GM-CSF-activated human dendritic cells promote type 1 T follicular helper cell polarization in a CD40-dependent manner.

T follicular helper (Tfh) cells regulate humoral responses and present a marked phenotypic and functional diversity. Type 1 Tfh (Tfh1) cells were recently identified and associated with disease severity in infection and autoimmune diseases. The cellular and molecular requirements to induce human Tfh1 differentiation are not known. Here, using single-cell RNA sequencing (scRNAseq) and protein validation, we report that human blood CD1c+ dendritic cells (DCs) activated by GM-CSF (also known as CSF2) drive the differentiation of naive CD4+ T cells into Tfh1 cells. These Tfh1 cells displayed typical Tfh molecular features, including high levels of PD-1 (encoded by PDCD1), CXCR5 and ICOS. They co-expressed BCL6 and TBET (encoded by TBX21), and secreted large amounts of IL-21 and IFN-γ (encoded by IFNG). Mechanistically, GM-CSF triggered the emergence of two DC sub-populations defined by their expression of CD40 and ICOS ligand (ICOS-L), presenting distinct phenotypes, morphologies, transcriptomic signatures and functions. CD40High ICOS-LLow DCs efficiently induced Tfh1 differentiation in a CD40-dependent manner. In patients with mild COVID-19 or latent Mycobacterium tuberculosis infection, Tfh1 cells were positively correlated with a CD40High ICOS-LLow DC signature in scRNAseq of peripheral blood mononuclear cells or blood transcriptomics, respectively. Our study uncovered a novel CD40-dependent Tfh1 axis with potential physiopathological relevance to infection. This article has an associated First Person interview with the first author of the paper.

Hubness reduction improves clustering and trajectory inference in single-cell transcriptomic data.

MOTIVATION: Single-cell RNA-seq (scRNAseq) datasets are characterized by large ambient dimensionality, and their analyses can be affected by various manifestations of the dimensionality curse. One of these manifestations is the hubness phenomenon, i.e. existence of data points with surprisingly large incoming connectivity degree in the datapoint neighbourhood graph. Conventional approach to dampen the unwanted effects of high dimension consists in applying drastic dimensionality reduction. It remains unexplored if this step can be avoided thus retaining more information than contained in the low-dimensional projections, by correcting directly hubness. RESULTS: We investigated hubness in scRNAseq data. We show that hub cells do not represent any visible technical or biological bias. The effect of various hubness reduction methods is investigated with respect to the clustering, trajectory inference and visualization tasks in scRNAseq datasets. We show that hubness reduction generates neighbourhood graphs with properties more suitable for applying machine learning methods; and that it outperforms other state-of-the-art methods for improving neighbourhood graphs. As a consequence, clustering, trajectory inference and visualization perform better, especially for datasets characterized by large intrinsic dimensionality. Hubness is an important phenomenon characterizing data point neighbourhood graphs computed for various types of sequencing datasets. Reducing hubness can be beneficial for the analysis of scRNAseq data with large intrinsic dimensionality in which case it can be an alternative to drastic dimensionality reduction. AVAILABILITY AND IMPLEMENTATION: The code used to analyze the datasets and produce the figures of this article is available from https://github.com/sysbio-curie/schubness. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Meta-Analysis of Human Cancer Single-Cell RNA-Seq Datasets Using the IMMUcan Database.

The development of single-cell RNA sequencing (scRNA-seq) technologies has greatly contributed to deciphering the tumor microenvironment (TME). An enormous amount of independent scRNA-seq studies have been published representing a valuable resource that provides opportunities for meta-analysis studies. However, the massive amount of biological information, the marked heterogeneity and variability between studies, and the technical challenges in processing heterogeneous datasets create major bottlenecks for the full exploitation of scRNA-seq data. We have developed IMMUcan scDB (https://immucanscdb.vital-it.ch), a fully integrated scRNA-seq database exclusively dedicated to human cancer and accessible to nonspecialists. IMMUcan scDB encompasses 144 datasets on 56 different cancer types, annotated in 50 fields containing precise clinical, technological, and biological information. A data processing pipeline was developed and organized in four steps: (i) data collection; (ii) data processing (quality control and sample integration); (iii) supervised cell annotation with a cell ontology classifier of the TME; and (iv) interface to analyze TME in a cancer type-specific or global manner. This framework was used to explore datasets across tumor locations in a gene-centric (CXCL13) and cell-centric (B cells) manner as well as to conduct meta-analysis studies such as ranking immune cell types and genes correlated to malignant transformation. This integrated, freely accessible, and user-friendly resource represents an unprecedented level of detailed annotation, offering vast possibilities for downstream exploitation of human cancer scRNA-seq data for discovery and validation studies. SIGNIFICANCE: The IMMUcan scDB database is an accessible supportive tool to analyze and decipher tumor-associated single-cell RNA sequencing data, allowing researchers to maximally use this data to provide new insights into cancer biology.

Large-scale analysis of cell-cell communication reveals angiogenin-dependent tumor progression in clear cell renal cell carcinoma.

Cellular crosstalk in the tumor microenvironment (TME) is still largely uncharacterized, while it plays an essential role in shaping immunosuppression or anti-tumor response. Large-scale analyses are needed to better decipher cell-cell communication in cancer. In this work, we used original and publicly available single-cell RNA sequencing (scRNAseq) data to characterize in-depth the communication networks in human clear cell renal cell carcinoma (ccRCC). We identified 50 putative communication channels specifically used by cancer cells to interact with other cells, including two novel angiogenin-mediated interactions. Expression of angiogenin and its receptors was validated at the protein level in primary ccRCC. Mechanistically, angiogenin enhanced ccRCC cell line proliferation and down-regulated secretion of IL-6, IL-8, and MCP-1 proinflammatory molecules. This study provides novel biological insights into molecular mechanisms of ccRCC, and suggests angiogenin and its receptors as potential therapeutic targets in clear cell renal cancer.

Context-Dependent Effects Explain Divergent Prognostic Roles of Tregs in Cancer.

Assessing cancer prognosis is a challenging task, given the heterogeneity of the disease. Multiple features (clinical, environmental, genetic) have been used for such assessments. The tumor immune microenvironment (TIME) is a key feature, and describing the impact of its many components on cancer prognosis is an active field of research. The complexity of the tumor microenvironment context makes it difficult to use the TIME to assess prognosis, as demonstrated by the example of regulatory T cells (Tregs). The effect of Tregs on prognosis is ambiguous, with different studies considering them to be negative, positive or neutral. We focused on five different cancer types (breast, colorectal, gastric, lung and ovarian). We clarified the definition of Tregs and their utility for assessing cancer prognosis by taking the context into account via the following parameters: the Treg subset, the anatomical location of these cells, and the neighboring cells. With a meta-analysis on these three parameters, we were able to clarify the prognostic role of Tregs. We found that CD45RO+ Tregs had a reproducible negative effect on prognosis across cancer types, and we gained insight into the contributions of the anatomical location of Tregs and of their neighboring cells on their prognostic value. Our results suggest that Tregs play a similar prognostic role in all cancer types. We also establish guidelines for improving the design of future studies addressing the pathophysiological role of Tregs in cancer.

Single-cell RNA sequencing of blood antigen-presenting cells in severe COVID-19 reveals multi-process defects in antiviral immunity.

COVID-19 can lead to life-threatening respiratory failure, with increased inflammatory mediators and viral load. Here, we perform single-cell RNA-sequencing to establish a high-resolution map of blood antigen-presenting cells (APCs) in 15 patients with moderate or severe COVID-19 pneumonia, at day 1 and day 4 post admission to intensive care unit or pulmonology department, as well as in 4 healthy donors. We generated a unique dataset of 81,643 APCs, including monocytes and rare dendritic cell (DC) subsets. We uncovered multi-process defects in antiviral immune defence in specific APCs from patients with severe disease: (1) increased pro-apoptotic pathways in plasmacytoid DCs (pDCs, key effectors of antiviral immunity), (2) a decrease of the innate sensors TLR9 and DHX36 in pDCs and CLEC9a+ DCs, respectively, (3) downregulation of antiviral interferon-stimulated genes in monocyte subsets and (4) a decrease of major histocompatibility complex (MHC) class II-related genes and MHC class II transactivator activity in cDC1c+ DCs, suggesting viral inhibition of antigen presentation. These novel mechanisms may explain patient aggravation and suggest strategies to restore the defective immune defence.