This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lee, H. Articles by Lisanti, M. P. Search for Related Content PubMed PubMed Citation Articles by Lee, H. Articles by Lisanti, M. P.

Constitutive and Growth Factor-Regulated Phosphorylation of Caveolin-1 Occurs at the Same Site (Tyr-14) in Vivo: Identification of a c-Src/Cav-1/Grb7 Signaling Cassette

Department of Molecular Pharmacology and The Albert Einstein Cancer Center (H.L., D.V., F.G., M.P.L.) Department of Cell Biology and The Albert Einstein Cancer Center (P.I., P.E.S.) Department of Pathology and The Albert Einstein Cancer Center (D.B.B.) Departments of Developmental & Molecular Biology (DMB) and Medicine; and the Albert Einstein Cancer Center (B.B., R.G.P.) Albert Einstein College of Medicine Bronx, New York 10461
Department of Pathology (D.M.L.) Washington University School of Medicine St. Louis, Missouri 63110
BD Transduction Laboratories (M.T.W., R.C.-G.) Lexington, Kentucky 40511

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Caveolin-1 was first identified as a phospho-protein in Rous sarcoma virus (RSV)-transformed chicken embryo fibroblasts. Tyrosine 14 is now thought to be the principal site for recognition by c-Src kinase; however, little is known about this phosphorylation event. Here, we generated a monoclonal antibody (mAb) probe that recognizes only tyrosine 14-phosphorylated caveolin-1. Using this approach, we show that caveolin-1 (Y14) is a specific tyrosine kinase substrate that is constitutively phosphorylated in Src- and Abl-transformed cells and transiently phosphorylated in a regulated fashion during growth factor signaling. We also provide evidence that tyrosine-phosphorylated caveolin-1 is localized at the major sites of tyrosine-kinase signaling, i.e. focal adhesions. By analogy with other signaling events, we hypothesized that caveolin-1 could serve as a docking site for pTyr-binding molecules. In support of this hypothesis, we show that phosphorylation of caveolin-1 on tyrosine 14 confers binding to Grb7 (an SH2-domain containing protein) both in vitro and in vivo. Furthermore, we demonstrate that binding of Grb7 to tyrosine 14-phosphorylated caveolin-1 functionally augments anchorage-independent growth and epidermal growth factor (EGF)-stimulated cell migration. We discuss the possible implications of our findings in the context of signal transduction.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Caveolae are small omega-shaped indentations of the plasma membrane that have been implicated in signal transduction and vesicular transport processes (1, 2, 3, 4). Caveolae are found in most cell types but are extremely abundant in terminally differentiated cell types: adipocytes (5, 6, 7), simple squamous epithelia (type I pneumocytes and endothelial cells) (8), smooth muscle cells (9), and fibroblasts (10).

Caveolin, a 21- to 24-kDa integral membrane protein, is a principal structural component of caveolae membranes in vivo (11, 12, 13, 14, 15). Recent studies have shown that caveolin is only the first member of a growing gene family of caveolin proteins; caveolin has been retermed caveolin-1. Three different caveolin genes (Cav-1, Cav-2, and Cav-3) encoding four different subtypes of caveolin have been described thus far (2, 3, 4, 16). There are two subtypes of caveolin-1 (Cav-1 and Cav-1ß) that differ in their respective translation initiation sites (17). The tissue distribution of caveolin-2 mRNA and protein is extremely similar to caveolin-1 (7, 18). In striking contrast, caveolin-3 mRNA and protein are expressed mainly in muscle tissue types (skeletal, cardiac, and smooth) (16, 19, 20).

Several independent lines of evidence suggest that caveolins may act as scaffolding proteins within caveolae membranes. In support of this assertion, 1) both the N-terminal and C-terminal domains of caveolin-1 face the cytoplasm (17, 21, 22, 23, 24); 2) caveolins exist within caveolae membranes as high molecular mass oligomers (18, 20, 23, 25, 26); 3) these caveolin oligomers have the capacity to self-associate in vitro to form larger structures that resemble caveolae (23); 4) caveolins copurify with cytoplasmic lipid-modified signaling molecules including heterotrimeric G proteins, Src family tyrosine kinases, and Ras-related GTPases (27, 28, 29, 30, 31); and 5) recombinant caveolin-1 interacts directly with heterotrimeric G proteins (32), c-Src (33) and H-Ras (30). Thus, we and others have proposed the "caveolae signaling hypothesis," which states that caveolins may serve as oligomeric docking sites for organizing and concentrating signaling molecules within caveolae membranes (1, 2, 3, 4, 23).

It has been suggested that fatty acylation may represent a common mechanism for targeting cytoplasmic signaling molecules to caveolae (1, 29, 34). In direct support of this hypothesis, many proteins (G protein -subunits, Src-family tyrosine kinases, Ras-related GTPases, and endothelial nitric oxide synthase) that copurify with caveolins undergo myristoylation, palmitoylation, prenylation, or dual acylation (1, 28, 35). These results indirectly suggest that acylation of lipid-modified signaling molecules may be required for their caveolar targeting in vivo.

Caveolins may facilitate this lipid-based targeting as: 1) caveolin-1 requires cholesterol for insertion into model lipid membranes (36, 37); 2) the membrane-spanning domain of caveolin-1 has a specific and high-affinity for fatty acyl moieties (38, 39); and 3) the C-terminal domain of caveolin-1 undergoes palmitoylation on three cysteines (residues 133, 143, and 156) (22). However, palmitoylation of caveolin-1 is not required for its membrane attachment or targeting to caveolae (22), suggesting that this modification may serve another role—perhaps in trapping other acylated proteins in the vicinity of caveolin molecules.

Historically, caveolin-1 was first identified as a major tyrosine phosphorylated protein in v-Src-transformed chicken embryo fibroblasts (40). Based on this observation, Glenney and co-workers (40) have proposed that caveolin-1 may represent a critical target during cellular transformation. The functional consequences of the phosphorylation of caveolin-1 on tyrosine are not yet known. At steady state, caveolin-1 is not phosphorylated on tyrosine (6). This is in contrast to v-Src-transformed cells where caveolin-1 is constitutively phosphorylated on tyrosine (11). However, caveolin tyrosine phosphorylation also occurs in normal cells but in a tightly regulated fashion (41). If 3T3-L1 adipocytes are serum starved and rapidly stimulated with insulin, caveolin transiently undergoes phosphorylation on tyrosine (41). It has been suggested that insulin-stimulated tyrosine phosphorylation of caveolin occurs indirectly via an endogenous Src-family tyrosine kinase, rather than directly via the insulin receptor. In support of this idea, caveolin-1 can be purified as part of a hetero-oligomeric complex that contains c-Src and other Src-family tyrosine kinases (27, 28, 30, 33).

Recently, we began to study the phosphorylation of caveolin-1 by Src family tyrosine kinases both in vitro and in vivo (42). Using purified recombinant components, we first reconstituted the phosphorylation of caveolin-1 by Src kinase in vitro. Microsequencing of Src-phosphorylated caveolin-1 revealed that phosphorylation occurs within the extreme N-terminal region of full-length caveolin-1, between residues 6–26. This region contains three tyrosine residues at positions 6, 14, and 25. Deletion mutagenesis demonstrated that caveolin-1 residues 1–21 are sufficient to support this phosphorylation event, implicating tyrosine 6 and/or 14. In vitro phosphorylation of caveolin-1-derived synthetic peptides and site-directed mutagenesis directly showed that tyrosine 14 is the principal substrate for Src kinase (42). In support of these observations, tyrosine 14 is the only tyrosine residue within caveolin-1 that bears any resemblance to the known recognition motifs for tyrosine kinases (KYVDSEGHLpY; [RK] - x(2, 3, 4) - [DE] - x(2, 3) - pY). To confirm or refute the in vivo relevance of these in vitro studies, we analyzed the tyrosine phosphorylation of endogenous caveolin-1 in v-Src-transformed NIH 3T3 cells. In vivo, two isoforms of caveolin-1 are known to exist: Cav-1 contains residues 1–178 and Cav-1ß contains residues 32–178. Only Cav-1 underwent tyrosine phosphorylation in v-Src-transformed NIH 3T3 cells, although Cav-1ß is well expressed in these cells. As Cav-1ß lacks residues 1–31 (and therefore tyrosine 14), these in vivo studies indirectly demonstrated the validity of our in vitro studies (42).

Here, we directly examine the phosphorylation of caveolin-1 on tyrosine 14 in vivo. For this purpose, we generated and extensively characterized a novel phospho-specific monoclonal antibody (mAb) probe that selectively recognizes only tyrosine 14-phosphorylated caveolin-1. Our results provide in vivo support for the hypothesis that certain tyrosine-kinase-mediated transmembrane signaling events are initiated at caveolae membranes and that caveolin-1 may function as a signaling molecule.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Characterization of a mAb Probe Specific for Tyrosine 14-Phosphorylated Caveolin-1
We previously showed that tyrosine 14 is the principal site of caveolin-1 phosphorylation by c-Src in vitro (42). To detect this phosphorylation event in vivo, we have generated a novel mAb probe that recognizes only caveolin-1 phosphorylated on tyrosine 14 (see Materials and Methods). A tyrosine-phosphorylated caveolin-1 peptide [SEGHL(pY)TVPI, residues 9–18] was used to immunize mice. Three positive clones were obtained (cl 34A, 49, and 56). Clone 56 gave the strongest immunoreactivity by Western blot analysis and was selected for detailed characterization.

Figure 1 shows the selectivity of antiphosphocaveolin-1 IgG. A glutathione-S-transferase (GST)-fusion protein carrying the N-terminal domain of caveolin-1 (residues 1–101) was purified from normal bacteria (BL21) or from a bacterial strain that harbors a tyrosine kinase (TKB1). Note that clone 56 recognized only tyrosine-phosphorylated caveolin-1 produced in the TKB1 strain, despite equal protein loading (Fig. 1A). We also reconstituted this tyrosine phosphorylation event in vivo. Cos-7 cells were transiently transfected with caveolin-1 and c-Src, alone or in combination. Figure 1B shows that antiphosphocaveolin-1 IgG only recognizes caveolin-1 when it is coexpressed with c-Src. Tyrosine-phosphorylated caveolin-1 migrated as multiple bands, with a major band at approximately 24–28 kDa. Normally, nonphosphorylated caveolin-1 migrates at approximately 22–24 kDa. In addition, we find that when tyrosine 14 is mutated to alanine (Y14A), preventing phosphorylation at this site, this immunoreactivity is completely abolished (Fig. 1B). We further tested the specificity of antiphosphocaveolin-1 IgG by peptide competition with caveolin-1-derived peptides (see Table 1 and Fig. 1C). Immunoreactivity was abolished by preincubation with a 100-fold molar excess of the antigenic peptide (PY14). Importantly, no inhibitory effect was observed with the nonphosphorylated version of the same peptide (Y14) or two irrelevant phosphopeptides (PY100 and PY148). In addition, when caveolins-1, -2, and -3 were cotransfected with activated c-Src (Y529F), only caveolin-1 was detectable with antiphosphocaveolin-1 IgG (data not shown). These results demonstrate the high selectivity of this novel mAb probe for tyrosine 14-phosphorylated caveolin-1.

View larger version (16K): [in this window] [in a new window]   Figure 1. Characterization of a Mouse mAb Probe (cl 56) That Only Recognizes Tyrosine 14-Phosphorylated Caveolin-1 A tyrosine-phosphorylated caveolin-1 peptide (SEGHL(pY)TVPI, residues 9–18) was used to immunize mice and generate a phosphocaveolin-1-specific mAb probe. A, GST-caveolin-1 fusion proteins. A GST-fusion protein carrying the N-terminal domain of caveolin-1 (residues 1–101) was purified from normal bacteria (BL21) or from a bacterial strain harboring a tyrosine kinase (TKB1). Note that clone 56 recognized only tyrosine- phosphorylated caveolin-1 produced in the TKB1 strain, despite equal protein loading. B, Caveolin-1 Y14A mutant. Cos-7 cells were transiently transfected with caveolin-1 and c-Src, alone or in combination. Note that antiphosphocaveolin-1 IgG only recognizes caveolin-1 when it is coexpressed with c-Src. Tyrosine-phosphorylated caveolin-1 migrated as multiple bands, with a major band at approximately 24–25 kDa. Importantly, when tyrosine 14 is mutated to alanine (Y14A), this immunoreactivity is completely abolished. Immunoblotting with mAb 2297 that recognizes total caveolin-1 is shown as a control for equal loading. The mobility of caveolin-1 (Y14A) may be shifted due to a conformational change. C, Peptide competition.Cos-7 cells were transiently transfected with caveolin-1 and c-Src in combination and subjected to preparative SDS-PAGE. After transfer, the nitrocellulose sheet was cut into strips and incubated with antiphosphocaveolin-1 IgG alone or in combination with peptides. The sequences of these four caveolin-1-based peptides (PY14, Y14, PY100, and PY148) are detailed in Table 1.

View larger version (24K): [in this window] [in a new window]   Figure 2. Wild-Type and Mutationally Activated Src Phosphorylate Caveolin-1 on Tyrosine 14 A, IP/Western of caveolin-1 (WT and Y14A). Cos-7 cells were transiently cotransfected with caveolin-1 (WT or Y14A), alone or in combination with c-Src. Thirty-six hours post transfection, the cells were processed for immunoprecipitation. Total caveolin-1 was retrieved using IgG directed against the caveolin-1 C terminus (pAb H-97). Immunoprecipitates were then probed with a mAb that recognizes phosphotyrosine (mAb PY20). Note that when tyrosine 14 is mutated to alanine (Y14A), reactivity of caveolin-1 with PY20 is completely abolished. LC denotes the IgG light chain. B, Activated and kinase-dead c-Src mutants. Cos-7 cells were transiently transfected with caveolin-1 and c-Src [WT, activated (Y529F), and kinase-dead (K297R)]. Tyrosine 14- phosphorylated caveolin-1 was detected by Western blot analysis with antiphosphocaveolin-1 IgG. Activated c-Src phosphorylated caveolin-1 to a much greater extent than wild-type, and kinase-dead c-Src showed no activity. Immunoblotting with mAb 2297 that recognizes total caveolin-1 is shown as a control for equal loading. C, v-Src cells/WB. Lysates were prepared from normal and v-Src-transformed NIH 3T3 cells and subjected to immunoblot analysis. Note that antiphosphocaveolin-1 IgG recognizes tyrosine 14-phosphorylated caveolin-1 in NIH 3T3 cells that stably overexpress the v-Src. In contrast, little or no immunoreactivity was observed in normal NIH 3T3 cells. A comparison between normal and v-Abl-transformed NIH 3T3 cells is also shown. Each lane contains equal amounts of total protein. Immunoblotting with mAb 2297 that recognizes total caveolin-1 is also shown. D, v-Src cells/IP. v-Src-transformed NIH 3T3 cells were lysed and immunoprecipitated with an excess of mAb 2234 that recognizes total caveolin-1 (residues 1–21) or an excess of mAb 56 that recognizes phosphocaveolin-1 (residues 9–18). After SDS-PAGE and transfer to nitrocellulose, the amount of caveolin-1 precipitated was estimated by Western blotting with rabbit anticaveolin-1 IgG (N-20). Note that that only a small fraction ( 5%) of caveolin-1 undergoes phosphorylation on tyrosine 14.

To roughly estimate the percentage of total cellular caveolin-1 that undergoes phosphorylation on tyrosine 14, we employed an immunoprecipitation approach. v-Src- transformed NIH 3T3 cells were lysed and subjected to immunoprecipitation with mAb 2234, which recognizes total caveolin-1 (residues 1–21), or mAb 56, which recognizes phosphocaveolin-1 (residues 9–18). The amount of caveolin-1 precipitated was estimated by Western blotting with rabbit anticaveolin-1 IgG. Using this approach, Fig. 2D shows that only a fraction ( 5%) of total caveolin-1 undergoes phosphorylation on tyrosine 14.

Membrane attachment of c-Src is mediated in part by N-terminal myristoylation of the protein product (44). Interestingly, this modification is required for the targeting of c-Src to caveolae membrane domains (45). Caveolin-1 also undergoes lipid modification (22). However, the functional role of caveolin-1 palmitoylation remains largely unknown. Thus, we next examined the possible requirement of these lipid modifications for the tyrosine phosphorylation of caveolin-1.

To evaluate the role of c-Src myristoylation, Cos-7 cells were cotransfected with caveolin-1 and c-Src [wild-type (WT) or myristoylation-deficient (Myr-minus)]. Figure 3A shows that myristoylation of c-Src is required for phosphorylation of caveolin-1 on tyrosine 14. These results are consistent with the observation that myristoylation of c-Src is required for its targeting to caveolae (45).

View larger version (9K): [in this window] [in a new window]   Figure 3. Lipid Modification Is Required for Src-Induced Phosphorylation of Caveolin-1 on Tyrosine 14 A, Myristoylation minus c-Src. Cos-7 cells were cotransfected with caveolin-1 and c-Src (WT or Myr-minus). Note that myristoylation of c-Src is required for phosphorylation of caveolin-1 on tyrosine 14. B, Palmitoylation-deficient caveolin-1. To evaluate the role of caveolin-1 palmitoylation, we cotransfected Cos-7 cells with wild-type c-Src and caveolin-1 (WT or Pal-minus). Note that palmitoylation of caveolin-1 is required for or greatly facilitates phosphorylation of caveolin-1 on tyrosine 14. In panels A and B, immunoblotting with mAb 2297 that recognizes total caveolin-1 is shown as a control for equal loading.

Colocalization of Tyrosine 14-Phosphorylated Caveolin-1 with Markers of Focal Adhesions in Cells Expressing Activated Src
To examine the localization of tyrosine-phosphorylated caveolin-1, we cotransfected Cos-7 cells with caveolin-1 alone and in combination with either wild-type c-Src or activated c-Src (Y529F). These cells were then doubly immunostained with antiphosphocaveolin-1 mouse mAb (cl 56) and an anticaveolin-1 rabbit polyclonal antibody (pAb) (N-20). These bound primary antibodies were visualized by using distinctly tagged fluorescent secondary antibodies (see Materials And Methods).

Figure 4 shows the distribution of caveolin-1 phosphorylated on tyrosine 14. In cells transfected with caveolin-1 alone, little or no immunostaining with antiphosphocaveolin-1 was observed (Fig. 4A). In cells cotransfected with caveolin-1 plus wild-type c-Src, immunostaining with antiphosphocaveolin-1 appeared as large fluorescent dots in the center and along the cell periphery (Fig. 4B). Immunostaining with anticaveolin-1 also appeared punctate, but the dots were of a much smaller size. These two labeling patterns appeared distinct, suggesting that tyrosine-phosphorylated caveolin-1 is present in a different region of the cell or that only a subpopulation of caveolae are tyrosine phosphorylated. In cells cotransfected with caveolin-1 plus activated c-Src (Y529F), immunostaining with antiphosphocaveolin-1 also appeared as large dots, but these dots were confined to the cell periphery and appeared to coincide with focal contacts or adhesions (Fig. 4C). A very similar staining pattern was observed in v-Src-transformed NIH 3T3 cells (Fig. 5A).

View larger version (69K): [in this window] [in a new window]   Figure 4. Localization of Tyrosine 14-Phosphorylated Caveolin-1 in Cos-7 Cells Transiently Transfected with c-Src To examine the localization of tyrosine-phosphorylated caveolin-1, we cotransfected Cos-7 cells with caveolin-1 and either WT or activated c-Src. Cells were doubly immunostained with antiphosphocaveolin-1 mouse mAb (cl 56; at left) and an anticaveolin-1 rabbit pAb (N-20; at right). Bound primary antibodies were visualized by using distinctly tagged fluorescent secondary antibodies. A, Caveolin-1 alone.In Cos-7 cells transfected with caveolin-1 alone, little or no immunostaining with antiphosphocaveolin-1 was observed. B, Caveolin-1 plus wild-type c-Src. In cells cotransfected with caveolin-1 plus WT c-Src, immunostaining with antiphosphocaveolin-1 appeared as large fluorescent dots in the central region of the cell and along the cell periphery. C, Caveolin-1 plus activated c-Src. In cells cotransfected with caveolin-1 plus activated c-Src (Y529F), immunostaining with antiphosphocaveolin-1 appeared as large dots confined to the cell periphery that were reminiscent of focal contacts or adhesions. In panels A-C, immunostaining with anticaveolin-1 IgG appeared punctate (right) and did not coincide with the labeling observed with antiphosphocaveolin-1.

View larger version (29K): [in this window] [in a new window]   Figure 5. Tyrosine 14-Phosphorylated Caveolin-1 Is Localized in Close Proximity to Focal Adhesions in NIH 3T3 Cells Stably Expressing v-Src Cells were doubly immunostained with two distinct primary antibodies as detailed below. Bound primary antibodies were visualized by using distinctly tagged fluorescent secondary antibodies. In panels A, B, and D, images were acquired with a MR600 confocal fluorescence microscope (Bio-Rad Laboratories, Inc.). In panels C and E, images were acquired with a IX80 microscope (Olympus Corp.) with a 60x Plan Neofluar objective and a Photometrics cooled CCD camera with a 35-mm shutter. A, Antiphosphocaveolin-1 (mAb) and anticaveolin-1 (pAb). Note that staining with antiphosphocaveolin-1 appeared as large dots, but these dots were confined to the cell periphery and appeared to coincide with focal contacts or adhesions. B, Antiphosphocaveolin-1 (mAb) and antiphospho tyrosine (pAb). Note that tyrosine-phosphorylated caveolin-1 is colocalized with the major sites of tyrosine phosphorylation. These sites of tyrosine phosphorylation activity are known to correspond to focal adhesions (48 ). C, Anti-phospho-caveolin-1 (mAb) and antipaxillin. Note that caveolin-1 and paxillin colocalize to a significant extent. These results suggest that caveolae in close proximity to focal adhesions are preferentially phosphorylated in v-Src-transformed cells. D, Anticaveolin-1 (pAb) and antiphosphotyrosine (PY20 mAb). Note that little or no colocalization is observed. These results indicate that the bulk of total caveolin-1 is not localized in proximity to the major sites of tyrosine phosphorylation in vivo. E, Color overlay. A color version of panel C is shown along with the merged image to better demonstrate the colocalization of phosphocaveolin-1 and paxillin.

In contrast, when v-Src-transformed NIH 3T3 cells were double-labeled with anticaveolin-1 (N-20 pAb) and antiphosphotyrosine (PY20 mAb), little or no colocalization was observed (Fig. 5D). These results indicate that in v-Src-transformed cells the bulk of total caveolin-1 is not localized in proximity to the major sites of tyrosine phosphorylation in vivo. This is consistent with the observation that only a fraction of total cellular caveolin-1 ( 5%) is tyrosine phosphorylated in v-Src-transformed cells (Fig. 2D).

Tyrosine-Phosphorylated Caveolin-1 Correctly Forms High Molecular Mass Oligomers and Is Targeted to Caveolae-Enriched Membrane Domains
As tyrosine-phosphorylated caveolin-1 was localized in close proximity to focal adhesions, we next assessed the biochemical properties of tyrosine-phosphorylated caveolin-1 using established assay systems. These approaches have been used previously to characterize the properties of total caveolin-1. Caveolin-1 normally forms stable high molecular mass homooligomeric complexes (7, 18). Once these homooligomers reach the plasma membrane, they become Triton insoluble due to their incorporation into caveolae membranes (49). This resistance to detergent solubilization is thought to reflect the local lipid microenvironment in which these caveolin homooligomers are embedded. In contrast, caveolin-1 associated with the Golgi complex remains Triton soluble (50). Thus, we next examined the effects of the tyrosine phosphorylation of caveolin-1 on its 1) Triton insolubility; 2) oligomeric state; and 3) caveolar targeting, using normal and v-Src- transformed NIH 3T3 cells.

In both normal and v-Src-transformed NIH 3T3 cells, total caveolin-1 was >95% Triton insoluble (not shown), formed high molecular mass oligomers of the correct size, and was targeted to caveolae-enriched membrane fractions (Fig. 6, A and B). Interestingly, virtually identical results were obtained with caveolin-1 phosphorylated on tyrosine 14, indicating that caveolin-1 remains caveolae associated even after tyrosine phosphorylation in v-Src-transformed cells. As the subcellular distribution of tyrosine- phosphorylated caveolin-1 coincides with focal adhesions by fluorescence microscopy (Fig. 6), these biochemical results are consistent with the idea that caveolae in close proximity to focal adhesions are preferentially targeted for phosphorylation in v-Src- transformed cells.

View larger version (15K): [in this window] [in a new window]   Figure 6. Tyrosine 14-Phosphorylated Caveolin-1 Correctly Forms High Molecular Mass Oligomers and Is Targeted to Caveolae-Enriched Membrane Domains In both panels, tyrosine phosphorylated caveolin-1 was detected with mAb 56 and total caveolin-1 was visualized using mAb 2297. A, Velocity gradient analysis. Normal and v-Src-transformed NIH 3T3 cells were solubilized, loaded atop a 5–40% sucrose density gradient, and subjected to centrifugation for 10 h. Twelve fractions were recovered and a 20 µl aliquot from each fraction was analyzed by SDS-PAGE/Western blotting. Arrows mark the positions of molecular mass standards. Note that caveolin-1 behaves as a high molecular mass oligomer of approximately 150–300 kDa. B, Caveolar targeting. Normal and v-Src-transformed NIH 3T3 cells were homogenized in a buffer containing 1% Triton X-100 and subjected to sucrose density gradient centrifugation. Twelve 1-ml fractions were collected, and an aliquot of each fraction ( 20 µl) was analyzed by SDS-PAGE/Western blotting. As expected, caveolin-1 is highly enriched in fractions 4–5, which represent the caveolae-enriched membrane fractions. Note that both total caveolin-1 and tyrosine 14- phosphorylated caveolin-1 form high molecular mass oligomers of the correct size and are targeted to caveolae-enriched membrane fractions in normal and v-Src-transformed cells.

This insulin-stimulated tyrosine phosphorylation of caveolin is thought to occur via an endogenous member of the Src family of tyrosine kinases (41). However, it remains unknown which tyrosine residue is phosphorylated in response to insulin stimulation. Also, as caveolins-1 and -2 form a tight heterooligomeric complex and stably coimmunoprecipitate (18, 26, 56, 57), it remains unclear whether caveolin-1 or caveolin-2 is actually the target of tyrosine phosphorylation in adipocytes.

To address these issues, we treated 3T3-L1 adipocytes with insulin or a variety of other growth factors. Figure 7 shows that a brief treatment (10 min) of 3T3-L1 adipocytes with insulin dramatically stimulated phosphorylation of caveolin-1 on tyrosine 14, while the other factors had little or no effect (see panels A and C). Interestingly, 3T3-L1 adipocytes showed an increased basal level of tyrosine phosphorylation of caveolin-1, as compared with Cos-7 cells or NIH 3T3 cells. Virtually identical results were obtained with 3T3-L1 fibroblasts, indicating that this event also occurs in the preadipocyte (Fig. 7A). Tyrosine phosphorylation of caveolin-1 was concentration dependent and increased from 5–150 nM insulin (not shown). Insulin-stimulated phosphorylation of caveolin-1 on tyrosine 14 was apparent after only 1 min of treatment, reached maximal levels at 5 min, and declined to basal levels by 120 min (Fig. 7B). These results are consistent with the idea that tyrosine phosphorylation of caveolin-1 is an early event in insulin receptor signal transduction.

View larger version (24K): [in this window] [in a new window]   Figure 7. Insulin-Stimulated Phosphorylation of Caveolin-1 on Tyrosine 14 in 3T3-L1 Adipocytes In panels A–C, cells were lysed and subjected to immunoblot analysis with antiphosphocaveolin-1 IgG. Immunoblotting with mAb 2297 that recognizes total caveolin-1 is shown as a control for equal loading. A, Growth-factor stimulation. 3T3-L1 adipocytes (left) or fibroblasts (right) were serum starved for 3 h and subsequently treated with insulin (150 nM) or a variety of other growth factors (PDGF (10 ng/ml), bFGF (5 ng/ml), TNF (10 nM), and IL6 (5 nM)). Note that a brief treatment (10 min) of 3T3-L1 adipocytes with insulin dramatically stimulated phosphorylation of caveolin-1 on tyrosine 14, while the other factors had little or no effect (left). Virtually identical results were obtained with 3T3-L1 preadipocytes (right). B, Insulin time course. 3T3-L1 adipocytes were serum starved for 3 h and subsequently treated with insulin. Note that insulin (150 nM)-stimulated tyrosine phosphorylation of caveolin-1 was apparent after only 1 min of treatment, reached maximal levels at 5 min, and declined to basal levels by 120 min. C, Insulin vs. EGF stimulation. 3T3-L1 adipocytes were serum starved for 3 h and subsequently treated with EGF (100 ng/ml) or insulin (150 nM). Note that a brief treatment (5 min) of 3T3-L1 adipocytes with insulin stimulated phosphorylation of caveolin-1 on tyrosine 14, while treatment with EGF had no effect.

EGF-Stimulated Phosphorylation of Caveolin-1 on Tyrosine 14 in A431 Cells
A431 cells are a human epidermoid carcinoma-derived cell line that has been used extensively to study EGF-mediated signal transduction. As A431 cells are known to express both EGF receptor (EGF-R) (58) and caveolin-1 (21, 59), we used these cells to evaluate the effects of EGF stimulation on the tyrosine phosphorylation of caveolin-1. For this purpose, A431 cells were serum starved and treated with EGF (100 ng/ml) for various times. As a positive control for these studies, we used an antibody that only detects the activated phosphorylated form of the EGF-R (58). Figure 8A shows that a brief treatment (5 min) of A431 cells with EGF activated the EGF-R and dramatically stimulated phosphorylation of caveolin-1 on tyrosine 14. This is in contrast to our results with 3T3-L1 adipocytes (Fig. 7C), which do not show tyrosine phosphorylation of caveolin-1 in response to EGF, despite the fact that these cells are known to express the EGF-R and are EGF responsive. Thus, growth factor-stimulated tyrosine phosphorylation of caveolin-1 may be dependent on cell-type specific coupling factors or simply dependent on the relative expression levels of different growth factor receptors in a given cell type.

View larger version (94K): [in this window] [in a new window]   Figure 8. EGF-Stimulated Phosphorylation of Caveolin-1 on Tyrosine 14 in A431 Cells A, EGF time course. A431 cells were serum-starved for 12–16 h and subsequently treated with EGF (100 ng/ml) for 0–4 h. Cells were lysed and subjected to immunoblot analysis with antiphosphocaveolin-1 IgG. As a positive control, we used an antibody that only detects the activated phosphorylated form of the EGF-R (mAb 74). Note that a brief treatment (5 min) of A431 cells with EGF dramatically activated EGF-R and stimulated the phosphorylation of caveolin-1 on tyrosine 14. Immunoblotting with phospho-independent antibodies to EGF-R (mAb 13) and caveolin-1 (mAb 2297) was used to visualize total EGF-R and total caveolin-1 levels. Each lane contains an equal amount of total protein. B, Immunolocalization. Confluent A431 cells were serum starved and incubated in the presence or absence of EGF (100 ng/ml) for 5 min. Note that tyrosine-phosphorylated caveolin-1 accumulated at the plasma membrane and was highest in areas of cell-cell contact. Importantly, little or no immunostaining was observed using antiphosphocaveolin-1 IgG if A431 cells were not treated with EGF. C, Colocalization with paxillin. Subconfluent A431 cells were serum-starved and subsequently incubated in the presence of EGF (100 ng/ml) for 5 min. Double labeling with antiphosphocaveolin-1 and antipaxillin revealed that caveolin-1 and paxillin colocalize to a significant extent. Arrowheads point to areas of colocalization. These results suggest that caveolae in close proximity to focal adhesions are preferentially phosphorylated in EGF-stimulated A431 cells.

Vanadate Treatment Induces the Accumulation of Tyrosine 14-Phosphoryated Caveolin-1 in Normal NIH 3T3 Cells
It remains unknown what ligands might stimulate tyrosine phosphorylation of caveolin-1 in other cell types, such as NIH 3T3 cells. For example, treatment of NIH 3T3 cells with various growth factors [EGF, PDGF, and basic fibroblast growth factor (bFGF)] had no apparent effect on the tyrosine phosphorylation of caveolin-1 (not shown). To investigate whether caveolin-1 undergoes tyrosine phosphorylation in normal NIH 3T3 cells, these cells were treated with vanadate (100 µM), a tyrosine phosphatase inhibitor. If caveolin-1 undergoes tyrosine phosphorylation, vanadate should prevent dephosphorylation and allow detection of accumulated tyrosine-phosphorylated intermediates. Figure 9A shows that treatment with vanadate for increasing times leads to the accumulation of tyrosine 14- phosphorylated caveolin-1. As in v-Src-transformed cells, several caveolin-1 bands of approximately 22–28 kDa were apparent (compare Figs. 2C and 9A).

View larger version (14K): [in this window] [in a new window]   Figure 9. Vanadate-Treatment Induces the Accumulation of Tyrosine 14-Phosphoryated Caveolin-1 in Normal NIH 3T3 Cells To investigate whether caveolin-1 undergoes tyrosine phosphorylation in normal NIH 3T3 cells, these cells were treated with vanadate (100 µM), a tyrosine phosphatase inhibitor. A, Western blot analysis. Note that treatment with vanadate for increasing times leads to the accumulation of tyrosine 14-phosphorylated caveolin-1. As in v-Src- transformed cells, several caveolin-1 bands of approximately 22–28 kDa are apparent. B, Immunofluorescence. NIH 3T3 cells were treated with vanadate for 2 h or left untreated. Note that tyrosine-phosphorylated caveolin-1 accumulated at the plasma membrane and was highest in areas of cell-cell contact (upper panels). Importantly, little or no immunostaining was observed using antiphosphocaveolin-1 IgG if normal NIH 3T3 cells were not treated with vanadate. In contrast, immunostaining of vanadate-treated NIH 3T3 cells with anti-caveolin-1 IgG demonstrated that caveolin-1 was predominantly localized to a perinuclear compartment and at areas of cell-cell contact (lower panels). Thus, tyrosine 14-phosphorylated caveolin-1 accumulated predominantly at the plasma membrane and did not strictly follow the distribution of total caveolin-1 in vanadate-treated NIH 3T3 cells. Images were acquired with an MR600 confocal fluorescence microscope(Bio-Rad Laboratories, Inc.).

Purified Caveolae-Enriched Membranes Contain a Kinase That Phosphorylates Caveolin-1 on Tyrosine 14 in Vitro
Purified caveolae membranes are dramatically enriched in caveolin-1 and have been shown to contain tyrosine kinase activity (6, 27, 41, 60, 61). Also, a number of Src-family tyrosine kinases (c-Src, c-Yes, Lyn, Fyn, Lck, and c-Fgr) and receptor tyrosine kinases (Ins-R, EGF-R, and PDGF-R) copurify with caveolae and caveolin-1 under these conditions (1, 27, 28, 30, 31, 61, 62, 63, 64, 65, 66, 67, 68). Thus, we examined the ability of caveolin-1 to be phosphorylated on tyrosine 14 using purified caveolae-enriched membrane domains.

Caveolae-enriched domains were purified from murine lung tissue using an established protocol (28) and incubated in kinase reaction buffer in the presence or absence of exogenous ATP (1 mM). Figure 10A (left panel) shows that addition of ATP dramatically induced phosphorylation of caveolin-1 on tyrosine 14. Thus, a kinase activity with the same specificity as is observed in vivo is present within purified caveolae membranes and copurifies with caveolin-1. Other tyrosine-phosphorylated proteins were visualized by immunoblotting with an antiphosphotyrosine mAb (PY20); the major tyrosine-phosphorylated proteins appeared in the 50- to 60-kDa and the 200-kDa range, consistent with the autophosphorylation of Src-family tyrosine kinases and receptor tyrosine kinases, respectively (Fig. 10A, right panel). [It should be noted that a 22- to 25-kDa band corresponding to caveolin-1 was also detectable with PY20 on longer exposures (not shown)].

View larger version (14K): [in this window] [in a new window]   Figure 10. Purified Caveolae-Enriched Membranes Contain a Kinase That Phosphorylates Caveolin-1 on Tyrosine 14 in Vitro A, In vitro kinase assay. Caveolae-enriched domains were purified from murine lung tissue using an established protocol (28 ) and incubated in kinase reaction buffer in the presence or absence of exogenous ATP (1 mM). After 10 min at 25 C, the reaction was halted by the addition of SDS sample buffer/boiling and samples were subjected to Western blotting. Note that addition of ATP dramatically induced phosphorylation of caveolin-1 on tyrosine 14 (left panel). The total pattern of tyrosine-phosphorylated proteins was visualized by immunoblotting with an antiphosphotyrosine mAb (PY20) (right panel). Immunoblotting with mAb 2297 that recognizes total caveolin-1 is shown as a control for equal loading (middle panel). B, Inhibitor profile. Note that pretreatment of caveolae-enriched membranes with either AG1478 or A9 (5 µM) had little or no effect (upper panel). In contrast, pretreatment with PP2 (5 µM) dramatically inhibited the phosphorylation of caveolin-1 on tyrosine 14. Immunoblotting with mAb 2297 that recognizes total caveolin-1 is shown as a control for equal loading (lower panel).

Identification of Grb7 as a pY14-Caveolin-1 Binding Partner in Vitro and in Vivo
One known function of tyrosine phosphorylation is to confer binding to SH2 domain- containing proteins. To identify SH2 domain proteins that bind to tyrosine-phosphorylated caveolin-1, we used an established in vitro binding approach. Briefly, we prepared nonphosphorylated and tyrosine-phosphorylated caveolin-1 from bacteria (Fig. 1A; see Materials and Methods). These purified GST-Cav-1 (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101) fusion proteins were then incubated with lysates prepared from normal NIH 3T3 cells. After extensive washing, the bound material was subjected to Western blot analysis with a panel of antibodies directed against SH2 domain-containing proteins (listed in Table 2).

View larger version (24K): [in this window] [in a new window]   Figure 11. Tyrosine 14-Phosphorylated Caveolin-1 Binds Grb7 in a Phosphorylation-Dependent Manner A, Grb7 binding in vitro. Nonphosphorylated and tyrosine-phosphorylated GST-Cav-1 (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 ) fusion proteins were prepared and incubated with cell lysates derived from normal NIH 3T3 cells. After binding, washing, and elution, the eluates were subjected to immunoblot analysis with antibodies directed against 20 different SH2 domain-containing proteins (listed in Table 2). Left panel, Note that only the binding of Grb7 was observed. Importantly, Grb7 bound to tyrosine-phosphorylated GST-Cav-1 (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 ), but not to GST alone, or nonphosphorylated GST-Cav-1 (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 ). Right panel, Other closely related proteins, such as Grb2 and Grb14, did not show any binding activity. B, Grb7 binding in vivo. Left panel, Cos-7 cells were co-transfected with cDNAs encoding caveolin-1 and Grb7, in the presence of c-Src WT or its corresponding empty vector (pUSE). Forty-eight hours post transfection, lysates were prepared and subjected to immunoprecipitation with antibodies directed against Grb7. These immunoprecipitates were then subjected to Western blot analysis with anticaveolin-1 IgG or antiphosphocaveolin-1 IgG. Note that both antibodies detect caveolin-1 coimmunoprecipitating with Grb7, but not with beads alone or an irrelevant IgG; and this binding occurs only in cells cotransfected with c-Src WT. Right panel, Note that this binding is strictly dependent on tyrosine 14, as caveolin-1 (Y14A) fails to coimmunoprecipitate with Grb7. C, Importance of tyrosine 14. Left panel, Cos-7 cells were cotransfected with cDNAs encoding caveolin-1 and Grb-7, in the presence of c-Src WT. Lysates were prepared and subjected to immunoprecipitation with antibodies directed against Grb7. Immunoprecipitates were then subjected to Western blotting with antiphosphocaveolin-1 IgG. Before immunoprecipitation, competing peptides were added to these lysates (detailed in Table 1). Note that the PY14 peptide completely blocks this interaction, while the corresponding nonphosphorylated peptide (Y14) or an irrelevant phosphopeptide (PY100) has no effect. Right panel, Cos-7 cells were transiently transfected with the cDNA encoding Grb-7. Lysates were prepared and incubated with biotinylated phosphopeptides prebound to streptavidin beads. After washing, these precipitates were subjected to Western blotting with anti-Grb7 IgG. Note that the caveolin-1-derived phosphopeptide containing tyrosine 14 (PY14) effectively pulls down Grb7, while another irrelevant caveolin-1-derived phosphopeptide containing tyrosine 100 (PY100) does not.

To further investigate the specific role of caveolin-1 tyrosine 14, we next used a peptide-based approach. Cos-7 cells were cotransfected with the cDNAs encoding caveolin-1 and Grb-7, in the presence of c-Src. Lysates were prepared and competing peptides were added (detailed in Table 1), and they were immunoprecipitated with antibodies directed against Grb7. These immunoprecipitates were subjected to Western blotting with antiphosphocaveolin-1 IgG. Figure 11C (left panel) shows that addition of the PY14 peptide completely blocks this interaction, while the addition of the corresponding nonphosphorylated peptide (Y14) or an irrelevant phosphopeptide (PY100) has no effect. These results again implicate phosphorylated tyrosine 14 in this binding event.

To assess whether phosphotyrosine 14 and its surrounding sequence is sufficient for Grb7 binding, Cos-7 cells were transiently transfected with the cDNA encoding Grb-7. Lysates were then incubated with biotinylated phosphopeptides bound to streptavidin beads. After washing, these precipitates were subjected to Western blot analysis with anti- Grb7 IgG. Figure 11C (right panel) shows that a caveolin-1-derived phosphopeptide containing tyrosine 14 (PY14) effectively pulls down Grb7, while another irrelevant caveolin-1-derived phosphopeptide (PY100) does not. Thus, caveolin-1 phosphotyrosine 14 and its surrounding sequence are sufficient for Grb7 binding.

Binding of Grb7 to Tyrosine 14-Phosphorylated Caveolin-1 Functionally Stimulates Anchorage-Independent Growth and Cell Migration
What is the function of tyrosine-phosphorylated caveolin-1 (Y14) ? As tyrosine-phosphorylated caveolin-1 is localized at focal adhesions, we speculated that it might play a role in regulating anchorage-dependent growth and/or cell migration. To test this hypothesis, we used 293T cells because they lack endogenous caveolin-1 expression and they undergo transfection with high efficiency, a prerequisite for these cotransfection studies. In this cellular context, we compared the activity of wild-type caveolin-1 with a mutant form of caveolin-1 (Y14A) that is unable to undergo tyrosine phosphorylation at residue 14.

Anchorage-Independent Growth Studies
Figure 12A shows that expression of caveolin-1 alone inhibited foci formation in 293T cells. This is consistent with previous reports showing that recombinant expression of caveolin-1 in other cell types inhibits anchorage-independent growth (72, 73, 74). Expression of c-Src alone had no effect on foci formation; it has been previously shown that overexpression of c-Src is not sufficient to mediate cell transformation (75). In contrast, coexpression of c-Src and caveolin-1 stimulated foci formation by approximately 2-fold, as compared with mock-transfected or cells transfected with Src alone. These results suggest that tyrosine phosphorylation of caveolin-1 can stimulate anchorage-independent growth.

View larger version (11K): [in this window] [in a new window]   Figure 12. Binding of Grb7 to Tyrosine 14-Phosphorylated Caveolin-1 Stimulates Both Anchorage-Independent Growth and Cell Migration 293T cells were cotransfected with the cDNAs encoding c-Src, Grb7, caveolin-1 (WT), and caveolin-1 (Y14A), alone or in combination. Cells were then subjected to focus formation or cell migration assays. A, Anchorage-independent growth. Note that coexpression of c-Src and caveolin-1 stimulated foci formation by approximately 2-fold, as compared with mock-transfected cells or cells transfected with Src alone. In addition, coexpression of c-Src, caveolin-1 (WT), and Grb7 dramatically stimulated foci formation by approximately 3.5-fold, as compared with mock-transfected cells, and by approximately 7-fold as compared with cells cotransfected with c-Src and Grb7. In striking contrast, coexpression of c-Src, caveolin-1 (Y14A), and Grb7 had no effect on foci formation. B, Cell migration. Note that coexpression of c-Src, caveolin-1 (WT), and Grb7 dramatically stimulated cell migration by approximately 2- to 3-fold, as compared with mock-transfected cells, cells transfected with caveolin-1 alone, or cells cotransfected with c-Src and Grb7. In contrast, coexpression of c-Src, caveolin-1 (Y14A), and Grb7 had little or no effect on cell migration.

Cell Migration Studies
293T cells are known to express endogenous EGF-R and to undergo EGF-stimulated cell migration. Thus, we next assessed the effect of the Src/Cav-1/Grb7 signaling cassette on this process (Fig. 12B). Interestingly, coexpression of c-Src, caveolin-1 (WT), and Grb7 dramatically stimulated cell migration by approximately 2- to 3-fold, as compared with mock-transfected cells, cells transfected with caveolin-1 alone, or cells transfected with c-Src and Grb7 alone. In contrast, coexpression of c-Src, caveolin-1 (Y14A), and Grb7 had little or no effect on cell migration, as compared with mock-transfected cells, and behaved the same as cells transfected with c-Src and Grb7 alone. Thus, binding of Grb7 to tyrosine 14-phosphorylated caveolin-1 augments both anchorage-independent growth and EGF-stimulated cell migration in 293T cells.

It should also be noted that expression of caveolin-1 alone (either WT or Y14A) had little or no effect on cell migration in 293T cells. This is in contrast to our previous results employing MTLn3 cells, a metastatic rat mammary adenocarcinoma cell line (74). In MTLn3 cells, caveolin-1 expression blocked EGF-stimulated cell migration (74). Thus, the negative regulatory effect of caveolin-1 on cell migration may be cell type specific.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
The oncogene v-Src arose by viral transduction of the normal cellular gene c-Src (76, 77). Thus, it is thought that viral tyrosine kinases largely induce transformation by intercepting cell-regulatory mechanisms that are normally under the control of tyrosine phosphorylation. In support of this notion, v-Src and c-Src appear to differ primarily in enzymatic activity, but not in their substrate specificity (78, 79). This difference in enzymatic activity can be attributed to the loss of tyrosine residue 527 within v-Src (residue 529 in human c-Src). Phosphorylation at this C-terminal site within c-Src by CSK normally inactivates c-Src (80, 81).

Few physiologically relevant v-Src substrates are known. Caveolin-1 is one of these substrates (11, 12, 13, 40). For example, caveolin-1 copurifies as a heterooligomeric complex with c-Src and other Src-family tyrosine kinases (27, 28, 30, 35). This is consistent with the general idea that v-Src phosphorylates the normal targets of c-Src or related Src-family tyrosine kinases, but in an unregulated fashion.

Here, we studied the phosphorylation of caveolin on tyrosine in vivo by employing a novel phosphospecific mAb probe that selectively recognizes tyrosine 14-phosphorylated caveolin-1. We reconstituted this event by cotransfecting caveolin-1 with c-Src. We observed that myristoylation of c-Src and palmitoylation of caveolin-1 are both required for Src-induced phosphorylation of caveolin-1 on tyrosine 14. In cells transfected with activated forms of Src [either c-Src (Y529F) or v-Src], tyrosine-phosphorylated caveolin-1 was localized mainly in close proximity to focal adhesions, the major cellular sites of tyrosine kinase-mediated signal transduction.

We also evaluated the tyrosine phosphorylation of caveolin-1 in other cell types. In 3T3-L1 adipocytes, insulin dramatically increased the phosphorylation of caveolin-1 on tyrosine 14, with maximal stimulation at 5 min of treatment. In A431 cells, EGF stimulation greatly increased the tyrosine phosphorylation of caveolin-1, which was localized near focal adhesions. Incubation of NIH 3T3 cells with vanadate (a tyrosine phosphatase inhibitor) drove the accumulation of tyrosine-phosphorylated caveolin-1. Also, purified caveolae membranes contain a kinase activity that phosphorylates caveolin-1 on tyrosine 14 in vitro. This phosphorylation event was inhibited by a selective inhibitor of Src-family tyrosine kinases (PP2). Importantly, these results demonstrate that constitutive (via activated Src) and regulated (via insulin, EGF, or vanadate treatment) phosphorylation of caveolin-1 occur at the same site, i.e. tyrosine 14, in vivo.

Caveolin-1 exists as a high molecular mass oligomer containing approximately 14–16 individual caveolin molecules (23, 25), and these caveolin oligomers have the capacity to self-associate into larger complexes (23). By analogy with other known tyrosine phosphorylation events, we speculated that tyrosine-phosphorylated caveolin-1 could potentially serve as an oligomeric docking-site for SH2-domain signaling molecules (23, 42)—much like activated growth factor receptors that oligomerize, undergo tyrosine phosphorylation, and recruit SH2 domain-containing proteins to the cytoplasmic surface of the plasma membrane (Fig. 13). In support of this hypothesis, we show here that phosphorylation of caveolin-1 on tyrosine 14 functionally confers binding to Grb7, an SH2 domain-containing adapt