Extended Data Fig. 9: The antitumor profile of cromolyn is dependent on its effect on mast cells.
(a) Growth curves and tumor weights at the endpoint in mice treated with 30 mg/kg fexofenadine (histamine antagonist) alone or in combination with 10 mg/kg anti-ACK2 neutralizing antibody (to deplete mast cells), anti-CSF1R neutralizing antibody (to deplete macrophages), or isotype control in the AT3 orthotopic tumor model (n = 5). (b, c) Growth curves and tumor weights at the endpoint in B6.Cg-KitW-sh mice (b, n = 5) or Cpa3CreERT2Cd74fl/fl mice (c, n = 5) treated with 30 mg/kg fexofenadine and 10 mg/kg anti-PD-1. (d) Growth curves and tumor weights at the endpoint in mice transferred with mast cells (CD45+CD117+), macrophages (CD45+F4/80+), dendritic cells (CD45+CD11c+), or other cells (CD45+CD117−F4/80−CD11c−, excluding mast cells, macrophages, and dendritic cells) pretreated with vehicle (Veh) or cromolyn (Cro). After cultured ex vivo for 3 days, cells were washed and transferred into mice bearing AT3 orthotopic tumors (n = 5). (e) Individual cells were sorted via flow cytometry according to classic surface markers and treated as indicated (under the condition of ± cromolyn and ± tumor lysate) for 12 h. After washing out, we cocultured these cells with healthy donor-derived T cells at 1:5 ratio for 24 h, which were then used for tumor-killing assay (1:5 ratio for 48 h). (f) Experimental groups of the study design. To avoid potential bias, we compared each cromolyn-treatment group versus vehicle-treatment group under the same condition (as per tumor/normal lysate and cell type). (g, h) Quantification of relative tumor viability from MDA-MB-231 (g) and Hs578T (h) cell lines, where MCF10A normal breast epithelial cells were used as normal control (n = 6). (i) Quantification of relative tumor viability from patient-derived organoids. In this study, each tumor was cut into small pieces and enzymatically digested. Then, digested tissues were equally mixed and separated into two parts for generating PDO and tumor lysate, respectively, and paired peritumoral tissues (≥2 cm from tumor region) were used as normal control (n = 10 per group, every point represents the mean value of 3 replicates). Data were normally distributed (Shapiro-Wilk test P > 0.05) and compared using two-sided paired t test. (j) Proliferation of indicated cells measured by CCK-8 reagent (n = 3). Two-way ANOVA was used for comparison. Data are mean ± s.e.m. (a–d, g–j). Two-sided unpaired Student’s t-test (a–d for tumor weight), paired Student’s t-test (g–i), two-way ANOVA (a–d for tumor volume, j).
