No abstract

Clusterin is a widely expressed, well conserved, secreted glycoprotein, which is highly induced in tissues regressing as a consequence of apoptotic cell death in vivo. It has recently been shown that clusterin expression is only confined to surviving cells following the induction of apoptosis in vitro, suggesting that it is involved in cell survival rather than death. In the hypothesis that clusterin may be implicated in cellular responses to stress, clusterin gene expression was analyzed in the A431 human epidermoid cancer cell line following heat shock and oxidative stress. Our results show that both a transient heat shock (20 min at 42 degrees C) and various oxidative stresses, including hydrogen peroxide, superoxide anion, hyperoxia and UVA exposure, induce a strong increase in clusterin mRNA levels as assessed by northern blot. Nuclear run-on analysis suggests that transcriptional activation is involved in inducing clusterin mRNA in response to heat shock. Using pulse-chase analysis of control and heat shocked cells, it is shown that clusterin mRNA is translated and secreted, thus resulting in increased extracellular levels of the protein following heat shock. To investigate the function of clusterin in response to these stresses, clusterin anti-sense transfectants that stably express virtually no clusterin at the mRNA and protein level were generated in A431 cells. These anti-sense transfectants are shown to be highly sensitive to apoptotic cell death induced by heat shock or oxidative stress compared with wild-type A431 cells or control transfectants. Taken together, our results show that clusterin gene expression is induced in response to heat shock and oxidative stress in human A431 cells, and confers cellular protection against heat shock and oxidative stress.

Dendritic cells play a key role in immune responses. There is growing evidence that reactive oxygen species participate in signaling pathways involving nuclear factor (NF)-kappaB, leading to expression of important immune system genes. We found that, unlike H2O2, reactive oxygen species generated by the reaction of oxidase on xanthine induced early phenotypic maturation of dendritic cells by upregulating specific markers CD80, CD83, and CD86 and downregulating mannose receptor-mediated endocytosis. Maturation induced by xanthine oxidase was prevented by allopurinol, an inhibitor of xanthine oxidase activity, and by N-acetylcysteine. The proteasome inhibitor MG-132, which blocks NF-kappaB activation, also inhibited CD86 upregulation, but not endocytosis downregulation by reactive oxygen species. Finally, xanthine-xanthine oxidase enhanced or blocked antigen presentation by dendritic cells depending on whether they had been prepulsed or not with the antigen. Taken together, these results demonstrate that oxidative stress induces phenotypic and functional maturation of dendritic cells, partly through an NF-kappaB-dependent mechanism.

The poor ability of respiratory epithelial cells to proliferate and differentiate in vitro into a pseudostratified mucociliated epithelium limits the general use of primary airway epithelial cell (AEC) cultures generated from patients with rare diseases, such as cystic fibrosis (CF). Here, we describe a procedure to amplify AEC isolated from nasal polyps and generate long-term cultures of the respiratory epithelium. AEC were seeded onto microporous permeable supports that carried on their undersurface a preformed feeder layer of primary human airway fibroblasts. The use of fibroblast feeder layers strongly stimulated the proliferation of epithelial cells, allowing the expansion of the cell pool with successive passages. AEC at increasing passage were seeded onto supports undercoated with airway fibroblasts and exposed to air. Either freshly isolated or amplified AEC could differentiate into a pseudostratified mucociliated epithelium for at least 10 mo. Thus, CF epithelia cultures showed elevated Na+ transport, drastic hyperabsorption of surface liquid, and absence of cAMP-induced Cl- secretion as compared with non-CF cultures. They were also characterized by thick apical secretion that hampered the movement of cell surface debris by cilia. However, CF respiratory epithelia did not show increased production of mucins or IL-8. The method described here is now routinely used in our laboratory to establish long-term cultures of well differentiated respiratory epithelia from human airway biopsies.

Exocytosis from Weibel-Palade bodies, the secretory granules of vascular endothelial cells, causes the rapid release of von Willebrand factor (vWF), an adhesive glycoprotein involved in primary hemostasis, and cell surface expression of P-selectin, a membrane protein involved in neutrophil binding. Thus, exocytosis may represent a link between hemostasis and inflammation. We investigated the effect of reactive oxygen intermediates (ROIs) on vWF secretion. Incubation of cultured endothelial cells with xanthine oxidase (XO), which generates superoxide anions (O2-), induces a potent, rapid secretory response. However, vWF release was not observed in response to H2O2. Extracellular, subendothelial vWF deposits typically seen after exocytosis from Weibel-Palade bodies were observed after exposure to XO. XO caused a rapid, sustained increase in intracellular free calcium concentration ([Ca2+]i). vWF secretion was markedly inhibited by BAPTA-AM, a cell-permeant calcium chelator. Removal of extracellular calcium did not inhibit vWF release, although the sustained phase of the [Ca2+]i increase was suppressed. These results suggest that XO-induced vWF release is mediated by the initial increase in [Ca2+]i which is caused by calcium mobilization from intracellular stores rather than by calcium influx. Exocytosis from Weibel-Palade bodies may contribute to the pathogenic effect of ROIs in atherosclerosis and inflammation.

To explore the level of regulation of the expression of the major antioxidant enzymes in response to hyperoxia, we exposed human umbilical vein endothelial cells to 95% O2 for 3 and 5 days and measured (1) the steady-state mRNA levels, (2) the activities, and (3) the immunoreactive content of CuZn and Mn superoxide dismutases (SOD), catalase (CAT), and glutathione peroxidase (GP). We found that a 3-day exposure to 95% O2 caused (1) an increase in CuZnSOD mRNA (by 41%), CAT mRNA (by 26%), and GP mRNA (by 173%); (2) an increase in CuZnSOD activity (by 30%), a decrease in CAT activity (by 37%), and an increase in GP activity (by 60%); and (3) an increase in CuZnSOD immunodetectable protein (by 26%) and a loss in CAT immunoreactive protein (by 27%). After a 5-day exposure to 95% O2, there was (1) a 93% increase in CuZnSOD mRNA, a 71% increase in CAT mRNA, and a 127% increase in GP mRNA; (2) a 56% increase in CuZnSOD activity, a 70% decrease in CAT activity, and an 89% increase in GP activity; and (3) a 35% increase in CuZnSOD immunoreactive protein and a 55% loss in CAT immunoreactive protein. There was no change in the steady-state MnSOD mRNA level after 3 days in 95% O2, but a 100% increase was observed on day 5 of oxygen exposure. MnSOD activity was unchanged in cells exposed to hyperoxia for 3 and 5 days. These data suggest that, in human umbilical vein endothelial cells, the regulation of antioxidant enzymes expression in response to O2 is complex and exerted at different levels.

Human peripheral blood monocytes (PBM) produce superoxide anions (O2-.) by a process involving electron transfer from NADPH to O2, catalyzed by the respiratory burst enzyme NADPH oxidase. We have previously shown that phagocytosis, while activating NADPH oxidase, induced in PBM the synthesis of heat shock (HS) proteins (HSP). The present study was undertaken to establish whether this increase in HSP expression was related to O2-. and/or to classical second messengers such as protein kinase C (PKC). Thus, the effects of the PKC activator phorbol 12-myristate 13-acetate (PMA) were compared with those of heat shock on the expression, in PBM, of the major HSP, hsp70 and hsp90, using biometabolic labeling, Western and Northern blotting, and gel mobility shift assays. PMA induced the accumulation of mRNA and an increased expression of hsp90 and, to a lesser extent, hsp/hsc70 (hsc is the cognate, constitutive form). This induction was also observed in PBM from patients with chronic granulomatous disease, a genetic defect in NADPH oxidase, and was abolished by the PKC inhibitors staurosporine and H-7. PMA did not cause activation of the HS factor, and the PMA-induced overexpression expression of HSP was not blocked by the transcriptional inhibitor actinomycin D. HSP-specific mRNA stability was increased after PMA exposure as compared with heat shock. These results suggest that O2-. is not involved in the PMA-mediated induction of hsp70 and hsp90 and that, in contrast to HS, PMA increases the expression of HSP as a result of PKC-induced mRNA stabilization rather than of transcriptional activation of HS genes.

In cells undergoing oxidative stress, DNA damage may result from attack by .OH radicals produced by the Fenton reaction, and/or by nucleases activated by nuclear calcium. In the present study, the participation of these two mechanisms was investigated in HeLa cells. Nuclear-targeted aequorin was used for selectively monitoring Ca2+ concentrations within the nuclei ([Ca2+]n), in conjunction with the cell-permeant calcium chelator bis-(o-aminophenoxy)ethane-N,N,N', N'-tetraacetic acid acetoxymethyl ester (BAPTA/AM), the lipid-soluble broad-spectrum metal chelator with low affinity for Ca2+ and Mg2+ N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN), and the high-affinity iron/copper chelator 1, 10-phenanthroline (PHE). In Ca2+-containing medium, H2O2 induced extensive DNA strand breaks and an increase in [Ca2+]n that was almost identical to that observed in the cytosol ([Ca2+]c). In cells bathed in Ca2+-free/EGTA medium, in which the increases in [Ca2+]n and [Ca2+]c due to H2O2 were significantly reduced, similar levels of DNA fragmentation also occurred. In cells preloaded with BAPTA/AM or TPEN, the small increase of [Ca2+]n normally elicited by H2O2 in Ca2+-free medium was completely buffered, and DNA damage was largely prevented. On the other hand, pretreatment with PHE did not affect the calcium response in the nuclei, but completely prevented DNA strand breakage induced by H2O2. Re-addition of 100 microM CuSO4 and 100 microM FeSO4 to TPEN- and PHE-treated cells prior to H2O2 challenge reversed the effect of TPEN and PHE, whereas 1 mM was necessary to negate the effect of BAPTA/AM. The levels of DNA strand breakage observed, however, did not correlate with the amounts of 8-hydroxy 2'-deoxyguanosine (8-OHdG): H2O2 did not produce 8-OHdG, whereas PHE alone slightly increased 8-OHdG levels. CuSO4 and FeSO4 enhanced the effects of PHE, particularly in the presence of H2O2. Exposure of cells to a mixture of CuSO4/FeSO4 also resulted in a significant increase in 8-OHdG levels, which was prevented in cells preloaded with BAPTA/AM. Similar results were obtained in a cell-free system using isolated calf thymus DNA exposed to CuSO4/FeSO4, regardless of whether H2O2 was present or not. These results suggest that BAPTA/AM prevents H2O2-induced DNA damage by acting as an iron/copper chelator. These data also indicate that caution must be exercised in using Ca2+ chelating agents as evidence for a role in cellular Ca2+ levels in experimental conditions in which transition-metal-ion-mediated oxidant production is also occurring.