Peroxiredoxin 6 fails to limit phospholipid peroxidation in lung from Cftr-knockout mice subjected to oxidative challenge

Abstract

Oxidative stress plays a prominent role in the pathophysiology of cystic fibrosis (CF). Despite the presence of oxidative stress markers and a decreased antioxidant capacity in CF airway lining fluid, few studies have focused on the oxidant/antioxidant balance in CF cells. The aim of the current study was to investigate the cellular levels of reactive oxygen species (ROS), oxidative damage and enzymatic antioxidant defenses in the lung of Cftr-knockout mice in basal conditions and as a response to oxidative insult.The results show that endogenous ROS and lipid peroxidation levels are higher in Cftr(-/-) lung when compared to wild-type (Cftr(+/+)) in basal conditions, despite a strong enzymatic antioxidant response involving superoxide dismutases, glutathione peroxidases and peroxiredoxin 6 (Prdx6). The latter has the unique capacity to directly reduce membrane phospholipid hydroperoxides (PL-OOH). A dramatic increase in PL-OOH levels in Cftr(-/-) lung consecutive to in vivo oxidative challenge by paraquat (PQ) unmasks a susceptibility to phospholipid peroxidation. PQ strongly decreases Prdx6 expression in Cftr(-/-) mice compared to Cftr(+/+). Similar results were obtained after P. aeruginosa LPS challenge. Two-dimensional gel analysis of Prdx6 revealed one main molecular form in basal conditions and a PQ-induced form only detected in Cftr(+/+) lung. Mass spectrometry experiments suggested that, as opposed to the main basal form, the one induced by PQ is devoid of overoxidized catalytic Cys47 and could correspond to a fully active form that is not induced in Cftr(-/-) lung. These results highlight a constitutive redox imbalance and a vulnerability to oxidative insult in Cftr(-/-) lung and present Prdx6 as a key component in CF antioxidant failure. This impaired PL-OOH detoxification mechanism may enhance oxidative damage and stress-related signaling, contributing to an exaggerated inflammatory response in CF lung.

Figure 1. Intracellular ROS and lipid peroxidation levels.

A) Endogenous ROS levels in ciliated tracheal cells. Epithelial cells freshly isolated from trachea were incubated with CM-H2DCF-DA and visualized by fluorescent confocal microscopy. Under identical imaging conditions, DCF fluorescence was markedly increased in ciliated epithelial cells from Cftr ^−/− mice (left) as confirmed by quantification of relative fluorescence (right). Values represent fluorescent intensity (arbitrary units, a.u.) and data correspond to the mean±S.E. of n = 90 for Cftr ^+/+ and n = 71 for Cftr ^−/− cells, obtained from five animals of each genotype. B) Lipid peroxidation in lung tissue. Levels of TBARS were significantly elevated in lung tissue from Cftr ^−/− mice. Values are presented as malondialdehyde equivalences (micromole per gram of lung protein), mean±S.E., n = 9 from 2 different experiments, in duplicate). * P≤0.05, *** P≤0.001

Figure 2. Antioxidant enzymatic activities.

Comparison between Cftr ^+/+ and Cftr ^−/− mice of SOD (A), GPx (B) and CAT (C) activities in lung homogenates. Enzymatic activities were measured by standard spectrophotometric methods. Lung homogenates from Cftr ^−/− mice displayed significantly higher SOD and GPx activities than those from Cftr ^+/+ mice. Both Cftr ^+/+ and Cftr ^−/− mice displayed a similar level of CAT activity. Values are presented as specific activity (mean±S.E., n = 5, in duplicate). * P≤0.05

Figure 3. Prdx6 protein and mRNA levels.

A) Prdx6 protein expression levels. Lung protein extracts from Cftr ^+/+ and Cftr ^−/− mice were subjected to immunoblot analysis using antibodies against Prdx6 (top) and α-tubulin (bottom). On the left, a representative blot obtained with three pairs of mice is presented. It shows that Prdx6 expression levels are higher in Cftr ^−/− lung compared to Cftr ^+/+, as confirmed on the right by quantitative analysis. Individual data were quantified as ratio of fluorescence intensity for Prdx6 bands to the intensity obtained for α-tubulin. Data are shown as percent of Cftr ^+/+ (mean±S.E., n = 9 from 2 different experiments). B) Prdx6 mRNA levels. Total lung RNA from Cftr ^+/+ and Cftr ^−/− mice was reversely transcribed and the amounts of cDNAs coding for Prdx6 and the reference protein β-actin were measured by quantitative Real-Time PCR. Prdx6 mRNAs are increased in Cftr ^−/− lung compared to Cftr ^+/+. Indicated values represent the ratio of Prdx6 mRNAs to β-actin mRNAs (mean±S.E., n = 8, in triplicate). C) Immunohistochemical analysis of Prdx6. Acetone-fixed lung cryosections from Cftr ^+/+ and Cftr ^−/− mice were incubated with an anti-Prdx6 antibody and visualized by confocal microscopy. Sections of bronchioles from Cftr ^+/+ (a) and Cftr ^−/− (c) mice and the alveolar region from Cftr ^+/+ (b) and Cftr ^−/− (d) mice are shown. Cftr ^−/− lung sections present higher Prdx6 staining intensity compared to Cftr ^+/+ mice. * P≤0.05, ** P≤0.01

Figure 4. Phosphatidylcholine hydroperoxides quantification.

Cftr ^+/+ and Cftr ^−/− mice were injected I.P. twice with paraquat (PQ) (+) or saline (−) at 24 h intervals and sacrificed 48 h after the first injection. After lipid extraction, phosphatidylcholines (PC) and their corresponding hydroperoxides (PC-OOH) were separated by HPLC and detected by UV and chemiluminescence. After PQ challenge PLPC-OOH (16∶0–18∶2), PAPC–OOH (16∶0–20∶4) and PDPC–OOH (16∶0–22∶6) were significantly higher in Cftr ^−/− lung compared to Cftr ^+/+. Data are expressed as the ratio of PC–OOH to PC peak heights (mean±S.E.) and are representative of 3 different experiments with at least n = 3 animals in each experiment. * P≤0.05, *** P≤0.001

Figure 5. Prdx6 protein and mRNA levels after paraquat exposure.

A) Prdx6 protein expression levels after paraquat (PQ) treatment. Lung protein extracts from Cftr ^+/+ and Cftr ^−/− mice treated with either PQ (+) or saline (−) were subjected to immunoblot analysis using antibodies against Prdx6. The decrease in Prdx6 protein after PQ exposure is greater in Cftr ^−/− mice than in Cftr ^+/+. On the left, a representative blot is presented. Individual data were quantified as ratio of fluorescence intensity for Prdx6 bands to the intensity obtained for α-tubulin. Data are shown as percent of Cftr ^+/+ treated with saline (mean±S.E., n = 5 for saline and n = 10 for PQ). B) Prdx6 mRNA levels after PQ treatment. Total lung RNA from Cftr ^+/+ and Cftr ^−/− mice treated with either PQ (+) or saline (−) was reversely transcribed and the amount of cDNAs coding for Prdx6 and the reference protein β-actin were measured by quantitative Real-Time PCR. The decrease in Prdx6 mRNA after PQ challenge is greater in Cftr ^−/− mice than in Cftr ^+/+. Indicated values correspond to the ratio of Prdx6 to β-actin mRNA (mean±S.E., n = 8 for saline and n = 5 for PQ, in triplicate). C) Immunohistochemical analysis of Prdx6 after PQ treatment. Acetone-fixed cryosections from Cftr ^+/+ and Cftr ^−/− mice treated with PQ or saline were incubated with anti-Prdx6 antibodies and visualized by confocal microscopy. Panels correspond to sections of bronchioles from Cftr ^+/+ mice treated with either saline (a) or PQ (b) and from Cftr ^−/− mice treated with saline (c) or PQ (d). Prdx6 staining intensity was lower after PQ exposure. * P≤0.05, *** P≤0.001

Figure 6. Prdx6 protein expression levels after LPS treatment.

Lung protein extracts from Cftr ^+/+ and Cftr ^−/− mice treated with either LPS (+) or saline (−) were subjected to immunoblot analysis using antibodies against Prdx6. The decrease in Prdx6 protein after LPS exposure is greater in Cftr ^−/− than in Cftr ^+/+ lung. On the left, a representative blot is presented. Individual data were quantified as ratio of fluorescence intensity for Prdx6 bands to the intensity obtained for α-tubulin. Data are shown as percent of Cftr ^+/+ treated with saline (mean±S.E., n = 5 for saline and n = 4 for LPS ). * P≤0.05

Figure 7. 2D immunoblot analysis of Prdx6 forms after paraquat treatment.

Lung protein extracts from Cftr ^+/+ and Cftr ^−/− mice treated with either paraquat (PQ) or saline (control) were subjected to immunoblot analysis after IEF/SDS-PAGE. Two forms (pI ∼5.9 and ∼6.3) are present in Cftr ^+/+ and Cftr ^−/− lung treated or not with PQ. An additional form of Prdx6 (pI ∼6.7) is only detected in lung extracts form Cftr ^+/+ mice treated with PQ. For both conditions, immunoblots are representative of n = 4 animals of each genotype.

Figure 8. Cysteine 47 overoxidation to sulfonic acid in Prdx6 extracted from pI 6.3 spot.

Fragmentation spectra of DFTPVC(SO3H)TTELGR peptide, corresponding to Prdx6 amino acid residues 42–53, unequivocally indicates oxydation of Cys47 to sulfonic acid. The most abundant peaks (singly charged) of the peptide are annotated as b- and y- series daughter ions, in red and blue respectively. Fragmentation spectra of the non-oxidized peptide show a similar fragmentation pattern, with a mass shift of 48 Da on peaks with oxidized cysteine (data not shown).