Intranasal mask for protecting the respiratory tract against viral aerosols

Abstract

The spread of many infectious diseases relies on aerosol transmission to the respiratory tract. Here we design an intranasal mask comprising a positively-charged thermosensitive hydrogel and cell-derived micro-sized vesicles with a specific viral receptor. We show that the positively charged hydrogel intercepts negatively charged viral aerosols, while the viral receptor on vesicles mediates the entrapment of viruses for inactivation. We demonstrate that when displaying matched viral receptors, the intranasal masks protect the nasal cavity and lung of mice from either severe acute respiratory syndrome coronavirus 2 or influenza A virus. With computerized tomography images of human nasal cavity, we further conduct computational fluid dynamics simulation and three-dimensional printing of an anatomically accurate human nasal cavity, which is connected to human lung organoids to generate a human respiratory tract model. Both simulative and experimental results support the suitability of intranasal masks in humans, as the likelihood of viral respiratory infections induced by different variant strains is dramatically reduced.

Fig. 1. Schematic and experimental design for MV@GEL as an intranasal mask to intercept viral aerosols and entrap virus.

a The intranasal mask (MV@GEL) was composed of engineered cell-derived microsized vesicles (MV) with viral receptor and thermosensitive hydrogel with positive charges. It could be sprayed into the nasal cavity at room temperature and quickly transformed from the liquid state to the gel state at body temperature. The viral receptor of vesicles could help vesicles entrap the virus, and the thermosensitive hydrogel could prolong the retention time of vesicles in the nasal cavity. Once the negative viral aerosols were inhaled, the intranasal mask could perform the protective effect in the following steps: Step 1, the positively charged hydrogel could intercept the negatively charged viral aerosols presenting in airflow; Step 2, those viral aerosols could fuse with MV@GEL and release viruses into MV@GEL; Step 3, the embedded MV in MV@GEL could entrap those released viruses. b The protective effect of the intranasal mask was investigated from the following three aspects. 1. Mouse model: MV@GEL conferred strong protection against viral aerosol infection in the mouse nose and downstream lung; 2. Digital human nasal model: based on computerized tomography (CT) images of the human nasal cavity, computational fluid dynamics (CFD) simulation supported that viral aerosols could be intercepted in the human nasal cavity under MV@GEL protection; 3. Human respiratory tract model: connecting a realistic human nasal apparatus with human lung organoids and providing respiratory airflow by the pump, the human respiratory tract model was constructed and utilized to demonstrate the good performance of MV@GEL in protecting the lung organoids from viral aerosols.

Fig. 2. Preparation of MV with high angiotensin-converting enzyme II (ACE2) expression and their entrapment effect on SARS-CoV-2 pseudovirus (SPV).

a Schematic for MV with high ACE2 expression (aMV) preparation and its mechanism of entrapping the SPV. b Confocal laser scanning microscopy (CLSM) images showing the ACE2 protein expression on wild-type 293 T cells transfected with control empty plasmid (WT-293T) or ACE2-plasmid (ACE2-293T). White: ACE2 antibody labeled ACE2 protein. Blue: 4′,6-diamidino-2-phenylindole (DAPI) labeled cell nuclei. c CLSM images of ACE2-293T cells after cytochalasin B treatment. Red: carboxyfluorescein succinimidyl ester labeled cytoplasm (CSFE). d CLSM and transmission electron microscope (TEM) images of aMV. Red: 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindodicarbocyanine,4-chlorobenzenesulfonate salt (DiD) labeled cell membrane. e CLSM images of ACE2 protein expression on vesicles that were obtained from WT-293T cells (nMV) and aMV. Red: DiD labeled cell membrane. White: ACE2 antibody labeled ACE2 protein. f Flow cytometry analysis of ACE2 protein expression on nMV and aMV. The ACE2 protein were labeled with the ACE2 antibody. g Stimulated emission depletion (STED) microscopy image of SPV (wild-type) internalized within aMV. Green: fluorescein isothiocyanate (FITC) labeled SPV. Red: DiD labeled cell membrane. h TEM images of ultrathin section of aMV after 24 h of challenge with SPV (wild-type). The green arrow indicated SPV. In the enlarged image (right), SPV was marked with false color (green). i CLSM images of ACE2-293T cells that were treated with the indicated agents and then challenged with SPV (wild-type), accompanied with the corresponding quantification of relative infection rate. Green: GFP was expressed in infected cells. Blue: DAPI labeled cell nuclei. j Relative infection rates of ACE2-293T cells that were treated with PBS or aMV and then challenged with the indicated SPVs with S protein mutation. Quantitative data in f, i, and j represent as means ± S.E.M., n = 3 biologically independent experiments. Statistical significance was calculated using two-tailed unpaired t-test (f, j) and one-way ANOVA with multiple comparison tests (i). All P-values are indicated. Source data are provided in the Source data file.

Fig. 3. Characterization and protective effect of chitosan/β-sodium glycerophosphate hydrogel loaded with aMV.

a Schematic for the viral aerosol interception mechanism of chitosan/β-sodium glycerophosphate hydrogel (GEL) loaded with aMV (aMV@GEL). b CLSM image of aMV@GEL showing aMVs were embedded in the network of GEL. Red: DiD labeled aMV; Cyan: FITC labeled GEL. c Scanning electron microscope (SEM) image of aMV@GEL. aMVs were marked with false color (red). d Photographic images of aMV@GEL showing aMV@GEL was in the liquid state at room temperature (RT, 25 °C) and transformed to the gel state at 37 °C. e Photographic image of spray characteristic of aMV@GEL and corresponding spray area formed on the paper 10 cm away from the sprayer. f Schematic design for evaluating the interception effect (left), fluorescence images of the tube precoated with PBS, aMV, GEL, or aMV@GEL after nebulizing cyanine 5 N-hydroxysuccinimide ester (Cy5) stained SPV (wild-type) aerosols into the left entrance end (middle), and corresponding fluorescence images of cut view site (indicated by the red arrow, right). g, h Fluorescence quantitative analysis of different sites in tubes of f (g) and relative passing rate of SPV aerosols through these tubes (h). i Schematic and corresponding CLSM images illustrating the detailed process of SPV (wild-type) aerosol interception and virus entrapment by aMV@GEL. (i) The drops of viral aerosols consisting of water and SPV. (ii) The top view of viral aerosols contacting the aMV@GEL (0 min). (iii) The top view of viral aerosols fusing with aMV@GEL (10 min). (iv) The entrapment of the released SPV by aMV inside GEL (60 min), which image was captured at a distance of 15 μm from the surface of the GEL. Purple: fluorescein sodium-stained water in viral aerosols; Green: cyanine 3 N-hydroxysuccinimide ester (Cy3) stained virus in viral aerosols; Red: Cy5 stained aMV; Cyan: FITC-stained gel. Quantitative data in h represents as mean ± S.E.M., n = 3 biologically independent experiments. Statistical significance was calculated using one-way ANOVA with multiple comparison tests. All significant P-values are indicated. Source data are provided in the Source data file.

Fig. 4. Distribution of aMV@GEL after intranasal administration and its in vivo protective effect against SPV aerosols.

a Schematic diagram for evaluating the distribution of free aMV and aMV@GEL after intranasal administration. b Representative in vivo fluorescence images of mice at the indicated time points after intranasal administration of free aMV or aMV@GEL. aMV in both groups were stained with Cy5. c Relative fluorescence intensity (F.I.) of free aMV or aMV@GEL in mouse nasal cavity in b (left) and corresponding half-life (right). d Representative frozen section images and enlarged views of the mouse nasal cavity at 8 h after intranasal administration of free aMV or aMV@GEL. Red: Cy5 stained aMV; Blue: DAPI stained cell nuclei. e Representative ex vivo fluorescence images of various tissues (nasal cavity, brain, lung, heart, liver, spleen, and kidney) at 8 h after intranasal administration of free aMV or aMV@GEL. aMV in both groups were stained with Cy5. f Schematic diagram for evaluating the in vivo protective effect of aMV@GEL against SPV aerosols. g Representative frozen section images and enlarged views of mouse nasal cavity (left) and relative level of GFP mRNA in the nasal cavity by using q-PCR detection (right) after 3 days of challenge with SPV (wild-type) aerosols. GFP was produced by SPV-infected cells; Blue: DAPI stained cell nuclei. h Representative frozen section images and enlarged views of mouse lung (left) and relative level of GFP mRNA in the lung tissue by using q-PCR detection (right) after 3 days of challenge with SPV (wild-type) aerosols. GFP was produced by SPV-infected cells; Blue: DAPI stained cell nuclei. i Relative level of GFP mRNA that was produced by SPV-infected cell in nasal cavity and lung using q-PCR detection after 3 days of challenge with SPV (S protein B.1.1.529 mutation (Omicron)) aerosols. Quantitative data in c, g, h, and i represent as means ± S.E.M., n = 3 biologically independent mice. Statistical significance was calculated using two-tailed unpaired t-test (c, i) and one-way ANOVA with multiple comparison tests (g, h). All significant P-values are indicated. Source data are provided in the Source data file.

Fig. 5. Preparation of GEL loaded with α-2,6 sialic acid-overexpressing MV (sMV@GEL) and its protective effect against H1N1-CA07 aerosols.

a Schematic for the preparation and H1N1 virus entrapment mechanism of MV with α-2,6 sialic acid-overexpression (sMV). b CLSM images showing SA expression on different cells. White: sialic acid (SA); Blue: nuclei. c CLSM image of CFSE stained ST-293T cells (red) after 30 min of cytochalasin B treatment. d CLSM image showing SA expression on sMV. White: SA; Red: sMV. e TEM images of ultrathin section of sMV after 24 h of challenge H1N1-CA07. The red arrow indicated H1N1-CA07. f Representative photographs of Madin-Darby canine kidney (MDCK) cells after 3 days of challenge with H1N1-CA07 under different dose of nMV or sMV protection. g Relative level of H1N1-CA07 mRNA in MDCK cells of different groups (PBS, 50 μg nMV, and 50 μg sMV). h CLSM images of sMV@GEL showing sMV were embedded in the network of gel. Cyan: GEL in sMV@GEL; Red: sMV in sMV@GEL. i Schematic diagram for evaluating the in vivo protective effect of sMV@GEL against H1N1-CA07 aerosols in mice. j Relative level of H1N1-CA07 mRNA in the nasal cavity (left) and lung (right) in the PBS group and sMV@GEL group after 7 days of challenge with low dose of H1N1-CA07 aerosols. k, l Representative IHC (upper) and H&E (lower) sections of the nasal cavity (k) and lung (l) in PBS group (left) and sMV@GEL group (right) of j. The red arrows: N protein of H1N1-CA07 (brown). m Schematic of H1N1-CA07 transmission model. n Relative level of H1N1-CA07 mRNA in the IR nasal cavity (left) and lung (right) after 3 days exposed to donor. o Number of infected TR in each cage after 3 days exposed to PBS IR (left) or sMV@GEL IR (right). Data in g, j and n represent as means ± S.E.M., n = 3 biologically independent experiments (g), n = 6 biologically independent mice (j, n). Statistical significance was tested with two-tailed unpaired t-test (j, n) and one-way ANOVA with multiple comparison tests (g). All significant P-values are indicated. Source data are provided in the Source data file.

Fig. 6. Prediction of viral aerosol interception effect of MV@GEL in human digital nasal cavity by using computational fluid dynamics-discrete particle simulation (CFD-DPS).

a Representative CT images of the human nasal cavity and its corresponding reconstructed 3D digital model. b Schematic diagram of the charge interaction between MV@GEL and aerosols in the nasal cavity during simulation calculation. Each tiny part on the nasal wall could generate a tiny Coulomb force to the specific viral aerosol. Meanwhile, the drag force of fluid, buoyancy, and gravity were considered, and the vector sum was the resultant force it received. c Distribution of inhaled viral aerosols (blue dot) at different time points (0.3 s, 0.6 s, 0.9 s, 1.2 s, and 1.5 s) of unprotection situation (upper) or MV@GEL situation (lower) after the beginning of inhalation (left), and the corresponding percentage of viral aerosols that retained in the nasal cavity or flowed into trachea at 1.5 s (right). d Flow field state of inhalation airflow (left) and the distribution of inhaled viral aerosols (blue dot) in different cross sections under unprotection situation or MV@GEL situation after 1.5 s inhalation (right). e Flow field state of exhalation airflow (left), the distribution of viral aerosols at 2.1 s (1.5 s inhalation followed by 0.6 s exhalation) under unprotection situation or MV@GEL situation (middle), accompanied with corresponding viral aerosols distribution in different cross sections (right). The inhaled viral aerosols were blue dots, and the exhalant viral aerosols were red dots.

Fig. 7. Confirmation of the protective effect of sMV@GEL against H1N1 viral aerosols by using the human respiratory tract model.

a Schematic diagram of the experimental design. We used the CT data from the volunteer and 3D printing technology to fabricate a realistic human nasal apparatus. Meanwhile, we cultured human lung organoids from the paracancerous tissue of a lung cancer patient. To imitate the human respiratory tract (HRT), we constructed an integrated HRT model by placing the lung organoids into a container with the following vents: one vent was connected to the nasal apparatus via a sterile tube, and the other was connected to a pump for providing respiratory airflow. b Full-view photograph of the integrated HRT model and magnified photographs of three important modules, showing the inhalation of the viral aerosols near the nasal apparatus (red frame), the airflow control by a pump (blue frame), and the lung organoids container for infection detection (green frame). c CLSM images showing FOXJ1 (ciliated cell marker), SCGB1A1 (club cell marker), P63 (basal cell marker), and SA (H1N1 receptor) expression in paracancerous tissue (upper) and lung organoids (lower). FOXJ1, SCGB1A1, and P63 were stained with corresponding antibodies and secondary Alexa Fluor® 647 fluorescent antibody, SA was stained with Cy3-SNL (red, false color). White: FITC-phalloidine stained cytoskeleton; Blue: DAPI stained cell nuclei. d Representative 3D reconstructed CLSM images of lung organoids in different groups after 24 h or 48 h of challenge with H1N1-CA07 aerosols. White: FITC-phalloidine stained cytoskeleton; Blue: DAPI stained cell nuclei. Green: the fluorescent antibody-stained N protein of H1N1-CA07. e Relative level of H1N1-CA07 mRNA in lung organoids of different groups after 24 h or 48 h of challenge with H1N1-CA07 aerosols. f Relative level of H1N1-PR8 mRNA in lung organoids of different groups after 24 h or 48 h of challenge with H1N1-PR8 aerosols. Data in e and f represent the means ± S.E.M., n = 3 biologically independent experiments. Statistical significance in e and f was tested with two-tailed unpaired t-test. All significant P-values are indicated. Source data are provided in the Source data file.