Infection of Primary Nasal Epithelial Cells Grown at an Air-Liquid Interface to Characterize Human Coronavirus-Host Interactions

Summary

The nasal epithelium is the primary barrier site encountered by all respiratory pathogens. Here, we outline methods to use primary nasal epithelial cells grown as air-liquid interface (ALI) cultures to characterize human coronavirus-host interactions in a physiologically relevant system.

Abstract

Three highly pathogenic human coronaviruses (HCoVs) - SARS-CoV (2002), MERS-CoV (2012), and SARS-CoV-2 (2019) - have emerged and caused significant public health crises in the past 20 years. Four additional HCoVs cause a significant portion of common cold cases each year (HCoV-NL63, -229E, -OC43, and -HKU1), highlighting the importance of studying these viruses in physiologically relevant systems. HCoVs enter the respiratory tract and establish infection in the nasal epithelium, the primary site encountered by all respiratory pathogens. We use a primary nasal epithelial culture system in which patient-derived nasal samples are grown at an air-liquid interface (ALI) to study host-pathogen interactions at this important sentinel site. These cultures recapitulate many features of the in vivo airway, including the cell types present, ciliary function, and mucus production. We describe methods to characterize viral replication, host cell tropism, virus-induced cytotoxicity, and innate immune induction in nasal ALI cultures following HCoV infection, using recent work comparing lethal and seasonal HCoVs as an example¹. An increased understanding of host-pathogen interactions in the nose has the potential to provide novel targets for antiviral therapeutics against HCoVs and other respiratory viruses that will likely emerge in the future.

Introduction

Seven human coronaviruses (HCoVs) have been identified to date and cause a range of respiratory diseases². The common or seasonal HCoVs (HCoV-NL63, -229E, -OC43, and -HKU1) are typically associated with upper respiratory tract pathology and cause an estimated 10%-30% of common cold cases annually. Though this is the typical clinical phenotype associated with the common HCoVs, these viruses can cause more significant lower respiratory tract disease in at-risk populations, including children, older adults, and immunocompromised individuals^3,4. Three pathogenic HCoVs have emerged and caused significant public health emergencies in the last 20 years, including severe acute respiratory syndrome (SARS)-CoV, Middle East respiratory syndrome (MERS)-CoV, and SARS-CoV-2. Lethal HCoVs are associated with more severe respiratory tract pathology, which is clearly illustrated by the >34% case-fatality rate associated with MERS-CoV cases (894 deaths from over 2,500 cases since its emergence in 2012)^5,6. It is important to note that the lethal HCoVs also cause a range of respiratory tract diseases, from asymptomatic infections to lethal pneumonia, as seen with the ongoing COVID-19 pandemic⁷.

HCoVs, like other respiratory pathogens, enter the respiratory tract and establish a productive infection in the nasal epithelium⁸. Spread to the lower airway is thought to be associated with aspiration from the oral/nasal cavity to the lung, where HCoVs cause more significant lower respiratory tract pathology^9,10,11. Thus, the nose serves as the initial portal for viral entry and is the primary barrier to infection with its robust mucociliary clearance machinery and unique innate immune mechanisms aimed at preventing further viral spread to the lower airway^12,13. For example, nasal epithelial cells have been reported to express higher than average basal levels of antiviral interferons and interferon-stimulated genes, indicating that nasal cells may be primed for early responses to respiratory viruses^14,15,16.

We have previously utilized patient-derived primary nasal epithelial cells grown at an air-liquid interface (ALI) to model HCoV-host interactions in the nose, where HCoV infections begin. Nasal ALI cultures are permissive to both pathogenic (SARS-CoV-2 and MERS-CoV) and common HCoVs (HCoV-NL63 and HCoV-229E) and offer various advantages over traditional airway epithelial cell lines such as A549 (a lung adenocarcinoma cell line)^16,17. After differentiation, nasal ALI cultures contain a heterogeneous cellular population and exhibit many of the functions expected of the in vivo nasal epithelium, such as mucociliary clearance machinery¹⁸. Nasal cells also offer advantages over lower airway culture systems (such as human bronchial epithelial cells, HBECs), as the acquisition of nasal epithelial cells via cytologic brushing is significantly less invasive compared with using techniques such as bronchoscopy for attaining HBECs^19,20,21.

This paper describes methods for utilizing this nasal ALI culture system to characterize HCoV-host interactions in the nasal epithelium. We have applied these methods in recently published works to compare SARS-CoV-2, MERS-CoV, HCoV-NL63, and HCoV-229E^1,16,17. Though these methods and representative results emphasize the study of HCoVs in this nasal cell model, the system is highly adaptable to other HCoVs, as well as other respiratory pathogens. Further, these methods can be applied more broadly to other ALI culture systems in order to investigate viral replication and cellular tropism, as well as cytotoxicity and innate immune induction following infection.

Protocol

The use of nasal specimens was approved by the University of Pennsylvania Institutional Review Board (protocol # 800614) and the Philadelphia VA Institutional Review Board (protocol # 00781).

1. Infection of nasal ALI cultures

NOTE: Acquisition of clinical specimens, as well as growth and differentiation of nasal ALI cultures, is outside the scope of this paper. Specific methods for culturing primary nasal epithelial cells can be found in recently published works utilizing these cultures^18,22,23. The below protocols can additionally be applied to commercially available nasal epithelial ALI cultures if desired. Protocols and volumes detailed below are applicable to 24-well plate transwell inserts (6.5 mm diameter, 0.33 cm² membrane surface area). If using ALI cultures grown on larger transwells (i.e., 12-well plates, 12 mm diameter, 1.12 cm² surface area), adjust the volumes proportionally to reflect the transwell size.

1. Day before infection:
   1. Wash ALI cultures 3x with phosphate-buffered saline (PBS) apically (add ~200 μL of warmed PBS, place in 37 °C incubator for 5 min, aspirate PBS, and repeat).
   2. Replace basal medium (500 μL).
   3. Allow cultures to equilibrate at the temperature at which infections will be conducted overnight (i.e., if infecting at 33 °C, place cultures in a 33 °C incubator after PBS washes).
      NOTE: HCoVs associated with the common cold such as HCoV-229E and HCoV-NL63 are reported to replicate more efficiently at 33 °C. Additionally, the temperature of the in vivo nasal epithelium is 33 °C (this differs from the temperature of the lung, which is 37 °C).
2. Dilute virus as needed in serum-free Dulbecco's Modified Eagle Medium (DMEM) to achieve the desired multiplicity of infection (MOI) in a total inoculum volume of 50 μL.
   NOTE: Infections have typically been conducted at MOI = 5 (high MOI); however, infections at MOI = 0.5 (low MOI) have also been used for SARS-CoV-2 and common cold-associated HCoVs, and these result in comparable peak viral titers, but with variable kinetics (either MOI is acceptable).
3. Add inoculum apically, and place cultures back into the incubator for 1 h.
4. Rock plates gently every 15 min during infection (holding plate firmly in both hands, rock forward and backward and side to side to ensure uniform adsorption of viral inoculum).
5. After 1 h of incubation, aspirate viral inoculum, and wash each infected culture 3x with PBS to ensure removal of viral inoculum (for each wash, add 200 μL of PBS, incubate for 5 min, and aspirate or remove with a pipette).
6. If desired, collect the third PBS wash to confirm adequate removal of input virus.
7. Replace the basal medium with fresh medium on infected ALIs every 72 h during infection.

2. Collection of apical surface liquid (ASL) and titration of shed virus

1. At predetermined time points following infection, add 200 μL of PBS to apical chamber of each infected transwell.
   NOTE: Relevant time points vary depending on HCoV of interest and range from 24 h to 192 h post infection (see the representative results section for viral replication data for various HCoVs).
2. Pipette PBS up and down 5x to ensure maximal collection of apically shed virus, and collect entire volume into a microcentrifuge tube (this is the ASL sample).
   NOTE: ASL includes shed viral particles in addition to mucus and other products secreted apically from ALI cultures.
3. Quantify infectious virus in ASL via standard viral plaque assay (serially dilute virus-containing samples in order to quantify the concentration of viral particles).
   NOTE: Cell types and incubation period used for plaque assay will depend on the virus used: SARS-CoV-2 (VeroE6 cells); MERS-CoV (VeroCCL81 cells); HCoV-NL63 (LLC-MK2 cells); HCoV-229E (Huh7 cells). Specifics on how to conduct viral plaque assays are beyond the scope of this manuscript but have been detailed previously in a Journal of Visualized Experiments (JoVE) publication²⁴.
4. If desired, collect the basal medium at various times post-infection to confirm the absence of basally released virus. HCoVs are typically released apically from nasal epithelial cells, but confirm this via plaque assay of undiluted basal medium.
5. Store ASL samples at −80 °C if quantification by plaque assay will not occur on the day of collection.

3. Quantification of intracellular virus

1. After collecting ASL sample, move each transwell into a clean 24-well plate preloaded with 500 μL of DMEM containing 2% fetal bovine serum (FBS) basally.
   NOTE: DMEM with 2% FBS is used to stabilize the virus during subsequent freeze-thaw cycles.
2. Wash each transwell 3x apically with PBS to ensure complete removal of apically shed virus.
3. After aspirating to remove the final PBS wash, add 100 μL of DMEM with 2% FBS to the apical compartment.
4. Move the plate containing transwells with both apical and basal media into a −80 °C freezer, and complete three consecutive freeze-thaw cycles to lyse the cells.
5. After the final freeze-thaw cycle, pool apical (100 μL) and basal (500 μL) media into a clean tube.
6. Centrifuge at 500 × g for 10 min at 4 °C to pellet any cellular debris.
7. Collect the supernatant. This is the intracellular virus sample for titration via standard plaque assay.
   NOTE: Three-fold dilution occurs during the collection process relative to ASL collection; ASL samples are collected in 200 μL of PBS, while intracellular virus sample is collected into a total volume of 600 μL.

4. Transepithelial electrical resistance (TEER) measurement

NOTE: For TEER measurement, PBS supplemented with calcium and magnesium (PBS + Ca²⁺/Mg²⁺) should be used. An epithelial volt/ohmmeter set to read in ohms is used (see the Table of Materials).

1. Clean, equilibrate, and blank the EVOM instrument according to the manufacturer's instructions; use an "empty" transwell with no nasal cells added for blanking. Record blank TEER measurement.
   NOTE: If using multiple viruses, the EVOM instrument must be cleaned stringently between conditions to avoid cross-contamination (washes with 70% ethanol followed by deionized water are sufficient).
2. Move each infected transwell into a prelabeled, clean 24-well plate with 500 μL of PBS + Ca²⁺/Mg²⁺ basally to wash residual basal medium from the transwells.
3. Add 200 μL of PBS + Ca²⁺/Mg²⁺ to the apical compartment of each transwell.
4. Add 1 mL of PBS + Ca²⁺/Mg²⁺ to the Endohm-6 measurement chamber.
5. Move each transwell into the Endohm-6 measurement chamber, and replace the lid of the chamber so that the apical electrode rests in the 200 μL of PBS in the apical compartment; the basal electrode is built into the bottom of the Endohm-6 chamber.
6. Allow the EVOM reading to stabilize, and record raw TEER measurement.
7. Collect ASL sample as described above after taking TEER measurement if titering is desired (collect in the 200 μL of PBS + Ca²⁺/Mg²⁺ that was added for TEER measurement).
   NOTE: ASL sample collection can induce micro-tears in the epithelial barrier that may confound TEER readings, so ASL must be collected after TEER measurement. Further, basal medium must not be changed immediately before TEER readings as this may also impact TEER values.
8. To convert raw TEER readings to final measurements in Ohms/cm², subtract blank TEER value, and multiply this value by the surface area of the transwell membrane using equation (1):
   TEER = [TEER reading - blank TEER value] × (surface area of transwell)    (1)
9. For HCoVs, evaluate TEER every 24 h or 48 h following infection; the kinetics of TEER changes often vary among viruses and necessitate troubleshooting at various time points.
10. When measuring TEER, always include mock-infected cultures and evaluate at each time point (negative control).
    NOTE: For mock cultures, TEER measurements should remain stable or may slightly increase from baseline after differentiation of the cultures is completed. Surface area of membrane supports will vary depending on size and manufacturer of transwells; 24-well transwells typically have a surface area of 0.33 cm².

5. Measurement of cytotoxicity during infection via lactate dehydrogenase (LDH) assay

NOTE: In this work, LDH content in ASL samples was quantified using a commercially available cytotoxicity detection kit. LDH signal in basal medium was often below the limit of detection and often less reproducible than LDH quantified in ASL samples from HCoV-infected cultures.

1. Prepare additional controls necessary for LDH assay.
   1. Background control: use PBS (use PBS containing calcium and magnesium if ASL samples were collected in this way).
   2. Positive control (ceiling value): treat ALIs apically with Triton X-100. Collect Triton control wells (typically use three ALI cultures for this at each time point).
      1. Add 200 μL of 2% Triton X-100 in PBS directly to the apical compartment of the transwell.
      2. Incubate for 10-15 min to allow cells to completely lyse.
      3. Collect the entire volume as a Triton ceiling sample.
   3. Negative control (low control/baseline LDH release): collect ASL from mock-infected ALI cultures derived from the same donor as infected cultures.
      1. Collect negative control mock ASL following ASL collection procedure above.
         NOTE: The same transwells can be used for this negative control for all time points, whereas fresh transwells will be necessary for each time point for the Triton control.
2. To allow for quantification of shed virus in addition to LDH readings from each ASL sample, load an optically clear flat-bottom black 96-well plate as follows:
   1. Dilute all samples in PBS, using 45 μL of sample and 55 μL of PBS (100 μL total volume).
   2. Treat Triton positive control and mock background control samples in the same manner.
      NOTE: This dilution allows for triplicate loading of each experimental sample (45 μL x 3) and enough residual ASL volume for titering via plaque assay.
   3. Load LDH plate in triplicate: Triton ceiling, mock background control, PBS control, and experimental samples.
   4. Prepare the reaction mixture as indicated by the manufacturer (dye solution + catalyst).
      1. Add 100 μL of reaction mixture to each well, and incubate the plate at room temperature for 20 min protected from light.
   5. After 20 min, measure absorbance at 492 nm.
   6. Calculate percentage cytotoxicity relative to the Triton ceiling value using equation (2).
      Cytotoxicity (%)      (2)

6. Preparation of nasal ALI cultures for immunofluorescence (IF) imaging

1. Move the transwell into a fresh 24-well plate, and wash 3x apically with PBS (collect the first PBS wash as ASL if titering; then, perform two additional PBS washes to remove excess shed virus that might hinder IF quality)
2. After the final PBS wash, cover the transwell apically and basally in 4% PFA.
3. Incubate for 30 min to fix in 4% PFA, and then remove and wash 3x with PBS.
4. Excise the transwell support containing the cells using a sharpened razor blade or scissors.
5. To increase the number of IF targets for each transwell, cut each membrane in half, and stain each half with different antibody combinations.
6. Permeabilize with 0.2% Triton X-100 in PBS for 10 min.
7. Block with 10% normal donkey serum and 1% BSA in PBST (PBS + 0.2% Triton X-100) for 60 min at room temperature.
   NOTE: Protect samples from light exposure from this step forward.
8. Incubate in primary antibody solution overnight at 4 °C. Dilute all antibodies 1:1,000 in blocking buffer.
   NOTE: Representative antibodies for HCoV nucleocapsid staining, as well as epithelial cell type markers and cytoskeletal stains, are listed below (see the Table of Materials for manufacturer information and catalog numbers).
   1. For HCoV antigen staining, use antibodies targeting SARS-CoV-2 nucleocapsid, MERS-CoV nucleocapsid, and HCoV-NL63 nucleocapsid.
   2. To identify epithelial cell types, use antibodies targeting goblet cell marker MUC5AC and ciliated cell marker type IV β-tubulin.
   3. For cytoskeletal markers, use antibodies directed against phalloidin (binds F-actin) and epithelial cell adhesion marker: EpCAM (CD326).
9. Incubate in secondary antibody solution for 60 min at room temperature; use secondary antibody dyes diluted 1:1,000 in blocking buffer.
10. After staining is complete, transfer the membrane onto a glass slide with a spatula, orient the transwell with the apical side toward the slide, and add mounting solution. Allow to settle for 15-30 min prior to applying clear nail polish around edges.
11. Acquire images using a confocal microscope (Z-axis step: 0.5 μm; sequential scanning)^1,16,17.
    ​NOTE: After fixation of cultures in 4% PFA and washing with PBS, fixed cultures can be stored at 4 °C for weeks to months prior to staining and preparation for imaging. After staining and mounting of membranes, samples can be stored long-term (>2 years) at 4 °C in the dark.

7. Collection of intracellular protein for western immunoblotting or RNA for RT-qPCR analysis

1. Collect ASL as described above if quantifying viral titers (protocol section 2).
2. Move the transwell to a clean 24-well plate, as scraping the membrane can lead to breaking the insert.
3. For western blot analysis, collect total protein lysate in 125 μL of RIPA buffer (50 mM Tris, pH 8, 150 mM NaCl, 0.5% deoxycholate, 0.1% SDS, 1% NP40) supplemented with protease inhibitors and phosphatase inhibitors.
4. Add 125 μL of RIPA buffer to the apical compartment, and incubate for 5-10 min.
5. Scrape the membrane using a P200 pipette tip to remove any remaining attached cells, and collect the entire volume (scrape vigorously across the entire surface of the membrane, and then pipette up and down multiple times to collect the entire sample).
6. Incubate protein lysate samples on ice for 10 min, and then centrifuge at maximum speed (20,000 × g) for 10 min at 4 °C.
7. Mix the supernatant with 4x Laemmli sample buffer with β-mercaptoethanol (reducing agent) according to manufacturer protocols.
8. Boil protein samples at 95 °C for 5 min, and then run using traditional western blotting protocols^1,16,17.
9. For collection of total RNA, use a commercially available RNA extraction kit of choice.
   NOTE: The apical compartment of 24-well transwell inserts has a maximal volume of ~200 μL; thus, we perform two sequential washes on each culture to reach total recommended volume. Details below correspond to recommended collection of RNA samples in a 350 μL total volume.
10. Add 200 μL of lysis buffer to the apical compartment of the infected transwell.
11. Let sit for 5-10 min, scrape any remaining cells from the membrane using a pipette tip, and collect the entire volume into a labeled microcentrifuge tube.
12. Add 150 μL of additional lysis buffer to the apical compartment of the transwell, and pipette up and down before collecting into same microcentrifuge tube.
13. Extract RNA according to the manufacturer's protocol.

Results

The representative figures are partially adapted from data that can be found in the manuscript Otter et al.¹. Nasal ALI cultures derived from four or six donors were infected with one of four HCoVs (SARS-CoV-2, MERS-CoV, HCoV-NL63, and HCoV-229E) according to the protocols described above, and the average apically shed viral titers for each virus are depicted in Figure 1A. While all four of these HCoVs replicate productively in nasal ALI cultures, SARS-CoV-2 and HCoV-229E replicate the most efficiently. Note that these are average viral titers, and that titers for each HCoV in individual donors were recently published¹. After washing nasal ALI cultures as described in protocol section 3, MERS-CoV titers in apical surface liquid (ASL) are compared to intracellular viral titers in Figure 1B. MERS-CoV intracellular and apically-shed viral titers are approximately the same at 48 h post infection (hpi).

Immunofluorescence imaging of infected nasal ALI cultures is a powerful tool that can be used to probe cellular tropism and other infection parameters, such as syncytia formation, cell-to-cell fusion, and damage to epithelial barrier integrity. Co-staining of infected cultures with antibodies specific to viral antigens (such as nucleocapsid or HCoV nonstructural proteins) and markers for ciliated or goblet cells (type IV β-tubulin and MUC5AC, respectively) can determine cellular tropism for an HCoV in nasal ALIs^25,26. Figure 2A shows representative images of nasal cultures infected with SARS-CoV-2, MERS-CoV, and HCoV-NL63.

SARS-CoV-2 and HCoV-NL63 infect primarily ciliated cells (evidenced by the colocalization of viral nucleocapsid staining with cilia marker type IV β-tubulin and lack of colocalization of viral antigen with goblet cell marker MUC5AC). MERS-CoV predominantly infects non-ciliated goblet cells in nasal ALI cultures, as MERS-CoV shows the opposite pattern, with colocalization of viral antigen staining with MUC5AC and minimal colocalization of viral nucleocapsid staining with type IV β-tubulin. Markers such as phalloidin, which binds to actin filaments, or EpCAM, epithelial cell adhesion molecule, can be used to visualize the epithelial cytoskeleton and detect a loss of epithelial barrier integrity²⁷. Figure 2B shows phalloidin staining in nasal ALI cultures; panel 1 shows intact cytoskeletal F-actin ultrastructure in a mock-infected culture, and panel 2 shows a loss of phalloidin integrity and suggests potential damage to epithelial barrier function in an HCoV-NL63-infected culture. Epithelial markers such as these can be applied to HCoV infections in order to characterize epithelial barrier dynamics during infection.

After infection of nasal ALI cultures with each of the four HCoVs, trans-epithelial electrical resistance (TEER) was monitored. Baseline TEER readings were recorded prior to infection, 0 hpi, and TEER was again evaluated for each transwell at 96 hpi and 192 hpi. Figure 3 depicts TEER changes for each of the HCoVs. Figure 3A,B show ΔTEER values (differences in TEER calculated for each transwell by subtracting baseline TEER at 0 hpi from TEER measured at any given point). For SARS-CoV-2, MERS-CoV, and HCoV-NL63, major changes in TEER occur late in the infection (192 hpi), and Figure 3A illustrates that SARS-CoV-2 and HCoV-NL63 infections result in negative ΔTEER values (192 hpi - 0 hpi), while MERS-CoV infection does not. Mock ΔTEER values are included for comparison and illustrate that changes in TEER following MERS-CoV infection are not significantly different than those seen during mock infection. Negative ΔTEER values indicate decreases in epithelial barrier integrity and compromised epithelial barrier function. Figure 3B shows ΔTEER values for HCoV-229E (calculated from 96 hpi and 192 hpi). HCoV-229E causes epithelial barrier dysfunction at the earlier time point (96 hpi), but recovery to mock levels occurs at the later time point (192 hpi). These data highlight how TEER kinetics can differ among viruses. Figure 3C,D show TEER traces, which depict raw TEER data for each infected transwell over time, thus allowing for visualization of TEER trends over the course of infection. If desired, treatment with the cytokine IL-13 can be included as a positive control during TEER experiments, as this impairs tight junctions, compromises epithelial barrier function, and thus, increases membrane permeability in airway epithelia; therefore, cytokine IL-13 is expected to result in decreases in TEER^28,29,30.

To complement TEER measurements in nasal cultures, cytotoxicity can be quantified via quantification of lactate dehydrogenase (LDH) released apically during infection. Figure 4 depicts average cytotoxicity data at 96 hpi and 192 hpi from cultures derived from 10 donors infected with each of the HCoVs assayed. SARS-CoV-2, HCoV-NL63, and HCoV-229E cause significant cytotoxicity in nasal cultures, while MERS-CoV does not.

Total protein or RNA collected from infected nasal cultures can be used to examine various HCoV-host interactions. We treated nasal cultures with type 2 cytokine IL-13 in order to induce goblet cell hyperplasia and model the tissue landscape of an asthmatic airway. Figure 5A shows qPCR data quantifying mRNA abundance of two major HCoV receptors following IL-13 treatment. DPP4 is the cellular receptor for MERS-CoV, and ACE2 is the cellular receptor for both SARS-CoV-2 and HCoV-NL63. IL-13 treatment results in dramatically increased DPP4 expression but no significant changes in ACE2 expression. Figure 5B shows western blot data to evaluate protein abundance following IL-13 treatment in uninfected and SARS-CoV-2-infected cultures. IL-13 treatment results in significantly increased MUC5AC (a goblet cell marker) and decreased type IV β-tubulin (a ciliated cell marker), reflecting the expected goblet cell hyperplasia caused by this cytokine treatment. Protein-level analysis shows that IL-13 treatment increases DPP4 expression but decreases ACE2 expression. Western blotting for SARS-CoV-2 nucleocapsid protein reveals that IL-13 treatment results in a slight decrease in viral antigen compared to sham-treated cultures. Similar analyses can be performed for any mRNA or protein of interest following manipulation of nasal ALI cultures and/or infection with HCoVs.

Figure 1: HCoV replication in nasal ALI cultures. Nasal cultures derived from six or four donors were infected in triplicate with SARS-CoV-2 (6), MERS-CoV (6), HCoV-NL63 (6), or HCoV-229E (4) at MOI = 5. Apical surface liquid was collected at 0 hpi, 48 hpi, 96 hpi, and 144 hpi, and infectious virus was quantified by plaque assay. (A) Averaged viral titers from all the donors infected with each HCoV are depicted. (B) Nasal cultures derived from a single donor were infected in triplicate at MOI = 5, and ASL was collected at 48 hpi. After ASL collection, the transwells were washed 3x with PBS and then lysed via freeze-thaw in order to quantify intracellular virus. Infectious virus in the apical shed compartment versus in the intracellular compartment was quantified by plaque assay. Data are displayed as mean ± SD. Abbreviations: ALI = air-liquid interface; MOI = multiplicity of infection; ASL = apical surface liquid; hpi = hours post infection; PBS = phosphate-buffered saline. This figure was constructed using data published in Otter et al.¹. Please click here to view a larger version of this figure.

Figure 2: Immunofluorescence imaging of nasal ALI cultures to characterize cellular tropism and epithelial integrity. (A) Nasal ALI cultures were infected with SARS-CoV-2, MERS-CoV, or HCoV-NL63 at MOI = 5 and fixed in 4% paraformaldehyde at 96 hpi. Cultures were prepared for immunofluorescence imaging, as described above in protocol section 6, using primary antibodies against each virus nucleocapsid protein (shown in red), ciliated cell marker type IV β-tubulin (green), or goblet cell marker MUC5AC (green). Of note, an antibody against MERS-CoV nonstructural protein 8 (nsp8) was used in place of an antibody against nucleocapsid protein due to species incompatibility with the MUC5AC antibody. Colocalization between viral antigen and each of the epithelial cell markers is visualized as orange/yellow color in merged images for each virus. (B) Nasal ALI cultures were stained using phalloidin (which stains actin cytoskeletal filaments) in order to visualize cytoskeletal integrity, as shown in pink. Hoescht staining is shown in blue. Panel 2B.1 shows crisp, intact phalloidin staining, indicating intact cytoskeletal architecture, while panel 2B.2 shows a loss of epithelial integrity and blurring of the phalloidin stain. Scale bars = (A) 50 µm, (B) 10 µm. Abbreviations: ALI = air-liquid interface; MOI = multiplicity of infection; hpi = hours post infection. This figure was constructed using data published in Otter et al. Please click here to view a larger version of this figure.

Figure 3: Measurement of trans-epithelial electrical resistance in ALI cultures during infection. (A) Nasal ALI cultures derived from 10 donors were either mock-infected or infected at MOI = 5 with SARS-CoV-2, MERS-CoV, or HCoV-NL63. ΔTEER was calculated for each transwell as TEER at 192 hpi minus TEER at 0 hpi (baseline TEER). Each bar illustrates the average ΔTEER value for each virus among triplicate cultures from each donor. (B) Nasal ALI cultures derived from 8 donors were mock-infected or infected with HCoV-229E at MOI = 5. ΔTEER from baseline TEER was calculated as in (A) using TEER at either 96 hpi or 192 hpi. (C,D) TEER trace data are depicted for mock-infected or HCoV-infected cultures derived from each of four donors. Each line represents TEER data from a single transwell over time (triplicate transwells from each donor were assayed). Donor numbers are color-coded and shown in the key to the right of each graph. Data are displayed as mean ± SD in panel A and panel B. Abbreviations: ALI = air-liquid interface; MOI = multiplicity of infection; hpi = hours post infection; TEER = trans-epithelial electrical resistance. This figure was constructed using data published in Otter et al.¹. Please click here to view a larger version of this figure.

Figure 4: Cytotoxicity quantification during HCoV infection of nasal ALI cultures. Nasal cultures derived from 10 donors infected with each HCoV in triplicate at MOI = 5 underwent LDH quantification in ASL samples, as described above. Averaged cytotoxicity among all donors tested is shown for each HCoV at 96 hpi and 192 hpi. Data are displayed as mean ± SD. Abbreviations: ALI = air-liquid interface; MOI = multiplicity of infection; hpi = hours post infection; TEER = trans-epithelial electrical resistance; ASL = apical surface liquid; LDH = lactate dehydrogenase. This figure was constructed using data published in Otter et al.¹. Please click here to view a larger version of this figure.

Figure 5: mRNA/protein analysis following infection of nasal cultures. Nasal cultures treated with type 2 cytokine IL-13 or sham-treated were used to evaluate changes in expression of HCoV receptors, as well as in ciliated and goblet cell markers. (A) mRNA expression of DPP4 and ACE2 was quantified via RT-qPCR. Data for cultures from three donors treated with IL-13 are shown as fold changes over sham-treated cultures derived from the same donors. Data are displayed as mean ± SD, with each point representing average fold change induction for a single donor. (B) Total protein was harvested from nasal cultures that were either sham- or IL-13 treated and either mock- or SARS-CoV-2-infected. Proteins were separated via SDS-PAGE and immunoblotted with antibodies against epithelial cell markers (type IV β-tubulin, MUC5AC), HCoV receptors (ACE2, DPP4), SARS-CoV-2 nucleocapsid protein, and GAPDH. Abbreviations: RT-qPCR = reverse-transcription-quantitative polymerase chain reaction; SDS-PAGE = sodium dodecylsulfate polyacrylamide gel electrophoresis. This figure was constructed using data published in Otter et al.¹. Please click here to view a larger version of this figure.

Discussion

The methods detailed here describe a primary epithelial culture system in which patient-derived nasal epithelial cells are grown at an air-liquid interface and applied to the study of HCoV-host interactions. Once differentiated, these nasal ALI cultures recapitulate many features of the in vivo nasal epithelium, including a heterogeneous cellular population with ciliated, goblet, and basal cells represented, as well as intact mucociliary function with robustly beating cilia and mucus secretion. This heterogeneous cell population is a key benefit of this culture system over traditional respiratory epithelial cell lines, as it allows for the characterization of host cell tropism during viral infection (as shown in Figure 2). Traditional airway epithelial cell lines also lack functional mucociliary clearance mechanisms, which play key roles in primary defense against respiratory viruses alongside innate immune pathways such as interferon production.

Critical steps within the protocol include the initial infection procedure; for example, the cultures must be equilibrated at the desired temperature prior to infection, and a consistent methodology should be used to start each infection. These steps are pivotal to standardize the method and ensure reproducibility in downstream findings.

Perhaps the most important portion of these methods is the quantification of apically released virus, as understanding the unique replication cycle of each HCoV helps to determine the time points that might be best for further analysis. The immunofluorescence techniques described above are also critical, as the ability to visualize viral infections in an in vitro nasal epithelium with a representative heterogeneous cellular population allows for accurate assessment of viral tropism and provides the ability to further query virus-host interactions in a physiologic setting.

The nasal epithelium is the primary barrier site encountered by all respiratory viruses, and, thus, it is likely that HCoV infections begin with primary infection of the nasal epithelium, with the potential to spread to the lower airway via an oral-lung aspiration axis. This likely plays a role in the development of severe pneumonia and respiratory disease during pathogenic HCoV infection (MERS-CoV, SARS-CoV-2), while the seasonal HCoVs tend to cause disease only in the upper airway. It is critical to understand HCoV-host interactions in this important immune sentinel site, and this nasal ALI system allows for characterization of viral replication, host cell tropism, as well as cytotoxicity and innate immune induction during HCoV infection.

This nasal ALI culture system is also highly adaptable and can be used to mirror many clinical disease states. Nasal specimens acquired from patients with genetic pulmonary diseases such as cystic fibrosis (caused by a chloride channel defect) or primary ciliary dyskinesias (in which ciliary beating mechanisms are dysfunctional) can be used to grow nasal ALI cultures representative of these pathologies to understand how these patients may be differentially impacted by HCoV infection^31,32. Additionally, various cytokine treatments can be used during the differentiation of nasal ALI cultures to recreate features of other disease states. For example, IL-13 treatment during differentiation induces goblet cell hyperplasia and can be used as a surrogate for studying asthmatic or allergic airways^33,34. Thus, this nasal ALI system is a powerful tool for probing HCoV-host as well as other virus-host interactions both in healthy and diseased nasal airways.

Though this system provides many advantages, some limitations also arise with the use of nasal ALI cultures. Though nasal cells require far less invasive techniques for acquisition than bronchial or lung cells, growing and differentiating ALI cultures requires significantly more time and resources than the use of traditional immortalized cell lines. Additionally, donor-to-donor variability in terms of permissiveness to infection as well as host responses to HCoV infection can make it more difficult to reproduce and interpret data (this is why experiments are often conducted with cultures derived from 5-10 donors, in order to mitigate these issues).

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Alexa Fluor secondary antibodies (488, 594, 647)	Invitrogen	Various
BSA (bovine serum albumin)	Sigma-Aldrich	A7906
cOmplete mini EDTA-free protease inhibitor	Roche	11836170001
Cytotoxicity detection kit	Roche	11644793001
DMEM (Dulbecco's Modified Eagle Media)	Gibco	11965-084
DPBS (Dulbecco's Phosphate Buffered Saline)	Gibco	14190136
DPBS + calcium + magnesium	Gibco	14040-117
Endohm-6G measurement chamber	World Precision Instruments	ENDOHM-6G
Epithelial cell adhesion marker (EpCAM; CD326)	eBiosciences	14-9326-82
Epithelial Volt/Ohm (TEER) Meter (EVOM)	World Precision Instruments	300523
FBS (Fetal Bovine Serum)	HyClone	SH30071.03
FV10-ASW software for imaging	Olympus	Version 4.02
HCoV-NL63 (Human coronavirus, NL63)	BEI Resources	NR-470
HCoV-NL63 nucleocapsid antibody	Sino Biological	40641-V07E
Hoescht stain	Thermo Fisher	H3570
Laemmli sample buffer (4x)	BIO-RAD	1610747
LLC-MK2 cells	ATCC	CCL-7	To titrate HCoV-NL63
MERS-CoV (Human coronavirus, Middle East Respiratory Syndrome Coronavirus (MERS-CoV), EMC/2012)	BEI Resources	NR-44260
MERS-CoV nucleocapsid antibody	Sino Biological	40068-MM10
MUC5AC antibody	Sigma-Aldrich	AMAB91539
Olympus Fluoview confocal microscope	Olympus	FV1000
Phalloidin-iFluor 647 stain	Abcam	ab176759
PhosStop easy pack (phosphatase inhibitors)	Roche	PHOSS-RO
Plate reader	Perkin Elmer	HH34000000	Any plate reader or ELISA reader is sufficient; must be able to read absorbance at 492 nm
RIPA buffer (50 mM Tris pH 8; 150 mM NaCl; 0.5% deoxycholate; 0.1% SDS; 1% NP40)	Thermo Fisher	89990	Can prep in-house or purchase
RNeasy Plus Kit	Qiagen	74134
SARS-CoV-2 (SARS-Related Coronavirus 2, Isolate USA-WA1/2020)	BEI Resources	NR-52281
SARS-CoV-2 nucleocapsid antibody	Genetex	GTX135357
Triton-X 100	Fisher Scientific	BP151100
Type IV β- tubulin antibody	Abcam	ab11315
VeroCCL81 cells	ATCC	CCL-81	To titrate MERS-CoV
VeroE6 cells	ATCC	CRL-1586	To titrate SARS-CoV-2