Fig 2: Cadherin-26 (CDH26) transcript variant A is expressed in bronchial epithelial cells.a Exon maps and patterns of gene expression for CDH26 variant A (blue) and variant B (red). Top plot: stacked bar plot of relative transcript abundance (in TPM = transcripts per million) for two CDH26 isoforms based on RNA-seq data for Epibank AECs (n = 141) grown at air–liquid interface (ALI) for at least 3 weeks. Bottom plot: bar plot of total library size (i.e., transcript abundances summed across all transcripts within a sample) for all AECs, arranged in the same order as the upper plot, demonstrating a lack of correspondence between increasing CDH26 expression and size of the dataset. b CDH26A transcript levels in a DNA gel for universal lung, fresh cell isolates from tracheas, AECs cells grown in culture, and clonal cell lines. GAPDH is use as the PCR reaction loading control. c Time-dependent gene expression (means and s.e.m.) for CDH26A in EpiBank AECs (n = 5 donors) grown under submerged and ALI conditions. Asterisks indicate significant difference from submerged cells: *p < 0.05, **p < 0.005 (Kruskal–Wallis test with Dunn’s Multiple Comparisons test). Time points correspond to 1, 3, 7, 14, and 21 days at ALI. d Time-dependent protein expression for CDH26A in a representative donor grown under ALI conditions. HeLa cell lysate loaded as a negative control for lack of CDH26A protein expression and GAPDH used as a protein loading control.

Fig 3: CDH26 variant A co-localizes with the actin cytoskeleton.a HeLa cells co-expressing Clover-N1 and actin marker mRuby2-LifeAct. b HeLa cells co-expressing CDH26-Clover-N1 and actin marker mRuby2-LifeAct show a co-localization pattern corresponding to cortical actin structures (yellow). Imaris 3D rendering to show relationships between fluorophors. ROI shows field selected to set threshold for co-localization. c Pearson’s correlation and Mander’s coefficient for the co-localization of Clover-mRuby2. Figures are representative of data from n = 3 separate experiments and 6 fields per condition. d Representative figure of the fluorescent intensities of the donor Clover presented as pre/post-photobleaching of the mRuby2 acceptor. Scale bar = 5 µm. e FRET efficiency measured in cells expressing Clover-N1 or CDH26-Clover-N1 as the donor and mRuby2-LifeAct as the acceptor fluorophore. Clover-10-mRuby2 as a positive control for FRET. Data from single cells measured, n = 8–10 fields at 60×, two separate experiments.

Fig 4: CDH26 variant A has functional cadherin domains.a Aggregation assay of CHO-K1 cells expressing Empty-GFP or CDH26-GFP in the presence or absence of calcium to test homotypic binding. Data are from 4 separate experiments with two replicates per experiment. Data presented as mean and s.e.m. *** indicates significantly different compared to empty-GFP CHO-K1 cells, p < 0.0001; +indicates significantly different compared to CDH26-GFP without calcium, p < 0.05. b Migration and adhesion assays of HeLa cells expressing Empty-GFP or CDH26-GFP. Data are from n = 4 separate experiments with three replicates per experiment. Data presented as mean and s.e.m., statistics pair t-test where **p < 0.005 and ***p < 0.0001 is significant to Empty-GFP HeLa cells. Bottom panel—10µg of HeLa cell lysate from Empty-GFP or CDH26-GFP cells blotted with anti-CDH26 to verify expression of CDH26-GFP. GAPDH used as a protein loading control. (c) Transepithelial resistance (TEER) measurements from untransfected HeLa cells or HeLa cells expressing CDH26-StrepII tag. n = 3 replicates and data presented as mean and s.e.m., statistics pair t-test where *p < 0.05 is significant to non-transfected HeLa cells. Bottom panel—10 µg of HeLa cell lysate from non-transfected or CDH26-StrepII cells blotted with anti-StrepII to verify expression of CDH26-StrepII. GAPDH used as a protein loading control. Recombinant CDH26 sedimentation assay in the absence of calcium (d) and in the presence of calcium (e) showing formation of multimeric protein aggregate structures. Top panels: in-gel staining for total protein. Bottom panels: blotted with anti-CDH26 to demonstrate sedimentation in each fraction. f Co-immunoprecipitation of CDH26 by actin from AECs visualized by in-gel total protein (upper) and western blot (lower) *Asterisks on the blot represent the IP bands pulled out in the co-IP reaction. Figure is representative of data obtained from n = 5 donor lysates from cells grown at ALI for at least 3 week in culture.

Fig 5: CDH26 expression is important for maintaining epithelial expression of planar cell polarity proteins.a CDH26 variant A gene expression by qPCR for samples at day 7 ALI confirming knockdown of CDH26A in n = 7 donors paired with scramble controls. Red bar-median expression of CDH26, dotted line represents the effects of knockdown within each individual donor. Statistics paired t-test where **p < 0.005 when compared to scramble controls. b Gene expression of proteins associated with planar cell polarity (PCP) by qPCR in the same n = 7 donors KD CDH26A, with data presented as mean and s.e.m. with statistics paired t-test where *p < 0.05 when compared to scramble controls. c Quantification of PCP protein CRB3 in AECs by immunofluorescence at different time points in ALI culture. d Expression and localization of CRB3 protein in scramble or KD CDH26A cells at 21 days ALI. e Expression and localization of centrin-1 protein in scramble or KD CDH26A cells at 21 days ALI. Images are representative from survey of n = 3 donors and 3 fields per condition, data presented as mean and s.e.m. with statistics non-parametric paired t-test with Wilcoxon matched-pairs signed rank test where *p < 0.05 and **p < 0.005 significant to scramble controls. Scale bar = 10 µm.