Cadherin-26 (CDH26) regulates airway epithelial cell cytoskeletal structure and polarity

Polarization of the airway epithelial cells (AECs) in the airway lumen is critical to the proper function of the mucociliary escalator and maintenance of lung health, but the cellular requirements for polarization of AECs are poorly understood. Using human AECs and cell lines, we demonstrate that cadherin-26 (CDH26) is abundantly expressed in differentiated AECs, localizes to the cell apices near ciliary membranes, and has functional cadherin domains with homotypic binding. We find a unique and non-redundant role for CDH26, previously uncharacterized in AECs, in regulation of cell–cell contact and cell integrity through maintaining cytoskeletal structures. Overexpression of CDH26 in cells with a fibroblastoid phenotype increases contact inhibition and promotes monolayer formation and cortical actin structures. CDH26 expression is also important for localization of planar cell polarity proteins. Knockdown of CDH26 in AECs results in loss of cortical actin and disruption of CRB3 and other proteins associated with apical polarity. Together, our findings uncover previously unrecognized functions for CDH26 in the maintenance of actin cytoskeleton and apicobasal polarity of AECs.

Institute Microscopy Core. Analysis was performed using ImageJ measurement tool of integrated density as a non-biased measurement of fluorescent intensity, as previously described [1]. 3D rendering was performed by Imaris 7.2 imaging software.

CHO-K1 Aggregation Assays
In brief, Empty-GFP or CDH26-GFP expressing cells were washed in Ca +2 /Mg +2 free HEPES and resuspended in Ca +2 -free HANKS buffered with HEPES and 1% BSA at a cell density of 5.0 x 10 4 /mL. Cells were made into a single cell suspension by passing through a 21g needle and plated in a 24 well plate at a density of 250,000 cells per well.
To determine if calcium enhanced aggregation, 2mM/L CaCl 2 solution was added and compared to wells containing no calcium. Cells were incubated for 90 mins at 37 °C at 5 x g on an orbital shaker. Assay was stopped by adding 5% PFA and large aggregates were photographed using green fluorescence filter on the ImageQuant LAS4000.
Aggregation presented as % change increase above empty-GFP control.

Recombinant CDH26 Protein Synthesis and Sedimentation Assay
The full-length native human CDH26 protein, herein indicated as rCDH26 (amino acid start MAMRS, end SGVPS) was generated with a C-terminal dual StrepII tag (GSSAWSHPQFEKGGGSGGGSGGSAWSHPQFEK) by cloning into the mammalian expression vector pAlphaH (Supplemental Text S2) under the control of a CAG promotor. The protein was expressed using the mammalian expression system using EXPI293 cells (ThermoFisher) according to the manufacturers supplied protocols via transient transfection and the cell pellet collected by centrifugation 48 hours post transfection. Cells were lysed by sonication and protein purified using manufacturer recommended protocols with a 1 mL of Streptactin Sepharose High Performance StrepTrap HP column (IBA Lifesciences). The sample was passed over an analytical superdex S200 10/300mm gel filtration column (GE Healthcare). The fractions containing rCDH26 were pooled elutions from this capture step and protein concentration determine ABS 280nm using calculated extinction coefficient calculated from the amino acid sequence. Final protein preparations were further purified by analytical size exclusion columns to generate a pool corresponding to the full-length protein of ~92kDa (see Supplemental Figure S3).
For samples in which protein sedimentation was assessed in calcium-free solutions, 15 µg of rCDH26 was concentrated in 3K Amicon Ultra filter units and buffer exchanged into calcium-magnesium free dPBS. Samples were subjected to incubation at room temperature for 1 h in the presence or absence of calcium to facilitate homotypic binding. Samples were then spun at 5,000g (slow), 20,000 g (fast) or 100,000g (UCultracentrifuge) at 22 °C for 20 mins. Supernatant liquid was removed and pellet left in the tube. Laemmli sample buffer was added to each, in-gel staining was done with Flamingo Total Protein Stain (Biorad) and western blots probed for 1:3000 anti-StrepII-HRP to detect StrepII-tagged rCDH26 in each fraction.

HeLa Adhesion and Migration Assays
For adhesion assays, Empty-GFP or CDH26-GFP cells were serum starved overnight.
In 24 well plates, rat tail collagen was gelled in 10X HBSS for 60 minutes at 37 °C.
Plates were blocked with 2% BSA for 30 minutes at 37 °C. Cells were seed at 100,000 cells per well in serum-free media and allowed to attach to the matrix for 1 hour at 37 °C. Unattached cells were removed by washing the surface of plate 3 times with PBS.
Cells were visualized with crystal violet staining and counted. Migration performed in transwell 3422 rat tail collagen coated plates. Plates were washed and blocked with 0.1% BSA for 1 hour at 37 °C. Cells were serum starved overnight, trypsinized and resuspended in DMEM with 0.1% BSA at a cell density of 1 x 10 6 cells/mL. Cells were plated at a density of 100,000/well in the upper chamber and allowed to migrate to serum media with 0.5% FCS in DMEM for 6 hours at 37 °C. The GFP-positive cells that migrated through the filter pores were quantified using green fluorescence filter on the ImageQuant LAS4000.

Vector Cloning and Expression Systems
Over-expression in cell lines was performed using the Neon electroporation system (ThermoFisher). CDH26 (Myc-DDK-tagged)-Human cadherin 26 (CDH26), transcript variant A vector was purchased from OriGene. The insert for CDH26 was cut out using AsiSI (SgfI) and MluI (New England Biolabs) and cloned into entry vector pCMV6-AC-mGFP vector to generate a c-terminal fused mGFP-CDH26 protein. For the StrepIItagged CDH26, CDH26 was synthesized from sequence optimized data and was cloned into vector backbone pAlphaH (formally known as pHLSec2) [2], by Genewiz using cloning sites KpnI and EcoRI (New England Biolabs). Using the pLL3.7 backbone [3,4], LifAct-mCherry was cloned in place of GFP using NheI and EcoRI (gift from Mallar Bhattacharya, UCSF). Vector maps and sequence information in S3.

Knockdown of CDH26 in Airway Epithelial Cells
To silence CDH26, Origene, hairpin sequences are detailed in S5, synthesized four unique 29mer shRNA constructs in a retroviral RFP vector and one custom designed 29mer shRNA construct in a retroviral GFP vector. A non-effective 29-mer scrambled shRNA cassette in pRFP-C-RS or pGFP-V-RS vector was used as a negative control.
Cells were transfected using the Neon electroporation system (Invitrogen). Briefly, primary airway epithelial cells were expanded in flasks and lifted with TrypLE (Gibco) at 80-90% confluence. Cells were resuspended in Buffer R at a density of 3.0 x 10 6 /mL. Cells were pulsed using 6 µg shRNA or scramble mix (see Supplemental Figure S4) using the Neon 100 µL tip at a single pulse of 1200 V width 40 ms. Cells were plated at a density of 600,000 cells/well on transwell 3460 submerged for 1-3 days in media supplemented with ROCK to allow cells to recover. After the submersion phase, cells were switched ALI and maintained in 50/550 DMEM/F12 supplemented with 2% Ultroser G [5]. Transepithelial resistance and voltage was measured using a chopstick voltmeter.

Soft Agar Assay for Colony Formation and Cell Proliferation
To determine if knockdown of CDH26 resulted in anchorage-independent proliferation or transformation, soft agar assays were performed as previously described in 96-well and 35 mm dish format as previously described [6]. Briefly, human AECs were transiently transfected with shRNA pool and plated at a density of 4000 cells/well in 96 well black plates with agar gel made in the base. Cells were overlaid with a feeder 0.5% agar layer. Cells were allowed to recover in plates for 96 hours and the number of viable cells was quantified on a plate reader by a Resazurin reduction assay per the manufacturer's instructions (ThermoFisher). To assay colony formation, cells were plated at a density of 500,000 cells per 35mm glass dish with a gel base and feeder overlay. Feeder layers were changed in dishes once a week. Cells were fixed, stained with crystal violet dishes imaged after three weeks to count colonies. Colony counting was performed using ImageJ analyze particles plugin.  A.

Supplemental Figures
C. B.