B08 antibody. HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the antibodies. The next day, the supernatants were analyzed by ELISA for their IL8 content as a measure of agonist activity. The results obtained are presented graphically in FIG. 28. The maximum response induced by C4-IgG1(N297A)-HC:C4 as a positive control is shown by the dotted line in the graph. Surprisingly, all reformatted tetravalent constructs showed good agonist activity, even though the original 005-B08 antibody is an antagonist of TNFR2, demonstrating the effect of the antibody format.

[0314] [Example 24] Varying linker lengths and scFv positions in tetravalent formats The possible effect of the length of the linker within the scFv and whether the "southern" scFv group is linked to the C-terminus of the light or heavy chain was examined for the "north" and "south" antibody structures shown in Figure 29 A. The different antibody variants differed in the length of the linker linking the variable heavy and light chains of the scFv, in the presence of either three ((G4S)3) or four ((G4S)4) repeats of the basic GGGS (G4S) linker, and whether the scFv group was linked to the antibody backbone light or heavy chain.

[0315] Gel filtration of purified C4-IgG1(Durv)-HC:scFvC4(G4S)4 and C4-IgG1(Durv)-LC:scFvC4(G4S) was performed and the results obtained are shown in FIG. 29B. The ability of the constructs to agonize TNFR2 was also tested using IL-8 release as a measure of agonistic activity and the results obtained are shown in FIG. 29C. To obtain these results, HT1080-Bcl2-TNFR2 cells were stimulated with supernatants from HEK293 cells transfected with the indicated constructs. The following day, the supernatants were analyzed for their IL8 content by ELISA. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. The tetravalent constructs worked well regardless of the linker length and whether the scFv group was attached to the heavy or light chain, illustrating the robust nature of the provided tetravalent antibody format.

[0316] [Example 25] Comparison of C4 variant sequences in the C4-IgG1(Durv)-HC / LC:scFvC4 format Variants of the C4 variable region were generated in silico and the sequence was optimized for humanization, deimmunization and removal of PTM instability to help further optimize the sequence. The C4 variant sequence was used in the C4-IgG1(Durv)-HC / LC:scFvC4 format. Protein was produced using HEK293 cells and purified via protein A and size exclusion chromatography. The purified variants were evaluated using gel filtration chromatography (results shown in FIG. 30, panel A) and IL-8 release as a measure of TNFR2 agonist activity (results shown in FIG. 30, panel B). To measure IL-8 release, HT1080-Bcl2-TNFR2 cells were stimulated with the indicated constructs. The parental construct C4-IgG1(N297A)-HC:scFvC4 was included for comparison. The following day, the supernatants were analyzed for their IL8 content by ELISA. The maximum response induced by the highly active oligomeric TNFR2 specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. As can be seen from the results shown in Figure 30, the C4 variant was highly effective in the IL-8 assay.

[0317] [Example 26] Generation and evaluation of VHH antibodies To provide further options for generating TNFR2 antibodies. A VHH discovery campaign was initiated. Two TNFR2 target proteins (human-ECD-Fc and mouse ECD-Flag-Fc) and a control protein were used for phage display screening of a single domain antibody (VHH) library (Creative Biolabs library CaVHHL-4). Alternative biopanning was performed for four rounds for both targets, and specific VHHs were identified for both targets, while negative binders for the control Fc tag were removed. DNA sequencing was performed on 153 clones, and 14 unique positive sequences were identified.

[0318] Eight of the 14 clones demonstrated specific binding to human TNFR2 expressed on the cell surface. The ability of VHH antibodies to bind TNFR2 was tested using VHH-Fc-GpL format proteins, each with a different specific VHH present, and the results obtained are shown in Figure 31. The GpL reporter domain was cloned C-terminal to the Fc domain of the antibody. Equilibrium binding assays of TNFR2-specific VHH-Fc-GpL sdAb constructs were performed at 37°C using cells expressing human TNFR2, and HEK293 cells were used as non-transfected control cells. The results obtained are shown in Figure 31. Specific binding (open circles) was obtained by subtracting the non-specific binding values ​​(Hek293 cells-cross) from the binding values ​​of transfectants expressing human TNFR2 (total binding-filled circles). Specific binding was analyzed by the "Non-linear regression curve for one-site specific binding" function of GraphPad Prism5 software. As can be seen from FIG. 31, the variants showed good specific binding to TNFR2.

[0319] Equilibrium experiments were also performed using TNFR2 from different species, namely Mus musculus (mouse), Macaca fascicularis (cynomolgus monkey), and Chlorocebus sabaeus (African green monkey), to test the ability of VHH regions to bind to TNFR2 from different species. The ability to cross-react with TNFR2 from different species can be useful to allow antibodies to be tested in various models before evaluation in human patients. The results obtained from the experiments are presented in Table 8 below, where "nsb" indicates no specific binding.

[0320] [Table 13]

[0321] [Example 27] Hexavalent antibodies using VHH antibody domains To examine the effect of antibody format on binding affinity, VHH domain antibodies were then used to generate hexavalent antibodies with 3xVHH-Fc-GpL format antibodies generated for multiple VHH antibody domains. All VHH domains in a given hexavalent antibody were the same. Equilibrium binding of those antibodies was then examined using the same approach as in Example 26. Figure 32 shows the equilibrium binding results at 37°C of the indicated TNFR2-specific 3xVHH-Fc-GpL sdAb constructs to cells expressing human TNFR2. The GpL reporter domain is cloned C-terminal to the Fc domain. The results obtained strikingly showed that the hexavalent format antibodies exhibited high affinity binding, with the variation in Kd between different VHHs being generally much smaller compared to the bivalent antibody results shown in Figure 31. The result of having multiple binding sites in close spatial proximity in hexavalent and higher valency formats may be more significant in terms of binding affinity than a specific (bivalent) VHH, illustrating the tight binding nature of the hexavalent format.

[0322] [Example 28] Evaluation of blocking TNFα binding to TNFR2 VHH-Fc(DANA) hexavalent format antibodies with VHH:C18, VHH:C74, VHH:C188 or VHH:C238 VHH domains as binding sites were compared for their ability to block TNFα binding to TNFR2. The results obtained are shown in Figure 33. HT1080-Bcl2-TNFR2 cells were pretreated with the indicated amounts of TNFR2-specific sdAb constructs or with the CD40-specific construct 3xVHH:CD40(V12t)-Fc(DANA) as a negative control specific for CD40 but not for TNFR2. 5ng / ml GpL-TNF was added to all samples and binding was then measured after 1 hour of incubation at 37°C. The left bar graph in Figure 33 shows the results of the positive control, i.e. in the absence of another construct, thus 100% binding of TNFα to TNFR2. The following bar graph shows the results for the indicated constructs with activities expressed as a percentage of the negative control value. The 3xVHH-Fc(DANA) variants of the hexavalent antibody format VHH:C18, VHH:C74, VHH:C188, and VHH:C238 all demonstrated the ability to potently block TNF binding to the TNFR2 receptor. V12t CD40-VHH is “A Bispecific Antibody Antagonizes Prosurvival CD40 Signaling and Promotes Vg9Vd2 T cell-Mediated Antitumor Responses in Human B-cell Malignancies A. C Iris de Weerdt, Roeland Lameris, George L. Scheffer, Jana Vree, Renate de Boer, Anita G. Stam, Rieneke van de Ven, Mark-David Levin, Steven T. Pals, Rob C. Roovers, Paul WHI Parren, Tanja D. de Gruijl, Arnon P. Kater, and Hans J. van der Vliet”.

[0323] [Example 29] Ability of hexavalent constructs to specifically bind to TNFR2 but not to TNFR1 HEK293 cells were transiently transfected with expression vectors encoding TNFR1 and TNFR2 and then analyzed for binding of various hexavalent constructs, with the results shown in FIG. 34. The hexavalent constructs evaluated are shown in FIG. 34. Binding to empty vector-transfected cells served as a negative control for non-specific binding. TNF-GpL served as a positive control for successful TNFR1 binding. By comparing the results in the graph on the left for TNFR1 binding with the corresponding graph on the right for TNFR2 binding, it can be seen that all hexavalent constructs show specific binding to TNFR2 compared to TNFR1, which may allow selective targeting to TNFR2.

[0324] [Example 30] VHH:C18, VHH:C188, and VHH:C238 bind to the CRD4 of TNFR2, and VHH:C74 binds to the CRD3. Black 96-well plates were coated with protein G and then loaded with 3xVHH-Fc (DANA) versions of the indicated TNFR2-specific VHHs. After removal of unbound proteins, the immobilized antibody constructs were analyzed for binding of GpL-tagged deletion mutants of TNFR2(ed) and shown in the Western blot in the right panel. Finally, the binding of these GpL-fusion proteins was quantified as shown in Figure 35. The binding of the VHH clones to the different CRD domains is represented diagrammatically in Figure 15.

[0325] [Example 31] Tetravalent N-terminal-only VHH-based constructs are potent TNFR2 agonists Next, the ability of the tetravalent "North only" TNFR2-specific sdAb variants to act as TNFR2 agonists was examined and the results obtained are shown in Figure 36. HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the indicated mono- or biparatopic constructs. The following day, supernatants were analyzed for their IL8 content by ELISA. Maximal responses induced by the highly active oligomeric TNFR2-specific TNF mutants TNF(143N / 145R) (top panel) or C4-IgG1(N297A)-HC:C4 (middle and bottom panels) serve as positive controls and are indicated by dotted lines. The tetravalent "North" only constructs were potent agonists.

[0326] [Example 32] Hexavalent N-terminal-only TNFR2-specific sdAb variants are potent TNFR2 agonists Next, the ability of the hexavalent "Nose only" VHH antibodies to act as TNFR2 agonists was investigated using both IL-8 release and cell viability as markers. The results obtained are shown in Figure 37. For the results shown in panel (A), HT1080-Bcl2-TNFR2 cells were stimulated with the indicated constructs. The following day, the supernatants were analyzed for their IL8 content by ELISA. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line as a positive control. For the results shown in panel (B), Kym-1 cells were stimulated with the various hexavalent variants and the following day cell viability was determined using the MTT assay. The maximum killing induced by 200ng / ml highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is again shown by the dotted line. This assay measures TNFR2 activation and signaling through a pathway that is distinct from the pathway assessed by measuring IL-8 release. Overall, the results show that the hexavalent "Nose only" format exhibits strong TNFR2 activity through the VHH domain.

[0327] [Example 33] Side-by-side comparison of agonist activity of VHH:C188-Fc, 2xVHH:C188-Fc-GpL, and 3xVHH:C188-Fc The potency of different antibody formats using the same VHH domain was compared. VHH:C188-Fc, 2xVHH:C188-Fc-GpL, and 3xVHH:C188-Fc formats, each using the C188 VHH domain, were compared and the results obtained are shown in Figure 38. Figure 38, panels (A) and (B) show the results of purified proteins analyzed by SDS-PAGE (A) and gel filtration (B). Panel (C) shows the results of IL-8 release. HT1080-Bcl2-TNFR2 cells were stimulated with the indicated constructs. The next day, the supernatants were analyzed for their IL8 content by ELISA. The maximum response induced by the highly active oligomeric TNFR2 specific TNF mutant TNF(143N / 145R) positive control is shown by the dotted line. Panel (D) shows the results of cell viability in Kym-1 cells stimulated with various VHH:C188 variants as indicated. The next day after stimulation, cell viability was determined using an MTT assay. The maximum killing induced by C4-IgG1(N297A)-HC:scFvC4 is indicated by the dotted line. The most potent of the three formats was the hexavalent 3xVHH:C188-Fc antibody format, which showed the greatest stimulation of IL-8 release and reduction in cell viability.

[0328] [Example 34] Side-by-side comparison of IL8 and cell death-inducing activities of VHH-Fc-GpL, 2xVHH-Fc-GpL, and 3xVHH-Fc-GpL variants of C18 and C74 Further comparison of different antibody formats with the same VHH binding domain was performed for the VHH-Fc-GpL, 2xVHH-Fc-GpL and 3xVHH-Fc-GpL variants of C18 and C74 using the same IL-8 and cell viability assays as in the previous examples. The results obtained are shown in Figure 39. HT1080-Bcl2-TNFR2 cells were stimulated with the indicated constructs. Panel (A) shows the results of the supernatants analyzed for their IL8 content by ELISA. Panel (B) shows the results of a cell viability assay in which Kym-1 cells were stimulated with the different variants and the following day cell viability was determined using an MTT assay. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown. Interestingly, the hexavalent format was the most active in both assays for the C18 variant. The hexavalent format of C74 was the best format for stimulating IL-8 release, but neither format induced a significant decrease in cell viability, suggesting that the C74 VHH antibody does not induce both TNFR2-mediated signaling pathways, but simply stimulates signaling in the pathway that leads to IL-8 release.

[0329] [Example 35] The length of the linker between the VHH domains of the 3xVHH:C188-IgG1(Durv) variant does not affect TNFR2 agonism Next, the effect of separating the VHH domains by progressively longer linkers was investigated for the 3xVHH:C188-IgG1(Durv) antibody format using progressively longer linkers. The different antibodies compared had a single G4S linker (GSSSS) between the VHH domains, three linkers between domain (G4S)3, or five linkers between domain (G4S)5. The results obtained are shown in Figure 40. Panel (A) shows the domain structure of the constructs. Panel (B) shows the SDS-PAGE analysis of the purified constructs and panel (C) shows the results of the gel filtration analysis. Panel (D) shows the results of an IL-8 release assay, where HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the indicated constructs. The following day, the supernatants were analyzed by ELISA for their IL8 content. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown by the dotted line. The parent molecule 3xVHH:C188-Fc(DANA) was included for comparison. As can be seen from Figure 40, panel (D) shows that increasing the overall linker length did not affect TNFR2 agonism as measured by IL-8 release, further illustrating the robust nature of the hexavalent format.

[0330] [Example 36] Constructs of different domain architectures with 2, 3, 4, 6, or 9 VHH:C188 domains induce strong TNFR2 agonism Different antibody formats were further compared using the C188 VHH domain. The results obtained are shown in Figure 41. Panel (A) shows the domain structure of the antibodies. TNC refers to the trimerization domain of tenascin-C. Panel (B) shows the results of the TNFR2 agonism activity as measured by IL-8 release. HT1080-Bcl2-TNFR2 cells were stimulated with the indicated constructs. The next day, the supernatants were analyzed by ELISA for their IL8 content. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown by the dotted line. Constructs with at least four VHH domains generally gave the best results, especially those antibodies in which at least two VHH domains are present in the same polypeptide.

[0331] [Example 37] Protein G enhances the agonist activity of a sequence-optimized variant of the anti-TNFR2 mAb C4 As shown in Figure 42, enhanced activity in the presence of protein G was obtained for C4 variant Abs, indicating the presence of potential agonism that can be achieved by multimerization of monoclonal antibodies. (A) Western blot analysis with anti-human IgG1 antibody of cell culture supernatants of HEK293 cells transiently transfected with expression plasmids encoding the light and heavy chains of the indicated antibody variants. (B) HT1080-TNFR2 cells were challenged with the indicated concentrations of C4 variants in the presence and absence of 500 ng / ml protein G. The next day, cell supernatants were analyzed by ELISA for IL8 production. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is indicated by the dotted line.

[0332] [Example 38] TNFR2 agonism of sequence-optimized variants of 3xVHH:C188-Fc(Durv) Fifteen variants were generated in silico by humanizing, deimmunizing, and optimizing sequences for removal of PTM instability. For one variant (variant 14), constructs with various linker lengths between the different VHHs were also generated. All variant constructs were successfully produced, as shown in Figure 43, panel (A): Western blot analysis with anti-human IgG1 antibody of cell culture supernatants of HEK293 cells transiently transfected with expression plasmids encoding the indicated 3xVHH:C188-Fc(Durv) variants. In the 3xVHH:C188-Fc(Durv) format, all variants were bioactive, as can be seen in panel (B) showing the results of an IL-8 release assay in which HT1080-Bcl2-TNFR2 cells were stimulated with supernatants containing the indicated constructs: HT1080-TNFR2 cells were challenged with the indicated concentrations of 3xVHH:C188-Fc(Durv) variants, and the parental 3xVHH:C188-Fc(DANA) molecule. The next day, cell supernatants were analyzed by ELISA for IL8 production. The maximum response induced by the highly active oligomeric TNFR2-specific TNF mutant TNF(143N / 145R) is shown by the dotted line.

[0333] [Example 39] TNFR2 agonists increase Tregs in human peripheral blood mononuclear cells (PBMCs) The ability of specific TNFR2 binding molecules to expand regulatory T cells and the ability of different TNFR2 binding sites to expand peripheral blood mononuclear cells was evaluated. Results obtained for tetravalent antibodies in Fab-HC-scFv format (C4-IgG1(N297A)-HC:scFvC4, C19-IgG1(N297A)-HC:scFvC19, and C40-IgG1(N297A)-HC:scFvC40) and hexavalent VHH constructs in 3xVHH-Fc format (3xVHH:C188-Fc(Durv)) are shown in Figure 44, and expansion data are shown in Figure 21. PBMCs were isolated from the indicated number of donors per condition and cells from each donor were evaluated individually. PBMCs were isolated from 10 cells / mL of donor-free donor cells, and 10 cells / mL of donor-free donor cells were evaluated. 7The cells were cultured at a density of 10 cells / ml for 2 days. Then, PBMCs were harvested and plated in multi-well plates at 10 cells / ml. 6 Cells were seeded at a density of 1000 / ml, 500,000 cells / well. After 4 days of culture, cells were stained with antibodies specific for CD3, CD8, CD4, CD25, and FoxP3 for flow cytometry analysis. Figure 44 shows the fold change in Treg frequency in human peripheral blood mononuclear cells (PBMCs) after 4 days of stimulation with the indicated TNFR2...

Claims

1. A binding molecule that specifically binds to TNFR2 but not to TNFR1, which is an FcγR-independent agonist of TNFR2 and has at least bivalency with respect to binding to TNFR2.

2. The binding molecule of claim 1 having at least four valencies.

3. (a) having 6 valency with respect to TNFR2; or (b) having tetravalency with respect to TNFR2; The binding molecule of claim 1.

4. The binding molecule of claim 1 , wherein at least one antigen-binding domain specific for TNFR2 of the binding molecule is a single domain binder (sdB) specific for TNFR2, preferably a single domain antibody (sdAb) specific for TNFR2.

5. The binding molecule of claim 4, wherein at least one sdBr (single domain binder) specific for TNFR2 is a VHH.

6. 2. The binding molecule of claim 1, comprising six TNFR2-specific antigen-binding sites, each of which is a VHH antigen-binding domain specific for TNFR2.

7. (a) comprises six VHH antigen-binding domains and an Fc region, and is preferably a hexavalent antibody; (b) comprising two polypeptides, each polypeptide comprising three VHH binding domains specific for TNFR2, preferably a hexavalent antibody; (c) comprising two polypeptides, each polypeptide comprising three VHH binding domains specific for TNFR2, each polypeptide further comprising a heavy chain constant region sequence at its C-terminal portion comprising CH2-CH3 but not CH1, preferably a hexavalent antibody; or (d) comprising two polypeptides, each polypeptide comprising three VHH binding domains specific for TNFR2, each polypeptide further comprising a heavy chain constant region sequence at its C-terminal portion comprising CH2-CH3 but not CH1, and comprising modifications meaning that the Fc domain has reduced effector function, preferably a hexavalent antibody; The binding molecule of claim 6.

8. (a) an antigen-binding site comprising a VHH domain comprising a set of three CDRs selected from the set of SEQ ID NOs: 213 / 214 / 215; 216 / 217 / 218; 219 / 220 / 221; 222 / 223 / 224; 225 / 226 / 227; 228 / 229 / 230; 231 / 232 / 233; 234 / 235 / 236; 237 / 238 / 239; 240 / 241 / 242; 243 / 244 / 245; 246 / 247 / 248; 249 / 250 / 251; 252 / 253 / 254; or a variant set of CDRs of any set thereof having up to five amino acid sequence changes in total, wherein the variant antigen-binding site retains the ability to bind TNFR2; (b) an antigen-binding site comprising a VHH domain comprising the three CDRs of SEQ ID NO: 154 / 155 / 156, or a variant set of three such CDRs having a total of up to five amino acid sequence changes, wherein the variant antigen-binding site retains the ability to bind TNFR2; (c) an antigen-binding site comprising a set of three CDRs selected from 154 / 155 / 156; 160 / 161 / 162; 163 / 164 / 165; 166 / 167 / 168; 169 / 170 / 171; 181 / 182 / 183; 187 / 188 / 189; 193 / 194 / 195, or any variant set thereof having up to five amino acid sequence changes, wherein the antigen-binding site retains the binding ability of TNFR2; (d) is hexavalent, containing six VHH antigen-binding sites which are all the same and which are antigen-binding sites defined in any of (a) to (c); (e) a hexavalent antibody comprising two polypeptides, each comprising three VHH antigen-binding sites, wherein all six antigen-binding sites are the same antigen-binding site as defined in any of (a) to (c); (f) is hexavalent, comprising six VHH antigen-binding sites, at least two of the VHH antigen-binding sites having different specificities for TNFR2, and at least one of those specificities is conferred by a VHH antigen-binding site defined in any of (a) to (c); (g) is hexavalent, comprising six VHH antigen-binding sites, at least three of which have different specificities for TNFR2, and at least one of those specificities is conferred by a VHH antigen-binding site defined in any of (a) to (c); or (h) is hexavalent, comprising six VHH antigen-binding sites, all six VHH antigen-binding sites having different specificities for TNFR2, at least one of which is conferred by a VHH antigen-binding site defined in any of (a) to (c); The binding molecule of claim 1.

9. (a) comprises at least one VHH antigen-binding domain of SEQ ID NOs: 198-211, or a variant thereof with at least 90% sequence identity which is still capable of binding to TNFR2; (b) comprises at least one VHH antigen-binding domain selected from SEQ ID NO: 140 or any variant thereof having at least 90% sequence identity; (c) comprises at least one VHH antigen-binding domain selected from the domains of SEQ ID NOs: 140, 142-145, 149, 151, 153, or any variant thereof having at least 90% sequence identity that is still capable of binding to TNFR2; (d) comprising two polypeptides each comprising at least one VHH antigen-binding domain as defined in (a), (b), or (c); (e) comprises two polypeptides each comprising three VHH antigen-binding domains as defined in (a), (b), or (c); (f) comprising two polypeptides each comprising three VHH antigen-binding domains as defined in (a), (b), or (c), wherein the antigen-binding domains of the binding molecules specific for TNFR2 are all identical; or (g) comprising two polypeptides each comprising three VHH antigen-binding domains as defined in (a), (b), or (c), wherein not all VHH antigen-binding domains of a binding molecule specific for TNFR2 are the same; The binding molecule of claim 1.

10. (a) each of which: (i) the amino acid sequence of SEQ ID NO: 197, except that the VHH domain corresponding to SEQ ID NO: 140 in that sequence is instead one of SEQ ID NOs: 198-211; or (ii) a variant amino acid sequence having at least 90% sequence identity to (i), which is still capable of binding to TNFR2; The polypeptide comprises two polypeptides having the following structure: (b) each comprising two polypeptides having the amino acid sequence of SEQ ID NO: 197, or a variant amino acid sequence with at least 90% sequence identity that is still capable of binding to TNFR2, preferably each polypeptide comprising three such VHH binding domains; (c) each of (i) the amino acid sequence of SEQ ID NO: 197, except that the VHH domain corresponding to SEQ ID NO: 140 in the sequence is instead one of SEQ ID NOs: 140, 142-145, 149, 151, or 153; (ii) an amino acid sequence having at least 90% sequence identity with (i) and still capable of binding to TNFR2; The polypeptide comprises two polypeptides having the following structure: (d) any one of (a)-(c), wherein the variant amino acid sequence has from 1 to 5 amino acid sequence changes compared to the specific sequence, and the binding molecule retains the ability to bind to TNFR2, preferably the variant has only 1, 2, or 3, more preferably 1 amino acid sequence changes compared to the specific sequence; or (e) any one of (a) to (c), wherein the variant amino acid sequence has the sequence of amino acids 302 to 405 of SEQ ID NO: 197 replaced with the amino acid sequence of SEQ ID NO: 327; The binding molecule of claim 1.

11. (a) a polypeptide having the amino acid sequence of SEQ ID NO: 330, or a variant thereof having at least 90% sequence identity that is still capable of binding to TNFR2; (b) a polypeptide having the amino acid sequence of SEQ ID NO: 331, or a variant thereof having at least 90% sequence identity that is still capable of binding to TNFR2; or (c) comprising three polypeptides of (a) or three polypeptides of (b); The binding molecule of claim 1.

12. The binding molecule of claim 1, which is a tetravalent antibody having binding sites towards both the "N-terminus" and the "C-terminus" of TNFR2.

13. (a) an antigen-binding site comprising a set of six CDRs: (i) the light chain CDRs are selected from a CDR1 / CDR2 / CDR3 set selected from SEQ ID NOs: 266 / 267 / 268; 269 / 270 / 271; 272 / 273 / 274; 275 / 276 / 277, and the heavy chain CDRs are selected from a CDR1 / CDR2 / CDR3 set selected from 282 / 283 / 284; 285 / 286 / 287; 288 / 289 / 290; or 291 / 292 / 293; or (ii) a variant of the set of six CDRs of (i), wherein the set of six CDRs has a total of five or fewer amino acid changes compared to the specific six CDRs, and the antigen-binding site retains the binding ability of TNFR2; (b) an antigen-binding site comprising a set of six CDRs; (i) the light chain CDR1 / CDR2 / CDR3 are SEQ ID NOs: 69 / 70 / 71 and the heavy chain CDR1 / CDR2 / CDR3 are SEQ ID NOs: 102 / 103 / 104; (ii) a variant of the set of six CDRs of (i), wherein the set of six CDRs has a total of five or fewer amino acid changes compared to the specific six CDRs, and the antigen-binding site retains the binding ability of TNFR2; (c) comprises an antigen-binding site comprising a set of three light chain CDRs and / or a set of three heavy chain CDRs from Table 5; (d) a set of six CDRs from the set of six set set forth in Table 5, or a variant set of six CDRs having five or fewer amino acid sequence changes that retain TNFR2 binding ability; (f) comprising the four antigen-binding sites of any one of (a) to (d), wherein all four antigen-binding sites are the same; (g) comprising four antigen-binding sites according to any one of (a) to (d), wherein all four antigen-binding sites are the same, and the antibody is tetravalent with respect to TNFR2; or (h) comprising four antigen-binding sites according to any one of (a) to (d), wherein all four antigen-binding sites are the same, and the antibody is tetravalent with respect to TNFR2 and comprises two N-terminally directed antigen-binding sites and two C-terminally directed antigen-binding sites. The binding molecule of claim 1.

14. (a)(i) two light chain polypeptides comprising an amino acid sequence having the sequence of SEQ ID NO:261, except that the amino acid sequence corresponding to SEQ ID NO:295 is replaced by a sequence selected from one of SEQ ID NOs:262-265; or (ii) two variant light chain polypeptides, each having at least 90% sequence identity to (i); and two light chain polypeptides, (b)(iii) two heavy chain polypeptides comprising an amino acid sequence having the sequence of SEQ ID NO:259, except that the amino acid sequence corresponding to SEQ ID NO:296 is replaced with a sequence selected from one of SEQ ID NOs:278-281; ​​or (iv) two variant heavy chain polypeptides, each having at least 90% sequence identity to (iii). and two heavy chain polypeptides, The binding molecule of claim 1.

15. (a) two light chain polypeptides having the sequence of SEQ ID NO:261, or an amino acid sequence having at least 90% amino acid sequence identity to SEQ ID NO:261; or (b) two heavy chain polypeptides having an amino acid sequence of SEQ ID NO:259, or an amino acid sequence having at least 90% amino acid sequence identity to SEQ ID NO:259; The binding molecule of claim 1 which is an antibody comprising:

16. (a) comprises a light chain of SEQ ID NO: 332 and a heavy chain of SEQ ID NO: 333, or a variant thereof having at least 90% sequence identity, and is still capable of acting as a TNFR2 agonist; (b) comprises a light chain of SEQ ID NO: 334 and a heavy chain of SEQ ID NO: 335, or a variant thereof having at least 90% sequence identity, and is still capable of acting as a TNFR2 agonist; (c) comprises a light chain of SEQ ID NO: 336 and a heavy chain of SEQ ID NO: 337, or a variant thereof having at least 90% sequence identity, and is still capable of acting as a TNFR2 agonist; or (d) comprises two light chains and two heavy chains, all as defined in (a), (b), or (c). The binding molecule of claim 1.

17. (a) oligomerizing TNFR2 on the surface of a cell expressing TNFR2, preferably clustering six or more TNFR2 molecules on the cell surface; (b) induces TNFR2 signaling; and / or (c) stimulating the proliferation of leukocytes, preferably T cells, more preferably Treg cells; The binding molecule of claim 1 .

18. (a) a moiety that modifies serum half-life, preferably comprising a moiety selected from the group consisting of an Fc tail, serum albumin, a moiety that is a conjugate of serum albumin, and PEG; (a) is bivalent with respect to TNFR2 binding; (b) is trivalent with respect to TNFR2 binding; (c) is tetravalent with respect to TNFR2 binding; or (d) having at least tetravalency with respect to TNFR2 binding; The binding molecule of claim 1.

19. (a) is an antibody; (b) an antibody comprising a ligand for TNFR2; (c) comprising at least a portion of an antibody constant region, preferably an Fc region; or (d) a ligand for TNFR2, preferably comprising a ligand derived from TNFα or lymphotoxin-alpha; Preferably, (a) an antibody that contains TNFR2 binding sites at both the N-terminus and C-terminus of one or more heavy chains of the antibody; or (b) an antibody that contains TNFR2 binding sites at both the N-terminus and C-terminus of one or more light chains of the antibody; The binding molecule of claim 1.

20. (a) the antibody is a bivalent antibody; (b) the antibody is a trivalent antibody; (c) the antibody is a tetravalent antibody; (d) the antibody is a hexavalent antibody; (e) the antibody is a heavy chain-only antibody; (f) the antibody is a single domain antibody, such as a single domain Ab (sdAb), VHH, or nanobody; (g) the antibody is a multimer of at least two antibodies; (h) the antibody comprises or is linked to a multimerizing moiety, preferably tenascin-c peptide (TNC), and is preferably a multimer of at least two antibodies associated through the tenascin-c peptide sequence; (i) the antibody is linked to one or more molecules derived from a ligand of TNFR2, preferably derived from TNF-α or lymphotoxin alpha; 20. The binding molecule of claim 19.

21. (a) the antibody lacks an Fc region; (b) the antibody comprises an altered Fc region that is unable to bind to an Fc receptor; or (c) the antibody comprises a heavy chain constant region that lacks CH1.

20. The binding molecule of claim 19.

22. - IgG with Fc modifications that eliminate Fc receptor binding; - scFv:IgG, where IgG has an Fc modification that eliminates Fc receptor binding; - IgG with a modification that eliminates Fc receptor binding, IgG-HC:TNC; - IgG with modifications that eliminate Fc receptor binding, scFv:IgG-HC:TNC; - 6-mer IgG; - 3 (VHH-TNC) an antibody in an antibody format selected from: - HC represents the heavy chain of IgG; - LC represents the light chain of IgG; - scFv: IgG represents an scFv linked to or forming part of an IgG molecule; - HC: TNC denotes a heavy chain (HC) comprising or linked to a tenascin-c trimerization domain peptide (TNC); - VHH-TNC denotes tenascin linked or bound to a VHH, The binding molecule of claim 1.

23. -Fab-LC: scFv; - IgG-HC:scFv, where the IgG has a modification that eliminates Fc receptor binding, preferably N297A; - IgG-LC:scFv, where the IgG has a modification that eliminates Fc receptor binding, preferably N297A; - IgG-LC:scFv-HC:scFv; - LC: scFv-HC: scFv-IgG-HC: scFv; - LC: scFv-HC: scFv-IgG-LC: scFv; - LC:scFv-HC:scFv-IgG-LC:scFv-HC:scFv; and - IgG-HC-TNC-scFv; - sdAb / VHH; -3 (VHH-TNC), an antibody in an antibody format selected from the group consisting of - HC represents the heavy chain of IgG; - LC represents the light chain of IgG; - LC: scFv denotes an scFv linked to or forming part of a light chain; - HC: scFv denotes the scFv linked to or forming part of the heavy chain; - HC: TNC denotes a heavy chain (HC) comprising or linked to tenascin (TNC); - TNC-scFv denotes tenascin linked or bound to a scFv; - TNC-VHH denotes tenascin linked or bound to a VHH; - sdAb stands for single domain antibody; - VHH denotes the variable region of a camelid antibody; - VHH-TNC denotes tenascin linked or bound to a VHH, Preferably, it is an antibody, -Fab-scFv; - IgG-HC: scFv; - LC: scFv-HC: scFv-IgG1; and - LC: scFv-HC: scFv-IgG1-TNC The antibody is an antibody in an antibody format selected from The binding molecule of claim 1.

24. (a) comprises at least two antigen-binding sites with different specificities for TNFR2; (b) all of the antigen-binding sites of the binding molecules specific for TNFR2 each have a different specificity for TNFR2; (c) two polypeptides, each polypeptide comprising three VHH antigen-binding domains specific for TNFR2, at least two of the three VHH antigen-binding domains having different specificities for TNFR2; or (d) two polypeptides, each polypeptide comprising three VHH antigen-binding domains specific for TNFR2, said three VHH antigen-binding domains having different specificities for TNFR2; The binding molecule of claim 1.

25. a binding molecule that specifically binds to TNFR2 but not to TNFR1, the binding molecule being an FcγR-independent agonist of TNFR2 and having at least bivalency with respect to binding to TNFR2, and comprising a mutant TNF-alpha that binds to TNFR2 but not to TNFR1, and a polypeptide comprising an Fc region, or a tenascin peptide sequence that oligomerizes the polypeptide, or both; Preferably, (a) does not contain antibody sequences; (b) a mutant TNF-alpha that binds to TNFR2 but not to TNFR1, and a polypeptide that includes a tenascin peptide; (c) the TNF-alpha mutant is TNF80; and / or (d) having at least trivalency with respect to TNFR2, preferably hexavalency with respect to TNFR2; A binding molecule, a binding molecule.

26. 26. A binding molecule according to any one of claims 1 to 25 for use as a medicament.

27. 26. A binding molecule according to any one of claims 1 to 25 for use in a method for treating or preventing an autoimmune or inflammatory disorder, preferably comprising (a) for use in a method wherein the disorder is graft-versus-host disease (GvHD), preferably wherein the binding molecule is administered prior to, simultaneously with, or following cell, tissue, or organ transplantation; or (b) the disorder is one involving impaired function or unwanted proliferation of leukocytes, preferably T cells, more preferably Treg cells; or (c) the binding molecule, wherein the disorder is selected from inflammatory bowel disease (e.g., ulcerative colitis, Crohn's disease, or celiac disease), atherosclerosis, lupus, multiple sclerosis, type I diabetes, myasthenia gravis, pemphigus vulgaris, and bullous pemphigoid.

28. 26. A method of stimulating cell proliferation comprising contacting a target cell expressing TNFR2 with a binding molecule according to any one of claims 1 to 25.

29. 26. A pharmaceutical composition comprising a binding molecule according to any one of claims 1 to 25 and a pharma- ceutically acceptable carrier.

30. 26. A method of detecting TNFR2 comprising contacting a test sample with a binding molecule of any one of claims 1 to 25 and detecting binding of the binding molecule to TNFR2, Preferably, the method, wherein said binding molecule is labeled and binding of the binding molecule to TNFR2 is detected via the label.

31. 26. A binding molecule according to any one of claims 1 to 25 for use in a method of diagnosis, the method comprising administering to a subject a binding molecule according to any one of claims 1 to 23 and detecting binding of the binding molecule to TNFR2, Preferably, said binding molecule is labeled and binding of the binding molecule to TNFR2 is detected via said label.

Images