Background: Transcriptomic data on bronchoalveolar lavage (BAL) from COVID-19 patients are currently scarce. Objectives: This case series seeks to characterize the intra-alveolar immunopathology of COVID-19. Method: BALs were performed on 14 patients (5 COVID-19, of which 3 mild and 2 largely asymptomatic, 9 controls). Controls included asthma (n = 1), unremarkable BALs (n = 3), infections with respiratory syncytial virus (n = 1), influenza B (n = 1), and infections with other coronaviruses (n = 3). SARS-CoV-2 RNA load was measured by quantitative nucleic acid testing, while the detection of other pathogens was performed by immunofluorescence or multiplex NAT. Results: Gene expression profiling showed 71 significantly downregulated and 5 upregulated transcripts in SARS-CoV-2-positive lavages versus controls. Downregulated transcripts included genes involved in macrophage development, polarization, and crosstalk (LGALS3, MARCO, ERG2, BTK, RAC1, CD83), and genes involved in chemokine signaling and immunometabolism (NUPR1, CEBPB, CEBPA, PECAM1, CCL18, PPARG, ALOX5, ALOX5AP). Upregulated transcripts featured genes involved in NK-T cell signaling (GZMA, GZMH, GNLY, PRF1, CD3G). Patients with mild COVID-19 showed a significant upregulation of genes involved in blood mononuclear cell/leukocyte function (G0S2, ANXA6, FCGR2B, ADORA3), coagulation (von Willebrand factor [VWF]), interferon response (IFRD1, IL12RB2), and a zinc metalloprotease elevated in asthma (CPA3) compared to asymptomatic cases. In-silico comparison of the 5 COVID-19 BAL cases to a published cohort of lethal COVID-19 showed a significant upregulation of “antigen processing and presentation” and “lysosome” pathways in lethal cases. Conclusions: These data underscore the heterogeneity of immune response in COVID-19. Further studies with a larger dataset are required to gain a better understanding of the hallmarks of SARS-CoV-2 immunological response.

Bronchoalveolar lavage (BAL) is a routine, invasive diagnostic procedure to characterize lower respiratory tract pathology mainly in the setting of infectious and interstitial lung diseases. The diagnostic yield for infectious causes in immunosuppressed patients ranges from 26 to 69% [1]. BAL is a safe procedure in non-critically ill patients while the complications are mostly self-limiting [1]. Patients with acute respiratory distress syndrome have a potentially increased risk for complications but, supported by non-invasive ventilation, BAL can demonstrate a feasible tool to circumvent a lung biopsy or resection [2‒6]. Microbiological analyses and differential cell counts in BAL samples may provide essential information in complex cases of pneumonia [7]. Therefore, BAL has also become an important component in COVID-19 diagnostic work-ups, to distinguish between infectious causes, superinfection, or non-infectious interstitial inflammation [8, 9].

Data on BAL fluid (BALF) from COVID-19 patients remain scarce, with early case series and studies reporting lymphocytosis and/or plasmacytosis in severe instances [10‒13]. Lymphocytosis in BAL was shown to correlate with longer hospitalization and mechanical ventilation time, with characteristic flow cytometry immunophenotyping profiles that may be linked to clinical outcomes [13]. Further studies to characterize the immunological fingerprint in the alveolar space are therefore required to better understand the pulmonary microenvironment response to SARS-CoV-2, especially on a genomic and molecular level.

Herein, we present a small case series of differential gene expression in SARS-CoV-2-positive BALs in comparison to a control cohort in an effort to better characterize the immunopathology of COVID-19 pneumopathies. We proceeded to compare differential gene expression according to disease severity and contrasted the results of this BAL cohort to a previously published autopsy cohort of lethal COVID-19.

Patient Selection and BAL

A workflow of this study is shown in Figure 1. A total of 14 patients with cytological specimens of BALF were retrospectively included (“BAL cohort”). Suspicion of pulmonary infection was defined by the presence of new or worsened pulmonary signs and symptoms (cough, sputum, breathlessness/tachypnea, fever, abnormal auscultation) and/or new or progressive radiological findings. BALs were performed at the University Hospital Basel (n = 13) and Cantonal Hospital Lucerne (n = 1) between March and May 2020. Five SARS-CoV-2 positive (quantitative nuclear acid testing, [QNAT], performed on lavage) and nine control BALs were analyzed. Out of the 5 patients in the SARS-CoV-2 arm, three presented with mild COVID-19-related pulmonary symptoms (case 1, 2, and 4), while two patients were asymptomatic (case 3 and 5). All patients with SARS-CoV-2-positive BALF were similarly tested positive on QNAT from nasal swab material. Control BALs were selected from a range of pulmonary findings, including asthma (n = 1), unremarkable BALs from patients investigated for a suspicious peripheral lung mass (n = 3), infections with respiratory syncytial virus (n = 1), influenza B-virus (n = 1), or other human coronaviruses (n = 3). Cases of lethal COVID-19 (n = 27) encompass a cohort of formalin-fixated and paraffin-embedded lung tissue from autopsies of patients who died of COVID-19-associated respiratory failure (lethal disease, “autopsy cohort”), which had been previously published by our group [14‒16].

Fig. 1.

Workflow of BAL processing: After performing lavage, a part of the sample was sent for cytological analysis. Another part was sent for microbiological analysis (culture, PCR for SARS-CoV-2, and RespiFinder). After processing, RNA was isolated from the smears and was sent for gene expression profiling according to standard HTG protocols. Parts of this artwork by Servier Medical Art©, licensed under a Creative Commons Attribution 3.0 Unported License.

Fig. 1.

Workflow of BAL processing: After performing lavage, a part of the sample was sent for cytological analysis. Another part was sent for microbiological analysis (culture, PCR for SARS-CoV-2, and RespiFinder). After processing, RNA was isolated from the smears and was sent for gene expression profiling according to standard HTG protocols. Parts of this artwork by Servier Medical Art©, licensed under a Creative Commons Attribution 3.0 Unported License.

Close modal

Sample Processing, Ancillary Testing, and QNAT

Processing of BALF in the cytopathology laboratory was performed according to routine procedures [6] with minor modifications. Total cell counts were obtained using a disposable counting chamber (Cell-Chip, Bioswisstec). Differential cell counts were acquired from manual enumeration of 200 cells on a May-Grunwald-Giemsa-stained smear. In addition, four cytological smears, fixed in Delaunay solution and stained according to Papanicolaou, were routinely performed with an automated stainer (HistoCore Spectra ST, Leica Biosystems). Dried and acetone fixed unstained cytocentrifugation preparations (CytoPro, ELITech Group, Puteaux, France) were assessed by immunofluorescence for the presence of Pneumocystis jirovecii (PCP) and cytomegalovirus (CMV). For PCP, a commercial test kit (MonoFluo™ Pneumocystis jirovecii, Bio-Rad, Hercules, CA, USA) and for CMV, a laboratory-developed assay with a mouse monoclonal anti‐CMV antibody (Cell Marque) and a fluorescein isothiocyanate-labeled anti-mouse secondary IgG antibody (Merck Millipore) were used. Fluorescence stains were used for detection of mycobacteria (Auramine-Rhodamine, Aerospray TB, ELITech Group, Puteaux, France) and fungal organisms (Fungiqual A, R&R Spezialchemikalien für die medizinische Diagnostik, Müllheim, Germany). Evaluation of cytological, immunofluorescence, and immunostaining samples were performed by experienced cytopathologists.

Microbiological and virological evaluations were performed on an aliquot of the BALF. Viral load of SARS-CoV-2-RNA was measured by QNAT as previously described [17, 18]. The presence of other viruses (respiratory syncytial virus, influenza, other human coronaviruses) was assessed by a commercial molecular testing panel (bioMérieux; RespiFinder, PathoFinder BV, Maastricht, NL).

Gene Expression Profiling

RNA was extracted from Papanicolaou-stained cytological smears after coverslips were removed by xylene, and cells were scraped from the glass slides. Sample and library preparation, quantitation, normalization, and next-generation sequencing was performed according to manufacturer’s protocols on the HTG EdgeSeq Processor (HTG Molecular Diagnostics, Tucson, AZ, USA) with the HTG EdgeSeq Immune Response Panel (detailed methodology as previously described [19]). Samples which did not meet the manufacturer’s quality control criteria (percentage of overall reads allocated to the positive process control probe per sample <28%, read depth ≥750,000, relative standard deviation of reads per probe within a sample >0.094) were excluded from the dataset.

Statistical Analysis

Data were analyzed by HTG EdgeSeq Reveal (https://reveal.htgmolecular.com/), with the DESeq2 package (version 1.30.1) from Bioconductor for differential gene expression using negative-binomial generalized linear models [20]. With empirical Bayes methods, dispersion and log2 (fold change) were estimated with pre-construed data-driven distributions. An adjustment for library size was made by the median ratio method [21], and dispersions were approximated by Cox-Reid-adjusted profile likelihood [22]. Normalized counts were log2-transformed via Tikhonov/ridge regularization with a zero-centered normal prior distribution to stabilize the variances (see DESeq2 documentation for details). Due to the limited size of this cohort, descriptive statistics on demographics and clinical findings were not performed. Asymptomatic and mild COVID-19 (BAL) cases were compared as a whole to controls and separately to each other. In addition, the ranked fold changes were used to test the enrichment of the Kyoto Encyclopedia of Genes and Genomes (KEGG) gene sets. The applied algorithm fgsea_1.20.0 is a common implementation of the Broad Institute’s GSEA algorithm. The KEGG collection of gene sets was also obtained from the Broad Institute’s Molecular Signatures database (https://www.gsea-msigdb.org/gsea/msigdb). Finally, cases of the BAL cohort were compared in-silico to the autopsy cohort after correcting for unwanted technical variation. For this purpose, additional variables were included into the design of the differential expression model: first, a categorical variable explicitly describing BAL and autopsy (formalin-fixed and paraffin-embedded) samples, and, second, a variable to capture any remaining batch effects estimated by the RUVSeq_1.28.0 package using positive control probes in the immune response panel. The effectiveness of the unwanted variation removal was confirmed by analysis of principle components and relative log expressions.

Clinical Characteristics and BAL Differential Counts

Details on clinical findings and differential cell counts are found in Tables 1 and 2.

Table 1.

Clinical findings

PatientGroupSexAgeClinical presentationComorbiditiesRadiological findingsMicrobiologySmoker
COVID-19, mild 54 Acute COVID-19 with headache and diarrhea, dizziness, and dyspnea Follicular lymphoma Bilateral reticular and ground glass opacities Never 
COVID-19, mild 68 Acute COVID-19 with respiratory symptoms MDS/MPN-U; S.p. allogeneic HCT Bilateral peribronchial and perivascular consolidations with ground glass opacities Pseudomonas aeruginosa Never 
COVID-19, asymptomatic 79 SARS-CoV-2 positive; asymptomatic S.p. hemicolectomy due to polyposis Bilateral patchy ground glass opacities Never 
COVID-19, mild 58 Acute COVID-19 with respiratory symptoms COPD, hypertension (initial presentation to the ER due to suspected anaphylactic shock) Bronchiectasis, mucous plugging, peribronchial consolidation Haemophilus influenzae Continued (40py) 
COVID-19, asymptomatic 28 SARS-CoV-2 positive; asymptomatic AML Consolidation of right lower lobe with subsegmental pulmonal arterial embolism Never 
Controls 52 Dyspnea Asthma bronchiale, obesity, hypertension Chronic bronchitis, no bronchiectasis, no consolidations Stopped 10 years ago (10py) 
Controls 78 n/a Pulmonary metastasis of rectum carcinoma Pulmonary consolidation in right upper lobe Stopped 12 years ago (40–50py) 
Controls 63 n/a Adenocarcinoma of the lung, post-radio and -chemotherapy and pembrolizumab Consolidation in lower left lobe Continued (80py) 
Controls 49 Febrile neutropenia and cough Triple-negative breast cancer, post-chemotherapy and pembrolizumab Pulmonary nodule in middle lobe with perifocal ground glass opacities RSV Never 
10 Controls 53 Tracheobronchitis, uveitis, fatigue Sarcoidosis Several pulmonary micronodules in upper and lower right lobes, bihilar, and mediastinal lymphadenopathy Influenza B Continued (50py) 
11 Controls 57 n/a B-ALL with allogeneic stem cell transplant (HSCT) and chronic pulmonary GvHD (initial diagnosis 2003) n/a (routine control) Coronavirus (other than SARS-CoV-2) Never 
12 Controls 71 n/a Squamous cell carcinoma of the lung, laryngeal carcinoma, COPD Pulmonary nodule in right upper lobe Continued (100py) 
13 Controls 20 Fever, cough AML-MRC with allogeneic HSCT Bilateral bronchopneumonia Coronavirus (other than SARS-CoV-2), Streptococcus pneumoniae Continued 
14 Controls 67 Chronic cough ANA/CCP- negative arthritis, COPD, hypertensive cardiopathy, obesity Pulmonary consolidation in right upper lobe Coronavirus (other than SARS-CoV-2), pneumococcus Continued (36py) 
PatientGroupSexAgeClinical presentationComorbiditiesRadiological findingsMicrobiologySmoker
COVID-19, mild 54 Acute COVID-19 with headache and diarrhea, dizziness, and dyspnea Follicular lymphoma Bilateral reticular and ground glass opacities Never 
COVID-19, mild 68 Acute COVID-19 with respiratory symptoms MDS/MPN-U; S.p. allogeneic HCT Bilateral peribronchial and perivascular consolidations with ground glass opacities Pseudomonas aeruginosa Never 
COVID-19, asymptomatic 79 SARS-CoV-2 positive; asymptomatic S.p. hemicolectomy due to polyposis Bilateral patchy ground glass opacities Never 
COVID-19, mild 58 Acute COVID-19 with respiratory symptoms COPD, hypertension (initial presentation to the ER due to suspected anaphylactic shock) Bronchiectasis, mucous plugging, peribronchial consolidation Haemophilus influenzae Continued (40py) 
COVID-19, asymptomatic 28 SARS-CoV-2 positive; asymptomatic AML Consolidation of right lower lobe with subsegmental pulmonal arterial embolism Never 
Controls 52 Dyspnea Asthma bronchiale, obesity, hypertension Chronic bronchitis, no bronchiectasis, no consolidations Stopped 10 years ago (10py) 
Controls 78 n/a Pulmonary metastasis of rectum carcinoma Pulmonary consolidation in right upper lobe Stopped 12 years ago (40–50py) 
Controls 63 n/a Adenocarcinoma of the lung, post-radio and -chemotherapy and pembrolizumab Consolidation in lower left lobe Continued (80py) 
Controls 49 Febrile neutropenia and cough Triple-negative breast cancer, post-chemotherapy and pembrolizumab Pulmonary nodule in middle lobe with perifocal ground glass opacities RSV Never 
10 Controls 53 Tracheobronchitis, uveitis, fatigue Sarcoidosis Several pulmonary micronodules in upper and lower right lobes, bihilar, and mediastinal lymphadenopathy Influenza B Continued (50py) 
11 Controls 57 n/a B-ALL with allogeneic stem cell transplant (HSCT) and chronic pulmonary GvHD (initial diagnosis 2003) n/a (routine control) Coronavirus (other than SARS-CoV-2) Never 
12 Controls 71 n/a Squamous cell carcinoma of the lung, laryngeal carcinoma, COPD Pulmonary nodule in right upper lobe Continued (100py) 
13 Controls 20 Fever, cough AML-MRC with allogeneic HSCT Bilateral bronchopneumonia Coronavirus (other than SARS-CoV-2), Streptococcus pneumoniae Continued 
14 Controls 67 Chronic cough ANA/CCP- negative arthritis, COPD, hypertensive cardiopathy, obesity Pulmonary consolidation in right upper lobe Coronavirus (other than SARS-CoV-2), pneumococcus Continued (36py) 

AML, acute myeloid leukemia; AML-MRC, acute myeloid leukemia with myelodysplasia-related changes; ANA, antinuclear antibody; CCP, cyclic citrullinated peptide; COPD, chronic obstructive pulmonary disease; COVID-19, coronavirus disease 2019; GvHD, graft versus host disease; HSCT, hematopoeitic stem cell transplant; MDS-MPN/U, myelodysplastic/myeloproliferative neoplasm, unclassifiable; py, pack years; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Table 2.

Differential cell count, BAL

PatientGroupTotal cell count, 106/LMacrophages, %Lymphocytes, %Neutrophilic granulocytes, %Eosinophilic granulocytes, %
COVID-19, mild 400 66 31 
COVID-19, mild 75 19 75 
COVID-19, asymptomatic 135 72 24 
COVID-19, mild n/a 
COVID-19, asymptomatic 15 95 
Controls 81 98 
Controls 93 98 
Controls 180 98 
Controls 81 60 31 
10 Controls 110 78 19 
11 Controls 97 76 19 
12 Controls 74 95 
13 Controls 506 91 
14 Controls 100 61 33 
PatientGroupTotal cell count, 106/LMacrophages, %Lymphocytes, %Neutrophilic granulocytes, %Eosinophilic granulocytes, %
COVID-19, mild 400 66 31 
COVID-19, mild 75 19 75 
COVID-19, asymptomatic 135 72 24 
COVID-19, mild n/a 
COVID-19, asymptomatic 15 95 
Controls 81 98 
Controls 93 98 
Controls 180 98 
Controls 81 60 31 
10 Controls 110 78 19 
11 Controls 97 76 19 
12 Controls 74 95 
13 Controls 506 91 
14 Controls 100 61 33 

Gene Expression Profiling of COVID-19 Patients versus Controls in BALF

Differential expression analysis revealed 71 downregulated and 5 upregulated genes in SARS-CoV-2 positive lavages (n = 14) compared to controls (n = 9) (Fig. 2a, see Supplementary Table 1 [for all online suppl. material, see https://doi.org/10.1159/000532057] for a full list of significant probes). A subsequently performed principal component analysis biplot did not reveal a distinct clustering of COVID-19 cases versus controls; however, all COVID-19 cases (purple) as well as 2 cases from the “other coronavirus” cohort (green) present with low values of component 1, indicating some overlaps of probe values (Fig. 2b).

Fig. 2.

Differential gene expression of SARS-CoV-2 positive BALs compared to controls. a Volcano plot reveals 76 probes, of which 71 were upregulated and 5 were downregulated in COVID-19 cases compared to controls. b A PCA biplot demonstrates no characteristic clustering between COVID-19 and controls, although all COVID-19 cases and 2 coronavirus (other) cases demonstrated markedly lower PC1 values. c Significantly upregulated genes involved in macrophage signaling, development, and T-helper cell crosstalk. d Significantly upregulated genes involved in chemokine signaling, oxidative stress, and immunometabolism. e Significantly downregulated genes involved in NK signaling. CD3G, a component of the CD3 receptor complex, was also significantly downregulated. Controls = orange, COVID-19 = light blue.

Fig. 2.

Differential gene expression of SARS-CoV-2 positive BALs compared to controls. a Volcano plot reveals 76 probes, of which 71 were upregulated and 5 were downregulated in COVID-19 cases compared to controls. b A PCA biplot demonstrates no characteristic clustering between COVID-19 and controls, although all COVID-19 cases and 2 coronavirus (other) cases demonstrated markedly lower PC1 values. c Significantly upregulated genes involved in macrophage signaling, development, and T-helper cell crosstalk. d Significantly upregulated genes involved in chemokine signaling, oxidative stress, and immunometabolism. e Significantly downregulated genes involved in NK signaling. CD3G, a component of the CD3 receptor complex, was also significantly downregulated. Controls = orange, COVID-19 = light blue.

Close modal

Figure 2c and d demonstrate a selection of significantly downregulated genes involved in macrophage development, polarization, and crosstalk with T-helper cells (Fig. 2c; LGALS3, MARCO, ERG2, BTK, RAC1, and CD83), genes influencing oxidative stress, immune metabolism, and chemokine signaling (Fig. 2d; NUPR1, CEBPB, CEBPA, PECAM1, CCL18, PPARG), and genes that play a key role in leukotriene synthesis (ALOX5 and ALOX5AP). Figure 2e shows upregulated genes associated with NK-signaling (GZMA, GZMH, GNLY, PRF1); the gene encoding for CD3G, a subunit of the T-cell receptor CD3 complex, was also significantly upregulated. See online suppl. Table 1 for a full list of significantly up- and downregulated genes with p values.

Hierarchical cluster analysis of all 76 significant probes showed that COVID-19 cases (pictured in light blue) cluster apart distinctly from the control group, except in 2 cases (case 6: asthma and case 13: other coronavirus infection with superinfection caused by Pneumococcus pneumoniae) (Fig. 3). Interestingly, the COVID-19 case, which clustered between the 2 controls also presented with a bacterial superinfection (case 4, SARS-CoV-2 and Haemophilus influenzae). Some notable genes influencing the hierarchical clustering included genes involved in class 2 membrane histocompatibility complex signaling (HLA-DRB and -DQB), which were significantly downregulated in COVID-19 cases compared to controls, as well as the genes involved in NK-signaling, which were significantly upregulated (as also shown in Fig. 2e).

Fig. 3.

Hierarchical cluster analysis of 76 up- and downregulated genes in SARS-CoV-2-positive BALs versus controls. Plotting all 76 significantly up- and downregulated genes in a heat map revealed distinct hierarchical clustering of all COVID-19 cases. 2 control cases (one with asthma, one with bacterial bronchopneumonia, and other coronavirus infection) clustered with the COVID-19 cases. Individual case numbers are denoted in the hierarchical clustering rows (compare with online suppl. Table 1). Controls = orange, COVID-19 = light blue.

Fig. 3.

Hierarchical cluster analysis of 76 up- and downregulated genes in SARS-CoV-2-positive BALs versus controls. Plotting all 76 significantly up- and downregulated genes in a heat map revealed distinct hierarchical clustering of all COVID-19 cases. 2 control cases (one with asthma, one with bacterial bronchopneumonia, and other coronavirus infection) clustered with the COVID-19 cases. Individual case numbers are denoted in the hierarchical clustering rows (compare with online suppl. Table 1). Controls = orange, COVID-19 = light blue.

Close modal

Gene Expression Profiling of COVID-19 Patients in BALF, according to Severity

Of the 5 COVID-19 patients included in the BAL cohort, 3 presented with mild disease and 2 were largely asymptomatic. Direct comparison between these 2 subgroups revealed an upregulation of genes involved in blood mononuclear cell/leukocyte function (G0S2, ANXA6, FCGR2B, ADORA3), coagulation (von Willebrand factor [VWF]), interferon response (IFRD1, IL12RB2), and a zinc metalloprotease that is elevated in human asthma patients (CPA3) (see online suppl. Table 2). Pathway enrichment analysis did not reveal any significant results.

Gene Expression Profiling of Mild Disease (BAL Cohort) Compared to Lethal Disease (Autopsy Cohort)

Upon comparing cases of the BAL cohort to the autopsy cohort (lethal COVID-19), pathway enrichment analysis showed “antigen processing and presentation” and “lysosome” pathways to be significantly upregulated in lethal cases. Genes involved in inflammation, histiocyte function, and M2 polarization (IFI6, TIMP1, CD163, S100A9, IL2R2, FCER1G, CEBPD) were significantly upregulated in lethal cases. LY6E, a gene involved in modulation of viral infection by coronaviruses, was similarly upregulated. Downregulated genes included JUN and FOS (AP-1 transcription complex genes), LAMP3 (involved in lysosomal function in dendritic cells), AQP4 (a water homeostasis gene), CXCL2 (a chemokine secreted by monocytes and macrophages that is chemotactic for polymorphonuclear leukocytes), NR4A2, NCOA3 (members of the nuclear steroid-thyroid hormone-retinoid receptor superfamily), and SOCS3 (a suppressor of the JAK-STAT pathway) (see online suppl. Table 3.1, 3.2 for a full list of genes and p values).

By analyzing BALs, this case series sought to gain an insight into the immunological transcriptome of the pulmonary microenvironment, in particular of the lower airways as it responds to acute SARS-CoV-2 infection. All lavages with SARS-CoV-2 positivity were performed early on in the pandemic (between March and May 2020), during a period in which all were affected with the same wild-type variant. The wild-type variant has been shown to have more cellular tropism for lower airways than later variants such as Delta and Omicron, which primarily target the upper airways [23]. This makes BALs performed in early 2020 suitable to specifically study the dynamics of host immune response to SARS-CoV-2 in the lung parenchyma.

Previous transcriptomic analyses of BALs performed on COVID-19 patients revealed distinct host inflammatory cytokine profiles [24], B-cell-driven immune responses [25], and stimulation of distinct myeloid lineages characteristic for macrophage activation syndrome [26]. These findings are mirrored in histomorphological and immunohistochemical analyses of mediastinal lymph nodes in COVID-19 patients too, showing extensive dysregulation in both the innate and adoptive immune responses [19]. The results of the present analysis show a distinct pattern of immunopathology, which are systematically discussed below.

A subset of genes downregulated in COVID-19 BALF cases versus control BALF plays an instrumental role in macrophage activation, polarization, and crosstalk (Fig. 2c). These include Rac family small GTPase1 (RAC1), related to the generation of reactive oxygen species in alveolar macrophages [27] and Bruton tyrosine kinase (BTK), which is not only a key regulator of B-cell receptor signaling but also of macrophage polarization [28]. Early growth response gene-2 (EGR2), an essential factor for the expression of the transcription factors CEBPβ and PPARγ in M2-polarized macrophages [29], is similarly downregulated; fittingly, PPARγ is downregulated in this dataset as well (Fig. 2d). Further downregulated genes include macrophage receptor with collagenous structure (MARCO) and LGALS3 (galectin-3), which among other functions were described to be involved in opsonization and phagocytosis [30, 31]. A paucity of such genes in the transcriptome of the alveolar space may thus be a reflection of the known dysregulated macrophage profiles in COVID-19 with a prevalence of HLA-DRlo and CD83lo monocytes in severe disease ( also both genes being downregulated in the here-reported series), which has similarly been shown in serological studies [32‒35]. Furthermore, higher levels of BTK in peripheral blood have been shown to be linked with increased disease severity, thus highlighting the potential importance of this marker in COVID-19 prognosis and the potential therapeutic utility of BTK inhibitors such as ibrutinib in severe disease [36, 37]. Indeed, BTK was downregulated in the BAL cohort (asymptomatic to mild disease) but not in our cohort of lethal COVID-19; further investigations are thus needed to compare its expression in airways and blood in the context of disease severity.

A downregulation of genes involved in immune metabolism and cytokine signaling in COVID-19 BALF versus controls is demonstrated in Figure 2d. CCAAT enhancer binding protein beta (CEBPB) is a key regulator of acute phase proteins as well as cytokines and reportedly is involved in the immune response of severe COVID-19 cases [38]. Platelet and endothelial cell adhesion molecule 1 (PECAM1), CD31, a cell adhesion molecule required for transendothelial migration of leucocytes, has previously been shown to be associated with more severe COVID-19 disease [39]. ALOX5 and ALOX5AP (arachidonate 5-lipoxygenase and its activating protein), both genes involved in leukotriene synthesis were similarly downregulated. The role of 5-lipoxygenase has been investigated in prior studies, which describe a crucial pathophysiological link between leukotriene release and the formation of a cytokine storm [40]. Paradoxically, these genes are significantly downregulated in our cohort. This may be due to transcriptomic heterogeneity between the alveolar and the interstitial space; although interestingly and analogous to our observations, low ALOX5AP expression levels were previously observed in the BALF of severe COVID-19 cases [41]. More comparative spatial transcriptomic studies of parenchyma versus BAL are required to further investigate this hypothesis.

An upregulation of genes encoding for granzymes (GZMA GZMH and GNLY) and perforin (PRF1), thus indicative of an increased NK-T cell activity, was observed between COVID-19 and control BALF (Fig. 2e, 3). In peripheral blood samples, an impaired cytotoxic activity has been described to be inversely proportional to IL-6 levels [42]. Additionally, higher mRNA levels of cytotoxicity-related genes were measured in blood samples of COVID-19 patients early after infection, fitting with our observations [43]. Furthermore, in a single-cell analysis involving samples from the upper and lower respiratory tract, an increased activity of cytotoxic T-cells (including an increased expression of granzymes among other genes) played a fundamental role in the immune response of COVID-19 patients with moderate and severe disease [44]. These results, together with the observations made in the present study, make cytotoxic activity a major hallmark of the immunological response against SARS-CoV-2.

Upon comparing differential gene expression between asymptomatic and mild COVID-19 cases in BALF, mild cases showed a significant upregulation of genes involved in blood mononuclear cell and leukocyte function (G0S2, ANXA6, FCGR2B, ADORA3) (online suppl. Table 2). For example, FCGR2B, an immune inhibitory receptor involved in the phagocytosis of immune complexes and the regulation of antibody synthesis, has been similarly shown to be upregulated in severe COVID-19, directly correlating with viral load [45]. An upregulation of these genes is thus indicative of an increased activity of blood mononuclear cell/leukocyte activity in more symptomatic disease. Interestingly, increased levels of VWF as observed in more symptomatic (as well as in lethal) cases may be indicative of a pro-inflammatory, hypercoagulable state caused by SARS-CoV-2-mediated endothelial damage [46, 47]. The critical role of VWF in SARS-CoV-2-associated thrombophilia has previously been elucidated by our group, using data from the same autopsy cohort [15].

Direct comparison of differential gene expression between COVID-19 BAL cases (asymptomatic to mild disease) and COVID-19 autopsies (severe/lethal) revealed a subset of upregulated genes related to inflammation, histocyte function, and M2 polarity in lethal cases (online suppl. Table 3). A deregulated myeloid compartment, in particular, an accumulation of M2-polarized macrophages has been previously described to be a hallmark of severe, prolonged COVID-19 [32, 48]. In our data, this is evidenced by an upregulation of genes such as CD163 and S100a9, and a downregulation of one of the suppressors of the JAK-STAT pathway (SOCS3) which, taken together, suggest an accumulation of M2-polarized macrophage population [48‒50]. Together with a broad body of evidence form the literature [51], these observations suggest potential for immunomodulatory agents, such as JAK inhibitors, to be administered in severe COVID-19, which is currently the subject of several clinical trials [52]. Furthermore, a downregulation of AP-1 transcription complex genes (FOS, JUN) in lethal disease was observed. JUN and FOS regulate the activity of the AP-1 transcription factor, which, in turn, impacts the glucocorticoid receptor in a negative feedback loop [53]. Previous studies have shown that high concentrations of circulating glucocorticoids decrease AP-1 activity [54]. Our comparative analysis indicates lower baseline levels of JUN and FOS in lethal disease, implying that decreased AP-1 function and its related effector genes play an instrumental role in lethal COVID-19.

In summary, our findings in this small case series feature dysregulated macrophage profiles and an increased cytotoxic activity as major hallmarks of the immune response in the pulmonary microenvironment of COVID-19. Further analyses with larger datasets that are normalized and grouped in accordance to quantitative microbiological results, disease severity, and microscopic compartments (spatial transcriptomics of the lung parenchyma and blood) are required to better characterize the immunopathology of SARS-CoV-2.

We would like to thank all patients included in this study and their relatives.

Written informed consent was obtained from all patients (or their parent/legal guardian/next of kin) to participate in the study and for publication of the details of their medical case and any accompanying images. This study was approved by the Ethics Committee of Northwestern and Central Switzerland (Approval ID 2020-00629), complies with the guidelines for human studies and was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

The authors declare that they have no conflicts of interest. A.T. is a co-editor of pathobiology but withdrew in all respective functions in connection to the present manuscript.

This study was supported by the Botnar Research Centre for Child Health, Grant No. FTC-2020-10.

Study design and completed the manuscript: A.T. and J.D.H. Clinical data collection: J.H., K.J., and A.S. K.J. performed BAL. Cytology analyses: S.S.P., J.H., and E.O. PCR assay validation and standardization: K.L. and H.H. Statistics: J.D.H. Critical revision of the manuscript: A.S., S.S.P., M.S.M., C.Z., E.O., H.H., K.J., J.H., K.L., J.D.H., and A.T. All authors read and approved of the manuscript.

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.

1.
Choo R, Naser NSH, Nadkarni NV, Anantham D. Utility of bronchoalveolar lavage in the management of immunocompromised patients presenting with lung infiltrates. BMC Pulm Med. 2019 Feb;19(1):51.
2.
Hisanaga J, Ichikado K, Kawamura K, Yoshioka M. The significance of bronchoalveolar or bronchial lavage in ARDS: validation in 180 patients. Eur Respir J. 2015 Sep;46(Suppl 59).
3.
Steinberg KP, Mitchell DR, Maunder RJ, Milberg JA, Whitcomb ME, Hudson LD. Safety of bronchoalveolar lavage in patients with adult respiratory distress syndrome. Am Rev Respir Dis. 1993 Sep;148(3):556–61.
4.
Korkmaz Ekren P, Basarik Aydogan B, Gurgun A, Tasbakan MS, Bacakoglu F, Nava S. Can fiberoptic bronchoscopy be applied to critically ill patients treated with noninvasive ventilation for acute respiratory distress syndrome? Prospective observational study. BMC Pulm Med. 2016 May;16(1):89.
5.
Öztürk MC, Küçük M, Uğur YL, Cömert B, Gökmen AN, Ergan B. The safety of fiberoptic bronchoscopy in airway pressure release ventilation mode in critically ill patients with severe acute respiratory distress syndrome: a preliminary study. Turk Thorac J. 2022 Nov;23(6):403–8.
6.
Meyer KC, Raghu G, Baughman RP, Brown KK, Costabel U, du Bois RM, et al. An official American Thoracic Society clinical practice guideline: the clinical utility of bronchoalveolar lavage cellular analysis in interstitial lung disease. Am J Respir Crit Care Med. 2012 May;185(9):1004–14.
7.
Leuzinger K, Stolz D, Gosert R, Naegele K, Prince SS, Tamm M, et al. Comparing cytomegalovirus diagnostics by cell culture and quantitative nucleic acid testing in broncho-alveolar lavage fluids. J Med Virol. 2021 Jun;93(6):3804–12.
8.
Pickens CO, Gao CA, Cuttica MJ, Smith SB, Pesce LL, Grant RA, et al. Bacterial superinfection pneumonia in patients mechanically ventilated for COVID-19 pneumonia. Am J Respir Crit Care Med. 2021 Oct;204(8):921–32.
9.
Søgaard KK, Baettig V, Osthoff M, Marsch S, Leuzinger K, Schweitzer M, et al. Community-acquired and hospital-acquired respiratory tract infection and bloodstream infection in patients hospitalized with COVID-19 pneumonia. J Intensive Care. 2021 Jan;9(1):10.
10.
Voiriot G, Fajac A, Lopinto J, Labbé V, Fartoukh M. Bronchoalveolar lavage findings in severe COVID-19 pneumonia. Intern Emerg Med. 2020 Oct;15(7):1333–4.
11.
Voiriot G, Fajac A, Gibelin A, Parrot A, Fartoukh M. Alveolar lymphocytosis with plasmacytosis in severe COVID-19. Respir Med Res. 2020 Nov;78:100784.
12.
Giani M, Seminati D, Lucchini A, Foti G, Pagni F. Exuberant plasmocytosis in bronchoalveolar lavage specimen of the first patient requiring extracorporeal membrane oxygenation for SARS-CoV-2 in europe. J Thorac Oncol. 2020 May;15(5):e65–6.
13.
Gelarden I, Nguyen J, Gao J, Chen Q, Morales-Nebreda L, Wunderink R, et al. Comprehensive evaluation of bronchoalveolar lavage from patients with severe COVID-19 and correlation with clinical outcomes. Hum Pathol. 2021 Jul;113:92–103.
14.
Pahima H, Zaffran I, Ben-Chetrit E, Jarjoui A, Gaur P, Manca ML, et al. Patients with coronavirus disease 2019 characterized by dysregulated levels of membrane and soluble cluster of differentiation 48. Ann Allergy Asthma Immunol. 2023 Feb;130(2):245–53.e9.
15.
van den Berg J, Haslbauer JD, Stalder AK, Romanens A, Mertz KD, Studt JD, et al. Von Willebrand factor and the thrombophilia of severe COVID-19: in situ evidence from autopsies. Res Pract Thromb Haemost. 2023 May;7(4):100182.
16.
Menter T, Haslbauer JD, Nienhold R, Savic S, Hopfer H, Deigendesch N, et al. Postmortem examination of COVID-19 patients reveals diffuse alveolar damage with severe capillary congestion and variegated findings in lungs and other organs suggesting vascular dysfunction. Histopathology. 2020;77(2):198–209.
17.
Leuzinger K, Gosert R, Søgaard KK, Naegele K, Bielicki J, Roloff T, et al. Epidemiology and precision of SARS-CoV-2 detection following lockdown and relaxation measures. J Med Virol. 2021 Apr;93(4):2374–84.
18.
Leuzinger K, Roloff T, Gosert R, Sogaard K, Naegele K, Rentsch K, et al. Epidemiology of severe acute respiratory syndrome coronavirus 2 emergence amidst community-acquired respiratory viruses. J Infect Dis. 2020 Sep;222(8):1270–9.
19.
Haslbauer JD, Zinner C, Stalder AK, Schneeberger J, Menter T, Bassetti S, et al. Vascular damage, thromboinflammation, plasmablast activation, T-cell dysregulation and pathological histiocytic response in pulmonary draining lymph nodes of COVID-19. Front Immunol. 2021 Dec;12:763098.
20.
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014 Dec;15(12):550.
21.
Anders S, Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010 Oct;11(10):R106.
22.
McCarthy DJ, Chen Y, Smyth GK. Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Res. 2012 May;40(10):4288–97.
23.
Hui KPY, Ho JCW, Cheung M, Ng K, Ching RHH, Lai K, et al. SARS-CoV-2 Omicron variant replication in human bronchus and lung ex vivo. Nature. 2022 Mar;603(7902):715–20.
24.
Xiong Y, Liu Y, Cao L, Wang D, Guo M, Jiang A, et al. Transcriptomic characteristics of bronchoalveolar lavage fluid and peripheral blood mononuclear cells in COVID-19 patients. Emerg Microbes Infect. 2020 Mar;9(1):761–70.
25.
Cavalli E, Petralia MC, Basile MS, Bramanti A, Bramanti P, Nicoletti F, et al. Transcriptomic analysis of COVID-19 lungs and bronchoalveolar lavage fluid samples reveals predominant B cell activation responses to infection. Int J Mol Med. 2020 Oct;46(4):1266–73.
26.
Daamen AR, Bachali P, Owen KA, Kingsmore KM, Hubbard EL, Labonte AC, et al. Comprehensive transcriptomic analysis of COVID-19 blood, lung, and airway. Sci Rep. 2021 Mar;11(1):7052.
27.
Osborn-Heaford HL, Ryan AJ, Murthy S, Racila AM, He C, Sieren JC, et al. Mitochondrial Rac1 GTPase import and electron transfer from cytochrome c are required for pulmonary fibrosis. J Biol Chem. 2012 Jan;287(5):3301–12.
28.
Ní Gabhann J, Hams E, Smith S, Wynne C, Byrne JC, Brennan K, et al. Btk regulates macrophage polarization in response to lipopolysaccharide. PLoS One. 2014;9(1):e85834.
29.
Veremeyko T, Yung AWY, Anthony DC, Strekalova T, Ponomarev ED. Early Growth response gene-2 is essential for M1 and M2 macrophage activation and plasticity by modulation of the transcription factor CEBPβ. Front Immunol. 2018;9:2515.
30.
Maler MD, Nielsen PJ, Stichling N, Cohen I, Ruzsics Z, Wood C, et al. Key role of the scavenger receptor MARCO in mediating adenovirus infection and subsequent innate responses of macrophages. mBio. 2017 Aug;8(4):e00670-17–.
31.
Rotshenker S. Galectin-3 (MAC-2) controls phagocytosis and macropinocytosis through intracellular and extracellular mechanisms. Front Cell Neurosci. 2022;16:949079.
32.
Schulte-Schrepping J, Reusch N, Paclik D, Baßler K, Schlickeiser S, Zhang B, et al. Severe COVID-19 is marked by a dysregulated myeloid cell compartment. Cell. 2020 Sep;182(6):1419–40.e23.
33.
Awasthi NP, Mishra S, Tiwari V, Agarwal J, Das PK, Jain P, et al. Monocyte HLADR and immune dysregulation index as biomarkers for COVID-19 severity and mortality. Indian J Clin Biochem. 2023 Apr;38(2):204–11.
34.
Giamarellos-Bourboulis EJ, Netea MG, Rovina N, Akinosoglou K, Antoniadou A, Antonakos N, et al. Complex immune dysregulation in COVID-19 patients with severe respiratory failure. Cell Host Microbe. 2020 Jun;27(6):992–1000.e3.
35.
Benlyamani I, Venet F, Coudereau R, Gossez M, Monneret G. Monocyte HLADR measurement by flow cytometry in COVID 19 patients: an interim review. Cytometry. 2020 Dec;97(12):1217–21.
36.
Roschewski M, Lionakis MS, Sharman JP, Roswarski J, Goy A, Monticelli MA, et al. Inhibition of Bruton tyrosine kinase in patients with severe COVID-19. Sci Immunol. 2020 Jun;5(48):eabd0110.
37.
Claude L, Martino F, Hermand P, Chahim B, Roger P, de Bourayne M, et al. Platelet caspase 1 and Bruton tyrosine kinase activation in patients with COVID-19 is associated with disease severity and reversed in vitro by ibrutinib. Res Pract Thromb Haemost. 2022 Nov;6(8):e12811.
38.
Ouyang Y, Yin J, Wang W, Shi H, Shi Y, Xu B, et al. Downregulated gene expression spectrum and immune responses changed during the disease progression in patients with COVID-19. Clin Infect Dis. 2020 Nov;71(16):2052–60.
39.
Li L, Huang M, Shen J, Wang Y, Wang R, Yuan C, et al. Serum levels of soluble platelet endothelial cell adhesion molecule 1 in COVID-19 patients are associated with disease severity. J Infect Dis. 2021 Jan;223(1):178–9.
40.
Ayola-Serrano NC, Roy N, Fathah Z, Anwar MM, Singh B, Ammar N, et al. The role of 5-lipoxygenase in the pathophysiology of COVID-19 and its therapeutic implications. Inflamm Res. 2021 Aug;70(8):877–89.
41.
Sahanic S, Löffler-Ragg J, Tymoszuk P, Hilbe R, Demetz E, Masanetz RK, et al. The role of innate immunity and bioactive lipid mediators in COVID-19 and influenza. Front Physiol. 2021 Jul;12:688946.
42.
Mazzoni A, Salvati L, Maggi L, Capone M, Vanni A, Spinicci M, et al. Impaired immune cell cytotoxicity in severe COVID-19 is IL-6 dependent. J Clin Invest. 2020 Sep;130(9):4694–703.
43.
Ramljak D, Vukoja M, Curlin M, Vukojevic K, Barbaric M, Glamoclija U, et al. Early response of CD8+ T cells in COVID-19 patients. J Pers Med. 2021 Dec;11(12):1291.
44.
Chua RL, Lukassen S, Trump S, Hennig BP, Wendisch D, Pott F, et al. COVID-19 severity correlates with airway epithelium-immune cell interactions identified by single-cell analysis. Nat Biotechnol. 2020 Aug;38(8):970–9.
45.
Saheb Sharif-Askari N, Saheb Sharif-Askari F, Mdkhana B, Al Heialy S, Alsafar HS, Hamoudi R, et al. Enhanced expression of immune checkpoint receptors during SARS-CoV-2 viral infection. Mol Ther Methods Clin Dev. 2021 Mar;20:109–21.
46.
Mei ZW, van Wijk XMR, Pham HP, Marin MJ. Role of von Willebrand factor in COVID-19 associated coagulopathy. J Appl Lab Med. 2021 Sep;6(5):1305–15.
47.
Stefanini GG, Montorfano M, Trabattoni D, Andreini D, Ferrante G, Ancona M, et al. ST-elevation myocardial infarction in patients with COVID-19: clinical and angiographic outcomes. Circulation. 2020 23;141(25):2113–6.
48.
Ziablitsev DS, Kozyk M, Strubchevska K, Dyadyk OO, Ziablitsev SV. Lung expression of macrophage markers CD68 and CD163, angiotensin converting enzyme 2 (ACE2), and caspase-3 in COVID-19. Med Mex. 2023 Apr;59(4):714.
49.
Schelbergen RF, Blom AB, de Munter W, Vogl T, Roth J, van den Berg WB, et al. Alarmins S100A8 and S100A9 stimulate production of pro-inflammatory cytokines in M2 macrophages without changing their M2 membrane phenotype. Ann Rheum Dis. 2012 Feb;20(Suppl 1):S233–4.
50.
Wilson HM. SOCS proteins in macrophage polarization and function. Front Immunol. 2014 Jul;5:357.
51.
van de Veerdonk FL, Giamarellos-Bourboulis E, Pickkers P, Derde L, Leavis H, van Crevel R, et al. A guide to immunotherapy for COVID-19. Nat Med. 2022 Jan;28(1):39–50.
52.
Kramer A, Prinz C, Fichtner F, Fischer AL, Thieme V, Grundeis F, et al. Janus kinase inhibitors for the treatment of COVID-19. Cochrane Database Syst Rev. 2022 Jun;6(6):CD015209.
53.
Alexaki VI, Henneicke H. The role of glucocorticoids in the management of COVID-19. Horm Metab Res. 2021 Jan;53(1):9–15.
54.
Bruscoli S, Puzzovio PG, Zaimi M, Tiligada K, Levi-Schaffer F, Riccardi C. Glucocorticoids and COVID-19. Pharmacol Res. 2022 Nov;185:106511.