Loss-of-function of the hippo transducer TAZ reduces mammary tumor growth through a myeloid-derived suppressor cell-dependent mechanism

Abstract

TAZ, one of the key effectors in the Hippo pathway, is often dysregulated in breast cancer, leading to cancer stemness, survival, and metastasis. However, the mechanistic bases of these tumor outcomes are incompletely understood and even less is known about the potential role played by the non-malignant cellular constituents of the tumor microenvironment (TME). Here, we revealed an inverse correlation between TAZ expression and survival in triple-negative breast cancer (TNBC), but not other subtypes of breast cancer. We found that TAZ knockdown in two murine TNBC tumor cell line models significantly inhibited tumor growth and metastasis in immune competent but not immune deficient hosts. RNA-seq analyses identified substantial alterations in immune components in TAZ knockdown tumors. Using mass cytometry analysis, we found that TAZ-deficiency altered the immune landscape of the TME leading to significant reductions in immune suppressive populations, namely myeloid-derived suppressor cells (MDSCs) and macrophages accompanied by elevated CD8⁺ T cell/myeloid cell ratios. Mechanistic studies demonstrated that TAZ-mediated tumor growth was MDSC-dependent in that MDSC depletion led to reduced tumor growth in control, but not TAZ-knockdown tumor cells. Altogether, we identified a novel non-cancer cell-autonomous mechanism by which tumor-intrinsic TAZ expression aids tumor progression. Thus, our findings advance an understanding of the crosstalk between tumor-derived TAZ expression and the immune contexture within the TME, which may lead to new therapeutic interventions for TNBC or other TAZ-driven cancers.

Fig. 1. TAZ is highly expressed and correlated with poor outcome of TNBCs.

A Comparison of TAZ mRNA expression between triple-negative breast cancer (TNBC) and non-TNBC in TCGA breast cancer datasets. TNBC; n = 160, non-TNBC; n = 862. B TAZ mRNA expression comparison by PAM50 classification in TCGA breast cancer datasets. Basal-like (Basal); n = 139, Her2; n = 67, Luminal A; n = 419, Luminal B; n = 192, normal-like (normal); n = 23. C Breast cancer disease-free survival Kaplan-Meier curves were generated using a TAZ target score in the whole cohort or each subtype of the TCGA breast cancer cohort. High TAZ target expression was significantly associated with high immune scores (D) and high stromal scores (E).

Fig. 2. Knockdown of Taz inhibits tumor growth.

A Immunoblotting detection of Taz knockdown in 4T1 cells; tumor growth in BALB/c mice as measured by caliper and tumor weights. GAPDH was used as a loading control. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: *p < 0.05; ***p < 0.001, n = 6. B Immunoblotting detection of Taz knockdown in EMT6 cells; tumor growth measured in BALB/c as in panel A. GAPDH was used as a loading control. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: *p < 0.05; **p < 0.01. n = 6. C Immunoblotting detection of Taz knockdown using shNT (Non-target) and two sequences independent shTaz in 4T1 cells; Tumor weights and quantification of lung tumor metastasis were measured for 4T1 shNT and shTaz 4T1 cells from BALB/c mice. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: ***p < 0.001. met = metastasis nodules n = 6.

Fig. 3. Knockdown of Taz alters the immune cell components in TME.

A Volcano plot shows significant gene expression alterations between sgTaz and sgCon 4T1 tumors. B Over-representation analysis (ORA) using pathway annotations. C Representative heatmap data of NanoString immunology panel analyses in sgCon and sgTaz 4T1 cells. D qRT-PCR analyses of Il33, TGF-β1, Ccl5, Il1a, and Cx3cl1 expression in sgCon and sgTaz 4T1 cells. Relative expression was normalized by GAPDH expression. Unpaired two-tailed student t-test: *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 4. Accumulation of immune suppressive cells in 4T1 sgCon or sgTaz tumors.

A Representative dot plots and quantification of PMN-MDSCs and M-MDSCs in the 4T1 sgCon or sgTaz primary tumors. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: **p < 0.01; ***p < 0.001. B Representative dot plots and quantification of CD11b⁺F4/80⁺ cells in the 4T1 sgCon or sgTaz primary tumors. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: **p < 0.01. C Representative dot plots and quantification of CD4⁺ T regulatory cells (CD25⁺CD39⁺) in the 4T1 sgCon or sgTaz tumors. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: *p < 0.05; ***p < 0.001. D Ratio of CD8⁺ T cells with PMN-MDSCs and macrophages in 4T1 sgCon or sgTaz 4T1 tumors. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: **p < 0.01.

Fig. 5. Anti-Gr-1 antibody treatment reduces tumor growth in 4T1 sgCon, but not sgTaz tumors.

A Schematic of anti-Gr-1 antibody or isotype control antibody treatment in 4T1 sgCon or sgTaz tumor-bearing mice. B Tumor growth was measured by caliper after implantation of 4T1 sgCon or sgTaz 4T1 cells, with or without the indicated antibody (A). Data are shown as the mean ± SD. Unpaired two-tailed student t-test: *p < 0.05; ***p < 0.001. n = 6. C Quantification of CD45⁺ cells. D CD4⁺ T cells. E CD8⁺ T cells. F CD11b⁺ cells. G Gr-1⁺ MDSCs. H PMN-MDSCs. I M-MDSCs. J Macrophages in the 4T1 sgCon or sgTaz tumors. Data are shown as the mean ± SD. Unpaired two-tailed student t-test: *p < 0.05; **p < 0.01.