Dysregulation of brain and choroid plexus cell types in severe COVID-19

Abstract

Although SARS-CoV-2 primarily targets the respiratory system, patients with and survivors of COVID-19 can suffer neurological symptoms^1-3. However, an unbiased understanding of the cellular and molecular processes that are affected in the brains of patients with COVID-19 is missing. Here we profile 65,309 single-nucleus transcriptomes from 30 frontal cortex and choroid plexus samples across 14 control individuals (including 1 patient with terminal influenza) and 8 patients with COVID-19. Although our systematic analysis yields no molecular traces of SARS-CoV-2 in the brain, we observe broad cellular perturbations indicating that barrier cells of the choroid plexus sense and relay peripheral inflammation into the brain and show that peripheral T cells infiltrate the parenchyma. We discover microglia and astrocyte subpopulations associated with COVID-19 that share features with pathological cell states that have previously been reported in human neurodegenerative disease^4-6. Synaptic signalling of upper-layer excitatory neurons-which are evolutionarily expanded in humans⁷ and linked to cognitive function⁸-is preferentially affected in COVID-19. Across cell types, perturbations associated with COVID-19 overlap with those found in chronic brain disorders and reside in genetic variants associated with cognition, schizophrenia and depression. Our findings and public dataset provide a molecular framework to understand current observations of COVID-19-related neurological disease, and any such disease that may emerge at a later date.

Extended Data Fig. 1 ∣. Characterization of human cortical and choroid plexi nuclei sequenced.

a, Total number of nuclei and median number of genes of each human sample sequenced in medial frontal cortex and choroid plexus. b, c, Quantification of the median number of genes detected per nuclei (b) and patient ages (c) in control (non-viral and influenza) and COVID-19 samples in medial frontal cortex (n = 8 control; n = 8 COVID-19, two-sided Mann-Whitey t-test; mean ± s.e.m.) and choroid plexus (n = 7 control; n = 7 COVID-19, two-sided Mann-Whitey t-test; mean ± s.e.m.). d, e, Bar graph presenting frequency of nuclei for control and COVID-19 medial frontal cortex (d) and choroid plexus (e) sample groups.

Extended Data Fig. 2 ∣. Gene expression variance analysis.

a, PVCA, displaying the gene expression variance explained by residuals (biological and technical noise) or experimental factors such as brain region, age, sex and respective combinations. n = 30 samples. b, Principal component (PC) analysis visualization of all samples, based on unscaled counts. c, UMAP projections of nuclei isolated from the medial frontal cortex (top) or choroid plexus (bottom), and split by disease group, showing no systematic batch effects.

Extended Data Fig. 3 ∣. Human brain cell-type markers.

a, Top cell-type-specific genes across the types of cells captured in the human cortex. The colour bar indicates gene expression from low (blue) to high (yellow). b, Example of top cell-type-specific genes across the types of cells captured in the human choroid plexus. Violin plots are centred around the median, with their shape representing cell distribution.

Extended Data Fig. 4 ∣. Cell-type-specific changes in gene expression and intercellular signalling in the brain of individuals with COVID-19.

a, Heat map displaying the number of significant biological pathways among the set of DEGs in each cell type (FDR < 0.05, Benjamini–Hochberg adjustment, hypergeometric test). Number of significant pathways is indicated in graded black (low) to yellow (high). b, Example upregulation of inflammatory and dysregulation of homeostatic genes in COVID-19 astrocytes. c, Comparison of the number of nuclei isolated per cell type and the number of predicted DEGs. Two-sided P-value indicates the significance of the correlation (Pearson, not significant).

Extended Data Fig. 5 ∣. Overlap between alternative snRNA-seq differential expression analysis methods.

a, b, Scatter plots demonstrating the strong correlation between the calculated effect sizes of two differential gene expression analysis methods (MAST (used here) and pseudobulk^,) across cell types in the human medial frontal cortex (a) and choroid plexus (b). Orange line denotes the trend line fitted with a generalized linear model, surrounded by a 95% confidence interval in purple. Spearman correlation is shown along with the significance by two-sided P-values.

Extended Data Fig. 6 ∣. DEGs in the brains of individuals with COVID-19 show no significant overlap with brain PMI-sensitive genes.

a, Comparison of post-mortem interval (PMI)-sensitive genes (left column, from a previous publication) and COVID-19 DEGs (all other columns). No statistically significant overlap is observed (Fisher’s exact test). b, The previous study categorized PMI-sensitive genes in two categories: glial genes upregulated and neural genes downregulated. Minimal overlap is seen with COVID-19 changes of the same category (for example, glial genes upregulated in COVID-19 versus glial genes upregulated with extended PMI). c, Heat map showing that PMI-sensitive genes are not the DEGs in COVID-19 and thus not driving the DEG-based findings of our study.

Extended Data Fig. 7 ∣. Expression of SARS-CoV-2 virus entry genes across cell types.

a, b, Expression of SARS-CoV-2 entry receptors, established and putative, across cell types in the human medial frontal cortex (a) and choroid plexus (b). Violin plots are centred around the median, with their shape representing cell distribution.

Extended Data Fig. 8 ∣. Choroid plexus inflammation in COVID-19.

Immunohistochemical staining for the macrophage activation marker CD68 (brown) in the choroid plexus of patients with COVID-19 and control individuals. Haematoxylin counterstain (blue). Scale bars, 20 μm.

Extended Data Fig. 9 ∣. No conclusive detection of SARS-CoV-2 neuroinvasion.

a, Summary of RNA-based assays to detect SARS-CoV-2 in the human cortex and choroid plexus. Aside from the 3A2 antibody, no other anti-SARS-CoV-2 antibody detected viral protein antigen in the brain or choroid plexus. b, qPCR detection of the SARS-CoV-2 genes N1 and N2 via CDC Emergency Use Authorization primers on choroid plexus samples (n = 6 non-viral control, n = 7 COVID-19; two-sided Mann–Whitney t-test; mean ± s.e.m.). c, Aberrant anti-SARS-CoV-2 spike (3A2) antibody reactivity (brown) in the frontal medial cortex of two patients with COVID-19 in tissue immediately adjacent to that used for snRNA-seq. Haematoxylin counterstain (purple). Scale bar, 20 μm. d, As in c, but for the choroid plexus and meninges in two patients with COVID-19. Scale bar, 20 μm. e, As in c, but using a different secondary antibody detection method (biotin–alkaline phosphatase (red)), recapitulating the specific vascular-localized signal. Scale bar, 20 μm. Immunohistochemical stains are representative of at least two independent experiments.

Extended Data Fig. 10 ∣. Cell communication analysis results for integrated choroid plexus and brain parenchyma cell types.

Circle plot showing the number of statistically significant intercellular signalling interactions for total signalling (over 30 ligand–receptor pathways) and the complement family of molecules in control individuals (non-viral and influenza) compared to patients with COVID-19 (permutation test, CellChat; n = 8 control, including influenza; n = 8 COVID-19 for cortex; and n = 7 control, including influenza; n = 7 COVID-19 for choroid plexus). Each circle (colour) represents one cell type, and edges connecting circles represent significant intercellular signalling inferred between those cell types. Circles and edges were normalized and scaled to display relative sizes, with the former proportional to the number of cells from a given cell type and the latter according to the inferred strength of signalling. Cell type labels correspond to signalling pathway increased in COVID-19.

Extended Data Fig. 11 ∣. Activation of parenchymal microglia and perivascular macrophages in COVID-19.

Immunohistochemical staining of microglia and macrophages by an antibody against the pro-inflammatory marker CD68 (immunoreaction in brown). Counterstained with haematoxylin for cell nuclei in blue. a, The frontal medial gyrus of patients with COVID-19 immediately adjacent to that used for snRNA-seq. A cluster of activated microglia up to single macrophages is immunostained in the parenchyma of the gyrus (subcortical white matter). Scale bar, 20 μm. b, A vessel of the medial frontal gyrus is surrounded by activated perivascular macrophages. Scale bar, 20 μm. c, The cortical surface is shown. The upper third of the figure contains the leptomeninges that cover the cortex. A dense infiltration by brown stained macrophages into the leptomeninges is visible. Scale bar, 20 μm. d, Summary of innate immune reactivity across eight patients with COVID-19, typically not observed in healthy brains at these levels, colour-coded and labelled by severity. A semiquantitative categorization for changes, as usual in the field of pathology, is used: mild = detectable microgliosis, atypical for healthy tissue; moderate = a pathological process typical of pathological changes; severe = a marked pathological process. Several clusters of microglia or macrophages were characterized as excessive beyond the severe category. Immunohistochemical stains are representative of at least two independent experiments.

Extended Data Fig. 12 ∣. Evaluation of COVID-19-enriched subpopulations in other parenchymal glia.

a, UMAP of astrocytes captured in the human frontal cortex, split by control individuals (including influenza, n = 8) and patients with COVID-19 (n = 8). Cells are coloured by cell-type subcluster. Genes upregulated in the COVID-19-enriched astrocyte cluster are labelled in green. b, Quantification of astrocyte cluster 1 as a proportion of total astrocytes (n = 8 control, including influenza; n = 8 COVID-19, two-sided Mann–Whitney t-test P = 0.0041; mean ± s.e.m.). Example genes upregulated in the COVID-19-associated astrocyte cluster are shown. c, Enriched biological pathways (Metascape) amongst upregulated gene markers of COVID-19 astrocytes. Enrichment is based on FDR-corrected cumulative hypergeometric P values (Bonferroni correction FDR < 0.05; MAST with default thresholds). d, UMAP projection of OPCs and trending but not significant emergence of a COVID-19-enriched subcluster. e, Quantification of the frequency of the COVID-19-enriched OPC subcluster as a proportion of all OPCs (n = 8 control, including 1 influenza and n = 8 COVID-19, two-sided Mann–Whitney t-test, P = 0.083; mean ± s.e.m., not significant). f, g, As in d, e, respectively, but for mature oligodendrocytes with P = 0.9591.

Fig. 1 ∣. Overview of diverse brain and choroid plexus cell types captured from post-mortem tissue from patients with COVID-19.

a, Study design. Coloured triangles denote the brain regions that were studied for each patient. IHC, immunohistochemistry. b, Uniform manifold approximation and projection (UMAP) of 38,217 nuclei from the medial frontal cortex of 8 control individuals (including 1 patient with influenza) and 8 patients with COVID-19. As in previous reports^,,, the ‘endothelial’ cluster also exhibits vascular mural cell markers and perivascular cells (perivascular fibroblast-like cells and perivascular macrophages) are not efficiently captured. exc., excitatory; in., inhibitory; OPC, oligodendrocyte precursor cell. c, Examples of DEGs in COVID-19 (n = 7 control individuals (without viral infection); n = 8 patients with COVID-19; MAST with default settings): excitatory neurons (exc. n.), inhibitory neurons (in. n.), astrocytes (ast.), oligodendrocytes (oli.), OPCs, and microglia and macrophages (mic./mac.). DEGs defined as log-transformed fold change > 0.25 (absolute value) and adjusted P value < 0.05 (Bonferroni correction). d, Cell-type specificity of cortical DEGs. UpSet plot showing a matrix layout of DEGs shared across and specific to each cell type. Each matrix column represents either DEGs specific to a cell type (single circle with no vertical lines) or DEGs shared between cell types, with the vertical line indicating the cell types that share that given DEG. Top, bar graph displays the number of DEGs in each combination of cell types. Right, bar graph displays the total number of DEGs for a given cell type. e, UMAP of 27,092 nuclei from the lateral choroid plexus of 14 individuals (n = 7 control individuals (including 1 patient with influenza); n = 7 patients with COVID-19; MAST with default settings). f, Expression profiles (counts per million reads mapped (CPM)) (circle size) and differential expression in patients with COVID-19 (average log-transformed fold change (avg log FC)) (colour) for genes relevant to SARS-CoV-2 entry into the brain. The highlighted region indicates the consistent upregulation of the antiviral defence gene IFITM3 in choroid and glia limitans brain-barrier cells.

Fig. 2 ∣. Brain-barrier inflammation in patients with COVID-19 does not require direct replicative infection.

a, Examples of inflammation-related DEGs in the choroid plexus of patients with COVID-19 (n = 6 control individuals (without viral infection); n = 7 patients with COVID-19; MAST with default settings). DEGs defined as log-transformed fold change > 0.25 (absolute value) and adjusted P value < 0.05 (Bonferroni correction). b, Validation of predicted choroid plexus DEGs by RT–qPCR (n = 6 control individuals (without viral infection), n = 7 patients with COVID-19; two-sided Mann–Whitney t-test; mean ± s.e.m.). Genes chosen for validation are either immediately related to SARS-CoV-2 (IFITM3) or genes with log-transformed fold changes similar to those of IFITM3 (NQO1), to assess the robustness of snRNA-seq thresholds. P values P = 0.0023 (IFITM3), 0.0484 (C7), 0.0350 (STAT3), 0.0140 (NQO1), 0.0082 (ZFP36) and 0.0734 (SDC4). c, snRNA-seq (left) or bulk RNA-seq (right) of choroid plexus and cortex from control individuals or patients with COVID-19 (no reads). snRNA-seq, n = 7 control, n = 7 COVID-19 (choroid plexus); n = 7 control, n = 7 COVID-19 (cortex). Bulk RNA-seq (after viral RNA isolation): n = 7 control, n = 4 COVID-19 (choroid plexus); n = 5 control, n = 4 COVID-19 (cortex). d, Circle plot showing the number of statistically significant intercellular signalling interactions for the CXCL and CCL pathway family of molecules in control individuals compared to patients with COVID-19 (permutation test, CellChat; n = 8 control individuals (including patients with influenza); n = 8 patients with COVID-19 (cortex); and n = 7 control individuals (including patients with influenza); n = 7 patients with COVID-19 (choroid plexus)). Each circle (colour) represents one cell type; edges connecting circles represent significant intercellular signalling inferred between those cell types. Circles and edges are normalized to the number of cells for a given cell type and inferred strength of signalling, respectively. Cell types labelled on the right correspond to signalling pathways increased in COVID-19. Endo., endothelial; epen., ependymal; epi., epithelial; mes., mesenchymal.

Fig. 3 ∣. A neuroinflammatory COVID-19 milieu marked by disease-associated microglia.

a, UMAP of immune cells captured in the human frontal cortex, split by control individuals (including a patient with influenza) (n = 8) (red) and patients with COVID-19 (n = 8) (light blue). Cells are coloured by cell-type subcluster (red cluster defined by homeostatic markers; light blue cluster defined by activation markers). b, Quantification of immune cell cluster 1 as a proportion of total immune cells (n = 8 control individuals (including a patient with influenza (circle marked as ‘flu’)); n = 8 patients with COVID-19, two-sided Mann–Whitney t-test P = 0.0098; mean ± s.e.m.). c, As in b, but for T cells. P=0.0003. d, e, As in a, b, respectively, but for MRC1⁻ parenchymal microglia. P = 0.0343. Unlike macrophages, microglia express low levels of MRC1 (CD206). Examples of genes that are upregulated in the microglial cluster associated with COVID-19 are shown in light blue. f, Pseudotime trajectory (Methods) indicated in graded purple (low) to yellow (high), plotting the emergence of the microglial cluster associated with COVID-19. Numbers indicate original source population (1) and the newly emerged population in COVID-19 (2). g, Immunohistochemical staining for the microglial activation marker CD68 (brown) in the frontal medial cortex of a patient with COVID-19, immediately adjacent to that used for snRNA-seq. Haematoxylin counterstain (blue). Scale bars, 20 μm. Immunohistochemical stains are representative of at least two independent experiments. h, Overlap (hypergeometric test) between marker genes of Alzheimer’s-disease-associated microglia (DAM, ARM and Mic1)^- and genes that are upregulated in the microglial cluster associated with COVID-19.

Fig. 4 ∣. Molecular dysfunction in upper-layer neurons and links to long-term symptoms.

a, Dot plot showing downregulation of synaptic vesicle components, especially in L2/3 excitatory neurons in patients with COVID-19 (n = 7 control individuals (without viral infection); n = 8 patients with COVID-19; MAST with default settings). FC, fold change. b, Diagram of cortical neurons captured in this study that have known layer localization. Neuron labels are colour-coded by layer localization as shaded in a. Figure layout adapted with permission from ref. . c, Overlap between COVID-19 DEGs and those in chronic CNS diseases (Methods). Dotted line indicates statistical significance (adjusted P value < 0.05, false-discovery rate (FDR) correction, cumulative hypergeometric test). d, Heat map showing the number of DEGs per cell type that overlap as GWAS risk variants across psychiatric and neurological diseases and traits from the GWAS catalogue (NHGRI-EBI). Significance of overlap is based on FDR-corrected cumulative hypergeometric P values (Benjamini–Hochberg correction) < 0.05; MAST with default thresholds). AD, Alzheimer’s disease; ADHD, attention deficit hyperactivity disorder; ALS, amyotrophic lateral sclerosis; MS, multiple sclerosis; MSA, multiple system atrophy; PD, Parkinson’s disease; PTSD, post-traumatic stress disorder.