To determine the effects of human rhinovirus infection on MYADM expression in epithelial cells, we first infected H1HeLa cells with rhinovirus-1B (RV-1B) and quantified MYADM+ cells by flow cytometry assay ( Figures 1A, B ). At the 0.5 multiplicity of infection (MOI), cells expressing MYADM increased twofold compared with those of non-infected controls. The increased expression was dose-dependent, with 4.7- and 6.8-fold increases at MOI of 1 and 5, respectively. MYADM upregulation was also replication-dependent as UV-irradiated RV-1B (UV-RV1B) did not increase MYADM-expressing cells.

MYADM gene expression in H1HeLa cells was further quantified by qRT-PCR ( Figure 1C ). At MOI 0.1, MYADM mRNA levels increased 2.2-fold compared with the control. MYADM mRNA levels were not increased with UV-RV1B challenge. Moreover, MYADM expression in differentiated human primary epithelial cells (AECs) at air–liquid interface (ALI) cultures also increased upon RV-1B infection ( Figures 1D, E ). Interestingly, another common respiratory virus, RSV, did not induce MYADM expression ( Supplement Figure 1 ).

We next examined MYADM expression in lung sections from a murine model of allergic airways disease. As shown in Figure 2A , the MYADM protein level (shown in green) in ovalbumin-treated lung sections from mice was increased ~25-fold compared with the saline controls. MYADM appeared to localize to the membrane of cytoplasmic vesicles, which were apically polarized toward the airway’s lumen ( Figure 2B ). In mouse airways, MYADM was mainly colocalized with cells staining positive for CC16 (club cells) and MUC5AC (goblet cells), whereas some staining was also detected in ciliated cells.

In human AECs, RV-1B infection led to increases in total ciliated and basal cell numbers ( Figure 2C ), as determined by flow cytometry, whereas the percentage of MYADM-expressing cells for each respective cell type remained consistent ( Supplement Figure 2 ).

To examine the role of MYADM in inflammation, we performed siRNA knockdown of MYADM in H1HeLa cells. As shown in Figure 3A , knockdown of MYADM reduced MYADM transcription by half (0.49-fold). The transcription levels of inflammatory cytokines, IL-6, IL-8, and TNFα, were also reduced by half (0.46, 0.50, and 0.54) in cells in which MYADM was knocked down, compared with controls. Transcription of calcium-activated chloride channel regulator 1 (CLCA1) and signal transducer and activator of transcription (STAT)-1 and -3 were also reduced to 0.62, 0.86, and 0.78, respectively. Additionally, cells expressing MYADM were quantified by flow cytometry ( Figure 3B ). The percentage of cells carrying MYADM was 26% for mock control (scramble siRNA) and 20% for the siRNA-treated cells. While MYADM knockdown at the protein levels was only modest, it still resulted in a significant decrease of IL-8 secretion into culture media compared with controls ( Figure 3C ).

We next sought to determine whether the induction of MYADM expression is associated with RV-1B replication by correlation analysis of gene transcripts ( Supplementary Figure 3 ). Increased MYADM expression significantly collerated with increased RV-1B replication in H1HeLa cells ( Supplementary Figure 3A ) and in AECs ( Supplementary Figure 3B ). For H1HeLa cells, several RV-1B-induced genes had significant positive associations with MYADM expression, including interferon-lambda (IFN-l), IL-8, STAT-3, and CLCA1. In differentiated AECs, the detectable RV-1B-induced genes that had significant positive association with MYADM expression included the IFN-related genes IFN-l, interferon-stimulated gene 15 (ISG15), ubiquitin-specific peptidase 18 (USP18), IFN-Λ receptor 1 (IFNΛ-R1), and IFN-β1; the cell receptors Toll-like receptor 3 (TLR3), low-density lipoprotein receptor (LDLR), ICAM-1, and melanoma differentiation-associated gene 5 (MDA5); and other inflammatory factors STAT1 and C–X–C motif chemokine ligand 11 (CXCL11). The correlation (R) of each gene transcript of interest to MYADM and the corresponding P-values are presented in Table 1 .

To further examine the role of MYADM during RV-1B infection, H1HeLa cells were transfected with MYADM siRNA followed by RV-1B infection. Quantitative RT-PCR revealed that both MYADM knockdown and control cells had comparable RV-1B replication ( Figure 4A ). The infection led to increased MYADM mRNA by 2.3-fold, whereas RV-1B infection was not able to induce further MYADM upregulation in MYADM-silenced cells. UV-RV1B-treated cells did not have a detectable change in MYADM expression, which were similar to controls. STAT3 was increased by RV-1B infection and was significantly decreased when MYADM was knocked down. IL-8 and IFN_Λ were increased significantly by RV1B infection; MYADM knockdown led to a slight, non-significant reduction in the induction. When assessing IL-8 protein levels, as shown in Figure 4B , whereas IL-8 protein concentrations in culture media were not altered by RV-1B infection, cells in which MYADM was knocked down during infection had a significant reduction in IL-8 protein secretion.

We next assessed whether TLR3 stimulation increased MYADM expression as we observed with live RV1B infection. For this, H1HeLa cells were incubated with poly I:C for 48 h and RT-PCR was performed. As shown in Figure 5 , poly I:C increased the transcription of MYADM, IL-6, IL-8, and TNFα, STAT3, TLR3, MDA5, and ICAM-1. Knockdown of MYADM led to a reduction in Poly I:C-induced expression of MYADM, IL-8, STAT-3, and TLR-3 but did not alter the expression of Poly I:C-induced IL-6, TNF-a, MDA5, or ICAM-1.

Human AECs were isolated from discarded bronchial segments from donor lungs that were collected for transplant according to the protocol approved by University of Arizona’s Institutional Review Board (under PI, Dr. Polverino). Upon arrival in the lab, the bronchus was rinsed with RPMI1640 and submerged in 1 mg/ml protease (Sigma-Aldrich, St. Louis, MO) in Ham’s F12 for 1 h at 37°C. After neutralizing with HyClone FBS (Cytiva, Marlborough, MA), AECs were gently scrapped out and incubated in Gibco Versene (Thermo Fisher Scientific, Waltham, MA). AECs were washed and seeded on a 100 mm collagen and fibronectin-coated tissue culture dish in PneumaCult EX Plus supplemented with 10 µM ROCK inhibitor (APExBIO, Houston, TX) until 90% confluency. Cells were collected using trypsin and preserved in PneumaCult EX Plus (STEMCELL Technologies, Vancouver, BC) containing 30% FBS and 10% DMSO at 10⁶ cells/ml and stored at −80°C.

The H1HeLa cell (CRL-1958) was purchased from ATCC and maintained in DMEM/F12 supplemented with 10% FBS and 1% penicillin–streptomycin (Thermo Fisher Scientific). Human AECs were cultured (26) in PneumaCult EX Plus until 80% confluency. Then, 2 × 10⁴ cells were seeded onto Transwells in a 24-well plate (Thermo Fisher Scientific) that was precoated with collagen and fibronectin. Once cells achieved 100% confluency, they were differentiated for 21 days with PneumaCult-ALI (STEMCELL Technologies, Vancouver, BC) media on a basolateral chamber only.

RV-1B (VR-1645) was obtained from ATCC propagated in a H1HeLa cell monolayer. First, after 1 h of viral attachment at MOI of 10, the infection was conducted in T25 flasks with DMEM/F12 10% NEAA at 33°C for 24 h. Then, the cell lysate supernatant was inoculated onto cells in T175 flasks. After 48 h, cells were lysed with 10% NP-40 (Sigma-Aldrich) and the supernatant was treated with RNase A (Thermo Fisher Scientific), lauryl sarcosine (Sigma-Aldrich), and β-mercaptoethanol. Next, the supernatant was overlaid on 30% sucrose and centrifuged at 40,000 RPM for 2 h. Finally, the particles were resuspended in 0.01%BSA Dulbecco’s phosphate-buffered saline (DPBS), filter-sterilized, and stored at −80°C until use. The purified RV1B were quantified by the plaque assay as described earlier (27) with some modification. First, the H1HeLa cell monolayers in the six-well plate were washed with DPBS with calcium and then incubated with various dilutions of RV-1B at room temperature for 1 h. The washed cells were subsequently overlaid with agarose and incubated at 33°C for 80 h. Plaques were fixed and stained with crystal violet. GFP-RSV was a generous gift of Drs. Mark Peeples and Peter Collins of the Ohio State University and propagated and used as previously detailed (28).

H1HeLa cells were cultured in a 24-well plate. Once cells were at 70% confluency, they were treated with 50 nM MYADM siRNA (Integrated DNA Technologies, Coralville, Iowa, Table 2 ), or scramble siRNA control in 1 µg/ml Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer’s instructions. Cells were harvested for RNA extraction or flow cytometry at 72 h after transfection.

For H1HeLa cells, 10⁵ cells/well were seeded in a 24-well plate for 24 h prior to infection. For MYADM knockdown, cells were treated with siRNA as described above for 24 h prior to infection. Cells were washed with DPBS and incubated with RV-1B 2 × 10⁴ PFU/well for 1 h. After 1 h, cells were washed and replaced with fresh media and incubated at 35°C for 48 h. The optimal temperature for RV-1B infection in human airway epithelial cells is 35⁰ C, which was used for these studies. For AECs, at day 21 of differentiation, cells were washed with DPBS and pretreated with 75 μl of media on an apical chamber and 500 µl on a basal chamber. On day 22, the cells were washed with DPBS and incubated with RV-1B 2 × 10⁴ PFU/well for 4 h at 35°C. Then, cells were washed, replaced with fresh media, and incubated at 35°C for 48 h.

H1HeLa cells were seeded onto a 24-well plate and cultured overnight. Poly I:C (Tocris, Bristol, UK) was diluted in the media to a final concentration of 10 μg/ml and incubated with H1HeLa cells in a 24-well plate for 48 h at 37°C. Optimal poly I:C stimulation is at 37⁰ C, which was used for these studies. Cells were harvested for qRT-PCR.

Cells were lysed with TRIzol reagent (Invitrogen, San Diego, CA), and extracted RNAs were quantified by a NanoDrop One spectrophotometer (Thermo Fisher Scientific). cDNAs were synthesized using iScript cDNA Synthesis Kit (Bio-Rad Laboratories, Hercules, CA). Gene amplification was performed on QuantStudio 3 (Thermo Fisher Scientific) with TaqMan Universal Master Mix II (Applied Biosystems) or PowerUp SYBR Green Master Mix (Applied Biosystems) with probes and primers (Integrated DNA Technologies), as shown in Table 2 . The target genes of interest are antiviral genes (interferon-lambda, IFN-Λ, interferon-beta 1, IFN-β1, interferon-stimulated gene 15, ISG15, IFN-Λ receptor 1, IFNΛ-R1, ubiquitin specific peptidase 18, USP18), inflammatory genes (interleukin-8, IL-8, signal transducer and activator of transcription 1, STAT1, STAT3, calcium-activated chloride channel regulator 1, CLCA1, C–X–C motif chemokine ligand 11, CXCL11), and receptors for cell signaling and rhinovirus entry (Toll-like receptor 3, TLR3, melanoma differentiation associated gene 5, MDA5, low-density lipoprotein receptor, LDLR, and ICAM-1).

Cells were harvested by trypsinization for 5 min and neutralized with FBS. After washing, cells were incubated in 1:100 anti MYADM antibody (Bioss Antibodies, Woburn, MA) in 2% FBS PBS at 4°C for 30 min. Then, the cells were washed and incubated with 1:100 anti-rabbit IgG Alexa Fluor 647 (Cell Signaling Technology, Danvers, MA). Next, cells were stained with Annexin V-FITC and PI using apoptosis detection kit (BD Pharmingen, San Diego, CA) and subsequently analyzed on Attune NxT Flow Cytometer (Thermo Fisher Scientific).

At the indicated time point, culture media were collected and kept at −80°C until further analysis. One day before the assay, a capture antibody provided by the human IL-8 ELISA kit (Invitrogen, Carlsbad CA) was coated onto a 96-well plate. The media were diluted 1:20 in an assay buffer and loaded into duplicated wells along with a detection antibody. Samples and standard IL-8 dilutions were incubated as per the manufacturer’s instruction. The absorbance at 450 nm was extrapolated into a concentration in picograms/ml using a standard IL-8 curve.

The protocol was approved from the Institutional Animal Care and Use Committee (IACUC for Ledford, 15-575) at the University of Arizona (29). Briefly, 8-week-old C57BL/6 mice (Jackson Laboratories, Bar Harbor, ME) were sensitized using 50 μg ovalbumin (OVA) in alum or sterile saline intraperitoneally (i.p.) on days 0 and 7 and challenged intranasally (i.n.) on days 14, 15, and 16. On day 18, left lung lobes were collected and fixed with 10% formalin. The formalin-embedded lung sections were prepared for immunohistochemistry.

After deparaffinized and antigen recovery, mouse lung sections were permeabilized, blocked, and incubated with 1:50 dilution of anti-MYADM (Bioss Antibodies) at 4°C overnight. Then, bound antibodies were captured with anti-rabbit IgG Alexa Fluor 488 (Cell Signaling Technology). Next, they were further incubated with 1:200 dilution of either anti-Ac-αTubulin (Cell Signaling Technology), anti-CYP2F2 (Santa Cruz Biotechnology, Santa Cruz, CA), anti-KRT5 (Invitrogen), anti-CC16 (Santa Cruz Biotechnology), or anti-MUC5AC (Invitrogen) for 2 h at 37°C. Bound antibodies were detected with anti-mouse IgG-Alexa Fluor 555 (Cell Signaling Technology). Images were taken using Leica DMI6000 with 40× magnification.

All analyses were done in GraphPad Prism software and were considered statistically significant when P-value < 0.05. Gene expression, IL-8 secretion, and flow cytometry data were presented as mean± SEM from three independent experiments, and they were analyzed by one-way ANOVA and Tukey’s post-hoc multiple comparisons. The intensity of immunohistochemistry staining was analyzed by unpaired t-test. Differential cell count of human AECs was analyzed using two-way ANOVA. Gene transcriptions of MYADM knockdown in H1HeLa cells were analyzed by unpaired t-test. The correlation of transcription profiles was examined using Pearson correlation coefficient analysis.