Background: Lung damage in severe COVID-19 is highly heterogeneous however studies with dedicated spatial distinction of discrete temporal phases of diffuse alveolar damage (DAD) and alternate lung injury patterns are lacking. Existing studies have also not accounted for progressive airspace obliteration in cellularity estimates. We used an imaging mass cytometry (IMC) analysis with a novel airspace correction step to more accurately identify the cellular immune response that underpins the heterogeneity of severe COVID-19 lung disease. Methods: Lung tissue was obtained at post-mortem from severe COVID-19 deaths. Pathologist-selected regions of interest (ROIs) were chosen by light microscopy representing the patho-evolutionary spectrum of DAD and alternate disease phenotypes were selected for comparison. Architecturally normal SARS-CoV-2-positive lung tissue and tissue from SARS-CoV-2-negative donors served as controls. ROIs were stained for 40 cellular protein markers and ablated using IMC before segmented cells were classified. Cell populations corrected by ROI airspace and their spatial relationships were compared across lung injury patterns. Results: Forty patients (32M:8F, age:22-98), 345 ROIs and >900k single cells were analysed. DAD progression was marked by airspace obliteration and significant increases in mononuclear phagocytes (MnPs), T and B lymphocytes and significant decreases in alveolar epithelial and endothelial cells. Neutrophil populations proved stable overall although several interferon-responding subsets demonstrated expansion. Spatial analysis revealed immune cell interactions occur prior to microscopically appreciable tissue injury. Conclusions: The immunopathogenesis of severe DAD in COVID-19 lung disease is characterised by sustained increases in MnPs and lymphocytes with key interactions occurring even prior to lung injury is established.

Diffuse alveolar damage (DAD) is the histological expression of acute respiratory distress syndrome and characterises lung pathology due to infection with SARS-CoV-2, and other respiratory pathogens of clinical significance. DAD reflects a time-dependent immunopathological process, progressing from an early/exudative stage through to an organising/fibrotic stage, yet within an individual these different stages of DAD may coexist. Understanding the progression of DAD is central to the development of new therapeutics to limit progressive lung damage. Here, we applied highly multiplexed spatial protein profiling to autopsy lung tissues derived from 27 patients who died from COVID-19 and identified a protein signature (ARG1, CD127, GZMB, IDO1, Ki67, phospho-PRAS40 (T246) and VISTA) that distinguishes early DAD from late DAD with good predictive accuracy. These proteins warrant further investigation as potential regulators of DAD progression.

Background Characterising and quantifying cell types within glioblastoma (GBM) tumours at scale will facilitate a better understanding of the association between the cellular landscape and tumour phenotypes or clinical correlates. We aimed to develop a tool that deconvolutes immune and neoplastic cells within the GBM tumour microenvironment from bulk RNA sequencing data. Methods We developed an IDH wild-type (IDHwt) GBM-specific single immune cell reference consisting of B cells, T cells, NK cells, microglia, tumour associated macrophages, monocytes, mast and DC cells. We used this alongside an existing neoplastic single cell-type reference for astrocyte-like, oligodendrocyte- and neuronal-progenitor like and mesenchymal GBM cancer cells to create both marker and gene signature matrix-based deconvolution tools. We applied single-cell resolution imaging mass cytometry (IMC) to ten IDHwt GBM samples, five paired primary and recurrent tumours, to determine which deconvolution approach performed best. Results Marker based deconvolution using GBM tissue specific markers was most accurate for both immune cells and cancer cells, so we packaged this approach as GBMdeconvoluteR. We applied GBMdeconvoluteR to bulk GBM RNAseq data from The Cancer Genome Atlas and recapitulated recent findings from multi-omics single cell studies with regards associations between mesenchymal GBM cancer cells and both lymphoid and myeloid cells. Furthermore, we expanded upon this to show that these associations are stronger in patients with worse prognosis. Conclusions GBMdeconvoluteR accurately quantifies immune and neoplastic cell proportions in IDHwt GBM bulk RNA sequencing data and is accessible here: https : // gbmdeconvoluter.leeds.ac.uk

Analysis of Imaging Mass Cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single cell segmentation and sub-optimal approaches for data visualisation and exploration. This can lead to inaccurate identification of cell phenotypes, states or spatial relationships compared to reference data from single cell suspension technologies. To this end we have developed the "OPTIMAL" framework to determine the best approaches for cell segmentation, parameter transformation, batch effect correction, data visualisation/clustering and spatial neighbourhood analysis. Using a panel of 27 metal-tagged antibodies recognising well characterised phenotypic and functional markers to stain the same FFPE human tonsil sample Tissue Microarray (TMA) over 12 temporally distinct batches we tested a total of four cell segmentation models, a range of different arcsinh cofactor parameter transformation values, five different dimensionality reduction algorithms and two clustering methods. Finally we assessed the optimal approach for performing neighbourhood analysis. We found that single cell segmentation was improved by the use of an Ilastik-derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using "classical" bi-variate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximised the statistical separation between negative and positive signal distributions and a simple Z-score normalisation step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out-performing Phenograph in terms of cell type identification. We also found that neighbourhood analysis was influenced by the method used for finding neighbouring cells with a "disc" pixel expansion outperforming a "bounding box" approach combined with the need for filtering objects based on size and image-edge location. Importantly OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates .FCS files from the segmentation output, allows for single cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists.

T cells play key protective but also pathogenic roles in COVID-19. We studied expression of long non-coding RNAs (lncRNAs) in COVID-19 T cell transcriptomes by integrating previously published single-cell RNA sequencing datasets. The long intergenic non-coding RNA MALAT1 was the most highly transcribed lncRNA in T cells, with Th1 cells demonstrating the lowest and CD8+ resident memory cells the highest MALAT1 expression, amongst CD4+ and CD8+ T cells populations, respectively. We then identified gene signatures that covaried with MALAT1 in single T cells. A significantly higher number of transcripts correlated negatively with MALAT1 than those that correlated. Enriched functional annotations of the MALAT1- anti-correlating gene signature included processes associated with T cell activation such as cell division, oxidative phosphorylation and response to cytokine. The MALAT1 anti-correlating gene signature shared by both CD4+ and CD8+ T cells marked dividing T cells in both lung and blood of COVID-19 patients. Focussing on the tissue, we used an independent patient cohort of post-mortem COVID-19 lung samples and demonstrated that MALAT1 suppression was indeed a marker of MKI67+ proliferating CD8+ T cells. Our results reveal MALAT1 suppression and its associated gene signature are a hallmark of human proliferating T cells.