Mitochondrial DNA integrity may be a determinant of endothelial barrier properties in oxidant-challenged rat lungs

Abstract

In cultured pulmonary artery endothelial cells and other cell types, overexpression of mt-targeted DNA repair enzymes protects against oxidant-induced mitochondrial DNA (mtDNA) damage and cell death. Whether mtDNA integrity governs functional properties of the endothelium in the intact pulmonary circulation is unknown. Accordingly, the present study used isolated, buffer-perfused rat lungs to determine whether fusion proteins targeting 8-oxoguanine DNA glycosylase 1 (Ogg1) or endonuclease III (Endo III) to mitochondria attenuated mtDNA damage and vascular barrier dysfunction evoked by glucose oxidase (GOX)-generated hydrogen peroxide. We found that both Endo III and Ogg1 fusion proteins accumulated in lung cell mitochondria within 30 min of addition to the perfusion medium. Both constructs prevented GOX-induced increases in the vascular filtration coefficient. Although GOX-induced nuclear DNA damage could not be detected, quantitative Southern blot analysis revealed substantial GOX-induced oxidative mtDNA damage that was prevented by pretreatment with both fusion proteins. The Ogg1 construct also reversed preexisting GOX-induced vascular barrier dysfunction and oxidative mtDNA damage. Collectively, these findings support the ideas that mtDNA is a sentinel molecule governing lung vascular barrier responses to oxidant stress in the intact lung and that the mtDNA repair pathway could be a target for pharmacological intervention in oxidant lung injury.

Fig. 1.

Distribution of hemagglutinin (HA)-tagged 8-oxoguanine DNA glycosylase 1 (Ogg1) fusion protein in subcellular fractions derived from perfused rat lung. Isolated rat lungs were perfused for 30 min with 220 nM Ogg1-HA fusion protein construct, after which lung tissue was homogenized and subcellular fractions were prepared as described in materials and methods. Western analysis of HA immunoreactivity was applied to total lysate (TL) derived from control lung tissue not treated with the fusion protein (C), as well as TL, nuclear (N), mitochondrial (M), and cytosolic (Cyt) fractions derived from lungs perfused with the Ogg1 fusion protein. Representative of 3 experiments.

Fig. 2.

Impact of Ogg1 and endonuclease III (Endo III) fusion proteins on glucose oxidase (GOX)-induced oxidative mtDNA damage in perfused rat lungs. A: representative quantitative Southern blot analyses of oxidative mitochondrial DNA (mtDNA) damage detected by alkali without (−) and with (+) formamidopyrimidine DNA glycosylase (Fpg) in isolated lungs perfused with the fusion protein vehicle (Veh), the Ogg1 or Endo III fusion proteins in the absence of an oxidant stress, or the 2 fusion proteins administered prior to challenging the lungs with GOX-generated H₂O₂. Note that diminished hybridization intensity relative to PSS or Veh controls is indicative. B: calculated changes in Fpg-detectable oxidative base lesion for the above-mentioned experimental groups. N = 4–6 per group. *Significantly different from Veh controls.

Fig. 3.

Impact of GOX and of Ogg1 and Endo III fusion proteins on mean DNA fragment length determined from intact perfused rat lung tissue. Mean fragment length of alkali-treated DNA was determined in isolated rat lungs perfused with the fusion protein vehicle (Veh), the Ogg1 or Endo III fusion proteins in the absence of an oxidant stress, or the 2 fusion proteins administered prior to challenging the lungs with GOX-generated H₂O₂. There were no detectable changes in mean DNA fragment length between the experimental groups. N = 4–6 per group.

Fig. 4.

Impact of Ogg1 and Endo III fusion proteins on GOX-induced changes the vascular filtration coefficient (K_f) in perfused rat lungs. Vascular filtration coefficient was determined in rat lungs perfused with Veh, the Ogg1 or Endo III fusion proteins in the absence of an oxidant stress, or the 2 fusion proteins administered prior to challenging the lungs with GOX-generated H₂O₂. Note that GOX caused large increases in K_f that were attenuated by either the Ogg1 or the Endo III fusion proteins. N = 4–6 per group. *Significantly increased compared with Veh control at P < 0.05. **Significantly decreased compared with Veh + GOX at P < 0.05.

Fig. 5.

Photomicrographs (magnification: ×10) of hematoxylin and eosin-stained lung tissue from control (Con) lungs, lungs challenged with GOX alone and in the presence of mt-targeted Ogg1. Arrow denotes presence of perivascular cuffs indicative of endothelial barrier dysfunction. Lungs treated with mt-targeted Ogg1 alone were indistinguishable from controls (data not shown). Photomicrographs representative of 3 lung preparations per experimental group.

Fig. 6.

Mitochondria-targeted Ogg1 reverses GOX-induced vascular barrier dysfunction and mtDNA damage in perfused rat lungs. Top: after baseline K_f measurement (0), either GOX or vehicle was added to the reservoir and K_f was assessed 60 min later (1). Mitochondria-targeted Ogg1 fusion protein or vehicle was then added to the reservoir and a final K_f was measured 45 min later (2). Control lungs were perfused with no additions for 105 min, with K_f determined at 90 min and termination of perfusion. N = 4–6 for each group. *Different from Veh at P < 0.05. Bottom: in separate experiments, lungs were perfused and treated as just described, and at the end of perfusion lung tissue DNA was isolated for determination of alkali and alkali + Fpg-detectable mtDNA damage by quantitative Southern blot analysis. In the representative Southern blot shown, note that hybridization intensities are reduced in GOX + Veh compared with Veh alone, indicating the persistence of mtDNA damage in GOX-treated lungs, and that addition of mitochondria-targeted Ogg1 partially restores hybridization intensity measured 45 min thereafter, indicating incomplete repair of mtDNA lesions. Representative of 3 experiments.