key: cord-0940945-jn2ewhdo
authors: Fließer, Elisabeth; Birnhuber, Anna; Marsh, Leigh M; Gschwandtner, Elisabeth; Klepetko, Walter; Olschewski, Horst; Kwapiszewska, Grazyna
title: Dysbalance of ACE2 levels – a possible cause for severe COVID‐19 outcome in COPD
date: 2021-05-12
journal: J Pathol Clin Res
DOI: 10.1002/cjp2.224
sha: 290c097246f4a70a44baac964d28135fb2670119
doc_id: 940945
cord_uid: jn2ewhdo

Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) poses a serious threat to healthcare systems worldwide. Binding of the virus to angiotensin‐converting enzyme 2 (ACE2) is an important step in the infection mechanism. However, it is unknown if ACE2 expression in patients with chronic lung diseases (CLDs), such as chronic obstructive pulmonary disease (COPD), idiopathic pulmonary arterial hypertension (IPAH), or pulmonary fibrosis (PF), is changed as compared to controls. We used lung samples from patients with COPD (n = 28), IPAH (n = 10), and PF (n = 10) as well as healthy control donor (n = 10) tissue samples to investigate the expression of ACE2 and related cofactors that might influence the course of SARS‐CoV‐2 infection. Expression levels of the ACE2 receptor, the putative receptor CD147/BSG, and the viral entry cofactors TMPRSS2 (transmembrane serine protease 2), EZR, and FURIN were determined by quantitative PCR and in open‐access RNA sequencing datasets. Immunohistochemical and single‐cell RNA sequencing (scRNAseq) analyses were used for localization and coexpression, respectively. Soluble ACE2 (sACE2) plasma levels were analyzed by enzyme‐linked immunosorbent assay. In COPD as compared to donor, IPAH, and PF lung tissue, gene expression of ACE2, TMPRSS2, and EZR was significantly elevated, but circulating sACE2 levels were significantly reduced in COPD and PF plasma compared to healthy control and IPAH plasma samples. Lung tissue expressions of FURIN and CD147/BSG were downregulated in COPD. None of these changes were associated with changes in pulmonary hemodynamics. Histological analysis revealed coexpression of ACE2, TMPRSS2, and Ezrin in bronchial regions and epithelial cells. This was confirmed by scRNAseq analysis. There were no significant expression changes of the analyzed molecules in the lung tissue of IPAH and idiopathic PF as compared to control. In conclusion, we reveal increased ACE2 and TMPRSS2 expression in lung tissue with a concomitant decrease of protective sACE2 in COPD patients. These changes represent the possible risk factors for an increased susceptibility of COPD patients to SARS‐CoV‐2 infection.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes acute respiratory tract infections, termed COVID-19, and leads to heterogeneous clinical manifestations of variable severity [1] . There is evidence that patients with chronic lung diseases (CLDs), such as chronic obstructive pulmonary disease (COPD), develop severe illness more frequently [2] [3] [4] . However, the underlying reasons and the contribution of pulmonary hypertension, a common complication in CLD patients, have remained elusive.

Similar to the prior coronavirus subtypes, SARS-CoV-1 and Middle East respiratory syndrome coronavirus, SARS-CoV-2 uses angiotensin-converting enzyme 2 (ACE2) as the main vehicle for cell entry [5, 6] . In the lungs, the receptor is reportedly expressed on various cell types, including type I and type II alveolar cells, smooth muscle cells, or macrophages, whereas reports on its expression on endothelial cells are still controversial [7, 8] . Alternatively, CD147 is debated as a putative host factor, interacting with SARS-CoV-2 [9, 10] . In addition, specific host proteases, such as Furin and transmembrane serine protease 2 (TMPRSS2), are needed to process the virus surface spike (S)-protein in order for the virus to fuse with the host cell membrane after engaging ACE2 [5, 11] . Once the virus has successfully infected an individual, the clinical symptoms and severity of the COVID-19 disease can be manifold. The most common symptoms are fever, cough, dyspnea, headache, and diarrhea [1] . The severity is very heterogeneous, ranging from no symptoms to severe acute respiratory distress syndrome [12] . In addition, cardiovascular complications and 'endothelialitis' have been described [13] . Patients with preexisting CLDs, affecting either the airways, the pulmonary parenchyma, or the vasculature, might therefore be at high risk of severe COVID-19 pneumonia.

Although more detailed evaluations show a lower prevalence of COVID-19 among asthma patients and do not suggest asthma as an individual risk factor for severe outcome, few studies have focused on the assessment of SARS-CoV-2-associated genes in CLD such as COPD, pulmonary fibrosis (PF), or idiopathic pulmonary arterial hypertension (IPAH) [14] [15] [16] [17] [18] . A detailed meta-analysis revealed COPD to be significantly associated with mechanical ventilation, intensive care unit (ICU) requirement, and death, although infection numbers are similar to the broad population [4] . Data on SARS-CoV-2 infections in IPAH patients are very limited, although early reports point toward an increased case fatality [19, 20] .

Here, our objective was to gain detailed insights into gene expression, localization, and shedding of SARS-CoV-2 relevant molecules ACE2, TMPRSS2, Furin, Ezrin (EZR), and CD147 (Basigin [BSG]) in the lung tissue and systemic circulation of COPD, IPAH, and PF as compared to controls. We also aimed to determine whether there is an association with pulmonary hemodynamics. By simultaneously analyzing this comprehensive set of costimulatory molecules as well as by correlating results with clinical parameters, we aimed to link molecular properties to susceptibility for severe COVID-19 pneumonia.

Lung samples from patients with COPD (n = 28), IPAH (n = 10), and PF (n = 10) as well as control lung tissue from downsized donor lungs (n = 10) were collected between 2011 and 2016. Patient and donor lungs were achieved during lung transplantation (LTX) at the Department of Surgery, Division of Thoracic Surgery, Medical University of Vienna, Austria. The protocol and tissue usage were approved by the institutional ethics committee and patient consent was obtained before LTX. Patient characteristics included age at the time of LTX, body mass index (BMI), sex, smoking status, oxygen requirement, partial pressure of CO 2 (pCO 2 ), partial pressure of O 2 (pO 2 ), mean pulmonary arterial pressure (mPAP), and pulmonary function tests, and are presented in Tables 1 and 2. The study cohort was partially described in Refs [21] [22] [23] . Healthy control samples were prospectively collected and stored in the Biobank of the Medical University of Graz between 2011 and 2015. Written Hemodynamics from right heart catheterization was available for 28, 10, and 10 patients with COPD, IPAH, and idiopathic PF (IPF), respectively.

Plasma samples from patients with COPD (n = 25), IPAH (n = 15), and PF (n = 10) were collected prospectively from an IPAH cohort of patients who were treated at the Medical University of Graz (Graz, Austria) when undergoing diagnostic or follow-up right heart catheterization as well as from end-stage transplant patients (COPD, IPAH, and IPF) from the Department of Surgery, Division of Thoracic Surgery, Medical University of Vienna, Austria. Patient characteristics included age at LTX, BMI, sex, smoking status, oxygen requirement, pCO 2 , pO 2 , mPAP, and pulmonary function tests, and are presented in Table 3 . The samples from the healthy controls (n = 35) were prospectively collected and stored in the Biobank of the Medical University of Graz between 2011 and 2014. Written informed consent was obtained from all subjects. The studies were approved by the Institutional Review Board of the Medical University of Graz (23-408 ex 10/11) as well as by the Institutional Review Board of the Medical University of Vienna (976/2010) in accordance with national law. Sample usage was approved by the institutional ethics committee and patient consent was obtained before LTX. 

Tissue samples were collected and cryopreserved in liquid nitrogen. Tissue was homogenized using mortar and pestle prior to RNA isolation. RNA was isolated using a peqGOLD Total RNA Kit (Peqlab, Erlangen, Germany). cDNA synthesis was performed using the qScript™ cDNA Synthesis Kit (Quantabio, Beverly, MA, USA) according to the manufacturer's instructions. Quantitative real-time PCR (qPCR) was performed using a LightCycler ® 480 System (Roche Applied Science, Vienna, Austria). The qPCRs were set up using a QuantiFast ® SYBR ® Green PCR kit (Qiagen, Hilden, Germany) using the following protocol: 5 min at 95 C (5 s at 95 C, 5 s at 60 C, and 10 s at 72 C) Â 45. The specificity of the reaction was confirmed based on melting curve analysis and gel electrophoresis. Primer sequences were as follows

B2M-rv TGTC GGATGGATGAAACCCAGACA; and HMBS-fw ACCCTAGAAACCCTGCCAGAGAA, HMBS-rv GCCGGGTGTTGAGGTTTCCCC.

B2M and PBGD/HMBS served as reference genes. The ΔCt values were calculated using the following formula:

ΔCt values of donors and patients were normalized to the mean gene expression of donors (ΔΔCt): ΔΔCt ¼ ΔCt diseased À mean ΔCt of all donors.

A sandwich high-sensitivity enzyme-linked immunosorbent assay (ELISA) kit was used to quantitatively detect ACE2 (#SEB886Hu; Dianova GmbH, Hamburg, Germany) protein levels in plasma of COPD, IPAH, and PF patients and corresponding controls. The ELISA was performed according to the manufacturer's instructions and absorption values were measured on a ClarioStar microplate reader (BMG Labtech, Ortenberg, Germany).

The microarray dataset GSE47460 using total RNA extracted from transplant lung homogenates was downloaded from GEO DataSets [24] . Expression [25] [26] [27] [28] [29] .

Single-cell RNA sequencing (scRNAseq) data from Adams et al's study [30] were downloaded from GEO (GSE136831), and the Feature-Barcode Matrices were imported and analyzed in Seurat 3.1.1 [31] . As previously described, lung tissue samples were collected from explanted lungs and processed for scRNAseq after removal of visible airway structures, vessels, blood clots, and mucin [30] . Ten donor (003C, 065C, 137C, 160C, 244C, 253C, 296C, 454C, 465C, and 483C) and 10 COPD samples (052CO, 056CO, 178CO, 186CO, 194CO, 207CO, 237CO, 238CO, 23CO, and 8CO) were taken for further analysis; cells with high mitochondrial percentage > 10% were removed and data were normalized using default parameters. Dimension reduction was performed by Principal Component Analysis (PCA) and Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) using default parameters. Cells were clustered based on metadata as provided in Adams et al's study [30] . To simplify readouts, the following clusters were manually concatenated to give dendritic cells (DC) (cDC1, cDC2, DC_Langerhans, DC_ Mature), Fibroblast (Fibroblast, Myofibroblast), innate lymphoid cells (ILC) (ILC_A, ILC_B), Monocyte (cMonocyte, ncMonocyte), T_cell (T, T_Cytotoxic, T_Regulatory), and Vascular_Endothelium (VE_Arterial, VE_Capillary_A, VE_Capillary_B, VE_Peribronchial, VE_Venous), all others retained original identity.

Formalin-fixed and paraffin-embedded lung tissue samples were cut into 2.5-μm thick serial sections for histological analysis. Sections were deparaffinized in ROTICLEAR ® (Carl Roth, Arlesheim, Switzerland), followed by decreasing concentrations of ethanol. Antigen retrieval was performed using citrate buffer (pH 9) in a 95 C hot water bath for 20 min. The primary antibodies were detected with the immPRESS anti-rabbit, anti-goat, or anti-mouse polymer detection kit using NovaRed peroxidase (HRP) substrate (Vector Laboratories, Burlingame, CA, USA). Hemalaun was used for nuclear counterstaining. Images were obtained using an Olympus VS120 slide scanning microscope (Olympus, Tokyo, Japan) at Â40 magnification.

Statistical analysis was performed in R (version number 4, www.r-project.org). Data are expressed as single data points with boxplot overlay indicating median and interquartile range. Multiple groups were compared using a nonparametric Kruskal-Wallis test. P values of <0.05 were considered as statistically significant.

We investigated a total of 49 explanted lungs and 10 donor lungs. While COPD and PF patients had major lung function limitation with a forced vital capacity (FVC) of 48 AE 18.1 and 43 AE 10.1, respectively, IPAH patients' lung function was more preserved (80 AE 9.2% predicted). As expected, IPAH patients had a severe increase of mPAP to 68 AE 27 mmHg. In COPD and PF, mPAP was elevated to 32 AE 11 mmHg in COPD and 41 AE 7) mmHg in PF patients. In all groups, hypoxemia was common with mean arterial pO 2 of 67 AE 12 mmHg in COPD, 70 AE 14 mmHg in IPAH, and 60 AE 20 mmHg in PF patients. General patient demographics and clinical characteristics are given in Table 1 .

Despite differences regarding age and sex (Table 1) , gene expression analysis of the SARS-CoV-2 receptor, ACE2, and interacting molecules showed a clear grouping within each patient cohort, indicating that expression of SARS-CoV-2-interacting molecules was highly group-dependent ( Figure 1A) . A striking increase of ACE2 and TMPRSS2 expression and a concomitant decrease of FURIN were detected in COPD patients compared to donor samples, while in IPAH and PF the expression levels were comparable to donor ( Figure 1A) . Similarly, EZR expression was highest in COPD compared to all other groups, but was also elevated in IPAH and PF compared to donor samples ( Figure 1A ). CD147/BSG, another suggested SARS-CoV-2 receptor, was significantly downregulated in COPD compared to donor, IPAH, and PF. ACE2 can be cleaved from the cell membrane and circulate in the plasma. As soluble ACE2 (sACE2) has been proposed to be beneficial in the fight against COVID-19 via neutralization of SARS-CoV-2 [32] , we investigated sACE2 levels in the plasma of COPD, IPAH, and PF patients and healthy controls ( Figure 1B) . COPD and PF patients showed significantly decreased circulating sACE2 compared to healthy controls. In IPAH, sACE2 levels were unaltered and comparable to controls ( Figure 1B ). Patient's clinical characteristics are shown in Table 2 .

To strengthen our investigations, we analyzed gene expression levels in publicly available microarray datasets of COPD lung tissue (GSE8500, GSE103174, GSE37768, and GSE47460) and PF lung tissue (GSE110147, GSE53845, and GSE24988) compared to control lung tissue. In accordance with our data, ACE2 expression was significantly elevated in COPD patients compared to control sample in three of the four analyzed COPD datasets, whereas BSG did not Expression profiling in lung biopsy tissue (A, dataset GSE8500), lung explant tissue (B, D, datasets GSE103174 and GSE47460, respectively), and lung cancer resection tissue (C, dataset GSE37768) compared between COPD patients and donors. Boxplots with single data points are shown. Statistical analysis was performed using a nonparametric Kruskal-Wallis test with post hoc comparison of multiple groups. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

show any significant changes between COPD patients and donors (Figure 2 ). Minor differences in the expression of TMPRSS2, FURIN, and EZR were inconsistent across different COPD cohorts (Figure 2 ). Furthermore, the comparison of gene expression levels between nonsmokers, smokers, and COPD patients (GSE37768) did not show any significant expression differences, except for EZR ( Figure 2C ).

By comparing expression levels between IPF patients and donors, we did not reveal consistent changes across the analyzed datasets GSE110147 and GSE53845 (see supplementary material, Figure S1A , B); however, significant gene expression regulation was observable for all analyzed genes. ACE2 expression was found to be opposite in the two datasets. Different types of control (donor transplant tissue versus cancer resection tissue) and IPF lung tissue from different disease stages (biopsy versus lung transplant) as well as region of sampling could represent key parameters for this contrary gene expression result. Importantly, no significant changes in gene expression levels were observed when different stages of PH were compared in PF patients (see supplementary material, Figure S1C ).

Due to the strong expression changes of ACE2 and TMPRSS2 observed in our COPD cohort and the consistent changes of ACE2 expression in the analyzed datasets, we decided to focus further investigations on this disease.

ACE2 expression levels showed a significant positive correlation with FVC in COPD patients (ρ = 0.43, p = 0.032; Figure 3A ). However, correlations with hemodynamic parameters, such as mPAP, were not significant ( Figure 3B ). Detailed characteristics of lung function, hemodynamic, and clinical parameters of patients are shown in Table 3 .

As SARS-CoV-2 receptor ACE2 and diverse proteases facilitating virus entry have to be in very close proximity or on the same target cell to allow viral entry, we investigated a publicly available scRNAseq dataset from Adams et al's study [30] (GSE136831) for cell type-specific expression levels in control samples and COPD patients ( Figure 4A ). ACE2 expression was mostly below the detection limit in control and COPD samples, indicating low expression levels which were not captured by the depth provided by the 10X Genomics (Pleasanton, CA, USA) platform; therefore, more sensitive analytical methods are required. Nevertheless, other genes investigated by scRNAseq were sufficiently captured and analyzed: TMPRSS2 expression was observed almost exclusively in epithelial cells of the lung, such as alveolar type (AT) II cells, ATI, and club cells (Figure 4B,C) . FURIN was expressed ubiquitously (also in epithelial cells) with the strongest signal in macrophages and monocytes in both COPD and control lungs ( Figure 4C ). EZR was expressed by most cell types but the average expression was most prominent in epithelial cells (ATI, ATII, goblet, club, and ciliated cells) and in inflammatory cell subsets (e.g. dendritic cells and monocytes) ( Figure 4C) .

We next substantiated these findings at the protein level using immunohistochemistry. ACE2 immunoreactivity was present in epithelial cells with the most prominent staining in bronchial and alveolar epithelial cells in COPD lungs (see supplementary material, Figure S2 ). Detailed analysis of ACE2, TMPRSS2, Furin, Ezrin, and CD147 localization was performed on corresponding serial sections ( Figure 5 ). In healthy donor tissue, ACE2 almost exclusively localized to the bronchial epithelial layer (black arrowheads) with light 

staining in ATII cells (black arrows), but no detectable staining in the vascular compartment. In COPD, a clear signal was visible in bronchial and ATII cells, as well as very weak staining in vascular endothelial cells (white arrowhead) which was in contrast to healthy donor tissue. Immunoreactivity against TMPRSS2 was observed ubiquitously in bronchial, parenchymal, and vascular regions in both donor and COPD lungs ( Figure 5 ). Similar to our scRNAseq analysis, very few cells stained positive for Furin, and immunoreactivity was almost undetectable in structural cells, while monocytes and macrophages showed clear immunoreactivity ( Figure 5, parenchymal panel, inset) . Ezrin was localized to the bronchial epithelium and in Figure 5 . Immunostaining reveals cellular localization of SARS-CoV-2 receptors and processing enzymes in COPD and donor lung tissue. High-magnification images showing ACE2, TMPRSS2, Furin, Ezrin, and CD147 immunostaining in bronchial, parenchymal, and vascular compartments on corresponding serial sections of donor and COPD tissue. Scalebar = 20 μm. Black arrowhead, bronchial epithelium; black arrow, alveolar epithelial type II cells; white arrowhead, vascular endothelium; gray arrowhead, erythrocytes; asterisks, inflammatory cells.

9 Linking COVID-19 severity and chronic obstructive pulmonary disease perivascular regions. CD147 localized mostly to inflammatory cells, such as macrophages (asterisks), and on the surface of erythrocytes (gray arrowheads). In the airway epithelium and submucosal cells, CD147 was partially observable. Cumulatively, our data point toward predominant expression of ACE2 and TMPRSS2 in bronchial as well as alveolar epithelial cells in COPD lungs.

Although knowledge about SARS-CoV-2 and related COVID-19 is growing, there are limited data on the expression of genes interacting with the virus and their effects on disease susceptibility and severity in CLD patients. In addition to binding to its main receptor ACE2 [5, 6] , SARS-CoV-2 entry additionally depends on the proteases Furin and TMPRSS2 and cofactor EZR [5, [33] [34] [35] . Besides ACE2, CD147 (BSG) has been proposed as a possible SARS-CoV-2-binding receptor in endothelial cells, but with contradictory and inconclusive results [9, 10] . In line with published data, we did not observe any detectable expression of CD147 protein in vascular or microvascular endothelium [9] . RNA expression levels were unaltered in IPAH and PF and decreased in COPD, indicating that altered CD147 expression may not explain any potential increase of SARS-CoV-2 infection risk in patients with CLDs. Furthermore, Neuropilin-1 (NRP1) has been debated as a putative host factor, interacting with SARS-CoV-2 [36, 37] . However, we did not observe any regulation of NRP1 in any of the investigated diseases in our cohorts or publicly available datasets (data not shown).

Several studies have used publicly available scRNAseq datasets to investigate SARS-CoV-2 virulence gene expression in healthy and diseased tissue. One common factor of all used datasets and also observed in this study is very limited detection of the main virus receptor ACE2 in scRNAseq datasets [38, 39] . While use of such datasets may have a huge potential for more abundant genes, detection of ACE2 is limited by the method's depth. To understand the actual viral interactions possibly occurring in the lungs of CLD patients, more sensitive methods, such as qPCR or analysis of protein levels, as performed in our study, are required. Using these methods, we observed a striking increase of ACE2 in COPD, as was also observed recently by other research groups [40, 41] . As viral infections are particularly relevant in this disease, where they can lead to systemic inflammation and recurrent exacerbations with slow recovery of symptoms [42] , it is highly relevant to further investigate new susceptibility factors to SARS-CoV-2 infections in COPD patients.

COPD appears to be underrepresented in the comorbidities reported for patients infected with COVID-19 [40, 43] , although several studies show an increased risk for a severe COVID-19 infection and increased mortality in patients with COPD [2, 40, 44] . Recent meta-analyses of early COVID-19 cohorts and more robust COVID-19 registry data from bigger centers [45] [46] [47] confirm that, although the infection rates do not differ between COPD and non-COPD, the risk of severe COVID-19 is increased in COPD, as indicated by increased hospitalization rates, ICU admission, and need for mechanical ventilation [45] [46] [47] . Interpretation of such studies is difficult because the general health status and comorbidities of COPD patients may be strongly associated with the outcome of SARS-COV-2 infection [48, 49] .

It has been suggested that increased COVID-19 mortality might be due to increased levels of ACE2 expression in the smaller airways of COPD patients and smokers [41] . Increased ACE2 expression was supported by our analysis, where a marked increase was found in various publicly available datasets of different COPD patient cohorts. In our explanted lung tissue, we could expand on these findings and show that not only ACE2 levels, but also TMPRSS2 and EZR are elevated in the lung tissue of COPD patients compared to control donor samples. Moreover, we could show increased ACE2 and TMPRSS2 expression levels not only in airway epithelial cells, but also in the lung parenchyma, including alveolar epithelial cells. While Leung et al [41] reported an inverse correlation of ACE2 levels with forced expiratory volume in 1 s (FEV 1 ) % predicted, our data indicated a positive correlation. This discrepancy might be due to the fact that Leung et al used ACE2 protein levels of small airway epithelium only, while we investigated ACE2 expression in the whole lung tissue.

The importance of the renin-angiotensin-aldosterone system (RAAS) in the pathophysiology of various pulmonary diseases has been established in recent years [50] [51] [52] [53] [54] [55] . In COPD, a dysbalance of vasoactive components of the RAAS has been postulated and therapeutically shifting the balance toward ACE2 provides beneficial effects in the patients [55, 56] . By binding angiotensin II, a key regulatory component of the RAAS, ACE2 acts in a vasodilative and anti-inflammatory manner, counterbalancing the proinflammatory, vasoconstrictive, and profibrotic ACE-angiotensin I axis [7] . The exact role of the ACE2-angiotensin II axis in the pathogenesis of COPD however remains unexplored. While an increase in pulmonary ACE2 might be a protective mechanism in response to the COPD-associated changes [57] , it could contribute to a worse COVID-19 outcome. Increased availability of SARS-CoV-2-susceptible cells expressing ACE2 together with TMPRSS2 and Ezrin in the distal bronchioles and lung parenchyma of COPD patients could facilitate more efficient entry and spread of SARS-CoV-2 viral particles within the distal parts of the lung, leading to severe disease progression. Therefore, administration of inhaled corticosteroids, an immunomodulatory medication, is currently being investigated to reduce disease susceptibility and mortality by downregulating ACE2 expression in COPD patients [58, 59] .

In addition, ACE2 can be released from the cell surface via proteases, including Adam17 and Adam10 [60, 61] . The soluble ectodomain of ACE2 (sACE2) maintains systemic protective effects and further binds and neutralizes SARS-CoV-2, thereby limiting virus propagation [60, 62] . The significant decrease of circulating sACE2 in COPD patients might therefore negatively influence disease outcome. The possible application of recombinant sACE2 is currently under investigation [32, 63, 64] . Taken together, our data showing increased local expression and decreased circulating levels of sACE2 could explain an elevated risk of severe COVID-19 pneumonia in COPD patients.

ACE2 expression did not correlate with mPAP or other hemodynamic parameters in COPD patients, indicating that the presence of PH per se might not be a risk factor for severe COVID-19. The analysis of a publicly available dataset of IPF patients with different degrees of PH did not show differences in the expression of any of the genes in our set between different PH levels. Recently, Sandoval et al investigated the ACE2-Ang(1-7) axis in human PAH and found increased serum ACE2 levels with a concomitant loss of ACE2 activity compared to healthy controls, suggesting a dysregulated RAAS in these patients [65] . However, in our IPAH patient group both locally in the lung or systemically in the plasma, we could not observe differences in expression levels regarding genes relevant for SARS-CoV-2 susceptibility and plasma ACE2 levels as compared to healthy controls.

In conclusion, our data offer a potential explanation for the increased risk of worse COVID-19 outcome in COPD patients, through increased expression levels of SARS-CoV-2 relevant genes ACE2 and TMPRSS2 in the distal lung with a concomitant decline of protective sACE2 levels in the circulation, independent of the presence or absence of PH.

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Chronic obstructive pulmonary disease is associated with severe coronavirus disease

Does comorbidity increase the risk of patients with COVID-19: evidence from meta-analysis

The impact of COPD and smoking history on the severity of COVID-19: a systemic review and meta-analysis

SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor

Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus

Pulmonary angiotensin-converting enzyme 2 (ACE2) and inflammatory lung disease

Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis

No evidence for basigin/CD147 as a direct SARS-CoV-2 spike binding receptor

Entry of SARS-CoV2 through the basal surface of alveolar endothelial cells -a proposed mechanism mediated by CD147 in COVID-19

Structure, function, and antigenicity of the SARS-CoV-2 spike glycoprotein

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and coronavirus disease-2019 (COVID-19): the epidemic and the challenges

Endothelial cell infection and endotheliitis in COVID-19

Age-dependent assessment of genes involved in cellular senescence, telomere, and mitochondrial pathways in human lung tissue of smokers, COPD, and IPF: associations with SARS-CoV-2 COVID-19 ACE2-TMPRSS2-Furin-DPP4 Axis

Distribution of ACE2, CD147, CD26, and other SARS-CoV-2 associated molecules in tissues and immune cells in health and in asthma, COPD, obesity, hypertension, and COVID-19 risk factors

Association of respiratory allergy, asthma, and expression of the SARS-CoV-2 receptor ACE2

Effect of asthma and asthma medication on the prognosis of patients with COVID-19

Asthma in COVID-19 hospitalizations: an overestimated risk factor?

COVID-19 in pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension: a reference centre survey

Could pulmonary arterial hypertension patients be at a lower risk from severe COVID-19?

Hypoxic vascular response and ventilation/perfusion matching in end-stage COPD may depend on p22phox

Distinct differences in gene expression patterns in pulmonary arteries of patients with chronic obstructive pulmonary disease and idiopathic pulmonary fibrosis with pulmonary hypertension

Compartment-specific expression of collagens and their processing enzymes in intrapulmonary arteries of IPAH patients

Integrative phenotyping framework (iPF): integrative clustering of multiple omics data identifies novel lung disease subphenotypes

Gene expression profiling in patients with chronic obstructive pulmonary disease and lung cancer

Multi-level immune response network in mild-moderate chronic obstructive pulmonary disease (COPD)

Comprehensive gene expression profiling identifies distinct and overlapping transcriptional profiles in non-specific interstitial pneumonia and idiopathic pulmonary fibrosis

Heterogeneous gene expression signatures correspond to distinct lung pathologies and biomarkers of disease severity in idiopathic pulmonary fibrosis

Gene expression profiling in the lungs of patients with pulmonary hypertension associated with pulmonary fibrosis

Single-cell RNA-seq reveals ectopic and aberrant lung-resident cell populations in idiopathic pulmonary fibrosis

Comprehensive integration of single-cell data

Human recombinant soluble ACE2 in severe COVID-19

A multibasic cleavage site in the spike protein of SARS-CoV-2 is essential for infection of human lung cells

Ezrin interacts with the SARS coronavirus spike protein and restrains infection at the entry stage

Emerging WuHan (COVID-19) coronavirus: glycan shield and structure prediction of spike glycoprotein and its interaction with human CD26

Neuropilin-1 facilitates SARS-CoV-2 cell entry and infectivity

Neuropilin1 as a new potential SARSCoV2 infection mediator implicated in the neurologic features and central nervous system involvement of COVID19

Single cell RNA-seq data analysis reveals the potential risk of SARS-CoV-2 infection among different respiratory system conditions

A single-cell RNA expression map of human coronavirus entry factors

Increased expression of ACE2, the SARS-CoV-2 entry receptor, in alveolar and bronchial epithelium of smokers and COPD subjects

ACE-2 expression in the small airway epithelia of smokers and COPD patients: implications for COVID-19

Human rhinovirus infection during naturally occurring COPD exacerbations

Do chronic respiratory diseases or their treatment affect the risk of SARS-CoV-2 infection?

Effect of chronic obstructive pulmonary disease and smoking on the outcome of COVID-19

COVID-19 and COPD

SARS-CoV-2 infection in the COPD population is associated with increased healthcare utilization: an analysis of Cleveland clinic's COVID-19 registry

Comorbidity and its impact on 1590 patients with COVID-19 in China: a nationwide analysis

Cardiovascular comorbidities in chronic obstructive pulmonary disease (COPD) -current considerations for clinical practice

Cardiovascular morbidity and mortality in COPD

Dysregulated renin-angiotensin-aldosterone system contributes to pulmonary arterial hypertension

A potential therapeutic role for angiotensin-converting enzyme 2 in human pulmonary arterial hypertension

Chronic activation of the renin-angiotensin system induces lung fibrosis

Angiotensin II and the fibroproliferative response to acute lung injury

Angiotensin-converting enzyme inhibition as an adjunct to pulmonary rehabilitation in chronic obstructive pulmonary disease

The association of renin-angiotensin system blockades and pneumonia requiring admission in patients with COPD

Targeting the renin-angiotensin system as novel therapeutic strategy for pulmonary diseases

Angiotensin-converting enzyme-2 overexpression attenuates inflammation in rat model of chronic obstructive pulmonary disease

Inhaled corticosteroids downregulate the SARS-CoV-2 receptor ACE2 in COPD through suppression of type I interferon

Risk of COVID-19-related death among patients with chronic obstructive pulmonary disease or asthma prescribed inhaled corticosteroids: an observational cohort study using the OpenSAFELY platform

Ectodomain shedding of angiotensin converting enzyme 2 in human airway epithelia

Tumor necrosis factoralpha convertase (ADAM17) mediates regulated ectodomain shedding of the severe-acute respiratory syndrome-coronavirus (SARS-CoV) receptor

Inhibition of SARS-CoV-2 infections in engineered human tissues using clinical-grade soluble human ACE2

Blockade of SARS-CoV-2 infection by recombinant soluble ACE2

Human recombinant soluble ACE2 (hrsACE2) shows promise for treating severe COVID-19

Angiotensin converting enzyme 2 and angiotensin (1-7) axis in pulmonary arterial hypertension

We would like to thank Elisabeth Blanz and Thomas Fuchs for excellent technical assistance. PhD candidate EF was funded by the Austrian Research Promotion Agency (FFG, project number: 34926649) and trained within the frame of the PhD program Molecular Medicine of the Medical University of Graz. LMM is supported by FWF grant (KLI 884-B) .

EF, AB and GK contributed to the conception and design of the study. EF, AB, EG and WK were responsible for sample and data acquisition. LMM performed scRNAseq analysis. EF, AB, LMM, HO and GK were involved in data analysis and interpretation. EF and AB drafted the first manuscript and all authors critically revised the manuscript.

 Figure S1 . Expression profiling of SARS-CoV-2 receptors and processing enzymes in publicly available datasets of IPF and control lungs Figure S2 . Immunostaining indicating tissue distribution of SARS-CoV-2 receptor ACE2 in COPD and donor lung tissue 13 Linking COVID-19 severity and chronic obstructive pulmonary disease