The role of transcriptional factor p63 in regulation of epithelial barrier and ciliogenesis of human nasal epithelial cells

Disruption of nasal epithelial tight junctions (TJs) and ciliary dysfunction are found in patients with chronic rhinosinusitis (CRS) and nasal polyps (NPs), along with an increase of p63-positive basal cells and histone deacetylase (HDAC) activity. To investigate these mechanisms, primary cultures of HNECs transfected with human telomerase reverse transcriptase (hTERT-HNECs) were transfected with siRNAs of TAp63 and ΔNp63, treated with the NF-kB inhibitor curucumin and inhibitors of HDACs, and infected with respiratory syncytial virus (RSV). In TERT-HNECs, knockdown of p63 by siRNAs of TAp63 and ΔNp63, induced claudin-1 and -4 with Sp1 activity and enhanced barrier and fence functions. The knockdown of p63 enhanced the number of microvilli with the presence of cilia-like structures. Treatment with curcumin and inhibitors of HDACs, or infection with RSV prevented expression of p63 with an increase of claudin-4 and the number of microvilli. The knockdown or downregulation of p63 inhibited phospho-p38MAPK, and the p38MAPK inhibitor downregulated p63 and upregulated the barrier function. Thus, epithelial barrier and ciliogenesis of nasal epithelium are regulated in a p63-negative manner in normal and upper airway diseases. Understanding of the regulation of p63/p38 MAPK/NF-κB may be important in the therapy for airway allergy and its drug delivery system.

A defective epithelial barrier with decreased expression of TJ proteins is found in patients with chronic rhinosinusitis (CRS) and nasal polyps (NPs) 19 . The nasal epithelial CLDN-4 is markedly upregulated by TGF-β, which is closely related to NPs, CRS and human respiratory syncytial virus (RSV)-infection 6,20,21 .
Transcriptional factor p63, which is a member of the p53 family and has two distinct isoforms, TAp63 and ΔNp63, plays an important role in the proliferation and differentiation of various epithelial basal cells 22 . It is known that p63 is upstream of IKKα in epidermal development 22 . p63 is also one of the regulators of various cellmatrix and cell-cell adhesion complexes in the epidermis 23 . It contributes to the formation and maintenance of differentiated pseudostratified bronchial epithelium 24 . ΔNp63 plays a critical role in epithelial stratification and in skin stem cell renewal 25 . Loss of ΔNp63 significantly reduces epithelial proliferation and increases E-cadherin expression in human airway epithelial cells 26 .
An increase in p63-positive cells is observed in the epithelium of NPs and the expression of p63 in multiple cell layers is an important pathologic phenomenon in the epithelial remodeling seen in NPs 27,28 . RSV infects p63 + airway basal cells in air-liquid interface cultures of human bronchial epithelial cells and influences p63 expression and differentiation 29 .
On the other hand, the ciliary dysfunction also occurs in chronic rhinosinsusitis 30 . Transcriptional factor Myb (+) cells are increased in chronic airways disease 31 . A p63(−) Myb(+) population arising from self-renewing p63 (+) Krt5 (+) epithelial progenitors in airway epithelial cells represents a distinct intermediate stage of differentiation towards ciliated cells under the influence of specific regulatory factors, including Notch and FOXJ1 31 . It is possible that the epithelial barrier created by TJ proteins and cilia formation in the upper airway are regulated via p63.
Expression of HDAC1 and HDAC9 is high in bronchial epithelial cells (HBECs) from asthmatic patients and the inhibition of HDAC activity reconstitutes a defective barrier by increasing TJ expression 34 . HDAC is an important epigenetic regulator in RSV-induced lung inflammation and treatment with inhibitors of HDAC inhibits RSV replication and decreases RSV-induced airway inflammation and oxidative stress 35 . HDAC inhibitors induce cell death, the cell cycle, senescence, differentiation, autophagy and tumor immunogenicity 36 . The HDAC inhibitor sodium butyrate significantly upregulates the protein levels of cingulin, ZO-1, and ZO-2 in Rat-1 fibroblasts, cingulin in COS-7 cells, and cingulin and occludin in HeLa cells 37 .
In the present study, p63, ΔNp63, HDAC1 and HDAC6 were upregulated and CLDN-1 and -4 were downregulated in the epithelium of sinusitis and NPs. In HNECs, knockdown of p63 and ΔNp63 induced expression of CLDN-1 and -4, enhanced barrier and fence functions, and increased the number of microvilli on the cell surface. Inhibitors of NF-κB, HDACs and p38 MPAK, and RSV infection, prevented p63 expression and induced TJ proteins, p63-negative regulation of the epithelial barrier and ciliogenesis of the nasal epithelium.

Methods
Ethics statement. The protocol for human study was reviewed and approved by the ethics committee of the Sapporo Medical University School of Medicine. Written informed consent was obtained from each patient who participated in the investigation. All experiments were carried out in accordance with the approved guidelines and with the Declaration of Helsinki.

Antibodies and reagents.
A mouse monoclonal anti-p63 (DAK-p63) antibody was obtained from Dako (Tokyo, Japan). A rabbit polyclonal anti-p63 antibody was from Abcam (Cambridge, MA, USA). A rabbit polyclonal anti-p40 (ΔNp63) antibody was from NICHIREI BIOSCIENCES INC. (Tokyo, Japan). A rabbit polyclonal anti-ΔNp63 antibody was from BioLegend (Tokyo, Japan). Mouse monoclonal anti-cytokeratin5 (CK5) and anti-cytokeratin-7 (CK7) antibodies were from Sigma Aldrich. Rabbit polyclonal anti-CLDN-1, -4, and -7, anti-occludin, anti-tricellulin and mouse monoclonal anti-CLDN-4 (3E2C1) antibodies were from Zymed Laboratories (San Francisco, CA). A rabbit polyclonal anti-LSR antibody was from Novus Biologicals (Littleton, CO, USA). Mouse monoclonal anti-HDAC1 (10E2) and rabbit polyclonal anti-HDAC6, anti-phospho-NFκB, anti-NFκB, anti-phospho-p38 MAPK and anti-p38 MAPK antibodies were from Cell Signaling Technology and microarray images were automatically analyzed using AROS TM , version 4.0 (Operon Biotechnologies, Tokyo, Japan). Immunohistochemical analysis. Human nasal tissues were obtained from each 10 patients with hypertrophic rhinitis or chronic sinusitis who underwent inferior turbinectomy at Sapporo Medical University, the Sapporo Hospital of Hokkaido Railway Company, or the KKR Sapporo Medical Center Tonan Hospital. Informed consent was obtained from all patients and this study was approved by the ethics committees of the above institutions.
The tissues were embedded in paraffin after fixation with 10% formalin in PBS. Briefly, 5-μm-thick sections were dewaxed in xylene, rehydrated in ethanol, and heated with Vision BioSystems Bond Max using ER2 solution (Leica) in an autoclave for antigen retrieval. Endogenous peroxidase was blocked by incubation with 3% hydrogen peroxide in methanol for 10 min. The tissue sections were then washed twice with Tris-buffered saline (TBS) and preblocked with Block Ace for 1 h. After washing with TBS, the sections were incubated with anti-p63, anti-ΔNp63, anti-claudin-1, anti-claudin-4, anti-claudin-7, anti-HDAC1 and anti-HDAC6 antibodies (1:400) for 1 h. The sections were then washed three times in TBS and incubated with Vision BioSystems Bond Polymer Refine Detection kit DS9800. After three washes in TBS, a diamino-benzidine tetrahydrochloride working solution was applied. Finally, the sections were counterstained with hematoxylin.
Cell culture and treatments. The cultured HNECs were derived from mucosal tissues of patients with hypertrophic rhinitis or chronic sinusitis who underwent inferior turbinectomy at Sapporo Medical University, the Sapporo Hospital of Hokkaido Railway Company, or the KKR Sapporo Medical Center Tonan Hospital. Informed consent was obtained from all patients and this study was approved by the ethics committees of the above institutions.

RSV experiment. RSV was grown in the human laryngeal carcinoma cell line HEp-2. For infection, HNECs
at 80% confluence were adsorbed at an RSV multiplicity of infection (MOI) of 1 for 60 min at 37 °C. After adsorption, the viral solutions were removed and the cells were rinsed twice with growth medium and incubated. The virus titers in the supernatant were determined by a plaque-forming assay with HEp-2 cells. Expression of RSV mRNA was confirmed by reverse transcription-PCR (RT-PCR). Some cells cultured without FBS were infected with RSV for 24 h 21 .
Luciferase reporter assay. Cells were seeded on 12-well plates in triplicate and allowed to grow overnight to reach 50-70% confluence. The cells were cotransfected with SP1 reporter and siRNAs of p63 and ΔNp63. After 48 h of transfection, Luciferase activity was measured using the Dual Luciferase Reporter Assay System kit (Promega) in a TECAN microplate reader (Infinite M1000 Pro, Tecan Japan Co., Ltd., Kawasaki, Japan). Luciferase activity was normalized to R. reniformis luciferase activity and plotted as mean ± SD from three independent experiments.
Scanning electron microscopy (SEM). Cells grown on coated coverslips were fixed with 2.5% glutaraldehyde/0.1 M PBS (pH 7.3) overnight at 4 °C. After several rinses with PBS, the cells were postfixed in 1% osmium tetroxide at 4 °C for 3 h and then rinsed with distilled water, dehydrated in a graded ethanol series, and freeze-dried. The specimens were sputter-coated with platinum and observed with a scanning electron microscope (S-4300, Hitachi; Tokyo, Japan) operating at 10 kV. Differences between groups were tested by ANOVA followed by a posthoc test and an unpaired two-tailed Student's t test and considered to be significant when p < 0.05. ΔNp63 and CLDN-1, -4, and -7 in the nasal epithelium of sinusitis and polyps. We performed immunohistochemical analysis for p63, ΔNp63 and CLDN-1, -4, -7 in normal, sinusitis and polyp (NP) tissues. In the normal nasal epithelium, p63 and ΔNp63 were positive in the nuclei of the basal cell type, and in the epithelium of sinusitis and NPs, p63 andΔNp63 were upregulated (Fig. 1). CLDN-1 and CLDN-4 were downregulated in the sinusitis and NPs, whereas no change of CLDN-7 was observed in sinusitis or NPs (Fig. 1).

Upregulation of tight junction proteins by siRNAs of TAp63 and ΔNp63 in HNECs.
To investigate the mechanisms involved in the regulation of TJs via p63 in HNECs, hTERT-HNECs cultured without FBS, were transfected with siRNAs of TAp63 and ΔNp63. hTERT-HNECs cultured without FBS were detected CK7, CK5, p63 and ΔNp63 (Fig. 2a). In Western blotting, knockdown of p63 by siRNAs of TAp63 and ΔNp63 induced CLDN-1, -4, TRIC and LSR (Fig. 2b). The immunocytochemical results showed that OCLN and CLDN-4 presented at the membranes of the cells, which reduced p63 expression by siRNAs of TAp63 and ΔNp63, whereas OCLN and CLDN-4 were not detected in control cells without FBS (Fig. 2c). Furthermore, tricellular tight junction proteins TRIC and LSR were presented at the membranes by siRNAs of TAp63 and ΔNp63, whereas they were not detected in control cells without FBS (Supplemental Fig. 1).

Upregulation of Sp1 by siRNAs of TAp63 and ΔNp63 in HNECs.
To investigate whether Sp1 activity contributed to induction of CLDN-1 and -4 by siRNAs of TAp63 and ΔNp63 in HNECs, we performed luciferase reporter assay for Sp1. Knockdown of p63 by siRNAs of TAp63 and ΔNp63 induced Sp1-luciferase activity (Fig. 3a).

Upregulation of epithelial barrier and fence functions by siRNAs of TAp63 and ΔNp63 in HNECs.
To investigate the epithelial barrier and fence functions affected via p63 in HNECs, we measured the TEER values of hTERT-HNECs transfected with siRNAs of TAp63 and ΔNp63 to determine the barrier function and examined diffusion of BODIPY-sphingomyelin for the fence function. Knockdown of p63 by siRNAs of TAp63 and ΔNp63 induced the epithelial barrier and maintained the fence function, whereas the TEER value was low and the fence function was absent in control cells cultured without FBS (Fig. 3b,c).

Increase of surface microvilli and presence of cilia-like structures induced by siRNAs of TAp63
and ΔNp63 in HNECs. To investigate the ciliogenesis mediated via p63 in HNECs, hTERT-HNECs transfected with siRNAs of TAp63 and ΔNp63 were examined by immunocytochemistry for Ac-tubulin and by SEM of the surface. Knockdown of p63 by siRNAs of TAp63 and ΔNp63 induced Ac-tubulin expression and enhanced the number of microvilli on the cell surface together with the presence of cilia-like structures. (Fig. 3e).

Upregulation of tight junction proteins via p63 by NF-κB inhibitor curcumin in HNECs. p65/
NF-κB regulates p63 expression 39 . In Western blotting, treatment with the NF-κB inhibitor curcumin prevented p63 expression and induced expression of CLDN-1, -4 and OCLN in hTERT-HNECs (Fig. 4a, Supplemental  Fig. 3A). The immunocytochemical results showed that OCLN and CLDN-4 presented at the membranes of the cells that had reduced p63 due to the treatment with curcumin, whereas OCLN and CLDN-4 were not detected in control cells without FBS (Fig. 4b).

Upregulation of tight junction proteins via p63 by RSV infection in HNECs.
We previously reported that OCLN and CLDN-4 were upregulated via NF-κB by infection with RSV in HNECs 21 . We therefore investigated whether RSV-infection upregulated OCLN and CLDN-4 via p63 in hTERT-HNECs. In Western blotting, RSV infection at MOI 1 for 24 h downregulated p63 and upregulated OCLN and CLDN-4 (Fig. 4c,  Supplemental Fig. 3B). The immunocytochemical results showed that OCLN and CLDN-4 presented at the membranes of the cells that were G-protein positive and had reduced p63 due to RSV infection (Fig. 4d).

Expression patterns of HDAC1 and HDAC6 in the nasal epithelium of sinusitis and polyps.
We performed immunohistochemical analysis for HDAC1 and HDAC6 in normal, sinusitis and NP tissues. The immunohistochemical results showed that HDAC1 was upregulated in the nasal epithelium of the sinusitis and NPs, and HDAC6 was upregulated in the sinusitis, whereas HDAC1 and HDAC6 were positive in the normal nuclei (Fig. 5a). and Ac-tubulin in Westen blotting and enhanced the epithelial barrier function in a dose-dependent manner (Fig. 5b,c, Supplemental Fig. 4). The immunocytochemical results showed that OCLN and CLDN-4 presented at the membranes of the cells in which p63 was reduced by treatment with 10 μM TSA (Fig. 6a). Treatment with TSA was found to induce Ac-tubulin together with reduction of p63 by immunocytochemistry and Western blotting

Upregulation of tight junction proteins, barrier function and surface microvilli by the inhibitors of HDAC1 and HDAC6 in HNECs.
To investigate which specific HDAC inhibitors affected tight junctions and microvilli, hTERT-HNECs cultured without FBS were treated with inhibitors of HDAC1 and HDAC6 at 1 and 10 μM for 24 h. In Western blotting, treatment with inhibitors of HDAC1 and HDAC6 at 10 μM reduced expression of p63 and enhanced that of CLDN-4 and Ac-tubulin (Fig. 7b, Supplemental Fig. 6). Immunocytochemical results showed that OCLN and CLDN-4 presented at the membranes of the cells in which p63 was reduced by treatment with inhibitors of HDAC1 and HDAC6 at 10 μM (Fig. 7c). In SEM analysis of the cells treated with inhibitors of HDAC1 and HDAC6 at 10 μM, the number of surface microvilli was increased compared to the control (Fig. 7c).
Downregulation of phospho-p38 MAPK by TAp63-siRNA, curcumin, TSA and HDAC1 inhibitor in HNECs. The epithelial barrier in HNECs is regulated via a distinct signal transduction pathway including p38 MAPK stress signaling 7,8 . We investigated whether p38 MPAK signaling closely contributed to downregulation of p63 by treatment with siRNAs of TAp63 and ΔNp63, curcumin, TSA and an HDAC1 inhibitor in HNECs. When hTERT-HNECs without FBS were treated with TAp63-siRNA, curcumin, TSA and the HDAC1 inhibitor, phospho-p38 MAPK was douwnregulated by Western blotting (Fig. 8a, Supplemental Fig. 7A). Upregulation of tight junction proteins and barrier function via p63 by the p38 MAPK inhibitor SB230580 in HNECs. It is also known that p63 is regulated via p38 MPAK 40 . When hTERT-HNECs cultured without FBS were treated with the p38 MAPK inhibitor SB230580, it downregulated p63 and upregulated CLDN-4 in Western blotting (Fig. 8b, Supplemental Fig. 7B). Immunocytochemistry revealed that treatment with SB230580 led to the presentation of OCLN and CLDN-4 in the membranes and significantly induced TEER, i.e. barrier function (Fig. 8c,d).

Discussion
In this study, we first found that downregulation of p63 regulated the epithelial barrier and ciliogenesis of the nasal epithelium in normal and diseased tissues. Inhibitors of HDACs, which were highly expressed in CRS and NPs, could induce the epithelial barrier and ciliogenesis via p63.
p63 regulates various cell-matrix and cell-cell adhesion complexes in the epidermis 23 . It contributes to the formation and maintenance of differentiated pseudostratified bronchial epithelium 24 and regulates the target genes by direct interaction with Sp1 41 . CLDN-1 is a p63 direct target gene in epithelial development and p63 deficiency leads to inhibition of CLDN-1 in p63-null mouse keratinocytes 42 . CLDN-1 and CLDN-4 are in part controlled by the Sp1-containing critical promoter region 16 . In HNECs in the present study, knockdown of p63 by siRNAs of TAp63 and ΔNp63 induced expression of CLDN-1 and -4 with an increase of Sp1 activity and enhanced the barrier and fence functions. Furthermore, expression of the tricellular tight junction proteins TRIC and LSR was increased by the knockdown of p63. These results indicated that p63 negatively regulated the nasal epithelial TJ proteins and their functions in HNECs. In fact, in the nasal epithelium of CRS and NPs with an increase of p63-positive basal cells, disruption of epithelial TJs was observed.
TAp63 is a transcriptional target of NF-κB, which may play a role in cell proliferation, differentiation and survival upon NF-κB activation by various stimuli 43 . The regulatory feedback loop between TAp63 and NF-κB is involved in the activation of the cell-death process of cancer cells 44 . In HNECs, the NF-κB inhibitor curcumin inhibits NF-κB activity and upregulates CLDN-4 and OCLN 45 . Furthermore, RSV-infection also inhibits NF-κB activity and upregulates CLDN-4 and OCLN in airway epithelial cells 21,29 . In the present study, treatment with curcumin and infection with RSV, downregulated p63 and upregulated CLDN-1 and CLDN-4, and CLDN-4 and OCLN, respectively, in HNECs. These findings suggested that the inhibition of NF-κB downregulated p63 expression and enhanced TJ proteins.
TJ proteins are upregulated in a kinase-dependent manner during cell differentiation induced by HDAC inhibitors 37 . The HDAC inhibitor TSA contributes to the activation of transcriptional factors p63 and Sp1 46 . IL-4-induced rat nasal epithelial barrier dysfunction can be blocked by TSA 47 . In the present study, HDAC1 was upregulated in the nasal epithelium of NPs, and HDAC6 was upregulated in sinusitis. In HNECs, treatment with TSA or specific inhibitors of HDAC1 and HDAC6 downregulated p63 expression and upregulated expression of some tight junction proteins, the barrier function and the numbers of surface microvilli. These results indicated that the HDAC inhibitors could protect against inhaled substances and pathogens through HNECs as a result of the barrier enhanced via p63.
The p38 MAPK/NF-κB signaling is involved in the epithelial barrier by means of various stimuli 48,49 . The nasal epithelial barrier is regulated via a distinct signaling pathway including p38 MAPK 3 . p63 is in part regulated via p38 MPAK 40 . The NF-κB inhibitor curucumin and the HDAC inhibitor TSA affect the epithelial barrier via p38 MAPK/NF-κB 46,50 . In the present study, the knockdown or downregulation of p63 by treatment with siRNAs of TAp63 and ΔNp63, curcumin, TSA and an HDAC1 inhibitor in HNECs inhibited the activity of phospho-p38MAPK. Conversely, the p38MAPK inhibitor SB23580 downregulated p63 expression and upregulated the epithelial barrier function with an increase of CLDN-4 expression. These results indicated that the bi-regulation between p63 and stress signal p38 MAPK/NF-κB was important in induction and maintenance of the nasal epithelial barrier.
Ciliary dysfunction is in part observed in chronic rhinosinsusitis 51,52 . IL-6/STAT3 promotes the differentiation of ciliated cells from basal stem cells in airway epithelium 53 . Furthermore, in airway epithelial cells, a p63 (−) Myb (+) population derived from self-renewing p63 (+) Krt5 (+) epithelial progenitors, becomes ciliated cells under the influence of specific regulatory factors, including Notch and FOXJ1 31 . In hTERT-HNECs in the present study, the knockdown of p63 by siRNAs of TAp63 and ΔNp63, curcumin and the HDAC inhibitors, induced Ac-tubulin and enhanced the number of microvilli on the cell surface. Cilia-like structures were also observed in the some p63-knockdown cells. IL-6 expression was increased in p63-knockdown cells compared to the control in DNA array analysis (Table 1). All nuclei of control hTERT-HNECs were p63/CK5-positive (Fig. 2a). The p63/CK5-posive cells were markedly decreased by siRNAs of TAp63 and ΔNp63 in immunocytochemistry (Supplemental Fig. 2), whereas no change of CK5 expression in total cells was observed by Western blotting (Fig. 2b). These results suggested that in HNECs, the p63 (−) population might become ciliated cells via promotion by IL-6, although the effects of Notch and FOXJ1 remained unclear in the present study.
In conclusion, downregulation of p63 regulates the epithelial barrier and ciliogenesis of the nasal epithelium. After knockdown of p63 in primary bronchial epithelial cells, they do not proliferate and show marked senescence 24 . RSV infects the p63-positive basal cells of human bronchial epithelium and alters the epithelial differentiation 29 . Thus, it is possible that downregulation of p63 by various stimuli may alter the proliferation and differentiation of nasal epithelial cells.