Fig 2: Cdh23 promotes cell migration and aggregation but not cell proliferation. (A) HEK293T cells were transfected with 100 and 200 pmol of Cdh23 siRNA; after 72 h, cell proliferation/viability was measured using MTT assay by monitoring the absorbance at 570/630 nm. The absorbance was then expressed as a percentage of cells in comparison with the control (scrambled siRNA, considered 100%; mean ± SD values for each time point were compared with regard to scrambled control using a two‐tailed paired t‐test). (B) Cell cycles were determined by quantitation of DNA content using PI as staining in flow cytometry after 72 h of transfection with 100 and 200 pmol of Cdh23‐siRNA. The data were expressed as a percentage of cells in different cell cycle phases. Mean ± SD values for each time point were compared with regard to scrambled control using a two‐tailed paired t‐test. (C) Transwell migration assay, and (D) hanging‐drop assay were performed on scrambled‐transfected HEK‐293T cells (control), Cdh23 over‐expressed HEK293T cells (pCdh23), 200 pmol Cdh23‐siRNA (siRNA)‐transfected HEK‐293T cells, and both Cdh23‐siRNA and pCdh23 (pCdh23 + Cdh23‐siRNA)‐transfected HEK‐293T cells (4× magnification). Scale bar: 1.2 mm. In transwell assay, the data were expressed as the percentage of cells that migrated through the membrane compared with control (100%) on the y‐axis and without pCdh23/siRNA on the x‐axis. Mean ± SD values were compared with the with regard to scrambled control using a two‐tailed paired t‐test. In the hanging‐drop assay, the HEK293T cells were treated as mentioned previously for the migration assay, marked in the figure, the treated cells observed for days, and each picture analyzed in gray scale using the free version of imagej. The change in the density was monitored and the spatial distribution of HEK293T cells with the number of days. To quantify the observed changes in the images, a LoR (red‐line) was plotted, and the RSD (RSD = SD in the intensity/mean‐intensity) along the LoR was estimated (mean ± SEM values with regard to respective controls were compared using two‐tailed Mann–Whitney Test) as shown in the righthand most side panel for day 1, day 3, and day 5 of given treatments (on x‐axis). (**P < 0.01, ***P < 0.001).

Fig 3: Silencing Cdh23 promotes cancer cell migration and inhibits cell aggregation. (A) Relative expression of Cdh23 mRNA (expressed as mean ± SEM) was estimated in various cancer cell lines by qRT‐PCR and plotted as a histogram. Based on their intrinsic expression of Cdh23, four cell‐lines, HeLa, L132, A549, and WRL68, HeLa being smallest and WRL68 the biggest, were sorted for the migration and aggregation assays. (B) Transwell cell migration assay was performed with the four sorted cell‐lines, HeLa, L132, A549, and WRL68, after treating with various doses of Cdh23‐siRNA and plotted as a percentage of cells (mean ± SD values were compared with regard to scrambled control using a two‐tailed paired t‐test) which migrated through the membrane. (C) Modified hanging‐drop assay was performed as previously with the four sorted cell‐lines 72 h after the treatment with three different doses of Cdh23‐siRNA (scrambled as control, 100, and 200 pmol). The quantitative estimation of the density and the distribution of cells on the spot area were measured by following the previously described method and plotted RSD (mean ± SEM values were compared with regard to scrambled control two‐tailed Mann–Whitney test) for individual cell lines. (D) Wound‐healing assay was performed after 12 h of transfection of scrambled control (Scram_C) and Cdh23‐siRNA (100 pmol, siRNA_100; 200 pmol, siRNA_200) to three cell‐lines (HeLa, L132, and A549) and the percentage decrease in the free space of wound area was calculated using T‐scratch (on the y‐axis) for Day 1, Day 2, and Day 3 (days on x‐axis; mean ± SD values were compared with regard to scrambled control for each time point using a two‐tailed paired t‐test, 10× magnification). Scale bar: 100 μm. (E) The transwell cell migration assay was performed for three cell‐lines (HeLa, L132, and A549) after treating with 200 pmol scrambled control (Control), 200 pmol Cdh23‐siRNA, 2 ng ECdh‐shRNA, and combination of Cdh23‐siRNA + ECdh‐shRNA, and expressed as percentage of cells migrated through the membrane compared with control (100%; mean ± SD values were compared with regard to control using a two‐tailed paired t‐test). (F) The hanging‐drop assay was repeated with A549 cells to compare the contribution of ECdh in aggregation. The cells were transfected with 200 pmol scrambled control (Control), 200 pmol Cdh23‐siRNA, 2 ng ECdh‐shRNA, and a combination of Cdh23‐siRNA + ECdh‐shRNA, and analyzed as before from the density and the distribution of cells on the drop‐spot. RSD values (mean ± SEM values were compared using two‐tailed Mann–Whitney Test) were plotted for Day 1 (24 h) and Day 3 (72 h) for different treatment conditions. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig 4: Cdh23 is uniformly expressed at the cell boundaries of various tissues. (A) Cdh23 mRNA expression in various mice tissues was measured by qRT‐PCR. To minimize the contribution of Cdh23 mRNA from fibroblasts, NIH3T3 fibroblast cells were used as a background control, and the relative expression of Cdh23 mRNA (expressed as mean expression ± SEM) for each tissue was obtained after subtracting the background control. (B) Paraffin‐embedded tissue sections of various mice tissues (kidney, muscle, brain, testes, and heart) were IHC‐stained with 3,3′‐diaminobenzidine (DAB) to indicate the expression of the Cdh23 protein (brown color, 20× magnification). Scale bar: 150 μm. (C) Cdh23 was over‐expressed in HEK293T cells (Green) using Cdh23‐GFP plasmid (pCdh23) and the protein expression was recorded by monitoring the GFP‐fluorescence using a confocal microscope (60× magnification). Scale bar: 2 μm. (D) Adenocarcinomic human alveolar basal epithelial cells (A549 cells) expressing intrinsic Cdh23 were immunocytochemically (IC) stained with fluorescent antibodies of β‐catenin with CF594 (green, second panel) and Cdh23 with CF633 (red, third panel). Co‐localization of Cdh23 with β‐catenin (Orange) was marked by yellow arrows (last right panel; 60× magnification). Scale bar: 5 μm). (E) TMA section of Lung (top row) and esophageal cancer (bottom row) were stained by IHC using DAB to show the expression of Cdh23 (brown color; normal tissues in 20× magnification). Scale bar: 150 μm. Carcinoma tissues in 4× magnification. Scale bar: 500 μm.

Fig 5: Promoter methylation regulates Cdh23 expression. (A) Dose‐response curves (four repeats) by MTT assay showing A549 cell viability with increasing concentrations of Aza with regard to 100% viability with no drug (control). (B) Relative expression of Cdh23 mRNA was measured by qRT‐PCR after treating with 5 and 50 μm of Aza (mean ± SD values were compared with control using two‐tailed paired t test. (C) Expression of Cdh23 in patients (n = 456, RSEM values) was correlated to methylation levels (TCGA, HM450) for LUAD (Spearman r = −0.360, 95% CI −0.4396 to −0.2749) and ESCC (Spearman r = −0.4586, 95% CI −0.5684 to −0.3327). (D) Comparison of relative methylation levels in Cdh23 gene at promoter, CpG region, 5′‐UTR, gene body and 3′‐UTR for LUAD tumor (n = 463) and normal (n = 32) samples as obtained in MethHC database (mean ± SD values were compared using the two‐tailed Mann–Whitney test). (E) Relative methylation levels (mean ± SD) as reported in TCGA Wanderer database at specific CpG island (Chromosome 10: 73156164–73157954) near the Cdh23 promoter for LUAD tumor (n = 463) and normal (n = 32) samples. and (F) Pattern of relative methylation levels in Cdh23 gene (Chromosome 10: 73156377–73575440) in normal (n = 32, green) and tumor (n = 463, red) samples. (**P < 0.01, ***P < 0.001).