Abstract
The inhibitory effect of caveolin on the cellular response to growth factor stimulation is well established. Given the significant overlap in signaling pathways involved in regulating cell proliferation and stress responsiveness, we hypothesized that caveolin would also affect a cell's ability to respond to environmental stress. Here we investigated the ability of caveolin-1 to modulate the cellular response to sodium arsenite and thereby alter survival of the human cell lines 293 and HeLa. Cells stably transfected with caveolin-1 were found to be much more sensitive to the toxic effects of sodium arsenite than either untransfected parental cells or parental cells transfected with an empty vector. Unexpectedly, the caveolin-overexpressing cells also exhibited a significant activation of the phosphatidylinositol 3-kinase (PI3K)/Akt pathway, which additional studies suggested was likely due to decreased neutral sphingomyelinase activity and ceramide synthesis. In contrast to its extensively documented antiapoptotic influence, the elevated activity of Akt appears to be important in sensitizing caveolin-expressing cells to arsenite-induced toxicity, as both pretreatment of cells with the PI3K inhibitor wortmannin and overexpression of a dominant-negative Akt mutant markedly improved the survival of arsenite-treated cells. This death-promoting influence of the PI3K/Akt pathway in caveolin-overexpressing cells appeared not to be unique to sodium arsenite, as wortmannin pretreatment also resulted in increased survival in the presence of H2O2. In summary, our results indicate that caveolin-induced upregulation of the PI3K/Akt signaling pathway, which appears to be a death signal in the presence of arsenite and H2O2, sensitizes cells to environmental stress.
Caveolins, 22- to 24-kDa integral membrane proteins (28, 30), are the main structural components of non-clathrin, flask-shaped invaginations named caveolae (little caves), originally described by Palade et al. and Yamada in the 1950s (24, 41). There are four caveolin isoforms present in mammalian cells: caveolin-1α, caveolin-1β, caveolin-2, and caveolin-3. While caveolin-3 is found mainly in muscle cells, caveolin-1 and -2 are coexpressed in most cells and share many physical properties, although the latter seems to lack full functional capacity to form caveolae (30). These specialized plasmalemmal microdomains, which are particularly abundant in endothelial, epithelial, and smooth muscle cells, are enriched in specific lipids (glycosphingolipids, sphingomyelin, and cholesterol) and important lipid-modified signaling molecules (2, 31).
Caveolae are thought to serve as sites for the gathering of signaling complexes, thereby facilitating the initiation and cross-talk of signaling events (23, 32). In fact, caveolin can modulate the function of many signal transducers integrated in caveolae, including epidermal growth factor (EGF) receptor, platelet-derived growth factor (PDGF) receptor, insulin receptor, Shc, Grb-2, mSOS-1, endothelial nitric oxide synthase (eNOS), H-Ras, Raf, MEK, and extracellular signal-regulated kinase 1 and 2 (ERK1/2) (31), by binding to their caveolin-binding motifs, an event that may occur preferentially when these molecules are in their inactive state, as is the case for H-Ras, Gα subunits, and Src (23). It has been suggested that oligomerized caveolin complexes (mainly 200 to 350 kDa) function to concentrate caveolin-interacting signaling molecules within caveolae (29, 33). Since its effects on signaling molecules are often inhibitory, caveolin is often viewed as a general kinase inhibitor or even a negative growth regulator (32). In support of this notion is the finding that caveolin expression is associated with reduced cell proliferation and is upregulated in senescent cells and tissues (26), which also exhibit a diminished ability to trigger signaling cascades in response to both growth factors and stressful stimulation (12, 13). Conversely, transformation results in the reduction of both caveolin and caveolae (15). Furthermore, the gene encoding caveolin-1 has been located within a locus deleted in many types of cancer that harbors a potential tumor suppressor gene (15, 27).
Stress signaling cascades share many early signaling components with those regulating the response to growth factors anchored in caveolae microdomains (6, 31, 32). The ERK and the phosphatidylinositol-3 kinase (PI3K)/Akt survival pathways are two such cascades, and their activation is often found to be a critical step for a favorable cellular outcome upon stressful stimulation (7, 9, 21, 38, 39). Caveolin has been shown to inhibit ERK activation by growth factors (10, 11). Moreover, treatment with the peptide corresponding to the caveolin scaffolding domain alone is sufficient to effectively inhibit EGF-induced signaling through the mitogen-activated protein (MAP) kinase cascade in vivo and in vitro (10). The influence of caveolin on the PI3K/Akt pathway is less understood. However, PI3K has been localized in caveolae in fibroblasts, endothelial cells, and myeloid-derived cells (16, 19, 42). This finding suggests that PI3K activity might be controlled by caveolin under certain conditions.
Here we have examined the influence of caveolin expression on the cellular response to treatment with the tumor promoter arsenite (4, 5, 14). Our results indicate that caveolin overexpression leads to an increase in Akt activation, possibly through the diminished synthesis of ceramide, a lipid that was previously shown to inhibit PI3K activity (43). Remarkably, this upregulation of Akt signaling in caveolin-overexpressing cells is found to promote cell death rather than to favor survival. These observations suggest the existence of a novel Akt-mediated death pathway that is dependent on high caveolin expression.
MATERIALS AND METHODS
Cell lines, reagents, and transfection.
The human cell lines 293 (kidney fibroblasts) and HeLa (cervical carcinoma) were maintained in high-glucose Dulbecco's modified Eagle medium (Invitrogen Life Technologies, Carlsbad, Calif.) supplemented with 10% fetal bovine serum (HyClone, Logan, Utah), 2 mM l-glutamine, HEPES (pH 7.4), and antibiotics. Cells were serum starved for 16 h and, unless otherwise indicated, were treated with 25 μM arsenite (Sigma, St. Louis, Mo.) for various time periods before being harvested for analysis. C2 ceramide and the inhibitors AG1296 (for PDGF receptor), AG1478 (for EGF receptor), suramin (for various growth factor receptors), U0126 (for MEK1/2), LY294002, and wortmannin (for PI3K) were purchased from Calbiochem-Novachem (San Diego, Calif.), and 14C-sphingomyelin was obtained from Perkin Elmer Life Sciences (Boston, Mass.). Cells were transfected with pcDNA3.1-caveolin-1 (generously provided by S. C. Park; see reference 26) by using Fugene6 (Roche Diagnostics Corp., Indianapolis, Ind.) following the manufacturer's instructions and were selected with 0.6 mg of G418/ml. Plasmid pUSE(K179M) was from Upstate Biotechnology, Inc. (Lake Placid, N.Y.).
Western blot analysis.
Cells were washed with cold phosphate-buffered saline (PBS) and were harvested in a lysis buffer containing 20 mM HEPES (pH 7.4), 2 mM EGTA, 50 mM β-glycerophosphate, 1 mM sodium vanadate, 5 mM sodium fluoride, 1% Triton X-100, 10% glycerol, 2 mM dithiothreitol, 2 μg of leupeptin/ml, and 2 μg of aprotinin/ml, vortexed extensively, and centrifuged at 25,000 × g for 15 min. Fifty-microgram protein aliquots were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 10% acrylamide) and were transferred onto polyvinylidene difluoride membranes (Millipore Corporation, Bedford, Mass.). Membranes were blocked for 1 h in TBST buffer (0.8% sodium chloride, 10 mM Tris [pH 8], and 0.07% Tween 20) containing 5% nonfat milk (Santa Cruz Biotechnology, Santa Cruz, Calif.) and then were incubated with antibodies recognizing caveolin-1 (BD Transduction Labs, San Diego, Calif.), Akt, p-Akt, p-JNK, p-p38, p-GSK-3 α/β (Cell Signaling Technology, Beverly, Mass.), p-ERK (Promega, Madison, Wis.), PTEN (Cascade Bioscience, Winchester, Mass.), and β-actin (Santa Cruz Biotechnology). After incubation with appropriate secondary antibodies, immunoreactive signals were visualized by using enhanced chemiluminescence (Perkin Elmer Life Sciences, Boston, Mass.).
Cell viability and colony survival assay.
Actively growing 293 and HeLa cells were treated with 5 or 10 μM arsenite for 24 h, whereupon all cells were replated at different dilutions for clonogenicity assay. Two weeks later colonies were fixed and stained with a solution containing 10% methanol and 0.4% crystal violet (Sigma); colonies with 50 or more cells were counted. In order to confirm if these differences in colony formation were due to true changes in survival and not merely a cytostatic effect, we treated cells with arsenite (25 μM) and stained them 24 h later with trypan blue.
Cellular fractionation and quantitation of membrane cholesterol.
Actively growing, subconfluent cells were washed with PBS and scraped into a small volume of cold PBS containing protease inhibitors. After three cycles of freezing/thawing in a solution containing 50 mM Tris (pH 7.4) and 0.5 mM EDTA, membrane fractions were separated from soluble fractions by centrifugation at 21,000 × g for 20 min and were washed three times with cold PBS. Membrane pellets were then assayed for cholesterol by using the Amplex Red Cholesterol Assay kit from Molecular Probes (Eugene, Oreg.). Membrane pellets were solubilized by sonication with a Cell Disruptor 350 (Sonifier) on ice for a total of 10 cycles per sample. Membranes were then centrifuged at 21,000 × g for 1 min, and supernatants were assayed according to the manufacturer's instructions. The fluorescence produced during the reaction, due to the presence of both free cholesterol and cholesterol esters, was measured with a Fluoroscan Ascent microplate reader (Labsystems) with excitation at 530 nm and fluorescence detection at 590 nm.
Akt kinase assay.
Akt activity was measured as previously described (39). Briefly, cells were rinsed with ice-cold PBS, lysed for 20 min at 4°C in lysis buffer (described in the Western blot analysis section), and centrifuged to remove insoluble material. Akt was immunoprecipitated with a solution containing goat polyclonal anti-Akt antibody (Santa Cruz Biotechnology) and 30 μl of a 50% slurry of protein G-Sepharose (Amersham Pharmacia Biotech, Piscataway, N.J.) for 4 h. The beads containing the immunoprecipitates were washed three times with lysis buffer, once with ice-cold water, and once with Akt kinase buffer (20 mM HEPES [pH 7.4], 10 mM MgCl2, 10 mM MnCl2). To measure kinase activity, immunoprecipitated Akt was added to 30 μl of kinase buffer containing 10 μCi of [γ-32P]ATP and 2 μg of histone H2B (Roche Molecular Biochemicals) as a substrate. 32P-labeled protein was separated by SDS-14% PAGE and was detected by a PhosphorImager (Molecular Dynamics, Sunnyvale, Calif.).
nSMase enzymatic assay.
Neutral sphingomyelinase (nSMase) activity in 293-V and 293-Cav25 cells was evaluated by following a method previously described (36), with some modifications. Briefly, actively growing cells were harvested by being scraped in cold PBS with protease inhibitors and 1 mM EDTA. Cells were centrifuged, and cell pellets were resuspended in 400 μl of a solution containing 0.5 mM Tris (pH 7.4) and 1 mM EDTA with protease inhibitors, vortexed, and subjected to 15 strokes of Dounce homogenization on ice. Forty microliters of 1% Triton X-100 and 40 μl of 2 M NaCl were then added to homogenates and vortexed. After 20 min on ice, lysates were centrifuged for 1 min, protein concentrations were determined, and enzyme activity was assessed in 200 μl of lysate plus 200 μl of substrate buffer (0.1% Triton X-100, 200 mM Tris [pH 7.5], 10 mM dithiothreitol, and 10 mM MgCl2). Reactions were carried out following the addition of 5 μl of 0.02 mCi of 14C-sphingomyelin (Perkin Elmer Life Sciences)/ml at 37°C. Two-hundred-microliter aliquots were taken after 20 and 40 min, and reactions were stopped by adding 400 μl of methanol-chloroform (1:1) followed by lipid extraction. The amount of radioactive phosphocholine formed was measured by subjecting the upper phase to scintillation counting.
Statistical analysis.
An unpaired Student's t test was used to assess differences between two groups. A P value of <0.05 was considered significant.
RESULTS
Caveolin-1 overexpression leads to elevated p-Akt.
To investigate the role of caveolin on stress signaling, 293 human kidney fibroblasts and HeLa human cervical carcinoma cells were stably transfected to overexpressed caveolin-1 (hereafter referred to as caveolin) (26), and multiple clones were isolated. Caveolin overexpression in 293 cells, which led to an increase in both monomeric (22-kDa) and oligomerized (ranging from 200- to 350-kDa) forms, resulted in elevated basal Akt phosphorylation as detected by Western blotting (Fig. 1). Interestingly, this correlation was particularly evident for the high-molecular-weight, oligomerized complexes, which are thought to be responsible for the concentration of caveolin-interacting molecules within caveolae (29). Clones 293-Cav2 and 293-Cav33, which expressed undetectable levels of caveolin, displayed minimal Akt phosphorylation, while 293-Cav25 and 293-Cav31 exhibited both highest caveolin and highest phosphorylated Akt levels. Thus, we chose 293-Cav25 and 293-Cav31 cells for subsequent analysis.
FIG. 1.
Caveolin overexpression in 293 cells leads to elevation in phospho-Akt. 293 Cells stably transfected to overexpress caveolin-1 were clonally selected by using G418, and phospho(p)-Akt levels were assessed by Western blotting. Shown are nine clonal isolates along with pooled 293 cells transfected with an insertless vector (pcDNA3.1). Fifty-microgram protein aliquots were resolved in 10% polyacrylamide SDS gels, transferred onto polyvinylidene difluoride membranes, and hybridized by using an anti-caveolin-1 antibody that recognizes both the monomeric and oligomeric forms of caveolin-1. Membranes were stripped and hybridized with an antibody recognizing phosphorylated Akt (p-Akt), and actin is shown as an internal control.
Since caveolin is know to bind and transport cholesterol into caveolae and since the presence of this lipid is key for caveolae function, we measured cholesterol levels in the plasma membranes of 293-V, 293-Cav25, and 293-Cav31 by using the fluorometric assay described in Materials and Methods. Membrane cholesterol was significantly increased, proportionally to the levels of caveolin (2.37 ± 0.73 ng of cholesterol/mg of protein for 293-V and 3.26 ± 0.74 ng of cholesterol/mg of protein for 293-Cav31), and was particularly elevated in the 293-Cav25 population (4.02 ± 0.36 ng of cholesterol/mg of protein), indicating that caveolin is likely to be more active in these cells.
Increased Akt phosphorylation is associated with reduced ceramide synthesis in caveolin-expressing cells.
Caveolae microdomains are highly enriched in lipids like cholesterol and sphingolipids, including sphingomyelin and ceramide (18, 35). Ceramide, a lipid synthesized by the cleavage of sphingomyelin by SMase, has been shown to effectively inhibit PI3K/Akt activity (43). Interestingly, nSMase activity is believed to be inhibited by the caveolin-scaffolding domain (36). Thus, we sought to test the hypothesis that caveolin overexpression in our model system inhibits nSMase activity, thereby diminishing the synthesis of ceramide and allowing the subsequent increase in PI3K/Akt basal activity. To this end, we first tested if the PI3K/Akt pathway was susceptible to inhibition by ceramide in the caveolin-transfected 293 cells. As shown in Fig. 2A and B, time- and dose-dependent decreases in Akt phosphorylation were seen in 293-Cav25 cells treated with C2 ceramide, while the levels of total Akt protein remained unchanged. Comparison of nSMase activity in a pooled population of vector (pcDNA3.1)-transfected cells (293-V) and in cells overexpressing caveolin (293-Cav25) revealed a 40 to 50% reduction in the cleavage of sphingomyelin (ceramide precursor) in 293-Cav25 compared to that attained with 293-V cells (Fig. 2C), with Km values of 8.3 and 2.6 μM, respectively. These observations support the view that heightened caveolin expression could contribute to elevating phosphorylated Akt by lowering the levels of ceramide, a natural inhibitor of the PI3K/Akt pathway.
FIG. 2.
Inhibition of ceramide synthesis by caveolin likely contributes to elevating Akt phosphorylation. (A) Time-dependent reduction of Akt phosphorylation in 293-Cav25 cells treated with 40 μM C2 ceramide. (B) Dose-dependent reduction of Akt phosphorylation in 293-Cav25 cells treated for 2 h with the indicated doses of C2 ceramide. Levels of Akt protein are also shown as an internal control. (C) nSMase (SM) enzymatic activity in 293-V and 293-Cav25 cells was assessed by monitoring cleavage of radiolabeled sphingomyelin, the precursor of ceramide, as described in Materials and Methods.
Arsenite-triggered activation of PI3K/Akt in 293 cells overexpressing caveolin.
Caveolin has been shown to modulate the activity of key components of several growth factor-initiated signal transduction pathways, particularly the MEK/ERK signaling cascade (10, 11). To further evaluate the role of caveolin in stress-related Akt signaling, clones 293-Cav25 and 293-Cav31 as well as 293-V cells were exposed to 25 μM arsenite, and Akt phosphorylation was assessed by Western blotting (Fig. 3A). A moderate time-dependent activation of Akt was observed in 293-V cells, while arsenite treatment caused no further Akt activation in 293-Cav25 and 293-Cav31 cells. No significant changes in the levels of total Akt protein were seen, indicating that the phospho-Akt levels are not due to uneven loading or to differences in the abundance of total Akt protein. We also evaluated the phosphorylation status of GSK-3 α/β, a known substrate of Akt. Indeed, the phosphorylation of GSK-3 correlated directly with the Akt phosphorylation status, supporting the view that the Akt pathway is functionally active. Similar results were seen with another Akt target, the forkhead transcription factor (data not shown). Akt activity was also directly assessed in 293-V and 293-Cav25 cells by kinase assay (Fig. 3B), which confirmed the Western blot analysis results shown in Fig. 3A. The effect of caveolin overexpression was also tested on other important stress signaling pathways, including ERK, JNK, and p38. As shown in Fig. 3A and in agreement with previous reports, ERK phosphorylation was significantly diminished by caveolin (approximately fivefold), particularly in 293-Cav25 cells, while we observed little or no change in the activation of JNK and p38.
FIG. 3.
Caveolin-induced changes in Akt and MAP kinase signaling pathways after exposure of 293 cells to arsenite. (A) 293-V, 293-Cav25, and 293-Cav31 cells were treated with 25 μM arsenite for the indicated times, whereupon Western blot analysis was carried out to assess the levels of oligomerized caveolin, monomeric caveolin, p-Akt, Akt, p-GSK-3 α/β, p-ERK, p-JNK, and p-p38 in 50 μg of whole-cell lysate. (B) 293-V and 293-Cav25 cells were harvested, and Akt was immunoprecipitated from lysates with a polyclonal anti-Akt antibody. Akt kinase activity was assessed by using histone H2B as a substrate.
In order to identify the upstream mediators of Akt activation, we first evaluated whether pretreatment with inhibitors of various growth factor receptors (AG1296, AG1478, Suramin), MEK1/2 (U0126), and PI3K (LY294002) (37) affected Akt phosphorylation in 293-V and 293-Cav25 cells. As shown in Fig. 4A, Akt phosphorylation was not changed by pretreatment with any of the inhibitors at doses previously shown to be effective (8), except for the PI3K-specific inhibitor LY294002, which greatly diminished Akt basal phosphorylation. Similar results were obtained with wortmannin, another selective inhibitor of PI3K (see Fig. 7A). These findings indicate that the elevated Akt activity is likely dependent on PI3K activity in caveolin-overexpressing cells. Importantly, the levels of PTEN (which catalyzes the reverse reaction as PI3K, thereby leading to Akt dephosphorylation) were not affected either by arsenite treatment or by caveolin overexpression (Fig. 4B). It is thus unlikely that differences in PTEN expression play a significant role in the elevation of Akt phosphorylation by caveolin.
FIG. 4.
Elevated Akt phosphorylation in caveolin-overexpressing cells is likely due to elevated PI3K activity and not to diminished PTEN expression. (A) Western blot analysis of p-Akt expression in 293-V and 293-Cav25 cells following either no treatment or treatment for 1 h with an inhibitor of growth factor receptors (20 μM AG1296, 1 μM AG1478, or 300 μM suramin), the MEK1/2 inhibitor U0126 (20 μM), or the PI3K inhibitor LY294002 (25 μM). (B) Western blot analysis of PTEN levels in 293-V, 293-Cav25, and 293-Cav31 cells treated with 25 μM arsenite for the times indicated.
FIG. 7.
Effect of wortmannin on the sensitivity of caveolin-overexpressing cells to arsenite and H2O2. (A) Western blotting showing the inhibition of Akt phosphorylation by 100 nM wortmannin (Wort) treatment for 1 h. Total Akt levels are also shown for comparison. (B) Photograph illustrating the different numbers of colonies formed by 293-C25 populations after treatment with 10 μM arsenite for 24 h in the absence or presence of wortmannin. Graphs depict the quantitation of clonogenicity assays when cells are stressed in the absence or presence of wortmannin (100 nM) for 293-V and 293-Cav25 cells exposed to 10 μM arsenite (C); HeLa-V and HeLa-C1 cells treated with 10 μM arsenite (D); 293-V and 293-Cav25 exposed to 10 μM H2O2 (E). Clonogenicity assays were carried out as described in Materials and Methods. An asterisk indicates that the result is not statistically significant by t test analysis (P > 0.05). Data represent the means and standard deviations from three independent experiments.
Caveolin-overexpressing cells show increased susceptibility to arsenite toxicity.
Activation of the PI3K/Akt signaling pathway has been shown to promote cell survival under many different conditions of stress (9). Considering the hyperactivation of Akt in caveolin-bearing cells, we sought to investigate its biological consequences for arsenite-treated cells. Survival was assessed both by the ability of cells to exclude trypan blue and their capacity to form colonies. Unexpectedly, a reduced capacity to form colonies was seen when caveolin-overexpressing cells were treated with low doses of arsenite for 24 h (5 and 10 μM doses are shown in Fig. 5A). Likewise, trypan blue exclusion assays after arsenite treatment (Fig. 5B) revealed significantly greater cell death of 293-Cav25 and 293-Cav31 cells compared to that of 293-V cells. That cell death occurred by apoptosis was confirmed by 4′,6′-diamidino-2-phenylindole (DAPI) staining and procyclic acidic repetitive protein cleavage (data not shown).
FIG. 5.
Decreased survival of caveolin-overexpressing 293 cells following arsenite treatment. (A) Clonogenicity assay to monitor the sensitivity of 293-V, 293-Cav25, and 293-Cav31 cells to 5 and 10 μM arsenite. Cells were treated with arsenite for 24 h, and surviving colonies were stained and scored as described in Materials and Methods. Approximately 1,000 cells were evaluated for each data point. (B) Trypan blue exclusion assay to monitor survival of 293 cells expressing different levels of caveolin following treatment with 25 μM arsenite for 24 h. Data shown in both panels are the means and standard deviations from three independent experiments.
The unanticipated influence of caveolin overexpression on arsenite-triggered cytotoxicity prompted us to extend our analysis to a different cell model. HeLa cells were thus stably transfected with caveolin (Fig. 6A), which also resulted in higher levels of p-Akt than in vector-transfected cells, although in this model system Akt activation was further induced by arsenite. In addition, we determined whether basal Akt activity was increased in caveolin-bearing cells. As shown in Fig. 6B, and in agreement with Akt phosphorylation levels shown in Fig. 6A, Akt kinase activity was significantly higher in HeLa-Cav1 cells relative to HeLa-V cells. These results are also in agreement with those seen for 293 cells (Fig. 3).
FIG. 6.
Differential sensitivity of arsenite-treated HeLa cells expressing different levels of caveolin. (A) HeLa cells transfected with either empty vector pcDNA3.1 (HeLa-V) or pcDNA3.1-caveolin-1 (HeLa-C1) to overexpress caveolin-1 were treated with arsenite, and caveolin-1 and p-Akt expression were subsequently evaluated by Western blot analysis. (B) HeLa-V and HeLa-Cav1 cells were harvested, and Akt was immunoprecipitated. Akt kinase activity was evaluated by using histone H2B as a substrate. (C) Graph depicting the quantitation of clonogenicity assays following exposure to the doses of arsenite shown for 24 h. (D) Photograph illustrating the different numbers of colonies formed by HeLa-V and HeLa-C1 populations after treatment with 10 μM arsenite for 24 h, replating for clonogenicity assay, and staining with crystal violet. Data represent the means and standard deviations from three independent experiments.
As seen with caveolin-overexpressing 293 cells, HeLa-Cav1 cells were extremely sensitive to arsenite treatment, yielding a markedly reduced number of colonies (>20-fold lower) following a 24-h treatment with 10 μM arsenite (Fig. 6C and D). Trypan blue exclusion assays confirmed that caveolin overexpression was associated with increased toxicity (data not shown).
Inhibition of Akt activity in caveolin-overexpressing cells increases survival after arsenite treatment.
From the results described above, we found that caveolin overexpression in 293 cells (Fig. 1) leads to elevated Akt phosphorylation. To determine its potential role in influencing survival, 293 and HeLa cells were pretreated with a 100 nM concentration of the PI3K-specific inhibitor wortmannin (3) prior to treatment with 10 μM arsenite for 24 h. Western blot analysis showed that wortmannin was effective in inhibiting Akt phosphorylation, as expected (Fig. 7A). Surprisingly, for both 293 and HeLa cells overexpressing caveolin exposure to wortmannin resulted in a significant increase in survival relative to cells exposed to arsenite alone (Fig. 7B to D). Interestingly, this effect was not restricted to arsenite, as wortmannin pretreatment also protected caveolin-overexpressing 293 cells from H2O2-induced toxicity (Fig. 7E).
To further confirm that Akt signaling was detrimental to cell survival, we transfected a myc-tagged dominant-negative Akt protein (myc-Akt DN) expression vector into 293-Cav25 and evaluated its consequences on cell survival after arsenite exposure. Unlike what we had consistently observed with other cells, where transfection of Akt DN often induced cell death, the enhanced viability of the 293-Cav25 cells bearing the myc-Akt DN construct was a strong indicator that, in this situation, Akt expression was clearly detrimental to cell survival. As shown in Fig. 8A, Cav25-DN23 and Cav25-DN28 as well as Cav25-DN58 (expressing high and low levels of myc-Akt DN, respectively) and a pool of vector-transfected 293-Cav25 cells (Cav25-V) were treated with low doses of arsenite, and cell survival was determined by colony formation assay (Fig. 8B). In order to determine that caveolin expression was not altered during the transfection and selection process, both oligomerized and monomeric forms of caveolin-1 were evaluated by Western blot (Fig. 8A). In support of the results described in the legend to Fig. 7, we observed a significant increase in the number of Cav25-DN23 and Cav25-DN28 colonies after arsenite treatment with respect to Cav25-V cells, particularly when using Cav25-DN28, the clonal population that expressed higher levels of myc-Akt DN. Likewise, other cells with lower levels of myc-Akt DN, such as Cav25-DN58, failed to display an increase in colony formation, reminiscent of the Cav25-V cells. Taken together, these results agree with those obtained above with wortmannin pretreatment and strongly support our hypothesis that Akt activation constitutes a death signal for caveolin-overexpressing cells.
FIG. 8.
Influence of dominant-negative Akt on the survival of 293-Cav25 cells. (A) Western blot analysis of the expression of myc-Akt DN in selected clones (Cav25-DN23, Cav25-DN28, and Cav25-DN58) obtained from expressing pUSE-Akt1 (K179M mutant) or the pUSE empty vector (Cav25-V). Anti-Akt antibodies allowed for the detection of both endogenous Akt and myc-Akt DN expression (the latter exhibiting higher molecular weight than that of endogenous Akt). In addition, oligomerized and monomeric forms of caveolin-1 are also shown. (B) Graph depicting the number of colonies formed by the Cav25 Akt DN clones selected and by a cellular pool expressing the pUSE vector after treatment with 5 and 10 μM arsenite (Ars) for 24 h. Colonies were stained and evaluated as described in Materials and Methods. Differences in the survival of Cav25-DN23 and Cav25-DN28 with respect to Cav25-V were statistically significant by t test analysis (P < 0.05). These results represent the means and standard deviations from three independent experiments.
DISCUSSION
Caveolin-1 is known to play an important role in regulating cell growth through its ability to modulate the activities of molecules involved in growth factor signaling. Since many environmental stresses can usurp these signaling pathways to influence cell fate, we sought to investigate the role of caveolin-1 in the cellular response to one such stressor, the tumor promoter sodium arsenite. We have demonstrated here that caveolin-1 overexpression greatly sensitizes cells to the toxic influence of arsenite treatment and have provided additional evidence suggesting that this effect is mediated in large part through the activation of Akt. Long known for its role in promoting cell survival and preventing apoptosis, our studies provide novel evidence that under certain conditions elevated Akt can likewise promote cell death.
A variety of studies have implicated caveolin-1 as a negative regulator of cell growth through the inhibition of various signaling proteins (10, 11, 23, 31). Notably, caveolin has been shown to suppress ERK activation in response to growth factors such as EGF (10, 26). Several laboratories have demonstrated that there is significant overlap in the signaling pathways involved in regulating growth and stress responsiveness (reviewed by Martindale and Holbrook [21]). We have shown, for example, that both hydrogen peroxide and sodium arsenite activate ERK in an EGF receptor-dependent manner (8; our unpublished findings). The PI3K/Akt pathway constitutes another important signaling pathway activated by growth factors that is critically involved in growth control and cell survival (9, 13, 39). We have demonstrated that Akt is activated by hydrogen peroxide in an EGF receptor-dependent manner (39). Given this prior evidence, we hypothesized that caveolin overexpression would lead to reduced activation of ERK and Akt in response to sodium arsenite treatment. As expected, we did observe a considerable reduction in ERK activation in response to arsenite treatment in cells overexpressing caveolin (Fig. 3A). Surprisingly, however, Akt activity was found not to be suppressed but rather was constitutively activated in both 293 and HeLa caveolin-expressing cells. Moreover, the relative Akt activity was roughly proportional to the relative levels of caveolin overexpression. While the basis for this activation is not entirely clear, our studies indicate that it is associated with a reduction in nSMase activity leading to lower levels of ceramide (Fig. 2). As ceramide has been shown to inhibit the PI3K/Akt pathway (43), its reduced synthesis would be expected to result in enhanced Akt activity. In support of this hypothesis, treatment with the nSMase inhibitor CoQ6 (20) resulted in a small but significant increase in Akt phosphorylation (data not shown).
The novel view that Akt activation promotes cell death rather than survival during conditions of stress contradicts a number of other reports, including several from our own laboratory, in which activation of Akt following oxidant injury was shown to enhance survival (13, 39). Our laboratory and others have also described instances in which the activation of ERK, another classical survival signaling, has been associated with increased cell death (1, 25, 34, 40). These exceptions suggest that the cellular outcome can indeed vary depending on both the type of stimuli and the signals they trigger (i.e., sustained versus transient signaling). Thus, it is likely that the influence of Akt on survival is also dependent on a number of factors, some of which could be contingent on the cellular levels of caveolin. That is, the death-promoting influence of Akt following arsenite treatment occurred only in caveolin-overexpressing cells and was not apparent in the vector control cells. This observation is in keeping with previous findings that treatment of HeLa cells with wortmannin led to greater sensitivity to hydrogen peroxide (29). Like the 293-V cells, normal HeLa cells express undetectable levels of caveolin. Hence, the death-promoting effects of Akt pathway activation during conditions of oxidative stress appear to rely on overexpression of caveolin. The presence of caveolin in these cells may in turn modulate the interactions of PI3K with other molecules of the PI3K/Akt cascade. However, it is important to note that this influence of caveolin overexpression on Akt activity may also be dependent on the stress condition, the cell type, or both. Indeed, two previous studies suggested that a downregulation of the PI3K/Akt pathway in caveolin-1-overexpressing cells sensitizes them to apoptotic stimuli (17, 44). The study by Zundel et al. reported that reduced PI3K activity in caveolin-overexpressing fibroblasts was associated with enhanced sensitivity to ceramide (44). Liu et al. reported that caveolin overexpression was found to reduce Akt activity in fibroblasts and epithelial cells and to enhance their sensitivity to staurosporine (17). In the same study, treatment of caveolin-1 antisense-expressing cells (which were resistant to staurosporine-induced apoptosis) with the PI3K inhibitor LY294002 increased their sensitivity to staurosporine. We believe that one of the disparities between the reports of Liu et al. (17), Zundel et al. (44), and ours could be the model systems used. In our investigation, overexpression of caveolin-1, which alone causes significant changes in nSMase and Akt activity, is followed by treatment with the stress stimulus arsenite. The sequence of events and assessment of the effects could also constitute the major differences with respect to the studies described by Zundel et al., since their approach was instead to examine the effect of ceramide (the stress) on the caveolin-1-mediated regulation on PI3K signaling. In addition, differences between our findings and those described by Liu et al., where cells were sensitized to staurosporine by a caveolin-1-mediated decrease in Akt phosphorylation (opposite to what we are reporting here) are unlikely to be due to the use of cancer cells in the Liu et al. study (T24 bladder carcinoma cells) compared to the use of normal cells (293 kidney fibroblasts) in our study, as we observed a similar phenomenon in HeLa cells bearing caveolin. Lastly, the reasons for the differences observed between our study and those of Liu et al. and Zundel et al. may reflect various activities of other stress-activated pathways, including those that involve the MAP kinases ERK, JNK, and p38 as well as p53.
While the downstream target(s) responsible for Akt's death-promoting effects remains to be identified, forkhead transcription factors are plausible candidates. The activity of forkhead transcription factors is negatively regulated via phosphorylation by Akt. Recently, Nemoto and Finkel (22) demonstrated that forkhead plays an important role in scavenging reactive oxygen species and increases resistance to oxidative stress. Its inactivation by hydrogen peroxide is believed to contribute to oxidant-induced toxicity. Elevated Akt activity in caveolin-overexpressing clones would be expected to result in increased forkhead phosphorylation and reduced forkhead activity. Indeed, we observed that caveolin overexpression is associated with elevated forkhead phosphorylation (data not shown). Given that arsenite's toxic effects are mediated in large part through oxidative stress (14), this could be an important factor in the increased susceptibility of caveolin-overexpressing cells.
Acknowledgments
We thank Dennis D. Taub, Ronald L. Wange, Placido Navas, and Roy Cutler for helpful discussions and Paritosh Ghosh for his advice and critical reading of the manuscript.
Sonsoles Shack and Xian-Tao Wang contributed equally to this work.
REFERENCES
- 1.Alessandrini, A., S. Namura, M. A. Molkowitz, and J. V. Bonaventre. 1999. MEK1 protein kinase inhibition protects against damage resulting from focal cerebral ischemia. Proc. Natl. Acad. Sci. USA 96:12866-12869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Anderson, R. G. W. 1998. The caveolae membrane system. Annu. Rev. Biochem. 67:199-225. [DOI] [PubMed] [Google Scholar]
- 3.Arcaro, A., and M. P. Wymann. 1993. Wortmannin is a potent phosphatidylinositol 3-kinase inhibitor: the role of phosphatidylinositol 3,4,5-trisphosphate in neutrophil responses. Biochem. J. 296:297-301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bagla, P., and J. Kaiser. 1996. India's spreading health crisis draws global arsenic experts. Science 274:174-175. [DOI] [PubMed] [Google Scholar]
- 5.Barchowsky, A., E. J. Dudek, M. D. Treadwell, and K. E. Wetterhan. 1996. Arsenic induces oxidant stress and NF-kappa B activation in cultured aortic endothelial cells. Free Radic. Biol. Med. 21:783-790. [DOI] [PubMed] [Google Scholar]
- 6.Carpenter, G. 1999. Employment of the epidermal growth factor receptor in growth factor-independent signalling pathways. J. Cell Biol. 146:697-702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chang, L., and M. Karin. 2001. Mammalian MAP kinase signalling cascades. Nature 410:37-40. [DOI] [PubMed] [Google Scholar]
- 8.Chen, W., J. L. Martindale, N. J. Holbrook, and Y. Liu. 1998. Tumor promoter arsenite activates extracellular signal-regulated kinase through a signaling pathway mediated by epidermal growth factor receptor and Shc. Mol. Cell. Biol. 18:5178-5188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Datta, S. R., A. Brunet, and M. E. Greenberg. 1999. Cellular survival: a play in three Akts. Genes Dev. 13:2905-2927. [DOI] [PubMed] [Google Scholar]
- 10.Engelman, J. A., C. Chu, A. Lin, H. Jo, T. Ikezu, T. Okamoto, D. S. Kohtz, and M. P. Lisanti. 1998. Caveolin-mediated regulation of signaling along the p42/44 MAP kinase cascade in vivo. A role for the caveolin-scaffolding domain. FEBS Lett. 428:205-211. [DOI] [PubMed] [Google Scholar]
- 11.Galbiatti, F., D. Volonte, J. A. Engelman, G. Watanabe, R. Burk, R. G. Pestell, and M. P. Lisanti. 1998. Targeted downregulation of caveolin-1 is sufficient to drive cell transformation and hyperactivate the p42/44 MAP kinase cascade. EMBO J. 17:6633-6648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Guyton, K. Z., M. Gorospe, X. Wang, Y. D. Mock, G. C. Kokkonen, Y. Liu, G. S. Roth, and N. J. Holbrook. 1998. Age-related changes in activation of mitogen-activated protein kinase cascades by oxidative stress. J. Investig. Dermatol. Symp. Proc. 3:23-27. [PubMed] [Google Scholar]
- 13.Ikeyama, S., G. Kokkonen, S. Shack, X. T. Wang, and N. J. Holbrook. 2002. Loss in oxidative stress tolerance with aging linked to reduced extracellular signal-regulated kinase and Akt kinase activities. FASEB J. 16:114-116. [DOI] [PubMed] [Google Scholar]
- 14.Kenyon, E. M., and M. F. Hughes. 2001. A concise review of the toxicity and carcinogenicity of dimethylarsinic acid. Toxicology 160:227-236. [DOI] [PubMed] [Google Scholar]
- 15.Koleske, A. J., D. Baltimore, and M. P. Lisanti. 1995. Reduction of caveolin and caveolae in oncogenically transformed cells. Proc. Natl. Acad. Sci. USA 92:1381-1385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Liu, J., P. Oh, T. Horner, R. A. Rogers, and J. E. Schnitzer. 1997. Organized endothelial cell surface signal transduction in caveolae distinct from glycosylphosphatidylinositol-anchored protein microdomains. J. Biol. Chem. 272:7211-7222. [DOI] [PubMed] [Google Scholar]
- 17.Liu, J., P. Lee, F. Galbiati, R. N. Kitsis, and M. P. Lisanti. 2001. Caveolin-1 expression sensitizes fibroblastic and epithelial cells to apoptotic stimulation. Am. J. Cell Physiol. 280:C823-C835. [DOI] [PubMed]
- 18.Liu, P., and R. G. W. Anderson. 1995. Compartmentalized production of ceramide at the cell surface. J. Biol. Chem. 270:27179-27185. [DOI] [PubMed] [Google Scholar]
- 19.Liu, P., Y. Ying, Y. G. Ko, and R. G. Anderson. 1996. Localization of platelet-derived growth factor-stimulated phosphorylation cascade to caveolae. J. Biol. Chem. 271:299-310. [DOI] [PubMed] [Google Scholar]
- 20.Martin, S. F., C. Gomez-Diaz, P. Navas, and J. M. Villalba. 2002. Ubiquinol inhibition of neutral sphingomyelinase in liver plasma membrane: specific inhibition of the Mg2+-dependent enzyme and role of isoprenoid chain. Biochem. Biophys. Res. Commun. 297:581-586. [DOI] [PubMed] [Google Scholar]
- 21.Martindale, J. L., and N. J. Holbrook. 2002. Cellular response to oxidative stress: signaling for suicide and survival. J. Cell. Physiol. 192:1-15. [DOI] [PubMed] [Google Scholar]
- 22.Nemoto, S., and T. Finkel. 2002. Redox regulation of forkhead proteins through a p66shc-dependent signaling pathway. Science 295:2450-2452. [DOI] [PubMed] [Google Scholar]
- 23.Okamoto, T., A. Schlegel, P. E. Scherer, and M. P. Lisanti. 1998. Caveolins, a family of scaffolding proteins for organizing “preassembled signaling complexes” at the plasma membrane. J. Biol. Chem. 273:5419-5422. [DOI] [PubMed] [Google Scholar]
- 24.Palade, G. E., and R. R. Bruns. 1968. Structural modification of plasmalemma vesicles. J. Cell Biol. 37:633-649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Park, S. A., H. J. Park, B. I. Lee, Y. H. Ahn, S. U. Kim, and K. S. Choi. 2001. Bcl-2 blocks cisplatin-induced apoptosis by suppression of ERK-mediated p53 accumulation in B104 cells. Brain Res. Mol. Brain Res. 93:18-26. [DOI] [PubMed] [Google Scholar]
- 26.Park, W. Y., J. S. Park, K. A. Cho, D. I. Kim, Y. G. Ko, J. S. Seo, and S. C. Park. 2000. Up-regulation of caveolin attenuates epidermal growth factor signaling in senescent cells. J. Biol. Chem. 275:20847-20852. [DOI] [PubMed] [Google Scholar]
- 27.Razani, B., A. Schlegel, J. Liu, and M. P. Lisanti. 2001. Caveolin-1, a putative tumour suppressor gene. Biochem. Soc. Trans. 29:494-499. [DOI] [PubMed] [Google Scholar]
- 28.Rothberg, K. G., J. E. Heuser, W. C. Donzell, Y. S. Ying, J. R. Glenney, and R. G. Anderson. 1992. Caveolin, a protein component of caveolae membrane coats. Cell 68:673-682. [DOI] [PubMed] [Google Scholar]
- 29.Sargiacomo, M., P. E. Scherer, Z. L. Tang, E. Kubler, K. S. Song, M. C. Sanders, and M. P. Lisanti. 1995. Oligomeric structure of caveolin: implications for caveolin membrane organization. Proc. Natl. Acad. Sci. USA 92:9407-9411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Schlegel, A., and M. P. Lisanti. 2001. Caveolae and their coat proteins, the caveolins: from electron microscopic novelty to biological launching pad. J. Cell. Physiol. 186:329-337. [DOI] [PubMed] [Google Scholar]
- 31.Shaul, P. W., and R. G. Anderson. 1998. Role of plasmalemmal caveolae in signal transduction. Am. J. Physiol. 275:L843-L851. [DOI] [PubMed] [Google Scholar]
- 32.Smart, E. J., G. A. Graf, M. A. McNiven, W. C. Sessa, F. A. Engelman, P. E. Scherer, T. Okamoto, and M. P. Lisanti. 1999. Caveolins, liquid-ordered domains, and signal transduction. Mol. Cell. Biol. 19:7289-7304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Song, K. S., Z. L. Tang, S. Li, and M. P. Lisanti. 1997. Mutational analysis of the properties of caveolin-1. A novel role for the C-terminal domain in mediating homo-typic caveolin-caveolin interactions. J. Biol. Chem. 272:4398-4403. [DOI] [PubMed] [Google Scholar]
- 34.Sutherland, C. L., A. W. Heath, S. L. Pelech, P. R. Young, and M. R. Gold. 1996. Differential activation of the ERK, JNK, and p38 mitogen-activated protein kinases by CD40 and the B cell antigen receptor. J. Immunol. 157:3381-3390. [PubMed] [Google Scholar]
- 35.Tomiuk, S., M. Zumbansen, and W. Stoffel. 2000. Characterization and subcellular localization of murine and human magnesium-dependent neutral sphingomyelinase. J. Biol. Chem. 275:5710-5717. [DOI] [PubMed] [Google Scholar]
- 36.Veldman, R. J., N. Maestre, O. M. Abduib, J. A. Medin, R. Salvayre, and T. Levade. 2001. A neutral sphingomyelinase resides in sphingolipid-enriched microdomains and is inhibited by the caveolin-scaffolding domain: potential implications in tumour necrosis factor signalling. Biochem. J. 355:859-868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Vlahos, C. J., W. F. Matter, K. Y. Hui, and R. F. Brown. 1994. A specific inhibitor of phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1benzopyran-4-one (LY294002). J. Biol. Chem. 269:5241-5248. [PubMed] [Google Scholar]
- 38.Wang, X., J. L. Martindale, Y. Liu, and N. J. Holbrook. 1998. The cellular response to oxidative stress: influences of mitogen-activated protein kinase signalling pathways on cell survival. Biochem. J. 333:291-300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wang, X., K. D. McCullough, T. F. Franke, and N. J. Holbrook. 2000. Epidermal growth factor receptor-dependent Akt activation by oxidative stress enhances cell survival. J. Biol. Chem. 275:14624-14631. [DOI] [PubMed] [Google Scholar]
- 40.Wang, X., J. L. Martindale, and N. J. Holbrook. 2000. Requirement for ERK activation in cisplatin-induced apoptosis. J. Biol. Chem. 275:39435-39443. [DOI] [PubMed] [Google Scholar]
- 41.Yamada, E. 1955. The fine structure of the gall bladder epithelium of the mouse. J. Biophys. Biochem. Cytol. 1:445-458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Yu, W., J. Cassara, and P. F. Weller. 2000. Phosphatidylinositide 3-kinase localizes to cytoplasmic lipid bodies in human polymorphonuclear leukocytes and other myeloid-derived cells. Blood 95:1078-1085. [PubMed] [Google Scholar]
- 43.Zundel, W., and A. Giaccia. 1998. Inhibition of the antiapoptotic PI(3)K/Akt/Bad pathway by stress. Genes Dev. 12:1941-1946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Zundel, W., L. M. Swiersz and A. Giaccia. 2000. Caveolin 1-mediated regulation of receptor tyrosine kinase-associated phosphatidylinositol 3-kinase activity by ceramide. Mol. Cell. Biol. 20:1507-1514. [DOI] [PMC free article] [PubMed] [Google Scholar]