A blueprint for tumor-infiltrating B cells across human cancers

• 成果类型: Article
• 署名作者: Ma, Jiaqiang; Wu, Yingcheng; Ma, Lifeng; Yang, Xupeng; Zhang, Tiancheng; Song, Guohe; Li, Teng; Gao, Ke; Shen, Xia; Lin, Jian; Chen, Yamin; Liu, Xiaoshan; Fu, Yuting; Gu, Xixi; Chen, Zechuan; Jiang, Shan; Rao, Dongning; Pan, Jiaomeng; Zhang, Shu; Zhou, Jian; Huang, Chen; Shi, Si; Fan, Jia; Guo, Guoji; Zhang, Xiaoming; Gao, Qiang
• 署名单位: Fudan University; Fudan University; Zhejiang University; Zhejiang University; Chinese Academy of Sciences; Shanghai Institute of Immunity and Infection, CAS; Shanghai Jiao Tong University; Fudan University
• 刊物名称: SCIENCE
• ISSN/ISSBN: 0036-10409
• DOI: 10.1126/science.adj4857
• 发表日期: 2024-05-03
• 关键词: tertiary lymphoid structures; immune cells; gene-expression; metabolic-control; plasma-cells; t-bet; landscape; microenvironment; memory; deconvolution

摘要

INTRODUCTION Tumor-infiltrating B cells have emerged as important players in cancer immunity and served as predictors of response to immunotherapy. These B cells display multiple functions, primarily through their ability to differentiate into plasma cells to produce antibodies, but vary spatiotemporally across different cancer types. Dissecting the abundance and differentiation states of B cells across diverse cancer types holds promise for improving the immunotherapeutic response. RATIONALE To compile a comprehensive pan-cancer B cell landscape, we performed single-cell RNA sequencing (scRNA-seq) on paired tumors, lymph node metastases, adjacent normal tissues, and peripheral blood from patients with various cancer types, as well as incorporating substantial published scRNA-seq datasets. After correction of the batch effect, this atlas consists of scRNA-seq data from 269 patients across 20 cancer types. We assembled B cell receptor (BCR) sequencing of individual B cells with gene-expression profiles to characterize the dynamic transition between B cells and antibody-secreting cells (ASCs). We integrated the single-cell chromatin accessibility landscape of B cells from different cancers to dissect the epigenomic regulation networks that function in fine-tuning B cell development. We spatially localized B cells in mature versus immature tertiary lymphoid structures (TLSs) and investigated the potential regulators that direct B cells into specific responses. RESULTS We revealed substantial heterogeneity within B and plasma cells, identifying 15 B cell subsets and 10 plasma cell subsets. We computationally derived and validated two independent developmental pathways to ASCs through canonical germinal center (GC) and alternative extrafollicular (EF) pathways and demonstrated an apparent cancer-type preference. Colon adenocarcinoma and liver hepatocellular carcinoma were the two representative types of cancer enriched for GC and EF pathways, respectively. We affirmed that EF-dominant cancers correlate with dysregulated immune responses and worse clinical outcomes. We then identified the dynamic metabolic-epigenetic-signaling networks engaged in fine-tuning tumor-infiltrating B cell differentiation and managing the balance between the EF and GC pathways. Atypical memory (AtM) B cells, the primary progenitors of EF-derived ASCs, exhibit an exhausted and bystander phenotype and develop independently of the GC pathway. We found that the AtM B cells reside in the center of immature TLSs and spatially relocate to the periphery during TLS maturation. Last, we mechanistically linked these findings to specific transcription factors and epigenomic regulations. We demonstrated that the glutamine-derived metabolite alpha-ketoglutarate (alpha-KG) could increase the expression of AtM B cell-associated transcription factors T-bet and BATF and promote their differentiation, accompanied by the activation of mammalian target of rapamycin complex 1 (mTORC1) signaling. Consequently, AtM B cells acquire an immunoregulatory function that dampens antitumor T cell responses and fosters an immunosuppressive microenvironment. CONCLUSION We compiled the blueprint of B cell heterogeneity and two dynamic differentiation pathways in human cancers, providing a fundamental reference of ASC differentiation trajectory for future studies. The systematic comparison between EF and GC pathways reveals the similarities and differences of B cell states across different cancer types, highlighting the unfavorable clinical outcome linked to the immunosuppressive microenvironment of EF pathway-associated AtM B cells. Metabolic-epigenetic networks are remarkably flexible and can reconfigure B cell fates in a way that will facilitate the development of B cell-targeted immunotherapies. Systematic analysis of a human pan-cancer B cell atlas. We analyzed 474,718 B cells from 269 patients across 20 cancer types using single-cell sequencing data. By combining gene expression profiles, BCR sequences, and chromatin accessibility, we investigated the diversity and plasticity of tumor-infiltrating B cells and performed a multilevel comparison of EF- and GC-responsive plasma cells among cancer types. We visualized their dynamic spatial locations along the maturation of TLSs and identified potential metabolic-epigenetic mechanisms in regulating B cell differentiation.