Prevalence and Mechanisms of Mucus Accumulation in COVID-19 Lung Disease

Abstract

Rationale: The incidence and sites of mucus accumulation and molecular regulation of mucin gene expression in coronavirus (COVID-19) lung disease have not been reported. Objectives: To characterize the incidence of mucus accumulation and the mechanisms mediating mucin hypersecretion in COVID-19 lung disease. Methods: Airway mucus and mucins were evaluated in COVID-19 autopsy lungs by Alcian blue and periodic acid-Schiff staining, immunohistochemical staining, RNA in situ hybridization, and spatial transcriptional profiling. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected human bronchial epithelial (HBE) cultures were used to investigate mechanisms of SARS-CoV-2-induced mucin expression and synthesis and test candidate countermeasures. Measurements and Main Results: MUC5B and variably MUC5AC RNA concentrations were increased throughout all airway regions of COVID-19 autopsy lungs, notably in the subacute/chronic disease phase after SARS-CoV-2 clearance. In the distal lung, MUC5B-dominated mucus plugging was observed in 90% of subjects with COVID-19 in both morphologically identified bronchioles and microcysts, and MUC5B accumulated in damaged alveolar spaces. SARS-CoV-2-infected HBE cultures exhibited peak titers 3 days after inoculation, whereas induction of MUC5B/MUC5AC peaked 7-14 days after inoculation. SARS-CoV-2 infection of HBE cultures induced expression of epidermal growth factor receptor (EGFR) ligands and inflammatory cytokines (e.g., IL-1α/β) associated with mucin gene regulation. Inhibiting EGFR/IL-1R pathways or administration of dexamethasone reduced SARS-CoV-2-induced mucin expression. Conclusions: SARS-CoV-2 infection is associated with a high prevalence of distal airspace mucus accumulation and increased MUC5B expression in COVID-19 autopsy lungs. HBE culture studies identified roles for EGFR and IL-1R signaling in mucin gene regulation after SARS-CoV-2 infection. These data suggest that time-sensitive mucolytic agents, specific pathway inhibitors, or corticosteroid administration may be therapeutic for COVID-19 lung disease.

Keywords: COVID-19; MUC5AC; MUC5B; SARS-CoV-2; airway mucin.

Figure 1.

A diagram showing acquisition, exclusion, and allocation of coronavirus disease (COVID-19) autopsy specimens. AB-PAS = Alcian blue and periodic acid–Schiff; IHC = immunohistochemistry; NIH = National Institutes of Health; RNA-ISH = RNA in situ hybridization; UNC = University of North Carolina.

Figure 2.

Goblet cell metaplasia and mucin gene expression in coronavirus disease (COVID-19) and control autopsy proximal airways. (A) Representative images from control (CTRL) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) autopsy (COVID-19) tracheobronchial specimens stained with hematoxylin and eosin (H&E) and Alcian blue and periodic acid–Schiff (AB-PAS) and probed by RNA in situ hybridization (RNA-ISH) for MUC5B and MUC5AC. (B) Morphometric quantification of AB-PAS staining signals in the tracheobronchial airways of control (n = 11) and COVID-19 autopsy specimens (n = 21). Triangles correspond to the same specimens shown in C and D. (C, D) RNA-ISH for (C) MUC5B and (D) MUC5AC signals in the tracheobronchial airways of control (n = 11) and COVID-19 autopsy specimens (n = 7). Histogram bars and error bars represent mean ± SEM. NS = not significant; *P < 0.05; ****P < 0.0001; Mann-Whitney U test. (E) Representative images from tracheas from acute and subacute/chronic COVID-19 autopsy lungs probed for SARS-CoV-2 RNA-ISH. One of five acute COVID-19 lungs (<20 d onset from symptoms to death) was SARS-CoV-2 RNA-ISH positive, whereas neither of two chronic (>20 d onset from symptoms to death) COVID-19 lungs was positive. Scale bars, 20 μm.

Figure 3.

MUC5B and MUC5AC RNA and protein expression in distal pulmonary regions of coronavirus disease (COVID-19) and control autopsy lungs. (A) Representative low-power images of control (CTRL) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)– infected (COVID-19) acute (3 d) and chronic (31 d) autopsy lung sections stained with Alcian blue and periodic acid–Schiff (AB-PAS). (B) Magnified images of boxed regions of each tissue section shown in A stained with hematoxylin and eosin (H&E) and AB-PAS; immunofluorescence staining for MUC5B and MUC5AC protein; and probed for MUC5B, MUC5AC, and XBP1S by RNA in situ hybridization (RNA-ISH). The control and acute COVID-19 airway structures were morphologically identified as bronchioles, whereas the structure in the chronic COVID-19 lung specimen was identified as a microcyst. Black arrows indicate the regions magnified in each inset. Scale bars, 5 mm (A); 100 μm (B). (C, i) Numbers of control versus COVID-19 autopsy subject lungs with mucus-obstructed airways. **P < 0.01; Fisher’s exact test. n = 5 for control and n = 15 for COVID-19. (C, ii) Percentage of distal airways within each autopsy lung specimen obstructed by mucus for control (n = 5) and COVID-19 (n = 15) lungs. Mann-Whitney U test. Total airway structures counted for control lungs = 31 (median, 6 per lung) and for COVID-19 lungs = 970 (median, 34 per lung). (C, iii–v) Morphometric quantification of RNA-ISH (C, iii) MUC5B, (C, iv) MUC5AC, and (C, v) XBP1S expression in distal airway structures within control (n = 5) and COVID-19 autopsy lungs (n = 14 for MUC5B and MUC5AC and n = 7 for XBP1S). Histogram bars and error bars represent mean ± SEM. Mann-Whitney U test. NS = not significant; *P < 0.05; ***P < 0.001. UNC = University of North Carolina.

Figure 4.

Mucus obstruction in coronavirus disease (COVID-19) bronchiolar and microcystic structures of COVID-19 autopsy lungs. (A, B) Representative images of Alcian blue and periodic acid–Schiff (AB-PAS) and Sirius red staining for acute (A; n = 7) and chronic (B; n = 8) phase COVID-19 autopsy lungs. (A) AB-PAS–stained acute COVID-19 lung with bronchioles scored outlined in green and depicted in the magnified image to the right. (B) AB-PAS–stained chronic COVID-19 autopsy lung with microcysts outlined in red and in the magnified image to the right. Scale bars, 1 mm. (C) Number of control (CTRL) and acute versus chronic COVID-19 autopsy lungs that exhibit morphologically defined bronchioles only, microcysts only, or a combination of bronchioles and microcysts. (D) Numbers of control versus COVID-19 autopsy lungs with identified bronchioles without (–) or with (+) bronchiolar mucus accumulation/obstruction. (E) Percentage of mucus-obstructed bronchioles in each control and COVID-19 autopsy lung shown in D. Numbers bronchioles counted = 31 from 5 controls (median, 6 per lung) and 95 from 10 COVID-19 autopsy lungs (median, 6.5 per lung) in which bronchioles were morphologically identified. (F) Fraction of COVID-19 lungs with morphologically identified microcysts that exhibited concomitant mucus accumulation. Note: No control lungs exhibited microcysts. (G) Percentage of microcyst structures within each microcyst-containing COVID-19 autopsy lung with significant (⩾50% lumen) mucus accumulation. No microcysts for mucus accumulation in controls. (H, I) MUC5B (H) and MUC5AC (I) RNA in situ hybridization (RNA-ISH)–stained area normalized to airway surface epithelia, subgrouped by the presence of microcysts (MC). Histogram bars and error bars represent mean ± SEM. Fisher’s exact test (D), Mann-Whitney U test (E, G–I). *P < 0.05; **P < 0.01. BV = blood vessel; NIH = National Institutes of Health.

Figure 5.

Identification of candidate mucin transcriptional regulators AREG (amphiregulin), IL1B, and IL6 via RNA expression in coronavirus disease (COVID-19) autopsy lungs. (A) AREG, IL1B, and IL6 RNA expression in representative control (CTRL) and COVID-19 bronchioles. Upper right and lower left insets show intraepithelial and lumen areas, respectively. (B) Confocal microscopic images of fluorescence RNA in situ hybridization (RNA-ISH) for AREG, IL1B, or IL6 (red) with a macrophage marker CD68 (green) and a neutrophil marker MPO (myeloperoxidase; gray) with differential interference contrast (DIC) imaging of distal bronchioles in COVID-19 autopsy lungs. Arrows indicate the regions magnified in insets. Scale bars, 100 μm (A); 20 μm (B). (C) Frequency of AREG, IL1B, and IL6 colocalization with CD68⁺, MPO⁺, or double-positive/double-negative cells in the COVID-19 autopsy lungs (n = 4 or 5). A total of 200 randomly selected AREG/IL1B/IL6-positive cells were counted.

Figure 6.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection of human bronchial epithelial (HBE) cultures over a 14-day interval. (A) Viral titers as plaque-forming units per milliliter (PFU/ml) over time. DPI = days postinfection. (B) Fold changes (virus vs. mock) in RNA transcripts per million (TPM) for the secretory mucins (MUC5B and MUC5AC) over time. (C) Histological sections of HBE cells at Days 3 and 14 after SARS-CoV-2 infection as visualized by hematoxylin and eosin (H&E) staining, Alcian blue and periodic acid–Schiff (AB-PAS) staining, and MUC5B and MUC5AC immunohistochemistry. Scale bar, 20 μm. (D) Fold changes (virus vs. mock) of HBE cell AB-PAS–positive areas normalized to basement membrane lengths at Days 3 and 14 after SARS-CoV-2 infection. (E) Fold changes (virus vs. mock) of MUC5B- and MUC5AC-positive areas normalized to HBE epithelial areas at Days 3 and 14 after SARS-CoV-2 infection. (F) IFN-β gene expression (expressed as transcripts/million [TPM]) in virus-infected versus mock cultures at Days 1, 3, 7, and 14 after inoculation. See Figure E11C for other IFN genes. (G) Fold changes (virus vs. mock) in epidermal growth factor receptor (EGFR) ligands (AREG [amphiregulin] and HBEGF) over time after infection. (H) Fold changes (virus vs. mock) in mucin-regulatory cytokine genes IL1A, IL1B, and IL-6. n = 14 individual HBE donors. Histogram bars and error bars represent mean ± SEM. One-sample Wilcoxon test (B, D, E, G, H) and Wilcoxon test (F). NS = not significant; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 7.

Effect of epidermal growth factor receptor (EGFR) inhibition on secretory mucin RNA expression and peak viral titers in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected human bronchial epithelial (HBE) cultures. (A, B) Mucin, EGFR ligand, cytokine (IL6, CXCL8) gene expression, and peak SARS-CoV-2 virus titers with EGFR tyrosine kinase inhibitor (gefitinib [Gefi]) (A) or EGFR monoclonal antibody (cetuximab [Cetu]) (B) administration versus vehicle. Data were derived from HBE cells from n = 6 or 7 donors. No correction has been made for multiple statistical comparisons. Histogram bars and error bars represent mean ± SEM. Wilcoxon test. NS = not significant; *P < 0.05.

Figure 8.

Effect of IL-1 receptor modulation on secretory mucin RNA expression and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral infectivity. (A) Comparison of SARS-CoV-2 infection of CRISPR/Cas9-mediated IL-1 receptor knockout (KO) versus negative control (NC) human bronchial epithelial (HBE) cells with respect to relative mRNA expression of MUC5B, MUC5AC, IL6, HBEGF, AREG, and CXCL8 normalized to TBP after 3- and 7-day inoculation. Data were derived from HBE cells from n = 5–7 donors. (B) Representative whole-mount immunofluorescence staining images for MUC5B (green), MUC5AC (red), α-tubulin (cilia; gray), and DAPI (nuclei) in NC or IL1R1 CRISPR KO HBE cells 7 days after SARS-CoV2 infection. Scale bar, 20 μm. (C) Quantification of MUC5B, MUC5AC, and α-tubulin staining area in IL1R1 KO versus NC HBEs normalized to DAPI-positive area. n = 5 donors. (D) SARS-CoV-2 viral titers in plaque-forming units per milliliter (PFU/ml) in NC versus IL1R1 CRISPR KO HBE cells after SARS-CoV2 infection over 7 days. n = 7 donors. DPI = days postinfection. (E) Linear regressions showing correlations between viral titers (PFU/ml) at Day 3 and MUC5B or MUC5AC RNA expression at Day 7. n = 5–7 donors. No correction has been made for multiple statistical comparisons. Scatterplot histograms present mean ± SEM. Mixed effect model (A, D), Mann-Whitney U tests (C), and Spearman’s rank-order correlation tests (ρ = Spearman’s rho) (E). *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 9.

Effect of dexamethasone treatment on severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–infected human bronchial epithelial (HBE) cultures and coronavirus disease (COVID-19) autopsy lungs. (A) Relative expression of mucin, SCGB1A1 (secretoglobin 1A1), SPDEF (sterile α-motif pointed domain epithelial specific transcription factor), epidermal growth factor receptor (EGFR) ligands, cytokines, and epithelial sodium channel (ENaC) subunit genes normalized to TBP, and peak virus titers. (B, i) Representative images of whole-mount vehicle versus dexamethasone (Dex)-treated HBE cultures stained for MUC5B (green), MUC5AC (red), cilia (α-tubulin; white), and nuclei (blue). Scale bars, 20 μm. (B, ii) Signal quantification in whole mounts of MUC5B and MUC5AC for SARS-CoV-2–infected HBE cultures treated with vehicle versus dexamethasone at 14 days after infection. Data were derived from HBE cells from n = 8 donors. (C) Percentage of airways obstructed/lung for COVID-19 autopsy lungs from patients treated (+) or not (–) with steroids during their clinical course (see Figure 4). (D) Relationships between MUC5B (D, i) and MUC5AC (D, ii) expression in COVID-19 autopsy lungs from patients not treated with (–) or treated with (+) steroids during their clinical course. RNA-ISH = fluorescence RNA in situ hybridization. Histogram bars and error bars represent mean ± SEM. Wilcoxon tests (B) and Mann-Whitney U tests (D). NS = not significant; *P < 0.05; **P < 0.01.