No abstract

BACKGROUND: DNA-PK (DNA-dependent protein kinase) is a stress-activated serine/threonine kinase that plays a central role in vascular smooth muscle cell proliferation and vascular proliferative disease processes such as neointimal formation. In this study, we link the activation of DNA-PK to the function of the transcription factor YB-1 (Y-box-binding protein). METHODS: To identify YB-1 phosphorylation by DNA-PK, we generated different YB-1-expressing vectors. YB-1 nuclear translocation was investigated using immunoblotting and immunofluorescence staining. For YB-1 activity, luciferase assays were performed. RESULTS: We show by mutational analysis and kinase assay that the transcriptional regulator YB-1 is a substrate of DNA-PK. Blockade of DNA-PK by specific inhibitors revealed its critical involvement in YB-1phosphorylation as demonstrated by inhibition of an overexpressed YB-1 reporter construct. Using DNA-PK-deficient cells, we demonstrate that the shuttling of YB-1 from the cytoplasm to the nucleus is dependent on DNA-PK and that the N-terminal domain of YB-1 is phosphorylated at threonine 89. Point mutation of YB-1 at this residue abrogated the translocation of YB-1 into the nucleus. The phosphorylation of YB-1 by DNA-PK increased cellular DNA repair after exposure to ionizing radiation. Atherosclerotic tissue specimens were analyzed by immunohistochemistry. The DNA-PK subunits and YB-1 phosphorylated at T89 were found colocalized suggesting their in vivo interaction. In mice, the local application of the specific DNA-PK inhibitor NU7026 via thermosensitive Pluronic F-127 gel around dilated arteries significantly reduced the phosphorylation of YB-1. CONCLUSIONS: DNA-PK directly phosphorylates YB-1 and, this way, modulates YB-1 function. This interaction could be demonstrated in vivo, and colocalization in human atherosclerotic plaques suggests clinical relevance of our finding. Phosphorylation of YB-1 by DNA-PK may represent a novel mechanism governing atherosclerotic plaque progression.

Mitochondrial functionality is crucial for the execution of physiologic functions of metabolically active cells in the respiratory tract including airway epithelial cells (AECs). Cigarette smoke is known to impair mitochondrial function in AECs. However, the potential contribution of mitochondrial dysfunction in AECs to airway infection and airway epithelial barrier dysfunction is unknown. In this study, we used an in vitro model based on AECs exposed to cigarette smoke extract (CSE) followed by an infection with Streptococcus pneumoniae (Sp). The levels of oxidative stress as an indicator of mitochondrial stress were quantified upon CSE and Sp treatment. In addition, expression of proteins associated with mitophagy, mitochondrial content, and biogenesis as well as mitochondrial fission and fusion was quantified. Transcriptional AEC profiling was performed to identify the potential changes in innate immune pathways and correlate them with indices of mitochondrial function. We observed that CSE exposure substantially altered mitochondrial function in AECs by suppressing mitochondrial complex protein levels, reducing mitochondrial membrane potential and increasing mitochondrial stress and mitophagy. Moreover, CSE-induced mitochondrial dysfunction correlated with reduced enrichment of genes involved in apical junctions and innate immune responses to Sp, particularly type I interferon responses. Together, our results demonstrated that CSE-induced mitochondrial dysfunction may contribute to impaired innate immune responses to Sp.

The NF-κB pathway is central pathway for inflammatory and immune responses, and IKKγ/NEMO is essential for NF-κB activation. In a previous report, we identified the role of glycogen synthase kinase-3β (GSK-3β) in NF-κB activation by regulating IKKγ/NEMO. Here, we show that NEMO phosphorylation by GSK-3β leads to NEMO localization into multivesicular bodies (MVBs). Using the endosome marker Rab5, we observed localization into endosomes. Using siRNA, we identified the AAA-ATPase Vps4A, which is involved in recycling the ESCRT machinery by facilitating its dissociation from endosomal membranes, which is necessary for NEMO stability and NF-κB activation. Co-immunoprecipitation studies of NEMO and mutated NEMO demonstrated its direct interaction with Vps4A, which requires NEMO phosphorylation. The transfection of cells by a mutated and constitutively active form of Vps4A, Vps4A-E233Q, resulted in the formation of large vacuoles and strong augmentation in NEMO expression compared to GFP-Vps4-WT. In addition, the overexpression of the mutated form of Vps4A led to increased NF-κB activation. The treatment of cells with the pharmacologic V-ATPase inhibitor bafilomycin A led to a dramatic downregulation of NEMO and, in this way, inhibited NF-κB signal transduction. These results reveal an unexpected role for GSK-3β and V-ATPase in NF-κB signaling activation.

Knowledge about normoxic hypoxia-inducible factor (HIF)-1α stabilization is limited. We investigated normoxic HIF-1α stabilization and its consequences using live cell imaging, immunoblotting, Bio-Plex multiplex immunoassay, immunofluorescence staining, and barrier integrity assays. We demonstrate for the first time that IL-8 and M-CSF caused HIF-1α stabilization and translocation into the nucleus under normoxic conditions in both human coronary endothelial cells (HCAECs) and HIF-1α-mKate2-expressing HEK-293 cells. In line with the current literature, our data show significant normoxic HIF-1α stabilization caused by TNF-α, INF-γ, IL-1β, and IGF-I in both cell lines, as well. Treatment with a cocktail consisting of TNF-α, INF-γ, and IL-1β caused significantly stronger HIF-1α stabilization in comparison to single treatments. Interestingly, this cumulative effect was not observed during simultaneous treatment with IL-8, M-CSF, and IGF-I. Furthermore, we identified two different kinetics of HIF-1α stabilization under normoxic conditions. Our data demonstrate elevated protein levels of HIF-1α-related genes known to be involved in the development of atherosclerosis. Moreover, we demonstrate an endothelial barrier dysfunction in HCAECs upon our treatments and during normoxic HIF-1α stabilization comparable to that under hypoxia. This study expands the knowledge of normoxic HIF-1α stabilization and activation and its consequences on the endothelial secretome and barrier function. Our data imply an active role of HIF-1α in vivo in the vasculature in the absence of hypoxia.