Angiotensin-converting enzyme 2 (ACE2) expression increases with age in patients requiring mechanical ventilation

Abstract

Mortality due to Covid-19 is highly associated with advanced age, owing in large part to severe lower respiratory tract infection. SARS-CoV-2 utilizes the host ACE2 receptor for infection. Whether ACE2 abundance in the lung contributes to age-associated vulnerability is currently unknown. We set out to characterize the RNA and protein expression profiles of ACE2 in aging human lung in the context of phenotypic parameters likely to affect lung physiology. Examining publicly available RNA sequencing data, we discovered that mechanical ventilation is a critical variable affecting lung ACE2 levels. Therefore, we investigated ACE2 protein abundance in patients either requiring mechanical ventilation or spontaneously breathing. ACE2 distribution and expression were determined in archival lung samples by immunohistochemistry (IHC). Tissues were selected from the specimen inventory at a large teaching hospital collected between 2010-2020. Twelve samples were chosen from patients receiving mechanical ventilation for acute hypoxic respiratory failure (AHRF). Twenty samples were selected from patients not requiring ventilation. We compared samples across age, ranging from 40-83 years old in the ventilated cohort and 14-80 years old in the non-ventilated cohort. Within the alveolated parenchyma, ACE2 expression is predominantly observed in type II pneumocytes (or alveolar type II / AT2 cells) and alveolar macrophages. All 12 samples from our ventilated cohort showed histologic features of diffuse alveolar damage including reactive, proliferating AT2 cells. In these cases, ACE2 was strongly upregulated with age when normalized to lung area (p = 0.004) or cellularity (p = 0.003), associated with prominent expression in AT2 cells. In non-ventilated individuals, AT2 cell reactive changes were not observed and ACE2 expression did not change with age when normalized to lung area (p = 0.231) or cellularity (p = 0.349). In summary, ACE2 expression increases with age in the setting of alveolar damage observed in patients on mechanical ventilation, providing a potential mechanism for higher Covid-19 mortality in the elderly.

Fig 1. ACE2 RNA expression increases with age and mechanical ventilation in the lung.

A) Human lung ACE2 expression is colored by decade of life when the individual died. In unadjusted linear regression, there was no significant effect of age on ACE2 expression (p = 0.227, n = 578 individuals). B) ACE2 expression in lung stratified by sex. Analysis adjusted for age and sex, revealed no significant effect for either variable (p = 0.228 for age, p = 0.920 for sex, multivariate regression, n = 395 males, n = 183 females). There was also no significant difference when compared by sex alone (p = 0.303, Wilcoxon rank-sum test). C) ACE2 expression in lung stratified by the Hardy scale. This scale indicates the length of time spent in the terminal phase before death, which is depicted above each grouping of data points. Within each Hardy scale group, the data are sub-stratified by age. (n = 26 for a score of 1 representing a violent and fast death lasting <10 minutes, n = 156 for a score of 2 representing a fast death by natural causes lasting 10 minutes– 1 hour, n = 31 for a score of 3 representing an intermediate rate of death lasting 1 hour– 24 hours, n = 64 for a score of 4 representing a slow death with a terminal phase lasting > 24 hours, n = 299 for a score of 0 representing donors supported by a ventilator preceding death, n = 2 with an unknown score). For all panels, each point represents a sample from a unique individual. Box plots indicate quartiles. A linear model fit to the data is inset, indicating the estimated coefficient for age (β₁) and its significance.

Fig 2. IHC for ACE2 highlights AT2 cells and alveolar macrophages in normal lung.

In normal human lung from a 23-year-old female using a 1x objective (left, scale bar 3mm), strong ACE2 expression is observed within bronchioles and alveoli. The region outlined by the red dashed box is shown magnified at low power (middle, scale bar 200μm) which highlights prominent ACE2 expression in AT2 cells (red arrowheads) along the alveolar septum and in alveolar macrophages (green arrowhead). At high power (right, scale bar 50μm), ACE2 staining can be seen concentrated along the membrane of AT2 cells (red arrowheads) and within the cytoplasm of an alveolar macrophage (green arrowhead). Sections were stained for ACE2 using DAB and counterstained with hematoxylin.

Fig 3. ACE2 protein expression increases with age in ventilated patients.

A) Representative images of lung stained for ACE2 from a 40-year-old man with acute lung injury are shown using a 1x objective (left, scale bar 3mm) with the region outlined by the red dashed box magnified at low power (middle, scale bar 200μm) and a second field at high power (right, scale bar 50μm). Clusters of reactive AT2 cells (red arrowheads) are present along the alveolar septum which exhibit low level ACE2 expression. B) Representative images of lung stained for ACE2 from a 67-year-old man with acute lung injury superimposed on fibrosing interstitial lung disease are shown using a 1x objective (left, scale bar 3mm) with the region outlined by the red dashed box magnified at low power (middle, scale bar 200μm) and a second field at high power (right, scale bar 50μm). Numerous reactive AT2 cells (red arrowheads) exhibiting nucleomegaly and abundant cytoplasm can be seen demonstrating strong ACE2 staining. C) Quantitative IHC for ACE2 was carried out on samples from ventilated patients (n = 12 samples from 11 patients). Total ACE2 expression from 5 low power fields is plotted relative to the patient’s age at the time of specimen collection. A linear fit to the data is indicated by the dashed line with the 95% confidence interval highlighted in grey. D) The same specimens quantitated in (C) were normalized by cellularity and the average ACE2 expression per cell is plotted along with a linear fit to the data and its 95% confidence interval. The red arrow in (C) and (D) indicates a patient providing 1 sample from the left lung and 1 sample from the right lung during the same procedure, utilized as a control for intra-individual reproducibility. The green and blue circle in (C) and (D) indicate the staining intensity of the samples depicted in Fig 5A and 5B, respectively. Sections were stained for ACE2 using DAB and counterstained with hematoxylin.

Fig 4. ACE2 protein expression does not change with age in non-ventilated patients.

A) Representative images of lung stained for ACE2 from a 26-year-old female who underwent blebectomy for a spontaneous pneumothorax are shown using a 1x objective (left, scale bar 3mm) with the region outlined by the red dashed box magnified at low power (middle, scale bar 200μm) and a second field at high power (right, scale bar 50μm). The normal alveolated lung parenchyma reveals AT2 cells which are positive for ACE2 (red arrowheads) scattered among largely ACE2 low type I pneumocytes. B) Representative images of lung stained for ACE2 from a 77-year-old man undergoing wedge resection for suspected metastatic angiosarcoma are shown using a 1x objective (left, scale bar 3mm) with the region outlined by the red dashed box magnified at low power (middle, scale bar 200μm) and a second field at high power (right, scale bar 50μm). Histologic findings in the normal alveolated parenchyma adjacent to the lesion were similar to those found in (A) with strong ACE2 staining in scattered AT2 cells (red arrowheads) and occasional alveolar macrophages. C) Quantitative IHC for ACE2 was carried out on samples from non-ventilated patients (n = 20 patients). Total ACE2 expression from 5 low power fields is plotted relative to the patient’s age at the time of specimen collection. A linear fit to the data is indicated by the dashed line with the 95% confidence interval highlighted in grey. D) The same specimens quantitated in (C) were normalized by cellularity and the average ACE2 expression per cell is plotted along with a linear fit to the data and its 95% confidence interval. Sections were stained for ACE2 using DAB and counterstained with hematoxylin.

Fig 5. Two patients on ACEI/ARB therapy exhibit intense endothelial ACE2 expression.

A) Representative images of lung stained for ACE2 from a 53-year-old man with organizing diffuse alveolar damage superimposed on fibrosing interstitial lung disease are shown using a 1x objective (left, scale bar 3mm) with the region outlined by the red dashed box magnified at low power (middle, scale bar 200μm) and second field at high power (right, scale bar 50μm). AT2 cell hyperplasia can be seen with lower level ACE2 expression present along the alveolar septum (green arrowheads). Strongly stained endothelial cells (red arrowheads) can be seen throughout the entire specimen from this patient who was on daily lisinopril at the time of specimen collection. B) Representative images of lung stained for ACE2 from an 83-year-old man in the organizing phase of diffuse alveolar damage are shown using a 1x objective (left, scale bar 3mm) with the region outlined by the red dashed box magnified at low power (middle, scale bar 200μm) and second field at high power (right, scale bar 50μm). In addition to high ACE2 expression in reactive AT2 cells (green arrowheads), strong staining can be seen in numerous vascular endothelial cells (red arrowheads) throughout the thickened septal area in this patient on valsartan 4 days prior to specimen collection. Sections were stained for ACE2 using DAB and counterstained with hematoxylin.