Fig 2: BoxA exerts therapeutic effects in a model of colon cancer Ecto‐calreticulin exposure (expressed as MFI) in CT26 cells treated with 800 nM BoxA or 1 µM MTX for 2 and 4 h. Error bars indicate standard deviation. Data were pooled from two different experiments.HMGB1 levels in the medium of CT26 cells treated with BoxA or 1 µM MTX for 24 h (n = 3).Ratio of phosphorylated eIF2α (p‐eIF2α) over eIF2α in CT26 cells treated with increasing concentrations of BoxA or 1 µM MTX for 24 h. β‐actin is shown as loading control (n = 3).Surface CD47 on CT26 cells after incubation for 24 h with the indicated concentrations of BoxA. The experiment shown is representative of 2 performed, in biological triplicate.Cell proliferation. CT26 cells (3 × 105) were seeded at t‐16 h, then exposed to increasing concentrations of BoxA or 1 µM MTX, and counted after further 24 h (n = 3).Western blot analysis of cleaved caspase‐3 in CT26 cells exposed or not to BoxA or MTX for 24 h. α‐tubulin was used for normalization (n = 2).Scheme of the experiment. Twenty‐two BALB/c mice were inoculated i.p. with 3.5 × 104 CT26‐LUC cells and treated with either 800 µg BoxA or PBS three times a week, 10 times in total. Yellow arrows represent BLI imaging.Tumor growth was detected by BLI. Lines that do not reach day 20 correspond to mice that were sacrificed for ethical reasons. One BoxA‐treated mouse survived without any BLI signal and was re‐challenged with CT26‐LUC cells at day 65; at day 79, it was sacrificed and no tumor masses were detected.Kaplan–Meier survival curves of mice shown in panel (H). Statistics: Gehan–Breslow–Wilcoxon test.BLI was measured 2 h after re‐challenge (upper images) and 1 week later (lower images). The survivor was imaged also at much higher sensitivity (note the different scales for radiance). Data information: In panels (A‐F), bars and error bars represent mean ± SD; statistics: One‐way ANOVA with Dunnett’s post‐test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Source data are available online for this figure.

Fig 3: Expression of CXCR4, RAGE, TLR4, and CD47 molecules mRNAs extracted from the indicated samples were subjected to qPCR with primers designed on Cxcr4, Rage, and Tlr4 transcripts. PCR products were loaded on a 2% agarose gel. Numbers below the lanes indicated CT values for the various samples.Representative confocal image of Proximity Ligation Assay performed on CD47, CXCR4, RAGE, and TLR4. MM cells were incubated over night with 400 nM BoxA or PBS and then permeabilized or not. Red dots represent physical contact of two identical receptor molecules. Nuclei are in blue (DAPI), phalloidin is green. Scale bar, 20 µm. Representative images from one experiment out of three performed are shown. Source data are available online for this figure.

Fig 4: BoxA does not induce CD47 degradation or shedding Western blot analysis of CD47 in the total lysate of MM cells exposed to BoxA or 1 µM MTX for 24 h. β‐actin was used for normalization.Western blot of CD47 assessed in the medium of MM cells exposed to BoxA for 24 h. Ponceau staining was used as loading control. Data information: The experiments shown are representative of 2 performed, in biological duplicate. Error bars indicate standard deviation. Statistics: one‐way ANOVA plus Dunnett’s post‐test, *P < 0.05. Source data are available online for this figure.

Fig 5: Engagement of CXCR4 by BoxA mediates CD47 internalization Confocal images of Proximity Ligation Assays performed on CD47 and CXCR4, CD47 and RAGE, CD47 and TLR4. Representative images from one experiment out of three performed are shown. MM cells were incubated overnight with 400 nM BoxA or PBS. Nuclei are in blue (DAPI), phalloidin is in green. Red dots represent physical contact of CD47 with CXCR4 or RAGE or TLR4. Scale bar, 20 µm. The intensity of red signal in PLA was quantified as described in Materials and Methods in individual cells (n as indicated). Insets are enlargements of the y axis. Mean and SD are indicated. Statistics: Kolmogorov–Smirnov test. Cxcr4 mRNA expression in MM cells transfected with shCxcr4 (pools 1 and 2) relative to MM cells transfected with shScramble (AB1 shCTR), one biological replicate, in technical triplicate.Cell proliferation. shCTR (n = 3) and shCxcr4 MM cells (pools 1+2, n = 6) were plated (4.4 × 104 at the start of the experiment, t‐16 h), treated or not with 800 nM BoxA for 24 h and counted. Statistics: t‐test.Flow cytometry analysis of surface CD47 in shCTR and shCxcr4 MM cells (pools 1+2) exposed or not to 800 nM BoxA for 24 h.Flow cytometry analysis of surface CD47 on MM cells after exposure to 800 nM BoxA and 100 nM AMD3100 for 24 h. Data information: In panels (D, E) bars and error bars represent mean ± SD; statistics: One‐way ANOVA plus Tukey’s post‐test. In all panels, error bars indicate standard deviation; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Source data are available online for this figure.