According to Austrian law and the ‘World Medical Association Declaration of Helsinki – Ethical Principles for Medical Research Involving Human Subjects’, the use of anonymised, non-identifiable cells obtained from a commercial provider do not require specific ethical approval. Primary human cells were purchased from Epithelix and were originally obtained with informed consent as part of processes that had ethical review and approval.

Submerged cell culture

HeLa Ohio- (supplied by ECACC, catalogue number 14J003), MRC-5- (supplied by ATCC, catalogue number CCL-171), MDCK- (supplied by ECACC, catalogue number CCL-34) and THP-1 NF-κB eGFP reporter- (provided by Prof. Peter Steinberger, Medical University of Vienna) cells were cultured in tissue culture-treated flasks at 37 °C, 5% CO₂, 85% humidity using cell-specific maintenance medium (see Supplementary Table S2). Adherent cells were passaged or used for experiments upon reaching 70–90% confluency and THP-1 cells were passaged 1–2 times per week.

Air-liquid interface cultures

HNEC (Epithelix Sàrl, EP51AB, donor no. AB0857 and AB060801), HBEC (Epithelix Sàrl, EP51AB, donor no. 02AB0859), HBEC3-KT (Evercyte, CkHT-004–0230, provided by LifeTeq, Austria) and Calu-3 cells (ATCC, HTB-55) were cultured at 37 °C, 5% CO₂, 85% humidity using cell-specific and differentiation-specific culture medium.

For submerged expansion culture in tissue culture-treated flasks, HNEC and HBEC were cultured in PneumaCult-Ex Plus medium, Calu-3 cells in Calu-3 medium and HBEC3-KT cells were grown in gelatine (0.1%) coated flasks using Keratinocyte-SFM medium. The cells were maintained until 70–90% confluency was reached, before detaching and seeding into 24-well plate transwell inserts. HNEC and HBEC were detached using an animal component-free dissociation kit, and HBEC3-KT and Calu-3 were detached using Accutase. Cells were seeded at a density of 6.0 × 10⁴ – 8.0 × 10⁴ cells per transwell insert. HNEC, HBEC and Calu-3 cells were maintained in the same medium as previously, while HBEC3-KT cells were transferred into in PneumaCult-Ex Plus medium.

At confluence, the cultures were airlifted by removing medium from the apical compartment and supplying cell-specific ALI medium to the basal compartment. HNEC, HBEC and HBEC3-KT cells were cultured in PneumaCult-ALI medium and Calu-3 cells in Calu-3 medium. The medium was refreshed every 2–4 days. Mucus was removed from HNEC, HBEC, and HBEC3-KT cultures through weekly apical washes with DPBS. Apical secretions were removed from Calu-3 ALI cultures every 3–4 days by aspiration.

Viral infection & 2-DGA treatment

All viral infections and drug treatments were performed at 34 °C, 5% CO₂, and 85% humidity.

To infect HeLa Ohio cells with RV-B14, the cells were plated at a density of 1.5 × 10⁵ cells/well in a 24-well plate in 500 µL HeLa Ohio experiment medium. The following day, the cells were washed with PBS and the medium was replaced before infecting with 1.5 × 10³ TCID₅₀/well of RV-B14. Simultaneously, treatment was initiated with 10 mM 2-DG, 2-FDG or 2-FDM diluted in culture medium, or with placebo (medium only). The cells were incubated for 7 h before washing with medium and collecting cell lysates.

ALI cultures were kept in their respective maintenance medium, except for HBEC, which were transferred into HBEC infection medium. ALI cultures were infected with 1 × 10⁴ TCID₅₀/well of RV-A1B, 1–5 × 10⁴ TCID₅₀/well of RV-A16, 0.75 × 10⁴ TCID₅₀/well of HCoV-229E or 1.5 × 10⁴ TCID₅₀/well of IAV-H1N1. For RV-A16, 1 × 10⁴ TCID₅₀/well was used in conventional solution treatment experiments, while 5 × 10⁴ TCID₅₀/well was used for nebulized treatment experiments. Infections were performed by adding 40 µL virus diluted in PBS, or PBS alone for mock infected controls, to the apical compartment of each ALI culture and incubating for 2 h. Afterwards, the cultures were washed six times with DPBS to remove non-internalised virus. Apical washes with DPBS were performed to collect samples for TCID₅₀ and viability assays, and the cultures were treated with cell lysis buffer to harvest cell lysates.

HNEC ALI cultures were treated with a 2-DG formulation intended for nasal application, consisting of 3.5% 2-DG (213 mM) in citrate buffer (50 mM). Treatments were performed by adding 40 µL of 2-DG or placebo (NaCl) to the apical compartment of each ALI culture. The treatment was either removed after 1 h or left on the cells until sample harvest.

For solution application experiments, HBEC ALI cultures were treated with 3.5% 2-DG (213 mM), 2-FDG (192 mM) or 2-FDM (192 mM) in PBS, or placebo (PBS). The HBEC were treated 1 h before infection and 3 h after infection by adding 40 µL 3.5% 2-DG or placebo to the apical compartment. Subsequent treatments were performed by adding an appropriate volume of 1 M 2-DGA (8.5 µL 2-DG or 7.7 µL 2-FDG/2-FDM) to achieve the same dose per well. This was done to avoid removing any released virus while also preventing diluting the treatment.

Solutions of 2-DG 2-FDG or 2-FDM in PBS were used for both aerosol and solution treatment of Calu-3 and HBEC ALI cultures in experiments involving nebulization. The placebo used was PBS alone. The Vitrocell Cloud a 12 aerosol exposure system paired with Aerogen Pro nebulizers was used for aerosol drug delivery. Aerosol exposure was performed at 34 °C, and one treatment consisted of four exposures (1 min nebulization, 10 min sedimentation). For each exposure, 200 µL of solution was nebulized in the 9-well chamber and 76.4 µL in the 3-well chamber. Basal medium cover adapter rings were utilized to ensure that the basal medium was not exposed to nebulized drug. The solution treatment was performed by adding 40 µL 2-DGA or placebo to the apical compartment of each transwell. The concentration of the 2-DGA solution was adjusted to match the expected deposited dose per transwell following nebulization, according to the experimentally determined deposition factor (as previously described⁴¹).

THP-1 activation and NF-κB translocation assay

The THP-1 NF-κB eGFP reporter human monocytic cell line has been genetically modified to express eGFP upon translocation of NF-κB to the nucleus following its activation⁴⁰. The THP-1 cells were seeded into 96-well flat-bottom plates in THP-1 medium at a density of 1.0 × 10⁵ cells/well. They were pre-treated with 30 mM 2-DG, 2-FDG or 2-FDM or placebo (THP-1 medium) for 1 h, before being activated with 30 µM R848 or treated with vehicle alone (unstimulated control). The 2-DGA treatment was continued after activation for 24 h.

Thereafter, the supernatant was collected for cytokine analysis and the cells were stained with the Zombie Violet Fixable Viability Kit. The cells were washed in PBS and stained in 50 µL viability dye diluted 1:500 in PBS, for 10 min at 4 °C. The THP-1 cells were washed and resuspended in PBS before being acquired on a BD LSRFortessa flow cytometer to assess NF-κB translocation. The population of interest was identified by excluding debris, doublets and dead cells, and the data was presented as fold change in eGFP median fluorescence intensity compared to the unstimulated control.

Cytokine analysis

The LEGENDplex COVID-19 Cytokine Storm Mix & Match kit was used to measure secreted cytokines and chemokines (IL-1b, IL-6, CCL2, CCL5, CXCL8, CXCL10 and TNFa). The analytes were quantified cell culture medium, in technical duplicates, according to the manufacturer’s instructions but adapted to use ¼ volumes.

RNA isolation, cDNA synthesis and qPCR

Total RNA was extracted from cell lysates using the innuPREP RNA Mini kit 2.0 or QIAGEN RNAeasy Mini Kit, following the manufacturers’ protocol. The RNA concentration of each sample was normalised to 200–400 ng by diluting in nuclease-free water and cDNA was synthesized using the First Strand cDNA Synthesis Kit, using oligo(dT)₁₈ primers, according to the manufacturer’s instructions (see Supplementary Table S3 for cycling conditions). qPCR was performed using the PowerTrack SYBR Green Master Mix according to the manufacturer’s instructions (see Supplementary Table S3 for cycling conditions and Supplementary Table S4 for primer sequences). The intracellular viral RNA and CXCL10 was measured by normalizing the gene expression to host housekeeping gene, hypoxanthine phosphoribosyltransferase 1 (HPRT; for HeLa Ohio) or β−2 microglobulin (B2M; for ALI cultures) using the 2^−DDCTmethod⁵⁵. The data was expressed as fold changes or percentages, compared to the average of a control group.

Median tissue culture infectious dose (TCID₅₀) assay

For TCID₅₀ assays, cells were seeded into 96-well plates in cell-specific infection medium and incubated overnight at 37 °C, 5% CO₂, 85% humidity. HeLa Ohio cells were detached using Accutase and seeded at a density of 1.1–1.5 × 10⁴ cells per well in 100 µL of HeLa Ohio infection medium. MRC-5 cells were detached with 0.5% trypsin-EDTA and seeded at a density of 1.5–2.0 × 10⁴ cells per well in 100 µL MRC-5 infection medium. MDCK cells were detached using 1x trypLE and seeded at a density of 1.5–2.0 × 10⁴ cells per well in 100 µL MDCK infection medium.

Viral supernatants were diluted in full logarithmic steps from 10⁻¹ to 10⁻¹¹ or half logarithmic steps from 10⁻¹ to 10⁻⁶ in specific infection medium (as described above). Titrations for RV, HCoV-229E and IAV-H1N1 were conducted in HeLa Ohio, MRC-5 and MDCK cells, respectively. The cells were infected by adding 50 µL of virus-containing supernatants and incubated at 34 °C, 5% CO₂. After 5–6 days the cells were examined under a microscope for cytopathic effect (CPE). For HCoV-229E and IAV-H1N1, CPE or no CPE was documented through microscopic observation. For RV-A16 and RV-A1B, supernatants were aspirated, the wells were washed with PBS, stained with crystal violet solution (0.05% crystal violet powder in 20% methanol/ddH₂O solution) for 30–60 min and washed three times with ddH₂O. After drying, 25% acetic acid was added to the wells, the plate was shaken for 2 min and the absorbance at 450 nm was recorded on a plate reader. CPE was defined as an absorbance decrease ≥25% in comparison to the average absorbance of uninfected control wells. The data was analysed according to Reed-Muench’s method⁵⁶.

Viability assay

The LDH-Glo Cytotoxicity Assay was used according to the manufacturer’s instructions to assess cell viability. Apical wash samples were diluted 3-20X in LDH storage buffer and stored at −20 °C until analysis. Maximum LDH release controls were prepared by treating one ALI culture with 40 µL of 10% Triton X-100 for 2 h at 37 °C. All samples were normalised to the maximum LDH release control sample, which was considered as 100% cytotoxicity.

Trans-epithelial electrical resistance (TEER) measurement

To assess barrier integrity, the TEER was measured. The culture medium was aspirated and 200 µL and 500 µL PBS was added to apical and basal compartment, respectively. The cells were incubated at room-temperature for 10 min to equilibrate the temperature. Resistance (R) was measured using an electrode connected to a Millicell-ERS Volt Ohm device. The electrode was held perpendicular and inserted into the transwell without touching the cell layer on the transwell membrane and values were recorded upon stabilisation. The resistance reading from a transwell without cells was used as a blank. The measured resistance values were used to calculate the TEER for each sample (Eq. 1):

2-FDG deposition factor determination

To determine the deposition factor of 2-FDG, 800 µl of 3.5% 2-FDG in PBS was nebulized over four exposures using the VitroCell Cloud a exposure chamber. The actual concentration of 2-FDG in the nebulized solution was quantified and used for subsequent calculations to ensure accuracy. The deposited mass was measured using a quartz crystal microbalance, as previously described⁵⁷, and corrected by subtracting the mass stemming from PBS. The deposition factor was calculated as described below (Eq. 2):

$$\:Deposition\:factor\:=\:\frac{D\text{m}}{(V\text{neb}\:\times\:\:C\text{sol}\text{)/}\text{A}\text{ch}}$$

(2)

D_m: Measured deposited 2-FDG (µg/cm²).

V_neb: Volume of nebulized 2-FDG (mL).

C_sol: Concentration of nebulized 2-FDG (µg/mL).

A_ch: Surface area of bottom of exposure chamber (cm²).

In vitro aerodynamic performance of 2-FDG solution

The aerodynamic performance tests were done using NGI equipped with a flow controller and a vacuum pump (Copley Scientific). All tests were done according to European Pharmacopeia and in triplicates, by using three independently prepared solutions. Briefly, 400 µL of 3.5% 2-FDG solution was nebulized using a vibrating mesh nebulizer Aerogen Ultra Solo (Aerogen) at the flow of 15 L/min. The solution was nebulized until no visible liquid could be observed in a nebulizer. The drug content in each part of the NGI was determined by the quantification method described in detail in paragraph: Quantification of 2-FDG and its metabolite 2-FDG-6P. The MMAD corresponds to the deposited particles diameter out of which 50% (w/w) have a lower- and 50% higher diameter. The FPF was considered as cumulative 2-FDG amount detected on stages 3 to 8, corrected for the cut off diameter on Stage 3 (5.39 μm for the flow of 15 L/min) (Eq. 3)⁴².

Cellular 2-FDG uptake assay

Calu-3 cells cultured at ALI were treated with nebulized 3.5% 2-FDG in PBS and the deposited mass was measured using a quartz crystal microbalance. The deposited 2-FDG dose per transwell was calculated by correcting the measured ng/cm²value for the mass stemming from PBS and multiplying by the transwell surface area (0.33 cm²). The calculated value was used to define the solution application dose. Accordingly, Calu-3 cells cultured at ALI were treated with 40 µL of a solution of 2-FDG in PBS, which was added to the apical compartment. The cells were incubated at 34 °C, 5% CO₂, 85% humidity, until metabolite extraction.

At the indicated time points, the solution treatment was removed, and the cells were washed twice with 4 °C PBS. Subsequently, 400 µL extraction buffer (50% methanol, 30% acetonitrile, 20% ddH₂O) was added to the apical compartment and the cells were incubated for 1 h at 4 °C. Finally, the extraction buffer was collected, flash frozen in liquid nitrogen and stored at 80 °C until analysis.

To determine the proportion of deposited 2-FDG that was taken up by the cells, a 1–1 conversion rate of 2-FDG to 2-FDG-6P was assumed. The measured amount of 2-FDG-6P in each sample was normalised to the 2-FDG treatment dose as described below (Eqs. 4 and 5).

C_2−FDG−6P: Measured 2-FDG-6P conc. (ng/µL)

V_2−FDG−6P: Extraction buffer volume (µL).

m_2−FDG: Deposited 2-FDG (ng/cm²).

A_insert: Area of transwell membrane (cm²)_.

C_2−FDG−6P: Measured 2-FDG-6P conc (ng/µL).

V_2−FDG−6P: Extraction buffer volume (µL).

C_2−FDG: 2-FDG treatment conc. (ng/µL)

V_2−FDG: 2-FDG treatment volume (µL).

Mucus penetration assay

Porcine intestine mucus (type III) was used as a substitute for airway mucus, to enable standardization across experiments. PBS was added to each well of the Vitrocell Cloud basal module and empty transwells for aerosol and solution application were placed in separate chambers. Empty transwells without mucus were included in the aerosol group as a control and 20 µL of porcine mucus in PBS (20 mg/mL) was added to the apical compartment of all other transwells. Next, 20 µL of 2-FDG in PBS was added on top of the mucus for the solution application samples. The 2-FDG solution concentration was adjusted to deliver 45 µg per transwell, which is equivalent to the deposited dose following nebulization. Immediately after the solution application, nebulization of 3.5% 2-FDG was initiated for the aerosol application samples. Upon treatment completion, 1000 µL of solution was collected from each basal module well and the samples were stored at −80 °C until analysis.

To determine the proportion of 2-FDG that had passed through the mucus layer into the basal compartment, the measured concentration of 2-FDG in each sample was normalised to the highest possible concentration, as described below (Eqs. 6 and 7):

Cm: 2-FDG conc. in mucus wells.

Ce: Mean 2-FDG conc. in empty wells.

$$\:Solution\:muc.\:pen.\:=\:\frac{Cb\times\:\:Vb}{Ca\:\times\:\:Va}$$

(7)

Cb: Basal 2-FDG conc.

Ca: Apical 2-FDG treatment conc.

Va: Apical 2-FDG treatment conc.

Quantification of 2-FDG and its metabolite 2-FDG-6P

Samples were analysed by hydrophilic interaction liquid chromatography (HILIC), coupled to tandem mass spectrometry (LC-MS/MS) to measure 2-FDG and 2-FDG-6P.

For 2-FDG measurements, all experimental and calibration curve samples were pretreated by adding 990 µL of extraction solvent (80% acetonitrile, 20% ddH₂O)) to 10 µL sample volume. The samples were centrifuged and the supernatants further diluted (1:10 − 1:100) to assure a concentration within the linear range of the instrument.

For both 2-FDG and 2-FDG-6P measurements, 1 µL of sample was injected onto a polymeric iHILIC-(P) Classic HPLC column and operated at a flow rate of 100 µL/min (see Supplementary Table S5 for gradients used for separation). An Ultimate 3000 HPLC system was directly coupled via electrospray ionization to a TSQ Quantiva mass spectrometer. Selected reaction monitoring was used for detection and quantification (see Supplementary Table S5 for details). For each transition and metabolite, authentic standards were used to determine optimal collision energies and for the validation of experimental retention times. All samples were measured in technical replicates in a randomized fashion. Injection of pure solvent was used to determine eventual carry over. Data interpretation was performed using TraceFinder (Thermo Fisher Scientific) and absolute values were derived by external calibration.

Software and statistical analysis

Graphs were generated and statistical analysis was performed using GraphPad Prism (GraphPad Software), flow cytometry data analysis was performed using FlowJo (BD), and experimental setup graphics were created in Inkscape.

Differences were considered statistically different when p < 0.05. The Shapiro-Wilk’s normality test was used to determine normality. Parametric statistical tests (e.g. Welch’s t-test) were used for normally distributed data. If the data did not have a normal distribution, or too few data points were available for normality testing, a nonparametric test (e.g. Kruskal-Wallis test) was used. Welch’s correction was applied for t-tests, as it corrects for unequal standard deviations but does not introduce error when standard deviations are equal. An appropriate multiple comparisons correction was used for all statistical tests that compare more than two groups. The applied statistical test, sample size (n) and independent experimental replicates (N) is indicated in each figure legend.