Superoxide dismutase protects against apoptosis and alveolar enlargement induced by ceramide

Abstract

The molecular events leading to emphysema development include generation of oxidative stress and alveolar cell apoptosis. Oxidative stress upregulates ceramides, proapoptotic signaling sphingolipids that trigger further oxidative stress and alveolar space enlargement, as shown in an experimental model of emphysema due to VEGF blockade. As alveolar cell apoptosis and oxidative stress mutually interact to mediate alveolar destruction, we hypothesized that the oxidative stress generated by ceramide is required for its pathogenic effect on lung alveoli. To model the direct lung effects of ceramide, mice received ceramide intratracheally (Cer(12:0) or Cer(8:0); 1 mg/kg) or vehicle. Apoptosis was inhibited with a general caspase inhibitor. Ceramide augmentation shown to mimic levels found in human emphysema lungs increased oxidative stress, and decreased, independently of caspase activation, the lung superoxide dismutase activity at 48 h. In contrast to their wild-type littermates, transgenic mice overexpressing human Cu/Zn SOD were significantly protected from ceramide-induced superoxide production, apoptosis, and air space enlargement. Activation of lung acid sphingomyelinase in response to ceramide treatment was abolished in the Cu/Zn SOD transgenic mice. Since cigarette smoke-induced emphysema in mice is similarly ameliorated by the Cu/Zn SOD overexpression, we hypothesized that cigarette smoke may induce ceramides in the mouse lung. Utilizing tandem mass spectrometry, we documented increased lung ceramides in adult mice exposed to cigarette smoke for 4 wk. In conclusion, ceramide-induced superoxide accumulation in the lung may be a critical step in ceramide's proapoptotic effect in the lung. This work implicates excessive lung ceramides as amplifiers of lung injury through redox-dependent mechanisms.

Fig. 1.

Effect of ceramide on oxidative stress in the mouse lung and modulation by Cu/Zn SOD overexpression. A: nitrotyrosine expression levels detected by immunohistochemistry (IHC) in the lung parenchyma 48 h after intratracheal administration of vehicle (n = 5) or ceramide 12:0 (Cer; 1 mg/kg; n = 5). Paraffin sections from lungs of mice were coded, and images of alveolar tissue were captured in a random blinded fashion. The number of alveolar cells expressing nitrotyrosine and the expression intensity were quantified by image software analysis and expressed as index (means + SD; *P = 0.002, ANOVA). B: lipid peroxide levels measured in the lungs of wild-type (WT; n = 4) or Cu/Zn SOD transgenic (Tg; n = 5) mice 48 h after Cer instillation compared with control (means + SE; *P = 0.050 vs. control). C: superoxide levels in bronchoalveolar (BAL) lavage of WT (n = 4) and Tg (n = 6) mice 48 h after intratracheal Cer instillation (means + SE; *P = 0.01 vs. WT Cer).

Fig. 2.

Antioxidant enzymatic activities in response to ceramide augmentation in the lung and modulation by caspase inhibitors. Lung SOD activity levels measured 24 h after intratracheal dihydroceramide (control; white bar) or ceramide (Cer) instillation (1 mg/kg) without (black bar) or with Z-Asp-CH2-DCB (3 mg/kg ip daily ×2; gray bar). SOD activity was measured in lung lysates after fractionation into cytosolic (A) and mitochondrial components (B), respectively, followed by normalization by protein concentration (means + SE; *P < 0.05 vs. control). C: catalase activity levels in the mouse lung measured 24 h after intratracheal dihydroceramide (control; n = 3) or ceramide (Cer; n = 4) instillation (1 mg/kg) without (black bar) or with Z-Asp-CH2-DCB (3 mg/kg ip daily ×2; gray bar). Results were expressed as mean fold change compared with control (dotted line) +SE.

Fig. 3.

SOD activity modulation by ceramides and Cu/Zn SOD transgenic overexpression. A: expression levels of human Cu/Zn SOD in the lung of wild-type (WT) mice (lanes 1–4) and transgenic (Tg) mice (lanes 5–10) detected by Western blot with species-specific Cu/Zn SOD antibody. B: SOD activity levels in the lungs of WT and Cu/Zn SOD transgenic mice measured 48 h after intratracheal vehicle (Veh) or ceramide (Cer) instillation. SOD activity was measured in lung lysates and normalized by protein concentration (means + SD; *P = 0.03 vs. Veh-Wt; **P < 0.05 vs. WT). Number of animals studied: WT + Veh, n = 5; WT + Cer, n = 4; Tg + Veh, n = 5; Tg + Cer, n = 8.

Fig. 4.

Effect of Cu/Zn SOD transgenic overexpression on ceramide-triggered caspase-3 activation in the mouse lung. A: kinetics of caspase-3 activity in mouse lung lysates incubated with a fluorescently labeled specific caspase substrate. Results were expressed in units of activity of recombinant human caspase-3, normalized by protein concentration; means ± SD. B: boxplot of caspase-3 activities measured in lung lysates from wild-type (WT) and Cu/Zn SOD transgenic (Tg) mice after vehicle (light boxes) or ceramide (Cer; dark boxes; n = 7) instillation. Horizontal line represents median activity normalized by protein concentration, with dots showing outliers and whiskers showing the 5th and 95th percentile values, respectively. (ANOVA followed by Bonferroni t-test; *P < 0.05 vs. Veh-Wt; **P < 0.05 vs. Cer-Wt). C: expression levels of active caspase-3 (brown, arrows) detected by IHC with specific active caspase-3 antibody in the lung parenchyma of wild-type and Cu/Zn SOD transgenic (Tg) mice 48 h after intratracheal instillation of vehicle or ceramide (1 mg/kg). D: quantification of caspase-3 immunostaining. Random sections were imaged from coded slides, followed by computation of the intensity and areas of caspase-3 activation by image analysis software, normalized by total nuclei (DAPI-stained; means + SE, *P < 0.05 compared with Cer-Wt. Ctl represented Veh-treated lungs). Number of animals studied: Wt + Veh, n = 5; Wt + Cer, n = 4; Tg + Veh, n = 5; Tg + Cer, n = 8.

Fig. 5.

Effect of Cu/Zn SOD overexpression on air space enlargement in ceramide-treated mice. A: microphotographs of hematoxylin-eosin-stained paraffin-embedded sections of mouse lung fixed at a constant inflation pressure. Compared with vehicle, ceramide instillation increased the air space size in wild-type mice (WT) but not Cu/Zn SOD transgenic (TG) mice at 48 h. B: standardized morphometric measurement of mean linear intercepts of lung parenchyma prepared as described in A and presented as boxplot with the horizontal line depicting median values, whiskers showing the 5th and 95th percentiles and dots depicting outliers. (Wt + Veh, n = 5; Wt + Cer, n = 4; Tg + Veh, n = 5; Tg + Cer, n = 8; P < 0.05, ANOVA).

Fig. 6.

Effect of Cu/Zn SOD overexpression on acid sphingomyelinase (ASM) activation in ceramide-treated mice. ASM activity was measured in wild-type mice (WT) or Cu/Zn SOD transgenic (Tg) mice 48 h after intratracheal ceramide instillation utilizing a fluorometric activity assay. The enzymatic activity was normalized by protein concentration, and results were expressed as average % change compared with the vehicle control for each genotype (% +SE; *P = 0.01, Student's t-test). Number of animals studied: Wt + Veh, n = 5; Wt + Cer, n = 4; Tg + Veh, n = 5; Tg + Cer, n = 8.

Fig. 7.

Lung ceramide levels in response to cigarette smoke exposure in mice. A: levels of endogenous ceramides measured by tandem mass spectrometry and normalized by lipid phosphorus in lungs of mice exposed to air (n = 5) or cigarette smoke (Csk; n = 5) for 4 wk (means ± SE; *P = 0.01). B: levels of endogenous dihydroceramides, immediate precursors of ceramides in the de novo synthetic pathway, measured by tandem mass spectrometry and normalized by lipid phosphorus in the lungs of mice exposed to air or Csk for 4 wk (means ± SE; *P = 0.003).