SARS-CoV-2 infection drives an inflammatory response in human adipose tissue through infection of adipocytes and macrophages

Abstract

Obesity, characterized by chronic low-grade inflammation of the adipose tissue, is associated with adverse coronavirus disease 2019 (COVID-19) outcomes, yet the underlying mechanism is unknown. To explore whether severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection of adipose tissue contributes to pathogenesis, we evaluated COVID-19 autopsy cases and deeply profiled the response of adipose tissue to SARS-CoV-2 infection in vitro. In COVID-19 autopsy cases, we identified SARS-CoV-2 RNA in adipocytes with an associated inflammatory infiltrate. We identified two distinct cellular targets of infection: adipocytes and a subset of inflammatory adipose tissue-resident macrophages. Mature adipocytes were permissive to SARS-CoV-2 infection; although macrophages were abortively infected, SARS-CoV-2 initiated inflammatory responses within both the infected macrophages and bystander preadipocytes. These data suggest that SARS-CoV-2 infection of adipose tissue could contribute to COVID-19 severity through replication of virus within adipocytes and through induction of local and systemic inflammation driven by infection of adipose tissue-resident macrophages.

Fig. 1.. SARS-CoV-2 RNA and immune infiltration are present in adipose tissue of autopsy samples from patients with COVID-19.

(A to E) RNA in situ hybridization (ISH) on epicardial fat from heart autopsies from patients with COVID-19 are shown. (A and B) show samples from autopsy #9. (C to E) show samples from autopsy #10. Assays were performed using probes against SARS-CoV-2 spike mRNA. Black arrowheads show ISH positive signals and gray arrows show inflammatory cells. (A and C) Overview of the heart tissue section (2mm), and (B and D) magnified view (20μm) of the represented region. (E) Interface of epicardial fat and myocardium (50μm). Note the inflammatory infiltration (gray arrows) only in the epicardial fat. Image has been rotated 90°.

Fig. 2.. Mature and in vitro differentiated adipocytes support SARS-CoV-2 infection in the absence of ACE2 protein.

Mature adipocytes (MA) from fresh human adipose tissue were isolated by collagenase digestion and infected with SARS-CoV-2 (USA-WA1/2020) at a MOI of 1, or mock-infected, for 1 hour followed by washing and culturing prior RNA collection. (A) Measurements of absolute SARS-CoV-2 (N gene) genome copy numbers of genomic (gRNA) and subgenomic (sgRNA) RNA are shown for mature adipocytes from subcutaneous (left; n=3) and visceral (right; n=2 omentum; n=1 epicardial; n=1 pericardial) adipose tissue at 24 hours post-infection (hpi). (B) SARS-CoV-2 N gene ΔCt values are shown at 24 hpi for adipocytes differentiated from pericardial adipose tissue for 0, 3, 6, and 13 days (Dif.0, 3, 6, 13). (C) SARS-CoV-2 N gene ΔCt values are shown at 24, 48, 72, and 96hpi for adipocytes differentiated from pericardial adipose tissue. In (B and C), each data point is an average of 2 technical replicates and are presented as mean ± s.e.m. (D) Plaque assay results are shown for supernatants collected from day 13 differentiated (Dif.13) adipocytes from pericardial adipose tissue that were either mock infected (top row, light green) or infected with SARS-CoV-2 (bottom row, dark green). Tested samples were either neat (undiluted) or serial 10-fold dilutions. Bar plot quantification of measured plaque forming units (PFU)/ml on the right. (E) Plaque assay results are shown for supernatant collected from day 12 differentiated (Dif.12) adipocytes from subcutaneous adipose tissue that were treated with heat inactivated (H.I) SARS-CoV-2 (middle, dark green), SARS-CoV-2 (right, dark green) or mock-infected (left, light green). Bar plot quantification of measured PFU/ml on the right. (F) ACE2 RNA expression was measured in A549 cells, A549-ACE2 cells, differentiated adipocytes, whole adipose tissue (WAT), and mature adipocytes from subcutaneous (MA-SAT) and visceral (MA-VAT) adipose tissue using primers targeting the exon 2 to 3 (Full length), exon 9a to 10 (isoform), and exon 17 to 18 (total) of ACE2. Each data point in A549 cells, A549-ACE2 cells, differentiated adipocytes, and whole adipose tissue are averages of technical replicates. Values in MA-SAT (n=3) and MA-VAT (n=4) are from independent donors. (G) ACE2 protein expression was evaluated by Western blot analysis on cell types as in (F), using antibodies against the N and C terminus of ACE2. GAPDH was used as a loading control. Absolute gene quantification in (A to C) was obtained using a standard curve generated with N gene insert. In all cases RTqPCR analyses, ΔCt values were obtained by 1-step RTqPCR using 18s as a housekeeping gene and data are presented as mean ± s.e.m. Color legend on the top right applies to all panels.

Fig. 3.. Exposure of stromal vascular cells from adipose tissue to SARS-CoV-2 leads to abortive infection in adipose tissue macrophages.

Stromal vascular cells (SVC) from fresh human adipose tissue was infected with SARS-CoV-2 (USA-WA1/2020) at an MOI of 1 or mock-infected, and SARS-CoV-2 N gene expression was obtained by 1-step RTqPCR. (A) N gene ΔCt at 24hpi or mock (subcutaneous, n=6; visceral, n=4), and (B) absolute gene copy numbers of genomic (gRNA) and subgenomic (sgRNA) viral RNA at 24hpi (subcutaneous, n=2; visceral, n=4) are shown. (C and D) N gene detection on SVC exposed to SARS-CoV-2, heat inactivated SARS-CoV-2, or mock for 96hpi are reported as N gene ΔCt for cells (C) or N-gene absolute copy numbers for supernatant (D). (E) N gene ΔCt was measured in infected or mock SVC that were maintained for 24, 48, 72, and 96hpi before RNA isolation. (F) Plaque assay results are shown for neat (undiluted) and diluted viral stock (top) or neat (undiluted) supernatants collected from subcutaneous (middle, blue) or visceral (bottom, red) adipose tissue SVC from mock (left, light colors), heat inactivated SARS-CoV-2 (middle, dark colors), or SARS-CoV-2 (left, dark colors) conditions. (G) Frequency of SARS-CoV-2 infected cells is shown based on SARS-CoV-2 N protein detection by flow cytometry of subcutaneous (n=2) and visceral (n=2) adipose tissue. Paired samples are connected by mock (left) and infected (right). (H) N gene measurements are shown for cells (left) and supernatant (right) from monocyte-derived macrophages (MDMs) infected with SARS-CoV-2 for 1hpi, 96hpi, or mock. (I) Plaque assay results are shown for neat supernatant from mock infected (left, light red), and SARS-CoV-2 infected MDMs for 1 hour (middle, dark red), and 96 hours (right, dark red). Data are presented as mean ± s.e.m., except in (D) and (H, right), where data are presented in geometric means. The color legend on the top right applies to all subfigures.

Fig. 4.. A subset of adipose tissue macrophages is infected with SARS-CoV-2.

(A to F) SVC were isolated from the SAT and VAT depots of three different participants and were mock infected or infected with SARS-CoV-2 at an MOI of 1. Each sample was collected for scRNA-seq at 24 hpi. (A to C) UMAP representation of the SVC from all participants (n=3) across 198,759 cells, colored by manually annotated cell type (A), infection and depot (B), and (C) SARS-CoV-2 cpm (log10) (C). (D) The box and whisker plot (box defines the inter-quartile range (IQR) between first (Q1) and third (Q3) quartiles with median quartile annotated by a line, whiskers define a distance of 1.5 * IQR from Q1 and Q3) reveals the proportion of clusters within the SARS-CoV-2 infected conditions that have greater than 10 viral unique molecular identifiers (UMIs) present, showing only the top four cell clusters with the highest composition of SARS-CoV-2+ cells. (E) UMAP projections of all macrophages from the scRNA seq dataset are shown, colored by log10 SARS-CoV-2 cpm (left) and colored by CD68, LYZ, IL1B and ACE2 expression (right). (F) Dotplot of the proportion of cells (dot size) in the macrophage clusters, split by infection condition, expressing genes relevant for SARS-CoV-2 entry and antiviral defense, as well as SARS-CoV-2 cpm and macrophage cluster markers, colored by scaled average expression. (G) ACE2 expression was measured on SVC from SAT and VAT using primers targeting exon 2 to 3 (Full length), exon 9a to 10 (isoform), and exon 17 to 18 (total) of ACE2. Values are shown for SVC-SAT (n=6) and SVC-VAT (n=7). (H) Barplots demonstrate the frequency of ACE2-expressing cells across various cell types in the SAT (left) and the VAT (right) using flow cytometry and quantifying ACE2 expression by comparing anti-ACE2 stained cells to isotype control.

Fig. 5.. The infected macrophage cluster is marked by increased chemokine expression.

(A) UMAP projections of a downsampled number of macrophages from the total scRNA seq dataset, such that each participant, infectious status, and depot contributed 1290 cells (the lowest number of cells across the group. Cells are colored by infection status, either Mock (black), Bystander (purple), or SARS-CoV-2+ (red). Proportion bars above each cluster are shown, representing the distribution of Infection Status across the cluster. (B) The volcano plot shows DEGs between macrophage clusters 2 (C2) and 12 (C12) across mock-infected samples. (C) A Venn diagram is shown comparing the significant DEGs (padj <.05, abs(log2FC) >= 0.6) and their direction of change across C2 versus C12 in ( 1 ) mock-infected, ( 2 ) all SAT, ( 3 ) all VAT, and ( 4 ) SARS-CoV-2-infected conditions. (D) The heatmap shows the most significant DEGs (padj <.001, abs(log2FC) >= 0.8) between SARS-CoV-2+ versus bystander macrophages within C2. (E) Reactome pathway analysis is shown for DEGs between SARS-CoV-2+ versus bystander C2 macrophages. Top ranked pathways by normalized enrichment score and padj < 0.05 are represented.

Fig. 6.. Preadipocytes respond to SARS-CoV-2 exposure.

(A and B) UMAP embedding is shown for all preadipocytes (n=140,867) colored by cluster (A) and sample and infection type (B) are shown. (C) A cell fraction bar plot is shown clustered by sample and infection type within each cluster. (D) Feature plots depicting expression of selected markers associated with preadipocyte cell states, cell types, and antiviral ISGs. (E) Box plots of average cytokine (top) and ISG (bottom) module scores are shown across the preadipocytes of each participant and depot in both mock and SARS-CoV-2 infection conditions. (F) Reactome pathway analysis was performed on the DEGs (padj < 0.1) between SARS-CoV-2-exposed versus mock preadipocytes by participant and cluster within SAT. Pathways that were represented and significant (padj < 0.05) in at least four of the participant-cluster subsets were included. Pathways were clustered by euclidean distance and split by the two major subtrees.