ZEB1 promotes chemoimmunotherapy resistance in pancreatic cancer models by downregulating chromatin acetylation of CXCL16
Shaobo Zhang, Yumeng Hu, Zhijun Zhou, Gaoyuan Lv, Chenze Zhang, Yuanyuan Guo, Fangxia Wang, Yuxin Ye, Haoran Qi, Hui Zhang, Wenming Wu, Min Li, Mingyang Liu
Shaobo Zhang, Yumeng Hu, Zhijun Zhou, Gaoyuan Lv, Chenze Zhang, Yuanyuan Guo, Fangxia Wang, Yuxin Ye, Haoran Qi, Hui Zhang, Wenming Wu, Min Li, Mingyang Liu
View: Text | PDF
Research Article Cell biology Oncology

ZEB1 promotes chemoimmunotherapy resistance in pancreatic cancer models by downregulating chromatin acetylation of CXCL16

Abstract

Pancreatic cancer (PC) is notoriously resistant to both chemotherapy and immunotherapy, presenting a major therapeutic challenge. Epigenetic modifications play a critical role in PC progression, yet their contribution to chemoimmunotherapy resistance remains poorly understood. Here, we identified the transcription factor ZEB1 as a critical driver of chemoimmunotherapy resistance in PC. ZEB1 knockdown synergized with gemcitabine and anti–PD-1 therapy, markedly suppressed PC growth, and prolonged survival in vivo. Single-cell and spatial transcriptomics revealed that ZEB1 ablation promoted tumor pyroptosis by recruiting and activating GZMA+CD8+ T cells in the tumor core through epigenetic upregulation of CXCL16. Meanwhile, ZEB1 blockade attenuates CD44+ neutrophil–induced CD8+ T cell exhaustion by reducing tumor-derived SPP1 secretion, which otherwise promotes exhaustion through activation of the PD-L1/PD-1 pathway. Clinically, high ZEB1 expression correlated with chemoresistance, immunosuppression, and diminished CXCL16 levels in patients with PC. Importantly, the epigenetic inhibitor mocetinostat (targeting ZEB1) potentiated the efficacy of chemoimmunotherapy, including anti–PD-1 and CAR T therapies, in patient-derived organoids, xenografts, and orthotopic models. Our study unveils ZEB1 as a master epigenetic regulator of chemoimmunotherapy resistance and proposes its targeting as a transformative strategy for PC treatment.

Authors

Shaobo Zhang, Yumeng Hu, Zhijun Zhou, Gaoyuan Lv, Chenze Zhang, Yuanyuan Guo, Fangxia Wang, Yuxin Ye, Haoran Qi, Hui Zhang, Wenming Wu, Min Li, Mingyang Liu

×

Figure 1

Blocking of Zeb1 enhances gemcitabine efficacy through activation of the PC immune microenvironment.

Options: View larger image (or click on image) Download as PowerPoint
Blocking of Zeb1 enhances gemcitabine efficacy through activation of the...
(A and B) Tumor images and weight of orthotopic allograft mouse model (immunocompetent and immunodeficient) established from KPC-shV and KPC-shZeb1 cells and treated with gemcitabine (GEM; 50 mg/kg) 3 times a week (n = 3). (C and D) Tumor images and weight of orthotopic allograft mouse model established from KPC-shV and KPC-shZeb1 cells in each treatment condition (n = 5). (E) Survival of orthotopic allograft mouse model established from KPC-shV and KPC-shZeb1 cells in each treatment condition (n = 6–10). (F and G) Tumor images and weight of orthotopic allograft mouse model established from KPC-shV and KPC-shZeb1 cells and treated with gemcitabine (50 mg/kg) and anti–PD-1 (10 mg/kg) 3 times a week (n = 6). (H) Uniform manifold approximation and projection (UMAP) plot of scRNA-Seq data derived from orthotopic allograft mouse model reveals the presence of 10 distinct cell types. Cell types are distinguished by color. (I) UMAP plot displays the distribution and subclustering of T and NK cell subsets. TSNE, t-distributed stochastic neighbor embedding. (J) Stacked histogram shows the proportion of each T/NK cluster in KPC-shV and KPC-shZeb1 mouse tumor tissues. (K) Circle plots depict the strength of cell-cell interactions between subclusters of T/NK cells and tumor cells, as identified through CellChat analysis. The edge weights and numerical values indicate the strength score of these interactions, while the direction of the arrows denotes the cell clusters responsible for signaling release and reception. *P < 0.05, **P < 0.01, ***P < 0.001, by unpaired, 2-tailed Student’s t test (B), 1-way ANOVA with Tukey’s multiple-comparison test (D and G), and log-rank test (E). Data are presented as mean ± SD in B, D, and G.