Anti-GAL9 immunoinhibitory binding molecule

Abstract

Inhibitory anti-GAL9 binding molecules, antibody constructs, pharmaceutical compositions comprising the binding molecules and antibody constructs, and methods for their use are presented.

Description

[Technical Field]

[0001] 1. CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit under 35 U.S.C. §119(e) of previously copending U.S. Provisional Patent Application No. 62 / 900,105, filed September 13, 2019, and U.S. Provisional Patent Application No. 62 / 855,590, filed May 31, 2019.

[0002] 2. Sequence Listing This application has been submitted via EFS-Web and contains a Sequence Listing, which is incorporated herein by reference in its entirety. The ASCII copy was created in XX month of 2020, is named XXXXXUS_sequencelisting.txt, and is X,XXX,XXX bytes in size. [Background technology]

[0003] 3. Background Autoimmune diseases arise from an imbalance within the immune system, resulting in immune-mediated attacks against the body's own cells and tissues. The current "gold standard" care for autoimmune diseases is systemic immunosuppression with immunosuppressants, including corticosteroids, anti-cytokine antibodies such as anti-TNF-α, anti-IL-1, anti-IL-5, anti-IL-6, anti-IL-17, and anti-IL-23 antibodies, and small molecule drugs that reduce inflammatory cytokine signaling, such as JAK / STAT inhibitors. However, nonspecific systemic immunosuppression can leave patients susceptible to infections and other serious side effects.

[0004] Immunotherapy holds great potential for the treatment of autoimmune diseases. Galectin-9 (GAL9) is an S-type lectin beta-galactoside-binding protein with N- and C-terminal carbohydrate-binding domains connected by a linker peptide. GAL9 has been implicated in regulating cell-cell and cell-matrix interactions. GAL9 has been shown to bind to soluble PD-L2, suggesting that at least some of the immunological effects of PD-L2 are mediated by the binding of multimeric PD-L2 to GAL9, rather than by PD-1 (WO 2016 / 008005, incorporated herein by reference in its entirety). However, the mechanisms by which GAL9 and PD-L2 influence immune effector function have not yet been fully characterized. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] International Publication No. 2016 / 008005 Brochure [Patent Document 2] US Patent Application Publication No. 2018 / 0118811 [Patent Document 3] U.S. Patent No. 6,407,213 [Patent Document 4] International Publication No. 2018075692 Brochure [Patent Document 5] U.S. Patent No. 8,053,562 [Patent Document 6] U.S. Patent No. 9,527,927 [Patent Document 7] International Publication No. 2017 / 011342 Brochure [Patent Document 8] International Publication No. 2017 / 106462 Brochure [Patent Document 9] U.S. Patent No. 5,821,333 [Patent Document 10] U.S. Patent No. 8,216,805 [Patent Document 11] U.S. Patent No. 8,592,562 [Patent Document 12] U.S. Patent No. 9,248,182 [Patent Document 13] U.S. Patent No. 9,358,286 [Patent Document 14] US Patent Application Publication No. 2018 / 0118811 [Patent Document 15] International Publication No. 2018 / 075692 Brochure [Patent Document 16] U.S. Patent No. 8,624,003 [Patent Document 17] U.S. Patent No. 8,163,888 [Patent Document 18] U.S. Patent No. 5,208,020 [Patent Document 19] U.S. Patent No. 8,337,856 [Patent Document 20] U.S. Patent No. 5,773,001 [Patent Document 21] U.S. Patent No. 7,829,531 [Patent Document 22] U.S. Patent No. 5,208,020 [Patent Document 23] U.S. Patent No. 7,745,394 [Patent Document 24] International Publication No. 2017 / 136623 Brochure [Patent Document 25] International Publication No. 2017 / 015502 Brochure [Patent Document 26] International Publication No. 2017 / 015496 Brochure [Patent Document 27] International Publication No. 2017 / 015495 Brochure [Patent Document 28] International Publication No. 2004 / 010957 Pamphlet [Patent Document 29] International Publication No. 2005 / 077090 Brochure [Patent Document 30] International Publication No. 2005 / 082023 Brochure [Patent Document 31] International Publication No. 2006 / 065533 Pamphlet [Patent Document 32] International Publication No. 2007 / 030642 Pamphlet [Patent Document 33] International Publication No. 2007 / 103288 Brochure [Patent Document 34] International Publication No. 2013 / 173337 Brochure [Patent Document 35] International Publication No. 2015 / 057699 Brochure [Patent Document 36] International Publication No. 2015 / 095755 Brochure [Patent Document 37] International Publication No. 2015 / 123679 Brochure [Patent Document 38] International Publication No. 2015 / 157286 Brochure [Patent Document 39] International Publication No. 2017 / 165851 Brochure [Patent Document 40] International Publication No. 2009 / 073445 Brochure [Patent Document 41] International Publication No. 2010 / 068759 Brochure [Patent Document 42] International Publication No. 2010 / 138719 Brochure [Patent Document 43] International Publication No. 2012 / 171020 Brochure [Patent Document 44] International Publication No. 2014 / 008375 Brochure [Patent Document 45] International Publication No. 2014 / 093394 Brochure [Patent Document 46] International Publication No. 2014 / 093640 Brochure [Patent Document 47] International Publication No. 2014 / 160360 Brochure [Patent Document 48] International Publication No. 2015 / 054659 Brochure [Patent Document 49] International Publication No. 2015 / 195925 Brochure [Patent Document 50] International Publication No. 2017 / 160754 Brochure [Patent Document 51] US Patent Application Publication No. 2017 / 0137530 [Patent Document 52] U.S. Patent No. 8,961,964 [Patent Document 53] U.S. Patent No. 8,945,865 [Patent Document 54] U.S. Patent No. 8,420,081 [Patent Document 55] U.S. Patent No. 6,685,940 [Patent Document 56] U.S. Patent No. 6,171,586 [Patent Document 57] U.S. Patent No. 8,821,865 [Patent Document 58] U.S. Patent No. 9,216,219 [Patent Document 59] U.S. Patent Application Serial No. 10 / 813,483 [Patent Document 60] International Publication No. 2014 / 066468 Brochure [Patent Document 61] International Publication No. 2011 / 104381 ​​Brochure [Patent Document 62] International Publication No. 2016 / 180941 Brochure [Patent Document 63] International Publication No. 2016 / 008005 Brochure [Non-patent literature]

[0006] [Non-licensed Document 1] www.imgt.org / IM GTScientificChart / Numbering / Hu_IGHGnber.html#refs [Non-licensed Document 2] Fukushima A, Sumi T, Fukuda K, Kumagai N, Nishida Tら(2008) Roles of galectin-9 in the development of experimental allergic conjunctivitis in mice. IntArch Allergy Immunol 146, Volume: 36~43 [Non-licensed Document 3] Bonischら(Protein Engineering, Design & Selection, 2017, pages 1~12) [Non-licensed Document 4] Stickler (Genes Immun. April 2011; Volume 12 (No. 3): 213-221) [Non-licensed Document 5] Merchantら(Nature Biotech (1998) Volume 16: Pages 677~681) [Non-licensed Document 6] Igawaら(Protein Eng. Des. Sel, 2010, Volume 23, Pages 667~677) [Non-licensed Document 7] Brinkmannら(MABS, 2017, Volume 9, No. 2, Pages 182~212) [Non-licensed Document 8] Storz (MAbs. November-December 2015; Volume 7 (No. 6): 989-1009) [Non-licensed Document 9] Lambertら(Adv Ther, 2017, 34 volumes: 1015 pages) [Non-licensed Document 10] Diamantisら(British Journal of Cancer, 2016, Volume 114, Pages 362~367) [Non-Patent Document 11] Carrico et al. (Nat Chem Biol, 2007. 3:321-2) [Non-Patent Document 12] We et al. (Proc Natl Acad Sci USA, 2009. Vol. 106:3000-5) [Non-Patent Document 13] Rabuka et al. (Curr Opin Chem Biol., 2011, Vol. 14:790-796) [Non-Patent Document 14] Hudak et al. (Angew Chem Int Ed Engl., 2012: 4161-5) [Non-Patent Document 15] Rabuka et al. (Nat Protoc., 2012, Vol. 7:1052-67) [Non-Patent Document 16] Agarwal et al. (Proc Natl Acad Sci USA. 2013, Vol. 110: 46-51) [Non-Patent Document 17] Agarwal et al. (Bioconjugate Chem., 2013, Vol. 24: 846-851) [Non-Patent Document 18] Barfield et al. (Drug Dev. and D., 2014, Vol. 14: 34-41) [Non-Patent Document 19] Drake et al. (Bioconjugate Chem., 2014, Vol. 25: 1331-41) [Non-Patent Document 20] Liang et al. (J Am Chem Soc., 2014, Vol. 136: 10850-3) [Non-Patent Document 21] Drake et al. (Curr Opin Chem Biol., 2015, Vol. 28:174-80) [Non-Patent Document 22] York et al. (BMC Biotechnology, 2016, Vol. 16(1):23) [Non-Patent Document 23] Armour et al. (Eur. J. Immunol. Vol. 29(8) (1999) pp. 2613-2624) [Non-Patent Document 24] Shields et al. (J. Biol. Chem. Vol. 276 (No. 9) (2001) pp. 6591-6604) [Non-Patent Document 25] Oganesyan et al. (Acta Cristallographica D64 (2008) pp. 700-704) [Non-Patent Document 26] Fang et al. (Biological Procedures Online, 2017, Vol. 19:11) [Non-Patent Document 27] Kunkel, T.A. (PNAS 1985, January 1, Vol. 82(2), pp. 488-492) [Non-patent document 28] Oomizu, S. et al., PLoS One Volume 7(Issue 11):e48574(2012); Doi: 10.1371 / journal.pone.0048574 [Non-Patent Document 29] Wykes, MN et al., Eur J Immunol. (2009) 39:2004-7 [Non-Patent Document 30] Karunarathne et al., Immunity (2016), August 16; 45(2): 333-45 [Non-Patent Document 31] Kumar and Singh (Developability of biotherapeutics: computational approaches. Boca Raton: CRC Press, Taylor & Francis Group, 2016) Summary of the Invention [Problem to be solved by the invention]

[0007] There remains a need for more targeted therapies that can re-establish balance in the immune system by modulating immune effector cells to establish a more clinically favorable cytokine profile. Such therapeutic agents may be useful for improved treatment of autoimmune and inflammatory diseases. [Means for solving the problem]

[0008] 4. Overview The present invention is based, in part, on the unexpected discovery that PD-L2 is overexpressed in autoimmune diseases and that inhibition of the galectin-9 / PD-L2 pathway modulates immune effector cells to produce a more clinically favorable cytokine profile.

[0009] Accordingly, disclosed herein are various GAL9 binding molecules, antigen-binding portions thereof, and antibodies that specifically bind to and antagonize human GAL9 (galectin-9). Inhibition of GAL9 using the anti-human GAL9 binding molecules disclosed herein reduces the secretion and production of pro-inflammatory cytokines, increases the secretion and production of anti-inflammatory cytokines, and reduces the surface expression of stimulatory molecules.

[0010] Also disclosed are pharmaceutical compositions comprising the GAL9-binding molecules. The anti-GAL9-binding molecules, antigen-binding portions thereof, and antibodies disclosed herein can be used by themselves as pharmaceutical compositions or in combination with other therapeutic agents or procedures for treating, preventing, and / or diagnosing conditions that evoke an inflammatory response, such as autoimmune diseases, inflammatory diseases, or infectious diseases. Anti-GAL9-binding molecules are particularly useful for diseases or conditions in which the GAL9 / PD-L2 interaction significantly contributes to the pathogenesis. Anti-GAL9-binding molecules are useful for treating or reducing inflammation, reducing autoimmune responses, prolonging remission, inducing remission, re-establishing immune tolerance, improving organ function, reducing disease progression, reducing the risk of developing secondary diseases, or increasing the overall survival of a subject.

[0011] In a first aspect, the present disclosure provides a Galectin-9 (GAL9) antigen-binding molecule comprising a first antigen-binding site (ABS) specific for a first epitope of a first GAL9 antigen, the first antigen-binding site being selected from the group consisting of P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-26, P9-27, P9-28, P9-29, P9-30, P9-31, P9-32, P9-33, P9-34, P9-35, P9-36, P9-37, P9-38, P9-39, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-47, P9-48, P9-49, P9-50, P9-51, P9-52, P9-53, P9-54, P9-55, P9-56, P9-57, P9-58, P9-59, P9-60, P9-61, P9-62, P9-63, P9-64, P9-65, P9-66, P9-67, P9-68, P9-69, P9-70, P9-71, P9-72, P9-73, P9-74, P9-75, P9-76, P9-77, P9-78, P9-79, P9-80, P9-81, P9-82, P9-83, P9-84, P9-85, P9-86, P9-87, P9-88, P9- and P9-51, P9-52, P9-53, P9-54, and P9-55. The present invention provides a GAL9 antigen-binding molecule comprising all three VH CDRs from any one of the ABS clones selected from the group consisting of P9-1, P9-2, P9-3, P9-4, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0012] In a second aspect, the disclosure provides a GAL9 antigen-binding molecule comprising a first antigen-binding site specific for a first epitope of a first galectin-9 (GAL9) antigen, wherein the first antigen-binding site is selected from the group consisting of P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, Provided are GAL9 antigen-binding molecules comprising all three VL CDRs from any one of ABS clones selected from P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0013] In a third aspect, the present disclosure provides a Galectin-9 (GAL9) antigen-binding molecule comprising a first antigen-binding site specific for a first epitope of a first GAL9 antigen, wherein the first antigen-binding site is selected from the group consisting of P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, Provided is a GAL9 antigen-binding molecule comprising all three VH CDRs and all three VL CDRs from any one of ABS clones selected from P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0014] In a fourth aspect, the present disclosure provides a GAL9 antigen-binding molecule comprising a first antigen-binding site specific for a first epitope of a first galectin-9 (GAL9) antigen, the first antigen-binding site comprising a VL sequence and a VH sequence derived from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0015] In some embodiments, the GAL9 antigen binding molecule comprises a complete immunoglobulin heavy chain "IgG1" sequence, including a VH sequence, and a complete immunoglobulin light chain sequence, including a VL sequence, wherein the VH sequence and VL sequence are derived from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0016] In some embodiments, the GAL9 antigen binding molecule comprises a complete immunoglobulin heavy chain "IgG4" sequence, including a VH sequence, and a complete immunoglobulin light chain sequence, including a VL sequence, wherein the VH sequence and VL sequence are derived from any one of the ABS clones selected from P9-01, P9-02A, PO-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0017] In some embodiments, the GAL9 antigen binding molecule may comprise a GAL9 antigen that is a human GAL9 antigen.

[0018] In some embodiments, the GAL9 antigen binding molecule may further comprise a second antigen binding site.

[0019] In certain embodiments, the second antigen-binding site is specific for the GAL9 antigen. In other embodiments, the second antigen-binding site is identical to the first antigen-binding site.

[0020] In other embodiments, the second antigen-binding site is specific for a second epitope of the first GAL9 antigen.

[0021] In some embodiments, the second antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from another ABS clone selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0022] In some embodiments, the second antigen-binding site comprises a VL sequence and a VH sequence derived from another ABS clone.

[0023] In some embodiments, the second antigen-binding site comprises a complete immunoglobulin heavy chain sequence comprising a VH sequence derived from another ABS clone, and a complete immunoglobulin light chain sequence comprising a VL sequence derived from another ABS clone.

[0024] In some embodiments, the second antigen-binding site is specific for an antigen other than the first GAL9 antigen.

[0025] In some embodiments, the first antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0026] In some embodiments, the first antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from any one of the ABS clones selected from P9-11, P9-24, P9-34, and P9-37.

[0027] In some embodiments, the first antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from any one of the ABS clones selected from P9-11, P9-24, and P9-34.

[0028] In some embodiments, the first antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from ABS clone P9-11.

[0029] In some embodiments, the first antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from ABS clone P9-24.

[0030] In some embodiments, the first antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from ABS clone P9-34.

[0031] In some embodiments, the first antigen-binding site comprises all three VH CDRs, all three VL CDRs, or all three VH CDRs and all three VL CDRs from ABS clone P9-37.

[0032] In some embodiments, the GAL9 antigen binding molecule comprises an antibody format selected from the group consisting of a full-length antibody, a Fab fragment, a F(ab)'2 fragment, an Fv, an scFv, a tandem scFv, a diabody, an scDiabody, a DART, a single-chain VHH camelid antibody, a tandAb, a minibody, and a B-body. B-bodies are described in U.S. Patent Application Publication No. 2018 / 0118811, which is incorporated herein by reference in its entirety.

[0033] In some embodiments, the GAL9 antigen binding molecule reduces TNF-α secretion by activated immune cells upon contact, the reduction being at least about a 30%, 35%, 40%, 45%, 50%, 55%, or 60% reduction compared to activated immune cells treated with a control agent.

[0034] In some embodiments, the GAL9 antigen binding molecule reduces IFN-γ secretion by activated immune cells upon contact, the reduction being at least about a 20%, 25%, 30%, 35%, 40%, 45%, or 50% reduction compared to activated immune cells treated with a control agent.

[0035] In some embodiments, the GAL9 antigen binding molecule increases IL-10 secretion by activated immune cells upon contact, wherein the increase is at least about 5%, 10%, 15%, 20%, 25%, 30%, 35%, or 40% compared to activated immune cells treated with a control agent.

[0036] In some embodiments, the GAL9 antigen binding molecule does not modulate PD-1 surface expression on activated immune cells relative to activated immune cells treated with a control agent.

[0037] In some embodiments, the GAL9 antigen binding molecule does not modulate PD-L1 surface expression on activated immune cells relative to activated immune cells treated with a control agent.

[0038] In some embodiments, the GAL9 antigen binding molecule does not modulate CTLA-4 surface expression on activated immune cells compared to activated immune cells treated with a control agent.

[0039] In some embodiments, the GAL9 antigen binding molecule does not modulate TIM3 surface expression on activated immune cells compared to activated immune cells treated with a control agent.

[0040] In some embodiments, the GAL9 antigen binding molecule does not modulate LAG3 surface expression on activated immune cells compared to activated immune cells treated with a control agent.

[0041] In some embodiments, the GAL9 antigen binding molecule is a control agent-treated activated CD8 + Compared with T cells, activated CD8 + Decreases 4-1BB surface expression on T cells.

[0042] In some embodiments, the GAL9 antigen binding molecule is a control agent-treated activated CD8 + Compared with T cells, activated CD8 + Reduces CD40L surface expression on T cells.

[0043] In some embodiments, the GAL9 antigen binding molecule is a control agent-treated activated CD8 + Compared with T cells, activated CD8 + Reduces OX40 surface expression on T cells.

[0044] In some embodiments, the control agent is a negative control agent or a positive control agent.

[0045] In some embodiments, the control agent is a control antibody.

[0046] In some embodiments, the control antibody is selected from the group consisting of ECA42 clone anti-GAL9 antibody, RG9.1 clone anti-GAL9 antibody, RG9.35 clone anti-GAL9 antibody, anti-PD1 antibody, 108A2 clone anti-GAL9 antibody, and a non-GAL9 binding isotype control antibody.

[0047] In some embodiments, the activated immune cells, activated CD8 + T cells or activated DCs were activated by peptide stimulation, anti-CD3, or dendritic cells.

[0048] In a fifth aspect, the present disclosure provides a GAL9 antigen binding molecule that reduces TNF-α secretion by activated immune cells, wherein the reduction is at least about a 30%, 35%, 40%, 45%, 50%, 55%, or 60% reduction compared to activated immune cells treated with a control agent.

[0049] In a sixth aspect, the present disclosure provides a GAL9 antigen binding molecule that reduces IFN-γ secretion by activated immune cells, wherein the reduction is at least about a 20%, 25%, 30%, 35%, 40%, 45%, or 50% reduction compared to activated immune cells treated with a control agent.

[0050] In a seventh aspect, the present disclosure provides a GAL9 antigen binding molecule that increases IF-10 secretion by activated immune cells, wherein the increase is at least about 5%, 10%, 15%, 20%, 25%, 30%, 35%, or 40% compared to activated immune cells treated with a control agent.

[0051] In an eighth aspect, the present disclosure provides a GAL9 antigen binding molecule that does not modulate PD-1 surface expression on activated immune cells relative to activated immune cells treated with a control agent.

[0052] In a ninth aspect, the present disclosure provides a GAL9 antigen binding molecule that does not modulate PD-L1 surface expression on activated immune cells relative to activated immune cells treated with a control agent.

[0053] In a tenth aspect, the present disclosure provides a GAL9 antigen binding molecule that does not modulate CTLA-4 surface expression on activated immune cells relative to activated immune cells treated with a control agent.

[0054] In an eleventh aspect, the present disclosure provides a GAL9 antigen binding molecule that does not modulate TIM3 surface expression on activated immune cells relative to activated immune cells treated with a control agent.

[0055] In a twelfth aspect, the present disclosure provides a GAL9 antigen binding molecule that does not modulate LAG3 surface expression on activated immune cells relative to activated immune cells treated with a control agent.

[0056] In a thirteenth aspect, the present disclosure provides a method for detecting activated CD8 + Activated CD8 compared with T cells + GAL9 antigen-binding molecules that reduce 4-1BB surface expression on T cells are provided.

[0057] In a fourteenth aspect, the present disclosure provides a method for detecting activated CD8 + Activated CD8 compared with T cells + GAL9 antigen-binding molecules that reduce CD40L surface expression on T cells are provided.

[0058] In a fifteenth aspect, the present disclosure provides a method for detecting activated CD8 + Activated CD8 compared with T cells + GAL9 antigen-binding molecules that reduce OX40 surface expression on T cells are provided.

[0059] In a sixteenth aspect, the present disclosure provides a GAL9 antigen binding molecule that exhibits one or more of the following properties: A) reduces TNF-α secretion by activated immune cells, the reduction being at least about a 30%, 35%, 40%, 45%, 50%, 55%, or 60% reduction compared to activated immune cells treated with a control agent; B) reducing IFN-γ secretion by activated immune cells, the reduction being at least about a 20%, 25%, 30%, 35%, 40%, 45%, or 50% reduction compared to activated immune cells treated with a control agent; C) increasing IL-10 secretion by activated immune cells, wherein the increase is at least about a 5%, 10%, 15%, 20%, 25%, 30%, 35%, or 40% increase compared to activated immune cells treated with a control agent; D) Failure to modulate PD-1 surface expression on activated immune cells compared to activated immune cells treated with a control agent; E) Failure to modulate PD-L1 surface expression on activated immune cells compared to activated immune cells treated with a control agent; F) failure to modulate CTLA-4 surface expression on activated immune cells compared to activated immune cells treated with a control agent; G) Failure to modulate TIM3 surface expression on activated immune cells compared to activated immune cells treated with a control agent; H) Failure to modulate LAG3 surface expression on activated immune cells compared to activated immune cells treated with a control agent; I) Activated CD8 treated with control agents + Compared with T cells, activated CD8 + reducing 4-1BB surface expression on T cells; J) Activated CD8 treated with control agents + Compared with T cells, activated CD8 + reducing CD40L surface expression on T cells; K) Activated CD8 treated with control agents + Compared with T cells, activated CD8 + Reducing OX40 surface expression on T cells.

[0060] In some embodiments, the control agent is a negative control agent or a positive control agent.

[0061] In some embodiments, the control agent is a control antibody.

[0062] In some embodiments, the control antibody is selected from the group consisting of ECA42 clone anti-GAL9 antibody, RG9.1 clone anti-GAL9 antibody, RG9.35 clone anti-GAL9 antibody, anti-PD1 antibody, 108A2 clone anti-GAL9 antibody, and a non-GAL9 binding isotype control antibody.

[0063] In some embodiments, the activated immune cells are activated by peptide stimulation, anti-CD3, or dendritic cells.

[0064] In some embodiments, the GAL9 antigen-binding molecules of the fifth to fifteenth aspects provided herein comprise a first antigen-binding site specific for a first epitope of a first GAL9 antigen, the first antigen-binding site being selected from the group consisting of P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-26, P9-27, P9-28, P9-29, P9-30, P9-31, P9-32, P9-33, P9-34, P9-35, P9-36, P9-37, P9-38, P9-39, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-47, P9-48, P9-49, P9-50, P9-51, P9-52, P9-53, P9-54, P9-55, P9-56, P9-57, P9-58, P9-59, P9-60, P9-61, P9-62, P9-63, P9-64, P9-65, P9-66, P9-67, P9-68, P9-69, P9-70, P9-71, P9-72, P9-73, P9-74, P9-75, P9-76, P9-77, P9-78, P9-79, P9-80, P9-81, P9-82, P9-83, P9-84, P9-85, P9-86, P9-87, P9-88, P9-89, P9 4, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0065] In some embodiments, the VL and VH sequences are derived from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0066] In some certain embodiments, the GAL9 antigen binding molecule comprises a complete immunoglobulin heavy chain sequence comprising a VH sequence, and a complete immunoglobulin light chain sequence comprising a VL sequence, wherein the VH sequence and the VL sequence are derived from any one of the ABS clones selected from P9-01, P9-02A, PO-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0067] In some embodiments, the GAL9 antigen is a human GAL9 antigen.

[0068] In some embodiments, the GAL9 antigen binding molecule further comprises a second antigen binding site.

[0069] In some embodiments, the second antigen-binding site is specific for the GAL9 antigen.

[0070] In some embodiments, the second antigen-binding site is identical to the first antigen-binding site.

[0071] In some embodiments, the second antigen-binding site is specific for a second epitope of the first GAL9 antigen.

[0072] In some embodiments, the second antigen-binding site comprises all three VH CDRs and all three VL CDRs from another ABS clone selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0073] In some embodiments, the second antigen-binding site comprises a VL sequence and a VH sequence derived from another ABS clone.

[0074] In some embodiments, the second antigen-binding site comprises a complete immunoglobulin heavy chain sequence comprising a VH sequence derived from another ABS clone, and a complete immunoglobulin light chain sequence comprising a VL sequence derived from another ABS clone.

[0075] In some embodiments, the second antigen-binding site is specific for an antigen other than the first GAL9 antigen.

[0076] In some embodiments, the first antigen-binding site comprises all three VH CDRs and all three VL CDRs from any one of the ABS clones selected from P9-11, P9-24, P9-34, and P9-37.

[0077] In some embodiments, the first antigen-binding site comprises all three VH CDRs and all three VL CDRs from any one of the ABS clones selected from P9-11, P9-24, and P9-34.

[0078] In some embodiments, the first antigen-binding site comprises all three VH CDRs and all three VL CDRs from ABS clone P9-11.

[0079] In some embodiments, the first antigen-binding site comprises all three VH CDRs and all three VL CDRs from ABS clone P9-24.

[0080] In some embodiments, the first antigen-binding site comprises all three VH CDRs and all three VL CDRs from ABS clone P9-34.

[0081] In some embodiments, the first antigen-binding site comprises all three VH CDRs and all three VL CDRs from ABS clone P9-37.

[0082] In some embodiments, the GAL9 antigen binding molecule comprises an antibody format selected from the group consisting of a full-length antibody, a Fab fragment, an Fv, an scFv, a tandem scFv, a diabody, an scdiabody, a DART, a tandAb, a minibody, and a B-body.

[0083] In a seventeenth aspect, the present disclosure provides a GAL9 antigen binding molecule that binds to the same epitope as a GAL9 antigen binding molecule according to any one of the preceding claims.

[0084] In an eighteenth aspect, the present disclosure provides a GAL9 antigen binding molecule that competes for binding with a GAL9 antigen binding molecule of any one of the preceding claims.

[0085] In some embodiments, the GAL9 antigen binding molecule is purified.

[0086] In a nineteenth aspect, the present disclosure provides a pharmaceutical composition comprising a GAL9 antigen binding molecule according to any one of the preceding claims and a pharmaceutically acceptable diluent.

[0087] In a twentieth aspect, the present disclosure provides a method for treating a subject having an autoimmune disease, the method comprising administering to the subject a therapeutically effective amount of a pharmaceutical composition as provided herein.

[0088] In some embodiments, subjects with an autoimmune disease exhibit increased expression of PD-L2 on dendritic cells compared to dendritic cells from healthy controls.

[0089] In some embodiments, the autoimmune disease is selected from the group consisting of inflammatory bowel disease, Crohn's disease, ulcerative colitis, colitis, celiac disease, rheumatoid arthritis, Behcet's disease, amyloidosis, psoriasis, psoriatic arthritis, systemic lupus erythematosus nephritis, graft-versus-host disease (GVHD), non-alcoholic steatohepatitis (NASH), and ankylosing spondylitis.

[0090] In some embodiments, administration of a therapeutically effective amount of the GAL-binding molecule itself or a pharmaceutical composition results in reduced inflammation, reduced autoimmune response, prolonged remission, induced remission, re-established immune tolerance, improved organ function, reduced disease progression, reduced risk of secondary disease progression or onset, or increased overall survival. [Brief explanation of the drawings]

[0091] 5. Brief description of the drawings [Figure 1A] FIG. 1 shows illustrative examples of various CDR and framework numbering systems—Chothia, Martin (ABA), and Kabat—applied to the P9-01 anti-human Gal9 candidate antibody provided herein. [Figure 1B] FIG. 1 shows illustrative examples of various CDR and framework numbering systems—Chothia, Martin (ABA), and Kabat—applied to the P9-01 anti-human Gal9 candidate antibody provided herein. [Figure 2] Density contour plot of the percentage of CD11c+ blood dendritic cells from Crohn's disease patients detected as positive for PD-L1 or PD-L2 expression compared to labeled isotype IgG control. [Figure 3A] FIG. 1 shows scatter plots of the percentage of blood dendritic cells expressing PD-L1 or PD-L2 in healthy controls or Crohn's disease patients. [Figure 3B] FIG. 1 shows scatter plots of the percentage of blood dendritic cells expressing PD-L1 or PD-L2 in healthy controls or Crohn's disease patients. [Figure 3C] Scatter plot of the geometric mean fluorescence (GMI) of PD-L1 or PD-L2 surface expression on blood dendritic cells in healthy controls or Crohn's disease patients. [Figure 3D] Scatter plot of the geometric mean fluorescence (GMI) of PD-L1 or PD-L2 surface expression on blood dendritic cells in healthy controls or Crohn's disease patients. [Figure 4]Figures 4A and 4B show representative confocal images of DNA (DAPI; blue), PD-L1 (green), and PD-L2 (red) expression on dendritic cells from two healthy control donors (4A) and three Crohn's disease patients (4B), rendered in grayscale in the accompanying figures. [Figure 5] Figures 5A-5C show the mean concentrations of cytokines secreted by PMBCs from Crohn's disease (CD) patients after treatment with either anti-CD3 and anti-PD-L2 (αPD-L2) to mimic TCR activation or an IgG control. Figures 5A-5B show the mean concentrations of TNF-α and IFN-γ in PMBCs from CD patients after treatment with anti-PD-L2 or an IgG control. Figure 5C shows the mean ratio of IL-10:TNF-α secretion in PMBCs from CD patients after treatment with anti-PD-L2 or an IgG control. [Figure 6] Figure 10 shows TNF-α secretion by anti-CD3-activated mouse CD4+ T cells after treatment with either sPD-L2 or both sPD-L2 and inhibitory anti-mouse anti-GAL9 (108A2). [Figure 7] Representative confocal images of DNA (DAPI; blue), PD-L1 (green), PD-1 (red), and OX40 (yellow) expression in CD4+ T cells from malaria-infected mice after treatment with mouse inhibitory anti-mouse GAL9 (108A2) and activating anti-mouse GAL9 (RG9.1) antibodies are shown, rendered in grayscale in the accompanying figures. [Figure 8] Figures 8A and 8B show bar graphs of the percentage of surviving mouse CD4+ and CD8+ T cells following treatment with either sPD-L2 or sPD-L2 and mouse inhibitory anti-GAL9 (108A2) antibody. [Figure 9] Figures 9A and 9B show bar graphs of INF-γ (9A) and TNF-α (9B) secretion from murine CD4+ T cells co-cultured with dendritic cells (stimulated) and treated with blocking anti-PD-L2 (clone Ty25) or inhibitory anti-GAL9 (108A2) murine antibodies, compared to control unstimulated CD4+ T cells. [Figure 10A] Figure 10A shows IFN-γ (10A) and TNF-α (10B) secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with various anti-human GAL9 candidates, known activating tool antibodies (tool mAbs (Tool mAbs)), anti-PD-1 antibodies, IgG control antibodies (IgG Ctrl), and vehicle control (PBS Ctrl). Black diamonds indicate secretion from activated PBMCs stimulated with tool mAbs and anti-PD-1 antibodies. [Figure 10B] Figure 10A shows IFN-γ (10A) and TNF-α (10B) secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with various anti-human GAL9 candidates, known activation tool antibodies (tool mAbs), anti-PD-1 antibodies, IgG control antibodies (IgG Ctrl), and vehicle control (PBS Ctrl). Black diamonds indicate secretion from activated PBMCs stimulated with tool mAbs and anti-PD-1 antibodies. [Figure 11A] FIG. 1 shows IFN-γ and TNF-α secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with anti-human GAL9 P9-1, P9-37, or P9-57 compared to an IgG control antibody (IgG). [Figure 11B] FIG. 1 shows IFN-γ and TNF-α secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with anti-human GAL9 P9-1, P9-37, or P9-57 compared to an IgG control antibody (IgG). [Figure 11C] FIG. 1 shows IFN-γ and TNF-α secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with anti-human GAL9 P9-1, P9-37, or P9-57 compared to an IgG control antibody (IgG). [Figure 12A] Figure 12A shows TNF-α (12A), IFN-γ (12B), and IL-10 (12C) secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with anti-human GAL9 candidates P9-11, P9-24, or P9-34 compared to an IgG control antibody (IgG). [Figure 12B] Figure 12A shows TNF-α (12A), IFN-γ (12B), and IL-10 (12C) secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with anti-human GAL9 candidates P9-11, P9-24, or P9-34 compared to an IgG control antibody (IgG). [Figure 12C] Figure 12A shows TNF-α (12A), IFN-γ (12B), and IL-10 (12C) secretion from PBMCs stimulated in vitro with HCMV peptides after treatment with anti-human GAL9 candidates P9-11, P9-24, or P9-34 compared to an IgG control antibody (IgG). [Figure 13] Figures 13A and 13B show bar graphs of the ratio of TNF-α:IL-10 secretion (13A) and the ratio of INF-γ:IL-10 secretion (13B) from anti-CD3-activated mouse CD3+ T cells after treatment with inhibitory anti-mouse GAL9 (108A2) and anti-human GAL9 P9-11, P9-24, or P9-34. DETAILED DESCRIPTION OF THE INVENTION

[0092] 6. Detailed Description 6.1. Definition Unless otherwise defined, all technical and scientific terms used herein have the meaning commonly understood by one of ordinary skill in the art to which this invention belongs. As used herein, the following terms have the meanings ascribed to them below.

[0093] "Antigen-binding site" or "ABS" refers to the region of a GAL9 binding molecule that specifically recognizes or binds a given antigen or epitope.

[0094] As used herein, the terms "treat" or "treatment" are used in their most widely accepted clinical sense. This term includes, but is not limited to, alleviating signs or symptoms of a disease; ameliorating signs or symptoms of a disease; alleviating symptoms; reducing the extent of a disease; a stable (i.e., not worsening) state of a disease; delaying or slowing the progression of a disease; reversing or alleviating a disease state; causing remission (whether partial or total), whether detectable or undetectable; curing; and prolonging survival compared to expected survival if not receiving treatment. Unless otherwise specified, "treat" or "treatment" is not intended to prevent or prevent disease.

[0095] "Subject" or "individual" or "animal" or "patient" or "mammal" means any subject, particularly a mammalian subject, for whom diagnosis, prognosis, or therapy is desired. Mammalian subjects include humans, farm animals, livestock, and zoo, sport, or pet animals such as dogs, cats, guinea pigs, rabbits, rats, mice, horses, cattle, and cows. Unless otherwise specified, by "patient" is intended a human "subject."

[0096] The term "sufficient amount" means an amount sufficient to produce a desired effect, for example, an amount sufficient to modulate protein aggregation in a cell.

[0097] The term "therapeutically effective amount" refers to an amount effective for ameliorating symptoms of the disease.

[0098] The term "prophylactically effective amount" refers to an amount effective in preventing symptoms of a disease.

[0099] 6.2. Other Interpretation Conventions Unless otherwise specified, all references to sequences herein are to amino acid sequences.

[0100] Unless otherwise specified, antibody constant region residue numbering is according to the Eu index as set forth at www.imgt.org / IMGTScientificChart / Numbering / Hu_IGHGnber.html#refs (accessed August 22, 2017), which is incorporated by reference in its entirety, and the residue number identifies the residue according to its position in the endogenous constant region sequence, regardless of the residue's physical location within the chain of the GAL9 binding molecules described herein.

[0101] Unless otherwise specified as a "Kabat CDR," a "Chothia CDR," a "Contact CDR," or an "IMGT CDR," all references to a "CDR" are CDRs defined using the Martin (ABA) definition.

[0102] "Endogenous sequence" or "native sequence" means any sequence, including both nucleic acid and amino acid sequences, that originates from an organism, tissue, or cell and has not been artificially modified or mutated.

[0103] Polypeptide chain numbers (e.g., "first" polypeptide chain, "second" polypeptide chain, etc., or polypeptide "chain 1," "chain 2," etc.) are used herein as unique identifiers of the particular polypeptide chains that form a binding molecule and are not intended to imply any order or quantity of the various polypeptide chains within the binding molecule.

[0104] In this disclosure, the words "comprises," "comprising," "containing," "having," "includes," "including," and linguistic variations thereof have the meanings ascribed to them in United States patent law and permit the presence of additional components other than those expressly recited.

[0105] As used herein, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Terms such as "include" and "such as" are intended to convey inclusion rather than limitation, unless specifically indicated otherwise.

[0106] Ranges provided herein are understood to be a shorthand for all values ​​within the range, inclusive of the recited endpoints. For example, a range of 1 to 50 is understood to include any number, combination of numbers, or subrange from the group consisting of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, and 50.

[0107] Unless otherwise specifically stated or otherwise apparent from the context, the term "about," as used herein, is understood to mean within normal tolerances in the art, e.g., within two standard deviations of the mean. It can be understood to be within about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05%, or 0.01% of the stated value.

[0108] 6.3. Basic overview The present disclosure provides GAL9 antigen-binding molecules, such as anti-galectin-9 (GAL9) antibodies and antigen-binding fragments thereof; compositions comprising GAL9-binding molecules; pharmaceutical compositions comprising GAL9-binding molecules; and methods for treating subjects with diseases or conditions using GAL9-binding molecules. In particular, the present disclosure provides various GAL9 antigen-binding molecules that are inhibitory, acting as inhibitors of the immune system, reducing the secretion and production of pro-inflammatory cytokines, increasing the secretion and production of anti-inflammatory cytokines in various immune cells, and reducing the surface expression of stimulatory molecules.

[0109] GAL9 antigen binding molecules are particularly useful for treating autoimmune or inflammatory diseases in a subject. In some embodiments, the compositions and methods are used to treat an infectious disease that triggers an inflammatory response in a subject. Anti-GAL9 binding molecules are particularly useful for treating diseases or conditions in which GAL9 / PD-L2 interaction significantly contributes to the pathogenesis. In some embodiments, the anti-GAL9 binding molecules are administered to a subject by themselves, as a pharmaceutical composition, or in combination with other therapeutic agents or procedures.

[0110] 6.4. GAL9 antigen binding molecule In a first aspect, an antigen-binding molecule is provided. In all embodiments, the antigen-binding molecule comprises at least a first antigen-binding site specific for the GAL9 antigen, and thus the binding molecule is referred to as a GAL9 antigen-binding molecule or GAL9-binding molecule.

[0111] The GAL9 antigen binding molecules described herein specifically bind to the GAL9 antigen.

[0112] As used herein, "GAL9 antigen" refers to galectin-9 family members and homologs. GAL9 is also known as LGALS9, HUAT, LGALS9A, tumor antigen HOM-HD-21, and ecalectin. In certain embodiments, a GAL9-binding molecule has an antigen-binding site that specifically binds to at least a portion of more than one GAL9 domain, such as the junction between a first GAL9 domain and a second GAL9 domain.

[0113] In certain embodiments, the GAL9 antigen is human. GenBank Accession No. NP_033665.1 describes the authentic human GAL9 protein, including its sequence and domain characteristics, and is incorporated herein by reference in its entirety. SEQ ID NO: 6 provides the full-length GAL9 protein sequence. [ka]

[0114] In various embodiments, the GAL9 binding molecule further specifically binds to at least one antigen in addition to the GAL9 antigen.

[0115] 6.4.1. Functional Properties of GAL9 Antigen-Binding Molecules In typical embodiments, upon contact therewith, the GAL9 antigen binding molecule modulates cytokine secretion (e.g., increases or decreases cytokine secretion) of an immune cell or activated immune cell. In some embodiments, the immune cell is a peripheral blood mononuclear cell (PBMC). In some embodiments, the immune cell is a T cell. In some embodiments, the T cell is an effector T cell. In some embodiments, the T cell is a CD8 + In embodiments, the T cells are CD4 + In some embodiments, the T cells are CD3 + T cells.

[0116] The effect of a GAL9 antigen binding molecule on immune cell cytokine secretion can be determined by any suitable means. For example, the effect of a GAL9 antigen binding molecule on immune cell cytokine secretion can be determined in vivo, ex vivo, or in vitro. In some embodiments, cytokine secretion is determined in activated immune cells contacted with a GAL9 antigen binding molecule compared to activated immune cells contacted with a control agent, such as a control antigen binding molecule or a vehicle control. Immune cells can be activated by peptide stimulation. For example, immune cells can be activated with a peptide or multiple peptides known to induce an immune response. The multiple peptides known to induce an immune response can be derived from a pathogen infection, such as a viral or bacterial infection.

[0117] The control agent may be a negative control or a positive control. In some embodiments, the GAL9 antigen-binding molecule increases cytokine secretion from immune cells compared to the negative control agent or negative control antigen-binding molecule. In some embodiments, the negative control antigen-binding molecule is an isotype control binding molecule that does not bind to GAL9. In some embodiments, the positive control antibody is an anti-PD1 antibody such as nivolumab. In some embodiments, the positive control antibody is a GAL9 control antibody. The GAL9 control antibody may be GAL9 antibody clone RG9.1 (Catalog No. BE0218, InVivoMab Antibodies) or RG9.35. Both RG9.1 and RG9.35 are described in Fukushima A, Sumi T, Fukuda K, Kumagai N, Nishida T et al. (2008) Roles of galectin-9 in the development of experimental allergic conjunctivitis in mice. IntArch Allergy Immunol 146:36-43. This document is incorporated herein by reference in its entirety. The GAL9 control antibody may be GAL9 antibody clone ECA42 (catalog number LS-C 179449, LifeSPan BioScience). The GAL9 control antibody may be GAL9 antibody clone 108A2 (BioLegend®, San Diego, CA). In some embodiments, the GAL9 antigen binding molecule reduces cytokine secretion of pro-inflammatory cytokines by immune cells compared to a control antibody. In some embodiments, the GAL9 antigen binding molecule increases cytokine secretion of inhibitory cytokines by immune cells compared to a control antibody.

[0118] Cytokine secretion by immune cells can be assessed by any suitable means. By way of example only, cytokine secretion by in vitro or ex vivo immune cell culture models can be assessed by analyzing the cytokine content of cultured cell supernatants, for example, by cytokine bead arrays.

[0119] In some embodiments, the cytokine is TNF-α. In some embodiments, the GAL9 antigen binding molecule reduces TNF-α secretion by activated immune cells by at least 1%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, or 90% compared to a control agent described herein. In some embodiments, the GAL9 antigen binding molecule reduces TNF-α secretion by activated immune cells by at least 1% to 5%, 5% to 10%, 10% to 15%, 15% to 20%, 20% to 25%, 25% to 30%, 30% to 35%, 35% to 40%, 40% to 45%, 45% to 50%, 50% to 55%, 55% to 60%, 60% to 65%, 70% to 75%, 75% to 80%, 80% to 85%, or 85% to 90% compared to a control agent described herein. In some embodiments, the GAL9 antigen binding molecule reduces TNF-α secretion by activated immune cells by about 30% to 50% compared to a control agent described herein.

[0120] In some embodiments, the cytokine is IFN-γ. In some embodiments, the GAL9 antigen binding molecule reduces IFN-γ secretion from activated immune cells by at least 1%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, or 75% compared to a control agent described herein. In some embodiments, the GAL9 antigen binding molecule reduces IFN-γ secretion from activated immune cells by at least 10-15%, 15-20%, 20-25%, 25-30%, 30-35%, 35%-40%, 40%-45%, 45%-50%, 50%-55%, 55%-60%, 60%-65%, or 70%-75% compared to a control agent described herein. In some embodiments, the GAL9 antigen binding molecule reduces IFN-γ secretion by activated immune cells by about 20% to 40% compared to a control agent described herein.

[0121] In some embodiments, the cytokine is IL-10. In some embodiments, the GAL9 antigen binding molecule increases IL-10 secretion by activated immune cells by at least 1%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, or 50% compared to a control agent described herein. In some embodiments, the GAL9 antigen binding molecule increases IL-10 secretion by activated immune cells by at least 1% to 5%, 5 to 10%, 10 to 15%, 15 to 20%, 20 to 25%, 25 to 30%, 30 to 35%, 35% to 40%, 40% to 45%, or 45% to 50% compared to a control agent described herein. In some embodiments, the GAL9 antigen binding molecule increases IL-10 secretion by activated immune cells by about 5% to 30% compared to a control agent described herein.

[0122] In some embodiments, upon contact therewith, the GAL9 antigen binding molecules do not modulate surface expression of immune checkpoint molecules (e.g., stimulatory or inhibitory checkpoint molecules) relative to activated immune cells treated with a control agent. The term "does not modulate" means that there is no substantial increase or decrease in expression of immune checkpoint molecules following treatment with a GAL9 binding molecule provided herein relative to the control agent. In some embodiments, no substantial increase in surface expression (e.g., no modulation of expression) is an increase in cell surface expression of no more than a 1.01x, 1.02x, 1.03x, 1.04x, 1.05x, 1.06x, 1.07x, 1.08x, 1.09x, 1.1x, 1.2x, or 1.3x fold change relative to activated immune cells treated with a control agent. In some embodiments, no substantial decrease in surface expression (e.g., no modulation of expression) is a decrease in cell surface expression of no more than 0.01×, 0.02×, 0.03×, 0.04×, 0.05×, 0.06×, 0.07×, 0.08×, 0.09×, 0.1×, or 0.2× fold change compared to activated immune cells treated with a control agent.

[0123] In some embodiments, no substantial increase in surface expression (e.g., no modulation of expression) refers to an increase in surface expression of about 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, or 15% compared to activated immune cells treated with a control agent. In some embodiments, no substantial decrease in surface expression (e.g., no modulation of expression) refers to an decrease in surface expression of about 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, or 15% compared to activated immune cells treated with a control agent.

[0124] In some embodiments, the absence of a substantial increase or decrease in surface expression is determined by comparing the level of surface expression to the noise level of the assay (e.g., in vivo, ex vivo, or in vitro). In some embodiments, the absence of a substantial increase or decrease in surface expression is determined by comparing the level of surface expression to the standard deviation of the assay (e.g., in vivo, ex vivo, or in vitro).

[0125] The effect of a GAL9 antigen binding molecule on the surface expression of one or more immune checkpoint molecules can be determined by any suitable means, for example, the effect of a GAL9 antigen binding molecule on the surface expression of one or more costimulatory molecules can be determined in vivo, ex vivo, or in vitro.

[0126] In some embodiments, the one or more immune checkpoint molecules are selected from PD-1, PD-L1, CTLA-4, TIM3, LAG3, TIGIT, and PVRIG. In some embodiments, the one or more checkpoint molecules are selected from PD-1, PD-L1, TIM3, and LAG3. In some embodiments, the immune checkpoint molecule is PD-1 or PD-L1. In various embodiments, the activated (e.g., stimulated) immune cells are selected from T cells, CD8 + T cells, CD4 + T cells, CD3 + T cells or PBMCs.

[0127] In some embodiments, the immune checkpoint molecule is PD-1. In some embodiments, activated CD8 treated with a GAL9 antigen-binding molecule. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in PD-1 surface expression of no more than 1.01×, 1.02×, 1.03×, 1.04×, 1.05×, 1.06×, 1.07×, 1.08×, 1.09×, 1.1×, 1.2×, or 1.3× fold change compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + exhibit a decrease in surface expression of less than 0.01×, 0.02×, 0.03×, 0.04×, 0.05×, 0.06×, 0.07×, 0.08×, 0.09×, 0.1×, or 0.2× fold change in PD-1 surface expression compared to T cells.

[0128] In some embodiments, activated CD8 treated with a GAL9 antigen binding molecule + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 +In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in PD-1 surface expression of about 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, or 15% or less compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + exhibits a reduction in PD-1 surface expression of no more than about 1% reduction, 2% reduction, 3% reduction, 4% reduction, 5% reduction, 6% reduction, 7% reduction, 8% reduction, 9% reduction, 10% reduction, 11% reduction, 12% reduction, 13% reduction, 14% reduction, or 15% reduction compared to T cells.

[0129] In some embodiments, the immune checkpoint molecule is PD-L1. In some embodiments, activated CD8 treated with a GAL9 antigen-binding molecule. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in PD-L1 surface expression that is no greater than a fold change compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in PD-L1 surface expression of no more than a 1.01×, 1.02×, 1.03×, 1.04×, 1.05×, 1.06×, 1.07×, 1.08×, 1.09×, 1.1×, 1.2×, or 1.3× fold change compared to GAL9 T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + exhibit a decrease in surface expression of PD-L1 relative to T cells of less than 0.01×, 0.02×, 0.03×, 0.04×, 0.05×, 0.06×, 0.07×, 0.08×, 0.09×, 0.1×, or 0.2× fold change in surface expression.

[0130] In some embodiments, activated CD8 treated with a GAL9 antigen binding molecule + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in PD-L1 surface expression of about 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, or 15% or less compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + exhibits a reduction in PD-L1 surface expression of no more than about 1% reduction, 2% reduction, 3% reduction, 4% reduction, 5% reduction, 6% reduction, 7% reduction, 8% reduction, 9% reduction, 10% reduction, 11% reduction, 12% reduction, 13% reduction, 14% reduction, or 15% reduction compared to T cells.

[0131] In some embodiments, the immune checkpoint molecule is CTLA-4. In some embodiments, activated CD8 treated with a GAL9 antigen-binding molecule. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in CTLA-4 surface expression of no more than a 1.01×, 1.02×, 1.03×, 1.04×, 1.05×, 1.06×, 1.07×, 1.08×, 1.09×, 1.1×, 1.2×, or 1.3× fold change compared to GAL9 T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 +Compared to T cells, they exhibit a decrease in surface expression of less than 0.01×, 0.02×, 0.03×, 0.04×, 0.05×, 0.06×, 0.07×, 0.08×, 0.09×, 0.1×, or 0.2× fold change in CTLA-4 surface expression.

[0132] In some embodiments, activated CD8 treated with a GAL9 antigen binding molecule + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in CTLA-4 surface expression of about 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, or 15% or less compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + exhibits a reduction in CTLA-4 surface expression of no more than about 1% reduction, 2% reduction, 3% reduction, 4% reduction, 5% reduction, 6% reduction, 7% reduction, 8% reduction, 9% reduction, 10% reduction, 11% reduction, 12% reduction, 13% reduction, 14% reduction, or 15% reduction compared to T cells.

[0133] In some embodiments, the immune checkpoint molecule is TIM3. In some embodiments, activated CD8 treated with a GAL9 antigen-binding molecule. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in TIM3 surface expression of no more than 1.01×, 1.02×, 1.03×, 1.04×, 1.05×, 1.06×, 1.07×, 1.08×, 1.09×, 1.1×, 1.2×, or 1.3× fold change compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8+ Compared to T cells, they exhibit a decrease in surface expression of less than 0.01×, 0.02×, 0.03×, 0.04×, 0.05×, 0.06×, 0.07×, 0.08×, 0.09×, 0.1×, or 0.2× fold change in TIM3 surface expression.

[0134] In some embodiments, activated CD8 treated with a GAL9 antigen binding molecule + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in TIM3 surface expression of about 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, or 15% or less compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + exhibits a reduction in TIM3 surface expression of no more than about 1% reduction, 2% reduction, 3% reduction, 4% reduction, 5% reduction, 6% reduction, 7% reduction, 8% reduction, 9% reduction, 10% reduction, 11% reduction, 12% reduction, 13% reduction, 14% reduction, or 15% reduction compared to T cells.

[0135] In some embodiments, the immune checkpoint molecule is LAG3. In some embodiments, activated CD8 treated with a GAL9 antigen-binding molecule. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in LAG3 surface expression of no more than a 1.01×, 1.02×, 1.03×, 1.04×, 1.05×, 1.06×, 1.07×, 1.08×, 1.09×, 1.1×, 1.2×, or 1.3× fold change compared to GAL9 antigen binding molecules. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8+ In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit a decrease in surface expression of LAG3 of less than or equal to 0.01×, 0.02×, 0.03×, 0.04×, 0.05×, 0.06×, 0.07×, 0.08×, 0.09×, 0.1×, or 0.2× fold change in surface expression compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit an increase in LAG3 surface expression of about 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, or 15% or less compared to T cells. + or CD4 + T cells were treated with control agents and activated CD4 + or CD8 + exhibits a reduction in LAG3 surface expression of no more than about 1% reduction, 2% reduction, 3% reduction, 4% reduction, 5% reduction, 6% reduction, 7% reduction, 8% reduction, 9% reduction, 10% reduction, 11% reduction, 12% reduction, 13% reduction, 14% reduction, or 15% reduction compared to T cells.

[0136] In some embodiments, the GAL9 antigen binding molecule reduces the surface expression of one or more costimulatory molecules on immune cells, e.g., human immune cells. In certain embodiments, the GAL9 antigen binding molecule reduces the surface expression of one or more costimulatory molecules on activated immune cells. In certain embodiments, the activated immune cells are T cells. In certain embodiments, the activated immune cells are CD8 + In some embodiments, the one or more costimulatory molecules are selected from 4-1BB, CD40L, and OX40. In some embodiments, the one or more costimulatory molecules are selected from 4-1BB and CD40L. In some embodiments, the costimulatory molecule is OX40.

[0137] The effect of a GAL9 antigen binding molecule on the surface expression of one or more costimulatory molecules can be determined by any suitable means, for example, the effect of a GAL9 antigen binding molecule on the surface expression of one or more costimulatory molecules can be determined in vivo, ex vivo, or in vitro.

[0138] In some embodiments, the GAL9 antigen binding molecule reduces the surface expression of one or more costimulatory molecules on activated immune cells compared to activated immune cells treated with a control agent. Exemplary control agents are described herein. In certain embodiments, the control agent is an isotype control binding molecule that does not bind to GAL9.

[0139] In some embodiments, the GAL9 antigen binding molecule is a control agent-treated activated CD8 + Compared with T cells, activated CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen-binding molecule are treated with a GAL9 antigen-binding molecule to reduce 4-1BB surface expression on T cells. + T cells were treated with control agents and activated CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit at least about a 0.1× reduction, 0.2× reduction, 0.3× reduction, 0.4× reduction, 0.5× reduction, or 0.6× reduction in 4-1BB surface expression compared to T cells. + T cells were treated with control agents and activated CD8 + They exhibit approximately a 0.1× to 0.2× reduction, a 0.2× to 0.3× reduction, a 0.3× to 0.4× reduction, a 0.4× to 0.5× reduction, or a 0.5× to 0.6× reduction in 4-1BB surface expression compared to T cells.

[0140] In some embodiments, the GAL9 antigen binding molecule is a control agent-treated activated CD8 + Compared with T cells, activated CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen-binding molecule are treated with a GAL9 antigen-binding molecule to reduce CD40L surface expression on the T cells. + T cells were treated with control agents and activated CD8 +In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit at least about a 0.1× reduction, 0.2× reduction, 0.3× reduction, 0.4× reduction, or 0.5× reduction in CD40L surface expression compared to T cells. + T cells were treated with control agents and activated CD8 + They exhibit at least about a 0.1× to 0.2× decrease, a 0.2× to 0.3× decrease, a 0.3× to 0.4× decrease, or a 0.4× to 0.5× decrease in CD40L surface expression compared to T cells.

[0141] In some embodiments, the GAL9 antigen binding molecule is a control agent-treated activated CD8 + Compared with T cells, activated CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule are treated with a GAL9 antigen binding molecule to reduce OX40 surface expression on the T cells. + T cells were treated with control agents and activated CD8 + In some embodiments, activated CD8 T cells treated with a GAL9 antigen binding molecule exhibit at least about a 0.1× reduction, 0.2× reduction, 0.3× reduction, 0.4× reduction, 0.5× reduction, or 0.6× reduction in OX40 surface expression compared to T cells. + T cells were treated with control agents and activated CD8 + They exhibit approximately a 0.1× to 0.2× reduction, a 0.2× to 0.3× reduction, a 0.3× to 0.4× reduction, a 0.4× to 0.5× reduction, or a 0.5× to 0.6× reduction in OX40 surface expression compared to T cells.

[0142] The present disclosure also provides GAL9 antigen binding molecules with various clinical benefits for improving the health of subjects with autoimmune or inflammatory diseases. The subject may be a mammal. The mammal may be a mouse. In some embodiments, the mammal is a human.

[0143] In some embodiments, the GAL9 antigen binding molecule reduces an autoimmune response in a subject. In some embodiments, the GAL9 antigen binding molecule reduces inflammation in a subject. The inflammation may be systemic or localized to an organ or tissue. In some embodiments, the GAL9 antigen binding molecule prolongs remission of a disease or condition in a subject. In some embodiments, the GAL9 antigen binding molecule induces remission in a subject. In some embodiments, the GAL9 antigen binding molecule re-establishes immune tolerance in a subject (e.g., improving the cytokine profile or environment). Re-establishment of immune tolerance may be a decrease in pro-inflammatory cytokines, an increase in inhibitory cytokines, or a combination thereof. In some embodiments, the GAL9 antigen binding molecule improves organ function in a subject. In some embodiments, the GAL9 antigen binding molecule reduces the risk / likelihood of disease progression or the development of a secondary disease, such as cancer or an infectious disease. In some embodiments, the GAL9 antigen binding molecule increases overall survival in a subject.

[0144] 6.4.2. Variable Area In exemplary embodiments, a GAL9 binding molecule comprises antibody variable region domain amino acid sequences, including VH and VL antibody domain sequences, which are described in more detail below in Sections 6.4.2.1 and 6.4.2.2, respectively.

[0145] 6.4.2.1. VH area In typical embodiments, the GAL9-binding molecules described herein comprise an antibody heavy chain variable domain sequence. In typical antibody configurations, both naturally occurring and in the GAL9-binding molecules described herein, a particular VH amino acid sequence associates with a particular VL amino acid sequence to form an antigen-binding site. In various embodiments, the VH amino acid sequence is a human sequence, a synthetic sequence, or a mammalian sequence, including a combination of non-human mammalian sequences, mammalian sequences, and / or synthetic sequences, as described in more detail above in Sections 6.4.2.3 and 6.4.2.4. In various embodiments, the VH amino acid sequence is a mutated sequence of a naturally occurring sequence.

[0146] 6.4.2.2. VL area Useful VL amino acid sequences for the GAL9-binding molecules described herein are antibody light chain variable domain sequences. In a typical configuration, both in natural antibodies and in the antibody constructs described herein, a particular VL amino acid sequence associates with a particular VH amino acid sequence to form an antigen-binding site. In various embodiments, the VL amino acid sequence is a human sequence, a synthetic sequence, or a mammalian sequence, including a combination of non-human mammalian sequences, mammalian sequences, and / or synthetic sequences, as described in more detail in Sections 6.4.2.3 and 6.4.2.4, below.

[0147] In various embodiments, the VL amino acid sequence is a mutant sequence of a naturally occurring sequence. In certain embodiments, the VL amino acid sequence is a lambda (λ) light chain variable domain sequence. In certain embodiments, the VL amino acid sequence is a kappa (κ) light chain variable domain sequence. In a preferred embodiment, the VL amino acid sequence is a kappa (κ) light chain variable domain sequence.

[0148] 6.4.2.3. Complementarity-Determining Regions The VH and VL amino acid sequences contain highly variable sequences called "complementarity-determining regions" (CDRs), typically three CDRs (CDR1, CDR2, and CDR3). In various embodiments, the CDRs are mammalian sequences, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the CDRs are human sequences. In various embodiments, the CDRs are naturally occurring sequences. In various embodiments, the CDRs are naturally occurring sequences that have been mutated to alter the binding affinity of the antigen-binding site for a particular antigen or epitope. In certain embodiments, naturally occurring CDRs have been mutated in vivo by affinity maturation and somatic hypermutation in a host. In certain embodiments, the CDRs have been mutated in vitro by methods including, but not limited to, PCR mutagenesis and chemical mutagenesis. In various embodiments, the CDRs are synthetic sequences comprising CDRs obtained from, but not limited to, random sequence CDR libraries and rationally designed CDR libraries. CDR boundaries were determined using the Martin numbering scheme. See Figures 1A-1B as applied to the P9-01 anti-human GAL9 candidate provided herein.

[0149] In various embodiments, CDRs identified as binding to an antigen of interest are further mutated (i.e., "affinity maturation") to achieve desired binding characteristics, such as increased affinity for the antigen of interest compared to the original CDR. For example, targeted diversity can be introduced into CDRs, including those identified as binding to an antigen of interest, using degenerate oligonucleotides. Various randomization schemes can be used. For example, "soft-randomization" can be used, which can provide a high bias against the identity of the wild-type sequence at a given amino acid position, such as allowing a given position in a CDR to vary among all 20 amino acids while biasing against the wild-type sequence by doping four bases at each codon position at unequal levels. As an illustrative example of soft randomization, if it is desired to achieve approximately 50% wild-type sequence, each base in each codon can be kept 70% wild-type, and other nucleotides can be kept at 10% each. Degenerate oligonucleotides can be used to generate a phage library focused around the selected CDR, and the resulting phage particles can be used for phage panning under various stringent selection conditions, as needed.

[0150] 6.4.2.4. Framework Region and CDR Grafting The VH and VL amino acid sequences comprise "framework region" (FR) sequences. FRs are generally conserved sequence regions that act as scaffolds for the interspersed CDRs (see Section 6.4.2.3) and are typically in the following arrangement (N- to C-terminus): FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. In various embodiments, the FRs are mammalian sequences, including but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the FRs are human sequences. In various embodiments, the FRs are naturally occurring sequences. In various embodiments, the FRs are synthetic sequences, including but not limited to, rationally designed sequences.

[0151] In various embodiments, both the FRs and CDRs are derived from the same naturally occurring variable domain sequence. In various embodiments, the FRs and CDRs are derived from different variable domain sequences, with the CDRs grafted onto the FR scaffold, and the CDRs providing specificity for a particular antigen. In certain embodiments, the grafted CDRs are all derived from the same naturally occurring variable domain sequence. In certain embodiments, the grafted CDRs are derived from different variable domain sequences. In certain embodiments, the grafted CDRs are synthetic sequences, including but not limited to CDRs obtained from random sequence CDR libraries and rationally designed CDR libraries. In certain embodiments, the grafted CDRs and FRs are derived from the same species. In certain embodiments, the grafted CDRs and FRs are derived from different species. In preferred grafted CDR embodiments, the antibody is "humanized," where the grafted CDRs are non-human mammalian sequences, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, and goat sequences, and the FRs are human sequences. Humanized antibodies are discussed in more detail in U.S. Patent No. 6,407,213, the entirety of which is incorporated herein by reference for all it teaches. In various embodiments, portions or specific sequences of FRs from one species are used to replace portions or specific sequences of FRs from another species.

[0152] 6.4.3. Exemplary Amino Acid Sequences of GAL9-Binding Molecules In various embodiments, the GAL9 binding molecule comprises a particular VH CDR3 (CDR-H3) sequence and a particular VL CDR3 (CDR-L3) sequence.

[0153] In some embodiments, the GAL9 binding molecule comprises a CDR-H3 and a CDR-L3 from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57. The VH CDR amino acid sequences of the ABS clones are disclosed in Table 3. The VL CDR amino acid sequences of the ABS clones are disclosed in Table 4. For clarity, each GAL9 ABS clone has been assigned a unique ABS clone number that will be used throughout this disclosure.

[0154] In one currently preferred embodiment, the GAL9 binding molecule comprises the CDR-H3 and CDR-L3 of ABS clone P9-11.

[0155] In some embodiments, the GAL9-binding molecule comprises all three VH CDRs from one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57. In one currently preferred embodiment, the GAL9-binding molecule comprises all three VH CDRs from the ABS clone P9-11.

[0156] In some embodiments, the GAL9-binding molecule comprises all three VL CDRs from one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57. In one currently preferred embodiment, the GAL9-binding molecule comprises all three VL CDRs from the ABS clone P9-11.

[0157] In some embodiments, the GAL9 binding molecule comprises all six CDRs from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57. In one currently preferred embodiment, the GAL9 binding molecule comprises all six CDRs from ABS clone P9-11.

[0158] In some embodiments, the GAL9 binding molecule comprises a VH amino acid sequence, a VL amino acid sequence, or a VH and VL amino acid sequence from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57. The complete immunoglobulin heavy and light chain sequences, and the VH and VL amino acid sequences are provided in Table 6. In one currently preferred embodiment, the GAL9 binding molecule comprises the VH amino acid sequence, the VL amino acid sequence, or the VH and VL amino acid sequences derived from ABS clone P9-11.

[0159] In some embodiments, the GAL9-binding molecule comprises the complete IgG heavy and light chain sequences from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57. In one currently preferred embodiment, the GAL9-binding molecule comprises the complete IgG heavy and light chain sequences from the ABS clone P9-11.

[0160] 6.4.4. Constant Region In some embodiments, a GAL9-binding molecule comprises an antibody constant region domain sequence. As described herein, a constant region domain amino acid sequence is the sequence of an antibody constant region domain. A constant region can refer to a CH1, CH2, CH3, CH4, or CL constant domain.

[0161] In various embodiments, the constant region sequence is a mammalian sequence, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the constant region sequence is a human sequence. In certain embodiments, the constant region sequence is derived from an antibody light chain. In particular embodiments, the constant region sequence is derived from a lambda or kappa light chain. In certain embodiments, the constant region sequence is derived from an antibody heavy chain. In particular embodiments, the constant region sequence is an antibody heavy chain sequence that is an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4, or IgM isotype. In particular embodiments, the constant region sequence is derived from an IgG isotype. In preferred embodiments, the constant region sequence is derived from an IgG1 isotype.

[0162] Exemplary constant regions and modifications thereof are described in WO2018075692, which is incorporated herein by reference in its entirety.

[0163] 6.4.4.1. CH1 and CL regions The CH1 amino acid sequence is the sequence of the second domain of an antibody heavy chain, as described herein, from N-terminus to C-terminus of the native antibody heavy chain architecture. In certain embodiments, the CH1 sequence is an endogenous sequence. In various embodiments, the CH1 is a mammalian sequence, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the CH1 sequence is a human sequence. In certain embodiments, the CH1 sequence is derived from an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4, or IgM isotype. In preferred embodiments, the CH1 sequence is derived from an IgG1 isotype. In preferred embodiments, the CH1 sequence is amino acids 1-98 of UniProt Accession Number P01857.

[0164] The CL amino acid sequences useful in the GAL9-binding molecules described herein are antibody light chain constant domain sequences based on natural antibody light chain architecture. In certain embodiments, the CL sequence is an endogenous sequence. In various embodiments, the CL sequence is a mammalian sequence, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In a preferred embodiment, the CL sequence is a human sequence.

[0165] In certain embodiments, the CL amino acid sequence is a lambda (λ) light chain constant domain sequence. In certain embodiments, the CL amino acid sequence is a human lambda light chain constant domain sequence. In a preferred embodiment, the lambda (λ) light chain sequence is UniProt accession number P0CG04.

[0166] In certain embodiments, the CL amino acid sequence is a kappa (κ) light chain constant domain sequence. In a preferred embodiment, the CL amino acid sequence is a human kappa (κ) light chain constant domain sequence. In a preferred embodiment, the lambda light chain sequence is UniProt accession number P01834.

[0167] In certain embodiments, the CH1 sequence and the CL sequence are both endogenous sequences. In certain embodiments, the CH1 sequence and the CL sequence separately comprise orthogonal modifications to the endogenous CH1 and CL sequences, respectively, as discussed in more detail in Section 6.4.4.1 below. The CH1 and CL sequences may also be portions thereof, either endogenous or modified, such that a domain having a CH1 sequence or a portion thereof can associate with a domain having a CL sequence or a portion thereof.

[0168] 6.4.4.2. CH1 and CL Orthogonal Modifications In certain embodiments, the CH1 and CL sequences separately comprise orthogonal modifications to the endogenous CH1 and CL sequences, respectively. Orthogonal mutations are generally described in more detail below in Sections 6.4.6.1-6.4.6.3.

[0169] In certain embodiments, the orthogonal modification in the endogenous CH1 and CL sequences is an engineered disulfide bridge selected from engineered cysteines at position 138 of the CH1 sequence and position 116 of the CL sequence, at position 128 of the CH1 sequence and position 119 of the CL sequence, or at position 129 of the CH1 sequence and position 210 of the CL sequence, numbering as discussed in more detail in U.S. Patent Nos. 8,053,562 and 9,527,927, each of which is incorporated herein by reference in its entirety. In a preferred embodiment, the engineered cysteines are at position 128 of the CH1 sequence and position 118 of the CL kappa sequence, numbering as per the Eu index.

[0170] In one set of preferred embodiments, the mutation providing a non-endogenous cysteine ​​amino acid is an F118C mutation in a CL sequence with a corresponding A141C in the CH1 sequence, or an F118C mutation in a CL sequence with a corresponding L128C in the CH1 sequence, or an S162C mutation in a CL sequence with a corresponding P171C mutation in the CH1 sequence, numbering as per the Eu index.

[0171] In various embodiments, the orthogonal mutations in the CL and CH1 sequences are charge-pair mutations. In particular embodiments, the charge-pair mutation is an F118S, F118A, or F118V mutation in the CL sequence with a corresponding A141L in the CH1 sequence, or a T129R mutation in the CL sequence with a corresponding K147D in the CH1 sequence, as numbered according to the Eu index, and as described in more detail in Bonisch et al. (Protein Engineering, Design & Selection, 2017, pp. 1-12). In a preferred set of embodiments, the charge-pair mutation is an N138K mutation in the CL sequence with a corresponding G166D in the CH1 sequence, or an N138D mutation in the CL sequence with a corresponding G166K in the CH1 sequence, as numbered according to the Eu index.

[0172] 6.4.4.3. CH2 area In the GAL9-binding molecules described herein, the GAL9-binding molecule may have a CH2 amino acid sequence. The CH2 amino acid sequence is the CH2 amino acid sequence of the third domain of an antibody heavy chain, measured from the N-terminus to the C-terminus of a native antibody heavy chain structure, as described herein. In various embodiments, the CH2 is a mammalian sequence, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the CH2 sequence is a human sequence. In certain embodiments, the CH2 sequence is derived from an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4, or IgM isotype. In preferred embodiments, the CH2 sequence is derived from an IgG1 isotype.

[0173] In certain embodiments, the CH2 sequence is an endogenous sequence. In a preferred embodiment, the sequence is amino acids 111-223 of UniProt Accession No. P01857.

[0174] In one series of embodiments, the GAL9-binding molecule has one or more paired sets of CH2 domains with CH2 sequences, where the first set has a CH2 amino acid sequence from a first isotype and one or more orthologous sets of CH2 amino acid sequences from another isotype. The orthologous CH2 amino acid sequences can interact with the CH2 amino acid sequence from the common isotype, as described herein, but do not significantly interact with CH2 amino acid sequences from another isotype present in the GAL9-binding molecule. In certain embodiments, all of the set of CH2 amino acid sequences are from the same species. In preferred embodiments, all of the set of CH2 amino acid sequences are human CH2 amino acid sequences. In other embodiments, the set of CH2 amino acid sequences are from different species. In certain embodiments, the CH2 amino acid sequences of the first set are from the same isotype as other non-CH2 domains in the GAL9-binding molecule. In certain embodiments, the first set has a CH2 amino acid sequence from an IgG isotype, and one or more of the orthologous sets have CH2 amino acid sequences from an IgM or IgE isotype. In certain embodiments, one or more of the set of CH2 amino acid sequences are endogenous CH2 sequences. In other embodiments, one or more of the set of CH2 amino acid sequences are endogenous CH2 sequences with one or more mutations. In certain embodiments, the one or more mutations are orthogonal knob-hole mutations, orthogonal charge-pair mutations, or orthogonal hydrophobic mutations. Orthologous CH2 amino acid sequences useful for GAL9-binding molecules are described in more detail in WO 2017 / 011342 and WO 2017 / 106462, which are incorporated herein by reference in their entireties.

[0175] 6.4.4.4. CH3 area The CH3 amino acid sequence is the sequence of the C-terminal domain of an antibody heavy chain, as described herein, relative to the N-terminus to C-terminus of the native antibody heavy chain architecture.

[0176] In various embodiments, the CH3 is a mammalian sequence, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the CH3 sequence is a human sequence. In certain embodiments, the CH3 sequence is derived from an IgA1, IgA2, IgD, IgE, IgM, IgG1, IgG2, IgG3, or IgG4 isotype, or the CH4 sequence is derived from an IgE or IgM isotype. In certain embodiments, the CH3 sequence is derived from an IgG isotype. In preferred embodiments, the CH3 sequence is derived from an IgG1 isotype.

[0177] In certain embodiments, the CH3 sequence is an endogenous sequence. In particular embodiments, the CH3 sequence is amino acids 224-330 of UniProt Accession Number P01857. In various embodiments, the CH3 sequence is a segment of an endogenous CH3 sequence. In particular embodiments, the CH3 sequence has an endogenous CH3 sequence lacking the N-terminal amino acids G224 and Q225. In particular embodiments, the CH3 sequence has an endogenous CH3 sequence lacking the C-terminal amino acids P328, G329, and K330. In particular embodiments, the CH3 sequence has an endogenous CH3 sequence lacking both the N-terminal amino acids G224 and Q225 and the C-terminal amino acids P328, G329, and K330. In preferred embodiments, the GAL9-binding molecule has multiple domains with CH3 sequences, and the CH3 sequence can refer to both the complete endogenous CH3 sequence and a CH3 sequence lacking the N-terminal amino acids, the C-terminal amino acids, or both.

[0178] In certain embodiments, the CH3 sequence is an endogenous sequence with one or more mutations, hi certain embodiments, the mutations are one or more orthogonal mutations introduced into the endogenous CH3 sequence to lead to a particular pair of particular CH3 sequences, as described in more detail below in Sections 6.4.6.1-6.4.6.3.

[0179] In certain embodiments, the CH3 sequence is engineered to reduce the immunogenicity of the antibody by replacing specific amino acids of one allotype with amino acids of another allotype, referred to herein as isoallotypic mutations, as described in more detail in Stickler et al. (Genes Immun. 2011 Apr;12(3):213-221), which is incorporated herein by reference for all it teaches. In certain embodiments, specific amino acids of the Glml allotype are replaced. In a preferred embodiment, the CH3 sequence has the isoallotypic mutations D356E and L358M.

[0180] In some embodiments, the IgG1 CH3 amino acid sequence comprises the following mutational changes: P343V; Y349C; and tripeptide insertions 445P, 446G, 447K. In other preferred embodiments, domain B has a human IgG1 CH3 sequence with the following mutational changes: T366K; and tripeptide insertions 445K, 446S, 447C. In yet other preferred embodiments, domain B has a human IgG1 CH3 sequence with the following mutational changes: Y349C and tripeptide insertions 445P, 446G, 447K.

[0181] In some embodiments, the IgG1 CH3 amino acid sequence comprises a 447C mutation that is otherwise incorporated into the endogenous CH3 sequence.

[0182] 6.4.5. Antigen binding site In some embodiments, the VL or VH amino acid sequence and the cognate VL or VH amino acid sequence associate to form a first antigen-binding site (ABS). The antigen-binding site (ABS) is capable of specifically binding to an epitope of an antigen. Antigen binding by an ABS is described in more detail in Section 6.4.5.1, below.

[0183] In alternative embodiments, for example, when the GAL9 binding molecule is a single domain antibody, the VH or VL amino acid sequence forms the first ABS.

[0184] In some embodiments, the GAL9 antigen-binding molecule comprises a second antigen-binding site. In some embodiments, the second ABS is specific for the same GAL9 antigen as the first ABS. In some embodiments, the second ABS specifically binds to the same epitope on the GAL9 antigen as the first ABS. In some embodiments, the second ABS is identical to the first ABS.

[0185] In some embodiments, the second ABS is specific for a different epitope of the first GAL9 antigen, for example, when the first ABS comprises CDRs or variable domains from any one of the ABS clones selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57. The second ABS may comprise CDRs or variable domains from another ABS clone selected from P9-01, P9-02A, P9-03, P9-06, P9-07, P9-11, P9-12, P9-14, P9-23, P9-24, P9-25, P9-29, P9-30, P9-34, P9-37, P9-38, P9-40, P9-41, P9-42, P9-43, P9-44, P9-45, P9-46, P9-50, P9-51, P9-52, P9-53, P9-56, and P9-57.

[0186] In some embodiments, the GAL9 antigen binding molecule is multispecific, e.g., a second ABS of the GAL9 antigen binding molecule specifically binds to a different GAL9 antigen than the first ABS specifically binds.

[0187] 6.4.5.1. Antigen binding by ABS ABSs, and GAL9-binding molecules that contain such ABSs, are said to "recognize" the epitope (or more generally, antigen) to which the ABS specifically binds, and the epitope (or more generally, antigen) is said to be the "recognition specificity" or "binding specificity" of the ABS.

[0188] An ABS is said to bind with a particular affinity to its specific antigen or epitope. As used herein, "affinity" refers to the strength of interaction of non-covalent intermolecular forces between one molecule and another. Affinity, or the strength of interaction, is determined by the dissociation equilibrium constant (K D ) and K D A lower value indicates a stronger interaction between the molecules. D Values ​​are measured by methods well known in the art, including, but not limited to, biolayer interferometry (e.g., Octet / FORTEBIO®), surface plasmon resonance (SPR) technology (e.g., Biacore®), and cell binding assays. For purposes herein, affinity is the dissociation equilibrium constant as measured by biolayer interferometry using Octet / FORTEBIO®.

[0189] "Specific binding," as used herein, refers to the affinity between an ABS and its cognate antigen or epitope, K D The value is 10 -6 M, 10 -7 M, 10 -8 M, 10 -9 M or 10 -10 It is less than M.

[0190] The number of ABSs in a GAL9-binding molecule as described herein defines the "valency" of the GAL9-binding molecule. A GAL9-binding molecule with a single ABS is "monovalent." A GAL9-binding molecule with multiple ABSs is said to be "multivalent." A multivalent GAL9-binding molecule with two ABSs is "bivalent." A multivalent GAL9-binding molecule with three ABSs is "trivalent." A multivalent GAL9-binding molecule with four ABSs is "tetravalent."

[0191] In various multivalent embodiments, the ABSs of the plurality all have the same recognition specificity. Such a GAL9-binding molecule is a "monospecific" or "multivalent" binding construct. In other multivalent embodiments, at least two of the ABSs of the plurality have different recognition specificities. Such a GAL9-binding molecule is multivalent and "multispecific." In multivalent embodiments in which the ABSs collectively have two recognition specificities, the GAL9-binding molecule is "bispecific." In multivalent embodiments in which the ABSs collectively have three recognition specificities, the GAL9-binding molecule is "trispecific."

[0192] In multivalent embodiments in which the ABSs collectively have multiple recognition specificities for different epitopes on the same antigen, the GAL9 binding molecule is "multiparatopic." Multivalent embodiments in which the ABSs collectively recognize two epitopes on the same antigen are "biparatopic."

[0193] In various multivalent embodiments, the multivalency of a GAL9-binding molecule improves the avidity of the GAL9-binding molecule for a particular target. As described herein, "avidity" refers to the overall strength of an interaction between two or more molecules, e.g., a multivalent GAL9-binding molecule for a particular target; avidity is the cumulative strength of the interaction provided by the affinity of multiple ABSs. Avidity can be measured by the same methods used to determine affinity, as described above. In certain embodiments, the avidity of a GAL9-binding molecule for a particular target is determined by the interaction being a specific binding interaction, and the avidity between two molecules is greater than 10. -6 M, 10 -7 M, 10 -8 M, 10 -9 M or 10 -10 K less than M D In certain embodiments, the avidity of a GAL9 binding molecule for a particular target has a K value such that the interaction is a specific binding interaction. DThe affinity of each ABS has a K value, and the affinity of each ABS is a K value that is sufficient for specific binding to the respective antigen or epitope. D In certain embodiments, avidity is the cumulative strength of interaction provided by the affinity of multiple ABSs for different antigens on a common specific target or complex, such as different antigens found on individual cells. In certain embodiments, avidity is the cumulative strength of interaction provided by the affinity of multiple ABSs for different epitopes on a common individual antigen.

[0194] 6.4.6. Orthogonal Modification In the GAL9 binding molecules described herein, the GAL9 binding molecules may have constant region domains that include orthogonal modifications. Constant region domain amino acid sequences are described in more detail in Section 6.4.4, above.

[0195] As used herein, an "orthogonal modification" or, synonymously, an "orthogonal mutation" is one or more genetically engineered mutations in the amino acid sequence of an antibody domain that increase the binding affinity of a first domain bearing the orthogonal modification for a second domain bearing the complementary orthogonal modification. In certain embodiments, the orthogonal modification decreases the affinity of a domain bearing the orthogonal modification for a domain lacking the complementary orthogonal modification. In certain embodiments, the orthogonal modification is a mutation in the endogenous antibody domain sequence. In various embodiments, the orthogonal modification is a modification of the N- or C-terminus of the endogenous antibody domain sequence, including, but not limited to, amino acid additions or deletions. In certain embodiments, the orthogonal modification includes, but is not limited to, genetically engineered disulfide bridges, knob-in-hole mutations, and charge pair mutations, as described in more detail in Sections 6.4.6.1-6.4.6.3, below. In certain embodiments, orthogonal modifications include combinations of orthogonal modifications selected from, but not limited to, engineered disulfide bridges, knobs-into-holes mutations, and charge-pair mutations. In certain embodiments, orthogonal modifications may be combined with amino acid substitutions that reduce immunogenicity, such as isoallotypic mutations, as described in more detail in Section 6.4.4.4, above.

[0196] 6.4.6.1. Orthogonal engineered disulfide bridges In various embodiments, the orthogonal modification comprises a mutation that generates an engineered disulfide bridge between the first and second domains. As described herein, an "engineered disulfide bridge" is a mutation that provides a non-endogenous cysteine ​​amino acid to two or more domains such that a non-native disulfide bond forms when the two or more domains associate. Engineered disulfide bridges are described in more detail in Merchant et al. (Nature Biotech (1998) 16:677-681), the entire contents of which are incorporated herein by reference for all that it teaches. In certain embodiments, the engineered disulfide bridge enhances orthogonal binding between certain domains. 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 mutation in the other CH3 domain. In a preferred embodiment, the engineered disulfide bridge-generating mutations are a S354C mutation in one of the first or second CH3 domains and a Y349C mutation in the other CH3 domain. In another preferred embodiment, the engineered disulfide bridge-generating mutations are a 447C mutation in both the first and second CH3 domains, provided by a C-terminal extension of the CH3 domains incorporating the KSC tripeptide sequence.

[0197] 6.4.6.2. Orthogonal Knob-Hole Mutations In various embodiments, the orthogonal modification comprises a knob-hole (synonymously, knob-in-hole) mutation. As described herein, a knob-hole mutation is a mutation that alters the surface conformation of a first domain such that the first domain preferentially associates with a second domain that has a complementary conformational mutation compared to association with a domain that does not have the complementary conformational mutation. Knob-hole mutations are described in more detail in U.S. Pat. Nos. 5,821,333 and 8,216,805, each of which is incorporated herein in its entirety. In various embodiments, a knob-hole mutation is combined with an engineered disulfide bridge, as described in more detail in Merchant et al. (Nature Biotech (1998) 16:677-681). In various embodiments, a knob-hole mutation, an isoallotypic mutation, and an engineered disulfide mutation are combined.

[0198] In certain embodiments, the knobs-in-hole mutations are a T366Y mutation in the first domain and a Y407T mutation in the second domain. In certain embodiments, the knobs-in-hole mutations are F405A in the first domain and T394W in the second domain. In certain embodiments, the knobs-in-hole mutations are a T366Y mutation and F405A in the first domain, and T394W and Y407T in the second domain. In certain embodiments, the knobs-in-hole mutations are a T366W mutation in the first domain and a Y407A mutation in the second domain. In certain embodiments, the combination of knobs-in-hole mutations and engineered disulfide mutations is S354C and T366W mutations in the first domain, and Y349C, T366S, L368A, and Y407V mutations in the second domain. In a preferred embodiment, the combination of knobs-in-holes mutations, isoallotypic mutations, and engineered disulfide mutations is S354C and T366W mutations in the first domain, and Y349C, D356E, L358M, T366S, L368A, and Y407V mutations in the second domain.

[0199] 6.4.6.3. Orthogonal Charge Pair Mutations In various embodiments, the orthogonal modification is a charge-pair mutation. As used herein, a charge-pair mutation is a mutation that affects the charge of an amino acid on the surface of a domain, such that the domain will preferentially associate with a second domain that has a complementary charge-pair mutation compared to association with a domain that does not have the complementary charge-pair mutation. In certain embodiments, the charge-pair mutation improves orthogonal association between certain domains. Charge-pair mutations are described in more detail in U.S. Pat. Nos. 8,592,562, 9,248,182, and 9,358,286, each of which is incorporated herein by reference for all that it teaches. In certain embodiments, the charge-pair mutation improves stability between certain domains. In a preferred embodiment, the charge-pair mutations are a T366K mutation in one domain and an L351D mutation in the other domain.

[0200] In certain embodiments, the orthogonal mutation is a charge-pair mutation at the VH / VL interface. In a preferred embodiment, the charge-pair mutation at the VH / VL interface is Q39E in VH and Q38K in the corresponding VL, or Q39K in VH and Q38E in the corresponding VL, as described in more detail in Igawa et al. (Protein Eng. Des. Sel. 2010, 23, 667-677), which is incorporated herein by reference for all it teaches.

[0201] 6.4.7. Trivalent and Tetravalent GAL9 Binding Molecules In another set of embodiments, the GAL9 binding molecule has three antigen-binding sites and is therefore referred to as "trivalent." In various embodiments, the GAL9 binding molecule has four antigen-binding sites and is therefore referred to as "tetravalent."

[0202] 6.5. GAL9-binding molecular architecture The antigen-binding sites described herein comprising specific CDR subsets can be formatted into any binding molecule architecture, including, but not limited to, full-length antibodies, Fab fragments, Fvs, scFvs, tandem scFvs, diabodies, scdiabodies, DARTs, tandAbs, minibodies, camelid VHHs, and other antibody fragments or formats known to those of skill in the art. Exemplary antibody and antibody fragment formats are described in detail in Brinkmann et al. (MABS, 2017, Vol. 9, No. 2, pp. 182-212), which is incorporated herein by reference for all that it teaches. The antigen-binding sites described herein comprising specific CDR subsets can also be formatted into a "B-body" format, as described in more detail in U.S. Patent Application Publication No. 2018 / 0118811 and WO 2018 / 075692, each of which is incorporated herein by reference in its entirety.

[0203] 6.6. Further Modifications In a further set of embodiments, the GAL9 binding molecule has additional modifications.

[0204] 6.6.1. Antibody-Drug Conjugates In various embodiments, the GAL9-binding molecule is conjugated to a therapeutic agent (i.e., a drug) to form a GAL9-binding molecule-drug conjugate. Therapeutic agents include, but are not limited to, chemotherapeutic agents, imaging agents (e.g., radioisotopes), immunomodulatory agents (e.g., cytokines, chemokines, or checkpoint inhibitors), and toxins (e.g., cytotoxic agents). In certain embodiments, the therapeutic agent is attached to the GAL9-binding molecule via a linker peptide, which is discussed in more detail in Section 6.6.3, below.

[0205] Methods for preparing antibody-drug conjugates (ADCs) that can be configured to conjugate drugs to the GAL9 binding molecules disclosed herein are described in the following documents: for example, U.S. Pat. No. 8,624,003 (Pot method), U.S. Pat. No. 8,163,888 (one-step method), U.S. Pat. No. 5,208,020 (two-step method), U.S. Pat. No. 8,337,856, U.S. Pat. No. 5,773,001, U.S. Pat. No. 7,829,531, U.S. Pat. No. 5, 208,020, U.S. Patent No. 7,745,394, WO 2017 / 136623, WO 2017 / 015502, WO 2017 / 015496, WO 2017 / 015495, WO 2004 / 010957, WO 2005 / 077090, WO 2005 / 082023, WO 2006 / 065533, WO International Publication No. 2007 / 030642, International Publication No. 2007 / 103288, International Publication No. 2013 / 173337, International Publication No. 2015 / 057699, International Publication No. 2015 / 095755, International Publication No. 2015 / 123679, International Publication No. 2015 / 157286, International Publication No. 2017 / 165851, International Publication No. 2009 / 073445, International Publication No. 2010 / 068759 Brochure, International Publication No. 2010 / 138719, International Publication No. 2012 / 171020, International Publication No. 2014 / 008375, International Publication No. 2014 / 093394, International Publication No. 2014 / 093640, International Publication No. 2014 / 160360, International Publication No. 2015 / 054659, International Publication No. 2015 / 195925, International Publication No. 2017 / 160754, Storz (MAbs.(Adv Ther, 2017, 34: 1015), Diamantis et al. (British Journal of Cancer, 2016, 114: 362-367), Carrico et al. (Nat Chem Biol, 2007, 3: 321-2), We et al. (Proc Natl Acad Sci USA, 2009, 106: 3000-5), Rabuka et al. (Curr Opin Chem Biol, 2011, 14: 790-6), Hudak et al. (Angew Chem Int Ed Engl., 2012, 4161-5), Rabuka et al. (Nat Protoc, 2012 (Proc Natl Acad Sci USA., 2013, 110:46-51), Agarwal et al. (Bioconjugate Chem., 2013, 24:846-851), Barfield et al. (Drug Dev. and D., 2014, 14:34-41), Drake et al. (Bioconjugate Chem., 2014, 25:1331-41), Liang et al. (J Am Chem Soc., 2014, 136:10850-3), Drake et al. (Curr Opin Chem Biol., 2015, 28:174-80), and York et al. (BMC Biotechnology, 2016, 16(1):23). Each of these documents is incorporated herein by reference in its entirety for all it teaches.

[0206] 6.6.2. Additional Binding Parts In various embodiments, the GAL9-binding molecule is modified to include one or more additional binding moieties. In certain embodiments, the binding moiety is an antibody fragment or antibody format, including, but not limited to, a full-length antibody, a Fab fragment, an Fv, an scFv, a tandem scFv, a diabody, an scdiabody, a DART, a tandAb, a minibody, a camelid VHH, and other antibody fragments or formats known to those of skill in the art. Exemplary antibody and antibody fragment formats are described in detail in Brinkmann et al. (MABS, 2017, Vol. 9, No. 2, pp. 182-212), which is incorporated herein by reference for all that it teaches.

[0207] In certain embodiments, the one or more additional binding moieties are attached to the C-terminus of the first or third polypeptide chain. In certain embodiments, the one or more additional binding moieties are attached to the C-terminus of both the first and third polypeptide chain. In certain embodiments, the one or more additional binding moieties are attached to the C-terminus of both the first and third polypeptide chain. In certain embodiments, individual portions of the one or more additional binding moieties are separately attached to the C-terminus of the first and third polypeptide chain such that they form functional binding moieties.

[0208] In certain embodiments, one or more additional binding moieties are attached to the N-terminus of any of the polypeptide chains (e.g., the first, second, third, fourth, fifth, or sixth polypeptide chain). In certain embodiments, individual portions of the additional binding moieties are separately attached to the N-terminus of different polypeptide chains such that they form functional binding moieties.

[0209] In certain embodiments, the one or more additional binding moieties are specific for different antigens or epitopes of the ABS within the GAL9-binding molecule. In certain embodiments, the one or more additional binding moieties are specific for the same antigen or epitope of the ABS within the GAL9-binding molecule. In certain embodiments, when the modification is two or more additional binding moieties, the additional binding moieties are specific for the same antigen or epitope. In certain embodiments, when the modification is two or more additional binding moieties, the additional binding moieties are specific for different antigens or epitopes.

[0210] In certain embodiments, one or more additional binding moieties are attached to the GAL9-binding molecule using in vitro methods, including, but not limited to, reactive chemistry and affinity tagging systems, as discussed in more detail in Section 6.6.3, below. In certain embodiments, one or more additional binding moieties are attached to the GAL9-binding molecule via Fc-mediated binding (e.g., Protein A / G). In certain embodiments, one or more additional binding moieties are attached to the GAL9-binding molecule using recombinant DNA techniques, such as encoding the nucleotide sequence of a fusion product of the GAL9-binding molecule and the additional binding moiety on the same expression vector (e.g., a plasmid).

[0211] 6.6.3. Functional / Reactive Groups In various embodiments, the GAL9 binding molecules have modifications that include functional or chemically reactive groups that can be used in downstream processing, such as conjugation with additional moieties (e.g., drug conjugates and additional binding moieties as discussed in more detail above in Sections 6.6.1. and 6.6.2.), and downstream purification processes.

[0212] In certain embodiments, the modification is a chemically reactive group, including, but not limited to, a reactive thiol (e.g., maleimide-based reactive group), a reactive amine (e.g., N-hydroxysuccinimide-based reactive group), a "click chemistry" group (e.g., a reactive alkyne group), and an aldehyde with formylglycine (FGly). In certain embodiments, the modification is a functional group, including, but not limited to, an affinity peptide sequence (e.g., HA, HIS, FLAG, GST, MBP, Strep system, etc.). In certain embodiments, the functional group or chemically reactive group comprises a cleavable peptide sequence. In certain embodiments, the cleavable peptide is cleaved by means including, but not limited to, photocleavage, chemical cleavage, protease cleavage, reducing conditions, and pH conditions. In certain embodiments, the protease cleavage is performed by an intracellular protease. In certain embodiments, the protease cleavage is performed by an extracellular or membrane-bound protease. ADC therapy utilizing protease cleavage is described in more detail in Choi et al. (Theranostics, 2012;2(2):156-178), which is incorporated herein by reference for all it teaches.

[0213] 6.6.4. Reduction of effector function In certain embodiments, the GAL9-binding molecule has one or more genetically engineered mutations in the amino acid sequence of the antibody domain that reduce effector functions naturally associated with antibody binding, including, but not limited to, cellular functions attributable to Fc receptors binding the Fc portion of an antibody, such as antibody-dependent cellular cytotoxicity (ADCC, also known as antibody-dependent cell-mediated cytotoxicity), complement fixation (e.g., Clq binding), antibody-dependent cell-mediated phagocytosis (ADCP), and opsonization. Exemplary engineered mutations that reduce effector function are described in more detail in U.S. Patent Application Publication No. 2017 / 0137530, Armour et al. (Eur. J. Immunol. 29(8) (1999) 2613-2624), Shields et al. (J. Biol. Chem. 276(9) (2001) 6591-6604), and Oganesyan et al. (Acta Cristallographica D64 (2008) 700-704), each of which is incorporated herein by reference in its entirety.

[0214] 6.7. Purification method Provided herein are methods for purifying GAL9-binding molecules. Purification steps include, but are not limited to, purifying GAL9-binding molecules based on protein characteristics, such as size (e.g., size exclusion chromatography), charge (e.g., ion exchange chromatography), or hydrophobicity (e.g., hydrophobic interaction chromatography). In one embodiment, cation exchange chromatography is performed. Other purification methods known to those skilled in the art may also be performed, including, but not limited to, the use of Protein A, Protein G, or Protein A / G reagents. Multiple iterations of a single purification method may also be performed. A combination of purification methods may also be performed.

[0215] 6.7.1. Complex Assembly and Purity In embodiments of the present invention, at least four distinct polypeptide chains associate together to form a complete complex, i.e., a GAL9-binding molecule. However, incomplete complexes that do not contain at least four distinct polypeptide chains can also be formed. For example, incomplete complexes can be formed that have only one, two, or three polypeptide chains. In other examples, incomplete complexes can contain more than three polypeptide chains but do not contain at least four distinct polypeptide chains. For example, incomplete complexes improperly associate with one or more copies of a distinct polypeptide chain. In methods of the present invention, complexes, i.e., fully assembled GAL9-binding molecules, are purified from incomplete complexes.

[0216] Methods for assessing the effectiveness and efficiency of purification steps are well known to those skilled in the art and include, but are not limited to, SDS-PAGE analysis, ion exchange chromatography, size exclusion chromatography, and mass spectrometry. Purity can also be assessed according to various criteria. Exemplary criteria include, but are not limited to, 1) assessing the percentage of total protein in the eluate provided by fully assembled GAL9-binding molecules; 2) assessing the enrichment factor or percent increase of the method for purifying the desired product, e.g., comparing the total protein provided by fully assembled GAL9-binding molecules in the eluate with the total protein in the starting sample; and 3) assessing the percentage of total protein or the percent decrease of undesired products, such as the incomplete complexes described above, including determining the percentage or decrease of specific undesired products (e.g., unassociated single polypeptide chains, dimers of any combination of peptide chains, or trimers of any combination of polypeptide chains). Purity can be assessed after any combination of the methods described herein.

[0217] 6.8. Manufacturing method The GAL9-binding molecules described herein can be readily produced by expression using standard cell-free translation, transient transfection, and stable transfection techniques currently used for antibody production. In certain embodiments, Expi293 cells (ThermoFisher) can be used to produce the GAL9-binding molecules using ThermoFisher protocols or reagents such as ExpiFectaine, or other reagents known to those of skill in the art, such as polyethyleneimine, as described in detail in Fang et al. (Biological Procedures Online, 2017, 19:11).

[0218] The expressed protein can be readily separated from undesired proteins and protein complexes using a variety of purification strategies, including but not limited to, the use of Protein A, Protein G, or Protein A / G reagents. Further purification can be achieved using ion exchange chromatography, as routinely used in the art.

[0219] 6.9. Pharmaceutical Compositions In another aspect, there is provided a pharmaceutical composition comprising a GAL9 binding molecule described herein and a pharmaceutically acceptable carrier or diluent. In typical embodiments, the pharmaceutical composition is sterile.

[0220] In various embodiments, the pharmaceutical composition comprises a GAL9-binding molecule at a concentration of 0.1 mg / ml to 100 mg / ml. In specific embodiments, the pharmaceutical composition comprises a GAL9-binding molecule at a concentration of 0.5 mg / ml, 1 mg / ml, 1.5 mg / ml, 2 mg / ml, 2.5 mg / ml, 5 mg / ml, 7.5 mg / ml, or 10 mg / ml. In some embodiments, the pharmaceutical composition comprises a GAL9-binding molecule at a concentration greater than 10 mg / ml. In certain embodiments, the GAL9-binding molecule is present at a concentration of 20 mg / ml, 25 mg / ml, 30 mg / ml, 35 mg / ml, 40 mg / ml, 45 mg / ml, or even 50 mg / ml or higher. In certain embodiments, the GAL9-binding molecule is present at a concentration greater than 50 mg / ml.

[0221] In various embodiments, pharmaceutical compositions are described in more detail in U.S. Patent Nos. 8,961,964, 8,945,865, 8,420,081, 6,685,940, 6,171,586, 8,821,865, 9,216,219, U.S. Patent Application No. 10 / 813,483, WO 2014 / 066468, WO 2011 / 104381, and WO 2016 / 180941, each of which is incorporated herein in its entirety.

[0222] 6.10. Treatment In another aspect, a method of treatment is provided that includes administering a GAL9-binding molecule as described herein to a patient (e.g., a subject) having a disease or condition in an amount effective to treat the patient (e.g., a therapeutically effective amount).

[0223] Target In some embodiments, the subject is a mammal. In some embodiments, the mammal is a mouse. In preferred embodiments, the mammal is a human. In some embodiments, immune cells of the subject exhibit increased PD-L2 expression compared to immune cells from a healthy individual (e.g., a healthy control), such as blood dendritic cells.

[0224] Combination Therapy GAL9-binding molecules can be used alone or in combination with other therapeutic agents or procedures to treat or prevent diseases or conditions. The GAL9-binding molecules can be administered either simultaneously or sequentially, depending on the disease or condition to be treated.

[0225] The anti-GAL9 binding molecules can be used in combination with agents or procedures that are used in the clinic or are within the current standard of care for treating or preventing the disease or condition. In some embodiments, the GAL9-binding molecule is administered in combination with a second immunosuppressant. In certain embodiments, the second immunosuppressant is a glucocorticoid (e.g., prednisone, dexamethasone, or hydrocortisone), a cytostatic, anti-cytokine antibody, including anti-TNF, anti-IL1, anti-IL5, anti-IL-6, IL-17, and anti-IL-23 antibodies, as well as a small molecule drug that reduces inflammatory cytokine signaling, such as a JAK / STAT inhibitor, methotrexate, hydroxychloroquine, chloroquine, an anti-CD25 or anti-CD52 antibody, or a drug that acts on an immunophilin (e.g., cyclosporine or sirolimus, or other drug known to inhibit or prevent immune system activity).

[0226] In some embodiments, the GAL9 binding molecule is administered in combination with one or more anti-inflammatory agents.

[0227] 6.10.3. Autoimmune or inflammatory diseases In some embodiments, the treatment comprises administering a GAL9 binding molecule as described herein to a subject with an autoimmune or inflammatory disease in an amount effective to treat the subject.

[0228] In some embodiments, the autoimmune disease is selected from the group consisting of amyotrophic lateral sclerosis (ALS), indolent rheumatoid arthritis, Addison's disease, adult Still's disease, agammaglobulinemia, alopecia areata, amyloidosis, ankylosing spondylitis, anti-GBM / anti-TBM nephritis, antiphospholipid syndrome, autoimmune angioedema, autoimmune dysautonomia, autoimmune encephalomyelitis, autoimmune hepatitis, autoimmune inner ear disease, autoimmune myocarditis, autoimmune oophoritis, autoimmune orchitis, autoimmune pancreatitis, autoimmune retinopathy, autoimmune urticaria, axonal and neuronal neuropathy (AMAN), Barrow's disease, Behcet's disease, benign mucosal ulcers, and ulcerative colitis. Pemphigus, bullous pemphigoid, Castleman's disease, celiac disease, Chagas' disease, chronic inflammatory demyelinating polyneuropathy, chronic relapsing multiple osteomyelitis, Churg-Strauss syndrome, eosinophilic granulomatosis, cicatricial pemphigoid, Cogan's syndrome, cold agglutinin disease, congenital heart block, Coxsackie myocarditis, CREST syndrome, Crohn's disease, dermatitis herpetiformis, dermatomyositis, Devic's disease (neuromyelitis optica), discoid lupus, Dressler's syndrome, endometriosis, eosinophilic esophagitis (EoE), eosinophilic fasciitis, erythema nodosum, essential mixed cryoglobulinemia, Evans' syndrome, fibromyalgia, fibromyalgia Alveolitis, giant cell arteritis (temporal arteritis), giant cell myocarditis, glomerulonephritis, Goodpasture's syndrome, granulomatosis with polyangiitis, Graves' disease, Guillain-Barré syndrome, Hashimoto's thyroiditis, hemolytic anemia, Henoch-Schönlein purpura (HSP), herpes gestationis or pemphigoid of pregnancy (PG), hidradenitis suppurativa (HS) (acne inversa), hypogammaglobulinemia, IgA nephropathy, IgG4-related sclerosing disease, immune thrombocytopenic purpura (ITP), inclusion body myositis, interstitial cystitis, juvenile arthritis, juvenile diabetes mellitus (type 1 diabetes) ), juvenile myositis, Kawasaki disease, Lambert-Eaton syndrome, leukocytoclastic vasculitis, lichen planus, lichen sclerosus, lignified conjunctivitis, linear immunoglobulin A disease (LAD), lupus, chronic Lyme disease, Meniere's disease, microscopic polyangiitis, mixed connective tissue disease (MCTD), Mooren's ulcer, Much-Habermann disease, multifocal motor neuropathy (MMN) or MMNCB, multiple sclerosis, myasthenia gravis, myositis, narcolepsy, neonatal lupus, neuromyelitis optica, neutropenia, ocular cicatricial pemphigoid, optic neuritis, relapsing rheumatoid arthritis (PR), PANDAS, paraneoplastic cerebellar degeneration (PCD),Paroxysmal nocturnal hemoglobinuria (PNH), Parry Romberg syndrome, peripheral uveitis (peripheral uveitis), Parsonage-Turner syndrome, pemphigus, peripheral neuropathy, perivenous encephalomyelitis, pernicious anemia (pa), POEMS syndrome, polyarteritis nodosa, polyglandular syndrome type I, II, or III, polymyalgia rheumatica, polymyositis, post-myocardial infarction syndrome, post-pericardiotomy syndrome, primary biliary cirrhosis, primary sclerosing cholangitis, progestational dermatitis, psoriasis, psoriatic arthritis, pure red cell aplasia, pyoderma gangrenosum, Raynaud's phenomenon, reactive arthritis, reflex sympathetic dystrophy The following conditions are considered to be pathological conditions: relapsing polychondritis, restless legs syndrome, retroperitoneal fibrosis, rheumatic fever, rheumatoid arthritis, sarcoidosis, Schmidt's syndrome, scleritis, scleroderma, Sjogren's syndrome, sperm and testicular autoimmunity, stiff-person syndrome, subacute bacterial endocarditis, Susac's syndrome, sympathetic ophthalmia, Takayasu's arteritis, temporal arteritis, giant cell arteritis, thrombocytopenic purpura, Tolosa-Hunt syndrome, transverse myelitis, type 1 diabetes, ulcerative colitis, undifferentiated connective tissue disease, uveitis, vasculitis, vitiligo, or Vogt-Koyanagi-Harada disease.

[0229] In some embodiments, the autoimmune disease is selected from the group consisting of inflammatory bowel disease, Crohn's disease, ulcerative colitis, colitis, celiac disease, rheumatoid arthritis, Behcet's disease, amyloidosis, psoriasis, psoriatic arthritis, systemic lupus erythematosus nephritis, graft-versus-host disease (GVHD), nonalcoholic steatohepatitis (NASH), and ankylosing spondylitis. In a preferred embodiment, the disease is Crohn's disease.

[0230] In some embodiments, the treatment comprises administering a GAL9-binding molecule as described herein to a subject at risk of transplant rejection in an amount effective to reduce transplant rejection. In some embodiments, the treatment comprises administering a GAL9-binding molecule as described herein to a subject with graft-versus-host disease in an amount effective to reduce GvHD. In some embodiments, the treatment comprises administering a GAL9-binding molecule as described herein to a subject with a post-traumatic immune response in an amount effective to reduce inflammation. In some embodiments, the treatment comprises administering a GAL9-binding molecule as described herein to a subject with ischemia in an amount effective to treat the subject. In some embodiments, the treatment comprises administering a GAL9-binding molecule as described herein to a subject who has suffered a stroke in an amount effective to treat the subject.

[0231] In some embodiments, the treatment comprises administering a GAL9-binding molecule to a subject having a viral infection in an amount effective to reduce acute respiratory distress syndrome and / or acute cytokine release syndrome (cytokine storm). In certain embodiments, the viral infection is infection with the SARS-CoV-2 virus and the disease is COVID-19.

[0232] Administration The GAL9-binding molecule can be administered to a subject by any route known in the art. For example, the GAL9-binding molecule can be administered to a human subject by, for example, intra-arterial, intramuscular, intradermal, intravenous, intraperitoneal, intranasal, parenteral, pulmonary, subcutaneous, topical, oral, sublingual, intratumoral, peritumoral, intralesional, intrasynovial, intrathecal, intracerebrospinal, or perilesional administration. The GAL9-binding molecule can be administered to a subject either by itself or as a pharmaceutical composition. Exemplary pharmaceutical compositions are described herein.

[0233] The anti-GAL9 binding molecules disclosed herein can be administered alone or in combination with other therapies or procedures to treat or prevent a disease or condition.

[0234] Depending on the condition or disease being treated, treatment with a GAL9-binding molecule can improve one or more endpoints in a subject. Examples of clinical endpoints that may be improved in a subject with a disease or condition include, but are not limited to, reduced inflammation, reduced autoimmune response, prolonged remission, induction of remission, re-establishment of immune tolerance, improved organ function, reduced risk of progression or development of a disease or condition, reduced risk of progression or development of a secondary disease, increased overall survival of the subject, or a combination thereof. [Example]

[0235] 6.11 Working Example The following examples are provided by way of illustration and not limitation, in particular the methods for expressing and purifying various antigen binding proteins and their use in the various assays described below are non-limiting and exemplary.

[0236] 6.11.1. Method 6.11.1.1. Expi293 Expression The various antigen-binding proteins tested were expressed using the Expi293 transient transfection system according to the manufacturer's instructions. Briefly, unless otherwise noted, plasmids encoding individual chains were mixed at a 1:1 mass ratio and transfected into Expi293 cells using the ExpiFectamine 293 transfection kit. Cells were cultured at 37°C, 8% CO2, and 100% humidity with shaking at 125 rpm. Transfected cells were fed once 16–18 hours after transfection. On day 5, cells were harvested by centrifugation at 2000 g for 10 minutes. The supernatant was collected for affinity chromatography purification.

[0237] 6.11.1.2. ExpiCHO Expression The various GAL9 antigen-binding proteins are expressed using the ExpiCHO transient transfection system according to the manufacturer's instructions. Briefly, the plasmids encoding the individual chains are mixed, for example, in a 1:1 mass ratio, and transfected into ExpiCHO using the ExpiFectamine CHO transfection kit.

[0238] Cells are grown at 37°C, 8% CO2, and 100% humidity with shaking at 125 rpm. Transfected cells are typically fed once 16-18 hours after transfection. Cells are harvested on day 5 by centrifugation at 2000g for 10 minutes. The supernatant is then collected for affinity chromatography purification.

[0239] Protein A Purification Clarified supernatants containing various antigen-binding proteins were separated using either Protein A (ProtA) resin or anti-CH1 resin in a Gravity flow purifier. In examples where direct comparisons were performed, the supernatants containing various antigen-binding proteins were split into two equal samples. For ProtA purification, a 1 mL Protein A column (GE Healthcare) was equilibrated with PBS (5 mM sodium potassium phosphate, pH 7.4, 150 mM sodium chloride). The sample was loaded onto the column at 5 mL / min. The sample was eluted using 0.1 M sodium acetate, pH 3.5. The eluate was monitored by absorbance at 280 nm, and the elution peaks were pooled for analysis. The eluate was monitored by absorbance at 280 nm, and the elution peaks were pooled for analysis.

[0240] 6.11.1.4. SDS-Page analysis Samples containing various isolated antigen-binding proteins were analyzed by reducing and non-reducing SDS-PAGE for the presence of complete and incomplete products and overall purity. 2 μg of each sample was added to 15 μL of SDS loading buffer. Reduced samples were incubated at 75°C for 10 minutes in the presence of 10 mM reducing agent. Non-reduced samples were incubated at 70°C for 5 minutes without reducing agent. Reduced and non-reduced samples were loaded onto a 4-15% gradient TGX gel (BioRad) with running buffer and run at 220 volts for 30 minutes. Upon completion of the run, the gel was washed with DI water and stained using GelCode Blue Safe Protein Stain (ThermoFisher). Prior to analysis, the gel was destained with DI water. Densitometry analysis of scanned images of the destained gel was performed using standard image analysis software to calculate the relative abundance of bands in each sample.

[0241] IEX Chromatography Samples containing various isolated antigen-binding proteins were analyzed for the ratio of complete product to incomplete product and impurities by cation exchange chromatography. The clarified supernatant was analyzed on an AKTA Purifier FPLC using 5 ml of MonoS (GE Lifesciences). The MonoS column was equilibrated with Buffer A, 10 mM MES pH 6.0. The sample was loaded onto the column at 2 ml / min. The sample was eluted over 6 CV using a 0-30% gradient of Buffer B (10 mM MES pH 6.0, 1 M sodium chloride). The eluate was monitored by absorbance at 280 nm, and the purity of the sample was calculated by peak integration to identify the abundance of the monomeric and contaminant peaks. The monomeric and contaminant peaks were pooled separately for analysis by SDS-PAGE as described above.

[0242] For analytical SEC chromatography, 1 mg / mL of each sample was loaded onto the column at 1 ml / min. Samples were eluted over 1.5 CV using an isocratic flow of PBS. The eluate was monitored by absorbance at 280 nm, and the eluted peaks were analyzed by peak integration.

[0243] 6.11.1.6. Mass spectrometry Samples containing various isolated antigen-binding proteins were analyzed by mass spectrometry to confirm the correct species by molecular weight. All analyses were performed by a third-party laboratory. Briefly, samples were treated with a cocktail of enzymes to remove glycosylation. To specifically identify each chain by molecular weight, both samples were run in a reducing format. To identify the molecular weight of all conjugates in the samples, all samples were run under non-reducing conditions. Mass spectrometry was used to identify the number of unique products based on molecular weight.

[0244] 6.11.1.7. Antibody Discovery by Phage Display Phage display of the human Fab library was performed using standard protocols. Human GAL9 protein was purchased from Aero Biosystems (human GAL9 His tag, catalog number LG9-H5244) and transfected with EZ-Link NHS-PEG using standard protocols. 12 The phage clones were biotinylated using -Biotin (ThermoScientific catalog number 21312). Phage clones were screened for their ability to bind to GAL9 protein by phage ELISA using standard protocols.

[0245] Briefly, a phage library in Fab format was constructed using an expression vector (also called a phagemid) capable of replication and expression in phage. Both heavy and light chains were encoded on the same expression vector, with the heavy chain fused to a truncated variant of the phage coat protein pIII. The light and heavy chain-pIII fusions were expressed as separate polypeptides and assembled in the bacterial periplasm, where redox potential allows disulfide bond formation, to form phage-displayed antibodies containing candidate Abs.

[0246] Libraries were generated using sequences derived from a specific human heavy chain variable domain (VH3-23) and a specific human light chain variable domain (Vκ-1). The screening library was diversified so that all three CDRs of the VH domain matched the positional amino acid frequencies of the CDRs along their lengths found in the human antibody repertoire. The generated light chain variable domains in the screening library were diversified only in the VL CDR3 (L3). The light chain VL CDR1 (L1) and CDR2 (L2) retained their human germline sequences.

[0247] The heavy chain scaffold (SEQ ID NO:2), light chain scaffold (SEQ ID NO:4), complete heavy chain Fab polypeptide (SEQ ID NO:1), and complete light chain Fab polypeptide (SEQ ID NO:3) used in the phage display library are shown below, where lowercase "x" represents the CDR amino acids that were altered to generate the library.

[0248] Phage display VH scaffold [SEQ ID NO: 2]: [ka]

[0249] Phage display VL scaffold [SEQ ID NO: 4]: [ka]

[0250] Phage-displayed heavy chain Fab polypeptide [SEQ ID NO: 1]: [ka]

[0251] Phage-displayed light chain Fab polypeptide [SEQ ID NO: 3]: [ka]

[0252] Diversity was generated by Kunkel mutagenesis using primers that introduced diversity into the VH CDR1 (H1), CDR2 (H2), CDR3 (H3), and VL CDR3, mimicking the diversity found in natural antibody repertoires, as described in more detail in Kunkel, TA (PNAS 1985, January 1, Vol. 82 (No. 2), pp. 488-492), which is incorporated herein by reference in its entirety. Briefly, single-stranded DNA was prepared from isolated phage using standard procedures, and Kunkel mutagenesis was performed. Chemically synthesized DNA was then electroporated into MC1061F-cells. The resulting phagemids, grown overnight, were digested with restriction enzymes (BamHI and XbaI) to remove wild-type sequences. The digested samples were electroporated into TG1 cells and subsequently harvested. The recovered cells were subcultured and infected with M13K07 helper phage to produce a phage library.

[0253] Phage panning was performed using standard procedures. Briefly, the first round of phage panning was performed using targets immobilized on streptavidin magnetic beads, followed by panning of approximately 5 × 10 phage from the prepared library in a 1 mL volume in PBST-2% BSA. 12 After 1 hour of incubation, the bead-bound phages were separated from the supernatant using a magnetic stand. The beads were washed three times to remove non-specifically bound phages, and then the OD 600 After 20 minutes, the infected cells were added to 5 mL of ER2738 cells with a pH of approximately 0.6. After 20 minutes, the infected cells were diluted with 25 mL of 2xYT. + Ampicillin and M13K07 helper phage (final concentration, approximately 10 10The cells were subcultured in 1000 μg / ml of phage culture medium and allowed to grow overnight at 37°C with vigorous shaking. The following day, phage were prepared by PEG precipitation using standard procedures. Prior to panning, specific phage were pre-depleted on SAV-coated beads. The second round of panning was performed using standard procedures with 100 nM bead-immobilized antigen using a KingFisher magnetic bead handler. In total, three to four rounds of phage panning were performed to enrich for phages displaying Fabs specific to the target antigen. Target-specific enrichment was confirmed using polyclonal and monoclonal phage ELISA. DNA sequencing was used to determine the isolated Fab clones containing candidate Fabs.

[0254] The VL and VH domains identified in the phage screen described above were reformatted into a bivalent, monospecific, native human full-length IgG1 architecture.

[0255] Native human full length IgG1 heavy chain architecture [SEQ ID NO: 5]: [ka]

[0256] Native human full-length IgG1 light chain architecture: Corresponds to the phage-displayed light chain Fab. See SEQ ID NO: 3.

[0257] 6.11.1.8. Octet Determination of Binding Kinetics To measure qualitative binding affinity in GAL9 binder discovery experiments, IgG1 reformatted conjugates were immobilized on the biosensor of an Octet (Pall ForteBio) Biolayer Interferometer.

[0258] Soluble GAF9 antigen was then added to the system and binding was measured. The weakest ( + ) to the strongest ( +++) qualitative binding affinity was assessed. A slow off-rate was indicated by a negligible drop in the dissociation phase of the sensogram, indicating tight antibody binding ( +++ To obtain accurate rate constants for monovalent affinity, at least five concentrations (approximately 10–20 × K) were used in the binding step. D to 0.1 x K D A dilution series containing the GAF9 analyte was measured over a range of values ​​(2-fold dilutions). During the dissociation phase, the sensor was immersed in a buffer solution without the GAF9 analyte, and the complex bound to the sensor surface in the buffer dissociated. Octet kinetic analysis software was used to calculate the rate and equilibrium binding constants based on the binding and dissociation curves. The analysis was performed globally (global fitting), and rate constants were derived simultaneously from all analyte concentrations included in the experiment.

[0259] 6.11.1.9. Epitope Binning Anti-GAL9 candidates formatted as bivalent, monospecific, native, full-length human IgG1s as described above were tested for GAL9 binding in a pairwise fashion using an octet-based "tandem" assay. Briefly, biotinylated GAL9 was immobilized on a streptavidin sensor, and two anti-GAL9 candidates were bound in tandem. Competitive blocking profiles were generated to determine whether a given anti-GAL9 candidate blocked the binding of a panel of other anti-GAL9 candidates to GAL9. Anti-GAL9 candidates that competed for the same or non-overlapping binding regions were grouped together and considered to belong to the same bin.

[0260] 6.11.1.10. PBMC Activation and Galectin-9 Antibody Treatment Individual aliquots of PepMix HCMVA(pp65) (>90%) Protein ID: P06725 (Cat. No. PM-PP65-2, JPT Peptide Technologies) were prepared according to the manufacturer's instructions. PepMix™ HCMVA(pp65) is a mixture of overlapping 15-mer peptides covering the entire protein, the 65 kDa phosphoprotein (pp65) (Swiss-Prot ID: P06725) of human cytomegalovirus (HHV-5), used for immunostimulation of immune cell responses.

[0261] Frozen human peripheral blood mononuclear cells (PBMCs) were thawed according to standard conditions and then resuspended in growth medium (10% FBS in RPMI).

[0262] Resuspend 5 x 10 PBMCs 5 Cells were seeded into 96-well plates and incubated with 2 μg / mL PepMix™ HCMVA (pp65) and 40 μg / mL candidate GAL9 antibody or control antibody in growth medium at 37° C., 5% CO for 24 hours.

[0263] 6.11.1.11. LEGENDplex Human Th Cytokine Assay Following PBMC activation and galectin-9 antibody treatment as described herein, cytokine secretion by PBMCs and immune cell subpopulations was assessed 24 and 72 hours after treatment with a cytokine bead array as follows.

[0264] 200 μl of cell culture supernatant was collected and centrifuged to pellet cell debris. The resulting supernatant was analyzed using the LEGENDplex™ Human Thl Panel (5-plex) (Cat. No. 740009, Biolegend). The LEGENDplex™ Human Thl Panel is a bead-based assay that allows for the simultaneous quantification of human cytokines IL-2, IL-6, IL-10, IFN-γ, and TNF-α using flow cytometry.

[0265] Briefly, cytokine standards and capture bead mixtures were prepared according to the manufacturer's instructions. An assay master mix of 1:1:1 capture bead mixture:biotinylated detection antibody; assay buffer was prepared.

[0266] 12.5 μl of supernatant sample or cytokine standard was incubated with 37.5 μl of assay master mix. The plate was sealed, covered with foil, and shaken at 600 rpm for 2 hours at room temperature. The wells were then incubated with streptavidin-phycoerythrin (SA-PE) for 30 minutes at room temperature with shaking at 600 rpm. The beads were then washed twice and resuspended before proceeding to flow cytometry analysis according to the manufacturer's instructions.

[0267] 6.11.1.12. PBMC Staining with Marker Antibodies Following PBMC activation and galectin-9 antibody treatment as described herein, PBMC immune cells were stained with marker antibodies according to the following procedure.

[0268] 5 x 10 cells 6 Cells were resuspended in growth medium (RPMI with 10% FBS) at 1000 cells / mL. 200 μL of resuspended cells were aliquoted into a 96-well plate and then incubated with Fixable Viability Dye eFluor® 780 for 30 minutes at 2-8°C to irreversibly label dead cells. Cells were then washed and then incubated with Human Fc Blocking Solution (Cat. No. 14-9161-73, eBiosciences) for 10 minutes at room temperature.

[0269] The antibody cocktail working solution was prepared according to the table below.

[0270] [Table 1]

[0271] Wells were incubated with 10 μL of diluted antibody cocktail for 30 min at 2–8° C. Cells were then washed, resuspended, and analyzed by flow cytometry analysis.

[0272] To analyze the immune stimulatory markers CD27, CD40L, ICOS, 4-1BB, and OX40, the same protocol provided above was followed, but cells were incubated with alternative antibody cocktails as shown in Table 2 below.

[0273] [Table 2]

[0274] Example 1 6.11.2. Example 1: Blood dendritic cells from Crohn's disease patients show increased PD-L2 expression Programmed death 1 (PD-1)-deficient mice develop a variety of autoimmune-like diseases, suggesting that the PD-1 receptor plays an important role in immunity and autoimmunity. PD-1 has two endogenous ligands, PD-L1 and PD-L2. PD-1 / PD-L1 interaction has been suggested to be involved in autoimmunity, but the role of PD-L2 in autoimmunity is poorly understood.

[0275] Crohn's disease (CD) is a chronic inflammatory disorder of the gastrointestinal tract. Although the specific cause of this disease is unclear, it is clear that an overactive immune system in CD patients leads to inflammation and damage in the gastrointestinal tract. This study was conducted to determine the expression of PD-L2 and PD-L1 on blood dendritic cells in patients with Crohn's disease.

[0276] research participants Peripheral blood samples were collected from 29 adults with colonoscopy-confirmed Crohn's disease. Patients were included at different stages of treatment, but those who had received anti-TNF-α therapy were excluded. Control samples were collected from 13 healthy adults who underwent colorectal cancer family history screening.

[0277] immunostaining Single-cell suspensions obtained from 10 ml of whole blood were incubated with Fc receptor-binding antibodies to block nonspecific Fc binding by specific antibodies. Fixable Viability Dye eFluor780 (ebioscience, San Diego, CA) was used to exclude dead cells from the analysis. Cells were assessed using the following anti-human monoclonal antibodies: HLA-DR PerCP-Cy5.5 (clone G46-6; BD Bioscience, San Jose, CA); lineage cocktail BV510 [CD3 (clone OKT3) / CD14 (clone M5E2) / CD16 (clone 3G8) / CD19 (clone HIB19) / CD20 (clone 2H7) and CD56 (clone HCD56)]; and CD11c BV605 (clone 3.9; BioLegend, San Diego, CA).

[0278] Anti-human PD-L2 monoclonal antibody (clone MIH18; BioLegend, San Diego, CA) and anti-human PD-L1 monoclonal antibody (clone 29E.2A3; BioLegend, San Diego, CA) or control IgG were labeled in-house using Lightning-Link Rapid DyLight 647 and Lightning-Link Rapid DyLight 488 (BioNovus Life Sciences, Cherry Brook, NSW, Australia). Cells were stained with anti-HLA-DR, anti-PD-L2, or anti-PD-L1, or IgG control, for 30 minutes at room temperature, then washed twice with PBS for 5 minutes before being fixed with 1% paraformaldehyde in PBS, pH 7.25.

[0279] Flow cytometry Cells were stained with Fixable Viability Dyes (FVD) and gated to capture only viable cells in the mononuclear cell region of a side scatter vs. forward scatter plot. Dendritic cells were identified as HLA-DR + and Lin +and subsequently CD11c in the total peripheral blood population. + For each donor, at least 1 × 10 4 Events were collected.

[0280] Cells were analyzed using a BD LSR Fortessa flow cytometer, and data were analyzed using BD FACSDiva software (Becton & Dickinson, Franklin Lakes, NJ), FCS express (De Novo software, Glendale, CA), or FlowJo software (Tree Star; a subsidiary of Becton, Dickinson and Company, Ashland, OR).

[0281] statistical analysis A two-tailed nonparametric Mann-Whitney U test was performed using GraphPad Prism (GraphPad Software).

[0282] Microscopic observation Microscopic specimens were prepared by mounting stained and sorted cells onto glass slides, and images were collected using a confocal microscope.

[0283] Results / Conclusion Figure 2 shows CD11c from a patient with Crohn's disease stained with either IgG control, anti-PD-L1, or anti-PD-L2. + Contour plots of dendritic cells (DC) cells are shown. We found that the anti-PD-L1 antibody bound 28.6% of DC cells to PD-L1, whereas the IgG control showed 2.23% non-specific binding to DC cells. + Similarly, in a second experiment, the IgG control was observed to stain as CD11c. + The anti-PD-L2 antibody bound to only 3.22% of DCs, whereas the anti-PD-L2 antibody bound to 62.7% of DCs. + was detected as.

[0284] Figures 3A-3B show CD11c from healthy control donors and CD patients. + PD-L1 in blood dendritic cells + The percentage of cells (Figure 3A) and CD11c + PD-L2 in blood dendritic cells + Scatter plots of the percentage of CD11c cells (Figure 3B) are shown. The horizontal bars of the scatter plots indicate the mean. Figures 3C-3D show the CD11c cells derived from healthy control donors and cloned patients. + Scatter plots of PD-L1 expression (GMI) (Figure 3C) and PD-L2 expression (GMI) on blood dendritic cells (Figure 3D) are shown. Horizontal bars in the scatter plots represent means. A single asterisk (*) indicates a P value of 0.0292. A double asterisk (**) indicates a P value of 0.0032.

[0285] Figures 4A-4B show representative immunostaining of dendritic cells (DC) derived from the blood of two healthy control donors and three Crohn's disease patients. DCs derived from healthy controls exhibit high PD-L1 (green) and PD-L2 (red) staining throughout the cell. These figures are rendered in grayscale. In contrast, DCs from Crohn's disease patients exhibit low PD-L1 expression and high levels of PD-L2, appearing to be aggregated. In some cells, we observed high staining for aggregated PD-L1.

[0286] The results show that PD-L2 protein is more highly expressed on blood dendritic cells derived from Crohn's patients compared to healthy control donors (P=0.0032), with a higher statistical difference than PD-L1 (P=0.0292). These results suggest that the PD-L2 pathway may play an important role in Crohn's disease and other autoimmune diseases.

[0287] Example 2 6.11.3. Example 2: Inhibition of PD-L2 in PBMCs from Crohn's Disease Patients Results in a Clinically Favorable Cytokine Profile This study was conducted to determine the inhibitory effect of PD-L2 protein on cytokine profiles in PBMCs derived from Crohn's disease (CD) patients compared to an IgG control.

[0288] research participants Blood samples were obtained from 14 different Crohn's disease patients. Peripheral blood mononuclear cells (PBMCs) were isolated by density centrifugation on Ficoll-Paque (Pharmacia, Freiburg, Germany) using heparinized blood. Isolated PBMCs from control and CD patients were plated onto anti-CD3 pre-coated wells (2 × 10 cells) containing R10 medium supplemented with penicillin (100 IU / ml), streptomycin (0.1 mg / ml), and L-glutamine (0.29 g / l). 5 Control IgG or blocking anti-PD-L2 (MIH18) antibody was added to the cultures at 20 μg / ml.

[0289] process Matched PBMC samples were treated with IgG control or anti-human PD-L2 antibody clone MIH18 (BioLegend) for 36 hours and then assayed.

[0290] Cytokine assay Concentrations of TNF-α, IFN-γ, and IL-10 were measured using a BD™ Cytometric Bead Array (CBA) according to the manufacturer's instructions.

[0291] statistical analysis A Wilcoxon matched-pairs signed rank test was performed using GraphPad Prism (GraphPad Software).

[0292] Results / Conclusion The mean concentrations of TNF-α and IFN-γ in matched samples are shown in Figures 5A-5B, respectively. Figure 5C shows the mean IL-10:TNF-α ratio. These results indicate that PD-L2 inhibition results in a clinically favorable cytokine profile in PMBCs derived from CD patients by reducing the levels of the pro-inflammatory cytokines TNF-α and IFN-γ and increasing the levels of the inhibitory cytokine IL-10.

[0293] Example 3 6.11.4. Example 3: Stimulation or Blockade of the GAL9 / PD-L2 Pathway Inhibits Murine CD4 + Regulates TNF-α secretion in T cells Previously, we have shown that GAL9 can bind to soluble PD-L2 and that some of the immunological effects of PD-L2 are mediated by the binding of multimeric PD-L2 to GAL9, rather than by PD-1 / PD-L1 (WO 2016 / 008005, which is incorporated herein by reference in its entirety). This study demonstrates that stimulation or blockade of the GAL9 / PD-L2 pathway significantly reduces the expression of murine CD4 + This was carried out to determine whether TNF-α secretion from T cells could be regulated.

[0294] animal C57BL6 / J mice were used in the study. All animals used in the study were housed and maintained in accordance with the National Health Medical Research Council (NHMRC) animal use guidelines.

[0295] sPD-L2 Soluble murine PD-L2 (sPD-L2) with human IgG1 Fc was custom-produced by Geneart (Germany).

[0296] antibody Treatments were performed with the inhibitory anti-mouse GAL9 antibody clone 108A2 (BioLegend®, San Diego, CA) or a rat IgG2a control antibody. The anti-mouse GAL9 clone (108A2) binds to the linker peptide of mouse galectin-9 (Oomizu, S. et al., PLoS One 7(11):e48574 (2012); doi: 10.1371 / journal.pone.0048574, incorporated herein by reference). Anti-CD3 (clone 145.201) (Aviva Systems Biology Corp., San Diego, CA) was used for stimulation.

[0297] CD4 + Cell isolation and stimulation of T cells Mouse splenocyte suspensions were prepared from five mice. + A Miltenyi Biotec Inc. (Auburn, CA) kit for T cells was used to measure CD4 + T cells were isolated. Mouse CD4 + T cells were stimulated with 5 μg / ml of anti-CD3 clone 145.2C11 (Aviva Systems Biology Corp., San Diego, CA). + T cells were treated with IgG control or sPD-L2 (20μg / ml), or sPD-L2 and anti-GAL9 mAb clone 108A2, both at 20μg / ml, and cultured for 36 hours.

[0298] Cytokine assay Thirty-six hours after treatment, TNF-α concentrations were measured using a BD™ Cytometric Bead Array according to the manufacturer's instructions.

[0299] statistical analysis Nonparametric Mann-Whitney U tests were performed using GraphPad Prism (GraphPad Software).

[0300] Results / Conclusion Figure 6 shows a bar graph of TNF-α concentration levels for each treatment group. + Treatment of T cells with sPD-L2 alone increased CD4 + The addition of an inhibitory anti-mouse GAL9 antibody (108A2) resulted in a significant increase in TNF-α secretion by T cells. *p-value <0.0001. The addition of an inhibitory anti-mouse GAL9 antibody (108A2) resulted in a significant increase in TNF-α secretion by T cells. + Activated CD4 + Significantly reduced TNF-α secretion from T cells. *p-value<0.0001.

[0301] sPD-L2 binding to GAL9 on T cells induces TNF-α secretion, whereas GAL9 inhibition inhibits CD4 + sPD-L2-mediated TNF-α secretion in T cells is blocked. These results suggest that the GAL9 / PD-L2 pathway mediates the activation of stimulated CD4 + It has been shown to regulate TNF-α levels in T cells.

[0302] Example 4 6.11.5. Example 4: Inhibitory anti-mouse GAL9 (108A2) antibodies are expressed in CD4+ cells derived from malaria-infected mice + It acts independently of PD-1 / PD-L1 in T cells, but not activating anti-GAL9 antibodies. This study was conducted to investigate the dependence of inhibitory and activating GAL9 antibodies on the PD-1 / PD-L1 pathway.

[0303] Malaria-infected mouse models can be used to study immune mechanisms and drug susceptibility. Wykes, MN et al., Eur J Immunol. (2009) 39:2004-7, which is incorporated herein by reference in its entirety. Furthermore, it has been shown that the malaria-causing Plasmodium parasite can utilize the PD-1 pathway to "inactivate" T cell function. The crucial role of PD-1 in malaria pathogenesis was demonstrated when PD-1-deficient mice were shown to rapidly and completely clear P. chabaudi infection. Therefore, malaria infection models can be used to understand the relative contributions of PD-1 and its ligands, PD-L1 and PD-L2, in immunity.

[0304] antibody In this study, an inhibitory anti-mouse GAL9 antibody (108A2) and an activating anti-mouse GAL9 antibody (RG9.1) (catalog no. BE0218, InVivoMab antibodies) were used.

[0305] Malaria-infected mouse model Cohorts of C57BL / 6 mice were infected with non-lethal malaria (P. yoelii 17XNL). 5 After intravenous injection of P. yoelii infected erythrocytes, mice were incubated for 7 days to allow infection to occur.

[0306] CD4 + Isolation and processing of T cells Miltenyi Biotec's untreated CD4 + Using a T cell isolation kit, CD4 + T cells were isolated from malaria-infected mice and then cultured and treated overnight with either a control IgG antibody, an inhibitory anti-mouse GAL9 antibody (108A2), or an activating anti-mouse GAL9 antibody (RG9.1).

[0307] Immunostaining and microscopy After treatment, cells were stained with DAPI (to detect DNA) and anti-OX40 (CD134), anti-PD-1, and anti-PD-L1 (BioXCell, Lebanon, NH) antibodies labeled using the Lightning-Link Rapid DyLight 647, 594, or 488 kits. Immunostaining was observed by confocal imaging.

[0308] Results / Conclusion Figure 7 shows the CD4+ cells treated with either an IgG control, an inhibitory anti-mouse GAL9 antibody (108A2), or an activating anti-mouse GAL9 antibody (RG9.1). + Representative confocal images of T cells are shown. Red staining indicates PD-1 receptor, green staining indicates PD-L1 ligand, yellow staining indicates OX40 receptor, and blue staining indicates DNA (DAPI). Rendered in grayscale in the accompanying figures.

[0309] We observed that treatment with activating anti-mouse GAL9 (RG9.1) antibody reduced the expression of PD-1 receptor (low level of staining) and PD-L1 ligand (very reduced level of staining). In contrast, we observed that treatment with inhibitory anti-GAL9 (108A2) had no effect on the expression of PD-1 receptor (staining level similar to IgG control level) or PD-L1 ligand (staining level similar to IgG control level). In addition, we observed that treatment with inhibitory anti-GAL9 (108A2) resulted in a decrease in the expression of OX40. These results suggest that GAL9 antibody inhibition reduces the expression of CD4 + This suggests that it acts independently of the PD-1 / PD-L1 pathway in T cells.

[0310] Example 5 6.11.6. Example 5: Treatment with Inhibitory Anti-Mouse GAL9 (108A2) Reduces CD4 Cells from Malaria-Infected Mice + and CD8 + Reduces PD-L2-mediated survival of T cells. This study used CD4 + and CD8+ This was carried out to determine the effect of an inhibitory anti-mouse GAL9 (108A2) antibody on PD-L2-mediated survival of T cells.

[0311] PD-L2 binds to parasite-specific CD4 + and CD8 + CD4 T cell proliferation in malaria-infected mice by increasing the number of T cells and protecting mice from lethal malaria infection. + and CD8 + It has been shown to mediate T cell survival. See Karunarathne et al., Immunity (2016), August 16;45(2):333-45, which is incorporated herein by reference in its entirety.

[0312] Malaria-infected mouse model Cohorts of five C57BL / 6 mice were infected with non-lethal malaria (P. yoelii 17XNL). 5 After intravenous injection of P. yoelii-infected red blood cells, mice were incubated for 7 days to allow infection to occur. All animals used in the study were housed and maintained in accordance with the National Health and Medical Research Council (NHMRC) animal use guidelines.

[0313] sPD-L2 As a positive control, CD4 + and CD8 + T cells were treated with soluble PD-L2, "sPD-L2," custom-made by Geneart (Germany).

[0314] Cell isolation, treatment, and viability assay Untreated CD4 + and CD8 + CD4 T cells were detected by FACS using a Miltenyi Biotec Inc. (Auburn, CA) kit. + and CD8 + T cells were isolated from infected mice and then cultured at 37°C for 36 hours. + and CD8+ T cells were treated with 20mg / ml sPD-L2 or 20mg / ml anti-mouse GAL9 (108A2). After treatment, cell viability was assayed using a viability dye and flow cytometry.

[0315] Results / Conclusion CD4 + T cells and CD8 + The results of the T cell viability assay are shown in Figure 8A and Figure 8B, respectively. Treatment with sPD-L2 significantly reduced the CD4 + and CD8 + In contrast, treatment with sPD-L2 and anti-GAL9 (108A2) increased the PD-L2-mediated survival of CD4 T cells. + and CD8 + These results suggest that PD-L2, acting together with GAL9, reduces PD-L2-mediated survival in both CD4 T cells and CD4 T cells. + and CD8 + This suggests that it mediates T cell survival.

[0316] Example 6 6.11.7. Example 6: Blockade of the GAL9 / PD-L2 Pathway Reduces Activated CD4 T Cells from Malaria-Infected Mice + Decreased pro-inflammatory cytokines in T cells This study demonstrated that blockade of the GAL9 / PD-L2 pathway with either a blocking anti-PD-L2 antibody or an inhibitory anti-mouse GAL9 (108A2) antibody significantly reduced activated CD4 T cells derived from malaria-infected mice. + This was carried out to determine whether pro-inflammatory cytokine secretion by T cells could be reduced.

[0317] Malaria-infected mouse model Cohorts of five C57BL / 6 mice were infected with the malaria strain P. yoelii 17XNL and incubated for 7 days to allow infection to occur. All animals used in the study were housed and maintained in accordance with NHMRC animal use guidelines.

[0318] antibody Blocking anti-mouse PD-L2 mAb clone TY25 (BioXCell, Lebanon, NH) or inhibitory anti-mouse GAL9 clone 108A2 (BioLegend®, San Diego, CA) were used.

[0319] Cell isolation and co-culture stimulation CD4 + Miltenyi Biotec kit (Auburn, CA) for T cell isolation and CD11c for DC isolation + By using beads, CD4 + T cells and DC cells were isolated from malaria-infected mice. 6 2 x 10 T cells in at least triplicate wells 5 DCs and then cultured with either 20 μg / ml anti-PD-L2 mAb or 20 μg / ml anti-GAL9 mAb for 36 hours.

[0320] Cytokine assay After treatment, the concentrations of IFN-γ or TNF-α were measured using a BD™ Cytometric Bead Array (CBA) according to the manufacturer's instructions.

[0321] statistical analysis Unpaired t-tests with Welch's correction were performed using GraphPad Prism (GraphPad Software).

[0322] Results / Conclusion Figure 9A shows a bar graph of the IFN-γ concentrations detected in each treatment group. Treatment with either anti-PD-L2 or anti-GAL9 (108A2) resulted in a significant reduction in IFN-γ levels compared to untreated co-culture controls.

[0323] Figure 9B shows a bar graph of the TNF-α concentrations detected in each treatment group. Treatment with either anti-PD-L2 or inhibitory anti-mouse GAL9 (108A2) resulted in a significant reduction in TNF-α levels compared to untreated co-culture controls. An asterisk "*" indicates statistical significance at p-values ​​<0.05 compared to control. Notably, treatment with anti-PD-L2 and anti-GAL9 (108A2) reduced IFN-γ and TNF-α to similar concentration levels.

[0324] Example 7 6.11.8. Example 7: Human GAL9 (anti-human GAL9) binding arm discovery experiments Using the monoclonal phage ELISA format described above, a chemically synthesized Fab phage library with diversity introduced into the Fab CDRs was screened against the GAL9 antigen. Phage clones expressing Fabs that recognize GAL9 were sequenced.

[0325] This experiment initially identified 52 GAL9-binding candidates (antigen-binding site clones). Functional assays performed after reformatting the variable regions of these clones into a bivalent, monospecific, human IgG1 format identified 30 antibodies with immunoinhibitory properties.

[0326] Table 3 lists the VH CDR1 / 2 / 3 sequences from 30 inhibitory ABS clones, showing only the CDR residues that were changed during library construction. Table 4 lists the VL CDR1 / 2 / 3 sequences of the identified ABS clones, showing only the CDR3 residues that were changed during library construction, with the light chain CDR1 and CDR2 sequences unchanged.

[0327] [Table 3]

[0328] [Table 4]

[0329] Table 5 shows the complete CDR sequences of human candidate inhibiting anti-GAL9 antibodies according to several art-accepted definitions.

[0330] [Table 5]

[0331] [Table 6]

[0332] [Table 7]

[0333] [Table 8]

[0334] [Table 9]

[0335] [Table 10]

[0336] [Table 11]

[0337] [Table 12]

[0338] [Table 13]

[0339] Table 14

[0340] Table 15

[0341] Table 16

[0342] Table 17

[0343] Table 18

[0344] Table 19

[0345] Table 20

[0346] Table 21

[0347] Table 22

[0348] Table 23

[0349] Table 24

[0350] Table 25

[0351] Table 26

[0352] Table 27

[0353] Table 28

[0354] Table 29

[0355] Table 30

[0356] Table 31

[0357] Table 32

[0358] Table 33

[0359] Table 34

[0360] [Table 35]

[0361] [Table 36]

[0362] [Table 37]

[0363] [Table 38]

[0364] Table 6 (Tables 39-54) shows the complete immunoglobulin heavy and light chain sequences and the VH and VL sequences of various ABS candidates formatted into a bivalent monospecific human full-length IgG1 architecture.

[0365] [Table 39]

[0366] [Table 40]

[0367] [Table 41]

[0368] [Table 42]

[0369] [Table 43]

[0370] Table 44

[0371] Table 45

[0372] Table 46

[0373] Table 47

[0374] Table 48

[0375] Table 49

[0376] Table 50

[0377] Table 51

[0378] Table 52

[0379] Table 53

[0380] [Table 54]

[0381] The binding properties of selected GAL9 binding candidates were analyzed: cross-reactive binding with mouse GAL9; qualitative binding; epitope binning (bin 2 - candidate bin with LSBio's commercial antibody clone ECA8 [LS-C179448]; bin 3 - candidate bin with LSBio's commercial antibody clone ECA42 [LS-C179449], the "tool antibody" referenced in Figure 10); and monovalent affinity binding. The analytical results are shown in Table 7.

[0382] [Table 55]

[0383] Selected GAL9-binding candidates were further analyzed for sequence motifs that may adversely affect antibody properties relevant to clinical development, such as stability, variability, and immunogenicity. Computational analysis was performed according to Kumar and Singh (Developability of biotherapeutics: computational approaches. Boca Raton: CRC Press, Taylor & Francis Group, 2016). The results of the analysis are presented in Table 8 and indicate a limited number of deleterious sequence motifs present in the listed clones, indicating the potential for further clinical development.

[0384] [Table 56]

[0385] Example 8 6.11.9. Example 8: Effect of Anti-Human GAL9 Candidates on Cytokine Production in Peripheral Blood Mononuclear Cells (PBMCs) Candidate anti-human GAL9 antigen-binding sites (ABSs) were formatted into bivalent, monospecific, native, full-length human IgG1 heavy and light chain architectures (SEQ ID NO:5 and SEQ ID NO:3, respectively) and tested for their effect on cytokine production by human PBMCs following peptide stimulation. PBMCs were stimulated essentially as described in Section 6.11.1 above. Briefly, PBMCs were obtained from human donors known to respond to human CMV virus (HCMV), placed in culture, stimulated with HCMV PepMix to prime an antigen-specific response, and treated with either control IgG, a comparator antibody anti-human GAL9 tool-activating mAb (clone ECA42, murine IgG2a), α-PD1 (nivolumab), or candidate anti-GAL9 antibodies formatted as bivalent, monospecific, full-length human IgG1 antibodies. Cytokine secretion was measured 24 and 72 hours after treatment using a bead cytokine array. Results for INF-γ and TNF-α are shown graphically in Figures 10A and 10B. The data shown in Figure 10 is described in more detail in Tables 9 and 10 provided below.

[0386] [Table 57]

[0387] [Table 58]

[0388] Example 9 6.11.10. Example 9: Treatment with anti-human GAL9 IgG1 antibodies P9-11, P9-37, or P9-57 reduces TNF-α and IFN-γ production in activated PBMCs. Selected inhibitory anti-human GAL9 candidates from Example 7, formatted as bivalent monospecific human IgG1 antibodies, were further tested in PBMCs from three additional human donors for their ability to inhibit cytokine production in PBMCs.

[0389] Stimulation of PBMCs Human primary PBMCs were collected from donors 19, RCB, and RG, who are known to respond strongly to human CMV virus (HCMV). PBMCs were stimulated essentially as described in Section 6.11.1 above. Briefly, PBMCs were collected from human donors known to respond to human CMV virus (HCMV), cultured, stimulated with HCMV PepMix to prime an antigen-specific response, and treated with P9-41, P9-42, P9-53, P9-11, P9-37, or P9-57 formatted as bivalent, monospecific, full-length human IgG1 antibodies, or a human IgG control.

[0390] Cytokine assay TNF-α and IFN-γ secretion was measured 24 and 72 hours after treatment using a BD™ Cytometric Bead Array (CBA) according to the manufacturer's instructions. Assays were performed in quadruplicate.

[0391] Results / Conclusion Representative data from 72 hours of treatment are shown in Figures 11A-C. Means are shown as horizontal bars in the scatter plots. Error bars indicate standard deviation.

[0392] Figures 11A-11B show scatter plots of TNF-α levels after treatment with a human IgG control (hIgG) and inhibitory anti-human GAL9 candidates. Treatment with P9-11, P9-37, or P9-57, formatted as human IgG1 antibodies, reduced TNF-α levels in PBMCs from all three human donors compared to the IgG control. Figure 11C shows scatter plots of IFN-γ levels after treatment with a human control IgG (hIgG) or anti-human GAL9 candidates. Treatment with either P9-11, P9-37, or P9-57 reduced IFN-γ levels in PBMCs compared to the control.

[0393] Treatment with either P9-41, P9-42, or P9-53 resulted in neutral or weak TNF-α and IFN-γ secretion (data not shown).

[0394] Example 10 6.11.11. Example 10: Treatment with anti-human GAL9 P9-11, P9-24, or P9-34 reduces TNF-α and IFN-γ production and increases IL-10 production in activated PBMCs. This study was conducted to determine the effect of selected inhibitory anti-human GAL9 candidates from Example 7 on the secretion of TNF-α, IFN-γ, and IL-10 by activated human PBMC.

[0395] Stimulation of PBMCs PBMCs were stimulated essentially as described above in Section 6.11.1. Briefly, PBMCs were obtained from human donors known to be highly responsive to human CMV virus (HCMV), placed in culture, stimulated with HCMV PepMix to prime an antigen-specific response, and treated with P9-11, P9-24, and P9-34 formatted as bivalent monospecific human IgG1 antibodies, or a human IgG control.

[0396] Cytokine assay Cytokine secretion of TNF-α, IFN-γ, and IL-10 was measured 72 hours after treatment using a BD™ Cytometric Bead Array (CBA) according to the manufacturer's instructions.

[0397] Results / Conclusion Figure 12A shows a bar graph of TNF-α levels after treatment with control IgG (hIgG) or inhibitory anti-human GAL9 candidates. Treatment with anti-human IgG1 P9-11, P9-24, or P9-34 resulted in decreased TNF-α secretion from PBMCs compared to the IgG control. Figure 12B shows a bar graph of IFN-γ levels after treatment with control IgG (hIgG) or inhibitory anti-human GAL9 candidates. Treatment with anti-human IgG9 antibodies P9-11, P9-24, or P9-34 resulted in decreased IFN-γ secretion from PBMCs compared to the IgG control. Figure 12C shows a bar graph of IL-10 levels after treatment with inhibitory anti-human GAL9 candidates or the IgG control. Treatment with P9-11, P9-24, or P9-34 antibodies increased IL-10 secretion from PBMCs compared to the control.

[0398] Example 11 6.11.12. Example 11: Activated CD3 + Treatment of T cells with anti-human GAL9 antibodies P9-11, P9-24, or P9-34 improved the cytokine profile, while anti-mouse GAL9 (108A2) completely blocked cytokine secretion. We measured IFN-γ, TNF-α, or IL-10 cytokine secretion and compared activated CD3 + The effects of anti-mouse GAL9 (clone 108A2) and anti-human GAL9 antibodies P9-11, P9-24, or P9-34 formatted as human IgG1 antibodies on the cytokine profile of T cells were determined.

[0399] Animals and CD3 + Isolation of T cells Five mice were used in each treatment group. All animals used in the study were housed and maintained in accordance with the National Health and Medical Research Council (NHMRC) animal use guidelines.

[0400] antibody Antibodies P9-11, P9-24, and P9-34 formatted as bivalent, monospecific human IgG1 antibodies were used, as well as a human IgG control. In addition, the inhibitory anti-mouse GAL9 clone 108A2 "mGAL9" (BioLegend™, San Diego, CA) was used.

[0401] CD3 + T cell simulation CD3 + T cells (CD90.2 + CD3 + ) was isolated from the spleen of a naive mouse. + T cells were stimulated with 5 μg / ml of anti-CD3 clone 145.2C11 (Aviva Systems Biology Corp., San Diego, CA). + T cells were treated with 20 μg / ml of IgG control or one of the inhibitory antibodies and cultured for 72 hours.

[0402] Cytokine assay After 72 hours of treatment, concentrations of IFN-γ, TNF-α, or IL-10 were measured using a BD™ Cytometric Bead Array (CBA) according to the manufacturer's instructions.

[0403] statistical analysis Non-parametric unpaired t-tests were performed using GraphPad Prism (GraphPad Software).

[0404] Results / Conclusion The results are shown in Figures 13A and 13B. A reduction in the ratio of TNF-α:IL-10 or IFN-γ:IL:10 indicates a decrease in the pro-inflammatory cytokine and an increase in the inhibitory cytokine, IL-10. Treatment with anti-mouse GAL9 (108A2) antibody significantly reduced the secretion of TNF-α, IFN-γ, and IL-10. See Figure 13A. In contrast, treatment with any of the anti-human GAL9 antibodies P9-11, P9-24, or P9-34 (human IgG1 Fc) did not reduce TNF-α or IFN-γ secretion, but significantly increased IL-10 secretion. See Figure 13B. An asterisk "*" indicates statistical significance at p-value <0.05 compared to the control.

[0405] Treatment with anti-human P9-11 and P9-24 antibodies, formatted as human IgG1 antibodies, resulted in amelioration of the inflammatory environment, reducing TNF-α and IFN-γ secretion and increasing IL-10 secretion. Notably, treatment with anti-mouse GAL9 (108A2) resulted in complete blockade of cytokine responses, including IL-10 secretion. The differences in the cytokine profiles generated by anti-human GAL9 and anti-mouse GAL9 (108A2) suggest that the anti-human GAL9 and anti-mouse GAL9 (108A2) antibodies have different mechanisms of action.

[0406] Example 12 6.11.13. Example 12: Treatment with Anti-Human GAL9 Reduces Stimulated CD4 + and CD8 + It does not substantially alter the expression of immune checkpoint molecules in T cells, and it inhibits CD8 + Decreases 4-1BB, CD40L, and OX40 costimulatory molecules in T cells. This study investigated stimulated CD8 + and CD4 + The effects of anti-human GAL9 candidates P9-11, P9-24, and P9-34 on the expression of selected checkpoint molecules in T cells and stimulated CD8 + This was carried out to determine the effect of anti-human GAL9 P9-11 on selected costimulatory molecules in T cells.

[0407] Stimulation and Treatment CD8 + or CD4 + PBMCs containing populations of T cells were stimulated as described above and treated with anti-human GAL9 P9-11, P9-24, P9-34 formatted as bivalent monospecific human IgG1 antibodies or a human IgG control.

[0408] immunolabeling PMBC, 5 x 10 6 Cells were resuspended in RPMI with 10% FBS at 1000 cells / mL. 200 μL of resuspended cells were aliquoted into a 96-well plate and then incubated with Fixable Viability Dye eFluor® 780 for 30 minutes at 2-8°C to irreversibly label dead cells. Cells were then washed and incubated with human Fc blocking solution (catalog no. 14-9161-73, eBiosciences) for 10 minutes at room temperature. Surface expression of PD-L1, PD-1, CTLA-4, TIM3, LAG3, 4-1BB, CD27, CD40L, ICOS, or OX40 was assessed by flow cytometry.

[0409] Flow cytometry Flow cytometry analysis was performed using a BD LSR Fortessa flow cytometer and BD FACSDiva software (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). At least 5 × 10 cells were collected for each sample. 5 Events were collected.

[0410] CD4 stained positive for immune checkpoint molecules + or CD8 + Representative data for the percentage of T cells are shown in Table 11 and Table 12 below. CD8 stained positive for costimulatory molecules + The T cell percentage data are shown in Table 13 below.

[0411] "% values" represent the % of cells with detectable levels of the indicated marker. "(x)" indicates the fold change compared to the human IgG control after treatment with the selected α-GAL9 antibody candidates.

[0412] [Table 59]

[0413] [Table 60]

[0414] [Table 61]

[0415] Results / Conclusion Stimulated CD8 + or CD4 + There was no substantial change in the expression of any of the immune checkpoint molecules in T cells. However, we found that stimulated CD8 + We observed a decrease in the costimulatory molecules 4-1BB, CD40L, and OX40 in T cells. These results suggest that the effect of anti-human GAL9 candidates on cytokine responses is driven by GAL9 inhibition and not by the PD-1 / PD-L1 immune checkpoint pathway or other checkpoint molecules such as CTLA-4, TIM3, or LAG3.

[0416] 7. Equivalents While various specific embodiments have been shown and described, the above specification is not limiting. It will be understood that various modifications can be made without departing from the spirit and scope of the invention. Numerous variations will become apparent to those skilled in the art upon review of this specification.

Claims

1. A GAL9 antigen-binding molecule comprising a first antigen-binding site (ABS) specific for a first epitope of a first galectin-9 (GAL9) antigen, wherein the first antigen-binding site is selected from the group consisting of: Table 1 A GAL9 antigen-binding molecule comprising all three VH CDRs and all three VL CDRs derived from any one of the ABS clones in

2. 1. A GAL9 antigen binding molecule comprising a first ABS specific for a first epitope of a first GAL9 antigen, wherein the first antigen-binding site is: 【Chemistry 1】 The GAL9 antigen-binding molecule of claim 1, comprising:

3. The GAL9 antigen-binding molecule of claim 1 or 2, wherein the first ABS comprises a first IgG heavy chain polypeptide and a first light chain polypeptide.

4. The GAL9 antigen-binding molecule of any one of claims 1 to 3, wherein the GAL9 antigen-binding molecule further comprises a second ABS.

5. The GAL9 antigen-binding molecule of claim 4, wherein the second ABS is specific for the GAL9 antigen.

6. The GAL9 antigen-binding molecule of any one of claims 1 to 5, wherein the GAL9 antigen-binding molecule comprises an antibody format selected from the group consisting of a full-length antibody, a Fab fragment, an Fv, an scFv, a tandem scFv, a diabody, an sc diabody, a DART, a tandAb, a minibody, and a B body.

7. The GAL9 antigen-binding molecule is upon contact, reduces TNF-α secretion by activated immune cells, said reduction being at least 30%, 35%, 40%, 45%, 50%, 55%, or 60% compared to activated immune cells treated with a control agent; and / or upon contact, reduces IFN-γ secretion by activated immune cells, said reduction being at least 20%, 25%, 30%, 35%, 40%, 45%, or 50% compared to activated immune cells treated with a control agent; and / or upon contact, increases IL-10 secretion by activated immune cells, said increase being at least 5%, 10%, 15%, 20%, 25%, 30%, 35%, or 40% compared to activated immune cells treated with a control agent; and / or does not modulate PD-1 surface expression on activated immune cells compared to activated immune cells treated with a control agent; and / or does not modulate PD-L1 surface expression on activated immune cells compared to activated immune cells treated with a control agent; and / or does not modulate CTLA-4 surface expression on activated immune cells compared to activated immune cells treated with a control agent; and / or does not modulate TIM3 surface expression on activated immune cells compared to activated immune cells treated with a control agent; and / or does not modulate LAG3 surface expression on activated immune cells relative to activated immune cells treated with a control agent; and / or CD8 treated with control agents + Compared to T cells, CD8 + reduce 4-1BB surface expression on T cells, and / or CD8 treated with control agents + Compared to T cells, CD8 + reduce CD40L surface expression on T cells, and / or CD8 treated with control agents + Compared to T cells, CD8 + The GAL9 antigen-binding molecule of any one of claims 1 to 6, which reduces OX40 surface expression on T cells.

8. The GAL9 antigen-binding molecule of claim 7, wherein the control agent is a negative control agent or a positive control agent, optionally a control antibody selected from ECA42 clone anti-GAL9 antibody, RG9.1 clone anti-GAL9 antibody, RG9.35 clone anti-GAL9 antibody, anti-PD1 antibody, 108A2 clone anti-GAL9 antibody, and a non-GAL9-binding isotype control antibody.

9. A GAL9 antigen-binding molecule, wherein the GAL9 antigen-binding molecule has the following properties: A) reducing TNF-α secretion by activated immune cells, said reduction being at least 30%, 35%, 40%, 45%, 50%, 55%, or 60% compared to activated immune cells treated with a control agent; B) reducing IFN-γ secretion by activated immune cells, said reduction being at least 20%, 25%, 30%, 35%, 40%, 45%, or 50% compared to activated immune cells treated with a control agent; C) increasing IL-10 secretion by activated immune cells, said increase being at least 5%, 10%, 15%, 20%, 25%, 30%, 35%, or 40% compared to activated immune cells treated with a control agent; D) Failure to modulate PD-1 surface expression on activated immune cells compared to activated immune cells treated with a control agent; E) Failure to modulate PD-L1 surface expression on activated immune cells compared to activated immune cells treated with a control agent; F) failure to modulate CTLA-4 surface expression on activated immune cells compared to activated immune cells treated with a control agent; G) Failure to modulate TIM3 surface expression on activated immune cells compared to activated immune cells treated with a control agent; H) Failure to modulate LAG3 surface expression on activated immune cells compared to activated immune cells treated with a control agent; I) Activated CD8 treated with control agents + Compared with T cells, activated CD8 + reducing 4-1BB surface expression on T cells; J) Activated CD8 treated with control agents + Compared with T cells, activated CD8 + reducing CD40L surface expression on T cells; or K) Activated CD8 treated with control agents + Compared with T cells, activated CD8 + Reducing OX40 surface expression on T cells The GAL9 antigen-binding molecule of any one of claims 1 to 8, wherein the GAL9 antigen-binding molecule exhibits one or more of the following:

10. The GAL9 antigen-binding molecule of any one of claims 1 to 9, which is purified.

11. A pharmaceutical composition comprising the GAL9 antigen-binding molecule of any one of claims 1 to 10 and a pharmaceutically acceptable diluent.

12. 12. The pharmaceutical composition of claim 11 for use in the treatment of an autoimmune disease.

13. 13. The pharmaceutical composition of claim 12, wherein subjects with an autoimmune disease exhibit increased PD-L2 expression on dendritic cells compared to dendritic cells from healthy controls.

14. 13. The pharmaceutical composition of claim 12, wherein the autoimmune disease is selected from the group consisting of inflammatory bowel disease, Crohn's disease, ulcerative colitis, colitis, celiac disease, rheumatoid arthritis, Behcet's disease, amyloidosis, psoriasis, psoriatic arthritis, systemic lupus erythematosus nephritis, graft-versus-host disease (GvHD), non-alcoholic steatohepatitis (NASH), and ankylosing spondylitis.

Images