Mitochondrial uncoupling proteins protect human airway epithelial ciliated cells from oxidative damage

Abstract

Apical cilia on epithelial cells defend the lung by propelling pathogens and particulates out of the respiratory airways. Ciliated cells produce ATP that powers cilia beating by densely grouping mitochondria just beneath the apical membrane. However, this efficient localization comes at a cost because electrons leaked during oxidative phosphorylation react with molecular oxygen to form superoxide, and thus, the cluster of mitochondria creates a hotspot for oxidant production. The relatively high oxygen concentration overlying airway epithelia further intensifies the risk of generating superoxide. Thus, airway ciliated cells face a unique challenge of producing harmful levels of oxidants. However, surprisingly, highly ciliated epithelia produce less reactive oxygen species (ROS) than epithelia with few ciliated cells. Compared to other airway cell types, ciliated cells express high levels of mitochondrial uncoupling proteins, UCP2 and UCP5. These proteins decrease mitochondrial protonmotive force and thereby reduce production of ROS. As a result, lipid peroxidation, a marker of oxidant injury, decreases. However, mitochondrial uncoupling proteins exact a price for decreasing oxidant production; they decrease the fraction of mitochondrial respiration that generates ATP. These findings indicate that ciliated cells sacrifice mitochondrial efficiency in exchange for safety from damaging oxidation. Employing uncoupling proteins to prevent oxidant production, instead of relying solely on antioxidants to decrease postproduction oxidant levels, may offer an advantage for targeting a local area of intense ROS generation.

Keywords: lung; metabolism; motile cilia; oxygen; reactive oxygen species.

Fig. 1.

Mitochondria are located apically in airway ciliated cells and power cilia beating. (A) Transmission electron micrograph of airway epithelia showing ciliated and secretory/goblet cells. The yellow dashed circle outlines mitochondria located directly below cilia. The scale bar indicates 10 µm. The Inset shows apical mitochondria. The scale bar indicates 1 µm. (B) Quantification of the mitochondria area in ciliated and secretory cells. Bars show mean ± SEM. The asterisk indicates P < 0.0001 by Student’s t test. (C) Ciliary beat frequency (CBF) in the presence of the indicated inhibitors of the electron transport chain compared to vehicle control. CBF measurements were performed at 20°C. Each set of data points and lines is from a different human donor. Bars and whiskers indicate mean ± SEM. The asterisk indicates P < 0.01 by paired Student’s t test.

Fig. 2.

Airway epithelia with abundant ciliated cells produce less ROS than epithelia with few ciliated cells. (A) Immunofluorescence images showing differentiated human airway epithelia with few and abundant ciliated cells (grown in USG media vs. Pneumacult ALI media, Method 1). β-tubulin IV immunostaining marks cilia (green). The scale bar indicates 10 µm. (B) Quantification of the percentage of airway surface covered by cilia determined by immunostaining with β-tubulin IV. (C and D) Levels of ROS in epithelia with few vs. abundant ciliation measured using CellROX-Green fluorescence assay (C) and ESR (D).

Fig. 3.

Airway ciliated cells express mitochondrial uncoupling proteins UCP2 and UCP5. (A) mRNA levels of indicated antioxidant enzymes in basal (Bas), secretory (Sec), and ciliated (Cil) cells of large airway tissue. Data are from a public single-cell mRNA database of human large airways (27). (B) Expression of UCP1-UCP5 mRNAs in airway epithelial cells from the same database as in panel A. (C) Ratios of UCP2 and UCP5 mRNA levels to the average of multiple Complex V mRNA levels of the electron transport chain (ETC). Data are from the same database as in panel A. (D) Ratios UCP2 and UCP5 mRNA levels to the average of mRNA levels of multiple Complex V proteins from a public single-cell mRNA database of pig large airways (29). In panels A–D, each data point is from a different human donor or pig. Bars and whiskers indicate mean ± SEM. Asterisks indicate *P < 0.05, **P < 0.01, and ***P < 0.001 by ANOVA.

Fig. 4.

UCP2 and UCP5 expression are directly related to ciliation of human airway epithelia. (A) RT-qPCR data showing relationships of UCP2 and UPC5 mRNA levels to FOXJ1 mRNA levels (indicating ciliated cells) in cultures of human airway epithelia. Epithelial cells were differentiated in USG vs. Pneumacult-ALI media (Method 1). (B) Methods were the same as in panel A except that epithelial cells were differentiated in Pneumacult-ALI media at either 0.5% vs. 18.5% O₂ (Method 2). (C) RT-qPCR data showing relationships of UCP2 and UPC5 mRNA levels to FOXJ1 mRNA. All epithelia were differentiated in USG media at 18.5% O₂, and variations are donor-to-donor differences. For panels A–C, each data point is from epithelia from a different human donor. Lines are linear least squares fit to the data; R² and P values are shown. (D) Immunoblot showing acetylated α-tubulin protein and UCP5 protein in epithelia with few vs. abundant ciliated cells (differentiated in USG vs. Pneumacult-ALI media, Method 1). Complex V of the ETC was used as the protein loading control. (E) Quantification of the UCP5/Complex V ratio from western blots like that shown in panel D. Each data point is from a different human donor. Bars and whiskers indicate mean ± SEM. The asterisk indicates P < 0.05 by paired Student’s t test.

Fig. 5.

UCP2 and UCP5 are located just beneath the cilia in ciliated airway epithelial cells. (A) Immunofluorescence images showing acetylated α-tubulin marking cilia (green), UCP2 (Top) and UPC5 (Bottom) (magenta), and DAPI marking nuclei (blue) in dissociated human airway epithelial cells. Scale bars indicate 10 µm. (B) Immunofluorescence images showing UCP2 (Left) and UPC5 (Right) (magenta) and DAPI marking nuclei (blue) in human large airway tissue. Scale bars indicate 100 µm. Merged panels include transmitted light marking tissue edge and ciliated cells marked with a yellow line. Scale bars indicate 10 µm. (C) Immunofluorescence images showing colocalization of UCP2 (Top) and UPC5 (Bottom) (magenta) and TOM70 marking mitochondria (magenta) in human large airway tissue. Merged panels include DAPI. Scale bars indicate 100 µm.

Fig. 6.

Epithelia with abundant ciliated cells have high levels of uncoupled mitochondrial respiration. (A) Schematic showing the strategy for differentiating cultures with few vs. abundant ciliated cells, sampling using 3-mm punches of epithelia, and measuring O₂ consumption. Studies were performed at 18.5% O₂. (B) Immunofluorescence images showing acetylated α-tubulin marking cilia (green) and actin staining (magenta) marking cell boundaries. Epithelia were grown in USG (Left) vs. Pneumacult-ALI (PC-ALI) (Right) (Method 1). Scale bars indicate 10 µm. (C) Same as Panel B except that epithelia were grown under 0.5% O₂ (Left) vs. 18.5% O₂ (Right); (Method 2). (D) Representative mitochondrial stress test showing responses after injection of oligomycin A (3 μM), FCCP (0.5 μM), and rotenone (1 μM) and antimycin A (1 μM) for epithelia differentiated under Method 1. Data are mean ± SEM of three to four technical replicates. (E) Same as Panel D except that epithelia were differentiated with Method 2. (F) Basal OCR in epithelia with few and abundant ciliated cells (Methods 1 and 2). (G) Uncoupled respiration as a percentage of total mitochondrial respiration in epithelia with few and abundant ciliated cells (Methods 1 and 2). For panels F and G, each set of data points and lines is from a different human donor. Bars and whiskers indicate mean ± SEM. The asterisk indicates P < 0.05 by paired Student’s t test.

Fig. 7.

Knockdown of UCP2 and UCP5 increased mitochondrial membrane potential, ROS production, and metabolic markers of lipid peroxidation. (A) UCP2 and UCP5 mRNA levels after control antisense oligonucleotides (Ctrl) and double knockdown (dKD) of UCP2 and UCP5. (B) Basal respiration by Ctrl and dKD treated airway epithelia. (C) Uncoupled respiration as a percentage of total mitochondrial respiration in Ctrl vs. dKD epithelia. (D) Schematic showing that a high mitochondrial membrane potential increases TMRM fluorescence. (E) TMRM fluorescence intensity of ciliated cells vs. all other cells (including basal, secretory, and rare cell types) assayed by flow cytometry. The uncoupling agent, FCCP served as a positive control. (F) TMRM fluorescence intensity of ciliated cells in Ctrl vs. UCP2 and UCP5 dKD treated ciliated cells assayed by flow cytometry. (G) Levels of ROS in Ctrl and dKD treated cells, as measured using CellRox-green (Left) and ESR (Right). ESR rates are normalized to time zero. (H) Levels of metabolic markers of lipid peroxidation (4-HNE, MDA, and 8-isoprostaglandin) in Ctrl vs. dKD treated epithelia. Ratios of oxidized to reduced glutathione (GSSG/GSH) in Ctrl vs. dKD treated epithelia. (I) CBF in Ctrl vs. dKD treated epithelia. Studies were done at 20 °C. In all panels, each set of data points and lines is from a different human donor. Asterisks indicate ***P < 0.001, **P < 0.01, and *P < 0.05 by paired Student’s t test. Bars and whiskers indicate mean ± SEM.

Fig. 8.

Model of ciliated cell mitochondrial function. Clustering of mitochondria and the relatively high O₂ levels make the apical region of ciliated cells a hotspot for ROS production. UCP2 and UCP5 on the inner mitochondrial membrane decrease the protonmotive force, thereby balancing mitochondrial efficiency with decreased ROS production.