JPWO2021055698A5 - - Google Patents

Description

Figures 1A-C show that the extracellular domain of VISTA contains an exceptionally high frequency of histidine residues, that a large number of these histidine residues are conserved, and that at least Some indicate that they may be involved in receptor-ligand binding. Figure 1A shows a graph of immunoglobulin domain-containing proteins, with the number of extracellular domain amino acid residues of each protein plotted on the x-axis and the frequency of histidine residues within the extracellular domain of each protein plotted on the y-axis. has been done. The size of each data point corresponds to the total number of histidine residues in the extracellular domain of each protein. FIG. 1B shows the aligned amino acid sequences of the extracellular domains of human (SEQ ID NO: 165) , cynomolgus macaque (SEQ ID NO: 166), and mouse (SEQ ID NO: 167) VISTA. Signal peptide (Sig) and transmembrane domain (TMD) sequence positions are indicated. Histidine residues that are conserved across all three species are shown in bold and underlined; histidine residues that are conserved across humans and cynomolgus monkeys are only shown in bold. Figure 1C shows a model of the three-dimensional structure of the human VISTA immunoglobulin domain. Histidine residues are represented as ball and stick traces. Figures 2A-B depict a model in which histidine residues in the extracellular domain of VISTA confer counterreceptor selectivity towards acidic pH over physiological pH. Figure 2A shows the equilibrium between the absence and presence of protonation of pyrrole ammonium groups (NH) in histidine residues. The pKa of histidine in solution is 6.5, indicating that histidine residues are more likely to be protonated and therefore have a positive charge below pH 6.5 than at higher pHs. FIG. 2B shows a model in which VISTA selectively associates with P-selectin glycoprotein ligand 1 (PSGL-1) or other counterreceptors and ligands (“VISTA-R”) at acidic pH. Therefore, antibodies that bind to the extracellular domain of VISTA at acidic rather than physiological pH may be important for inhibiting or modulating VISTA activity. Figure 3 shows VISTA surface expression (anti-VISTA antibody staining) on tumor-infiltrating macrophages, dendritic cells, neutrophils, CD4+ effector T cells, CD4+ regulatory T cells, CD8+ T cells, natural killer (NK) cells, and B cells. The level of mean fluorescence intensity (MFI) is shown. VISTA is expressed on many tumor-infiltrating leukocytes, especially myeloid cells. The tumor microenvironment is often acidic, allowing VISTA to associate with counter-receptors and ligands. Figures 4A-G show that VISTA binds selectively to leukocytes and to PSGL-1 at acidic pH, with little or no binding at neutral pH, and that this binding can be blocked by anti-VISTA antibodies. show. Figure 4A on the left shows a representative histogram of fluorescently conjugated recombinant VISTA multimers binding to activated human CD4+ T cells. Dark to light gray filled histograms represent binding at pH 7.0, 6.5, 6.4, 6.3, 6.1 and 6.0. Some histograms have their corresponding pH displayed. Non-VISTA control multimer binding at pH 6.0 is shown as an unfilled histogram. On the right, the average MFI of VISTA (circles) and control (triangles) multimers binding to activated human CD4+ T cells from two donors at different pH is graphed. Figure 4B shows representative histograms of recombinant VISTA multimers binding to peripheral blood mononuclear cells (PBMC) at pH 6.0 and pH 7.4. Dark gray to light gray filled histograms represent binding to CD19+ B cells, CD4+ T cells, CD8+ T cells, CD56+ NK cells and CD14+ monocytes at pH 6.0. Unfilled, solid border and dotted border histograms represent binding at pH 7.4 to total PBMC lymphocytes and monocytes, respectively. FIG. 4C shows representative recombinant VISTA multimers binding to activated human CD4+ T cells in the presence of anti-VISTA blocking antibodies (squares) or non-VISTA-specific isotype-matched control antibodies (circles). Antibody concentration is plotted on a logarithmic scale. Nonlinear regression is also shown. Triangles represent background signal from activated human CD4+ T cells that were not stained with recombinant VISTA multimers. Figure 4D shows recombinant VISTA multimer binding at pH 6.0 to heparan sulfate-deficient Chinese hamster ovary (CHO) cells (strain pGSD-677, American Type Culture Collection) transfected to express human PSGL-1. A representative two-dimensional flow cytometry plot is shown. Multimer binding was performed in the presence and absence of anti-VISTA blocking antibodies shown in Figure 4C. Cells that remain unstained by recombinant VISTA multimers are shown as a control. PSGL-1 antibody staining is plotted on the y-axis and VISTA multimer staining is plotted on the x-axis. Figure 4E shows a representative histogram of recombinant mouse VISTA-Fc fusion protein binding to mouse splenocytes at pH 6.0 and pH 7.4. The dark gray to light gray filled histogram represents binding to CD8+ T cells, CD11b+ myeloid cells and CD4+ T cells at pH 6.0. Unfilled histograms represent binding to total splenocytes at pH 7.4. Figures 4F and G show that VISTA multimers bind to monocytes and neutrophils, respectively, more strongly at pH 6.0 than at pH 7.4. Figures 5A-D show that VISTA preferentially mediates T cell suppression and cell:cell adhesion at acidic pH, and that both effects can be reversed using anti-VISTA blocking antibodies. Figure 5A shows the relationship between 293T cells expressing hVISTA or vector control (plotted on the y-axis) and CHO cells endogenously expressing cell surface heparan sulfate on the x-axis at pH 6.0 and 7.0. Representative cells: Cell conjugate formation is shown. FIG. 5B is a graph of the frequency of cell conjugates formed at pH 6.0 between the same cells in the presence of anti-VISTA blocking antibody, anti-VISTA non-blocking antibody or isotype-matched non-VISTA-specific control antibody. Figure 5C shows Jurkat cells expressing the NFkB luciferase reporter ( A representative plot of luciferase activity produced by human T-cell line) cells is shown. Anti-VISTA blocking antibodies (squares) or isotype-matched non-VISTA-specific control antibodies (circles) were added to co-cultured cells. In FIG. 5D, the data shown in FIG. 5A is plotted as the fold increase in luciferase signal ("effect size") with anti-VISTA antibody treatment compared to control. Figures 6A-G show that VISTA can be found in intracellular endosomes, particularly Rab11+ recycling endosomes, and can be recycled to and from the cell surface by endosomal transport. Figure 6A shows the colocalization of VISTA, Rab5 (early endosome marker), Rab7 (late endosome marker) and Rab11 (recycling endosome marker) in 293T cells expressing human VISTA. Figure 6B shows colocalization of VISTA and Rab11 within human monocytes. Intracellular VISTA colocalizes with Rab11+ recycling endosomes. A non-VISTA binding control antibody of the same isotype as the VISTA antibody (“cAb”) does not detectably bind to monocytes. Figure 6C shows pH 7.4 (black), 6.7 (dark gray) and 6. (Light gray) shows the binding of three anti-VISTA antibodies to recombinant VISTA. Figure 6D shows VISTA expression upon killing by the same anti-VISTA antibodies 1 (inverted triangles), 2 (circles), 3 (squares) or a non-VISTA-specific control antibody (triangles) with a cathepsin B-sensitive linker and cytotoxic payload. Figure 2 shows the sensitivity of acute myeloid leukemia (AML) cell lines. Cell viability (CellTiter-Glo LU) is plotted on the y-axis and antibody concentration on the x-axis. Figure 6E compares hVISTA binding of anti-VISTA antibody 3 to that of an engineered variant ("VISTA mAb 3c") that shows no improved binding at acidic pH. Figure 6F shows an antibody drug conjugate assay comparing the efficacy of anti-VISTA antibodies 3 (squares) and 3c (diamonds). Figure 6G shows a schematic diagram of endosomal transport, with VISTA recycling to and from the cell surface via early endosomes and recycling endosomes. Figures 7A-F illustrate how anti-VISTA antibody variant libraries are designed and screened to obtain acidic pH-selective antibodies. Figure 7A (SEQ ID NO: 168-181, 696-737, respectively) was performed in the VH CDR3 of anti-human VISTA antibody clone P1-061029 ('029 for short) to create the '029 library for screening. The following amino acid substitutions are shown. To potentially improve binding to the histidine-rich region of VISTA at acidic pH, the library tolerated substitutions of the negatively charged amino acids aspartate and glutamic acid and pH-responsive histidine. X=H, D or E. Sequences in parentheses were removed from the synthesis to avoid introducing penalties. A total of 647 unique sequences of P1-061029 HCDR3 with 1-2 mutations were synthesized. Figure 7B shows the procedure in which the '029 library is iteratively screened and selected for acidic pH-selective antibody variants. R represents selection round. Figure 7C shows representative two-dimensional flow cytometry plot data showing the mutant pool after nine rounds of selection. VISTA binding is plotted on the y-axis and variant antibody expression is plotted on the x-axis. Binding data are shown at various antibody concentrations and pH. Figure 7D shows a diagram of P1-061029 and its progeny clones binding human VISTA at pH 6.0 and 7.4. FIG. 7E shows a diagram of the off-rate of P1-061029 and its progeny clones against human VISTA at pH 6.0. Figure 7F shows SPR binding data for antibodies P1-068761, P1-068767 and P1-061029 with human VISTA at pH 6.0 and pH 7.4. Figures 8A-F show acidic pH selective cell binding, blocking and effector activity of VISTA antibodies P1-068761 and P1-068767. Figures 8A and 8B show the average fluorescence intensity of acidic pH selective antibodies P1-068761 (Figure 8A) and P1-068767 (Figure 8B) binding to Raji cells ectopically expressing human VISTA. Cells have a pH of approximately 6.0 (circle; highest curve in Figure 8A), 6.1 (square; third highest curve), 6.2 (triangle; second highest curve), 6.4 (inverted triangle; 6.6 (diamond; 4th curve from the bottom), 7.0 (circle; 3rd curve from the bottom), 7.2 (square; 2nd curve from the bottom) curve) and 8.1 (unfilled triangle; bottom curve in Figure 8A). Binding was detected using a fluorescently conjugated anti-human IgG secondary antibody. Figure 8C shows P1-068767 (circles) and isotype-matched non-specific control antibody (triangles) binding to Raji cells ectopically expressing human VISTA at 3125 ng/mL at various pH. The "pH ₅₀ ", the pH at which 50% of P1-068767 binding is lost, is approximately 6.6. FIG. 8D shows isotype-matched non-specific control antibodies (filled and unfilled circles at pH 7.0 and 6.0, respectively), anti-VISTA mAb 2 (“control”), anti-VISTA mAb 2 (“control”) binding to human monocytes; See Figure 6C, filled and unfilled squares at pH 7.0 and 6.0, respectively), P1-068761 (filled and unfilled squares at pH 7.0 and 6.0, respectively). Shown are the mean fluorescence intensities (MFI) of P1-068767 (filled and unfilled inverted triangles at pH 7.0 and 6.0, respectively). Binding was detected with a fluorescently conjugated anti-human IgG secondary antibody. Figure 8E shows comparable blockade of recombinant VISTA multimer binding to activated human CD4+ T cells at pH 6.0 by P1-061029 (squares), P1-068761 (triangles) and P1-068767 (inverted triangles). On the other hand, a non-VISTA-specific control antibody (circle) did not block VISTA binding. Figure 8F shows the reduced efficacy of P1-068761 (triangles) and P1-068767 (inverted triangles) in mediating antibody-dependent cellular cytotoxicity (ADCC) at physiological pH. P1-061029 (squares), non-VISTA specific positive control antibody (circles) and non-VISTA specific negative control antibody (diamonds) are also shown. NK cell-specific lysis of target cells as a percentage of total target cells is plotted on the y-axis and antibody concentration is plotted on the x-axis. Nonlinear regression is also shown. Figure 9 shows enhanced pharmacokinetics (PK) of acidic pH-selective anti-VISTA antibodies in cynomolgus monkeys. The figure shows VISTA antibody 2 (“control”, circles, see Figure 6C), VISTA antibody 3 (“acidic pH sensitive”, squares, see Figure 6C) or P1-068767 (triangles) treated Serum antibody concentrations over time in cynomolgus monkeys are shown. Figures 10A and 10B (P1-061029 - SEQ ID NO: 67 positions 26-35, 50-66, 99-110) demonstrate the combined effect of mutations in acidic pH-selective anti-VISTA antibodies '761 and '767. show. FIG. 10A shows the kinetic binding data of the P1-068761 revertant at pH 7.4, pH 6.7 and pH 6.0 and the position of the revertant relative to P1-068761. FIG. 10B shows the kinetic binding data of the P1-068767 revertant at pH 7.4, pH 6.7 and pH 6.0 and the position of the revertant relative to P1-068767. Figures 11A-C show epitope binning and mapping of various anti-VISTA antibodies. FIG. 11A shows VISTA epitope competition for P1-068761 and P1-068767 compared to P1-061029 and VISTA antibody controls. Figures 11B and 11C show the representation of epitopes among all residues of the blocking hVISTA antibody (Figure 11B) as listed in Table 14 compared to the non-blocking hVISTA antibody (mAb1; Figure 11C). Amino acid residues 66 (H) and 162 (A) are shown to represent the orientation of the molecule. Histidine residues are gray and epitope residues are black. Figures 12A-C show imaged capillary isoelectric focusing (icIEF) data for the following: Figure 12A: P1-061029, Figure 12B: P1-068761 and Figure 12C: P1-068767. Isoelectric points (pI major) of the main species as well as pI markers are indicated. Figures 13A and B show alignments of the variable regions of '029 and '015 progeny clones. Figure 13A (SEQ ID NO: 67) shows an alignment of the amino acid sequences of the variable regions of '029 and its progeny clones. Figure 13B (SEQ ID NO: 95) shows an alignment of the amino acid sequences of the variable regions of '015 and its progeny clones. FIG. 14 shows an alignment of the VH sequences of P1-068761 with and without the K16R and T84A substitutions. Double underlined residues indicate positions 16 and 84 of the framework region, and shaded areas indicate CDRs. Figures 15A-O show wild type C57BL6 mice implanted with MC38 tumors, unbound isotype matched control antibody (black squares), murine VISTA blocking antibody VISTA. 10 (triangles pointing up), treated with mouse PD-1 blocking antibody (squares) or a combination of VISTA and PD-1 blocking antibodies (triangles pointing down). All antibodies were of the mouse IgG1-D265A (Fc-inactive) isotype (see Figures 15A-D). These data are representative of three independent experiments. Figures 15A-D show tumor volume over time. n=10 per group. "TF" indicates a mouse that rejected its tumor. Figures 15E and F show the frequency of intratumoral CD8+ and CD4+ T cells 7 days after the start of treatment. n=5 per group. One-way ANOVA with Dunnett's multiple comparisons, P=0.0001. Figures 15G and H show the results for the individual mice shown in Figures 15A-D. Figure 15I: VISTA knockout mice and wild-type littermates were implanted with MC38 tumors and unbound isotype-matched control antibodies (top two curves 0/7 TF and 0/5 TF, marked with circles and downward triangles). ) or with a mouse PD-1 blocking antibody (lower two curves 0/5 TF and 5/8 TF, marked with squares and downward triangles). Median tumor growth and number of tumor-free (TF) mice at the end of the study relative to the total number of mice are shown adjacent to each curve (eg, 0/7 TF). These data are representative of two independent experiments. Error bars represent interquartile range. Figures 15J-M show MC38 tumors implanted with non-binding isotype matched control antibody (Figure 15J), mouse PD-1 blocking antibody (Figure 15K), mouse PD-1 blocking antibody and non-pH selective human VISTA blocking antibody P1. -061029 (Figure 15L) or the combination of murine PD-1 blocking antibody and acidic pH-selective human VISTA blocking antibody P1-068767 (Figure 15M). show. All antibodies were of mouse IgG1-D265A isotype. Tumor volume is shown over time. n=5-8 per group. These data are representative of one independent experiment. Figure 15N shows human VISTA KI and wild-type littermates ( WT) Mouse serum antibody concentration is shown. The calculated serum mean residence times (MRT) of P1-061029 and P1-068767 in KI mice are estimated to be 4.1 and 71 hours, respectively. n=4 KI mice and 1-2 WT mice per antibody. These data are representative of a single experiment. Figure 15O shows 5mg/kg VISTA. Cynomolgus monkey serum antibody concentrations are shown after intravenous injection of P1-068767 (circles) or P1-068767 (squares). VISTA. The calculated serum mean retention times (MRT) of 4 and P1-061029 are estimated to be 7.6 hours and 717 hours, respectively. n=1 macaque per antibody. These data are representative of a single experiment. Error bars represent standard error of the mean unless otherwise indicated. Figures 16A-C show representative histograms of intratumoral CD8+ T cell expression of PD-1 (Figure 16A), LAG-3 (Figure 16B) and TIM-3 (Figure 16C) 7 days after initiation of treatment. Error bars represent standard error of the mean. Figures 17A-C show that VISTA binds PSGL-1 at acidic pH and that this interaction is consistent with VISTA antibodies P1-061029, P1-068761, P1-068767 and VISTA. 4 indicates that it is blocked. Figure 17A shows BLI binding sensorgrams of P-selectin-Fc and VISTA-Fc binding to captured PSGL1 at pH 6.0 and pH 7.4. Figure 17B shows antibodies P1-061029, P1-068761, P1-068767 and VISTA. 4 is a histogram showing that binding of PSGL-1 to hVISTA is inhibited. FIG. 17C shows VISTA. 4 (upward triangles) and by anti-PSGL-1 antibody KPL-1 (circles). These data are representative of two independent experiments. Error bars represent standard error of the mean. Figures 18A-E show P1-068767 Fab and hVISTA or (in Figure 18E) non-blocking antibody VISTA. 5 shows a representation of the co-crystal structure of 5 and hVISTA. The VISTA IgV domain is characterized by an unusual, histidine-rich extension of its central β-sheet. The VISTA IgV domain was co-crystallized with P1-068767 fragment antigen binding (Fab). The crystal structure of the VISTA+P1-068767 complex was determined at 1.6 Å resolution. Figure 18A shows the VISTA IgV domain: P1-068767 Fab co-crystal structure. Figure 18A shows the overall structure of the VISTA IgV domain in complex with P1-068767 Fab (heavy chain, dark grey; light chain, light grey). Figure 18B shows the superposition of VISTA and PD-L1 IgV domains. VISTA histidine residues are represented in stick representation. Figure 18B shows that the IgV domain of VISTA has an unusual, histidine-rich β-sheet extension. FIG. 18C shows the molecular surface of the P1-068767 epitope (light gray electrostatic surface) as revealed by the VISTA+P1-068767 crystal structure. Figure 18C shows that the blocking antibody binds to the histidine-rich β-sheet extension of VISTA. Figure 18D shows VISTA (gray ribbon schematic with epitope residues H121, H122 and H123 represented in stick representation) and P1-068767 (represented as an electrostatic surface with its residues E100 and D102 in stick representation). ) is shown. Figure 18D shows that acidic pH-selective P1-068767 associates with VISTA histidine with acidic residues. Figure 18E shows non-blocking antibody VISTA. 5 shows that 5 binds in different regions of hVISTA from P1-068767. FIG. 19 shows VISTA. as determined by MS-HDX (MS trace). 4 epitopes are shown. Figure 20 (SEQ ID NO: 738) shows VISTA in the amino acid sequence of hVISTA based on the data in Figure 19. The location of epitope 4 is shown. Residues 57-68, 86-97 and 148-165 highlighted in FIG. 19 are also represented by lighter gray text and underlined in FIG. 20. Figures 21A and 21B show the antibody VISTA. 4 (triangle), VISTA. VISTA multimer binding to activated human CD4+ T cells at pH 6.0 in the presence of 5 (squares) and non-VISTA binding (control, circles). Figure 21B shows the blocking efficiency of each antibody compared to unblocked T cells. One-way ANOVA with Dunnett's multiple comparisons, ^*** , P<0.001. These data are representative of more than four independent experiments. Error bars represent standard error of the mean. Figure 22 shows that antibodies that block VISTA binding at acidic pH are functional. The blocking antibody VISTA. 4 (square), non-blocking antibody VISTA. Effect of 5 (triangles) and non-VISTA-conjugated (control, circles) antibodies. One-way ANOVA with Dunnett's multiple comparisons, ^* , P<0.05. These data are representative of more than four independent experiments. FIG. 23 shows Jurkat T cell activation after co-culture with 293T-OKT3-VISTA cells and VISTA cells at various pH. Figure 4 shows the effect of 4 blockade (as measured by NF-kB inhibition). These data are representative of a composite of three independent experiments. Figure 24 shows the effect of pH on VISTA suppression of human CD4+ T cells. Cells were transferred to VISTA. 4 (upward triangle), VISTA. Plate-coated OKT3 and VISTA-Fc were stimulated at the indicated pH in the presence of 5 (downward triangles) or non-VISTA-binding antibodies (antibody control, squares). Also shown are cells stimulated with plate-coated OKT3 and control IgG (VISTA control, black circles) or without OKT3 (no OKT3, gray diamonds). These data are representative of one independent experiment. Figures 25A-E show that VISTA:PSGL-1 binding specificity is determined by histidine and sulfotyrosine residues. As shown in Figure 25A, human PSGL-1 19-mer Fc recombinant protein was produced in cells with or without sialyl Lewis X decoration (SLX+ and SLX-, respectively). BLI binding magnitude at pH 6.0 (white) and 7.4 (black) is shown for VISTA-Fc and P-selectin-Fc as indicated. Data are representative of a single independent experiment. As shown in Figure 25B, human PSGL-1 19-mer Fc glycopeptides produced with sialyl Lewis It was separated into fractions with sulfation (poor in sY). BLI binding magnitude at pH 6.0 (white) and 7.4 (black) is shown for VISTA-Fc and P-selectin-Fc as indicated. These data are representative of a single independent experiment. Histidine residues at positions 153-155 remain unchanged (WT VISTA) or alanine (H2A mutant), aspartate (H2D mutant) or arginine (H2R mutant), as provided in Figures 25C-25D. A human VISTA-Fc recombinant protein was produced, which was replaced by a human VISTA-Fc mutant. Figure 25C shows the BLI binding magnitude of wild type and mutant VISTA-Fc protein binding to captured PSGL-1 at pH 6.0 and 7.4. These data are representative of a single experiment. Figure 25D shows CHO-PSGL at pH 6.0 of WT VISTA (circles), H2A mutant (squares), H2D mutant (downward triangles) and H2R mutant (gray upward triangles) and control (diamonds). 1 shows VISTA-Fc binding to -1 cells. These data are representative of two independent experiments. Figure 25E shows a computational model of the PSGL-1 19-mer glycopeptide (top) in complex with the histidine-rich ligand interface of VISTA (gray ribbon, bottom). VISTA residues H98, H100, H153 and H154 are marked. PSGL-1 residues Y46, Y48, E56, T57 and Y58 are also noted. Figures 26A-F: Figure 26A shows the BLI binding magnitude of GP1BA-his (squares) and PSGL-1 19mer-Fc (circles) with captured VISTA-Fc at the indicated pH. These data are representative of one experiment. Figure 26B shows the VISTA multimer binding histogram with human platelets. Binding was determined in the presence of non-VISTA binding control antibodies at pH 7.4 and pH 6.0 or in the presence of VISTA. 4 was performed in the presence of blocking antibodies. Unstained platelets (gray filled histogram) are also shown. These data are representative of two independent experiments. Figure 26C shows the BLI binding magnitude of VSIG-3-Fc binding to captured VISTA-Fc at the indicated pH. These data are representative of two independent experiments. Figure 26D shows the BLI binding magnitude of the indicated concentrations of VSIG-3-Fc binding to captured VISTA-Fc at pH 6.0. Competition was with buffer alone (leftmost bar), unbound isotype matched control antibody, human PSGL-1 19mer-Fc, P1-061029, P1-061767 or VISTA. Provided by 5 (far right). These data are representative of one independent experiment. Figure 26E shows VSIG-3-Fc binding to activated human PBMCT cells at pH 6.0 (circles) or pH 7.4 (squares). Binding of isotype-matched control antibodies at pH 6.0 (black diamonds) and pH 7.4 (gray triangles) is also shown. These data are representative of two independent experiments. Error bars represent standard error of the mean. Figure 26F shows BLI binding magnitude of VISTA-Fc (left) and PSGL-1 19mer-Fc (right) binding to captured VISTA-Fc at the indicated pH. These data are representative of one independent experiment. Figures 27A-F: Figures 27A-D show antibody VISTA at pH 7.4 (left) and pH 6.0 (right). 4 and P1-068767 are shown. These data are representative of two independent experiments. Figure 27A: VISTA. Human VISTA sensorgram of 4. Figure 27B: Human VISTA sensorgram of P1-068767. Figure 27C: VISTA. 4 cynomolgus monkey VISTA sensorgram. Figure 27D: Cynomolgus VISTA sensorgram of P1-068767. Figure 27E shows the levels of labeled P1-061029 or P1-068767 antibodies in various organs, blood or tumors in human VISTA knock-in mice implanted with MC38 tumors and treated with the respective antibodies. Figure 27F shows tumor growth in mice treated with P1-061029 alone (left, downward triangles) or with P1-068767 alone (right, upward triangles). n=16 per group. These data are a composite of two independent experiments. Error bars represent standard error of the mean. Figure 28 shows the results of an experiment in which human VISTA knock-in mice were implanted with MC38 tumors and treated with fluorescently labeled P1-061029 (left) or P1-068767 (right). The radiation efficiency (×10 ⁹ ) in the indicated organs 51 hours after injection is shown. These data are representative of a single experiment. Figures 29A-C show the results of an SPR assay of anti-hVISTA antibody binding to hVISTA. Figure 29A shows the relative hVISTA binding (%Rmax) at pH 6.0 (black bars) and pH 7.4 (gray bars) by SPR of the P1-068744 revertant compared to the binding of the parent P1-061015 antibody. show. Figure 29B shows the relative hVISTA binding (%Rmax) at pH 6.0 (black bars) and pH 7.4 (gray bars) by SPR of the P1-068748 revertant compared to the binding of the parent P1-061015 antibody. show. Figure 29C (positions 26-35, 50-66, 99-110 of SEQ ID NO: 95) shows a summary of the results as an alignment of antibody heavy chain CDRs. The deliniated amino acid residues of P1-068744 and P1-068748 in bold indicate that their reversion to the corresponding residue in P1-061015 maintains pH selectivity and that P1-068744 or P1-068748 The k _d is improved at pH 6.0 compared to the antibody. The amino acid residues with double underlines are those whose reversion to the corresponding residue in P1-061015 caused the loss of pH selectivity of P1-068744 or P1-068748 and thus the pH with hVISTA. is an important residue in these antibodies for selective binding. Amino acid residues with jagged underlines are those whose reversion to the corresponding residue in P1-061015 was detrimental to the kinetics and/or binding of P1-068744 or P1-068748 to hVISTA, and thus , are important residues in these antibodies for binding to hVISTA. Alternative names for the depicted amino acid residues in the P1-068744 and P1-068748 antibodies are indicated. Figures 30A-D show anti-hVISTA antibodies (10 μg/ml) engineered to express the hVISTA extracellular domain in HBSS/MES buffer at pH 6.0 (black bars) and pH 7.4 (white bars). Results from binding studies with 293T cells are shown. Results are shown as percent binding normalized to P1-061029 binding set to 100%. Figures 30A, 30B, 30C and 30D each show the results for a particular set of antibodies as shown in each bar graph below.

[Example 30]

Claims (83)

An antibody that specifically binds to human VISTA (hVISTA), wherein the antibody has 1 to 5 glutamic acids (E), aspartic acids (D), or The amino acid sequences of the HV CDR 1, VH CDR2 and VH CDR3 of the antibody as a whole have been engineered by replacing the histidine (H) amino acid with the amino acid residue found at the same position in antibody P1-061015. is an antibody different from that of antibody P1-061015. VH CDR1, CDR2 and/or CDR3 of P1-061015 or a variant thereof comprising 1-3 or 1-2 amino acid differences in the VH CDR1, CDR2 and/or CDR3 compared to antibody P1-061015, The antibody according to claim 1. 3. The antibody of claim 2, wherein the single VH CDR does not contain more than two amino acid differences compared to the corresponding VH CDR of P1-061015. 4. The amino acid difference according to claim 2 or 3, wherein the amino acid difference includes a substitution of an amino acid residue in P1-061015 with a glutamic acid residue (E), an aspartic acid residue (D), or a histidine residue (H). Antibodies . - VH CDR1 comprising GFTFSX ₁ X ₂ AMH (where X ₁ is E, H or S and X ₂ is H or Y)
-X ₃ IWYDGSNX ₄ X ₅ X ₆ ADSVKG (where X ₃ is H or I, X ₄ is D or K, X ₅ is D or Y, and X ₆ is E or -DSGFYX ₇ SYYFDX ₈ (where X ₇ is D, E or S and X ₈ is E or Y).
5. The antibody according to any one of claims 2 to 4, comprising: - VH CDR1 comprising GFTFSX ₁ YAMH (where X ₁ is E or S),
-X ₂ IWYDGSNKYX ₃ ADSVKG (where X ₂ is H or I and X ₂ is E or Y) and/or -DSGFYX ₄ SYYFDX ₅ (where X ₄ is E or S, and X ₅ is E or Y)
5. The antibody according to any one of claims 2 to 4, comprising: - VH CDR1 comprising GFTFSX ₁ X ₂ AMH (where X ₁ is H or S and X ₂ is H or Y),
- VH CDR2 _containing -IIWYDGSNX ₃ X ₄ YADSVKG ( _where X _{3 is} D or K and or S)
5. The antibody according to any one of claims 2 to 4, comprising: P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_H50I_E59Y, P1-068 744_E59Y_E100S, P1-068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S , P1-068744_H50I_E102Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1 -068748_H31S_H32Y, P1-068748_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1 - VH CDR1, CDR2 and/or CDR3 of 068748_H31S_D58Y, P1-068748_H31S_D100S, P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S 8. The antibody according to any one of claims 1 to 7, comprising: P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_H50I_E59Y, P1-068 744_E59Y_E100S, P1-068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S , P1-068744_H50I_E102Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1 -068748_H31S_H32Y, P1-068748_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1 - Contains VH CDR1, CDR2 and CDR3 of 068748_H31S_D58Y, P1-068748_H31S_D100S, P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S , the antibody according to any one of claims 1 to 7. The antibody according to any one of claims 8 or 9, comprising the VL CDR1, CDR2 and CDR3 of the VL of P1-061015. P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_H50I_E59Y, P1-068 744_E59Y_E100S, P1-068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S , P1-068744_H50I_E102Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1 -068748_H31S_H32Y, P1-068748_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1 - at least 90%, 95%, 97%, 98% or Contains VHs that contain amino acid sequences that are 99% identical ,
or P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_ modified by 1, 2, 3, 4 or 5 amino acid substitutions H50I_E59Y, P1-068744_E59Y_E100S, P1 -068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S, P1-068744_H50I_E10 2Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1-068748_H31S_H32Y, P1-068748_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1-068748_H31S_D58Y, P1-068748_H31S_D100S , P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S. Antibodies described in Section. P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_H50I_E59Y, P1-068 744_E59Y_E100S, P1-068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S , P1-068744_H50I_E102Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1 -068748_H31S_H32Y, P1-068748_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1 -068748_H31S_D58Y, P1-068748_H31S_D100S, P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S;
or P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-06 modified by 1, 2, 3, 4 or 5 amino acid substitutions in the framework region 8744_H50I_E59Y, P1-068744_E59Y_E100S, P1-068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E 100S, P1-068744_H50I_E102Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S , P1-068748_H31S_H32Y, P1-068748_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1-068748_H31S_D58Y , P1-068748_H31S_D100S, P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S, from claim 1 10. The antibody according to any one of 10. 13. The antibody of any one of claims 1-12, comprising a VL comprising an amino acid sequence that is at least 90%, 95%, 97%, 98% or 99% identical to that of P1-061015. 13. The antibody according to any one of claims 1 to 12 , comprising the VL of P1-061015. The antibodies are P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_H50I_E59Y, P1- 068744_E59Y_E100S, P1-068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S, P1-068744_H50I_E102Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58 Y, P1-068748_D100S, P1-068748_H31S_H32Y, P1-068748_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1- 068748_H31S_D57K, P1-068748_H31S_D58Y, P1-068748_H31S_D100S, P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S. 15. An antibody according to any one of claims 1 to 14, comprising the VH and VL of P1-061015. 14. The antibody of claim 13, comprising a VL comprising the amino acid sequence of the VL of P1-061015 modified by a T85V substitution. 17. The antibody according to any one of claims 1 to 16 , comprising a heavy chain constant region comprising the amino acid sequence of SEQ ID NO: 182, 183 or 184. 18. The antibody of any one of claims 1 to 17 , comprising the LC of P1-061015 or the LC of P1-061015 modified by a T85V substitution. 19. An antibody according to any one of claims 1 to 18 , which binds at or near a histidine-rich region of hVISTA , such as a histidine-rich β-sheet extension. 20. An antibody according to claim 19 , which binds at or near a histidine-rich region of hVISTA, such as a histidine-rich β-sheet extension, in conditions having a pH of 6.0 to 6.5. For binding to hVISTA, as determined by competitive biolayer interferometry (BLI) epitope binning assay or using competitive surface plasmon resonance (SPR) epitope binning , P1 -061015, P1-068744, P1-068748, P1-068744_E31S, P1-68744_H50I, P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50 I, P1-068744_H50I_E59Y, P1-068744_E59Y_E100S, P1-068744_E100S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S, P1-068744_H50I_E102Y, P1-068744_E59Y_E102Y, P1-068748_H31S, P1 -068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1-068748_H31S_H32Y, P1-068748_H32Y_D57K, P1- 068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1-068748_H31S_D58Y, P1-068748_H31S_D100S, P1-068748_H32Y_D58Y, P1 -068748_H32Y_D100S, or P1-068748_D57K_D100S). 21. An antibody according to any one of claims 1 to 20 , which competes or cross-competes with the antibody . 22. An antibody according to any one of claims 1 to 21 , which binds to epitope group A as determined by a competitive biolayer interferometry (BLI) epitope binning assay. As determined by yeast mutational analysis, the following amino acid residues: T35, Y37, K38, T39, Y41, R54, T61, F62, Q63, L65, H66, L67, H68, H69, F97, L115, 23. Does not bind significantly to hVISTA modified in that one or more of V117, I119, H121, H122, S124, E125, R127 has been mutated. Antibodies . 24. The antibody (Ab) according to any one of claims 1 to 23 , which specifically binds hVISTA under acidic conditions,
- Inhibits the interaction between hVISTA and (a) T cells and/or (b) PSGL-1 and/or (c) VSIG-3 (e.g. H153 and H154 of hVISTA with SEQ ID NO: 1 and PSGL -1 tyrosine Y46 and Y48),
- enhance T cell activation by enhancing T cell proliferation, enhancing IFN-γ production from T cells and/or stimulating T cell receptor-mediated NF-kB signaling;
one or more (e.g., at least 1 to 3, 1 to 5 , 1 ~10, 5-10, 5-15 or all) contact hVISTA by energetically important contact residues Y37, T39, R54, F62, H66, V117, I119 or S124,
- binds to the histidine-rich β-sheet extension of hVISTA, as determined by crystallography ;
- contacts H121, H122 and/or H123 of mature hVISTA (distance of 4.0 angstroms (Å) or less), as determined by crystallography ;
Binding may be strongest with region 2, as determined by M S-HDX, region 1 of hVISTA with SEQ ID NO: 1: ₅₇ LGPVDKGHDVTF ₆₈ (SEQ ID NO: 566), region 2: ₈₆ RRPIRNLTFQDL ₉₇ (SEQ ID NO: 567) and region 3: ₁₄₈ VVEIRHHSEHRVHGAME ₁₆₅ (SEQ ID NO: 568),
- For binding to hVISTA , one or more antibodies , P1-061015, P1-068744, P1-068748, P1-061029, P1-068761, P1-068767 and VISTA. 4 (two-way placement conflict),
- contacts hVISTA by at least one or more glutamic acid, aspartic acid or histidine residues located in the VH CDR1, CDR2 or CDR3;
- Has low target-mediated drug disposal , leading to a mean residence time (MRT) of at least 100, 200, 300, 400, 500, 600 or 700 hours - In hVISTA knock-in mice bearing MC38 tumors, lung, liver and/or Accumulates preferentially in tumor tissue compared to lung, liver and spleen in subjects to whom the antibody was administered.
Antibody (Ab). specifically with hVISTA under acidic conditions with a _KD (and/or koff) that is at least 10-fold, 100-fold or 1000-fold lower than the _KD or koff of binding to hVISTA under neutral or physiological pH. 25. An antibody (Ab) according to any one of claims 1 to 24 , which binds to
・Specifically binds to hVIST A at acidic pH.
- Lacks significant binding to hVIST A at physiological or neutral pH;
・Specifically binds to cyno VIST A at acidic pH .
・Lack significant binding to cyno VIST A at physiological or neutral pH;
- Compared to hVISTA ECD with SEQ ID NO: 2, the following amino acids: T35, Y37, K38, T39, Y41, R54, T61, F62, Q63, L65, H66, L67, H68, H69, F97, L115, V117 , I119, H121, H122, S124, E125, R127.
- cross-competes with P1-061015, P1-068744 and/or P1-068748 for binding to hVISTA;
・Inhibits the binding of hVISTA to human T cells expressing VISTA at acidic pH,
- inhibits the binding of hVISTA to PSGL-1 at acidic pH , PSGL-1 has or does not have siayl lewis X, the tyrosine is preferably sulfotyrosine,
- has a mean residence time (MRT) in cynomolgus monkeys of at least 100, 200, 300, 350, 400, 450, 500, 600 or 700 hours.
- stimulate T cell activation by enhancing T cell proliferation, enhancing IFN-γ production from T cells and/or stimulating T cell receptor-mediated NF-kB signaling;
・VISTA-mediated cell: inhibits cell adhesion,
- binds specifically to hVISTA in a sample of human tumor cells or a sample of inflamed human tissue that expresses VISTA;
- One or more energetically important contact residues Y37, T39, the numbering of which is that of mature hVISTA (SEQ ID NO: 2), as determined using yeast surface display and NGS assays , R54, F62, H66, V117, I119 or S124,
Binding may be strongest with region 2, as determined by M S-HDX, region 1 of hVISTA with SEQ ID NO: 1: ₅₇ LGPVDKGHDVTF ₆₈ (SEQ ID NO: 566), region 2: ₈₆ RRPIRNLTFQDL ₉₇ (SEQ ID NO: 567) and region 3: ₁₄₈ VVEIRHHSEHRVHGAME ₁₆₅ (SEQ ID NO: 568),
- binds to the histidine-rich β-sheet extension of hVISTA, as determined by crystallography ;
- contact H121, H122 and/or H123 of mature hVISTA (distance of 4.0 angstroms (Å) or less) as determined by crystallography ;
Contact hVISTA by at least one or more glutamic acid, aspartic acid or histidine residues located in VH CDR1, CDR2 or CDR3
Antibody (Ab). 26. An antibody according to any one of claims 1 to 25 , having an isoelectric point (pI) of between 6.5 and 6.8, as measured by i cIEF. 27. The antibody according to any one of claims 1 to 26 , which is an IgG antibody. 28. The antibody according to claim 27 , which is an IgG1, IgG2 or IgG4 antibody (IgG4 optionally having S228P). 29. An antibody according to any one of claims 1 to 28 , which is an effector-free antibody and /or an antibody that does not significantly bind one or more FcγRs . 1 in the wild-type heavy chain constant region that reduces the effector function of the antibody and/or the ability of the constant region to bind one or more FcγRs compared to that of the corresponding wild-type heavy chain constant region. 30. The antibody of claim 29 , comprising ˜5 mutations. 31. An antibody according to any one of claims 1 to 30 , wherein the constant region of the antibody is IgG1.3, IgG1.1, or IgG1 with a P238K substitution. 32. An antibody according to any one of claims 1 to 31 , which has effector functions and/or binds one or more FcγRs . 33. The antibody of claim 32 , which is defucosylated. 32 or 33 , wherein the constant region comprises 1 to 5 mutations that enhance the effector function of the antibody and/or the ability to bind one or more FcγRs compared to the corresponding wild type constant region. Antibodies described in. 35. The antibody of any one of claims 1 to 34 , which is a full-length antibody or an antibody comprising a full-length heavy chain (with or without a C-terminal lysine) and a full-length light chain. 35. The antibody according to any one of claims 1 to 34 , which is an antigen-binding fragment of an antibody. 37. The antibody according to any one of claims 1 to 36 , which is a multimeric antibody . 38. An antibody according to any one of claims 1 to 37 , which is linked to another molecule. 39. The antibody of claim 38 , wherein the other molecule is a label. 40. The antibody according to claim 38 or 39 , wherein the other molecule is a peptide. 41. An antibody according to any one of claims 1 to 40 , which is an antibody drug conjugate (ADC) or an activatable antibody. 42. A nucleic acid encoding an antibody according to any one of claims 1 to 41 . 42. A nucleic acid encoding the heavy chain and/or light chain of the antibody according to any one of claims 1 to 41 . 42. A composition comprising an isolated nucleic acid encoding a heavy chain of an antibody according to any one of claims 1 to 41 and a nucleic acid encoding a light chain of an antibody. 42. A cell comprising an isolated nucleic acid encoding an antibody or encoding a heavy chain and/or light chain of an antibody according to any one of claims 1 to 41 . 46. A method of preparing antibodies, the method comprising culturing the cells of claim 45 under conditions in which the antibodies are expressed. 42. A composition comprising an antibody , nucleic acid, composition or cell according to any one of claims 1 to 41 and a pharmaceutically acceptable carrier. 48. The composition of claim 47 , comprising a second therapeutic agent. 49. The composition of claim 48 , wherein the second therapeutic agent is an immunostimulant or a chemotherapeutic agent. 50. The composition of claim 49 , wherein the second therapeutic agent is an antagonist of PD -1/PD-L1, CTLA-4 and/or LAG-3 or an agonist of G ITR and/or OX40. 42. An antibody according to any one of claims 1 to 41 for use in a method of treating cancer in a subject. 52. The antibody of claim 51 , wherein the cancer comprises VISTA positive cells. 53. The antibody of claim 52 , wherein the VISTA-positive cells are VISTA-positive infiltrating lymphoid or myelomonocytic cells. 54. An antibody according to any one of claims 51 to 53 , wherein the subject is first tested for the presence of VISTA positive cells in a tumor. 55. The antibody of any one of claims 51-54 , wherein the second therapy is administered to the subject . 56. The antibody of claim 55 , wherein the second therapy is chemotherapy, radiation therapy, surgery, or administration of a second agent. 57. The antibody of claim 56 , wherein the second therapy is a second agent, and the second agent is an immunostimulant or a chemotherapeutic agent. 57. The second therapeutic agent is an immunostimulant that is an antagonist of PD -1/PD-L1, CTLA-4 and/or LAG-3 or an agonist of G ITR and/or OX40. antibodies . 42. An antibody according to any one of claims 1 to 41 for use in a method of treating an infectious disease in a subject. 42. An antibody according to any one of claims 1 to 41 for use in a method of treating inflammation, inflammatory conditions and autoimmune diseases, graft-versus-host disease or diseases that benefit from a reduced immune response. A method of improving the antitumor efficacy of an antibody (Ab) that specifically binds to the human VISTA extracellular domain (hVISTA-ECD), the method comprising:
a. providing Ab P1-061015, P1-068744 or P1-068748;
b. 1 to 5 amino acid residues in the heavy or light chain of the Ab with glutamic acid , aspartic acid or histidine residues, or in the case of P1-06844 and P1-06848, the corresponding residues of P1-061015 and 1 to 5 amino acid residues are contact residues with hVISTA-ECD,
c. determining whether the Ab obtained in (b) has a higher affinity for hVISTA-ECD at pH 6.5 or below compared to the Ab of (a); and d. repeating steps (a) to (c) for several rounds sufficient to obtain an Ab that binds hVISTA-ECD with a K _D of 10 ⁻⁷ M or less at pH 6.5 or less; Method. A method of improving the antitumor efficacy of an antibody (Ab) that specifically binds to the human VISTA extracellular domain (hVISTA-ECD), the method comprising:
a. providing Ab P1-061015, P1-068744 or P1-068748;
b. A library of variants of the Ab of (a) is prepared, each variant containing glutamic acid , aspartic acid or histidine residues of 1 to 5 amino acid residues in the heavy or light chain of the Ab; In the case of P1-06844 and P1-06848, including a substitution with the corresponding residue of P1-061015, 1 to 5 amino acid residues are contact residues with hVISTA-ECD;
c. Selecting an Ab of the library of mutants of (b) that binds hVISTA-ECD with a K _D of 10 ⁻⁷ M or less at pH 6.5 or less; and d. Optionally, a method comprising testing the anti-tumor efficacy of the Ab of (c) in a tumor model. A method of improving the efficacy of an antibody that binds human VISTA (hVISTA), the method comprising A method comprising increasing the number of residues to enhance binding of an antibody to hVISTA at acidic pH. 64. The method of any one of claims 61-63, further comprising counterselecting for antibodies that bind VISTA at physiological pH. 65. The method of any one of claims 61-64, further comprising selecting an antibody that is either a VISTA antagonist or a VISTA agonist. 66. Any one of claims 61 to 65, further comprising selecting an antibody that inhibits the interaction between VISTA and the VISTA co- receptor and/or the interaction between VISTA and T cells or myelomonocytic cells. The method described in section. 67. The method of any one of claims 61-66, further comprising selecting an antibody having one or more of the properties of antibody P1-068744 or P1-068748. A method of improving the antitumor efficacy of an antibody (Ab) that specifically binds to the human VISTA extracellular domain (hVISTA-ECD), the method comprising:
a. providing Ab P1-061015, P1-068744 or P1-068748;
b. 1 to 5 amino acid residues in the heavy chain or light chain of the Ab are replaced with different amino acid residues , and 1 to 5 amino acid residues are contact residues with hVISTA-ECD,
c. determining whether the Ab obtained in (b) has a higher affinity for hVISTA-ECD at pH 6.5 than at pH 7.0 compared to the Ab of (a); and/or whether the Ab obtained in (b) has similar or higher affinity at pH 6.5 compared to the antibody in (a), and/or compared to the antibody in (a). determining whether it has a lower affinity at pH 7.0; and d. Sufficient numbers of Abs to obtain an Ab that binds hVISTA-ECD with a K _D of 10 ⁻⁷ M or less at pH 6.5 and an Ab that binds hVISTA-ECD with a K _D of 10 ⁻⁶ M or more at pH 7.0. repeating steps (a) to (c) during said round. 42. A method of detecting VISTA in a sample, the method comprising contacting the sample with a VISTA antibody according to any one of claims 1 to 41 . 70. The antibody, composition , use or method of any one of claims 1-69, wherein the antibody inhibits the interaction between hVISTA extracellular domain (ECD) and PSGL-1. 71. The antibody, composition , use or method of any one of claims 1-70 , wherein the antibody competes with an antibody described herein for binding to hVISTA ECD. 72. The antibody, composition , use of any one of claims 1-71, wherein the antibody interacts with the histidine triad (H121, 122 and 123 of mature hVISTA) as determined by X -ray crystallography. Or how. An antibody that specifically binds human VISTA in acidic conditions for use in a method of selectively targeting an anti-VISTA Ab to a tumor in a subject. 74. The antibody of claim 73, wherein the anti-VISTA Ab has one or more of the following properties:
a. Anti-VISTA Ab preferentially accumulates in tumor cells compared to bone marrow cells, compared to blood, and/or compared to liver, lung and spleen cells ;
b. the anti-VISTA Ab accumulates to at least 2-fold higher levels in the tumor compared to blood 24-51 hours after administration of the Ab to the subject ;
c. The anti-VISTA Ab does not accumulate significantly more in the subject's liver or lungs compared to the subject's blood ;
d. Anti-VISTA Ab is shown to preferentially accumulate in tumors over blood in VISTA knock-in mice;
e. Accumulation of anti-VISTA Ab in tumors is sustained over time ;
f. Accumulation of anti-VISTA Ab in the tumor is determined by administering to the subject Ab labeled with a PET tracer ;
g. Anti-VISTA Ab specifically binds VISTA at pH 6.5, but not significantly at pH 7.0 ;
h. the anti-VISTA Ab accumulates preferentially in tumor tissue of hVISTA knock-in mice bearing MC-38 tumors compared to blood and/or compared to lung, liver and spleen ; and/or
i. The anti-VISTA Ab is VISTA. 4, P1-061015, P1-061029, P1-068757, P1-068759, P1-068761, P1-068763, P1-068765, P1-068767, P1-068769, P1-068771, P1-068773, P1-068775 , P1-069059, P1-069061, P1-069063, P1-069065, P1-069067, P1-069069, P1-069071, P1-069073, P1-069075, P1-069077, P1-061015, P1-068736, P 1- 068738, P1-068740, P1-068742, P1-068744, P1-068766, P1-068748, P1-068750, P1-068752 P1-068754, P1-068761_E55A, P1-068761_H100G, P1- 068761_E56N, P1-068761_E55A_E56N, P1 -068761_E30D, P1-068761_E30D_E55A, P1-068761_E56N_H100G, P1-068761_E30D_H100G, P1-068761_E30D_E56N, P1-068761_E100fF, P1-068761 _E55A_E100fF, P1-068761_H100G_E100fF, P1-068761_E30D_E100fF, P1-068761_E56N_E100fF, P1-068761_E32Y, P1-068761_E32Y_E55A, P1-06876 1_E32Y_E56N , P1-068761_E30D_E32Y, P1-068761_E32Y_H100G, P1-068761_E32Y_E100fF, P1-068767_D52N_D102V, P1-068767_D52N, P1-068767_D52N_E55A, P 1-068767_E55A_D102V, P1-068767_D102V, P1-068767_E55A, P1-068767_E30D_D52N, P1-068767_E30D_D102V, P1-068767_E30D, P1 -068767_E30D_E55A, P1-068767_E100fF_D102V, P1-068767_E55A_E100fF, P1-068767_D52N_E100fF, P1-068767_E100fF, P1-068767_E30D_E100 fF. I , P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_H50I_E59Y, P1-068744_E59Y_E100S, P1-068744_E10 0S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S, P1-068744_H50I_E102Y, P1- 068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1-068748_H31S_H32Y, P1-0687 48_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1-068748_H31S_D58Y, P1-068748_H31S_D100S, any one of P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S, and in the case of P1-061029 or its progeny, the VH may include one or both of the K16R and T84A substitutions; P1-061015 or its progeny, the VL may include the T85V substitution . The cells to which hVISTA binds are human leukocytes, human PBMCs, or human T cells, the cells are stimulated with anti-CD3/CD28 bead stimulation for 72-96 hours, and the hVISTA in the binding protocol is a VISTA multimer or VISTA-Fc. 42. The antibody according to any one of claims 1 to 41, which is a chimeric protein. 76. The antibody of claim 75, which inhibits hVISTA binding to cells by at least 50%, at least 70%, at least 80%, at least 90%, at least 95%, at least 98%, at least 99% or 100%. Claim 75 or 76 , wherein the antibody inhibits the binding of hVISTA on the first cell to human VSIG-3 on the second cell, and the second cell is a cell expressing human VSIG-3. Antibodies described in. The first cell is a HEK293 cell that ectopically expresses hVISTA, the second cell is a CHO cell that ectopically expresses human VSIG-3, and the hVISTA expressed by the first cell is 78. The antibody of claim 77 , wherein is a VISTA multimer or a VISTA-Fc chimeric protein. 50%, at least 70%, at least 80%, at least 90%, at least 95%, at least 98%, at least 99% or 100% inhibits binding of the first cell to the second cell. 78 . An antibody that specifically binds human VISTA (hVISTA) , as determined using competitive SPR epitope binning, the antibody VISTA. 4, P1-061015, P1-061029, P1-068757, P1-068759, P1-068761, P1-068763, P1-068765, P1-068767, P1-068769, P1-068771, P1-068773, P1-068775 , P1-069059, P1-069061, P1-069063, P1-069065, P1-069067, P1-069069, P1-069071, P1-069073, P1-069075, P1-069077, P1-061015, P1-068736, P 1- 068738, P1-068740, P1-068742, P1-068744, P1-068766, P1-068748, P1-068750, P1-068752 P1-068754, P1-068761_E55A, P1-068761_H100G, P1- 068761_E56N, P1-068761_E55A_E56N, P1 -068761_E30D, P1-068761_E30D_E55A, P1-068761_E56N_H100G, P1-068761_E30D_H100G, P1-068761_E30D_E56N, P1-068761_E100fF, P1-068761 _E55A_E100fF, P1-068761_H100G_E100fF, P1-068761_E30D_E100fF, P1-068761_E56N_E100fF, P1-068761_E32Y, P1-068761_E32Y_E55A, P1-06876 1_E32Y_E56N , P1-068761_E30D_E32Y, P1-068761_E32Y_H100G, P1-068761_E32Y_E100fF, P1-068767_D52N_D102V, P1-068767_D52N, P1-068767_D52N_E55A, P 1-068767_E55A_D102V, P1-068767_D102V, P1-068767_E55A, P1-068767_E30D_D52N, P1-068767_E30D_D102V, P1-068767_E30D, P1 -068767_E30D_E55A, P1-068767_E100fF_D102V, P1-068767_E55A_E100fF, P1-068767_D52N_E100fF, P1-068767_E100fF, P1-068767_E30D_E100 fF. I , P1-68744_E59Y, P1-068744_E100S, P1-068744_E102Y, P1-068744_E31S_H50I, P1-068744_H50I_E59Y, P1-068744_E59Y_E100S, P1-068744_E10 0S_E102Y, P1-068744_E31S_E102Y P1-068744_E31S_E59Y, P1-068744_E31S_E100S, P1-068744_H50I_E100S, P1-068744_H50I_E102Y, P1- 068744_E59Y_E102Y, P1-068748_H31S, P1-068748_H32Y, P1-068748_D57K, P1-068748_D58Y, P1-068748_D100S, P1-068748_H31S_H32Y, P1-0687 48_H32Y_D57K, P1-068748_D57K_D58Y, P1-068748_D58Y_D100S, P1-068748_H31S_D57K, P1-068748_H31S_D58Y, P1-068748_H31S_D100S, An antibody that blocks or cross -blocks the binding of any one of P1-068748_H32Y_D58Y, P1-068748_H32Y_D100S, or P1-068748_D57K_D100S to hVISTA. An anti- PSGL-1 antibody that inhibits the binding of human VISTA (hVISTA) to cells, the cells being cells to which hVISTA binds. The cells to which hVISTA binds are human leukocytes, human PBMCs or human T cells, the cells are stimulated with anti-CD3/CD28 bead stimulation for 72-96 hours, and the hVISTA in the binding protocol is either VISTA multimer or VISTA- 82. The anti -PSGL-1 antibody according to claim 81 , which is an Fc chimeric protein. 83. The anti-PSGL of claim 81 or 82 , which inhibits hVISTA binding to cells by at least 50%, at least 70%, at least 80%, at least 90%, at least 95%, at least 98%, at least 99% or 100%. -1 antibody.