Increase of mast cells in COVID-19 pneumonia may contribute to pulmonary fibrosis and thrombosis

Abstract

Aims: Lung tissue from COVID-19 patients shares similar histomorphological features with chronic lung allograft disease, also suggesting activation of autoimmune-related pathways in COVID-19. To more clearly understand the underlying spectrum of pathophysiology in COVID-19 pneumonia, we analysed mRNA expression of autoimmune-related genes in post-mortem lung tissue from COVID-19 patients.

Methods and results: Formalin-fixed, paraffin-embedded lung tissue samples of 18 COVID-19 patients and eight influenza patients were used for targeted gene expression profiling using NanoString technology. Multiplex immunofluorescence for tryptase and chymase was applied for validation. Genes related to mast cells were significantly increased in COVID-19. This finding was strengthened by multiplex immunofluorescence also showing a significant increase of tryptase- and chymase-positive cells in COVID-19. Furthermore, receptors for advanced glycation end-products (RAGE) and pro-platelet basic protein (PPBP) were up-regulated in COVID-19 compared to influenza. Genes associated with Type I interferon signalling showed a significant correlation to detected SARS-CoV2 pathway-related genes. The comparison of lung tissue samples from both groups based on the presence of histomorphological features indicative of acute respiratory distress syndrome did not result in finding any specific gene or pathways.

Conclusion: Two separate means of measuring show a significant increase of mast cells in SARS-CoV-2-infected lung tissue compared to influenza. Additionally, several genes involved in fibrosis and thrombosis, among which are RAGE and PPBP, are up-regulated in COVID-19. As mast cells are able to induce thrombosis and fibrosis, they may play an important role in the pathogenesis of COVID-19.

Keywords: COVID-19; autoimmunity; fibrosis; gene expression profiling; genomics; mast cells; thrombosis.

Figure 1

Histology of lung tissue of severe COVID‐19 and influenza. An overview of COVID‐19 lung tissue (A) displays features of acute DAD consisting of intra‐alveolar dispositions of hyaline membranes (B) and scattered ‘balls’ of fibrin indicative for AFOP (C). Lung tissue of influenza samples show features of acute and organising DAD. Extensive intra‐alveolar haemorrhage and macrophages (D) are indicative for acute DAD, while patchy distribution of fibroblastic proliferations (E) and squamous metaplasia (F) are features fitting organising DAD. Displayed features indicative for acute DAD, organising DAD and AFOP were interchangeable between COVID‐19 and influenza. G–I, Thrombi of various sizes from COVID‐19 samples. Microthrombi (G) as well as thrombi in small or large arteries (H, I) were more often seen in COVID‐19 than influenza. AFOP, acute fibrinous and organising pneumonia; DAD, diffuse alveolar damage.

Figure 2

Volcano plot displaying each gene's –log₁₀ (P‐value) and log₂ fold change. Twelve of the 43 differentially expressed genes had an adjusted P‐value of <0.50. Genes with a negative fold change are up‐regulated in influenza (blue), whereas genes with positive fold change are up‐regulated in COVID‐19 (orange). [Color figure can be viewed at wileyonlinelibrary.com]

Figure 3

Cell type scores and mRNA expression of mast cell related genes. A, Plots of cell scores in influenza and COVID‐19. Mast cells were the only types of cells which were found to be differentially expressed between influenza and COVID‐19 with statistical significance (P = 0.0172). B, Distribution of normalised data of genes related to mast cells in influenza (blue) and COVID‐19 (orange). Included genes for this comparison were TPSAB1/B2, CPA3 and HDC. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4

Volcano plots displaying each gene's –log₁₀ (P‐value) and log₂ fold change based upon predominant DAD phase (A) and presence of AFOP (B). Genes with a negative fold change (left side) are up‐regulated in groups with either acute DAD or AFOP. Highlighted genes have an adjusted P‐value >0.50. AFOP, acute fibrinous and organising pneumonia; DAD, diffuse alveolar damage. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 5

Multiplex immunofluorescence of tryptase and chymase. A, Haematoxylin and eosin staining of lung tissue. This case depicts lung tissue affected by COVID‐19. B, Multiplex immunofluorescence of the same area. Tryptase is stained green and chymase is stained white. Cell nuclei are stained blue by 4′,6‐diamidino‐2‐phenylindole (DAPI). The yellow square depicts the hot‐spot selected for automated cell count. C, Enhanced view of hot‐spot. Cells detected by QuPath are encircled. Yellow border: tryptase‐positive cells. Pink border: chymase‐positive cells. Red border: tryptase‐ and chymase‐positive cells. Grey border: tryptase‐ and chymase‐negative cells. D, Box‐plot of tryptase‐ and chymase‐positive cells; the y‐axis depicts the percentage of double‐positive cells from the total cells.

Figure 6

Examples of multiplex immunofluorescence from different influenza (A–C) and COVID‐19 (D–F) tissue samples. Tryptase and chymase are stained green and white, respectively; nuclei are stained blue by 4′,6‐diamidino‐2‐phenylindole (DAPI). Note that the vast majority of tryptase‐positive cells in COVID‐19 samples are also positive for chymase, while a substantial amount of tryptase‐positive cells are negative for tryptase in influenza samples. [Color figure can be viewed at wileyonlinelibrary.com]