Figure 1.
Binding of anti-CCR4 scFv fragments to avian DT40 cells transfected with human CCR4.
The results of titration on DT40/CCR4 and non-transfected DT40 cells are shown for scFv 17G (a), 9E (b), 1O (c) and 11F (d). Binding of scFv fragments in the presence of increasing concentrations of CCR4 ligands CCL22 and CCL17 is shown in panels (e) and (f), respectively. As a control, a scFv fragment derived from hybridoma KM2160 was used. Cell binding was analyzed by flow cytometry; bound scFv fragments were detected with anti-cMyc antibody followed by anti-human-PE immunoconjugates. Median fluorescence intensity is plotted against the scFv concentration (µg/mL); the results of representative experiments from three repeats are shown.
Figure 2.
Dose-dependent binding of anti-CCR4 IgG antibodies to CCR4+ and CCR4− cell lines.
Different concentrations of 17G, 9E and KM3060var antibodies were tested in flow cytometry for binding to avian DT40 cells stably transfected with human CCR4 (a), to human HEK-293 cells transiently transfected with human CCR4 (b) or to human T-cell leukemia line CCRF-CEM naturally expressing CCR4 (c). Binding to CCR4-negative non-transfected cells is also shown in panels (a) and (b). The results of representative experiments from three repeats are shown.
Figure 3.
Effect of antibodies on ligand-induced signaling using calcium flux assay.
CCR4-positive CCRF-CEM cells loaded with the Ca2+ sensing dye Fluo-4 were pre-incubated with 10 µg/mL of IgGs KM3060var, 17G or 9E for 15 min before adding CCR4-specific ligands, CCL17 or CCL22 at final concentrations of 10 ng/mL and 2.5 ng/mL, respectively. As a negative control, an irrelevant chemokine, stromal cell-derived factor-1α (SDF-1α/CXCL12), specific for CXCR4 receptor was used at a concentration of 2.5 ng/mL. Time course data from calcium flux assays were expressed in terms of relative fluorescent units (RFU). The black line in all panels represents the signal from the non-treated cells (control). Effect of antibodies on cell signaling induced by the ligands CCL22, CCL17 and CXCL12 is shown in panels (a), (b) and (c), respectively. Effect of antibodies alone at concentrations 10 µg/mL and 100 µg/mL on cell signaling is shown in panels (d) and (e), respectively.
Table 1.
Effect of anti-CCR4 antibodies on calcium mobilization induced by CCR4-specific ligands CCL22 and CCL17 and by irrelevant ligand CXCL12.
Figure 4.
ADCC activity of anti-CCR4 antibodies.
Representative experiments demonstrating dose dependent killing of CCRF-CEM cells is shown. (a) Comparison of ADCC activity of human IgG1 variants 17G and 9E. (b) Comparison of killing activity of the human IgG1 variant 9E with a comparator antibody KM3060var. Mean and SD values of quadruplicates are plotted.
Table 2.
Comparative ADCC activity of anti-CCR4 antibodies.
Figure 5.
Affinity assessment of the improved anti-CCR4 antibodies.
(a, b) Analysis of dissociation rates in the cell surface retention assay in vitro on DT40/CCR+ cells. Either scFv molecules cross-linked with PE-labeled anti-cMyc antibody (a) or human IgG1 molecules labeled with Alexa Fluor 488 (b) were used for cell binding. In all experiments the unlabeled 9E10J IgG1 molecules were added to the cells to prevent rebinding of labeled molecules; decay of the fluorescence signal was recorded. Values are expressed as a percentage of the initial median fluorescence intensity. (c–e) Inhibition of biotinylated antibodies 9E (c, d) and 9E10J (e) binding to DT40/CCR4 (c, e) or CCRF-CEM cells (d) in the presence of increasing concentrations of unlabeled IgGs 9E, 9E10J, 306, 406, 503 and control antibody KM3060var. (f, g) Inhibition of labeled ligands, CCL22 (f) or CCL17 (g), binding to DT40/CCR4 cells in the presence of increasing concentrations of unlabeled IgG 9E10J, 306, 406, 503, control antibody KW-0761var and of the corresponding unlabeled ligand.
Table 3.
Apparent affinities of anti-CCR4 antibodies (human IgG1 format) as determined in cell binding experiments.
Table 4.
Cell surface retention of anti-CCR4 antibodies.
Figure 6.
Analysis of species cross-reactivity in cell-binding experiments.
(a–c) Human IgG1 antibody 503 (1 µg/mL) demonstrates binding to HEK-293 cells transiently expressing either human CCR4 (huCCR4; a) or the receptor from rhesus macaque (reCCR4; b), or from mouse (muCCR4; c). As negative controls, binding of the antibody 503 to mock-transfected HEK-293 cells is shown. (d) Human IgG1 antibody 503 (2 µg/mL) specifically binds to the mouse kidney cancer cell line Renca. (e) Dose-dependent binding of biotinylated IgG1 503 to activated mouse splenocytes. In addition, binding patterns of the isotype control and anti-human CCR4 commercial antibody 1G1 are shown.
Figure 7.
Effect of human anti-CCR4 antibody 9E10J on platelet aggregation.
(a) Binding of 9E10J IgG1 at concentration 10 µg/mL to human platelets isolated from the fresh donor blood. As negative controls, binding histograms for an isotype control antibody (Isotype) and of a secondary antibody alone (anti-human-PE) are shown. (b) Effect of anti-CCR4 antibody 9E10 and of the isotype control antibody on platelet aggregation in comparison with ADP. (c) Platelet aggregation induced by CCR4 ligands, CCL17 and CCL22, in comparison with ADP. (d, e) Ligand-induced aggregation of platelets pre-incubated with either the isotype control antibody (d) or anti-CCR4 antibody 9E10J (e).
Figure 8.
Dose-dependent effect of anti-CCR4 antibodies on CCL17-induced signaling using calcium flux assay.
The results of four independent experiments on CCRF-CEM cells loaded with Fluo-4 using different concentration intervals are shown in panels (a–d), (e–h), (i) and (j). The cells were pre-incubated with human IgG1 variants 9E, 9E10J, 306, 406, 503 or with the control antibodies KM3060var and KW-0761var for 15 min before adding a CCR4-specific ligand CCL17 at a concentration of 10 ng/mL. The IgG concentrations were either 1 (a), 10 (b) and 100 ng/mL (c) or 0.1 (e), 1.0 (f) and 10 µg/mL (g) in the first and second set of experiments, respectively. The areas under the curves (AUC) were integrated using software PRISM (GraphPad) and plotted as percentage of AUC for maximal stimulation with CCL17 alone against antibody concentrations, as shown in panels (d) and (h) for the first and second set of experiments, respectively. The results of other two independent experiments using broader range of antibody concentrations are shown in panels (i) and (j).
Figure 9.
Dose-dependent effect of anti-CCR4 antibodies on CCL17 or CCL22-induced chemotaxis of CCR4+ CCRF-CEM cells.
Human T-cell leukemia CCRF-CEM cells were induced to migrate in the transwell plates, where the CCR4 ligands were placed in the lower chamber and the cells were co-incubated with either CCR4-specific antibodies or isotype control antibody in the upper chambers. The cells migrated to the lower chamber were detected and counted by flow cytometry. (a, c) Inhibition of CCL17-induced migration. (b, d) Inhibition of CCL22-induced migration. Mean and SD values of triplicates are plotted.
Figure 10.
ADCC activity of anti-CCR4 antibodies using different tumor cell lines and autologous Treg cells as targets.
(a, b) Comparison of the ADCC activity on T-cell leukemia CCRF-CEM cells of the affinity improved variant 9E10J to its parent antibody 9E (a) and a comparator antibody KM3060var (b). (c, d) Analysis of kifunensine effect (indicated with K) on ADCC activity of the human antibody 9E10J (c) and of the comparator antibodies KM3060var (not treated with kifunensine) and defucosylated KW-0761var (d) using CCRF-CEM as target cells. (e–g) ADCC activity of the defucosylated affinity improved anti-CCR4 antibodies 306 K, 406 K, 503 K (K means kifunensine treatment) and of the comparator antibody KW-0761var, prepared under the same conditions, on CCRF-CEM cells (e), Hodgkin’s lymphoma L428 cells (f) and cutaneous T-cell lymphoma HUT78 cells (g). (h) Comparative ADCC activity of anti-CCR4 antibodies on isolated autologous Treg cells. The IgG1 molecules were used at a concentration of 3.5 nM. Cytotoxicity was normalized to a maximum release (100% cell lysis) in presence of Triton X-100. The killing activity (%) is shown in brackets. Typical experiments from 3–8 repeats are shown. Mean and SD values of quadruplicates are plotted.
Figure 11.
ADCP activity of defucosylated anti-CCR4 IgG1 antibody 503 K in the presence of isolated human monocytes.
Dose-dependent phagocytosis of T-cell leukemia CCRF-CEM cells (a) and human renal cell cancer 786-O cells (b) incubated with monocytes at E:T ratio 5∶1 for 1, 2 or 3 hrs is shown. Dotted lines indicate the levels of spontaneous phagocytosis.
Figure 12.
Kaplan-Meier survival plots of nude mice bearing CCRF-CEM tumors.
Indicated treatments were given i.v. twice weekly when the tumors became palpable. (a) Comparison of different isotypes of anti-CCR4 antibody 9E10J (human IgG1 and murine IgG2a). (b) Comparison of affinity improved anti-CCR4 variants 306, 406 and 503 as defucosylated human IgG1. The overlay plots comparing the treated and control groups are shown separately for each treatment group. The mean survival times (MST) ± SD for each group are shown in the insets; the median survival times are shown in the brackets. Significant differences between the experimental groups are indicated by *(P<0.05), **(P<0.01), ***(P<0.001), or ns (not significant).