key: cord-0752467-aea534dm
authors: Finney, Lydia J.; Glanville, Nicholas; Farne, Hugo; Aniscenko, Julia; Fenwick, Peter; Kemp, Samuel V.; Trujillo-Torralbo, Maria-Belen; Loo, Su Ling; Calderazzo, Maria Adelaide; Wedzicha, Jadwiga A.; Mallia, Patrick; Bartlett, Nathan W.; Johnston, Sebastian L.; Singanayagam, Aran
title: Inhaled corticosteroids downregulate the SARS-CoV-2 receptor ACE2 in COPD through suppression of type I interferon
date: 2020-10-15
journal: J Allergy Clin Immunol
DOI: 10.1016/j.jaci.2020.09.034
sha: 97acf17564239fe170cc68c9514daf5909f7d999
doc_id: 752467
cord_uid: aea534dm

Background The mechanisms underlying altered susceptibility and propensity to severe Coronavirus disease 2019 (COVID-19) disease in at-risk groups such as patients with chronic obstructive pulmonary disease (COPD) are poorly understood. Inhaled corticosteroids (ICS) are widely used in COPD but the extent to which these therapies protect or expose patients to risk of severe COVID-19 is unknown. Objective The aim of this study was to evaluate the effect of ICS upon pulmonary expression of the SARS-CoV-2 viral entry receptor angiotensin-converting enzyme (ACE)-2. Methods We evaluated the effect of ICS administration upon pulmonary ACE2 expression in vitro in human airway epithelial cell cultures and in vivo in mouse models of ICS administration. Mice deficient in the type I interferon-α/β receptor (Ifnar1 −/−) and exogenous interferon-β administration experiments were used to study the functional role of type-I IFN signalling in ACE2 expression. We compared sputum ACE2 expression in patients with COPD stratified according to use or non-use of ICS. Results ICS administration attenuated ACE2 expression in mice, an effect that was reversed by exogenous interferon-β administration and Ifnar1 −/− mice had reduced ACE2 expression, indicating that type I interferon contributes mechanistically to this effect. ICS administration attenuated expression of ACE2 in COPD airway epithelial cell cultures and in mice with elastase-induced COPD-like changes. COPD patients taking ICS also had reduced sputum expression of ACE2 compared to non-ICS users. Conclusion ICS therapies in COPD reduce expression of the SARS-CoV-2 entry receptor ACE2. This effect may thus contribute to altered susceptibility to COVID-19 in patients with COPD.

widely used in COPD but the extent to which these therapies protect or expose patients to risk of 48 severe COVID-19 is unknown. reported in smokers and subjects with COPD 14 and is postulated to be a factor predisposing these 131 individuals to adverse outcome from COVID-19. Conversely, ACE2 is downregulated in asthma 15 , an 132 effect that may be due to suppressive effects of type 2 cytokines 16 or related to ICS use 17 . Emerging 133 evidence also indicates that ACE2 expression co-localizes with immune genes involved in interferon 134 J o u r n a l P r e -p r o o f signalling pathways. 18 

Pulmonary ACE2 mRNA expression is reduced in COPD patients taking inhaled corticosteroids.

Recent data indicate that sputum expression of ACE2 mRNA is reduced in asthmatic subjects taking 212 ICS 17 but it is unclear whether similar suppression occurs in the context of COPD. We therefore 213 initially used a community-based cohort of 40 COPD subjects 20 to determine whether ICS use affects 214 ACE2 expression in COPD. 36 out of 40 subjects had sufficient sample for evaluation and were 215 stratified according to current use (n=18) or non-use (n=18) of ICS. There were no significant 216 differences between these groups in terms of age, disease severity, smoking status or other 217 comorbidities known to affect ACE2 expression and/or associated with increased risk of COVID-19 ( 218 Table 1 ). Sputum cell ACE2 mRNA expression was detectable in 22/36 COPD subjects (61.1%) and, 219 consistent with prior observations in asthma 17 , significantly reduced in ICS users compared to non-220 users (Fig 1a) . This effect was consistently observed in a sub-group analysis following exclusion of 221 patients with significant bronchodilator reversibility (>12%) (Supplementary Fig 1) . Sputum cell 222 expression of the serine protease TMPRSS2 that is used by SARS-CoV-2 for mucosal entry 24 was 223 detectable in all subjects with no significant difference observed between ICS users and non-users 224 (Fig 1b) . Similarly, the alternative SARS-CoV receptor CD147 25 (gene: Basigin BSG) was detectable in 225 all subjects with no difference observed between ICS users and non-users (Supplementary Fig 2) .

Given that cause and effect cannot be inferred from a cross-sectional human study, we next have also previously shown that significant GR activation occurs 8 ). This effect persisted at 24 hours 234 post-administration but had resolved from 48 hours onwards (Fig 2b) . Consistent with effects 235 J o u r n a l P r e -p r o o f observed in human sputum, FP administration had no effect expression of Tmprss2 or Bsg in mouse 236 lung (Supplementary Fig 3) . Suppression of Ace2 by FP occurred in a dose-dependent manner with 237 loss of suppression at a ten-fold lower concentration (2µg) (Fig 2c) , a dose at which effects on GR 238 activation are also lost 8 . We observed similar suppression of lung Ace2 mRNA expression with 239 administration of 20µg of other commonly used ICS budesonide and beclomethasone, suggesting 240 that the effect of ICS on Ace2 is not class dependent (Fig 2d) . To corroborate the effects observed on 241 Ace2 mRNA expression, we subsequently measured protein levels in lung homogenate of ICS-treated 242 mice by ELISA. We observed similar suppression of total lung ACE2 protein occurring at 24 hours 243 post-administration, consistent with effects observed at the mRNA level (Fig 2e) . (Fig 3a) . Given our 252 previous data showing that COPD patients treated with ICS have reduced basal airway expression of 253 IFNβ 8 , we hypothesized that downregulation of ACE2 by FP may be functionally related to its 254 suppressive effects on type I IFN signalling. Accordingly, recombinant IFN-β administration (Fig 3b) 255 could reverse FP-mediated suppression of Ace2 mRNA and ACE2 protein (Fig 3c) , indicating that the 256 effect of FP on ACE2 expression is functionally related to suppressive effects on type I IFN.

Ifnar -/mice have reduced basal expression of ACE2

To further confirm the functional importance of type I IFN in regulating pulmonary ACE2, we 260 evaluated basal pulmonary expression levels in mice deficient in IFN signalling (Ifnar -/-). Compared to 261 wild-type control mice, Ifnar -/mice had a small, but statistically significant, reduction in lung ACE2 262 mRNA expression (Fig 4a) with a concomitant trend (P=0.15) towards reduced lung ACE2 protein 263 levels (Fig 4b) . These observations further confirm the key regulatory role played by type I IFN 264 signalling in pulmonary expression of ACE2.

The suppressive effect of ICS upon ACE2 expression occurs at the bronchial epithelium

Existing data indicate that ACE2 is expressed primarily in the nasal and bronchial epithelium and is 268 absent from immune cells 16 . Given our prior data indicating that FP also exerts its inhibitory effects 269 J o u r n a l P r e -p r o o f on immunity principally at the pulmonary epithelium 8, 22 , we next assessed whether suppressive 270 effects on ACE2 were also observed following ex vivo ICS administration in cultured COPD bronchial 271 epithelial cells (BECs, Fig 5a) . Baseline characteristics of the subjects included in these analyses are 272 shown in table 2. In keeping with recent in situ expression studies in COPD patients 27 , we found that 273 basal expression of ACE2 was increased by ~3 fold in BECs from COPD patients compared to healthy 274 non-smokers (Fig 5b) . Consistent with our findings in human ICS users and in the mouse model of ICS 275 administration (Figs 1&2) , FP administration (at a clinically relevant concentration of 10nM) induced 276 ~75% suppression in ACE2 expression (Fig 5c) . There was no effect of FP administration on TMPRSS2 277 expression in COPD BECs (Supplementary Fig 4) . (Fig 6a) . and lung ACE2 expression was measured at 10 days after administration (timepoint at which 284 COPD-like disease features are established 28 ) and a further 7 days later. In keeping with our findings 285 in human COPD cells, elastase-treated mice had significantly increased (~5-fold) lung Ace2 286 expression at 10 days with further enhancement to >15-fold at 17 days (Fig 6b) . Administration of a 287 single dose of FP at 10 days attenuated the significant upregulation of lung Ace2 mRNA (Fig 6c) and 288 ACE2 protein concentrations measured 24 hours later in elastase-treated mice (Fig 6d) but had no 289 effect upon Tmprss2 (Supplementary Fig 5) . Therefore, the suppressive effects of ICS on ACE2 also 

Currently, a direct causal link between ACE2 and increased susceptibility to acquisition of SARS-CoV-354 2 or subsequent severity has not been proven and we cannot conclude unequivocally that 355 downregulation of ACE2 by ICS is an effect that would confer protection clinically. In asthma, a 356 disease where ICS are more commonly prescribed than in COPD, ACE2 expression is reduced 357 compared to healthy subjects 15 and also further attenuated in ICS users 17 . In contrast to COPD, 358 asthma has not been shown to be associated with increased COVID-19 mortality 43 and the more 359 widespread use of ICS with associated suppression of ACE2 could theoretically be one factor driving 360 this. Conversely, It is important to note that there is evidence to suggest that downregulation of 361 ACE2 could also theoretically worsen outcome. In mouse models of experimentally-induced acid 362 aspiration and sepsis, genetic deletion of Ace2 worsens acute lung injury, an effect that is partially 564  565  566  567  568  569  570  571  572  573  574  575  576  577  578  579  580  581  582  583  584  585  586  587  588  589  590  591  592  593 

422 guidance-on-social-distancing-and-for-vulnerable-people/guidance-on-social-423 distancing-for-everyone-in-the-uk-and-protecting-older-people-and-vulnerable-424 adults

Do chronic respiratory diseases or 428 their treatment affect the risk of SARS-COV-2 infection?

Factors 431 associated with COVID-19-related death using OpenSAFELY

Severity and Mortality associated with COPD and Smoking in patients 434 with COVID-19: A Rapid Systematic Review and Meta-Analysis

Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary 438 disease

The 440 prevention of chronic obstructive pulmonary disease exacerbations by 441 salmeterol/fluticasone propionate or tiotropium bromide

Corticosteroid suppression of antiviral immunity increases bacterial loads and mucus 445 production in COPD exacerbations

Glucocorticosteroids enhance 447 replication of respiratory viruses: effect of adjuvant interferon

Human 449 Metapneumovirus Infection in Chronic Obstructive Pulmonary Disease: Impact of 450 Glucocorticosteroids and Interferon

Inhibitory effects of glycopyrronium, formoterol, and budesonide on coronavirus 453

HCoV-229E replication and cytokine production by primary cultures of human nasal 454 and tracheal epithelial cells

The inhaled 456 corticosteroid ciclesonide blocks coronavirus RNA replication by targeting viral 457 NSP15

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically 460 Proven Protease Inhibitor

ACE-2 462 expression in the small airway epithelia of smokers and COPD patients: implications 463 for COVID-19

465 Association of Respiratory Allergy

Type 2 468 and interferon inflammation strongly regulate SARS-CoV-2 related gene expression 469 in the airway epithelium

Sputum Cells in Asthma: Relationship to Demographic Features 473 and Corticosteroids

SARS-CoV-2 475 entry factors are highly expressed in nasal epithelial cells together with innate 476 immune genes

Receptor ACE2 Is an Interferon-Stimulated Gene in Human Airway Epithelial Cells 479 and Is Detected in Specific Cell Subsets across Tissues

COPD patients with frequent exacerbations

Role 484 of airway glucose in bacterial infections in patients with chronic obstructive 485 pulmonary disease

487 Inhaled corticosteroid suppression of cathelicidin drives dysbiosis and bacterial 488 infection in chronic obstructive pulmonary disease

Beclomethasone has lesser suppressive effects on inflammation and anti-bacterial 491 immunity than Fluticasone or Budesonide in experimental infection models

SARS-CoV-2 Cell entry 494 depends on ACE2 and TMPRSS2

healthy control subjects were cultured ex vivo and treated with 10nM fluticasone propionate (FP) or 637 medium control. Cell lysates were collected. ACE2 mRNA expression was measured at baseline (b) 638 and at 24h after FP administration (c) in COPD BECs. Data shown as median (+/-IQR)

Figure 6 ACE2 expression is increased in a mouse model of COPD and suppressed by fluticasone 642 administration. (a) C57BL/6 were treated intranasally with a single dose of elastase or PBS as control 643 and lung tissue harvested at 10 or 17 days later

Lung Ace2 645 mRNA expression in elastase-versus PBS-treated mice at 10 and 17 days after treatment. (c) At 10 646 days after elastase treatment, C57BL/6 mice were treated with a single dose of 20µg fluticasone 647 propionate (FP) or vehicle (VEH) control. Lung tissue was harvested at 24 hours after FP 648 administration. ACE2 mRNA was measured by qPCR

Data shown mean (+/-) SEM of 5 mice per treatment group from a single 650 experiment. representative of at least two independent expreriments. Data analysed by one way 651 ANOVA with Bonferroni post-test

 Supplementary Fig 1: Sputum gene expression of ACE2 in COPD subjects stratified according to use or non-use of inhaled corticosteroids excluding subjects with bronchodilator reversibility. Sputum samples were taken from a cohort of patients with COPD when clinically stable for at least 6 weeks. Patients were stratified according to current use or non-use of inhaled corticosteroids (ICS),(n=17 ICS users, n=15 non ICS users). Sputum cell mRNA expression of ACE2 was measured by quantitative PCR. Box-and-whisker plots show median (line within box)), interquartile range (IQR; box) and minimum to maximum (whiskers). Statistical comparisons made using Mann Whitney U test. *P<0.05Supplementary Figure 2 : Sputum gene expression of BSG in COPD subjects stratified according to inhaled corticosteroid use. Sputum samples were taken from a cohort of patients with COPD when clinically stable for at least 6 weeks. Patients were stratified according to current use or non-use of inhaled corticosteroids (ICS) (n=18 per group). Sputum cell mRNA expression of BSG was measured by quantitative PCR. Box-and-whisker plots show median (line within box)), interquartile range (IQR; box) and minimum to maximum (whiskers). Statistical comparisons made using Mann Whitney U test.