henhaobio
Sign in

bioRxivBioinformaticsPreprint

Integrated analyses of single-cell atlases reveal age, gender, and smoking status associations with cell type-specific expression of mediators of SARS-CoV-2 viral entry and highlights inflammatory programs in putative target cells

Christoph MuusiD, Malte D. LueckeniD, Gökçen EraslaniD, Avinash WaghrayiD, Graham HeimbergiD, Lisa SikkemaiD, Yoshihiko KobayashiiD, Eeshit Dhaval VaishnaviD, Ayshwarya SubramanianiD, Christopher Smilie, Karthik A. JagadeeshiD, Thu Elizabeth DuongiD, Evgenij FiškiniD, Elena Torlai TrigliaiD, Meshal AnsariiD, Peiwen CaiiD, Brian LiniD, Justin BuchananiD, Sijia CheniD, Jian ShuiD, Adam L. HaberiD, Hattie ChungiD, Daniel T. MontoroiD, Taylor Sterling AdamsiD, Hananeh AlieeiD, , Wenjun Yan

PDFDOI
In-depth full-text digest available · Jump to digest ↓

Abstract

The COVID-19 pandemic, caused by the novel coronavirus SARS-CoV-2, creates an urgent need for identifying molecular mechanisms that mediate viral entry, propagation, and tissue pathology. Cell membrane bound angiotensin-converting enzyme 2 (ACE2) and associated proteases, transmembrane protease serine 2 (TMPRSS2) and Cathepsin L (CTSL), were previously identified as mediators of SARS-CoV2 cellular entry. Here, we assess the cell type-specific RNA expression of ACE2, TMPRSS2, and CTSL through an integrated analysis of 107 single-cell and single-nucleus RNA-Seq studies, including 22 lung and airways datasets (16 unpublished), and 85 datasets from other diverse organs. Joint expression of ACE2 and the accessory proteases identifies specific subsets of respiratory epithelial cells as putative targets of viral infection in the nasal passages, airways, and alveoli. Cells that co-express ACE2 and proteases are also identified in cells from other organs, some of which have been associated with COVID-19 transmission or pathology, including gut enterocytes, corneal epithelial cells, cardiomyocytes, heart pericytes, olfactory sustentacular cells, and renal epithelial cells. Performing the first meta-analyses of scRNA-seq studies, we analyzed 1,176,683 cells from 282 nasal, airway, and lung parenchyma samples from 164 donors spanning fetal, childhood, adult, and elderly age groups, associate increased levels of ACE2, TMPRSS2, and CTSL in specific cell types with increasing age, male gender, and smoking, all of which are epidemiologically linked to COVID-19 susceptibility and outcomes. Notably, there was a particularly low expression of ACE2 in the few young pediatric samples in the analysis. Further analysis reveals a gene expression program shared by ACE2+TMPRSS2+ cells in nasal, lung and gut tissues, including genes that may mediate viral entry, subtend key immune functions, and mediate epithelial-macrophage cross-talk. Amongst these are IL6, its receptor and co-receptor, IL1R, TNF response pathways, and complement genes. Cell type specificity in the lung and airways and smoking effects were conserved in mice. Our analyses suggest that differences in the cell type-specific expression of mediators of SARS-CoV-2 viral entry may be responsible for aspects of COVID-19 epidemiology and clinical course, and point to putative molecular pathways involved in disease susceptibility and pathogenesis.

In-depth summary

Claude Opus 5.5 · xhigh effort · 17 Oct 2026 · from the full text of version 2 (21 Apr 2020) · CC BY-ND 4.0

Across 1,176,683 airway and lung cells from 22 datasets, ACE2 rises with age, in males and in smokers' multiciliated cells, and is lowest in young children.

Age, male sex and smoking track with COVID-19 severity, but no single scRNA-seq study is large enough to test whether they also track with expression of the viral entry machinery in specific cell types. Pooling 107 single-cell and single-nucleus studies maps ACE2, TMPRSS2 and CTSL co-expression across organs and yields the gene programs those cells run, while 22 nasal, airway and lung datasets support what the authors call the first single-cell meta-analysis relating expression to donor covariates.

Key results

  1. ShownIn airway and alveolar epithelium, ACE2 expression goes up with age, is higher in males, and is strongly higher in smokers, with smoking the largest effect.

    Poisson models on 761,693 cells (165 samples, 77 donors) gave robust trends: ACE2 up with age in basal and multiciliated cells, up in males in secretory and AT2 cells, and up in past or current smokers in multiciliated cells (log FC 1.52). Multiciliated coefficients: 1.52 smoking, 0.51 sex, 0.0107 age; a non-robust indication put the largest age slope, 0.020 per year, in AT2. FDR 5%, pseudo-bulk and holdout checks. · ACE2 and TMPRSS2 expression in airway epithelial…; Integrated analysis for association of ACE2 and…

  2. ShownCells co-expressing ACE2 with TMPRSS2 or CTSL are concentrated in nasal goblet and multiciliated cells, alveolar AT2 cells and gut enterocytes, but also appear in organs linked to extrapulmonary disease.

    Across 92 human studies, dual-positive cells appeared in cornea and conjunctiva, olfactory sustentacular cells, cardiomyocytes, heart pericytes (ACE2 in 30% in one heart dataset versus at most 21.7% in kidney), renal tubule, liver, pancreas and brain oligodendrocytes. scATAC-seq agreed: ACE2 accessible in 10.8% (364 of 3,371) of AT2 cells versus 6.2% (199 of 3,215) ACE2+ by scRNA-seq in the same patient. · A cross-tissue survey identifies dual-positive…; Validation of ACE2, TMPRSS2 and CTSL mRNA and…; Integrated co-expression analysis of high…

  3. SuggestedExpression of the entry machinery is very low in lungs of newborns and young children, offering a candidate correlate of milder paediatric disease.

    ACE2 was high in third-trimester preterm lung (dual-positive AT2 cells) but very low in newborns and in roughly 3-month-old and 3-year-old lungs; scTHS-seq at day 1, 14 months, 3 and 9 years (one donor each) showed no accessibility at birth. In placenta ACE2 reached 1.4% of cells, TMPRSS2 only 0.2%, CTSL 56%, and ACE2+CTSL+ cells 1.3%. · Distinctive changes in ACE2+TMPRSS2+ expression…

  4. ShownDual-positive cells in nasal, lung and gut tissue share an inflammatory and antiviral gene program, including interleukin-6 signalling, tumour necrosis factor response genes and complement.

    Random forest tissue programs and per-cell-type differential expression recovered CEACAM5, SLPI, PIGR, MHC class I and II genes, complement C3 and C4BPA, and TNF-pathway genes (RIPK3, BIRC3, CCL20, JUN, NFKB1); lung and gut programs include IL6ST, the AT2 program IL6 and IL1R1. Enterocyte ACE2 was reported lower with anti-TNF treatment in ulcerative colitis (means 0.22 and 0.13 log2(TP10K+1), adjusted P < 1e-3). · An immune gene program associated with…; Identification of gene programs using feature…; Identification of gene programs enriched in DP…

  5. ShownMouse experiments reproduce both the cell-type pattern and the smoking effect seen in human data, making mice usable for follow-up on expression if not on binding.

    In healthy mouse airway, Ace2+Tmprss2+ cells were mainly club (5.1%) and multiciliated (2.8%) cells, Ace2+Ctsl+ club cells 11.1%; Ace2 rose over a 2-month ageing course in club (p=1.16e-06) and goblet cells (p=0.01911). Two months of daily cigarette smoke raised Ace2 in airway secretory but not AT2 cells, matching the human contrast, with a dose-dependent rise over 5 months by bulk RNA-seq. · Conserved expression patterns in mouse models

Methods, data and reproducibility

Published and unpublished single-cell and single-nucleus RNA-seq datasets were harmonised to a shared cell-type ontology and scored for ACE2, TMPRSS2 and CTSL co-expression; Poisson regression with interaction terms tested age, sex and smoking per cell type, backed by pseudo-bulk and holdout checks, scATAC-seq, in situ hybridisation, immunohistochemistry and mouse smoke exposure.

  • Integrated survey of single-cell and single-nucleus RNA-seq studies of lung, airways and other organs · human · n = 107 studies: 22 lung/airway (16 unpublished) and 85 other
  • Harmonised nasal, airway and lung meta-analysis cohort for covariate modelling · human · n = 1,176,683 cells, 282 samples, 164 donors
  • GEO-derived published 10x datasets used for the broad co-expression survey · human · n = 2,333,199 cells from 469 samples in 64 datasets
  • Compendium with high-resolution annotations across tissues · human · n = 2,433,890 cells from 21 tissues or organs
  • Developmental and paediatric lung samples, plus scTHS-seq chromatin accessibility at four ages · human · n = 12 first-trimester, 29 second-trimester, 13 postnatal samples
  • scATAC-seq of primary carina and subpleural parenchyma · human · n = 3 samples per location from 1 patient
  • Mouse airway and lung scRNA-seq for ageing and cigarette-smoke exposure, plus sorted-population and ENCODE bulk RNA-seq · mouse · n = 2-month ageing course; 2 months daily smoke exposure
  • Declined-donor transplant lung dataset used for protease co-expression and cell-cell interactions · human · n = 41 samples, 10 patients, 2-6 locations each
Methods & tools
Scanpy, diffxpy, statsmodels, Harmony, Signac and ChromVAR, Seurat, CellRanger ATAC, MAST, scikit-learn random forest, CellphoneDB, lme4, gprofiler and EnrichR
Reagents & environment
Package versions are given throughout: Scanpy 1.4.5 and 1.4.6, harmony-pytorch 0.1.1, diffxpy 0.7.3 with batchglm 0.7.4, statsmodels 0.11.1, scikit-learn 0.21.3, Signac v0.1.6, Seurat 3.1.1, CellRanger ATAC 1.1.0, MAST 1.8.2, lme4 1.1.21, CellphoneDB v2.0.0, louvain 0.6.1, networkx 2.2.
Reproducibility gaps
The code statement points to a Terra workspace and the data statement to the Single Cell Portal, but both URLs are missing here. For unpublished datasets only counts for the three genes, total UMIs, annotations and limited metadata were shared.
Limitations (6)
  • The regression treats each cell as an independent observation, so p-values can be inflated where few donors contribute most cells of a type. · stated by the authors
  • Several associations are only indications: they disappear when one large contributing dataset is held out, and only one dataset had enough donors across age, sex and smoking to fit the full model. · stated by the authors
  • Developmental and paediatric conclusions rest on very few samples, often one donor per age, so the authors ask that they be read with caution. · stated by the authors
  • The anti-TNF association in enterocytes could not be adjusted for disease severity, which is itself strongly linked to treatment. · stated by the authors
  • Three ACE2 antibodies gave inconsistent staining, so some are probably non-specific, and harmonised annotations still aggregate heterogeneous subtypes. · stated by the authors
  • Everything is measured in healthy or non-infected tissue at the RNA level; no infection or viral entry is tested, so target cells remain putative. · noted in this summary

Read this if you work on respiratory viral entry, want cell-type-resolved ACE2 and protease expression across human organs, or need a worked example of meta-analysis across single-cell datasets.

Citations
236

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISYin '20Radzikowska '20Aguiar '20Inde '20Han '20Barker '20
81% semantic matchPathology
J. Yin, B. Kasper, F. Petersen+1 more·bioRxiv·Aug 6, 2020