Cilia

Mucus was harvested from primary human bronchial epithelial (HBE) cell cultures as previously described [18] (link), [19] , [21] . Briefly, excess surgical tissue was procured by the UNC Tissue Core Facility. Normal human bronchial epithelial cells were cultured on a 0.4 mm pore-sized Millicell (Millipore, Billerica, MA) coated with collagen and maintained in air-liquid interface media (UNC Tissue Core) as described in [29] (link). Over a period of 6 weeks, confluent cultures developed cilia, generated and established a periciliary liquid (PCL) layer surrounding the cilia, a mucus layer, and the HBE culture transported mucus. Washings from >100 cultures were pooled and then concentrated against Spectra/Gel to the desired wt%. Concentrated mucus was dialyzed against PBS to insure isotonicity as previously described [18] (link), [21] .

Ventricles were dissected according to Mirzadeh et al., 2010a (link), pre-extracted with 0.1% Triton X in PBS for 1 min, then fixed in 4% paraformaldehyde (PFA) or ice cold methanol for at least 24 h at 4°C, followed by permeabilization in PBST (0.5% Triton X-100) for 20-min room temperature. Tracheas were dissected and cut longitudinally into two, pre-extracted in for 30 s on ice in PEM (0.1 M PIPES (1,4-Piperazinediethanesulfonic acid disodium salt) pH 6.8, 2 mM EGTA (ethylene glycol tetraacetic acid), 1 mM MgSO₄) prior to fixing in ice cold methanol on ice for at least 24 hr. Ventricles and tracheas were blocked in 10% donkey serum in TBST (0.1% Triton X) or 4% bovine serum albumin (BSA) in PBST (0.25% Triton X-100) for 1 hr at room temperature, then placed cilia layer down in primary antibodies (Supplementary file 3) in 4% BSA PBST (0.25% Tween-20) or 1% donkey serum in TBST (0.1% Triton X) for at least 12 hr. Ventricles and tracheas were washed in PBS 3 × 10 min and secondaries (Supplementary file 4) in 4% BSA in PBST (0.25% Triton X-100) or 1% donkey serum in TBST (0.1% Triton X) were added at 4°C for at least 12 hr. Ventricles and tracheas were washed in PBS 3 × 10 min, and ventricles were mounted on glass bottom dishes (Nest, 801002) in Vectashield (VectorLabs), immobilized with a cell strainer (Greiner Bio-One, 542040). Tracheas were mounted on slides with Prolong Gold.

Image analysis was performed in NIS Elements, FIJI (Schindelin et al., 2012 (link)), QuPath (Bankhead et al., 2017 (link)), CellProfiler (Stirling et al., 2021 (link)), or Imaris. All analysis tools have been made available on GitHub (https://github.com/IGC-Advanced-Imaging-Resource/Hall2022_Paper; Murphy, 2022 ). Cerebellum and ventricle area was measured from PAS stained sagittal brain sections in QuPath. The number of cilia in E18.5 ribs was calculated using Batch Pipeline in Imaris, segmenting DAPI and cilia as surfaces. The number of ependymal cells with multiple basal bodies was calculated by segmenting FOP staining and cells in 2D using a CellProfiler pipeline. Briefly, an IdentifyPrimaryObjects module was used to detect the nuclei, followed by an IdentifySecondaryObjects module using the tubulin stain to detect the cell boundaries. Another Identify Primary objects module was used to detect the basal bodies and a RelateObjects module was used to assign parent–child relationships between the cells and basal bodies. The percentage of ciliated ependymal cells, and the number of ependymal cells with rosette-like FOP staining, and elongated FOP-positive structures were counted by eye using NIS Elements Counts Tool. Analysis of cultured ependymal cells (beat frequency, number of cilia, coordinated beat pattern) and beat frequency determination in mTECs and trachea was assessed in FIJI by eye while blinded to genotype. The number of centrioles and cilia in cultured ependymal cells was manually calculated using Imaris. CEP131 and MIB1 intensity at satellites was calculated in FIJI using a macro which segmented basal bodies with Gamma Tubulin, then drew concentric rings, each 0.5 μm wider than the previous and calculated the intensity of MIB1 and CEP131 within these rings. CP110 intensity in MEFs was calculated by manually defining mother and daughter centrioles in FIJI, CP110 intensity in ependyma and tracheas was calculated by segmenting FOP in 3D in Imaris and calculating CP110 intensity within this volume. Image quantification in RPE1 cells were performed using CellProfiler as described previously (Kumar et al., 2021 (link)). Images were prepared for publication using FIJI, Imaris, Adobe Photoshop, Illustrator, and InDesign.

MEFs were maintained as previously published (Hall et al., 2013 (link)). SNAP labeling was performed as previously described (Quidwai et al., 2021 (link)). Ependymal cells were isolated and cultured as published in Delgehyr et al., 2015 (link). mTECs were isolated and cultured as described in Eenjes et al., 2018 (link); You et al., 2002 (link). RPE1-hTERT (female, human epithelial cells immortalized with hTERT, Cat. No. CRL-4000) from ATCC were grown in Dulbecco's Modified Eagle Medium (DMEM, Life Technologies) or DMEM/F12 (Thermo Fisher Scientific, 10565042) supplemented with 10% fetal bovine serum at 37°C with 5% CO₂. For live imaging, the membrane was cut out and placed cilia down on a glass dish (Nest, 801002) in a drop of media. PCM1^−/− RPE1 cells were generated as described previously (Kumar et al., 2021 (link)) (all figures except for Figure 8—figure supplement 1, in which case they were generated as in Gheiratmand et al., 2019 (link)). hTERT-RPE1: Source ATCC, confirmed mycoplasma negative and verified by STR profiling. Two PCM1^−/− RPE1 cell lines were generated using single guide RNAs (Supplementary file 1). Loss of PCM1 was confirmed by genotyping, immunoblotting, and immunofluorescence. Monoclonal PCM1^−/− RPE1 cell lines stably expressing eGFP or eYFP-PCM1 (plasmid a gift from Bryan Dynlacht; Wang et al., 2016 (link)) were generated using lentiviruses and manually selected based on fluorescence. To synchronize cells in G1/S aphidicolin (Sigma) was added to the culture medium at 2 μg/ml for 16 hr. To arrest cells in mitosis, taxol (paclitaxel; Millipore-Sigma) was added to the culture medium at 5 μM for 16 hr prior to rounded up cells being collected by mitotic shake-off. For arrest in G0, cells were washed 2× with phosphate-buffered saline (PBS; Gibco) and 1× with DMEM (without serum) before being cultured in serum-free DMEM for 16 hr. To disrupt cytoplasmic microtubules, cells were treated with 20 μM nocodozole (Sigma, SML1665) for 1–2 hr prior to fixation.