Submersion and hypoxia inhibit ciliated cell differentiation in a notch-dependent manner

Abstract

The epithelium that lines the conducting airways is composed of several distinct cell types that differentiate from common progenitor cells. The signals that control fate selection and differentiation of ciliated cells, a major component of the epithelium, are not completely understood. Ciliated cell differentiation can be accomplished in vitro when primary normal human bronchial epithelial (NHBE) cells are cultured at an air-liquid interface, but is inhibited when NHBE cells are cultured under submerged conditions. The mechanism by which submersion prevents ciliogenesis is not understood, but may provide clues to in vivo regulation of ciliated cell differentiation. We hypothesized that submersion creates a hypoxic environment that prevents ciliated cell differentiation by blocking the gene expression program required for ciliogenesis. This was confirmed by showing that expression of multicilin and Forkhead box J1, key factors needed for ciliated cell differentiation, was inhibited when NHBE cells were cultured in submerged and hypoxic conditions. Multicilin and Forkhead box J1 expression and ciliated cell differentiation were restored in submerged and hypoxic cells upon treatment with the γ-secretase inhibitor, N-[(3,5-difluorophenyl)acetyl]-L-alanyl-2-phenyl]glycine-1,1-dimethylethyl ester (DAPT), which suggested that Notch signaling was involved. Overexpression of Notch intracellular domain inhibited differentiation in the presence of DAPT, confirming the role of Notch signaling. These results indicate that submersion and hypoxia prevent ciliated cell differentiation by maintaining Notch signaling, which represses genes necessary for ciliogenesis. These data provide new insights into the molecular mechanisms that control human bronchial differentiation.

Keywords: Forkhead box J1; Notch; hypoxia; multicilin; multiciliogenesis.

Figure 1.

Forkhead box J1 (FOXJ1) expression and ciliated cell differentiation are reduced by submersion in an apical volume–dependent manner. (A–H) Representative extended focus confocal immunofluorescent images of NHBE cells cultured for 21 days with apical media volumes of 0 ml (A and E), 0.125 ml (B and F), 0.25 ml (C and G), and 0.5 ml (D and H) stained for nuclei (Hoechst, blue) and cilia (acetylated α-tubulin, white; A–D) or FOXJ1 (green; E–H). (I) Percent FOXJ1-positive (FOXJ1⁺; white bars) and ciliated (black bars) cells at the apical media volumes, indicated on the x axis, showing a significant increase at 0.125 ml and a significant decrease at 0.5 ml. (J) FOXJ1 messenger RNA (mRNA) levels after 21 days of differentiation at the apical media volumes, indicated on the x axis, normalized to glyceraldehyde phosphate dehydrogenase (GAPDH) mRNA, showing that the amount of FOXJ1 mRNA increases at 0.125 ml and decreases significantly at higher apical volumes, which is similar to changes in FOXJ1⁺ cells. (K) Quantification of nuclear density after 21 days submerged with different volumes of media showing a significant reduction in cells in 0.5 ml. Scale bar, 50 μm. Data shown are means (± SEM). One-way ANOVA (*P < 0.05; n = 3).

Figure 2.

Submersion and hypoxia induce hypoxia-inducible factor (HIF) 1α and HIF-2α proteins. NHBE cells cultured 72 hours at an air–liquid interface (ALI) in air (21% O₂) then switched to an ALI in hypoxia (0.5%; Hyp.) or submerged (0.5 ml; Sub.) for 5 hours. (A) Western blots of whole-cell protein lysates with antibodies to HIF-1α, HIF-2α, and β-actin showing both HIF-1α and HIF-2α induction in submerged and hypoxic conditions compared with ALI (Air) control. (B) Densitometry quantification of the Western blot data from two lungs. The data show increases in HIF proteins, indicating that submersion causes a hypoxic environment. β-actin was used as a loading control. Data shown are means (± SEM).

Figure 3.

γ-secretase inhibition restores FOXJ1 expression and ciliated cell differentiation in submerged and hypoxic conditions. (A–J) Representative extended-focus confocal immunofluorescent images of NHBE cells after 21 days of culture (A and B) submerged (0.5 ml), (C and D) submerged (0.5 ml) with 10 μM N-[(3,5-difluorophenyl)acetyl]-L-alanyl-2-phenyl]glycine-1,1-dimethylethyl ester (DAPT), (E and F) ALI in hypoxia (0.5% O₂), (G and H) ALI in hypoxia (0.5% O₂) with 10 μM DAPT, or (I and J) ALI in air, and stained for nuclei (Hoechst, blue) and cilia (acetylated α-tubulin, white) or FOXJ1 (green). Scale bars, 50 μm. (K) Graph displaying quantification of the percent FOXJ1⁺ in each condition showing a significant decrease in submerged and hypoxic conditions compared with the air control, and showing that DAPT restores ciliogenesis to control ALI levels. One-way ANOVA (*P < 0.05; n = 3). (L) Western blots of whole-cell lysates (40 μg) from NHBE cells cultured 24 hours at an ALI in air, submerged (0.5 ml) with and without 10 μM DAPT (top panel), or 48 hours at an ALI in air, and ALI in hypoxia (0.5% O₂) with and without 10 μM DAPT (bottom panel) probed for activated-Notch1 (Notch intracellular domain [NICD] 1). The data show that NICD1 levels do not increase over ALI (Air), and that DAPT decreases NICD1 protein levels in submersion and hypoxia. β-actin was used as the loading control. (M) Quantitative RT-PCR (qRT-PCR) Notch target gene, hairy/enhancer-of-split related with YRPW motif 1 (HEY1), mRNA from NHBE cells cultured for 21 days in submerged and ALI in hypoxia (0.5% O₂) (Hyp.) with and without 10 μM DAPT. HEY1 mRNA was normalized to the housekeeping gene, β2-microglobin (B2M), and shows that DAPT inhibits Notch signaling. Data shown are means (± SEM). Student’s t test (*P < 0.05; n = 3).

Figure 4.

Multicilin (MCI) expression is inhibited by submersion in a γ-secretase–dependent manner. Expression of (A) MCI and (B) FOXJ1 mRNAs during NHBE differentiation at an ALI in air (21% O₂; orange), submerged (0.5 ml; blue), or submerged (0.5 ml) with 10 μM DAPT (green). MCI and FOXJ1 mRNAs are normalized to the housekeeping gene, B2M. The graphs show that expression of MCI and FOXJ1 mRNAs is inhibited by submersion, but that inhibition is overcome by adding DAPT (n = 3). Data shown are means (± SEM).

Figure 5.

Expression of influenza hemagglutinin epitope–tagged (HA)-NICD1 increases Notch signaling and inhibits FOXJ1 in DAPT-treated submerged and hypoxic cells. (A) qRT-PCR of HEY1 mRNA from NHBE cells transduced with lentiviruses expressing either HA-tagged NICD1 (HA-NICD1) or vector control (HA-vector), showing a significant increase in the Notch target gene, HEY1 mRNA, in HA-NICD1 transduced cells, indicating that HA-NICD1 induces Notch signaling. HEY1 mRNA was normalized to the housekeeping gene, B2M mRNA. Student’s t test (*P < 0.05; n = 3). (B–M) Representative extended-focus confocal immunofluorescent images of NHBE cells transduced with HA-NICD1 or HA-vector lentiviruses. Cells were grown for 3 weeks in submerged conditions with 10 μM DAPT or in hypoxia (0.5% O₂) with 10 μM DAPT, and then stained for nuclei (Hoechst, blue), FOXJ1 (green), or HA (red). (B–E) Hoechst and FOXJ1 merged images, (F–I) images of HA alone, and (J–M) HA and FOXJ1 merged images. The images show cytoplasmic HA staining colocalizing with FOXJ1 in HA-vector transduced cells, whereas HA-NICD1 transduced cells have nuclear HA staining and significantly less colocalization with FOXJ1 staining. Quantification of percent FOXJ1⁺ HA-NICD1 or HA-vector transduced (HA⁺) and nontransduced (HA⁻) NHBE cells in submerged plus DAPT (N) and hypoxia plus DAPT (O) conditions showing significant reduction of FOXJ1⁺ cells in HA-NICD1 transduced cells compared with nontransduced cells, and to HA-vector control transduced cells in DAPT-treated submerged and hypoxic conditions. These results indicate that expression of HA-NICD1 inhibits ciliated cell differentiation in the presence of DAPT, and suggest that DAPT promotes ciliogenesis by blocking Notch signaling. A minimum of 500 cells from 3 different lung donors was counted for each group. Scale bar, 50 μm. Data shown are means (± SEM). One-way ANOVA (*P < 0.05; n = 3).

Figure 6.

DAPT enables FOXJ1 expression and ciliated cell differentiation of NHBE cells in submerged culture on plastic in 96-well plates. Representative three-dimensional (3D) opacity renderings of 40× confocal immunofluorescent images of NHBE cells cultured in wells of a 96-well plate submerged for 21 days (top) or submerged with 10 μM DAPT (middle); a Z-stack of submerged conditions with 10 μM DAPT is also shown (bottom). Cells were stained for cilia (acetylated α-tubulin, white), FOXJ1 (green), and nuclei (Hoechst, blue). The percent FOXJ1⁺ cells are indicated in the upper right corner of the upper two panels. The 3D images were cropped and rotated along the z axis to view cells from above the apical surface. Scale bar, 20 μm (n = 3).

Figure 7.

The proposed model for the regulation of ciliated cell differentiation by oxygen and Notch signaling. Submersion and/or hypoxia cause a low-oxygen environment that potentiates Notch signaling and prevents MCI and FOXJ1 expression, factors required for ciliated cell differentiation. Possible mechanisms for hypoxia potentiation of Notch signaling are indicated by the dashed arrow, and include induction of ligand expression, induction of Notch receptor expression, and direct interaction with NICD. Inhibition of γ-secretase with DAPT prevents Notch cleavage into the activated NICD form, and inhibits Notch signaling, which permits MCI and FOXJ1 expression and ciliated cell differentiation.