Fig 2: Coculture with CD8+ T cells promotes estrogen-independent growth of HR+ breast cancer cells.(A and B) MCF7 (A) and T47D (B) cells with or without T-ALL-104 cells in Transwell inserts were seeded in EFM as described in Methods. Data represent the mean ± SD of MCF7, and T47D cell numbers were evaluated 6 days after seeding. Experiments were performed at least 3 times in triplicate wells. Statistical differences were assessed using 1-way ANOVA with Dunnett’s correction for multiple testing; P < 0.05 was considered statistically significant. (C–E) ELISA analysis of CXCL11 (C), CXCL9 (D), and CXCL10 (E) levels in MCF7 with or without T-ALL-104 cells as measured by ELISA. MCF7 cells with or without T-ALL-104 cells were seeded in EFM as in A and B. Media conditioned by MCF7 cells with or without T-ALL-104 cells were collected after 6 days. Experiments were performed at least 3 times in triplicate wells. Data represent the mean ± SD. Statistical differences were assessed using 1-way ANOVA with Dunnett’s correction for multiple testing; P < 0.05 was considered statistically significant. (F) CXCL11 expression in PanCK+, CD45+, PanCK–CD45– compartments evaluated using spatial transcriptomics. Kruskal-Wallis test and Benjamini-Krieger-Yekutieli FDR adjustment for multiple comparisons was applied to compare cell-type distributions; q < 0.05 was considered statistically significant. (G) CXCL11 expression inferred from CIBERSORT. Kruskal-Wallis test and Benjamini-Hochberg FDR adjustment for multiple comparisons was applied to compare CXCL11 expression across cell types and time points; q < 0.05 was considered statistically significant.

Fig 3: CXCL11 promotes resistance to estrogen suppression.(A) Expression of chemokines associated with antitumor activity upregulated in ED-resistant (versus. ED-sensitive) tumors after (on the left) and prior to (on the right) treatment with letrozole. P adj, adjusted P value. (B and C) Growth curves of MCF7 (B) or T47D (C) cells, seeded in EFM, alone or in the presence of CXCL9, CXCL10, and CXCL11, acquired using IncuCyte live cells. The mean ± SD of the number of cells is shown (note that SDs are not always visible when smaller than the size of the symbol). *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, by unpaired t test. Experiments were performed at least 3 times in triplicate wells. (D and E) MCF7 (D) or T47D (E) cell proliferation evaluated by IncuCyte, as in B and C. Cells were seeded as in B and C and treated with PBS plus 1% BSA (control) or with CXCL9, CXCL10, and CXCL11 (10 nM each) or double (CXCL9 + CXCL10, CXCL9 + CXCL11, CXCL10 +CXCL11) or triple (CXCL9 + CXCL10 + CXCL11) combinations. On day 5, all the treatment arms were compared with vehicle using 1-way ANOVA with Dunnett’s correction for multiple testing; P < 0.05 was considered statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001). Experiments were performed at least 3 times in triplicate wells. (F and G) Immunoblot analysis of MCF7 (F) and T47D (G) levels in cell lysates. Cells were seeded overnight in serum-starved medium, and the lysates were collected 15 minutes after adding 10 nM CXCL11 and then probed with the indicated antibodies. tot, total.

Fig 4: Coculture with CD8+ T cells promotes estrogen-independent growth of HR+ breast cancer cells.(A and B) MCF7 (A) and T47D (B) cells with or without T-ALL-104 cells in Transwell inserts were seeded in EFM as described in Methods. Data represent the mean ± SD of MCF7, and T47D cell numbers were evaluated 6 days after seeding. Experiments were performed at least 3 times in triplicate wells. Statistical differences were assessed using 1-way ANOVA with Dunnett’s correction for multiple testing; P < 0.05 was considered statistically significant. (C–E) ELISA analysis of CXCL11 (C), CXCL9 (D), and CXCL10 (E) levels in MCF7 with or without T-ALL-104 cells as measured by ELISA. MCF7 cells with or without T-ALL-104 cells were seeded in EFM as in A and B. Media conditioned by MCF7 cells with or without T-ALL-104 cells were collected after 6 days. Experiments were performed at least 3 times in triplicate wells. Data represent the mean ± SD. Statistical differences were assessed using 1-way ANOVA with Dunnett’s correction for multiple testing; P < 0.05 was considered statistically significant. (F) CXCL11 expression in PanCK+, CD45+, PanCK–CD45– compartments evaluated using spatial transcriptomics. Kruskal-Wallis test and Benjamini-Krieger-Yekutieli FDR adjustment for multiple comparisons was applied to compare cell-type distributions; q < 0.05 was considered statistically significant. (G) CXCL11 expression inferred from CIBERSORT. Kruskal-Wallis test and Benjamini-Hochberg FDR adjustment for multiple comparisons was applied to compare CXCL11 expression across cell types and time points; q < 0.05 was considered statistically significant.