key: cord-321427-bwcpd6im
authors: Yee, Min; Cohen, E. David; Haak, Jeannie; Dylag, Andrew M.; O’Reilly, Michael A.
title: Neonatal hyperoxia enhances age-dependent expression of SARS-CoV-2 receptors in mice
date: 2020-07-22
journal: bioRxiv
DOI: 10.1101/2020.07.22.215962
sha: 
doc_id: 321427
cord_uid: bwcpd6im

The severity of COVID-19 lung disease is higher in the elderly and people with pre-existing co-morbidities. People who were born preterm may be at greater risk for COVID-19 because their early exposure to oxygen at birth increases their risk of being hospitalized when infected with RSV and other respiratory viruses. Our prior studies in mice showed how high levels of oxygen (hyperoxia) between postnatal days 0–4 increases the severity of influenza A virus infections by reducing the number of alveolar epithelial type 2 (AT2) cells. Because AT2 cells express the SARS-CoV-2 receptors angiotensin converting enzyme (ACE2) and transmembrane protease/serine subfamily member 2 (TMPRSS2), we expected their expression would decline as AT2 cells were depleted by hyperoxia. Instead, we made the surprising discovery that expression of Ace2 and Tmprss2 mRNA increases as mice age and is accelerated by exposing mice to neonatal hyperoxia. ACE2 is primarily expressed at birth by airway Club cells and becomes detectable in AT2 cells by one year of life. Neonatal hyperoxia increases ACE2 expression in Club cells and makes it detectable in 2-month-old AT2 cells. This early and increased expression of SARS-CoV-2 receptors was not seen in adult mice who had been administered the mitochondrial superoxide scavenger mitoTEMPO during hyperoxia. Our finding that early life insults such as hyperoxia enhances the age-dependent expression of SARS-CoV-2 receptors in the respiratory epithelium helps explain why COVID-19 lung disease is greater in the elderly and people with pre-existing co-morbidities.

of ACE2 was examined in the lungs of mice between PND4 and 2 years of age by 110 immunohistochemistry so as to better understand the temporal spatial pattern of its expression. ACE2 111 was primarily detected in airway epithelial cells with minimal staining seen in the alveolar space 112 (Figure 1a) . The intensity of ACE2 staining increased steadily in the airway epithelium throughout the 113 life of the mouse. A rare ACE2-positive alveolar cells (arrows) was first observed on PND7 and then 114 steadily increased in number between 6 and 24 months of age. Western blotting for ACE2 confirmed 115 that the abundance of ACE2 protein became progressively enriched in the whole lungs of 12-and 24-116 month-old mice relative to those of mice harvested at 2 months of age (Figure 1b) . ACE2 mRNA 117 levels were similarly increased in the whole lungs of 24-month-old mice than in those of mice 118 harvested at 2 months of age (Figure 1c) .

showed extensive co-localization along the airways at both 2 and 12 months of age (Figure 2a) , but 121 the intensity of ACE2 staining was significantly higher at 12 months of age than at 2 months of age 122 (Figure 2b) . Co-staining for ACE2 and the AT2 cell marker proSP-C revealed that the vast majority of 123 ACE2+ cells in the alveoli were AT2 cells (Figure 2c ). Approximately 20% of proSP-C+ AT2 cells 124 expressed ACE2 at 2 months while 80% of proSP-C+ AT2 cells expressed it at 12 months ( Figure   125 2d). These findings reveal that ACE2 is primarily expressed by the airway Club cells of young adult 126 mice but becomes increasingly expressed by AT2 cells as mice age.

40% oxygen was not sufficient to induce Ace2 mRNA, the levels of Ace2 expression was significantly 139 higher in mice exposed to 60% and 80% oxygen relative to controls. Exposing mice to a low chronic 140 dose of oxygen (40% for 8 days) that does not alter alveolar development 24 also failed to increase cumulative dose of oxygen than 60% for 4 days, these findings suggest that oxygen alone may not be 143 stimulating Ace2 expression.

Immunohistochemistry was used to further understand how hyperoxia affected ACE2 145 expression in the adult lung. While neonatal hyperoxia increased intensity of ACE2 staining in the 146 airway, it most obviously increased the number of alveolar cells with detectable ACE2 (Figure 4a ).

When quantified, neonatal hyperoxia increased the number of alveolar cells expressing ACE2 by 148 approximately 50% at 2, 6 and 12 months of age (Figure 4b) . The increased alveolar expression 149 seen at 2 months of age was primarily attributed to increased expression by proSP-C+ AT2 cells; 150 however, this difference resolved at 6 and 12 months of age as more AT2 cells in control lungs began 151 to express ACE2 (Figure 4c ).

Anti-oxidants block oxygen-dependent changes in ACE2 expression. Prior studies by us and other 154 investigators showed that administering the mitochondrial superoxide scavenger mitoTEMPO to mice 155 during exposure to hyperoxia (Figure 5a ) prevents the alveolar simplification and cardiovascular 156 disease observed when these mice reach adulthood 25-27 . qRT-PCR revealed administering 7 staining in control mice, it reduced the numbers of alveolar ACE2+ cells in the lungs of hyperoxia-162 exposed mice lower than controls.

Neonatal hyperoxia stimulates age-dependent changes in TMPRSS2. TMPRSS2 is an 165 endoprotease expressed by respiratory epithelial cells that facilitates viral entry of coronaviruses into epithelial cells 9 . The levels of Tmprss2 mRNA and protein were examined in the lungs of 2-, 12-and 18-month-old mice that were exposed to neonatal hyperoxia and room air from PND0-4 by qRT-PCR 168 and western blotting. Tmprss2 mRNA was readily detected in the lungs of 2-month-old mice, and 169 increased ~5-fold at 12 months and ~8-fold at 18 months (Figure 6a) . Neonatal hyperoxia further showed that the levels of TMPRSS2 protein were higher in the whole lung lysates of mice exposed to 172 neonatal hyperoxia than in those of control mice (Figure 6b ). As observed for Ace2 expression, 173 exposure to ≥ 60% oxygen from PND4-0 was required to significantly increase the levels of Tmprss2 174 mRNA in the lungs of mice at 2 months of age (Figure 6c ). Exposure to 40% oxygen from PND0-8 175 also failed to change Tmprss2 expression in adult mice (data not shown) while the administration of 176 mitoTEMPO to mice during exposure blunted the effects of neonatal hyperoxia on Tmprss2 mRNA 177 (Figure 6d) . Together, these findings suggest age and neonatal hyperoxia have similar effects on 178 increasing TMPRSS2 as they do for ACE2.

180 . CC-BY-ND 4.0 International license was not certified by peer review) is the author/funder. It is made available under a The copyright holder for this preprint (which this version posted July 22, 2020. . https://doi.org/10.1101/2020.07.22.215962 doi: bioRxiv preprint expanded rapidly to become one of the worst pandemics to ever challenge the modern world. While cancer 28 . Those with multiple co-morbidities have a higher rate of mortality. People born preterm may 186 also be at great risk for COVID-19 because they often suffer from multiple co-morbidities due, in part,

to their lungs being exposed to oxygen too soon or to super-physiological concentrations used to 188 maintain appropriate blood oxygen saturations. It is unclear whether co-morbidities increase disease 189 by changing spatial and temporal expression of SARS-CoV-2 receptors or the immune response that 190 leads to a lethal cytokine storm 1 . In this study, we present evidence that expression of the SARS-

CoV-2 co-receptors ACE2 and TMPRSS2 increase in the respiratory epithelium of mice as they age 

We found that ACE2 was primarily expressed by airway Club cells during early postnatal life.

The intensity of ACE2 staining increased in the airways of mice with age and became detectable in 202 the alveoli of young adult mice. Co-localization with proSP-C revealed that most, but not all alveolar The copyright holder for this preprint (which this version posted July 22, 2020. . https://doi.org/10.1101/2020.07.22.215962 doi: bioRxiv preprint expression continues to increase as mice age. We also found that Tmprss2 mRNA expression 208 increases as mice age and this expression was similarly enhanced by neonatal hyperoxia. While AT2 209 cells have previously been shown to express TMPRSS2 11 , we were not able to detect it in the mouse 210 lung using commercially available antibodies. However, we did find that the abundance of Tmprss2 211 mRNA and protein abundance increased with age and neonatal hyperoxia, and was reduced by 212 mitoTEMPO similar to that of Ace2. The higher expression of these genes as mice age is in 213 agreement with recent review that discussed two unpublished studies deposited in bioRxiv showing 214 how expression of Ace2 and Tmprss2 mRNA increases with age in human respiratory epithelium 1 . receptors may be responsible for increasing the severity of COVID-19 lung disease in elderly people.

It is important to recognize the normal functions of ACE2 and TMPRSS2 because that may 218 help explain why their expression steadily increases with age 32 . ACE2 is perhaps best known for its 219 role in controlling blood pressure in the renin-angiotensin system 33 . ACE1 converts the 10-amino acid 220 angiotensin I to an 8-amino acid vasoconstrictive peptide called angiotensin II. ACE2 accumulates in 221 people with pulmonary hypertension and hydrolyzes Angiotensin II to Ang(1-7), which has 222 vasodilation properties. Over-expressing ACE2 also protects against right ventricular hypertrophy 34 .

Hence, higher levels of ACE2 seen as the lung ages may reflect an adaptive response designed to 

TMPRSS2 is a serine protease that is localized to the apical surface of secretory cells such as Club 233 and AT2 cells of the lung 39 . Its expression is highly regulated by androgens in the prostate gland and 234 . CC-BY-ND 4.0 International license was not certified by peer review) is the author/funder. It is made available under a The copyright holder for this preprint (which this version posted July 22, 2020. . https://doi.org/10.1101/2020.07.22.215962 doi: bioRxiv preprint may be similarly responsive to androgens in the lung, suggesting it may play a role in sex-dependent 235 differences in the lung.

Our study also found that neonatal hyperoxia increased or accelerated expression of Ace2 237 mRNA, ACE2 protein, and Tmprss2 mRNA as mice age. Significant changes were seen with 60% or 238 more FiO2 at 8 weeks (2 months) of age and persisted as mice age. How hyperoxia regulates 239 expression of these proteins is conflicting and remains to be better understood. One study using 240 human fetal IMR-90 fibroblasts found that hyperoxia does not change expression of ACE2 40 .

However, ACE2 was depleted when cells returned to room air presumably because it was being 242 proteolyzed and shed into the media. In contrast, another study found higher levels of ACE2 in 243 newborn rats exposed to 95% oxygen for the first week of life and then recovered in 60% oxygen for 244 the next two weeks 41 . In our hands, changes in Ace2 or Tmprss2 mRNA were first detected in 8-

week-old mice exposed to hyperoxia between PND0-4. We did not detect changes at the end of 246 oxygen exposure (PND4). In fact, we recently deposited an RNA-seq analysis of AT2 cells isolated 247 from PND4 mice exposed to room air versus hyperoxia that shows hyperoxia modestly inhibits Ace2 

60% for 4 days but not 40% for 4 or 8 days), we speculate that they occur as an adaptive response to 253 the alveolar simplification and cardiovascular disease as mice exposed to neonatal oxygen age. The 

Genetic studies in mice suggest mutant forms of SP-C that activate the UPR are not sufficient by 281 themselves to cause fibrotic lung disease. However, they can predispose the lung to fibrotic disease 282 following viral infections 47 . Familial forms of IPF that activate the UPR in AT2 cells may therefore 283 accelerate the age-dependent susceptibility of AT2 cells to SARS-CoV-2 infections.

In summary, we found that neonatal hyperoxia increases or accelerates the age-dependent relative to control mice exposed to room air. (c) qRT-PCR was used to measure Tmprss2 mRNA in 542 total lung homogenates of 2 month mice exposed to room air, 40%, 60%, or 80% oxygen between 543 PND0-4. (d) qRT-PCR was used to measure Tmprss2 mRNA in control and 2-month-old mice 544 exposed to room air or hyperoxia and vehicle or mitoTEMPO between PND0-4 N=4-5 mice per group.

Statistical significance is comparisons for all pairs using Tukey-Kramer HSD test with *P≤0.05; 546 **P≤0.01.

. CC-BY-ND 4.0 International license was not certified by peer review) is the author/funder. It is made available under a The copyright holder for this preprint (which this version posted July 22, 2020. . https://doi.org/10.1101/2020.07.22.215962 doi: bioRxiv preprint 

Neonatal hyperoxia causes pulmonary vascular disease and shortens life span 445 in aging mice

Genetic toxicology of oxygen

COVID-19 spike-host cell 449 receptor GRP78 binding site prediction

Signal integration in the endoplasmic reticulum unfolded protein 452 response

Idiopathic pulmonary fibrosis: a disorder of