Intersubject Variation in ACE2 Protein Expression in Human Airway Epithelia and Its Relationship to Severe Acute Respiratory Syndrome Coronavirus 2

Abstract

Background

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2 ) initiates entry into airway epithelia by binding its receptor, angiotensin-converting enzyme 2 (ACE2).

Methods

To explore whether interindividual variation in ACE2 abundance contributes to variability in coronavirus disease 2019 (COVID-19) outcomes, we measured ACE2 protein abundance in primary airway epithelial cultures derived from 58 human donor lungs.

Results

We found no evidence for sex- or age-dependent differences in ACE2 protein expression. Furthermore, we found that variations in ACE2 abundance had minimal effects on viral replication and induction of the interferon response in airway epithelia infected with SARS-CoV-2.

Conclusions

Our results highlight the relative importance of additional host factors, beyond viral receptor expression, in determining COVID-19 lung disease outcomes.

In primary cultures of human airway epithelia, individual variation in ACE2 protein levels does not correlate with sex or age and has little influence on SARS-CoV-2 replication subsequent to infection.

The coronavirus disease 2019 (COVID-19) pandemic has caused a global health crisis and disrupted economies across the world. Infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes a spectrum of respiratory and other symptoms, with severity that varies widely between individuals. The factors determining variability in outcomes are incompletely understood. Male sex and older age, as well as other comorbidities that can compromise immune function, are important risk factors. One candidate host factor is the viral receptor, angiotensin-converting enzyme 2 (ACE2) [1], which is expressed in ciliated and secretory cells of the conducting airways and type II pneumocytes of the alveoli [2–4]. ACE2 is also expressed widely in other organs, particularly the heart, gastrointestinal tract, eye, and olfactory neuroepithelium [3–5], likely contributing to the broad range of clinical manifestations.

Although ACE2 is a determinant of viral entry, it is unclear whether individual variability in ACE2 abundance in the airways impacts SARS-CoV-2 infection and the risk of developing severe lung disease. There is speculation that sex differences in ACE2 expression could explain the increased risk of severe COVID-19 and mortality in males, although evidence for sex-dependent differences in ACE2 in human lung tissue is inconsistent [6–10]. It has also been hypothesized that age-dependent increases in ACE2 abundance contribute to morbidity and mortality in older individuals. In addition, ACE2 abundance could influence disease outcomes through effects on viral entry and/or the innate immune response. In this study, we investigate the relationship between ACE2 variation and other risk factors for severe COVID-19, by quantifying ACE2 protein in airway epithelia. We studied well differentiated primary cultures of human airway epithelia (HAE) grown at an air-liquid interface (ALI), a culture model that closely resembles the surface epithelium of the conducting airways [11]. Much of the data regarding individual variability in ACE2 expression is derived from ribonucleic acid (RNA)-sequencing datasets; because messenger RNA (mRNA) abundance does not always correlate with protein expression, we quantified ACE2 protein directly by enzyme-linked immunosorbent assay (ELISA).

METHODS

Cell Culture

Primary airway epithelia (passage 0) were isolated from human donor bronchi and grown at an air-liquid interface on collagen-coated, semipermeable membranes with a 0.4-μm pore size (Costar Transwell; surface area 0.33 cm²; Corning) as reported previously [12]. Human bronchial epithelial cultures were grown in Dulbecco’s modified Eagle’s medium (DMEM)/F12 with 2% Ultroser G media at 37°C with 5% CO₂. All cell preparations were well differentiated (>3 weeks old; resistance >1000 Ohm × cm²). Using this culture method, we find that each Transwell filter typically comprises approximately 300 000 cells. The study was approved by the Institutional Review Board at the University of Iowa. Calu-3 cells were first cultured in minimal essential medium supplemented with 20% fetal bovine serum (FBS), 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 2 mM l-glutamine, 1% penicillin and streptomycin, and 0.15% NaHCO₃ at 37°C with 5% CO₂. When cells reached 80% confluence in submerged culture conditions, the Calu-3 cells were transferred to collagen-coated Transwell membranes as described [12] to generate ALI cultures. Vero E6 cells were maintained in DMEM with 10% FBS and 1% penicillin and streptomycin at 37°C with 5% CO₂.

Angiotensin-Converting Enzyme 2 Enzyme-Linked Immunosorbent Assay

Human airway epithelia and Calu-3 cells were rinsed apically with sterile phosphate-buffered saline (PBS) before preparing cell lysates. To make lysates, cells were resuspended in lysis buffer containing 1% NP-40 (100 µL per Transwell culture), then spun in a tabletop centrifuge for 5 minutes at 1000 rpm to pellet cell debris. Supernatants were assayed for ACE2 protein using the human ACE2 ELISA kit (Abcam catalog number ab235649), according to the manufacturer’s directions.

Severe Acute Respiratory Syndrome Coronavirus 2 Infection in Human Airway Epithelia

All infection experiments with SARS-CoV-2 were performed at the Biosafety Level 3 (BSL3) facility at the University of Iowa. To infect HAE cultures, HAE were first rinsed apically with 200 µL sterile PBS to remove shed ACE2 from the cell surface. Next, SARS-CoV-2 (2019-nCoV/USA-WA1/2020) was diluted in serum-free medium and applied to the apical surface of ALI cultures at a multiplicity of infection (MOI) of 0.1, in a total volume of 100 µL. Cultures were incubated at 37°C, 5% CO₂ for 1 hour, then unbound virus was removed and cultures were rinsed apically 3 times with 200 µL PBS containing Ca²⁺ and Mg²⁺ to remove residual virus. Infected cultures were maintained at 37°C, 5% CO₂ for the duration of the experiments. At the indicated time points, apical washes were performed in a volume of 60 µL sterile PBS/Transwell filter, to collect released virions from the apical compartment. Apical washes were stored at −80°C before titration. Cells were collected in TRIzol Reagent (Life Technologies) and stored at −80°C until RNA isolation.

Severe Acute Respiratory Syndrome Coronavirus 2 Plaque Assay

Severe acute respiratory syndrome coronavirus 2 was titrated by plaque assay as described previously [13]. In brief, thawed apical wash samples were serially diluted in serum-free DMEM and applied to the surface of Vero E6 cells. Cells were incubated with virus at 37°C, 5% CO₂, with gentle rocking. After 1 hour, virus was removed and replaced by a 1.2% agarose overlay. Cells were incubated for 3 days at 37°C, 5% CO₂. To visualize plaques, cells were fixed with 25% formaldehyde and stained with 0.1% crystal violet. Titers are expressed as plaque-forming units per milliliter. Plaque assays were performed at the BSL3 facility at the University of Iowa.

Quantitative Reverse-Transcription Polymerase Chain Reaction to Measure Viral and Host Genes

Infected cells were suspended in TRIzol Reagent and total RNA was extracted using the Direct-zol RNA MiniPrep kit (Zymo Research, Irvine, CA). A DNase treatment step was included. Total RNA (200 ng) was used as a template for first-strand complementary deoxyribonucleic acid (cDNA) synthesis with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, catalog number 4368814). The resulting cDNA was subjected to amplification of selected genes by real-time quantitative polymerase chain reaction (PCR) using Power SYBR Green PCR Master Mix (Applied Biosystems, catalog number 4367695). Samples were run in duplicate and averaged. These averages were then used to calculate the relative abundance of transcripts normalized to hypoxanthine-guanine phosphoribosyltransferase and presented as 2^−ΔCT. The primers used are listed in Supplementary Table 1.

Statistical Analysis

Statistical tests were performed using GraphPad Prism 7. Unpaired, 2-tailed Student’s t tests and one-way analysis of variance were used to analyze differences in mean values between groups, as indicated. Simple linear regression was used to test for correlations between ACE2 protein abundance and other variables. All results are expressed as mean ± SE, and P < .05 were considered significant.

RESULTS

We used ELISA to measure ACE2 protein levels in cell lysates from primary human airway epithelia derived from lung donors. Before preparing cell lysates, HAE cultures were washed extensively to remove shed ACE2, ensuring that measurements represent membrane-bound ACE2. We detected ACE2 in cell lysates from all 58 donor specimens, with abundance ranging from ~10⁴ pg/mL to almost 10⁶ pg/mL (Figure 1A). In contrast, the mean ACE2 protein abundance in cell lysates from polarized Calu-3 cells, a human lung cell line frequently used for in vitro SARS-CoV-2 studies, was ~10-fold lower. We noted no significant difference in ACE2 protein abundance in epithelia derived from male and female donors (Figure 1B). We also failed to observe a relationship between ACE2 protein levels and age of the HAE donor (Figure 1C). This latter finding contrasts with earlier reports that ACE2 mRNA increases with advancing age in human lung and nasal epithelium [7, 14], although this observation has not always been replicated [8, 15]. Our data suggest that, at the protein level, individual differences in ACE2 abundance in airway epithelia do not correlate significantly with sex and/or age.

Figure 1.

Angiotensin-converting enzyme 2 (ACE2) protein expression in human airway epithelia (HAE). (A) The ACE2 protein abundance in human airway epithelial donors (n = 58) or air-liquid interface cultures of Calu-3 cells (n = 4). For each donor, the result represents the ACE2 protein concentration in a lysate prepared from a single Transwell filter (see Methods for additional details). (B) Epithelial cell ACE2 abundance analyzed by sex (male n = 42, female n = 16). Unpaired 2-tailed t test used to test for significant differences. (C) Linear regression analysis of the relationship between ACE2 abundance and HAE donor age.

To further explore how variations in ACE2 abundance influence SARS-CoV-2 infection and replication, we infected HAE with SARS-CoV-2 (MOI 0.1) and measured viral progeny release (Figure 2A). The HAE donors were selected to represent the range of ACE2 protein expression observed in our larger dataset. There was a positive correlation between ACE2 abundance and SARS-CoV-2 virion production from individual donors at 1 day postinfection (R² = 0.4611, P = .0152). Reasoning that HAE with higher ACE2 protein levels might support greater SARS-CoV-2 replication, we monitored plaque titers and viral RNA accumulation over a timecourse (Figure 2B and C). The HAE donors were sorted into 2 groups according to ACE2 protein abundance: an “Above-Mean” group, donors with ACE2 greater than the average in this experiment, and a “Below-Mean” group, donors with ACE2 abundance below the average. We observed similar levels of viral release in both groups (Figure 2B), and no significant differences in viral RNA production were observed between the Above- and Below-Mean ACE2 groups at any time point postinfection (Figure 2C).

Figure 2.

Variability in angiotensin-converting enzyme 2 (ACE2) abundance has limited impact on severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) replication and interferon response in human airway epithelia (HAE). (A) The HAE were infected with SARS-CoV-2 (multiplicity of infection [MOI] 0.1), and progeny virions in the apical compartment at 1 day postinfection were quantified by plaque assay (n = 12 donors). The ACE2 abundance and virus titers were correlated by linear regression. (B) The HAE were infected with SARS-CoV-2 (MOI 0.1) and viral progeny release quantified at the indicated time points postinfection. Above-Mean ACE2 group, n = 4; Below-Mean ACE2 group, n = 4; limit of detection. (C) The SARS-CoV-2 ribonucleic acid (RNA) in infected HAE quantified by real-time quantitative polymerase chain reaction (qRT-PCR). In (B) and (C), statistical significance was assessed using unpaired 2-tailed t tests, with the Holm-Sidak correction for multiple comparisons. (D) The HAE were infected with SARS-CoV-2 (MOI 0.1), and messenger RNA (mRNA) expression was assayed by qRT-PCR. The mRNA abundance of the indicated genes was normalized to hypoxanthine phosphoribosyl-transferase (HPRT). Above-Mean ACE2 group, n = 8; Below-Mean ACE2 group, n = 8. One-way analysis of variance, followed by Sidak’s multiple comparisons test, was used to test data for significant differences relative to the corresponding uninfected control cells (Day 0) within each group. #, P_adj < .05; ##, P_adj < .01; ####, P_adj < .0001. To test for differences between the 2 groups at each timepoint, we performed unpaired 2-tailed t tests, with the Holm-Sidak correction for multiple comparisons. *, P_adj < .05; **, P_adj < .01. In all graphs, data represent mean ± standard error. HRPT, hypoxanthine-guanine phosphoribosyltransferase; IFN, interferon, PFU, plaque-forming units.

We next used quantitative reverse-transcription PCR to probe for ACE2-dependent differences in host antiviral responses (Figure 2D). Transcripts for interferon-α, interferon-β, and interferon-λ were significantly increased at 1 day postinfection and generally remained elevated through 7 days postinfection. Significant differences in interferon expression were evident only at the latest timepoint (7 days postinfection), when donors in the Above-Mean ACE2 group exhibited elevated interferon relative to the Below-Mean group. Although this finding suggests that higher levels of ACE2 protein may be associated with a more robust interferon response, the implications of this result are difficult to interpret in light of the fact that viral burden did not differ in the Above- and Below-Mean ACE2 groups at this timepoint (Figure 2B and C). We also observed upregulation of several interferon-stimulated and other host defense genes over the course of infection with SARS-CoV-2, including interferon-stimulated gene 15 (ISG15), interferon-induced transmembrane protein 3 (IFITM3), 2’-5’-oligoadenylate synthetase 1 (OAS1), the proinflammatory cytokines interleukin (IL)-1α, IL-1β, IL-6, and the chemokine IP-10 (Supplementary Figure 1). Induction of these genes was largely similar in the Above- and Below-Mean ACE2 groups; however, we did note higher expression of ISG15, IFITM3, and OAS1 in the Below-Mean ACE2 donors approximately 3–5 days postinfection and significantly greater expression of IL-10 in the Above-Mean ACE2 donors at 7 days postinfection. More detailed studies will be required to more fully understand the relationship between ACE2 levels and the kinetics of the innate immune response in HAE.

DISCUSSION

Overall, these findings do not support the hypothesis that age- and/or sex-dependent differences in ACE2 abundance influence SARS-CoV-2 infection of airway epithelia. One advantage of this study was that we surveyed ACE2 variation in primary epithelia from a significant number of lung donors (n = 58). We acknowledge that further studies are needed to assess whether ACE2 protein expression is influenced by other comorbidities including obesity, chronic respiratory illness, smoke exposure, etc. An additional limitation is that our ACE2 measurements were made in an ex vivo model, and this study does not address whether age- or sex-dependent differences in ACE2 abundance are present in other tissues or cell types relevant to COVID-19 pathophysiology. We found that variation in ACE2 protein abundance had limited effects on viral load and interferon expression in infected HAE, suggesting that, although ACE2 is necessary for SARS-CoV-2 to bind and enter epithelia, the subsequent course of the infection is relatively ACE2 independent.

CONCLUSIONS

These findings emphasize the complexity of host responses to SARS-CoV-2 and point to the importance of additional host factors as determinants of COVID-19 severity and disease outcomes.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. We thank Brajesh Singh and Katarina Kulhankova for careful review of the manuscript.

Financial support. This work was funded by the National Institutes of Health (P01 AI060699). We also acknowledge the support of the Cell Morphology Core and Pathology Core, partially supported by the Center for Gene Therapy for Cystic Fibrosis (National Institutes of Health P30 DK054759) and the Cystic Fibrosis Foundation. P. B. M. is funded by the Roy J. Carver Charitable Trust.

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.