Cellular Stress Induces the Tyrosine Phosphorylation of Caveolin-1 (Tyr) via Activation of p38 Mitogen-activated Protein Kinase and c-Src kinase EVIDENCE FOR CAVEOLAE, THE ACTIN CYTOSKELETON, AND FOCAL ADHESIONS AS MECHANICAL SENSORS OF OSMOTIC STRESS*

Environmental stressors have been recently shown to activate intracellular mitogen-activated protein (MAP) kinases, such as p38 MAP kinase, leading to changes in cellular functioning. However, little is known about the downstream elements in these signaling cascades. In this study, we show that caveolin-1 is phosphorylated on tyrosine 14 in NIH 3T3 cells after stimulation with a variety of cellular stressors (i.e. high osmolarity, H2O2, and UV light). To detect this phosphorylation event, we employed a phosphospecific monoclonal antibody probe that recognizes only tyrosine 14-phosphorylated caveolin-1. Since p38 MAP kinase and c-Src have been previously implicated in the stress response, we next assessed their role in the tyrosine phosphorylation of caveolin-1. Interestingly, we show that the p38 inhibitor (SB203580) and a dominant-negative mutant of c-Src (SRC-RF) both block the stress-induced tyrosine phosphorylation of caveolin-1 (Tyr(P)). In contrast, inhibition of the p42/44 MAP kinase cascade did not affect the tyrosine phosphorylation of caveolin-1. These results indicate that extracellular stressors can induce caveolin-1 tyrosine phosphorylation through the activation of well established upstream elements, such as p38 MAP kinase and c-Src kinase. However, heat shock did not promote the tyrosine phosphorylation of caveolin-1 and did not activate p38 MAP kinase. Finally, we show that after hyperosmotic shock, tyrosine-phosphorylated caveolin-1 is localized near focal adhesions, the major sites of tyrosine kinase signaling. In accordance with this localization, disruption of the actin cytoskeleton dramatically potentiates the tyrosine phosphorylation of caveolin-1. Taken together, our results clearly define a novel signaling pathway, involving p38 MAP kinase activation and caveolin-1 (Tyr(P)). Thus, tyrosine phosphorylation of caveolin-1 may represent an important downstream element in the signal transduction cascades activated by cellular stress. Eukaryotic cells adapt to environmental changes by activating cellular mechanisms that detect and control diverse forms of stress. In many cell types, osmotic stress affects cell volume and intracellular ion homeostasis. Cells shrink when exposed to hyperosmotic stress and swell in hypo-osmotic medium. Usually, cells do not tolerate significant perturbations of cell volume and intracellular ion concentrations. In fact, they manage to restore their normal cell integrity and function. When DNA or protein damage occur, cells activate repair processes and protection systems to compensate for osmotic stress. When osmotic stress is too intense or when cells are exposed to osmotic stress for too long, cells activate programmed cell death cascades (apoptosis) (1–5). Mitogen-activated protein (MAP) kinases are key elements for signal transduction events from the cell surface to the nucleus. MAP kinases have central roles in cellular proliferation, apoptosis, and differentiation. Also, MAP kinases are activated in response to cellular stress. It has been shown that osmotic shock can activate p38, stress-activated protein kinase (c-Jun N-terminal kinase), and Jak/STAT pathways in mammalian cells (6–9). Additionally, the p38 MAP kinase pathway is activated by hydrogen peroxide and UV light (10–13). Thus, it has become evident that physico-chemical stressors trigger signal transduction cascades leading to changes in cellular functions. Caveolae are 50–100-nm invaginations of the plasma membrane that have been implicated in vesicular trafficking events and signal transduction processes (14–17). Caveolin-1, a 21– 24-kDa integral membrane protein, is a principal component of caveolae membranes in vivo (18–22). Several independent lines of evidence suggest that caveolin family members (caveolin-1, -2, and -3) function as scaffolding proteins (23) to concentrate and organize specific lipids (cholesterol and sphingolipids (24–26)) and lipid-modified signaling molecules within caveolae membranes (26–31). Caveolin-1 was first identified as a major tyrosine-phosphorylated protein in v-Src-transformed chicken embryo fibroblasts (18). However, the functional significance of the phosphorylation of caveolin-1 on tyrosine is not known yet. Using both in vitro and in vivo approaches, we have previously shown that Src induces phosphorylation of caveolin-1 on tyrosine 14. Recently, we have characterized a novel monoclonal antibody * This work was supported by grants from the National Institutes of Health (NIH), the Muscular Dystrophy Association, the American Heart Association and the Komen Breast Cancer Foundation (to M.P.L.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. § Recipient of Telethon-Italia Fellowship 470/bi. i Supported in part by grants NIH Grants R01-CA70897, R01CA75503, and P50-HL56399. Recipient of the Irma T. Hirschl award and an award from the Susan G. Komen Breast Cancer Foundation. ** To whom correspondence should be addressed: Dept. of Molecular Pharmacology and the Albert Einstein Cancer Center, Albert Einstein College of Medicine, 1300 Morris Park Ave., Bronx, NY 10461. Tel.: 718-430-8828; Fax: 718-430-8830; E-mail: lisanti@aecom.yu.edu. 1 The abbreviations used are: MAP, mitogen-activated protein; MAPK, MAP kinase; STAT, signal transducers and activators of transcription; mAb, monoclonal antibody; pAb, polyclonal antibody; ERK, extracellular signal-regulated kinase; IPTG, isopropyl-1-thio-b-D-galactopyranoside (dioxane-free); PBS, phosphate-buffered saline. THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 11, Issue of March 16, pp. 8094–8103, 2001 © 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

Eukaryotic cells adapt to environmental changes by activating cellular mechanisms that detect and control diverse forms of stress. In many cell types, osmotic stress affects cell volume and intracellular ion homeostasis. Cells shrink when exposed to hyperosmotic stress and swell in hypo-osmotic medium. Usually, cells do not tolerate significant perturbations of cell volume and intracellular ion concentrations. In fact, they manage to restore their normal cell integrity and function. When DNA or protein damage occur, cells activate repair processes and protection systems to compensate for osmotic stress. When osmotic stress is too intense or when cells are exposed to osmotic stress for too long, cells activate programmed cell death cascades (apoptosis) (1)(2)(3)(4)(5).
Mitogen-activated protein (MAP) 1 kinases are key elements for signal transduction events from the cell surface to the nucleus. MAP kinases have central roles in cellular proliferation, apoptosis, and differentiation. Also, MAP kinases are activated in response to cellular stress. It has been shown that osmotic shock can activate p38, stress-activated protein kinase (c-Jun N-terminal kinase), and Jak/STAT pathways in mammalian cells (6 -9). Additionally, the p38 MAP kinase pathway is activated by hydrogen peroxide and UV light (10 -13). Thus, it has become evident that physico-chemical stressors trigger signal transduction cascades leading to changes in cellular functions.
Caveolin-1 was first identified as a major tyrosine-phosphorylated protein in v-Src-transformed chicken embryo fibroblasts (18). However, the functional significance of the phosphorylation of caveolin-1 on tyrosine is not known yet. Using both in vitro and in vivo approaches, we have previously shown that Src induces phosphorylation of caveolin-1 on tyrosine 14. Recently, we have characterized a novel monoclonal antibody probe that is specific for tyrosine 14-phosphorylated caveolin-1 (32). We have demonstrated that insulin and epidermal growth factor stimulate phosphorylation of caveolin-1 on tyrosine 14. Also, we have shown that activated Src induces constitutive phosphorylation of caveolin-1 on tyrosine 14, which is localized in close proximity with focal adhesions (32).
Here, we examine if phosphorylation of caveolin-1 on tyrosine 14 is modulated by extracellular stressors. We show that hyperosmotic shock promotes phosphorylation of caveolin-1 on tyrosine 14 through the activation of p38 MAP kinase and c-Src kinase. Our results suggest that phosphorylation of caveolin-1 may represent a key element in the signal transduction cascades activated by cellular stress.

Cell Culture
NIH 3T3 cells were grown in Dulbecco's modified Eagle's medium supplemented with glutamine, antibiotics (penicillin and streptomycin), and 10% donor bovine calf serum, as previously described (34,35). NIH 3T3 cells stably transfected with an IPTG-inducible vector expressing a dominant negative mutant of c-Src (SRC-RF) were also grown in Dulbecco's modified Eagle's medium supplemented with glutamine, antibiotics (penicillin and streptomycin), and 10% donor bovine calf serum (36 -38). These cells were treated with 5 mM IPTG for 72 h to induce SRC-RF expression.

Exposure of NIH 3T3 Cells to Cellular Stress
Hyperosmotic Shock and Recovery-NIH 3T3 cells were treated with the indicated concentration of sucrose for the indicated period of time.
In recovery experiments, after treatment with 600 mM sucrose for 10 min, cells were grown without sucrose for the indicated period of time. Sucrose was diluted into complete growth medium.
Urea Treatment-NIH 3T3 cells were treated with 600 mM urea alone or in combination with 600 mM sucrose for 10 min. Urea was diluted into complete growth medium.
Sorbitol Treatment-NIH 3T3 cells were treated with 600 mM sorbitol for 10 min. Sorbitol was diluted into complete growth medium.
H 2 O 2 Treatment-NIH 3T3 cells were treated with the indicated concentration of H 2 O 2 for the indicated period of time. H 2 O 2 was diluted into complete growth medium.
UV Radiation Treatment-NIH 3T3 cells were irradiated with different doses of UV-C light (10,20,60,120,360, and 600 J/m 2 ). During irradiation, cells were deprived of growth medium.
Heat Shock-NIH 3T3 cells were incubated at 43°C for 30 min.

Cytoskeletal Disrupting Agents
NIH 3T3 cells were treated with 5 g/ml cytochalasin D for 15 min or 1.5 g/ml nocodazole for 1 h, prior to hyperosmotic shock with 600 mM sucrose for 10 min (in the presence of cytochalasin D or nocodazole) (39).

Immunofluorescence Microscopy
NIH 3T3 cells grown on glass coverslips were washed three times with PBS and fixed for 30 min at room temperature with 2% paraformaldehyde in PBS. Fixed cells were rinsed with PBS and permeabilized with PBS containing 0.1% Triton X-100 and 0.2% bovine serum albumin for 10 min. Then cells were treated with 25 mM NH 4 Cl in PBS for 10 min at room temperature to quench free aldehyde groups. Cells were rinsed with PBS and incubated with the primary antibody for 1 h at room temperature (ϳ1000-fold diluted in PBS with 0.1% Triton X-100 and 0.2% bovine serum albumin). After three washes with PBS (10 min each), cells were incubated with the secondary antibody for 1 h at room temperature: lissamine rhodamine B sulfonyl chloride-conjugated goat anti-rabbit antibody (5 g/ml) and/or fluorescein isothiocyanate-conjugated goat anti-mouse antibody (5 g/ml). Finally, cells were washed three times with PBS (10 min each wash), and slides were mounted with Slow-Fade anti-fade reagent (Molecular Probes, Inc., Eugene, OR) and observed under a Bio-Rad MR 600 confocal microscope. Note that both anti-phosphocaveolin-1 IgG (mAb cl 56; directed against residues 9 SEGHLpYTVPI 18 (where pY represents phosphotyrosine)) and anticaveolin-1 IgG (pAb N-20; directed against residues 2 SGGKYVDSEGH-LYTVPIREQ 21 ) recognize a very similar epitope that is present in Cav-1␣ (residues 1-178) but lacking in Cav-1␤ (residues 32-178).

p38-and p42/44-MAPK Pathway Inhibitor Treatments
NIH 3T3 cells were treated for 16 h with the indicated concentration of p38 MAP kinase inhibitor, SB203580 (Calbiochem) or the p42/44 MAP kinase pathway inhibitor, PD98059 (Calbiochem). Cells were then subjected to hyperosmotic shock or H 2 O 2 treatment (see "Exposure of NIH 3T3 Cells to Cellular Stress") in the presence of the inhibitor.

Hyperosmotic Shock Induces the Phosphorylation of Caveolin-1 on Tyrosine 14
To examine if hyperosmotic stress induces the phosphorylation of caveolin-1 on tyrosine 14, NIH 3T3 cells were treated with different concentrations of sucrose (1, 10, 50, 100, and 600 mM) for 10 min (Fig. 1, A and B) or with 600 mM sucrose for different periods of time (1,5,10,20, and 30 min) (Fig. 2, A and  B). Cell lysates were prepared and subjected to immunoblot analysis with a phosphospecific mAb probe that recognizes only caveolin-1 phosphorylated on tyrosine 14 (mAb cl 56). Fig. 1A shows that caveolin-1 is phosphorylated on tyrosine 14 in NIH 3T3 cells treated with 600 mM sucrose for 10 min (upper panel). Phosphorylation of caveolin-1 was undetectable in control cells (CTL). As shown in Fig. 2A, caveolin-1 phosphorylation induced by hyperosmotic stress occurred transiently, since phosphorylation of caveolin-1 on tyrosine 14 was maximal after stimulation with 600 mM sucrose for 10 min, and progressively decreased after longer stimulation (upper panel). Interestingly, total caveolin-1 expression did not change upon hyperosmotic stress (Fig. 1A, lower panel, and Fig. 2A, lower panel), indicating that sucrose treatment modulates the phosphorylation state of caveolin-1 rather than caveolin-1 protein expression.
Hyperosmotic stress is known to activate the p38 MAP kinase pathway in a variety of different cellular contexts (6 -9).
Thus, we verified in our experimental model that hyperosmotic stress activated this intracellular pathway. Fig. 1B shows that treatment with 600 mM sucrose for 10 min is sufficient to induce the phosphorylation of p38 MAP kinase (upper panel). Phosphorylation of p38 MAP kinase results in the activation of this enzyme. Fig. 2B shows that, like caveolin-1, phosphorylation of p38 MAP kinase in NIH 3T3 cells becomes detectable after stimulation with 600 mM sucrose for 10 min. Interestingly, p38 MAP kinase remained activated if the sucrose stimulation persisted for 30 min (upper panel). Total p38 MAP kinase protein expression was not affected by hyperosmotic stress, as demonstrated using an antibody that recognizes both phosphorylated and nonphosphorylated forms of p38 (Fig. 1B, lower panel, and Fig.  2B, lower panel). These findings suggest that the p38 MAP kinase pathway remains activated as long as the cells are subjected to hyperosmotic shock, while the tyrosine phosphorylation of caveolin-1 represents a transient event.
To further explore this idea, we examined if p38 MAP kinase activation and phosphorylation of caveolin-1 could be reversed by removing the hyperosmotic stimulus from the medium of hyperosmotic shocked NIH 3T3 cells. NIH 3T3 cells were shocked with 600 mM sucrose for 10 min and then switched to normal medium (without sucrose) for different periods of time (10 min, 30 min, 2 h, and 4 h) (Fig. 3). Note that phosphorylation of caveolin-1 (Fig. 3A, upper panel) and p38 MAP kinase (Fig. 3B, upper panel) are completely reversed when the hyperosmotic stimulus is removed from the growth medium. Importantly, total caveolin-1 (Fig. 3A, lower panel) and p38 MAP Blots were probed with a phosphospecific caveolin-1 monoclonal antibody (mAb clone 56) that recognizes only the tyrosine 14-phosphorylated form of caveolin-1 (upper panel). Note that only treatment with 600 mM sucrose for 10 min induces the phosphorylation of caveolin-1 on tyrosine 14. Immunoblotting with mAb 2297, that recognizes total caveolin-1 (lower panel), indicated that caveolin-1 protein expression was not modified by sucrose treatment. B, p38 MAP kinase activation. Blots were probed with a phosphospecific p38 polyclonal antibody that recognizes only the active phosphorylated form of p38 MAP kinase (upper panel). Similarly, p38 MAP kinase was activated only when cells were treated with 600 mM sucrose for 10 min. Immunoblotting with anti-p38 MAP kinase pAb (lower panel), which recognizes total p38 MAP kinase, indicated that total p38 MAP kinase protein expression was not modified by sucrose treatment.

Activation of the p38 MAP Kinase Pathway, but Not the p42/44 MAP Kinase Pathway, Is Required to Induce the Tyrosine Phosphorylation of Caveolin-1
Since hyperosmotic shock induced activation of the p38 MAP kinase pathway and phosphorylation of caveolin-1, we next tested the hypothesis that p38 MAP kinase is an up-stream element in the cascade that induces the tyrosine phosphorylation of caveolin-1. We pretreated NIH 3T3 cells for 16 h with different concentrations (2, 10, and 50 M) of a well characterized p38 MAP kinase inhibitor (SB203580) prior to hyperosmotic stimulation for 10 min (in the presence of the inhibitor). Fig. 4A shows that we successfully inhibited the activation of the p38 MAP kinase pathway. In fact, treatment with as little as 2 M SB203580 was sufficient to inhibit p38 MAP kinase phosphorylation (upper right panel). Interestingly, inhibition of the p38 MAP kinase pathway also resulted in the inhibition of caveolin-1 tyrosine phosphorylation (Fig. 4A, upper left panel). These results support the idea that caveolin-1 phosphorylation follows activation of the p38 MAP kinase pathway in the signaling cascade induced by hyperosmotic shock. Total caveolin-1 and p38 MAP kinase protein expression did not change under these experimental conditions (Fig. 4A, lower right and left panels).
We next examined if hyperosmotic shock can induce activation of a different MAP kinase pathway, the p42-44 MAP kinase pathway, and if inhibition of the p42-44 MAP kinase pathway results in inhibition of caveolin-1 tyrosine phosphorylation. We pretreated NIH 3T3 cells for 16 h with a well known inhibitor of the p42-44 MAP kinase pathway (PD98059; 10 M) prior to hyperosmotic shock in the presence of inhibitor.

Recombinant Expression of a Dominant Negative Form of c-Src Blocks the Tyrosine Phosphorylation of Caveolin-1 Induced by Hyperosmotic Shock
We and others have recently demonstrated that caveolin-1 (Tyr 14 ) is a specific tyrosine kinase substrate for activated Src both in vitro and in vivo (32). Thus, we next examined the possible requirement of c-Src kinase activity for the tyrosine phosphorylation of caveolin-1 in response to hyperosmotic shock. For this purpose, we employed an NIH 3T3 cell line stably transfected with an IPTG-inducible vector encoding a dominant-negative mutant of c-Src (termed SRC-RF). SRC-RF is a double mutant that is kinase-dead (K295R) and lacks the C-terminal regulatory tyrosine (Y527F). Induction of SRC-RF expression results in a strong reduction of endogenous c-Src kinase activity (36 -38).
SRC-RF expression was induced for 72 h with 5 mM IPTG before the cells were subjected to hyperosmotic shock (in the presence of IPTG). Fig. 5 shows that induction of this dominant negative form of c-Src results in the inhibition of caveolin-1 tyrosine phosphorylation in response to hyperosmotic shock (upper panel). Total caveolin-1 protein expression did not change under these experimental conditions (Fig. 5, lower panel). These data indicate that endogenous c-Src kinase activity is also important for mediating the tyrosine phosphorylation of caveolin-1 in response to cellular stress. These results provide the first evidence that endogenous c-Src is involved in the tyrosine phosphorylation of caveolin-1 in vivo.

What Is the Localization of Tyrosine 14-Phosphorylated
Caveolin-1 after Hyperosmotic Shock?
To examine the localization of tyrosine 14-phosphorylated caveolin-1 after hyperosmotic shock, we stimulated NIH 3T3 cells with 600 mM sucrose for 10 min and doubly immunostained the cells with anti-phosphocaveolin-1 mouse mAb (clone 56) and anti-caveolin-1 rabbit pAb (N-20). These bound primary antibodies were visualized using distinctly tagged fluorescent secondary antibodies (see "Experimental Procedures"). Importantly, little or no immunostaining was ob- Note that inhibition of the p42/44 MAP kinase pathway does not prevent phosphorylation of caveolin-1 in NIH 3T3 cells treated with 600 mM sucrose for 10 min. In A and B, immunoblotting with mAb 2297, which recognizes total caveolin-1, and anti-p38 MAP kinase pAb, which recognizes total p38 MAP kinase, indicated that total caveolin-1 and p38 MAP kinase protein expression did not change under these experimental conditions. Each lane contains equal amount of total proteins. served using anti-phosphocaveolin-1 IgG when cells where not subjected to hyperosmotic shock (Fig. 6A, upper left panel). Conversely, caveolin-1 tyrosine phosphorylation levels increased after cells were treated with 600 mM sucrose for 10 min (Fig. 6A, lower left panel). Total caveolin-1 protein expression did not change after osmotic shock (Fig. 6A, central  panels). These results are consistent with our results obtained by Western blot analysis (this report), which indicates that tyrosine phosphorylation of caveolin-1 is induced by hyperosmotic shock.
To verify that tyrosine 14-phosphorylated caveolin-1 is indeed localized near focal adhesions after hyperosmotic shock, we next performed double labeling experiments with antibodies directed against a widely used marker for focal adhesions, paxillin. After hyperosmotic shock, NIH 3T3 cells were double-labeled with anti-phosphocaveolin-1 mAb (left panel) and anti-paxillin (LRSC-conjugated mAb) (middle panel). Fig. 6B shows that tyrosine 14-phosphorylated caveolin-1 and paxillin colocalize in NIH 3T3 cells, indicating that hyperosmotic shock induces the tyrosine phosphorylation of caveolin-1 near focal adhesions. A color overlay is shown to better illustrate this colocalization (right panel).

Cell Shrinkage Is Required to Induce Caveolin-1 Phosphorylation and to Activate the p38 MAP Kinase Pathway
We next examined whether hyperosmolarity by itself or the resulting cell shrinkage is responsible for p38 MAP kinase activation and caveolin-1 phosphorylation. To increase the hypertonicity of the medium, we incubated NIH 3T3 cells with 600 mM urea for 10 min, which is well known not to affect cell volume (9). Interestingly, Fig. 7 shows that sucrose treatment, but not urea treatment, induces the tyrosine phosphorylation of caveolin-1 (Fig. 7A, upper panel) and activation of the p38 MAP kinase pathway (Fig. 7B, upper panel). When cells were incubated with 600 mM urea and 600 mM  6. Tyrosine 14-phosphorylated caveolin-1 is localized near focal adhesions after hyperosmotic stimulation. A, NIH 3T3 cells were left untreated (Ϫshock; upper panels) or treated for 10 min with 600 mM sucrose (ϩshock; lower panels). Cells were then doublelabeled with anti-phosphocaveolin-1 IgG (mAb 56) (left panels) and anti-caveolin-1 IgG (pAb N-20) (central panels). Note that after hyperosmotic stress, phosphocaveolin-1 staining is enriched at discrete sites near the plasma membrane and shows partial colocalization with total caveolin-1 staining. A color overlay is shown to better illustrate the partial colocalization between phosphocaveolin-1 and total caveolin-1 (right panels). B, NIH 3T3 cells were treated for 10 min with 600 mM sucrose and double-labeled with antibodies specific for phosphocaveolin-1 (left panel) and paxillin (middle panel), a marker for focal adhesions. Note that phosphocaveolin-1 colocalizes with paxillin at the level of focal adhesions. A color overlay (right panel) is shown to better illustrate the colocalization between phosphocaveolin-1 and paxillin. sucrose at the same time, caveolin-1 phosphorylation and activation of p38 MAP kinase were detected. Total caveolin-1 (Fig. 7A, lower panel) and p38 MAP kinase (Fig. 7B, lower panel) protein expression did not change under these experimental conditions. In addition, cell shrinkage was morphologically observed after incubation with sucrose but not urea (not shown). From these results, we conclude that cell shrinkage is the trigger for p38 MAP kinase activation and caveolin-1 tyrosine phosphorylation.

Cytoskeletal Disrupting Agents Potentiate the Tyrosine Phosphorylation of Caveolin-1 Induced by Osmotic Shock
We have shown that cell shrinkage is crucial for caveolin-1 phosphorylation and activation of p38 MAP kinase (Fig. 7). Since the actin cytoskeleton and microtubules are important for maintenance of cell shape, we next asked if disruption of these important intracellular elements affects caveolin-1 tyrosine phosphorylation. To address this issue, cells were pretreated with either cytochalasin D (to disrupt the actin cytoskeleton) or nocodazole (to disrupt microtubules) prior to osmotic shock with 600 mM sucrose for 10 min. Fig. 8 shows that treatment with cytochalasin D or nocodazole alone did not affect caveolin-1 tyrosine phosphorylation. However, pretreatment of cells with these cytoskeletal dis-ruptors dramatically potentiates the effects of osmotic stress induced by treatment with 600 mM sucrose (Fig. 8A). Interestingly, cytochalasin D pretreatment had the largest effect (ϳ4-fold stimulation). This is consistent with the observation that tyrosine-phosphorylated caveolin-1 is localized near focal adhesions, which are known to be linked to the actin cytoskeleton. In contrast, disruption of the cytoskeleton did not potentiate the effect of osmotic stress on p38 activation (Fig. 8B). One possible explanation for these differences is that pretreatment with cytoskeletal disruptors prevents the rapid dephosphorylation of caveolin-1 rather than stimulating its overall phosphorylation. Nevertheless, these results demonstrate a clear link between caveolin-1 tyrosine phosphorylation and the integrity of the cytoskeleton.
Also, Fig. 8 shows that both sucrose and sorbitol have similar effects in promoting caveolin-1 phosphorylation and activation of p38 MAP kinase. This result indicates that, independently of the sugar employed, hyperosmotic shock induces caveolin-1 phosphorylation on tyrosine 14 and activation of p38 MAP kinase. Importantly, total caveolin-1 and p38 protein expression did not change under these experimental conditions (Fig.  8, A and B, lower panels). . Note that only sucrose, and not urea treatment, induces caveolin-1 phosphorylation. Immunoblotting with mAb 2297, which recognizes total caveolin-1 (lower panel), indicated that caveolin-1 protein expression was not modified by either sucrose or urea treatment. B, p38 MAPK activation. Blots were probed with a phosphospecific p38 polyclonal antibody (upper panel). Similarly, p38 MAP kinase was activated only when cells were treated with 600 mM sucrose for 10 min but not with 600 mM urea for 10 min. Immunoblotting with anti-p38 MAP kinase pAb (lower panel), which recognizes total p38 MAP kinase, indicated that total p38 MAP kinase protein expression was not modified by either sucrose or urea treatment.
FIG. 8. Cytoskeletal disrupting agents potentiate the tyrosine phosphorylation of caveolin-1 induced by osmotic shock. NIH 3T3 cells were left untreated, treated with 5 g/ml cytochalasin D for 15 min, or treated with 1.5 g/ml nocodazole for 1 h prior to hyperosmotic shock with 600 mM sucrose for 10 min (in the presence of cytochalasin D or nocodazole). Also, cells were treated with 600 mM sorbitol for 10 min. Cells were then subjected to SDS-PAGE/Western blotting analysis. A, caveolin-1 tyrosine phosphorylation. Blots were probed with a phosphospecific caveolin-1 monoclonal antibody (mAb clone 56) (upper panel). Note that pretreatment with cytochalasin D or nocodazole potentiates caveolin-1 phosphorylation. Immunoblotting with mAb 2297, which recognizes total caveolin-1 (lower panel), indicated that caveolin-1 protein expression was not modified by these treatments. B, p38 MAPK activation. Blots were probed with a phosphospecific p38 polyclonal antibody (upper panel). Note that pretreatment with cytochalasin D or nocodazole did not affect the activation of p38 MAP kinase. Immunoblotting with anti-p38 MAP kinase pAb (lower panel), which recognizes total p38 MAP kinase, indicated that total p38 MAP kinase protein expression was not modified by these treatments.

The Introduction of Oxidized Cholesterol into the Plasma Membrane of Living Cells Prevents the Tyrosine Phosphorylation of Caveolin-1
Cholesterol and sphingolipids are major components of caveolae membranes. The role of these lipids in signal transduction mechanisms is still poorly understood. It has been previously reported that cholesterol oxidase, which converts caveolae cholesterol to cholestenone, induces accumulation of caveolin-1 in intracellular membrane compartments (40). Also, the presence of oxidized cholesterol (cholestenone) in caveolae uncouples active platelet-derived growth factor receptors from tyrosine kinase substrates (40). Here, we introduced cholestenone into living cells and observed its effect on caveolin-1 phosphorylation induced by hyperosmotic shock. NIH 3T3 cells were incubated with or without 900 mM cholest-4-en-3-one in the presence of 0.2% methyl-␤-cyclodextrin for 1 h prior to hyperosmotic shock with 600 mM sucrose for 10 min (in the presence of cholest-4-en-3-one and methyl-␤-cyclodextrin). Also, as important internal controls, we incubated NIH 3T3 cells with methyl-␤-cyclodextrin alone or in combination with normal cholesterol prior to sucrose treatment. Cell lysates were then subjected to immunoblot analysis with anti-phosphocaveolin-1 mAb (clone 56) and antiphospho-p38 MAPK antibodies. Fig. 9 shows that only when cells were preincubated with cholest-4-en-3-one in the presence of methyl-␤-cyclodextrin, phosphorylation of caveolin-1 (Fig. 9A, upper panel) and p38 MAP kinase activation (Fig.  9B, upper panel) were significantly reduced. These results suggest that the caveolar lipid composition is very important for activation of the p38 MAP kinase pathway and phosphorylation of caveolin-1.

Multiple Cellular Stressors (Hyperosmotic, Oxidative, and UV) That Activate the p38 MAP Kinase Pathway Also Induce the Phosphorylation of Caveolin-1 on Tyrosine 14
We have shown here that activation of the p38 MAP kinase pathway by hyperosmotic shock is required to promote the tyrosine phosphorylation of caveolin-1. However, other cellular stresses, such as hydrogen peroxide and UV radiation, have also been shown to activate the p38 MAP kinase pathway (10 -13). Thus, we next asked if oxidative stress (hydrogen peroxide) and UV irradiation can induce the phosphorylation of caveolin-1 on tyrosine 14.
Hydrogen Peroxide Treatment-NIH 3T3 cells were treated for 10 min with different concentrations of hydrogen peroxide (1 and 5 mM) (Fig. 10A) or with 5 mM hydrogen peroxide for different periods of time (5, 10, or 20 min) (Fig. 10B). Cells were then subjected to immunoblot analysis with anti-phos- Note that treatment with 900 M cholest-4-en-3-one in the presence of 0.2% methyl-␤-cyclodextrin significantly reduced the phosphorylation of caveolin-1 on tyrosine 14 induced by hyperosmotic shock; in contrast, normal cholesterol had no effect. Immunoblotting with mAb 2297, which recognizes total caveolin-1 (lower panel), indicated that caveolin-1 protein expression was not modified by these treatments. B, p38 MAP kinase activation. Blots were probed with a phosphospecific p38 polyclonal antibody (upper panel). Similarly, only the treatment with 900 M cholest-4-en-3-one in the presence of 0.2% methyl-␤-cyclodextrin significantly reduced activation of p38 MAP kinase induced by hyperosmotic shock. Immunoblotting with anti-p38 MAP kinase pAb (lower panel), which recognizes total p38 MAP kinase, indicated that total p38 MAP kinase protein expression was not modified by these treatments. Cells were then subjected to SDS-PAGE/Western blot analysis using antibodies specific for phosphocaveolin-1 (upper panel) and total caveolin-1 (lower panel). Note that treatment with 5 mM hydrogen peroxide for 5 min is sufficient to induce the tyrosine phosphorylation of caveolin-1 in NIH 3T3 cells. Importantly, total caveolin-1 expression was not affected by hydrogen peroxide treatment. Each lane contains an equal amount of total proteins. phocaveolin-1 IgG (mAb clone 56) and anti-caveolin-1 IgG (pAb N-20). We observed that treatment with either 1 or 5 mM hydrogen peroxide induced the tyrosine phosphorylation of caveolin-1 (Fig. 10A, upper panel). Fig. 10B (upper panel) shows that treatment with 5 mM hydrogen peroxide for 5 min is sufficient to promote caveolin-1 phosphorylation. Total caveolin-1 protein expression was not affected by treatment with hydrogen peroxide (Fig. 10, A and B, lower panels).
We next tested if activation of the p38 MAP kinase pathway was necessary to induce caveolin-1 phosphorylation after treatment with hydrogen peroxide. NIH 3T3 cells were left untreated or treated with 5 mM hydrogen peroxide for 20 min with or without SB203580 inhibitor and subjected to immunoblot analysis. Fig. 11 indicates that hydrogen peroxide activates the p38 MAP kinase pathway and that inhibition of this cellular pathway with the SB203580 inhibitor blocks caveolin-1 phosphorylation.
UV Irradiation-We also examined if phosphorylation of caveolin-1 was induced by UV irradiation. NIH 3T3 cells were irradiated with varying doses of UV-C light and then subjected to immunoblot analysis. Fig. 12 shows that UV-C irradiation also induces the tyrosine phosphorylation of caveolin-1 (Fig. 12A, upper panel) and activates the p38 MAP kinase pathway (Fig. 12B, upper panel). Importantly, UV-C irradiation did not modify total caveolin-1 and p38 protein expression (Fig. 10, A and B, lower panels).
Thus, our results indicate that multiple cellular stressors (hyperosmotic, oxidative, and UV) that activate the p38 MAP kinase pathway also induce the phosphorylation of caveolin-1 on tyrosine 14.

Heat Shock Does Not Induce Caveolin-1 Tyrosine Phosphorylation and Does Not Activate p38 MAP Kinase in NIH 3T3 Cells
In mammalian cells, the heat shock response is essential to protect cells against increases in temperature. The heat shock response promotes the activation and translocation to the nucleus of heat shock transcription factors, which control the expression of heat shock proteins. Heat shock proteins are responsible for cytoprotective effects. To establish whether heat shock can promote phosphorylation of caveolin-1 on tyrosine 14, we incubated NIH 3T3 cells at 43°C for 30 min. Cell lysates were then subjected to immunoblot analysis with anti-phosphocaveolin-1 mAb (clone 56) and anti-phospho-p38 MAPK antibody. Fig. 13 shows that heat shock does not promote phosphorylation of caveolin-1 (Fig. 13A, upper panel) and activation of p38 MAP kinase (Fig. 13B, upper panel). Conversely, hyperosmotic stress induced by sucrose significantly induces caveolin-1 phosphorylation and activation of p38 MAP kinase. Total caveolin-1 (Fig. 13A, lower panel) and p38 (Fig. 13B, lower panel) protein expression remains constant under these experimental conditions.
These results indicate that the p38 MAP kinase/c-Src/phosphocaveolin-1 signaling pathway is not involved in the signal transduction events activated by heat shock in NIH 3T3 cells. DISCUSSION In recent years, it has become evident that a number of extracellular stressors activate specific intracellular pathways, leading to changes in cellular functioning. Hyperosmotic stress, for example, has important implications in human physiology. In fact, the induction of osmoprotective genes has been shown to be part of the protective response of the kidney medulla tubular cells to exposure to hypertonicity during urinary concentration (42). It has been shown that environmental stres- Note that UV radiation of NIH 3T3 cells induces progressive activation of the p38 MAP kinase pathway and tyrosine phosphorylation of caveolin-1. Immunoblotting with mAb 2297, which recognizes total caveolin-1 (A, lower panel), and anti-p38 MAP kinase pAb (B, lower panel), which recognizes total p38 MAP kinase, indicated that total caveolin-1 and p38 MAP kinase protein expression did not change under these experimental conditions. Each lane contains an equal amount of total protein.
p38 MAP kinase, a serine/threonine protein kinase, is phosphorylated by MKK-6, its upstream activator. When activated, p38 MAP kinase phosphorylates downstream elements, leading to modulation of gene transcription. In the present study, we have now linked the activation of the p38 MAP kinase pathway to tyrosine phosphorylation of caveolin-1. Using a phosphospecific monoclonal antibody probe that we previously generated and characterized (32), we demonstrate that hyperosmotic shock transiently stimulates phosphorylation of caveolin-1 on tyrosine 14. Using a specific p38 MAP kinase inhibitor, SB203580, we demonstrate that phosphorylation of caveolin-1 occurs specifically through activation of the p38 MAP kinase pathway. Importantly, when we employed a specific inhibitor of the p42/44 MAP kinase pathway (PD98059), we did not observe any reductions in caveolin-1 tyrosine phosphorylation. These results clearly indicate that activation of the p42-44 MAP kinase pathway is not involved in stimulating caveolin-1 tyrosine phosphorylation. Interestingly, phosphorylation of caveolin-1 and activation of p38 MAP kinase, induced by hyperosmotic shock, are reversible. In fact, when we incubated sucrosestressed NIH 3T3 cells in normal medium for 30 min, caveolin-1 and p38 MAP kinase were no longer phosphorylated. These results suggest that the p38 MAP kinase pathway is activated and caveolin-1 is phosphorylated only after stimulation with the extracellular stressor.
Caveolin-1 was first identified as a major tyrosine-phosphorylated protein in v-Src-transformed chicken embryo fibroblast (18), and we recently reported that activated c-Src constitutively phosphorylates caveolin-1 (32). Here, we demonstrate that c-Src is responsible for caveolin-1 phosphorylation on tyrosine 14 induced by hyperosmotic shock. By employing NIH 3T3 cells stably transfected with an IPTG-inducible vector expressing a dominant negative mutant of c-Src (SRC-RF), we show that induction of this dominant negative form of c-Src, which blocks endogenous c-Src activity, prevents the phosphorylation of caveolin-1 induced by sucrose treatment. Importantly, these results define a novel signaling pathway that is activated by hyperosmotic shock, where p38 MAP kinase, c-Src, and phosphocaveolin-1 represent key elements in this cascade.
Focal adhesions are major cellular sites of tyrosine kinasemediated signal transduction. Interestingly, phosphocaveolin-1 is localized in close proximity to focal adhesions after hyperosmotic shock, as demonstrated by colocalization with paxillin, a marker for focal adhesions. This result is consistent with our recent findings, which demonstrate that when caveolin-1 is constitutively phosphorylated on tyrosine 14 by activated c-Src, it is also localized to focal adhesions (32). Most importantly, these results provide evidence that caveolae near focal adhesions are important intracellular sites involved in the signal transduction mechanisms activated by hyperosmotic stress.
Incubation of cells with sucrose results in an increased intracellular osmolarity and cell shrinkage. Interestingly, when NIH 3T3 cells were incubated with urea, which readily diffuses into the cell and does not change cell volume (44), we did not observe caveolin-1 and p38 MAP kinase phosphorylation. This observation indicates that cell shrinkage is required for induction of caveolin-1 tyrosine phosphorylation and p38 MAP kinase activation, and that increases in intracellular osmolarity are not sufficient to activate this intracellular signaling pathway. Similarly, sorbitol, but not urea, activates the Jak/STAT pathway in COS-7 cells (9). This is also consistent with the notion of focal adhesions as mechanical sensors of osmotic stress.
Microtubules and the actin-cytoskeleton are clearly important for the maintenance of cell shape. To further explore the idea that cell shrinkage is the initial trigger for the activation of p38 MAP kinase and phosphorylation of caveolin-1, we next decided to disrupt microtubules or the actin cytoskeleton by incubating NIH 3T3 cells with nocodazole or cytochalasin D, respectively. Interestingly, treatment with cytochalasin D (and to a lesser extent with nocodazole) dramatically potentiated the tyrosine phosphorylation of caveolin-1 induced by hyperosmotic stress without affecting the activation of p38 MAP kinase. However, treatment with these cytoskeletal disruptors alone was not sufficient to induce the tyrosine phosphorylation in the absence of the osmotic stressor. Since focal adhesions are membrane-attached anchoring points for the actin cytoskeleton, these results are also consistent with a role for focal adhesions in this process.
Cholesterol and glycosphingolipids are essential components of caveolae. Sterol-binding agents, such as filipin, bind to cholesterol and disrupt caveolar structure (45). Also, treatment of living cells with cholesterol oxidase effectively reduces cellular cholesterol (by driving the oxidation of cholesterol to cholest-4en-3-one) and causes the translocation of caveolin-1 from the plasma membrane to intracellular membrane compartments (40). As a consequence, alterations in the lipid components of caveolae membranes could potentially result in profound changes in caveolar functioning. Here, we show that incorporation of cholest-4-en-3-one into the plasma membrane of living cells prevents the activation of p38 MAP kinase and phosphorylation of caveolin-1 induced by hyperosmotic stress, without FIG. 13. Heat shock does not induce caveolin-1 tyrosine phosphorylation and does not activate p38 MAP kinase in NIH 3T3 cells. NIH 3T3 cells were left untreated (CTL), treated with 600 mM sucrose for 10 min, or incubated at 43°C for 30 min. Cell lysates were then subjected to SDS-PAGE/Western blotting analysis. A, caveolin-1 phosphorylation. Blots were probed with a phosphospecific caveolin-1 monoclonal antibody (mAb clone 56) (upper panel). Note that heat shock does not significantly induce the phosphorylation of caveolin-1 on tyrosine 14. Immunoblotting with mAb 2297, that recognizes total caveolin-1 (lower panel), indicated that caveolin-1 protein expression was not modified by heat shock. B, p38 MAP kinase activation. Blots were probed with a phosphospecific p38 polyclonal antibody (upper panel). Similarly, heat shock does not significantly induce the activation of p38 MAP kinase. Immunoblotting with anti-p38 MAP kinase pAb (lower panel), which recognizes total p38 MAP kinase, indicated that total p38 MAP kinase protein expression was not modified by heat shock. affecting their total protein expression (Fig. 9). Thus, introduction of oxidized cholesterol into the plasma membrane is sufficient to uncouple hyperosmotic stress from intracellular signaling. These results are consistent with recent findings that show that introduction of cholest-4-en-3-one into caveolae membranes uncouples platelet-derived growth factor autophosphorylation from tyrosine phosphorylation of neighboring proteins (40). Thus, caveolae lipids are important for organizing and regulating the molecular interactions of multiple signaling pathways.
In different cell types, including hepatocytes and fibroblasts, oxidative stress produces significant damage to the cytoskeleton as well as the plasma membrane (46,47). Interestingly, we demonstrate here that a brief incubation of NIH 3T3 cells with hydrogen peroxide is sufficient to promote caveolin-1 phosphorylation and activate p38 MAP kinase. Thus, perturbation of cytoskeletal elements, by using either cytochalasin D or oxidative stress, is responsible for potentiation and/or activation of this novel signal transduction pathway, leading to activation of p38 MAP kinase and phosphorylation of caveolin-1.
UV irradiation induces apoptosis through activation of the p38 MAP kinase pathway and phosphorylation of the p53 tumor suppressor protein by p38 MAP kinase. Incubation with the p38 MAP kinase inhibitor, SB203580, reduces UV-induced apoptosis, p53 DNA binding activity, and p53-dependent transcription (12,13,48,49). These results suggest that activation of p38 MAP kinase plays a prominent role in UV radiationmediated apoptosis. It has been shown that overexpression of caveolin 1 alone is sufficient to sensitize fibroblasts to ceramide-induced cell death (43), suggesting a possible role for caveolin-1 in mediating apoptosis. In this paper, we demonstrate that UV-C irradiation, in the dose range of 20 J/m 2 , is sufficient to activate p38 MAP kinase and phosphorylate caveolin-1 in NIH 3T3 cells. One possibility is that the tyrosine phosphorylation of caveolin-1 might represent an important step in such UV-mediated apoptosis.
In mammalian cells, the heat shock response is essential to protect cells against increases in temperature. For example, the heat shock response promotes the activation and translocation to the nucleus of heat shock transcription factors, which control the expression of heat shock proteins. Heat shock proteins then confer special cytoprotective effects. However, we show here that heat shock does not promote the activation of p38 MAP kinase or the tyrosine phosphorylation of caveolin-1. These results are consistent with the idea that this novel signaling pathway (p38 MAP kinase 3 c-Src 3 caveolin-1 (Tyr(P) 14 )) is selectively responsive to acute modifications in cell shape and/or disorganization of the cytoskeleton.