key: cord-0777748-8qykn0nr
authors: Xu, G.; Qi, F.; Li, H.; Yang, Q.; Wang, H.; Wang, X.; Liu, X.; Zhao, J.; Liao, X.; Liu, Y.; Amit, I.; Liu, L.; Zhang, S.; Zhang, Z.
title: The differential immune responses to COVID-19 in peripheral and lung revealed by single-cell RNA sequencing
date: 2020-08-17
journal: nan
DOI: 10.1101/2020.08.15.20175638
sha: 6534983caa7929d993d00f4b932394737fa41d5b
doc_id: 777748
cord_uid: 8qykn0nr

Understanding the mechanism that leads to immune dysfunction induced by SARS-CoV2 virus is crucial to develop treatment for severe COVID-19. Here, using single cell RNA-seq, we characterized the peripheral blood mononuclear cells (PBMC) from uninfected controls and COVID-19 patients, and cells in paired broncho-alveolar lavage fluid (BALF). We found a close association of decreased dendritic cells (DC) and increased monocytes resembling myeloid-derived suppressor cells (MDSC) which correlated with lymphopenia and inflammation in the blood of severe COVID-19 patients. Those MDSC-like monocytes were immune-paralyzed. In contrast, monocyte-macrophages in BALFs of COVID-19 patients produced massive amounts of cytokines and chemokines, but secreted little interferons. The frequencies of peripheral T cells and NK cells were significantly decreased in severe COVID-19 patients, especially for innate-like T and various CD8+ T cell subsets, compared to health controls. In contrast, the proportions of various activated CD4+ T cell subsets, including Th1, Th2 and Th17-like cells were increased and more clonally expanded in severe COVID-19 patients. Patients' peripheral T cells showed no sign of exhaustion or augmented cell death, whereas T cells in BALFs produced higher levels of IFNG, TNF, CCL4 and CCL5 etc. Paired TCR tracking indicated abundant recruitment of peripheral T cells to the patients' lung. Together, this study comprehensively depicts how the immune cell landscape is perturbed in severe COVID-19.

samples from 2 mild and 5 severe patients. The integration analysis of BALF and circulating myeloid 131 cells showed clusters of neutrophil (FCGR3B), mDC (CD1C), monocyte-macrophages (CD14, 132 FCGR3A and CD68) ( Figure 3A and Figure S3A ). Macrophage subset classification markers, 133 including FCN1, SPP1 and FABP4, were differentially expressed by circulating and BALF monocyte-134 macrophages from patients with mild or severe COVID-19( Figure 3B ). Analysis of differentiation 135 trajectory of circulating and BALF monocyte-macrophages from the same patient revealed a consensus 136 blood-toward-BALF course ( Figure 3C and Figure S3B ), consistent with the recruitment to peripheral 137 monocytes into inflammatory tissues as expected. 138 Next, we performed transcriptome analysis of circulating and BALF monocyte-macrophages to 139 understand their functional status. Among the DEGs, there were 524 shared upregulated genes and 140 501 downregulated genes in BALF monocyte-macrophages versus those in blood, identified from both 141 mild and severe COVID-19 patients ( Figure 3D and Table S2 ). Such a large number of DEGs 142 suggested significant difference existed between the peripheral and lung monocyte-macrophages. 143 Indeed, the GO analysis revealed broad activation of multiple immune pathways in BALF monocyte-144 macrophages, including response to IFNs and cytokines, neutrophil activation and leukocyte migration, 145 while the pathway involved in myeloid cell differentiation, ATP metabolism etc. were enriched in 146 blood monocytes ( Figure 3E ). In addition, these comparisons revealed perturbed pathways in BALF 147 monocyte-macrophages relevant to severe COVID-19. e.g. responses to hypoxia, high temperature, 148 metal ion, wounding and Fc receptor signaling pathways were specially upregulated ( Figure 3E ), while 149 pathways related to alveoli macrophage functions were downregulated, including lipid metabolism, 150 apoptotic cell clearance and antigen presentation ( Figure 3E ). The representative DEGs involved in 151 those pathways were shown in Figure S3C . 152 Monocyte-macrophages were thought to play key roles in driving the cytokine storm underlying the 153 development of severe COVID-19 25 . Therefore, we examined the cytokine and chemokine levels in 154 monocyte-macrophages in paired blood and BALF samples from the same patient. We found that all 155 types of IFNs (IFNAs, IFNB, IFNG and IFNLs) were minimally expressed by monocyte-macrophages, 156 whereas cytokines (IL1A, IL1B, IL1R2, IL1RN, IL18, IL6, TNF, IL10 and TGFB1) and multiple paired blood samples ( Figure 3F ). Intriguingly, we observed significantly higher levels of antiinflammatory cytokines (IL1R2, IL1RN and TGFB1) and lower levels of IL18 in BALF cells from 160 severe COVID-19 than mild cases, whereas classical pro-inflammatory cytokines (IL1A, IL1B, IL6 161 and TNF) were comparable between the two groups ( Figure 3F ). In contrast, as shown in our earlier 162 studies, monocytes-and neutrophils-recruiting chemokines (CCL2, CCL3, CCL4, CCL7, CCL8, 163 CXCL1, CXCL2, CXCL3 and CXCL8) recruiting monocytes and neutrophils were highly expressed, 164 whereas T cell recruiting chemokines (CXCL9 and CXCL16) recruiting T cells were less expressed by 165 monocyte-macrophages in BALFs of severe COVID-19 monocyte-macrophages than those in mild 166 cases ( Figure 3F ). The higher levels of cytokines (IL-1β and IL-6 etc.) and IL-8 in BALFs than paired 167 plasma was further validated at the protein levels, particularly exemplified by the extremely high levels 168 of IL-8 in BALFs ( Figure 3G ). Thus, these paired analyses revealed a restricted involvement of tissue 169 monocyte-macrophages in cytokine storms during severe COVID-19, through producing chemokines 170 and recruiting more monocytes and neutrophils, but unlikely attribute to increased production of pro-171 inflammatory cytokines. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. . https://doi.org/10.1101/2020.08. 15.20175638 doi: medRxiv preprint Cell density UMAP projections revealed an obviously perturbed T cell landscape in severe COVID- 185 19 compared to mild COVID and control groups ( Figure 4C ). Percentages of innate-like T cells, 186 including MAIT and NKT cells were significantly lower in severe COVID-19 than those in mild 187 COVID-19. Percentages of CD8-Naive, CD8-GZMK and CD8-GZMB subsets were also lower in 188 severe COVID-19 patients than those in mild cases, although the difference in CD8-GZMB 189 comparison was not statistically significant ( Figure 4D ). In contrast, the percentages of several CD4 +

T cell subsets, including CD4-Naive, CD4-LTB, CD4-ICOS, Treg-CTLA4, as well as cycling T cells, 191 were significantly increased in severe COVID-19 than those in mild COVID-19. The percentages of 192 CD4-GATA3 and CD4-CCR6 also showed an increasing trend in severe COVID-19 ( Figure 4D ). In Figure S4C and Table S3 ). There was no evidence of T cell exhaustion, 202 activation of cell death pathways and cytokine productions in those cells from COVID-19 patients, 203 although the pathways related to virus infection and IFNs responses were identified ( Figure S4D ).

Similar findings were also noticed by other groups 16 ; thus, exhaustion and cell death are unlikely the 205 major causes for T cell loss during COVID-19.

The scRNA-seq has been recently been applied to study host immune response in COVID-19 by us 239 and others 8, 9, 13, [17] [18] [19] . Although these studies helped to reveal several aspects of the COVID-19 240 pathogenesis, a complete picture has yet been generated. Here, we integrated scRNA-seq analysis of Previously, several studies have reported that MHC class II molecules were downregulated in blood 256 monocytes 4, 14, 18 in patients with severe COVID-19, however, there was still no consensus on the 257 upregulated genes. We found here that genes related to neutrophil activation, including S100A8, 258 S100A9 and S100A12, were expressed at higher levels in severe COVID-19 patients than those in mild 259 cases. Interestingly, these upregulated and downregulated genes marking monocytic MDSCs whose 260 frequencies are known to increase during various inflammatory conditions 24, 25 . Consistent with this 261 immune suppressive status, we also noticed that genes related to cytokine and IFN production were perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. . https://doi.org/10.1101/2020.08.15.20175638 doi: medRxiv preprint associated with lymphopenia and inflammation markers, we speculate a crucial role of MDSCs in 266 dampening immune response and amplifying COVID-19 pathogenesis, which need further functional 267 validation and clinical investigations to evaluate its significance.

The lymphopenia is another prominent feature of immune perturbation of severe COVID-19 2,3,16 . Here, 269 we further revealed that the COVID-19 associated lymphopenia included not only depletion of CD8 +

T cells, but also significant loss of innate-like T cells, including MAIT, γδ T and NKT cells, similarly 271 reported by another study 13, 28 . In contrast, we noticed that the frequencies of CD4 + T cells among all 272 CD3 + T lymphocytes were actually increased, like Th2, Th17 and Tfh cells. Those CD4 + T cell subsets 273 were also more clonally expanded, suggesting their activation status. We suspected that the disturbed perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in People's Hospital (2020-207). All participants provided written informed consent for sample collection and subsequent analyses.

Thirteen COVID-19 patients were enrolled in this study at the Shenzhen Third People's Hospital.

Metadata and patients' samples were collected similarly as previously described 8 : The severity of COVID-19 was categorized to be mild, moderate, severe and critical according to the "Diagnosis and Treatment Protocol of COVID-19 (the 7th Tentative Version)" by National Health Commission of China (http://www.nhc.gov.cn/yzygj/s7653p/202003/46c9294a7dfe4cef80dc7f5912eb1989.shtml). In this study, we grouped patients with mild and moderate COVID-19 as the mild group, and included those with severe and critical diseases as the severe group. Three healthy subjects were enrolled as the control group.

In clinical practice, nasal swab, throat swab, sputum, anal swab or BALF could be collected for the SARS-CoV-2 nucleic acid assays. Total nucleic acid was extracted from the samples using the QIAamp RNA Viral kit (Qiagen) and the qRT-PCR was performed using a China Food and Drug Administration-approved commercial kit specific for SARS-CoV-2 detection (GeneoDX). Each qRT-PCR assay provided a threshold cycle (Ct) value. The specimens were considered positive if the Ct value was ≤ 37, and otherwise it was negative. Specimens with a Ct value > 37 were repeated. The specimen was considered positive if the repeat results were the same as the initial result or between 37 and 40. If the repeat Ct was undetectable, the specimen was considered to be negative.

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. . https://doi.org/10.1101/2020.08.15.20175638 doi: medRxiv preprint For harvesting BALF cells, freshly obtained BALF was placed on ice and processed within 2 hours in BSL-3 laboratory. By passing BALF through a 100 µm nylon cell strainer to filter out cell aggregates and debris, the remaining fluid was centrifuged and the cell pellets were re-suspended in the cooled RPMI 1640 complete medium. For PBMC isolation, immune cells from peripheral blood were isolated by ficoll-hypaque density gradient centrifugation protocol. For subsequent study, the isolated cells were counted in 0.4% trypan blued, centrifuged and re-suspended at the concentration of 2 × 10 6 /ml.

Twelve cytokines including IL-1β, IL-6 and IL-8 etc. were detected according to the instruction (Unimedica, Shenzhen, China, Cat. No. 503022). In brief, the supernatant was taken from BALF after 10 min centrifugation at 1, 000g. Afterwards, 25 µl Sonicate Beads, 25 µl BALF supernatant or plasma, and 25µl of Detection Antibodies were mixed and placed on a shaker at 500 rpm for 2 hours at room temperature. Then 25 µl of SA-PE was added to each tube directly. The tubes were then placed on a shaker at 500 rpm for 30 minutes. The data were obtained by flow cytometry (Canto II, BD) and were analyzed use LEGENDplex v8.0 (VigeneTech Inc.).

For cell-surface labeling, 1x10 6 cells were blocked with Fc-block reagent (BD Biosciences). Then, the following antibodies were added and incubated for 30 min, including anti-CD3 (BioLegend, HIT3a), anti-CD14 (BioLegend, 63D3), anti-HLA-DR (BioLegend, L243), and anti-CD45 (BioLegend, 2D1).

After incubation, the samples were washed and reconstituted in PBS for flow cytometric analysis on a FACSCanto II flow cytometer.

The scRNA-seq libraries were prepared with Chromium Single Cell VDJ Reagent Kits v3 (10x perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. . and library construction. The resulting library products consist of Illumina adapters and sample indices, allowing pooling and sequencing of multiple libraries on the next-generation short read sequencer.

We aligned the sequenced reads against GRCh38 human reference genome by Cell Ranger (version 3.1.0, 10x genomics). To remove potential ambient RNAs, we used the remove-background function in CellBender 30 , which removes ambient RNA contamination and random barcode swapping from the raw UMI-based scRNA-seq data. Quality of cells were further assessed by the following criteria: 1)

The number of sequenced genes is 200 to 6,000; 2) The total number of UMI per cells is greater than 1,000; 3) The percentage of mitochondrial RNA is less than 15% per cell.

Data integration, cell clustering and dimension reduction were performed by Seurat (version 3) 31 . First, we identified 2,000 highly variable genes (HVGs) which were used for the following analysis using FindVariableFeatures function. Next, we integrated different samples by IntegrateData function, which eliminates technical or batch effect by canonical correlation analysis (CCA). Using those HVGs, we calculate a PCA matrix with the top 50 components by RunPCA function. The cells were then clustered by FindClusters function after building nearest neighbor graph using FindNeighbors function.

The cluster-specific marker genes were identified by FindMarkers function using MAST algorithm.

The clustered cells were then projected into a two-dimension space for visualization by a non-linear dimensional reduction method RunUMAP in Seurat package.

For cells in PBMCs, we integrated the myeloid compartment including mDC and monocytes, or NK and T cells using the similar aforementioned procedure. We re-clustered the myeloid or NK and T cells using the top 20 dimensions of PCA with default parameters. To obtain high resolution cell clusters for each subset, we set the parameter resolution to 1.2. The cell clusters were annotated by canonical markers.

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. . https://doi.org/10.1101/2020.08. 15.20175638 doi: medRxiv preprint Myeloid or NK and T cells from PBMC and BALF were integrated separately. For myeloid cells, we extracted macrophage and mDC cells in BALF, and monocyte and mDC in PBMC from the corresponding raw count matrix. The extracted cells were integrated using CCA in Seurat (version 3) as mentioned above. For clustering, the resolution parameter was set to 0.6. Similarly, we extracted NK and T cells in BALF and in PBMC from the corresponding raw count matrix. The extracted cells were integrated using CCA in Seurat (version 3) as mentioned above. For clustering, the resolution parameter was set to 1.5. 

ClusterProfiler 32 in R was used to perform GO term enrichment analysis for the significantly upregulated and down-regulated genes. Only GO term of Biological Process (BP) was displayed.

Composite signature scores of MHC class II molecules and calprotectin proteins of each peripheral CD14 + monocytes were calculated using "AddModuleScore" function implemented in the Seurat package. MHC class II score was calculated using the following genes, i. Calprotectin protein score was calculated using genes including S100A1, S100A2, S100A3, S100A4, S100A5, S100A6, S100A7, S100A7A, S100A7L2, S100A7P1, S100A7P2, S100A8, S100A9, S100A10, S100A11, S100A12, S100A13, S100A14, S100A15A, S100A16, S100B, S100G, S100P, and S100Z. The All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. . https://doi.org/10.1101/2020.08.15.20175638 doi: medRxiv preprint correlations between MDSC-like scores ("Calprotectin protein score" minus "MHC class II score") and plasma IL6, CRP levels, and the blood neutrophil / CD3 / CD4 / CD8 cell counts, were calculated using GraphPad Prism 8.4.2. Lines were fitted using the simple linear regression method.

Trajectories analysis was performed using slingshot 33 for monocyte-macrophages and T cells separately. For T cells in PBMC, naive CD4 and CD8 T cells were set as the start point for CD4 + T cells and CD8 + T differentiation trajectory, respectively. For integrated analysis of monocytemacrophages in PBMC and BALF, we deduced the cell trajectory for each individual using the peripheral monocytes as the start point.

The amino acid and nucleotide sequence of TCR chains were assembled and annotated by cellranger vdj function in CellRanger (version 3.1.0). Only cells with paired TCRα and TCRβ chains were included in clonotype analysis. Cells sharing the same TCRα-and TCRβ-CDR3 amino acid sequences were assigned to the same TCR clonotype. We assessed the TCR expansion, TCR transition among cell types and TCR migration between PBMC and BALF using R package STARTRAC (version 0.1.0) 34 . TCR migration between PBMC and BALF were shown using circos software 35 .

The Student's t-test (t.test in R, two-sided, unadjusted for multiple comparisons) was used for pairwise comparisons of the cell proportions between different groups. Statistical difference of TCR expansion index and migration index, between mild and severe disease group, were calculated using t.test in R. Statistical difference of cytokines level between BALF and Blood in Figure 3G were calculated using two-sided wilcox.test in R.

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. (G) The Pearson correlation of "MDSC-like signature score" and plasma CRP, IL-6 levels, blood neutrophil, CD3 + , CD4 + and CD8 + T cell counts.

(H) Left panel shows the representative flow cytometric data of HLA-DR expression on CD14 + and CD14 -PBMCs.

Right plot shows the summarized data from more subjects (two-sided Student's t-test). perpetuity.

preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in

The copyright holder for this this version posted August 17, 2020. . https://doi.org/10.1101/2020.08.15.20175638 doi: medRxiv preprint colored bar).

(F) Heatmaps show the expression of selected interferon, cytokine and chemokine genes in paired blood and BALF monocyte-macrophages derived from the same patient. Stars indicate that the genes are differentially expressed in BALF monocyte-macrophages between mild and severe COVID-19. Purple and green stars show the gene expression are significantly upregulated in severe COVID-19 and mild COVID-19 group respectively (MAST; p < 0.01).

(G) The levels of selected cytokines and chemokines in paired BALF and plasma samples were measured by CBA (two-sided Wilcoxon test between BALF and PBMC of severe patients). (D) Comparisons of percentages of each peripheral NK and T cell types between the two COVID-19 groups and controls (two-sided Student's t-test, *p＜0.05, **p＜0.01, ***p＜0.001).

(E) Clonally expanded T cells from COVID-19 patients and controls are projected into UMAP from (A).

(F) Clonal expansion indexes of T cell subsets from COVID-19 patients and controls, are separately displayed.

(G) T cell state transition status among any two clusters is inferred by their shared TCR clonotypes. Each T cell cluster is represented by a unique color. The numbers above the bar indicate the percentages of cells sharing TCRs in those two clusters. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 17, 2020. . https://doi.org/10.1101/2020.08.15.20175638 doi: medRxiv preprint

Clinical Characteristics of Coronavirus Disease 2019 in China

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Longitudinal hematologic and immunologic variations associated with the progression of COVID-19 patients in China

Effective treatment of severe COVID-19 patients with tocilizumab

The trinity of COVID-19: immunity, inflammation and intervention

Inhibition of Bruton tyrosine kinase in patients with severe COVID-19

Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19

Single-cell landscape of bronchoalveolar immune cells in patients with COVID-19

Host-Viral Infection Maps Reveal Signatures of Severe COVID-19 Patients

Neutralizing antibody responses to SARS-CoV-2 in a COVID-19 recovered patient cohort and their implications

Detection of SARS-CoV-2-Specific Humoral and Cellular Immunity in COVID-19 Convalescent Individuals

Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans with COVID-19

Immune cell profiling of COVID-19 patients in the recovery stage by single-cell sequencing

Complex Immune Dysregulation in COVID-19 Patients with Severe Respiratory Failure

Differential Redistribution of Activated Monocyte and Dendritic Cell Subsets to the Lung Associates with Severity of COVID-19

The UPenn COVID Processing Unit

Deep immune profiling of COVID-19 patients reveals patient heterogeneity and distinct immunotypes with implications for therapeutic interventions

Cross-talk between the airway epithelium and activated immune cells defines severity in COVID-19

Catherine A. Blish A single-cell atlas of the peripheral immune response in patients with severe COVID-19

Targeted Immunosuppression Distinguishes COVID-19 from Influenza in Moderate and Severe Disease

Clinical course and mortality risk of severe COVID-19

Myeloid-derived suppressor cells as regulators of the immune system

Myeloid-derived suppressor cells coming of age

The CD14(+)HLA-DR(lo/neg) Monocyte: An Immunosuppressive Phenotype That Restrains Responses to Cancer Immunotherapy

Recommendations for myeloid-derived suppressor cell nomenclature and characterization standards

Pathological inflammation in patients with COVID-19: a key role for monocytes and macrophages

Haiming Wei Pathogenic T cells and inflammatory monocytes incite inflammatory storm in severe COVID-19 patients

Impaired type I interferon activity and exacerbated inflammatory responses in severe Covid-19 patients

Robin Patients with Coronavirus Disease

CellBender remove-background: a deep generative model for unsupervised removal of background noise from scRNA-seq datasets

Comprehensive Integration of Single-Cell Data

clusterProfiler: an R package for comparing biological themes among gene clusters

Slingshot: cell lineage and pseudotime inference for single-cell transcriptomics

Lineage tracking reveals dynamic relationships of T cells in colorectal cancer

Circos: an information aesthetic for comparative genomics