COVID-19 and influenza infections mediate distinct pulmonary cellular and transcriptomic changes

SARS-CoV-2 infection can cause persistent respiratory sequelae. However, the underlying mechanisms remain unclear. Here we report that sub-lethally infected K18-human ACE2 mice show patchy pneumonia associated with histiocytic inflammation and collagen deposition at 21 and 45 days post infection (DPI). Transcriptomic analyses revealed that compared to influenza-infected mice, SARS-CoV-2-infected mice had reduced interferon-gamma/alpha responses at 4 DPI and failed to induce keratin 5 (Krt5) at 6 DPI in lung, a marker of nascent pulmonary progenitor cells. Histologically, influenza- but not SARS-CoV-2-infected mice showed extensive Krt5+ “pods” structure co-stained with stem cell markers Trp63/NGFR proliferated in the pulmonary consolidation area at both 7 and 14 DPI, with regression at 21 DPI. These Krt5+ “pods” structures were not observed in the lungs of SARS-CoV-2-infected humans or nonhuman primates. These results suggest that SARS-CoV-2 infection fails to induce nascent Krt5+ cell proliferation in consolidated regions, leading to incomplete repair of the injured lung.


Supplemental Figure 5 .
Representative image shows the colocalization of Krt-5 and NGFR in pulmonary krt5+ "pod" structure of Flu infected mice at 14 DPI.Lung sections are stained by DAPI (White), Krt-5 (green) and NGFR (Red).At 14 DPI, Flu infected mouse exhibits Krt5+ "pods" in the lung.The Krt-5 signals colocalize with NGFR (the highest magnification image in lowest panel).

Supplemental Figure 6 .
Serial sections show the presence of AT1 cells in pulmonary krt-5+ "pod" area and AT2 cells at the edge of pulmonary krt5+ pod area at 14 days post Flu infection (n=4).a-c) Representative images of Krt-5, E-cad, and pro-SPC IF staining.Similar region of lung as determined by DAPI staining.(a) Krt-5+ pod is located by Krt5 fluorescence staining.(b) AT1 cells are stained by E-cad.(c) pro-SPC staining reveals AT2 cells locate at the edge of Krt-5+ pod. of Fibrosis in Consolidated Lung versus Normal lung in Flu and COVID.Picrosirius Red staining for the comparison and quantification of fibrosis in regions of SARS-CoV-2-and Flu-associated pulmonary consolidation.Left) PSR staining in combination with polarized light microscopy identifies collagen by red staining and birefringence on merged images (brightfield and polarized, black arrows).Middle) Birefringence of collagen (white arrows) is easily seen when viewed with only polarized light and differentiates from background red staining of the merged image.Right) Quantification was performed with pattern recognition software trained to recognize the dual expression of red staining and birefringence (blue, black arrows).Regions of interest were drawn to exclude areas of the lung that normally contain collagen within these regions (large vessels and airways).Red = glass, green= Background.

Supplemental Figure 8 :
Collagen deposited in Flu mouse model and COVID mouse model.(a) Masson's trichrome staining showing collagen deposition in lungs of 1 x10 4 TCID 50 SARS-CoV-2-infected K18 mice at 7, 14 and 21 DPI and Flu infected mice at 5, 7, 10 and 14 DPI.(b) Comparison of collagen deposition in regions of consolidated lung to normal lung in the same Flu-infected and COVID mice.Data are shown as mean ± SEM.Two-way analysis of variance (ANOVA) was used to compare collagen deposition level changes over time.One-tailed unpaired Student's t-test was performed to test the difference between two groups at one time point.* * * indicates p < 0.001.
Method of histopathologic quantification of Krt5, smooth muscle actin, and collagen deposition in regions of consolidation.Lung, influenza 14 DPI, region of consolidation.Computer software was used to quantify the deposition of collagen (a, blue), SMA (b, green), or Krt-5 (c, red) in regions of consolidation and normal lung.Detection of each was determined based on intensity of staining in brightfield (collagen) or fluorescence (SMA & Krt-5).The same region of consolidation (box, top right) is used to demonstrate the analysis results for each marker.Arrows highlight foci of marker detection in paired images.Data generated from these analysis for all normal and consolidated regions are shown in graphs below (mean ± SEM). .Representative histopathology of COVID in human patient 3. (a) Emphysematous changes characterized by coalescing and expanded alveolar space.(b) Organizing pneumonia with fibrocollagenous aggregates ("Masson bodies") plugging the alveoli.(c) Fat embolus in a small artery.
. No Krt5+ progenitor cell "pod" in COVID human patient 2 and patient 3. Krt5+ cells are only found in the basal layer of airways without any evidence of proliferation.

Flu
Supplemental Figure 12.Quantification of AT2 cells in the lungs of SARS-CoV-2and Flu-infected mice.Lung sections are from naïve B6 mice (n = 3), SARS-CoV2infected z mice at 7 (n = 3) , 14 (n = 3), 21 (n = 3) and 45 (n = 2 )DPI, Flu-infected B6 mice at 7 (n = 3) and 14 ( n = 4) DPI.a-c) Representative image of AT2 cells in the lung of mice at 21 days post flu infection, 21 days post SARS-COV2 infection and naïve mice respectively.(d) Quantified data for the percentage of AT2 cells among total pulmonary cells in the lung tissue.AT2 cells are counted by Halo software.Mixed-effect analysis is used for flu-SARS comparison.Two-tailed unpaired t test is performed to compare the difference at 21 DPI.** indicates p < 0.01.

Table 1 . Mice information
DPI: Days post infection.Supplemental Table 2.