Deletion of caveolin-1 protects hyperoxia-induced apoptosis via survivin-mediated pathways

Abstract

Hyperoxia-induced lung injury is an established model that mimics human acute respiratory distress syndrome. Cell death is a prominent feature in lungs following prolonged hyperoxia. Caveolae are omega-shaped invaginations of the plasma membrane. Caveolin-1 (cav-1), a 22-kDa transmembrane scaffolding protein, is the principal structural component of caveolae. We have recently shown that deletion of cav-1 (cav-1^−/−) protected against hyperoxia-induced cell death and lung injury both in vitro and in vivo; however, the mechanisms remain unclear. Survivin, a member of the inhibitor of apoptosis protein family, inhibits apoptosis in tumor cells. Although emerging evidence suggests that survivin is involved in wound healing, especially in vascular injuries, its role in hyperoxia-induced lung injury has not been investigated. Our current data demonstrated that hyperoxia induced apoptosis via suppressing survivin expression. Deletion of cav-1 abolished this suppression and subsequently protected against hyperoxia-induced apoptosis. Using “gain” and “loss” of function assays, we determined that survivin protected lung cells from hyperoxia-induced apoptosis via the inhibition of apoptosis executor caspase-3. Overexpression of survivin by deletion of cav-1 was regulated by Egr-1. Egr-1 functioned as a negative regulator of survivin expression. Deletion of cav-1 upregulated survivin via decreased Egr-1 binding of the survivin promoter region. Together, this study illustrates the effect of hyperoxia on survivin expression and the role of survivin in hyperoxia-induced apoptosis. We also demonstrate that deletion of cav-1 protects hyperoxia-induced apoptosis via modulation of survivin expression.

Acute Lung Injury (ALI) and its more severe manifestation, acute respiratory distress syndrome (ARDS), are devastating conditions that account for high morbidity and mortality among critically ill patients (28, 44). Exposure to high O₂ tension (hyperoxia) has been shown in animal models to induce lung injury that closely resembles ARDS. Hyperoxia triggers an extensive inflammatory response in the lung that is typically followed by severe damage of the alveolar-capillary barrier. This damage results in impaired gas exchange and pulmonary edema (10, 30, 47). Cell death of pulmonary capillary endothelial cells and alveolar epithelial cells is the major pathological change in hyperoxia-injured lungs (1, 13, 14, 26). Compromised epithelial and endothelial cell function leads to fluid and macromolecule accumulation in the air space and can cause clinical respiratory failure and death (1, 13, 14, 26, 47). The mechanisms underlying lung injury and cell death in response to hyperoxia remain incompletely understood, although recent studies have implicated that both necrosis and apoptosis pathways play important roles in hyperoxic lung injury (1, 13, 14, 26, 47).

Caveolae are one subset of lipid rafts that are enriched in cholesterol and sphingomyelin and are characterized by their omega-shaped invaginations (50–100 nm) in the plasma membrane (36). These structures participate in a number of cellular processes, including endocytosis, transcytosis, and intracellular signal transduction (2, 25, 36, 39). Caveolin-1 (cav-1), a 22-kDa transmembrane scaffolding protein, serves as the principal structural component of caveolae (2, 25, 36, 39). Recent data indicate that cav-1 mediates ALI via a number of different mechanisms, such as increasing hyperoxic cell death and vascular permeability (16, 20, 22). It is known that deletion of cav-1 protects hyperoxia-induced cell death and ALI both in vitro and in vivo (22); however, the mechanisms remain unclear.

Cell death including apoptosis, necrosis, and oncosis, are features commonly found in ALI and are thought to be one of the mechanisms responsible for ALI/ARDS (10, 28, 30, 44, 47). Apoptosis results from the activation of the family of caspases. Caspases are enzymes that cleave cellular proteins at aspartic acid residues (40). Both the intrinsic and extrinsic apoptotic pathways use caspase-3 as the “executioner” caspase, which cleaves the essential substrates for cell survival (6, 11, 40). Among the antiapoptotic factors, there is a family of proteins named inhibitor of apoptosis proteins (IAP). Survivin, a 16.5-kDa protein, is the smallest member of the IAP family (45). Survivin contains a baculovirus inhibitor of apoptosis repeat protein domain (32, 34, 45) that inhibits apoptosis either directly or indirectly by interfering with the function of caspases (32, 34, 45). Survivin is expressed abundantly in embryonic organs, but only negligible levels can be found in the majority of well-differentiated adult normal tissues (9). Survivin is, however, highly reexpressed in various human cancers (9). Due to its importance in inhibiting cell death and cell division, survivin has been studied extensively as a useful diagnostic marker of cancer and a potential target for cancer treatment (9). Despite the demonstration of a variety of roles attributed to survivin in tumor biology, the function of this protein in normal tissue remains unclear.

Recently, emerging evidence suggests that survivin is involved in tissue injury and wound healing. In vascular injuries, survivin plays multiple roles, including the regulation of vascular cell responses, PDGF gene expression, and vein graft remodeling (5). In neighboring endothelial cells, survivin functions as a critical regulator of angiogenesis (5). In addition, survivin was found to regulate smooth muscle cell growth and may play an important role in the prevention of restenosis following revascularization procedures (19). Although a significant number of studies have explored the role of survivin in cancer research and carcinogenesis, thus far, to our knowledge, no studies have explored the role of survivin in hyperoxia-induced lung injury or have determined whether hyperoxia mediates survivin expression and gene regulation.

Our current study explores the role of deletion of cav-1 in hyperoxic apoptosis and the potential apoptotic pathways involved in lung cells. We report here the discovery of a novel target of hyperoxia in the oxidative apoptosis pathways in lung cells and discuss the potential functions of survivin in cav-1-mediated hyperoxic apoptosis.

MATERIALS AND METHODS

Cell culture and treatments.

Human bronchial epithelial cells (Beas-2B), primary mouse lung fibroblasts, and pulmonary endothelial cells were cultured as described (42) and used for experiments after reaching subconfluent monolayers (usually between culture passages 7 and 17). Primary cells were cultured from the lungs of wild-type C57BL/6 mice as previously described (23). Beas-2B lung epithelial cells were purchased from American Type Culture Collection (Manassas, VA) and cultured in the defined medium, DMEM (Cambrex, East Rutherford, NJ). All cells were grown in humidified incubators containing an atmosphere of 5% CO₂ and 95% air at 37°C. Cell cultures were exposed to hyperoxia in modular exposure chambers as described (23), using 95% oxygen with 5% CO₂.

Reagents.

Survivin antibodies and siRNA were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and Applied Biosystems (Foster City, CA). Survivin overexpression clones were obtained from Genecopoeia (Germantown, MD). Caspase activity kits were purchased from Calbiochem (Gibbstown, NJ). All other reagents and chemicals were purchased from Sigma (St. Louis, MO).

Animal exposures.

Wild-type C57BL/6 mice, 8–12 wk, were maintained in laminar flow cages in a pathogen-free facility at the University of Pittsburgh. All procedures were performed in accordance with and approved by the Council on Animal Care at the University of Pittsburgh and the National Research Council's Guide for the Humane Care and Use of Laboratory animals. The C57BL/6 mice were purchased from Jackson Laboratories (Bar Harbor, Maine). The animals were exposed to room air or hyperoxia (95% O₂, 5% N₂).

Transfections and cell viability assays.

Beas-2B cells were transfected with survivin siRNA and control siRNA using transfection reagent (Santa Cruz, CA). Overexpression vectors were transfected with lipoD293 (Signagen, Gaithersburg, MD). Transiently transfected cells were incubated for an additional 24 h and exposed to hyperoxia. After 48 h, cell viability assays were performed using the CellTiter-Glo Luminescent or CellTiter-Blue Cell Viability Assay according to the manufacturer's protocol (Promega, Madison, WI). Briefly, cells were plated into 96-well plates. After transfection and exposure to hyperoxia, cells were washed twice with cold PBS. One-hundred microliters of PBS was added into each well, followed by 100 μl of CellTiter-Glo Substrate. Cells were incubated at room temperature for at least 10 min; luminescent or fluorescent signal was then measured using an Lmax luminometer or Fluorometer (Molecular Devices, Sunnyvale, CA).

Western blot analysis, immunocytochemistry, and immunoprecipitation.

The following antibodies were used for immunoprecipitation and immunoblotting: monoclonal anti-caspase-3 (BD Transduction Laboratories), mono- and polyclonal anti-survivin (Santa Cruz), and polyclonal anti-caspase-3 (Santa Cruz). Western blot analysis or immunocytochemistry were performed as described previously (23).

EMSA.

Double-stranded probe was made from equal amounts of complementary biotin-labeled oligonucleotides (Integrated DNA Technologies, Coralville, IA). Gel mobility shift assays were performed using the Lightshift chemiluminescent electrophoretic mobility shift assay kit (Pierce, Rockford, IL) according to the manufacturer's instructions. Briefly, 4 μg of nuclear proteins were incubated with 50 fmol of biotin end-labeled double-stranded probe for 20 min at room temperature. Electrophoresis was then carried out on 4% native polyacrylamide gels. The protein-DNA complexes were transferred and crosslinked to a nylon membrane. After incubating membranes in blocking buffer, conjugated/blocking buffer, washing buffer, substrate equilibration buffer, and substrate working solution, the membrane was finally exposed to Biomax MR film (Kodak, Rochester, NY). The sequences used for Egr-1 probe are CGTCAGCTCCCCCGCGATCCAC and GCAGTCGAGGGGGCGCTAGGTG.

Quantitative RT-PCR.

RNA was extracted from cells using the TRIzol method (Invitrogen) and was converted to cDNA using the high-capacity cDNA archive kit (Applied Biosystems). Quantitative RT-PCR was performed as described previously (21), whereas probes and primers for Cyr61 and TaqMan Master Mix for gene expression assays were obtained from Applied Biosystems. Gene expression was analyzed by the ΔΔCt method, with 18s rRNA and βGUS as the endogenous control.

Statistical analysis.

All values were expressed as means ± SD from at least three independent experiments. Differences in measured variables between experimental and control groups were assessed by using the Student's t-test (Statview II statistical package; Abacus Concepts, Berkeley, CA). Statistically significant difference was accepted at P < 0.05.

RESULTS

Deletion of cav-1 protected against hyperoxia-induced apoptosis in vitro.

Our previous studies demonstrated that deletion of cav-1 protected against hyperoxia-induced cell death (22). Previous studies published by others have established that hyperoxia induces cell death via both necrosis and apoptosis (47). In this study, we showed that deletion of cav-1 protected murine fibroblasts from hyperoxia-induced apoptosis ( Fig. 1). In wild-type cells, hyperoxia induced caspase-3 activity in a time-dependent manner; these activities were blunted, however, in cav-1^−/− fibroblasts. To verify our observation in epithelial cells, we transfected human bronchial Beas-2B cells with scramble control siRNA and cav-1 siRNA and determined caspase-3 activity following 48 h of hyperoxia. Our results (Fig. 1B) demonstrated that cells transfected with control siRNA exhibited significantly elevated caspase-3 activity, which was not observed in cells transfected with cav-1 siRNA.

Fig. 1.

Deletion of caveolin-1 (cav-1) protected hyperoxia-induced apoptosis. Cells were treated with hyperoxia (time course), and caspase-3 activity was measured as described in materials and methods. A: wild-type and cav-1^−/− fibroblasts were exposed to hyperoxia. Caspase-3 activity was measured. A time course was determined. Open bars, wild-type cells; filled bars, cav-1^−/− cells. B: Beas-2B cells were treated with cav-1 siRNA and then exposed to hyperoxia. After 48 h, caspase-3 activity was measured. Open bars, cells exposed to room air; filled bars, cells exposed to hyperoxia. Three independent assays are represented. *+#P < 0.05.

Deletion of cav-1 augmented survivin expression both in vitro and in vivo.

First, we found that hyperoxia suppressed survivin expression in Beas-2B cells in a time-dependent manner (Fig. 2A). We then explored survivin expression in wild-type and cav-1^−/− fibroblasts. Interestingly, we found that deletion of cav-1 (cav-1^−/− fibroblasts) increased survivin expression at basal levels and after hyperoxia (Fig. 2B). In contrast to decreased survivin expression in wild-type cells following hyperoxia, we observed that hyperoxia had no effects on survivin expression in cav-1^−/− fibroblasts (Fig. 2B). To determine if our observation was not exclusive to fibroblasts, we compared survivin expression with and without hyperoxia in homogenized mouse lung tissue. Again, hyperoxia suppressed survivin in a time-dependent manner in wild-type mice, whereas this effect was significantly reduced in cav-1^−/− mice (Fig. 2C).

Fig. 2.

Effect of hyperoxia on survivin expression. A: Beas-2B cells were exposed to hyperoxia. At each time point, cell lysates were collected, analyzed by Western blot analysis, and blotted with anti-survivin or β-actin as described in materials and methods. B: mouse lung primary fibroblasts were exposed to hyperoxia. After 24 h, cell lysates were collected, and Western blot analyses were performed using anti-survivin, anti-cav-1, or β-actin. C: C57BL/6 mice were exposed to hyperoxia. At each time point, mice lung tissue was collected and homogenized, and Western blot analyses were performed as above. All blots represent at least 3 repeats.

Survivin protected cells from hyperoxic cell death and apoptosis in vitro.

We next explored whether survivin played a role in hyperoxic cell death and apoptosis by “loss” and “gain” of function assays. Initially, we knocked down survivin by transfection of survivin siRNA. Cell viability was evaluated in cells transfected with survivin siRNA after 48 h of hyperoxia (48 h). As shown in Fig. 1A, knocking down survivin significantly enhanced the hyperoxic cell death in Beas-2B cells (Fig. 3A) and mouse lung fibroblasts (data not shown). We confirmed our initial observation by overexpressing survivin in Beas-2B cells. At 48 h posttransfection, cells overexpressing survivin and cells transfected with control vectors were exposed to hyperoxia as described previously. Cell viability was measured using both CellTiter-Glo and trypan blue assays. Significant resistance to hyperoxic cell death was found in cells overexpressing survivin (Fig. 3B), with increased cell proliferation (data not shown).

Fig. 3.

Effect of survivin on hyperoxia-induced cell death and apoptosis. The effect of survivin on hyperoxia-induced cell death was evaluated by 1) loss of function assay using siRNA and 2) gain of function assay using survivin overexpressing vectors. A: Beas-2B human bronchial epithelial cells were treated with control siRNA or survivin siRNA. After hyperoxia (48 h), cell viability was measured using CellTiter-blue assay as described in materials and methods. Inset: Western blot analysis indicates successful knockdown of survivin. B: Beas-2B cells were transfected with control vector and survivin overexpressing vector. After hyperoxia (48 h), cell viability was measured as above. Inset: Western blot analysis indicates successful overexpression of survivin. Open bar, cells exposed to room air; filled bar, cells exposed with hyperoxia. *#P < 0.05. Three similar experiments are represented. The effect of survivin on hyperoxia-induced caspase-3 activation was evaluated by 1) loss of function assay using siRNA and 2) gain of function assay using survivin overexpressing vectors, as described in materials and methods. C: Beas-2B cells were treated with control siRNA and survivin siRNA. After hyperoxia (48 h), caspase-3 activity was measured using a colormetric assay (Calbiochem, Gibbstown, NJ) as described in materials and methods. Inset: Western blot analysis indicates successful knockdown of survivin. D: Beas-2B cells were transfected with control vector and survivin overexpressing vector. After hyperoxia (48 h), caspase-3 activity was measured as above. Inset: Western blot analysis indicates successful overexpression of survivin. Open bars, cells exposed to room air; filled bars, cells exposed with hyperoxia. *#P < 0.05. Three similar experiments are represented. E: the effect of cav-1 deficiency in survivin mediated survival after hyperoxia was evaluated. Open bars, wild-type fibroblasts; filled bars, cav-1^−/− cells. Cells were transfected with control and survivin siRNA, respectively, and then exposed to hyperoxia. After 48 h, cells were evaluated for caspase-3 activity. *#P < 0.05. At least 3 repeats are represented.

Survivin protected cells against hyperoxia-induced apoptosis.

To examine the role of survivin in hyperoxia-induced apoptosis, we measured caspase-3 activity. As shown in Fig. 3C, caspase-3 activity was induced 48 h after hyperoxia in Beas-2B cells. Cells transfected with survivin siRNA exhibited higher caspase-3 activity compared with cells transfected with control siRNA. Furthermore, overexpression of survivin in these cells blunted the elevation of caspase-3 activity induced by hyperoxia, shown in Fig. 3D. We further evaluated whether survivin overexpression prevents the release of cytochrome c. First, we overexpressed survivin in Beas-2B cells. After hyperoxia, the release of cytochrome c was examined. Survivin overexpression did not significantly modulate cytochrome c release (data not shown).

We next examined whether the elevated survivin in cav-1^−/− cells was responsible for the observed resistance to hyperoxic apoptosis in these cells. As illustrated in Fig. 3E, wild-type and cav-1^−/− cells were transfected with control siRNA or survivin siRNA and then exposed to hyperoxia. As expected, hyperoxia induced caspase-3 activity in wild-type cells treated with control siRNA. Transfecting cells with survivin siRNA further augmented this effect. In cav-1^−/− cells, hyperoxia-induced apoptosis was reduced, as shown previously. Notably, on day 2 and day 3, the protective effects attributed to an absence of cav-1 were abolished by transfecting survivin siRNA (Fig. 3E).

Hyperoxia regulated interaction between survivin and caspase-3 in cav-1^−/− cells.

After determining that survivin protected against hyperoxia-induced apoptosis, we sought to examine the mechanism by which survivin mediated this cytoprotective effect. We found that in cav-1^−/− cells, hyperoxia increased the physical binding between caspase-3 and survivin at 4 h after hyperoxia (Fig. 4, A and B). Interestingly, at 12 h, survivin expression was significantly decreased, as shown in Fig. 2A.

Fig. 4.

Interaction and colocalization of caspase-3 and survivin, with and without hyperoxia. A: colocalization between survivin and caspase-3 in wild-type and cav-1^−/− fibroblasts. Cells were exposed to hyperoxia. After 4 h, cells were fixed in 2% paraformaldehyde followed by immunostaining with anti-caspase-3 (rabbit polyclonal antibodies, Santa Cruz) and anti-survivin (mouse monoclonal antibodies, Santa Cruz). Cells were further stained with Hoechst for nucleus definition. Red, survivin; green, caspase-3; blue, nucleus; yellow, merge. Red arrow, colocalized caspase-3 and survivin. At least 3 identical experiments are represented. B: cell lysates were immunoprecipitated with rabbit polyclonal anti-caspase-3 antibodies and then blotted with mouse monoclonal anti-survivin antibodies. Blot represented 3 independent experiments.

Hyperoxia suppressed survivin expression by downregulation of mRNA.

Next, we sought to determine the mechanisms by which hyperoxia suppressed survivin expression. By Taqman real-time PCR assay, hyperoxia suppressed survivin production via transcriptional regulation, as evidenced by a decrease in survivin mRNA (Fig. 5A) in wild-type fibroblasts. Similar results were found in human epithelial Beas-2B cells (data not shown). Conversely, decreases in mRNA survivin levels were not observed in cav-1^−/− cells following hyperoxia (Fig. 5A).

Fig. 5.

Effect of hyperoxia on the transcriptional regulation of survivin. A: Beas-2B cells were exposed to room air and hyperoxia. After 24 h, RNA was collected, and reverse transcription was performed as described in materials and methods. Taqman real-time PCR was then performed using the manufacturer's protocol. Relative folds of mRNA transcription were indicated. *+#P < 0.05. At least 3 experiments are represented. B: effect of cav-1 deficiency on Egr-1 binding activities by EMSA, with and without hyperoxia. Wild-type and cav-1⁻/⁻ fibroblasts were exposed to room air or hyperoxia. After 4 h, nucleus extract was determined for protein concentration, and binding activity to Egr-1 consensus sequence was determined as described in materials and methods. At least 3 independent experiments are represented. C: expression and translocation of Egr-1 into cell nucleus after hyperoxia in wild-type and cav-1^−/− fibroblasts. Wild-type and cav-1^−/− fibroblasts were exposed to hyperoxia. After 4 h, cell nucleus was extracted, and Western blot analysis was used to determine Egr-1 translocation. Blots represent at least 3 independent experiments.

Previously published studies indicate that Egr-1 is a negative regulator in the promoter region of survivin (46). We evaluated whether hyperoxia affected Egr-1 binding activity and whether deletion of cav-1 interfered with this process. First, hyperoxia increased Egr-1 binding activity in a time-dependent manner in Beas-2B cells (data not shown). Next, we further found that deletion of cav-1 (cav-1^−/−) decreased Egr-1 binding activity at both the basal level and after hyperoxia (Fig. 5B). In addition, there was a higher Egr-1 nuclear translocation in wild-type cells compared with cav-1^−/− cells (Fig. 5C).

Egr-1 is a negative regulator of survivin expression in lung cells.

Given that an absence of cav-1 (cav-1^−/−) decreased Egr-1 binding activity at both the basal level and after hyperoxia, and that cav-1^−/− cells exhibited elevated survivin expression, we explored the possibility that Egr-1 negatively regulated survivin expression in these cells. As shown in Fig. 6A, using Egr-1 knockout cells (Egr-1^−/−), we found elevated survivin expression. We then determined whether hyperoxia affected Egr-1 levels. As shown in Fig. 6B, hyperoxia induced Egr-1 in a time-dependent manner. To confirm if Egr-1 negatively regulated survivin mRNA, we treated Beas-2B cells with scrambled control siRNA or Egr-1 siRNA and then exposed these cells to hyperoxia. After 8 h, survivin mRNA was evaluated by Taqman real-time PCR as described previously. Figure 6C showed that knocking down Egr-1 by siRNA enhanced survivin mRNA transcription after hyperoxia. Given that Egr-1 negatively regulated survivin expression and survivin protected hyperoxia-induced cell death, we were interested in examining the viability of Egr-1^−/− cells after hyperoxia. As shown in Fig. 6D, as expected, Egr-1^−/− cells resisted hyperoxia-induced cell death compared with wild-type cells. Furthermore, we studied whether an antioxidant prevented the hyperoxia-induced upregulation of Egr-1 in wild-type cells. In Beas-2B cells pretreated with 20 mM N-acetyl-cysteine, hyperoxia-induced Egr-1 expression was significantly reduced compared with untreated cells (Fig. 6E).

Fig. 6.

Effect of hyperoxia on Egr-1 expression and effect of Egr-1 deficiency on survivin expression. A: effect of Egr-1 deficiency on survivin expression. Wild-type and Egr-1^−/− fibroblasts were exposed to hyperoxia. After 24 h, cell lysates were collected and blotted with anti-survivin antibodies. B: effect of hyperoxia on Egr-1 expression. Beas-2B cells were exposed to hyperoxia. After indicated time of exposure, cell lysates were analyzed by Western blot to evaluate Egr-1 expressions. All blots represent at least 3 independent experiments. C: hyperoxia upregulated Egr-1 mRNA. Beas-2B cells were exposed to room air and hyperoxia. After 24 h, RNA was collected, and reverse transcription was performed as described in materials and methods. Taqman real-time PCR was then performed using the manufacturer's protocol. Relative folds of mRNA transcription were indicated. *P < 0.05. At least 3 experiments are represented. D: effect of Egr-1 deficiency on cell viability after hyperoxia. Wild-type and Egr-1^−/− fibroblasts were exposed to hyperoxia. After 48 h, cell viability was evaluated as described in materials and methods. *P < 0.05. Three independent assays are represented. E: effect of N-acetyl-cysteine (NAC) on hyperoxia-induced Egr-1 expression. Beas-2B cells were pretreated with 20 mM NAC, followed by exposure to hyperoxia. After 12 h, cell lysates were analyzed by Western blot to evaluate Egr-1 expressions. Blot represents at least 3 independent experiments.

We next examined the survivin-caspase-3 interaction in Egr-1^−/− cells. Colocalization and interaction of survivin-caspase-3 in Egr-1^−/− cells were higher than those in wild-type cells (Fig. 7).

Fig. 7.

Effect of Egr-1 deficiency (Egr-1^−/−) on colocalization and interaction between caspase-3 and survivin with and without hyperoxia. Wild-type and Egr-1^−/− cells were exposed to hyperoxia. After 4 h, cells were analyzed by confocal microscopy or co-IP assays as described in materials and methods. A: colocalization of survivin and caspase-3 in the absence and presence of hyperoxia. Red, survivin; green, caspase-3; yellow, merge. At least 3 identical experiments are represented. B: cell lysates were immunoprecipitated with rabbit polyclonal anti-caspase-3 antibodies and then blotted with mouse monoclonal anti-survivin antibodies. Blot represented 3 independent experiments.

DISCUSSION

Cell death including apoptosis, necrosis, and oncosis, are features commonly found in ALI and are thought to be one of the mechanisms responsible for ALI/ARDS (10, 28, 30, 44, 47). Our group has previously shown that deletion of cav-1 protects against ALI. Deletion of cav-1 protected both hyperoxia-induced lung injury in vivo and hyperoxic cell death in vitro (22). The mechanisms of this protective effect remain unexplored. In this study, for the first time, we show that deletion of cav-1 protected against hyperoxic apoptosis by modulating the caspase-dependent pathways via upregulation of survivin. Survivin protected hyperoxia-induced cell death by binding to caspase-3. However, this protective effect is likely to delay rather than prevent cell death after hyperoxia. This is based on our observation that overexpressing survivin failed to reduce the release of cytochrome c after hyperoxia (data not shown).

We examined the mechanisms by which deletion of cav-1 induced survivin expression. The early growth response gene (Egr-1) is activated by hyperoxia (24) and has a binding site in the promoter region of survivin (46). Our results indicated that survivin was negatively regulated by Egr-1 after hyperoxia. Deletion of cav-1 reduced Egr-1 nuclear translocation and binding activity, resulting in upregulation of survivin. This was consistent with previous reports showing that survivin is downregulated following exogenous expression of wild-type Egr-1 (46).

Survivin protected against hyperoxia-induced apoptosis. Deletion of cav-1 protected apoptosis via upregulation of survivin by blocking the nuclear translocation of Egr-1 after hyperoxia and subsequently decreasing Egr-1 binding activity. Egr-1 was a negative regulator of survivin and was induced by hyperoxia in a time-dependent manner (Fig. 5C). Hyperoxia induced both Egr-1 translocation and expression. This can partially explain why hyperoxia decreased survivin expression as early as after 12 h of exposure. Besides transcriptional regulation, regulation of survivin can occur at various levels, including protein degradation and intracellular sequestration (17, 33, 43, 48). Whether cav-1 is involved in these levels of regulation requires further exploration.

Several other questions remain to be clarified. First, besides Egr-1, there are many other transcriptional factors involved in survivin regulation (17, 33, 43, 48), such as signal transducer and activator of transcription 3 (STAT3) family (43), the NF-κB (34), the phosphatidylinositol 3-kinase/Akt pathway (37), and insulin-like growth factor I/mTOR signaling (48). These pathways have been reported to upregulate survivin via regulation of mRNA translation (17, 24, 33, 37, 43, 48). The Ras oncogene family and the antiapoptotic factor Wnt-2 (41) also have been reported to play roles in regulating survivin expression. Furthermore, survivin has been reported to be repressed at the transcriptional level by wild-type p53 (29). Whether hyperoxia and cav-1 have any interactions with these pathways to regulate survivin requires further investigation. In addition, previous studies have shown that the mitogen-activated protein kinase kinase (MEK)/extracellular signal-regulated kinase (ERK), p38 MAPK, and PI3-kinase pathways contributed to the activation of Egr-1 in response to hyperoxia (4, 7, 38, 49). These pathways are likely to regulate cav-1-mediated apoptosis and downregulation of survivin, given that ERK/PI3-kinase are altered in cav-1^−/− (12). Second, does deletion of cav-1 and hyperoxia affect other IAPs, such as XIAP, IAP1, or IAP2? To address this question will help to illustrate whether survivin plays a unique role in hyperoxia-induced apoptosis and further help to determine whether survivin will be a specific target for ALI therapy. Third, cav-1 plays a relatively central role in regulating cell death and survival (35); thus, upregulation of survivin may be one, but not the only, mechanism involved in cav-1-mediated apoptosis after hyperoxia.

While we are focusing on cell survival and death, another aspect is cell proliferation. Previous reports have raised a theory that growth arrest is an important defense against oxidative damage (31). Lung cancer cells arrested in G₁ phase had less DNA injury and higher cell survival (31). Interestingly, the synthesis and degradation of survivin has been reported to be modulated in a cell cycle-dependent manner. Survivin transcription is most active in G₂-M phase. It was reported to stabilize the microtubules and contribute to bipolar spindle formation (3, 8, 27). Survivin has important roles in mitotic regulation (3), and its synthesis and degradation is tightly controlled by cell cycle (3, 8, 27). Therefore, the major known functions of survivin are the regulation of cell cycle and the inhibition of apoptosis (3). More interestingly, both hyperoxia and cav-1 have been reported to be associated with cell cycle. Cav-1 causes cell cycle arrest at G₁/S phase and so does hyperoxia (15, 18). While in G₁/S phase, less survivin transcription occurred. Therefore, hyperoxia and cav-1 may control cell cycles via regulating survivin.

Last, our current work is conducted in human bronchial epithelial cells and mouse primary fibroblasts. Whether the same observation will be found in other types of cells remains unexplored. Published data (16, 20, 22) and our previous work had indicated that deletion of cav-1 protects from hyperoxia-induced lung injury (22). Although survivin seems to mediate the protective effect of cav-1^−/− in hyperoxia-induced apoptosis in vitro, the importance of survivin in hyperoxia-induced lung injury in vivo remains unclear.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grant 5K08-HL-085601-02 (Y. Jin) and an ATS unrestricted research grant (Y. Jin).