Fig 2: PTX008 inhibits binding of Galectin-1 to the surface of ALL cells. LAX56 cells (2 × 104) were plated without OP9 stroma and treated for 1 h with DMSO or 10 µM PTX008. The ability of human recombinant Galectin-1 or human recombinant GST-Galectin-3 or control GST to cause agglutination of the LAX56 cells was then quantitated by light microscopy. a Representative images of cells. Left two panels show controls of cells incubated with GST only or DMSO only. b quantification of number of aggregates in each condition. (*p < 0.05)

Fig 3: Galectin-1-mediated adhesion to fibronectin is inhibited by PTX008. a-c Representative analysis of (a) US7, (b) LAX56 or (c) LAX57 cells adhering to fibronectin (left panels) or poly-L-lysine coated plates (middle panels) in the presence of increasing concentrations of PTX008 or 5 and 50 mM lactose. Error bars: standard deviation. Cell surface staining using FACS (Genetex antibodies) for Galectin-1 on these samples (right panels). Numbers, percentage of cells in the upper left quadrants. *p=0.05 (95% CI, 1-way ANOVA). d Summary fold difference in adhesion of LAX56 and LAX57 to fibronectin and poly-L-lysine coated plates. Percent adhesion was normalized to DMSO control for 3 replicate experiments with 3 or more replicate wells. Error bars, standard error of mean (**p<0.01, 1-way ANOVA)

Fig 4: Galectin-1 inhibition reduces ALL cell migration. US7 (a) or LAX56 (b) cells treated with DMSO or 10 µM PTX008 were allowed to migrate toward aMEM + 2%FBS (control), 200 ng/mL SDF1a, or a confluent layer of irradiated OP9 cells. Left panels, representative experiment. Error bars, standard deviation (**, p<0.01, ***, p<0.001, 95% CI, 2-way ANOVA). Right panels, summary data from 3 replicate experiments with 3 replicate samples each. Error bars, standard error of mean (**, p<0.01, ***, p<0.001, 95% CI, 2-way ANOVA)

Fig 5: Galectin-1 inhibition combined with vincristine results in increased cytotoxicity to ALL cells in vitro. a and c. Representative analysis of proliferation (left panels, cell numbers) and viability (right panels) of LAX56 (a) and LAX57 (c) treated for 72 h with PTX008 alone (top panels) or PTX008 plus vincristine (bottom panels) in the presence of OP9 stromal support. Error bars, standard deviation (*p<0.05, 95% CI, 1-way ANOVA). b and d Summary fold difference LAX56 (b) and LAX57 (d). Cell number and viability was normalized to DMSO or vincristine only control for 3 replicate experiments with 3 replicate samples. Error bars represent standard error of mean (*, p<0.05; **, p<0.01; ***, p<0.001, two-tailed t-test). e Cell counts (left panel) and viability (right panel) of LAX56 in combination treatment for 72 h with a fixed amount of 10 µM PTX008 (- 1 and - 2: drug from different sources) and the indicated concentration of vincristine in the presence of mitotically inactivated (mitomycin C-treated) OP9 stroma. Error bars, SD. One experiment, triplicate wells (*, p<0.05; ***, p<0.001, ****, p<0.0001, compared to vincristine only or DMSO [0 nM vincristine]). Two way ANOVA, Dunnett’s multiple comparison test