Search

CD73-derived adenosine suppresses anti-cancer immunity and CD73 inhibitors are currently evaluated in several clinical trials. Here we have assessed enzyme kinetics of all key purinergic ectoenzymes in five cancer cell lines (Hodgkin lymphoma, multiple myeloma, pancreas adenocarcinoma, urinary bladder carcinoma, glioblastoma) under normoxia and hypoxia. We found that adenosine metabolism varied considerably between individual cancer types. All cell lines investigated exhibited high ecto-adenosine deaminase (ADA) activity, which critically influenced the kinetics of adenosine accumulation. Combining kinetics data with single-cell RNA sequencing data on myeloma and glioblastoma cancerous tissue revealed, that purine metabolism is not homogeneously organized, but differs in a cancer type-specific fashion between malignant cells, stromal cells, and immune cells. Since purine metabolism in cancerous tissue is most likely spatially heterogeneous and differs between the various cell types, diffusion distances in the microenvironment as well as ADA activity may be important variables that influence the level of bioactive adenosine.

Dendritic cells (DC) have the unique capacities to induce primary T-cell responses. In mice, CD8α+DC are specialized to cross-prime CD8+T cells and produce interleukin-12 (IL-12) that promotes cytotoxicity. Human BDCA-3+DC share several relevant characteristics with CD8α+DC, but the capacities of human DC subsets to induce CD8+T-cell responses are incompletely understood. Here we compared CD1c+myeloid DC (mDC)1, BDCA-3+mDC2, and plasmacytoid DC (pDC) in peripheral blood and lymphoid tissues for phenotype, cytokine production, and their capacities to prime cytotoxic T cells. mDC1 were surprisingly the only human DC that secreted high amounts of IL-12p70, but they required combinational Toll-like receptor (TLR) stimulation. mDC2 and pDC produced interferon-λ and interferon-α, respectively. Importantly, mDC1 and mDC2 required different combinations of TLR ligands to cross-present protein antigens to CD8+T cells. pDC were inefficient and also expressed lower levels of major histocompatibility complex and co-stimulatory molecules. Nevertheless, all DC induced CD8+memory T-cell expansions upon licensing by CD4+T cells, and primed naive CD8+T cells following appropriate TLR stimulation. However, because mDC1 produced IL-12, they induced the highest levels of cytotoxic molecules. In conclusion, CD1c+mDC1 are the relevant source of IL-12 for naive T cells and are fully equipped to cross-prime cytotoxic T-cell responses.

CCR6 is a chemokine receptor expressed on Th17 cells and regulatory T cells that is induced by T-cell priming with certain cytokines, but how its expression and stability are regulated at the molecular level is largely unknown. Here, we identified and characterized a noncoding region of the human CCR6locus that displayed unmethylated CpG motifs (differentially methylated region [DMR]) selectively in CCR6+lymphocytes. CCR6 expression on circulating CD4+T cells was stable on cytokine-induced proliferation but partially down-regulated on T-cell receptor stimulation. However, CCR6 down-regulation was mostly transient, and the DMR within the CCR6locus remained demethylated. Notably, in vitro induction of CCR6 expression with cytokines in T-cell receptor-activated naive CD4+T cells was not associated with a demethylated DMR and resulted in unstable CCR6 expression. Conversely, treatment with the DNA methylation inhibitor 5′-azacytidine induced demethylation of the DMR and led to increased and stable CCR6 expression. Finally, when cloned into a reporter gene plasmid, the DMR displayed transcriptional activity in memory T cells that was suppressed by DNA methylation. In summary, we have identified a noncoding region of the human CCR6gene with methylation-sensitive transcriptional activity in CCR6+T cells that controls stable CCR6 expression via epigenetic mechanisms.