Krt14 and Krt15 differentially regulate regenerative properties and differentiation potential of airway basal cells

Keratin expression dynamically changes in airway basal cells (BCs) after acute and chronic injury, yet the functional consequences of these changes on BC behavior remain unknown. In bronchiolitis obliterans (BO) after lung transplantation, BC clonogenicity declines, which is associated with a switch from keratin15 (Krt15) to keratin14 (Krt14). We investigated these keratins’ roles using Crispr-KO in vitro and in vivo and found that Krt14-KO and Krt15-KO produce contrasting phenotypes in terms of differentiation and clonogenicity. Primary mouse Krt14-KO BCs did not differentiate into club and ciliated cells but had enhanced clonogenicity. By contrast, Krt15-KO did not alter BC differentiation but impaired clonogenicity in vitro and reduced the number of label-retaining BCs in vivo after injury. Krt14, but not Krt15, bound the tumor suppressor stratifin (Sfn). Disruption of Krt14, but not of Krt15, reduced Sfn protein abundance and increased expression of the oncogene dNp63a during BC differentiation, whereas dNp63a levels were reduced in Krt15-KO BCs. Overall, the phenotype of Krt15-KO BCs contrasts with Krt14-KO phenotype and resembles the phenotype in BO with decreased clonogenicity, increased Krt14, and decreased dNp63a expression. This work demonstrates that Krt14 and Krt15 functionally regulate BC behavior, which is relevant in chronic disease states like BO.


Primary cell isolation and culture
Primary Cell Isolation: Primary surface epithelial and submucosal gland cells were isolated from murine tracheas using enzymatic digestion. Tracheas from 3-6 mice were dissected longitudinally and digested in 3 mg/ml Pronase (Roche), which was dissolved in F12 media (Gibco), for 45 min at 37ºC on a rocker. This was done to dissociate surface airway epithelial (SAE) cells. Detached SAE cells were then separated from the remaining tissue by passing through the 100 µm strainer. Cells were then washed twice in excess of complete DMEM (DMEM with 10% FBS, 1% Penicillin/Streptomycin). The remaining tissue containing submucosal gland (SMG) cells was then incubated in 1X Collagenase/Hyaluronidase (STEMCELL Technologies, Inc.) dissolved in F12 for 20 min at 37ºC on a rocker. 1 ml of Trypsin/EDTA (0.025% Trypsin, 0.01% EDTA; Thermo Fisher Scientific) was then added to 3 ml of Collagenase/ Hyaluronidase solution for an additional 5-10 min at 37ºC. After the digestion detached SMG cells were passed through a 100 µm cell strainer and washed in an excess of complete DMEM before plating on 804G-coated plastic plates.

Crispr/Cas9 Knockout
Crispr/Cas9 knockouts were performed in primary SAE cells isolated from H11-Cas9: Rosa-TG mice. Cells were grown to 60-80% confluence in 6-well plates and transfected using 0.3 nMoles of gRNA against the LoxP or Tomato and 0.3 nMoles of gRNA against the gene of interest (See Table 2 for details). Transfection was done using 10µl of Lipofectamine RNAiMax. FACS was performed 2-3 days after transfection with LoxP gRNA and 7-10 days after transfection with Tomato gRNA. DNA was isolated from sorted cells using DNeasy Blood and Tissue Kit (Qiagen) and the gene locus was amplified by PCR. PCR product was either sent for Sanger sequencing with a subsequent peak de-convolution using ICE analysis (Synthego), or was TOPO cloned into a plasmid vector with a subsequent sequencing of plasmid DNA from individual bacterial colonies. ICE analysis and TOPO cloning approaches showed nearly identical results; therefore, ICE analysis was the primary method for determining knockout efficiency.

Immunofluorescence
Freshly dissected murine tracheas were fixed in 4% PFA for at least 2h at room temperature (RT) or for at least 24h at 4ºC. Samples were paraffin embedded by University of Iowa Comparative Pathology Laboratories. 6 µm paraffin sections were 'baked' onto glass slides at 45ºC overnight. After de-paraffinization through xylenes and a stepwise EtOH gradient (100% à 0%), samples were 'boiled' in Sodium Citrate buffer (10mM Na3C6H9O9, 0.05% Tween 20, pH=6.0) using a pressure cooker for 5 min to retrieve antigens. Slides were then incubated in blocking buffer containing 20% normal donkey serum, 0.3% Triton X-100, and 1 mM CaCl2 in PBS for 1h at RT. Primary antibodies (See Table S1 for details) were diluted in the diluent buffer (1% normal donkey serum, 0.3% Triton X-100, and 1 mM CaCl2 in PBS) and incubated on the sample slides overnight at 4°C. Slides were washed in PBS 3 times for 5-10 min and incubated with secondary antibodies and Hoechst 33342 (Invitrogen) in diluent buffer for 2h at room temperature.
EdU staining was done using Click-iT™ Plus EdU Cell Proliferation Kit for Imaging (ThermoFisher Scientific) according to the provided protocol. ProLong Diamond media (Invitrogen) was used to mount the coverslips on the samples.
Cultures at the assay endpoints (both CFE, ALI and monolayer cultures) were fixed in 4% PFA for 20 minutes at RT, and subsequently washed 3 times in PBS for 10 minutes each. Primary and secondary antibody incubations were performed as described above.
Zeiss LSM 880 line-scanning confocal microscope (Carl Zeiss, Germany) or a Leica DM6 B upright microscope (Leica Mircosystems, Inc., United States) were used to acquire fluorescent images.

Image analysis
Composited immunofluorescent tile scans were quantified using the Metamorph Software, Label-retaining basal cell scoring in Figure 7 was done manually in a blinded manner by 2 people, scoring using Metamorph was not feasible due to technical limitations of the software. Epithelial coverage of the explants was scored in ImageJ by dividing the area stained with epithelial markers (Krt5, Krt14, Krt15) over the total explant area. Colony number for CFE assays was scored manually, while staining area in CFE and ALI was quantified automatically after setting an appropriate threshold in ImageJ.

Ex vivo scratch assays and whole mount ferret trachea staining
Tracheas from adult wild-type ferrets were used in ex vivo scratch assays. Freshly for 2h at RT and then mounted between 2 glass slides, edges were sealed with superglue, and clamped with binder clips prior until the glue set. Explants were imaged using Zeiss LSM 880 line-scanning confocal microscope (Carl Zeiss, Germany).

Western Blot analysis
Protein samples were collected from primary SAE cells that have been edited with a specified gRNA, fluorescently sorted, and expanded in SAGM or cultured in Pneumacult ALI for a specified number of days. Cells were lysed either using RIPA buffer (Sigma) for 10 min on ice or using a fractionation kit NE-PER (Thermo Fisher Scientific) following the manufacturer's protocol. Protein concentration was measured using Pierce BCA assay kit (Thermo Fisher Scientific). Samples were run on a SDS PAGE gel under reducing conditions and transferred onto Amersham Protran 0.45 nitrocellulose membrane (GE healthcare), which was subsequently probed using the specified antibodies and imaged using Ai600 imager (GE healthcare). Band intensity was analyzed using ImageJ and normalized to loading control and to the WT control.

Quantitative PCR analysis
RNA was isolated from cells using RNeasy kit (Qiagen) according to the provided protocol. cDNA was synthesized using high-capacity cDNA synthesis kit (Applied Biosystems) according to the manufacturer's instructions. qPCR reactions were set up using 10nM primers (see Table 2 for details), cDNA, water and POWER SYBR master mix (Applied Biosystems) and run on CFX connect Real-Time PCR detection system (Bio-Rad). The expression data was normalized using delta-delta-CT method.

Transwell migration assays
Migration assays were performed using primary mouse SAE or SMG cells grown in SAGM. Cells were seeded at a density of 3x10 5 cell/per well (d=6.5 mm, pore size 8 µm. Corning, processed in parallel to competitive migration assays. Cells were fixed in 4% PFA 12h after seeding and the underside of the membrane was subsequently stained for Krt5 and Krt14. Keratin staining area was quantified using ImageJ software.

Proximity ligation assay
Passage 2 primary Human SAE from 3 donors were seeded on permanox chamber slides (Lab-Tek, Cat# 177445). 12h later, cells were washed and fixed for 10 minutes in 4% PFA and blocked using the blocking buffer provided in the PLA kit (Millipore Sigma, Cat# DUO92202). The assay was carried out according to the provided protocol. Primary antibodies were incubated o/n at 4°C in individual chambers while all the subsequent steps of the protocol were carried out after removing the chamber dividers. We used the following primary antibodies at the specified concentration: Krt14 (RB-9020-P1) 1:500; Sfn (1433s01) 1:50; Krt15 (HPA024554) 1:300. Experimental samples and negative controls (no primary antibody for each one or both target proteins) were run on the same slides. PLA signal was calculated by dividing the total PLA puncta (determined by intensity thresholding and object count module in Fiji) by the number of nuclei in maximum intensity projection images (24-26 z-planes spanning 24-26 µm of depth). At least 100 cells per donor were analyzed in each group.