Mice

All of the experiments were approved by the Institutional Animal Care and Use Committee of MD Anderson Cancer Center. Syt2 conditionally deleted mice with the second exon flanked by loxP recombination sites (floxed, Syt2F/F) were obtained from T. C. Südhof24. These were generated on a mixed 129/Sv:C57BL/6 background and backcrossed by us for ten generations onto a C57BL/6J background. To delete Syt2 in airway epithelial cells, Syt2F/F mice were crossed with mice in which a Cre recombinase optimized for mammalian codon usage was knocked into the secretoglobin 1A1 locus (Scgb1a1Cre) (refs. 25,26). Half of the progeny that resulted from crossing Syt2F/F mice with Syt2F/F mice that were also heterozygous for the Scgb1a1Cre allele were Syt2 deletant (Syt2D/D) mice, and the other half were Syt2F/F mice that served as littermate controls for the mucin-secretion experiments. Genotyping was performed by PCR using the oligonucleotide primers for Syt2 WT and mutant alleles described in ref. 24. Deletion of Syt2 in airway epithelial cells was confirmed by immunohistochemical staining (Extended Data Fig. 1) using primary rabbit polyclonal antibodies against Syt2 (Abcam, ab113545, 1:1,000) and secondary donkey anti-rabbit IgG polyclonal antibodies conjugated to horseradish peroxidase (Jackson ImmunoResearch, 711-005-152, 1:5,000). Peroxidase activity was localized using a diaminobenzidine substrate kit (Vector Laboratories, SK-4100), and the slides were counterstained with haematoxylin. C57BL/6J mice were purchased from the Jackson Laboratory and used as controls to be certain the Syt2F/F allele was not hypomorphic in airway epithelium. As Syt2F/F mice did not differ from WT Syt2 mice at the baseline or in the degree of mucous metaplasia and efficiency of stimulated secretion (Fig. 1), they were used as the primary comparator for Syt2D/D mice to minimize environmental and off-target genetic differences. Mice of both sexes were used aged 6–26 weeks. The animals were housed in specific pathogen free conditions under a 12 h–12 h light–dark cycle with food and water ad libitum. Group sizes were calculated to detect between-group differences with a power of 90% and two-tailed significance of 5%, based upon effect sizes from our numerous prior studies of mucin secretion and lumenal mucus accumulation. The number of animals used was the minimum that is consistent with scientific integrity and regulatory acceptability, with consideration given to the welfare of individual animals in terms of the number and extent of procedures to be carried out on each animal. Mice from the appropriate genotypes were randomly assigned to groups for all of the conditions. Investigators were blinded to the mouse group allocation during data collection and analysis. Furthermore, mouse airway images were analysed by investigators who were blinded to the genotype and treatment of the animals.

Mucin secretion and airway mucus occlusion in mice

The efficiency of stimulated mucin secretion and the extent of mucus accumulation in the airway lumen of Syt2-mutant mice were measured as described previously22. In brief, to increase intracellular mucin content (that is, induce mucous metaplasia), 3 µg IL-13 (BioLegend) in 40 µl PBS was instilled every other day for a total of 3 times into the posterior pharynx of mice under isoflurane anesthesia to be aspirated during inhalation. Three days after the last instillation, mucin secretion was stimulated by exposing mice for 10 min to an aerosol of 100 mM ATP in 0.9% NaCl, then the lungs were collected 20 min later. Transverse sections of bronchial airways of mice were stained with PAFS to demonstrate mucin with red fluorescence. Fractional mucin secretion was calculated as the percentage reduction in intracellular mucin content (see below for quantification) of individual mice after sequential treatment with IL-13 and ATP or methacholine compared with the group mean mucin content of mice of the same genotype treated with only IL-13. To measure intracellular airway epithelial mucin content, the lungs were inflated through the trachea with 10% neutral buffered formalin to 20 cm water pressure for 24 h at 4 °C, then embedded in paraffin. A single transverse 5 µm section was taken through the axial bronchus of the left lung between lateral branches 1 and 2, deparaffinized, rehydrated and stained with PAFS reagent.

Images were acquired using an upright microscope (Olympus BX 60) with a ×40 objective lens (NA 0.75), and intracellular mucin was measured around the circumferential section of the axial bronchus using ImagePro-5.1 (Media Cybernetics). Images were analysed by investigators who were blinded to mouse genotype and treatment (Supplementary Table 1).

Quantification of mucin secretion was performed according to previous protocols22,57. First, using images that were acquired in the red channel alone, the total area and fluorescence intensity of intracellular staining above the basement membrane were measured. Second, using the images captured under both red and green fluorescence, the total surface area of the epithelium and the length of the basement membrane in each field were measured. The volume density of mucin staining in the airway epithelium was then calculated stereologically as the ratio of surface area of staining to total surface area of the epithelium divided by a boundary length measurement, which is a product of the total epithelial surface area, the basement membrane length and the geometric constant 4/π. As a result, data are presented as the epithelial mucin volume density, signifying the measured volume of mucin overlying a unit area of epithelial basal lamina.

To measure airway lumenal mucus content in Syt2-mutant mice, mucous metaplasia was induced as above, then mucin secretion and bronchoconstriction were induced by exposure for 10 min to an aerosol of 150 mM methacholine. Lungs were collected and fixed by immersion for 48 h at 4 °C to avoid displacement of lumenal mucus and using methanol-based Carnoy’s solution (methacarn) for fixation to minimize changes in mucus volume. A single transverse 5 µm section was taken through the axial bronchus of the left lung between lateral branches 1 and 2 and stained with PAFS as above to evaluate intracellular mucin to ensure secretion had been stimulated. Then, six 5 µm sections of the paraffin blocks caudal to the initial section were taken at 500 μm intervals and stained with PAFS. Mucus in the lumens of airways was identified manually and the area summed for all twelve sections using ImagePro-5.1 (Media Cybernetics)22.

To measure the efficiency of stimulated mucin secretion and the extent of mucus accumulation in the airway lumen of WT mice treated with peptides, the same procedures as those used for analysis of Syt2-mutant mice were followed, except that both outcomes were measured in the same mouse because of peptide expense by fixation of the left lung through inflation with formalin to measure mucin secretion and fixation of the right lung through immersion in methacarn to measure mucus accumulation. Stimulation of mucin secretion with a single methacholine aerosol was used for both outcomes. A MicroSprayer Aerosolizer (Penn-Century) was used for intratracheal peptide delivery (50 μl) to the airways, secretion was measured in the left axial bronchus at a site 3 mm distal to the site used in the Syt2-mutant mice because of greater intracellular peptide uptake in more distal airways (Extended Data Fig. 9a), and mucus accumulation was measured in the right caudal lobe taking 8–10 sections of 5 µm thickness at 500 µm intervals.

Details about experimental repeats, the number of data points n (that is, mice) and the number of images analysed for each mouse are provided in Supplementary Table 1. The experiments, involving all the experimental groups, were performed twice on two different occasions, weeks apart. Lung tissues from the mice from each of those independent experiments were processed separately.

Epithelial cell uptake of peptides in mice

To measure the uptake of SP9 and control peptides conjugated to CPP in vivo, peptides were labelled with Cy3 by conjugation at the C-terminal cystine residue, as described below. Labelled peptides (200 μM microsprayer concentration) or PBS were introduced into the airways as aerosols using a Penn-Century MicroSprayer inserted into the distal trachea under direct visualization with a laryngoscope. After 30 min, the mice were euthanized and the lungs were fixed by inflation with 10% neutral buffered formalin as described above for the measurement of mucin secretion. Transverse sections were made of the left axial bronchus, and sections were stained with DAPI to demonstrate nuclei. Fractional uptake of labelled peptides was measure as red Cy3 fluorescent staining over the number of blue DAPI-labelled nuclei in individual cells. To illustrate the relative peptide uptake by secretory and ciliated cells (Extended Data Fig. 9d), immunofluorescence staining of secretory cells was performed using primary goat polyclonal antibodies against CCSP (Millipore Sigma, ABS1673, 1:1,000), secondary donkey anti-goat IgG polyclonal antibodies conjugated to Alexa Fluor 488 (Jackson ImmunoResearch, 705-545-147, 1:1,000) and staining of nuclei with DAPI. Details about the experimental repeats, the number of data points n (that is, cells) and the number of images analysed for each mouse are provided in Supplementary Table 1.

Protein expression and purification

We used the same constructs and protocols to purify cysteine-free Stx1A, SNAP-25A, VAMP2 and Syt1 as described in ref. 41. We used the same constructs and protocols to purify NSF, and αSNAP as described in ref. 58. The protein sample concentrations were measured by ultraviolet light absorption at 280 nm, and aliquots were flash-frozen in liquid nitrogen and stored at −80 °C.

Stx3

Full-length human STX3 was expressed in Escherichia coli strain BL-21 (DE3) with an N-terminal, TEV-protease-cleavable hexa-histidine tag. The expression and purification protocols were mostly identical to that of Stx1A. The protein was expressed overnight at 30 °C in 8 l of autoinducing medium. Cell pellets from 8 l of culture were suspended in 1× PBS, 1 mM EDTA, 1 mM PMSF and 8 EDTA-free protease inhibitor tablets (Roche) supplemented with lysozyme and DNase I (Sigma-Aldrich). The cells were lysed using a sonicator (Thermo Fisher Scientific) and an M-110EH microfluidizer (Microfluidics). Inclusion bodies were removed by centrifugation for 30 min at 13,000 rpm in a JA-14 rotor (Beckman Coulter), and the supernatant was centrifuged at 43,000 rpm for 1.5 h in a Ti-45 rotor (Beckman Coulter) to pellet the membrane. Membranes were resuspended in 20 mM HEPES pH 7.5, 300 mM NaCl, 10% glycerol and centrifuged at 43,000 rpm for 1 h. The pellet was resuspended once more in the same buffer, dodecylmaltoside (Anatrace) was added to 2% and the sample was stirred for 1.5 h at 4 °C. The solubilized membrane was centrifuged at 40,000 rpm for 35 min, and the supernatant was loaded onto a 5 ml column of Nickel-NTA agarose (Qiagen). The column was washed with 20 mM HEPES pH 7.5, 20 mM imidazole, 300 mM NaCl, 110 mM octyl glucoside (Anatrace), 10% glycerol and the protein was eluted with wash buffer supplemented with 450 mM imidazole and 1 M NaCl. The protein fractions were pooled, digested with 110 µg TEV protease and dialysed overnight against 20 mM HEPES pH 7.5, 50 mM NaCl, 110 mM octyl glucoside (OG) (Anatrace) and 10% glycerol. The fractions were loaded onto a MonoQ 4.6/100 PE column (GE Healthcare) that had been equilibrated with dialysis buffer. The protein was eluted with a gradient of 50 mM to 1 M NaCl over 30 column volumes. Protein concentration was measured by absorption at 280 nm and aliquots were flash-frozen in liquid nitrogen and stored at −80 °C.

SNAP-23

The expression and purification protocols were mostly identical to those of SNAP-25A. Cysteine free SNAP-23, in which all cysteine residues were changed to serine, was expressed in BL21(DE3) cells using autoinducing medium from a pGEX vector as an N-terminal GST tag with a thrombin protease cleavage site to remove the tag. Cells from 4.0 l of the induced culture were resuspended in 250 ml of buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 4 mM DTT, 10% glycerol) containing 1 mM PMSF and 5 EDTA-free protease inhibitor tablets. Cells were lysed by sonication. The lysate was clarified by centrifugation in a Ti45 rotor for 35 min at 40,000 rpm. The supernatant was bound to 10 ml of Glutathione Sepharose beads (GE Healthcare) for 1 h with stirring at 4 °C. The beads were collected by centrifugation, poured into a column and washed with 100 ml of buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 4 mM DTT, 10% glycerol). Thrombin (10 µl of 5 mg ml−1) was added to the washed beads along with 5 ml of buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 4 mM DTT, 10% glycerol) and the mixture was rocked overnight at 4 °C to remove the GST tag. The cleaved SNAP-23 sample was washed out of the column using buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 4 mM DTT, 10% glycerol) and concentrated to 5 ml. The sample was injected onto a Superdex 200 (16/60) column (GE Healthcare) equilibrated in 20 mM HEPES pH 7.5, 100 mM NaCl and 10% glycerol. Protein-containing fractions were combined and concentrated to around 100 µM SNAP-23. The protein concentration was measured by absorption at 280 nm and aliquots were flash-frozen in liquid nitrogen and stored at −80 °C.

VAMP8

VAMP8 was expressed in E. coli strain BL-21 (DE3) with an N-terminal, TEV protease-cleavable, GST tag. The protein was expressed overnight at 25 °C in 8.0 l of autoinducing medium. Cell pellets were suspended in 20 mM HEPES pH 7.5, 300 mM NaCl, 2 mM DTT, 1 mM EDTA and 8 EDTA-free protease inhibitor tablets supplemented with lysozyme and DNase I. The cells were lysed using a sonicator (Thermo Fisher Scientific) and an M-110EH microfluidizer. Cell debris was removed by centrifugation for 30 min at 13,000 rpm in a JA-14 rotor, and the supernatant was centrifuged at 43,000 rpm for 1 h in a Ti-45 rotor to pellet the membrane. The pellet was resuspended in 20 mM HEPES pH 7.5, 300 mM NaCl, 2 mM DTT, 1 mM EDTA and 1.5% DDM and solubilized at 4 °C with stirring for 2.5 h. The solubilized membrane was centrifuged at 43,000 rpm for 35 min, the supernatant was mixed with 5 ml of Glutathione Sepharose 4B and incubated overnight at 4 °C with end-over-end mixing. The beads were washed with 20 CV of 20 mM HEPES pH 7.5, 300 mM NaCl, 2 mM DTT, 1 mM EDTA and 110 mM OG. The protein was cleaved off the column by resuspending the beads in 2 ml of wash buffer supplemented with 110 µg of TEV protease, and incubating at 4 °C for 1 h. After digestion, the column was drained and the flow through (containing cleaved VAMP8) was injected onto a Superdex 200 10/300 Increase column (GE Healthcare) equilibrated with 20 mM HEPES pH 7.5, 300 mM NaCl, 2 mM DTT and 110 mM OG. The fractions containing protein were pooled, and protein concentration was measured by absorption at 280 nm. Aliquots were flash-frozen in liquid nitrogen and stored at −80 °C.

Syt2

Syt2 was expressed in BL21(DE3) cells using autoinducing medium from a pGEX vector as an N-terminal GST tag with a thrombin protease cleavage site to remove the tag. Cells from 4 l of induced culture were resuspended in 200 ml of buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 1 mM EDTA and 2 mM DTT), containing 4 EDTA-free protease inhibitor tablets. Cells were lysed by three passes through the Emulsiflex C5 homogenizer (Avestin) at 15,000 p.s.i. Unlysed cells and debris were removed by centrifugation in a JA-14 rotor for 10 min at 8,000 rpm, the supernatant was centrifuged again in the same rotor for 10 min at 8,000 rpm to remove any final debris. The supernatant from the second centrifugation was then centrifuged in a Ti45 rotor for 1 h at 40,000 rpm to collect the membranes. Membranes were resuspended using a Dounce homogenizer in 100 ml of buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 1 mM EDTA and 2 mM DTT) and n-dodecylmaltoside was added to a final concentration of 2% (w/v) to solubilize the membranes overnight at 4 °C with stirring. The extract was clarified by centrifugation using the Ti45 rotor at 40,000 rpm for 35 min. The extract was applied to a 5 ml bed of Glutathione Sepharose by stirring at 4 °C for 2 h. The column was washed with buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 1 mM EDTA, 2 mM DTT) containing 110 mM β-octyl-glucoside and eluted with buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 1 mM EDTA, 2 mM DTT) containing 110 mM β-octyl-glucoside and 20 mM reduced Glutathione. The GST tag was removed by cleavage with 10 µl of 5 mg ml−1 thrombin for 2 h and the Syt2 sample was purified on a monoS column equilibrated in 20 mM HEPES pH 7.5, 100 mM NaCl, 110 mM β-octyl-glucoside and 2 mM DTT (monoS buffer). After washing the column with monoS buffer (20 mM HEPES pH 7.5, 100 mM NaCl, 110 mM β-octyl-glucoside, 2 mM DTT), the protein was eluted using a linear gradient from 100 mM to 1 M NaCl. Protein-containing fractions were combined, the protein concentration was measured by absorption at 280 nm and aliquots were flash-frozen in liquid nitrogen and stored at −80 °C.

Syt1 C2B, Syt2 C2B and Syt1 C2B(QM)

The Syt1 C2B (amino acid range 271–421), Syt2 C2B (amino acid range 272–422) domains and the Syt1 C2B(QM) (amino acid range 271–421, R281A, E295A, Y338W, R398A, R399A) mutant were expressed as GST-tagged fusion proteins in E. coli BL21 (DE3) cells at 30 °C overnight. After collecting the cells by centrifugation, the sample was resuspended in lysis buffer containing 50 mM HEPES-Na pH 7.5, 300 mM NaCl, 2 mM DTT and EDTA-free protease inhibitor cocktail, and then sonicated and centrifuged. The supernatant was incubated with Glutathione Sepharose beads. The resin was extensively washed with 50 ml of wash buffer I containing 50 mM HEPES-Na pH 7.5, 300 mM NaCl and 1 mM DDT, followed by 50 ml of wash buffer II containing 50 mM HEPES-Na pH 7.5, 300 mM NaCl, 1 mM DTT and 50 mM CaCl2. The GST tag was cleaved overnight at 4 °C with PreScission protease (GE Healthcare) in cleavage buffer containing 50 mM HEPES-Na pH 7.5, 300 mM NaCl, 1 mM DTT and 2 mM EDTA. The cleaved proteins were purified by monoS column and gel filtration on Superdex 75 (GE Healthcare). The protein concentration was measured by absorption at 280 nm and aliquots were flash-frozen in liquid nitrogen and stored at −80 °C.

Munc13-2*

The Munc13-2* fragment of Munc13-2 (amino acid range 451–1407, that is, including the C1, C2B and the C-terminally truncated MUN domains, but excluding residues 1326–1343) was cloned into a pFastBac HTB vector with a GST tag and a PreScission cleavage site. The deletion of residues 1326–1343 in this construct prevents aggregation39,59, and the C-terminal truncation improves solubility. Cells from 8 l of SF9 cell culture were resuspended in 200 ml resuspension buffer (RB) (50 mM Tris, pH 8.0, 500 mM NaCl, 1 mM EDTA, 0.5 mM TCEP, 10% glycerol) containing 6 EDTA-free protease inhibitor tablets. The cells were lysed by three passes through the Avestin C5 homogenizer at 15,000 p.s.i. The lysate was clarified by centrifugation for 35 min at 40,000 rpm in a Ti45 rotor. The supernatant was mixed with 15 ml Glutathione Sepharose beads at 4 °C with stirring for 2 h. The beads were washed using the Akta Prime system (GE Healthcare) with 20 ml RB, 90 ml RB + 1% Triton X-100, then eluted with RB + 50 mM reduced Glutathione. Peak fractions were pooled and then 100 µl of 10 mg ml−1 PreScission protease was added and incubated overnight. The cleaved proteins were purified by gel filtration on the Superdex 200 column. The protein concentration was measured by absorption at 280 nm and aliquots were flash-frozen in liquid nitrogen and stored at −80 °C.

Munc18-2

Munc18-2 (amino acid range 1–594) was cloned into a pFastBac HTB vector with an N-terminal hexa-histidine tag and a TEV-cleavage site. Cells from 4.0 l of a SF9 cell culture were resuspended in 100 ml resuspension buffer (RB) (20 mM sodium phosphate, pH 8.0, 300 mM NaCl, 2 mM DTT, 10% glycerol with 1 mM PMSF) containing 6 EDTA-free protease inhibitor tablets. The cells were lysed via 3 passes through the Avestin C5 homogenizer at 15,000 p.s.i. The lysate was clarified by centrifugation for 35 min at 40,000 rpm in a Ti45 rotor. The supernatant was mixed with 3 ml Ni-NTA beads at 4 °C stirring for 1 h. The beads were washed using an Akta Prime system with 20 ml each of RB, then eluted with RB + 300 mM imidazole. Peak fractions were pooled and then 100 μl of 11 mg ml−1 TEV protease was added. The mixture was dialysed overnight against 1 l of 500 ml of 20 mM HEPES pH 7.5, 300 mM NaCl, 2 mM DTT, 10% glycerol. The TEV cleaved protein was injected on a Superdex 200 column. Peak fractions were combined and the protein concentration was measured by ultraviolet light absorption at 280 nm. Aliquots of 100 μl were flash-frozen in liquid N2 and stored at −80 °C.

Peptide synthesis

The stapled peptide SP9 and the non-stapled peptide P0 (Fig. 2c), as well as peptide chimeras with conjugated CPPs or biotin at the N terminus, and/or conjugated fluorescent dye Cy3 labels at the C terminus (SP9–Cy3, biotin–SP9–Cy3, PEN–SP9–Cy3, PEN–P9–Cy3, TAT–SP9–Cy3 and TAT–P9–Cy3 (Fig. 4a); P0–Cy3 (Fig. 2f)) were synthesized and purified by Vivitide (formerly New England Peptide). Peptide synthesis was carried out using solid-phase peptide synthesis and Fmoc chemistry. The peptides were cleaved using trifluoroacetic acid and standard scavengers. The peptides were purified using reverse-phase high-pressure liquid chromatography (RP-HPLC). For the stapled peptides, α,α-disubstituted non-natural amino acids of olefinic side chains were synthesized (S5–S stereochemistry, bridging 5 amino acids).

The hydrocarbon-staple was made using Grubbs catalyst36. For all stapled peptides, the N termini were acetylated and the C termini were amidated. For example, SP9 was synthesized at a 0.2 mmol scale using Rink amide resin on a Liberty Blue instrument (CEM). Standard protecting groups were used for all amino acids. All amino acids were coupled using 5 equivalents of amino acid/HBTU/DIEA relative to resin loading; amino acids after S5 were triple coupled using the same molar excess. Fmoc deprotection was performed with 20% piperidine in dimethylformamide (DMF). After final Fmoc deprotection, the N terminus was acetylated using 0.8 M acetic anhydride and 0.43 M N-methyl-2-pyrrolidone in DMF. Ring-closing metathesis was performed using first-generation Grubbs catalyst in dichloroethane (DCE); the reaction was allowed to proceed overnight protected from light. The resin was then rinsed with DCE, followed by 1% sodium diethyldithiocarbamate trihydrate in DMF (4 × 30 min). The resin was then rinsed with DMF and dichloromethane. The peptide was cleaved and deprotected using trifluoroacetic acid:H2O:ethane-1,2-dithiol:thioanisole/ethylmethylsulfide (84:4:4:4:4) for 3 h, precipitated in ether and centrifuged to pellet. The pellet was resuspended in ether and centrifuged, after which the solvent was decanted. The pellet was dissolved in 1:1 acetonitrile:H2O and lyophilized. The crude peptide was purified by RP-HPLC using a C18 column (10 µm, 120 Å, 25 × 250 mm), and a gradient of 42–58% buffer B (0.1% trifluoroacetic acid in acetonitrile) for 140 min.

The biotin-labelled stapled peptide, biotin–SP9–Cy3, was biotinylated at the N terminus by cross-linking biotin through the carbon spacer 6-aminohexaonic.

For the specified peptides, the C-terminal cystine residue was conjugated to Cy3 fluorescent dyes through maleimide reaction chemistry at pH 7.4 and a 1–2 molar ratio of dye to peptide, and after conjugation the Cy3-labelled peptides were purified again. For example, purified SP9 was dissolved in PBS:acetonitrile:DMSO (2:1:1). Cy3-maleimide (1 equivalent) was dissolved in DMSO and added to the peptide solution; the reaction was allowed to proceed for 1 h in the dark. The conjugated peptide was purified by RP-HPLC using a C18 column (10 µm, 120 Å, 25 × 250 mm), and a gradient of 46–66% buffer B for 140 min.

All of the peptides were purified to >90–95% and quality control was performed by liquid chromatography coupled with mass spectrometry (LC–MS) by the manufacturer (HPLC chromatograms and LC–MS data for SP9, TAT–SP9–Cy3, PEN–SP9–Cy3, PEN–P9–Cy3, TAT–P9–Cy3, SP9–Cy3 and P0 are provided in Supplementary Figs. 28, respectively). Subsequently, the peptides were lyophilized and shipped. The LC–MS quality-control data indicate that the peptides have the predicted molecular mass according to their chemical composition. Moreover, 1H 1D and 2D HSQC, HMBC, ROESY NMR experiments of 5 mg SP9 dissolved in DMSO (Supplementary Fig. 9) show that the staple (S5)-residues are at the expected positions in the peptide sequence and are connected with the neighbouring residues. Although there is some spectral overlap for some of the resonances, the data are that are consistent with formation of two S5-S5 pairs. Taken together, the data show that SP9 has the expected sequence and chemical configuration.

For each group of experiments, aliquots of peptide powder were directly dissolved in the specified buffers at ~1 mM concentration using a vortexer, and then diluted to the specified peptide concentrations. For example, to prepare a stock solution of SP9, 2 mg SP9 peptide powder was weighed out. Considering the molecular mass of 2222 g mol−1 of SP9 (Supplementary Fig. 2), this corresponds to 9 × 10−7 mol. For the desired concentration of 1 mM SP9, we added 9 × 10−7 mol 1−1 × 10−3 mol l−1 = 0.9 ml buffer. The concentration of the stock solution was confirmed by absorption measurement at 205 nm using a Nanodrop instrument (Thermo Fisher Scientific).

CD spectroscopy

CD spectra were measured using the AVIV stop-flow CD spectropolarimeter at 190–250 nm using a cell with a 1 mm path length. The sample containing 100 μM of synthesized peptides in PBS buffer (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4 and 2 mM KH2PO4, pH 7.4) was measured at 20 °C. For the correction of the baseline error, the signal from a blank run with PBS buffer was subtracted from all the experimental spectra. The α-helical content of each peptide was calculated by dividing the mean residue ellipticity [φ]222obs by the reported [φ]222obs for a model helical decapeptide60.

Cryo-electron microscopy

PM and SG vesicles were separately vitrified on lacey carbon grids using the Vitrobot (Thermo Fisher Scientific), and imaged using the FEI Tecnai F20 transmission cryo-electron microscope with a field emission gun (FEI) operated at 200 kV. Images were recorded on a Gatan K2 Summit electron-counting direct detection camera (Gatan) in electron-counting mode61. Nominal magnifications of ×5,000 and ×9,600 (corresponding to pixel sizes of 7.4 Å and 3.8 Å) were used for airway PM and SG vesicles, respectively (Extended Data Fig. 6a). The diameters of the vesicles (Extended Data Fig. 6b) were measured using EMAN2 (ref. 62).

Bulk fluorescence anisotropy measurements

In the bulk fluorescence anisotropy experiments, P0 and SP9 were labelled with the fluorescent dye Cy3 at the C terminus. The fluorescence anisotropy was measured using the Tecan Infinite M1000/PRO fluorimeter using an excitation wavelength of 530 ± 5 nm and emission wavelength of 580 ± 5 nm at 27.2 °C. The fluorescent-dye-labelled samples were diluted to 10 nM concentration in TBS (20 mM Tris, pH 7.5, 150 mM NaCl, 0.5 mM TCEP) for optimal read out.

Molecular dynamics simulations

The starting point for all of the molecular dynamics simulations was the crystal structure of the SNARE–Syt-1–complexin-1 complex at 1.85 Å resolution (PDB: 5W5C)15. Before the simulations, the Syt1 C2A domain, the crystallographic water molecules, Mg2+ and glycerol molecules were deleted from the crystal structure. Specifically, the following residues were included in the simulations of the primary interface: Syt1 C2B (amino acid range 270–419), synaptobrevin-2 (amino acid range 29–66), Stx1A (amino acid range 191–244), SNAP-25A (amino acid ranges 10–74 and 141–194). Complexin-1 was not included in the simulations. For the primary interface (SNARE–Syt1 C2B) simulations, the Syt1 C2B molecule that produces the primary interface was used.

For the simulations of Syt1 C2B–P9, Syt1 C2B (amino acid range 270–419) and residues 37–53 of SNAP-25 were used (corresponding to the P9 sequence: EESKDAGIRTLVMLDEQ). For the simulations of Syt1 C2B–SP9, the Syt1 C2B–P9 complex was used as a starting point and the staples for SP9 were created by using CHARMM topology and parameter files for S5 and the covalent bond between S5 residues63. Initial coordinates for the S5 residues were generated by mutating the native residues into Lys using PyMol v.2.5.1 (Schrödinger), and then using the VMD mutate command64 to change Lys into S5. The SP9 and P9 peptides were simulated with an acetylated N terminus, and an amidated C terminus. For all of the simulations, the NAMD program was used65.

As a control, five 1 μs molecular dynamics simulations of the primary interface were performed in a solvated environment (Extended Data Fig. 2d, e). For these simulations, the starting models were placed in a 113 × 125 × 116 Å periodic boundary condition box. The empty space in the box was filled with 50,420 water molecules using the VMD solvate plugin. The system has a total of 157,833 atoms. The system was charge-neutralized and ionized by addition of 155 potassium and 138 chloride ions, corresponding to a salt concentration of ~145 mM using the VMD autoionize plugin.

For the simulations with P9 and SP9, the starting models were placed in a 80 × 80 × 80 Å periodic boundary condition box. The empty space in the box was filled with ~15,200 water molecules using the VMD solvate plugin. The system has a total of 48,486 atoms. The system was charge-neutralized and ionized by the addition of 42 potassium and 44 chloride ions, corresponding to a salt concentration of ~145 mM using the VMD autoionize plugin.

The CHARMM22 (P9–Syt1 C2B and SP9–Syt1 C2B simulations) or CHARMM36 (primary interface simulations) all-hydrogen force fields and parameters66 were used with a non-bonded cut-off of 11 Å. A constant-pressure method was used by adjusting the size of the box. The particle mesh Ewald method was used to accelerate the calculation of long-range electrostatic non-bonded energy terms. Langevin dynamics (with a friction term and a random force term) was used to maintain the temperature of the simulation. All hydrogen-heavy-atom bonds were kept rigid using the Rattle method as implemented in NAMD.

For the simulations with stapled peptides, in the relaxation step, dihedral angle restraints were added to restrain the S5–S5 CE–CE double bond in the cis conformation, the S5 olefinic side chains in the trans conformation, and all α-helices in the α-helical conformation (using the ssrestraints plugin for VMD). In all of the subsequent steps (heating steps and production runs), all these dihedral angle restraints were turned off. For all of the other simulations with peptides without staples, in the relaxation step, α-helical (secondary structure) restraints were added for all α-helices (using the ssrestraints plugin for VMD). In all of the subsequent steps (heating steps and production runs), all these dihedral angle restraints were turned off. The system was equilibrated by the following procedure: (1) relaxation step, ramping up the temperature from 0 to 50 K for 50 ps with a 1 fs time step; (2) first heating step, ramping up the temperature from 50 to 100 K for 50 ps with a 1 fs time step; (3) second heating step, ramping up the temperature from 100 to 250 K for 150 ps with a 1 fs time step. For all simulations, 1 ns chunks were run at a temperature of 300 K with a time step of 1 fs. Five independent 1 μs simulations were performed for each system (primary interface, SP9–Syt1 C2B, P9–Syt1 C2B) by using different initial random number seeds. As expected, the primary interface is stable in these simulations.

One simulation of P9–Syt1 C2B resulted in a dissociation event (Fig. 2g (green, right)). Interestingly, the dissociated peptide P9 is highly dynamic, revealing a variety of distorted, partially helical conformations. Presumably, the increased dynamics of the non-stapled P9 peptide resulted in the destabilization of the interactions with Syt1 C2B, producing the rather different binding poses of P9 (Fig. 2g).

All of the simulations were performed on the Stanford Sherlock Cluster using 4 nodes, each node consisting of dual ten-core CPU 2.4 Ghz Intel processors, that is, a total of 80 CPUs were used for each simulation. The MPI-parallel NAMD2 2.14b1 executable was used. To visualize the results, only protein components are shown, and all of the structures were fitted to each other, and displayed using PyMOL v.2.5.1.

Vesicle reconstitution

For the ensemble lipid mixing assay, the lipid composition of the SV vesicles was phosphatidylcholine (PC) (46%), phosphatidylethanolamine (PE) (20%), phosphatidylserine (PS) (12%), cholesterol (20%) and 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindodicarbocyanine perchlorate (DiD) (Invitrogen) (2%); for the both neuronal and airway PM vesicles, the lipid composition was brain total lipid extract supplemented 3.5 mol% PIP2, 0.1 mol% biotinylated PE and 2 mol% 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI) (Invitrogen). All the lipids are from Avanti Polar Lipids.

For single-vesicle content mixing assay, the lipid composition of the SV, SG, VAMP2, or VAMP8 vesicles was PC (48%), PE (20%), PS (12%) and cholesterol (20%); for both the neuronal and airway PM vesicles, the lipid composition was brain total lipid extract supplemented 3.5 mol% PIP2 and 0.1 mol% biotinylated PE.

The reconstitution method for neuronal PM and SV vesicles is described in detail in refs. 41,67,68. The same methods were used for airway PM, SG, VAMP2 and VAMP8 vesicles. Dried lipid films were dissolved in 110 mM OG buffer containing purified proteins at protein-to-lipid ratios of 1:200 for VAMP2 and Stx1A for SV and neuronal PM vesicles, respectively (or 1:200 for VAMP8 and Stx3 for SG and airway PM vesicles, respectively), and 1:800 for Syt1 for SV vesicles (or 1:1,200 for Syt2 for SG vesicles).

A three to fivefold excess of SNAP-25A or SNAP-23 (with respect to Stx1A or Stx3) was added to the protein–lipid mixture for neuronal or airway PM vesicles. Detergent-free buffer (20 mM HEPES pH 7.4, 90 mM NaCl and 0.1% 2-mercaptoethanol) was added to the protein–lipid mixture until the detergent concentration was at (but not lower than) the critical micelle concentration of 24.4 mM, that is, vesicles did not yet form. For the preparation of SV, SG, VAMP2,or VAMP8 vesicles for the single-vesicle content mixing assay, 50 mM sulforhodamine B (Thermo Fisher Scientific) was added to the protein–lipid mixture. The vesicles subsequently formed during size-exclusion chromatography using a Sepharose CL-4B column, packed under near constant pressure by gravity with a peristaltic pump (GE Healthcare) in a 5.5 ml column with a ~5 ml bed volume that was equilibrated with buffer V (20 mM HEPES pH 7.4 and 90 mM NaCl) supplemented with 20 µM EGTA and 0.1% 2-mercaptoethanol. The eluent was dialysed into 2 l of detergent-free buffer V supplemented with 20 µM EGTA, 0.1% 2-mercaptoethanol, 5 g of Bio-beads SM2 (Bio-Rad) and 0.8 g l−1 Chelex 100 resin (Bio-Rad). After 4 h, the buffer was exchanged with 2 l of fresh buffer V supplemented with 20 µM EGTA, 0.1% 2-mercaptoethanol and Bio-beads, and the dialysis was continued overnight for another 12 h. We note that the chromatography equilibration and elution buffers did not contain sulforhodamine, so the effective sulforhodamine concentration inside SV, SG, VAMP2 or VAMP8 vesicles is considerably (up to tenfold) lower than 50 mM. For the ensemble lipid mixing assay, the reconstitution method is the same as that for the single-vesicle content mixing assay, except that 50 mM sulforhodamine B was omitted for all the steps.

As described previously67, the presence and purity of reconstituted proteins in the airway system was confirmed by SDS–PAGE of the vesicle preparations and the directionality of the membrane proteins (facing outward) was assessed by chymotrypsin digestion followed by SDS–PAGE gel electrophoresis. The size distributions of the airway PM and SG vesicles were analysed by cryo-electron microscopy (Extended Data Fig. 6a, b) as described previously69.

Single-molecule counting experiments with SP9–Cy3

PEG-coated flow chambers were prepared using the same protocol as for the single-vesicle content mixing experiments. Freshly synthesized SP9–Cy3 powder was first dissolved in an imaging buffer (20 mM HEPES pH 7.4, 90 mM NaCl and 0.5 mM TCEP) at a concentration of 50 μM, then centrifuged at around 16,000 g for 10 min to remove potential insoluble materials. The sample’s concentration was remeasured by absorption by Cy3 at 550 nm before serial dilution to concentrations of 100 nM, 10 nM, 1 nM and 0.5 nM. Diluted sample (5 μl) was injected into a flow chamber on the quartz slides followed by an immediate (~500 μl) wash with imaging buffer. There is some degree of non-specific binding of SP9–Cy3 to the imaging surface, enabling the counting of molecules in fluorescent spots by observing single-molecule photobleaching events70. After quickly focusing, the sample stage was moved to a fresh location within the same sample chamber distant from prior illumination and the recording was started before exciting SP9–Cy3 by green (532 nm) laser light at an excitation power of ~8 mW. Multiple recordings were performed at fresh locations within the same chamber.

The number of SP9–Cy3 molecules in a fluorescent spot was counted by observing sequential stepwise photobleaching events. Fluorescent spots were automatically detected by smCamera and time traces for each spot generated. The time traces were automatically analysed by Hidden Markov modelling71,72 using a script written for MATLAB. We applied constraint-based clustering to initiate the Hidden Markov model (HMM) and calculated the probability matrices of transition and emission iteratively. The most probable state sequences were then reconstructed with a standard Viterbi algorithm. The time traces and automatic HMM fits were manually inspected. For many traces, there were distinct stepwise decreases in fluorescence intensity where the stepwise decreases were approximately as recognized by HMM (Extended Data Fig. 3c). Traces were selected that showed distinct stepwise fluorescence intensity decreases and that had undergone complete photo-bleaching at the end of the observation period. Histograms of the number of photo-bleaching steps (also known as the number of SP9–Cy3 molecules per fluorescent spot) were then generated (Extended Data Fig. 3d).

Single-vesicle content mixing experiments

All single-vesicle fusion experiments were performed on a prism-type total internal reflection fluorescence microscope using 532 nm (green) laser (CrystaLaser) and 637 nm (red) laser (OBIS) excitation. Two observation channels were created by a 640 nm single-edge dichroic beamsplitter (FF640-FDi01-25×36, Semrock): one channel was used for the fluorescence emission intensity of the content dyes and the other channel for that of the Cy5 dye that is part of the injected Ca2+ solution. The two channels were recorded on two adjacent rectangular areas (45 × 90 μm2) of a charge-coupled device camera (iXon+ DV 897E, Andor Technology). The imaging data were recorded and analysed using the smCamera program73 developed by K. Suk Lee and T. Ha. Fluorescent peaks were automatically detected using smCamera and time traces were saved in smCamera format as well as in plain text (scripts to convert the smCamera files to tiff and text files were provided by M. Hyn Jo). Candidates for fusion events in the time traces were detected using a script written for MATLAB and then confirmed by manual inspection.

Flow chambers were assembled by creating a ‘sandwich’ consisting of a quartz slide and a glass coverslip that were both coated with polyethylene glycol (PEG) molecules, including 0.1% (w/v) biotinylated-PEG except when stated otherwise, and using double-sided tape to create up to five flow chambers. Coating the surface with PEG molecules alleviates non-specific binding of vesicles. The same protocol and quality controls (surface coverage and non-specific binding) were used as described previously67,74 except that PEG-SVA (Laysan Bio) instead of mPEG-SCM (Laysan Bio) was used as it has a longer half-life. The flow chambers were incubated with neutravidin for 30 min (0.1 mg ml−1).

For the single-vesicle fusion experiments described in Fig. 3 and Extended Data Figs. 4, 5 and 7, biotinylated neuronal or airway PM vesicles (100× dilution) were tethered to the imaging surface by incubation at room temperature (25 °C) for 30 min followed by three rounds of washing with 120 µl buffer V to remove unbound neuronal or airway PM vesicles; each buffer wash effectively replaces the 3 µl flow chamber volume more than 100 times.

For the complete reconstitution (Fig. 3), to form airway SM vesicles with reconstituted Stx3–Munc18-2 complex, we added the ‘disassembly factors’ (1 μM Munc18-2, 0.5 μM NSF, 5 μM αSNAP, 3 mM ATP and 3 mM Mg2+) to tethered airway PM vesicles (Fig. 3b), according to previous work with neuronal proteins41. This procedure results in tethered SM vesicles. Next, the flow chamber with the tethered SM vesicles was washed with buffer V along with 0.5 μM Munc-13-2* and 2 μM SNAP-23.

For all of the reconstitution experiments, after the start of illumination and recording of the fluorescence from a particular field of view of the flow chamber, SV, SG, VAMP2 or VAMP8 vesicles (diluted 100 to 1,000 times; including peptides at the specified concentration, 0.5 μM Munc-13-2* and 2 μM SNAP-23, if applicable) were loaded into the flow chamber to directly monitor vesicle association of SG, SV, VAMP2 or VAMP8 vesicles to neuronal or airway PM vesicles for 1 min. When peptide was included in a particular experiment, it was mixed with the SV, SG, VAMP2 or VAMP8 vesicles before loading into the flow chamber. Thus, the peptide would have a chance to bind to Syt1 or Syt2 in the SV or SG vesicles before loading them into the flow chamber. While continuing the recording, the flow chamber was washed three times with 120 µl of buffer V (including peptides at the specified concentration, 0.5 μM Munc-13-2* and 2 μM SNAP-23, if applicable) to remove unbound vesicles.

For the complete reconstitution (Fig. 3), note that Munc13-2* will catalyse the transfer of Stx3 from the Stx3–Munc18-2 complex into the ternary SNARE complex with SNAP-23 and VAMP8; we therefore call the tethered vesicles again as PM vesicles after this transfer (Fig. 3b).

Subsequently, we continued recording for another minute to monitor spontaneous fusion events. To initiate Ca2+-triggered fusion events within the same field of view, a solution consisting of buffer V, 500 µM Ca2+ or 50 µM Ca2+, 500 pM Cy5 dye molecules (used as an indicator for the arrival of Ca2+ in the evanescent field) and, if applicable, peptide was injected into the flow chamber. The injection was performed at a speed of 66 µl s−1 by a motorized syringe pump (Harvard Apparatus) using a withdrawal method similar to the one described previously74.

Multiple acquisition rounds and repeats for the single-vesicle content mixing experiments

To increase the throughput of the assay and make better use of the vesicle samples, after intensive washing (3 × 120 µl) with buffer V (which includes 20 μM EGTA to remove Ca2+ from the sample chamber), we repeated the entire acquisition sequence (SV, SG, VAMP2 or VAMP8 vesicle loading, counting the number of freshly associated vesicle-vesicle pairs, monitoring of Ca2+-independent fusion, Ca2+-injection and monitoring of Ca2+-triggered fusion) in a different imaging area within the same flow chamber. Five such acquisition rounds were performed with the same sample chamber. SV, SG, VAMP2 or VAMP8 vesicles were diluted 1,000× for the first and second acquisition rounds, 200× for the third and fourth acquisition rounds, and 100× for the fifth acquisition round to offset the slightly increasing saturation of the surface with SG, SV, VAMP2 or VAMP8 vesicles. The entire experiment (each with five acquisition rounds) was then repeated several times (Supplementary Table 2) (referred to as repeat experiment). Among the specified number of repeats, there are at least three different protein preparations and vesicle reconstitutions, so the variations observed in the bar charts reflect sample variations as well as variations among different flow chambers. At least two independent reconstitutions were performed for each condition, and multiple technical repeats were performed using different imaging areas, so the number n refers to the number of repeats combining at least two independent reconstitutions for each condition; all of the repeats were successful, and the number of repeats was deemed to be sufficient to reach significance between different conditions.

Cell culture

Primary HAE cells from several donors were obtained from Promocell at passage 2 or isolated from fresh tissues that were obtained during tumour resections or lung transplantation with fully consent of patients (Ethics approval: ethics committee Medical School Hannover, project no. 2701-2015). Cells were isolated according to the protocol by ref. 57, aliquots were maintained in liquid nitrogen until use. HAE cells from individual donors were thawed and expanded in a T75 flask (Sarstedt) inAirway Epithelial Cell Basal Medium supplemented with Airway Epithelial Cell Growth Medium Supplement Pack (both Promocell) and with 5 μg ml−1 Plasmocin prophylactic, 100 μg ml−1 Primocin and 10 μg ml−1 Fungin (all from InvivoGen). Growth medium was replaced every two days. After reaching 90% confluence, HAE cells were detached using DetachKIT (Promocell) and seeded into 6.5 mm Transwell filters with a 0.4 μm pore size (3470, Corning Costar). The filters were precoated with collagen solution (StemCell Technologies) overnight and irradiated with ultraviolet light for 30 min before cell seeding for collagen cross-linking and sterilization. Cells (3.5 × 104) in 200 µl growth medium were added to the apical side of each filter, and an additional 600 µl of growth medium was added basolaterally. The apical medium was replaced after 48 h. After 72–96 h, when cells reached confluence, the apical medium was removed and basolateral medium was switched to differentiation medium ± 10 ng ml−1 IL-13 (IL012; Merck Millipore). Differentiation medium consisted of a 50:50 mixture of DMEM-H and LHC Basal (Thermo Fisher Scientific) supplemented with Airway Epithelial Cell Growth Medium SupplementPack as previously described75 and was replaced every 2 days. Air lifting (removal of apical medium) defined day 0 of ALI culture, and cells were grown at ALI conditions until experiments were performed at day 25 to 28. To avoid mucus accumulation on the apical side, HAE cell cultures were washed apically with Dulbecco’s phosphate buffered solution (DPBS) for 30 min every 3 days from day 14 onwards.

Mucin-secretion assay in HAE cells

Mucin-secretion experiments under static, that is, non-perfused, conditions were conducted as described previously51,52,76 with modifications for the peptide treatments. In brief, for the 24 h peptide treatment, 20 µl of DMEM ± 100 µM peptides was added to the apical surface 24 h before stimulation. On the day of the assay, cells were washed five times with 100 µl DMEM for 1 h for each wash on the apical side. Apical supernatants were collected after every wash (wash 1–5), then 100 µl of DMEM ± 100 µM peptides was added to the apical surface and HAE cells incubated for 15 min before collecting the supernatant (baseline wash). HAE cells were then incubated for an additional 15 min with 100 µl DMEM ± 100 µM ATP (Sigma-Aldrich) before collecting the supernatants (experimental washes) (Fig. 4e). After sample collection, cells were lysed in 100 µl of lysis buffer (lysate) containing 50 mM Tris-HCl pH 7.2, 1 mM EDTA, 1 mM EGTA, 1% Triton-X (Sigma-Aldrich), protease inhibitor cOmplete mini EDTA-free and phosphatase inhibitor PhosSTOP (Roche). The protocol was adapted for 30 min peptide treatment as follows. After wash 5, 100 µl of DMEM ± 10 µM or 100 µM of peptides (Fig. 4e) was added to the apical surface and HAE cells incubated for 30 min (baseline wash). HAE cells were then incubated for 30 min with 100 µl DMEM ± 100 µM ATP to collect experimental washes.

All of the samples were diluted 1:10 in PBS (washes and cell lysates) and 50 µl of each sample was vacuum-aspirated onto a 0.45 µm pore nitrocellulose membrane using the Bio-Dot MicrofiltrationApparatus (Bio-Rad). Subsequently, membranes were incubated with Intercept blocking buffer (Li-Cor) for 1 h before probing with anti-MUC5AC (MA1–21907, Invitrogen) added at 1:250 in Intercept blocking buffer for 1 h. Membranes were then washed four times for 10 min in PBS-Tween-20 (PBST) before incubation with the IRDye secondary antibodies (926–33212 or 926–68072; Li-Cor) diluted at 1:10,000 in Intercept blocking for 1 h. All of the steps were performed at room temperature. Fluorescent signals were acquired using the Odyssey Fc Imaging System (Li-Cor) and quantified using ImageJ (v.2.0.0; NIH). Equal volumes of samples were loaded on the gels for control and peptide treatments (all of the raw gels are provided in Supplementary Fig. 1). Differences in total MUC5AC signal result from differences in IL-13 induced metaplasia between individual filters. Stimulated secretion was therefore normalized to baseline secretion within individual filters to account for filter-to-filter heterogeneities.

To account for donor heterogeneity, all of the relevant experiments were performed in HAE cell ALI cultures generated from at least four individual donors. Complete sets of control and experimental conditions were conducted in ALI cultures from the same donor. Donors were selected randomly from our depository. Individual ALI cultures from the same donor were then randomly allocated to a control treatment groups. Thus, covariates including sex, age and clinical history were identical in all of the conditions. No blinding was performed. Donor numbers are indicated in the respective figure legends. Donor participant sex, age and smoking status is listed in Supplementary Table 3.

Immunofluorescence staining in HAE cells for CPP uptake experiments

HAE cells grown on Transwell filters were incubated with 20 µl of DMEM ± 100 µM specified peptides (Fig. 4a) on day 28 of establishing ALI. Then, 24 h later, cells were fixed for 20 min in 2% paraformaldehyde in DPBS. Cells were then permeabilized for 10 min with 0.2% saponin and 10% FBS (Thermo Fisher Scientific) in DPBS. Cells were washed twice with DPBS and stained with anti-MUC5AC (45M1, MA1-21907, Thermo Fisher Scientific) antibodies diluted 1:100 in DPBS, 0.2% saponin and 10% FBS overnight at 4 °C. Subsequently, cells were washed twice with DPBS and incubated for 1 h at room temperature in DPBS, 0.2% saponin and 10% FBS containing AlexaFluor-488-labelled anti-mouse secondary antibodies (1:500; Thermo Fisher Scientific) and DAPI (1:5,000; Thermo Fisher Scientific). Images were taken on an inverted confocal microscope (Leica TCS SP5) using a ×40 lens (Leica HC PL APO CS2 40×1.30 OIL). Images for the blue (DAPI), green (AlexaFluor 488) and red (Cy3) channels were taken in sequential mode using appropriate excitation and emission settings.

Image analysis for analysis of CPP uptake in HAE cells

Serial sections of images along the basolateral to apical cell axis (z axis) were acquired with a 0.28 µm distance between individual z-sections to analyse the distribution of intracellular Cy3 fluorescence (Extended Data Fig. 8). Fluorescence intensity profiles along the z axis in individual cells were calculated for all of the channels using the Lecia LAS X software (Leica). In brief, fluorescence intensities of DAPI, AlexaFluor 488 (MUC5AC) and Cy3 were analysed within individual cells at each z-section, normalized and fluorescence intensity traces were calculated along the basolateral to apical cell axis. Traces were exported to GraphPad Prism 7 for graph plotting. For quantitative analysis of intracellular Cy3 fluorescence intensities, maximum projections of all z-sections were calculated using the Leica LAS X software and average fluorescence intensities were analysed for individual MUC5AC+ cells. Experiments to analyse peptide uptake were performed in HAE cell ALI cultures from two individual donors and complete sets of experimental conditions were conducted in ALI cultures from both donors.

For binding of biotin–SP9–Cy3 to bacterial toxins, C2 and CRM197 were conjugated to streptavidin. Biotin–SP9–Cy3 and streptavidin-conjugated toxins were mixed at a 10:1 ratio at 30 °C for 30 min before adding to cells.

Quantification and statistical analysis

Origin, MATLAB and Prism were used to generate all curves and graphs. The fusion experiments were conducted at least three times with different protein preparations and vesicle reconstitutions, and properties were calculated as the mean ± s.e.m. Two-tailed Student’s t-tests were used to test statistical significance in Figs. 1, 3 and 5 and Extended Data Figs. 4, 5 and 7 with respect to the specified reference experiment. Statistical significance in Fig. 4 and Extended Data Fig. 8 was assessed using ANOVA followed by post hoc Dunnett’s test or by two-tailed Student’s t-tests, where appropriate.

Box plots are defined as follows: the whiskers show the minimum and maximum values (excluding outliers), the box limits show the 25% and 75% percentiles, the square point denotes the mean, and the centre line denotes the median.

Software and code

The HAE cell data collection was performed using the Leica LAS X v3.1.5.16308, Li-Cor Odyssey Fc Imaging System v.5.2. The data for the single-vesicle fusion experiments and single molecule counting experiments were collected by the smCamera program developed by T. Ha.

Data analysis was performed for the HAE cell experiments using MS Excel for Mac v.16.36, GraphPad Prism v.7 and NIH ImageJ v.2.0.0-rc69. For the mouse experiments, ImagePro-5.1 (Media Cybernetics) was used. For the single-vesicle fusion experiments, single-molecule counting experiments, fluorescence anisotropy experiments and circular dichroism experiments, OriginPro 8 and MATLAB-2021b were used. EMAN2-2.91 was used to analyse the Cryo-EM images in Extended Data Fig. 6a. NAMD2 v.2.14b1 was used for the molecular dynamics simulations. Pymol v.2.5.1 was used for modelling mutations and visualization.

Animal statement

All the mouse work was conducted in accordance with the UT MD Anderson Cancer Center IACUC guidelines, and under the IACUC supervision; protocol no. 00001214-RN02.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this paper.



Source link