And Sepsis-Induced Lung Inflammation and Mediates Myd88

The Journal of Immunology

Caveolin-1 Tyr14 Phosphorylation Induces Interaction with TLR4 in Endothelial Cells and Mediates MyD88-Dependent Signaling and Sepsis-Induced Lung Inﬂammation

Hao Jiao,*,†,1 Yang Zhang,*,†,1 Zhibo Yan,* Zhen-Guo Wang,* Gongjian Liu,† Richard D. Minshall,*,‡ Asrar B. Malik,‡ and Guochang Hu*,‡

Activation of TLR4 by the endotoxin LPS is a critical event in the pathogenesis of Gram-negative sepsis. Caveolin-1, the signaling protein associated with caveolae, is implicated in regulating the lung inflammatory response to LPS; however, the mechanism is not understood. In this study, we investigated the role of caveolin-1 in regulating TLR4 signaling in endothelial cells. We observed that LPS interaction with CD14 in endothelial cells induced Src-dependent caveolin-1 phosphorylation at Tyr14. Using a TLR4-MD2- CD14–transfected HEK-293 cell line and caveolin-1–deficient (cav-12/2) mouse lung microvascular endothelial cells, we demonstrated that caveolin-1 phosphorylation at Tyr14 following LPS exposure induced caveolin-1 and TLR4 interaction and, thereby, TLR4 activation of MyD88, leading to NF-kB activation and generation of proinflammatory cytokines. Exogenous expression of phosphorylation-deficient Y14F caveolin-1 mutant in cav-12/2 mouse pulmonary vasculature rendered the mice resistant to LPS compared with reintroduction of wild-type caveolin-1. Thus, caveolin-1 Y14 phosphorylation was required for the interaction with TLR4 and activation of TLR4-MyD88 signaling and sepsis-induced lung inflammation. Inhibiting caveolin-1 Tyr14 phosphorylation and resultant inactivation of TLR4 signaling in pulmonary vascular endothelial cells represent a novel strategy for preventing sepsis-induced lung inflammation and injury. The Journal of Immunology, 2013, 191: 6191–6199.

epsis from severe bacterial infection is the major cause of recruits a number of cytoplasmic adaptor proteins containing TIR morbidity and mortality in critically ill patients (1). LPS domains, including MyD88, TIR-associated protein/MyD88 adaptor- S (also known as endotoxin), the key structural component like, TRIF, and TLR-associated molecule (9, 10). Recruitment of the of the cell wall of Gram-negative bacteria, is a highly potent adaptor protein MyD88 initiates early activation of NF-kB, whereas trigger of cytokine release through TLR4 (2). Release of LPS into recruitment of TRIF activates a MyD88-independent pathway, the blood activates a generalized inflammatory response, leading leading to delayed activation of NF-kB (11) and activation of IFN to multiple organ dysfunction syndrome (2–4). The binding of regulatoryfactor3(12,13).UponactivationofTLR4withLPS, LPS to TLR4 in endothelial cells plays a particularly important MyD88 induces the association with IRAK-4 and IRAK-1 and re- role in the inflammatory response. LPS activates endothelial cells cruitment of TNFR-associated factor 6 (TRAF6) to IRAK-1 (14-16). through a receptor complex consisting of TLR4, CD14, and MD-2 The IRAK-4/IRAK-1/TRAF6 complex dissociates from TLR4 and (5, 6). The outcome of activation of this signaling pathway is the interacts with the TGF-b–activated kinase 1 complex, which, in turn, production of proinflammatory mediators by endothelial cells, activates IkB kinases, leading to phosphorylation and degradation of upregulation of endothelial adhesion molecules, loss of endothe- IkB and release and translocation of NF-kB to the nucleus, as well as lial barrier integrity, leukocyte recruitment, and, ultimately, cel- induces expression of proinflammatory genes (9). lular and organ injury (7, 8). LPS engagement of TLR4 also Caveolin-1 (Cav-1), a 21–24-kDa protein, is the principal structural and signaling component of caveolae that oligomerizes, and the oligomers are inserted into caveolae membrane (17, 18). Cav-1 *Department of Anesthesiology, University of Illinois at Chicago College of Medi- functions as a scaffolding protein and interacts with multiple mole- cine, Chicago, IL 60612; †Department of Anesthesiology, Xuzhou Medical Col- cules (19, 20). Recent studies implicated a key role for Cav-1 in ‡ lege, Xuzhou 221002, China; and Department of Pharmacology, University of regulating innate immunity and inflammation (21–23), but the results Illinois at Chicago College of Medicine, Chicago, IL 60612 have been inconclusive as to how Cav-1 functions. Downregulation 1H.J. and Y.Z. equally contributed to this work. of Cav-1 expression in murine alveolar and peritoneal macrophages Received for publication April 2, 2013. Accepted for publication October 15, 2013. using small interfering RNAs (siRNAs) increased LPS-induced This work was supported by National Science Foundation of China Grant 81070058 production of proinflammatory cytokines (TNF-a and IL-6) (24, and National Institutes of Health/National Heart, Lung, and Blood Institute Grant HL104092 (both to G.H.). 25), indicating an anti-inflammatory role for Cav-1. Studies using 2/2 Address correspondence and reprint requests to Dr. Guochang Hu, Department of Cav-1–deficient (cav-1 ) mice demonstrated that Cav-1 induced Pharmacology (m/c 868), University of Illinois at Chicago College of Medicine, 835 upregulation of NF-kBactivityandincreasedlunginflammatory South Wolcott Avenue, Chicago, IL 60612. E-mail address: [email protected] response following LPS stimulation (26). NF-kB activity and lung The online version of this article contains supplemental material. inflammation and injury following LPS challenge in cav-12/2 mouse Abbreviations used in this article: Cav-1, caveolin-1; eNOS, endothelial NO syn- lungs were reduced compared with controls (26). This was ascribed thase; HEK-TLR4, HEK-293 cell line stably expressing TLR4 and MD2 in combination with CD14; m, mouse; MLMVEC, mouse lung microvascular endothelial cell; to greater production of NO due to deinhibition of endothelial NO ROS, reactive oxygen species; SEAP, secreted embryonic alkaline phosphatase; synthase (eNOS) in the Cav-1–deleted mice. eNOS activation sec- siRNA, small interfering RNA; TRAF6, TNFR-associated factor 6; WT, wild-type; ondary to Cav-1 deficiency also impaired IRAK-4 activity through Y14F–Cav-1, nonphosphorylatable caveolin-1 Y14F mutant. nitration of IRAK-4 in response to LPS (27). Although the above Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 studies described a role for increased eNOS activity following the www.jimmunol.org/cgi/doi/10.4049/jimmunol.1300873 6192 CAVEOLIN-1 PHOSPHORYLATION ENHANCES TLR4 SIGNALING deletion of Cav-1 in attenuating inflammatory signaling pathways, (Cell Biologics, Chicago IL) (28). Briefly, lung tissues were minced and they did not address the important signaling function of Cav-1, which digested with collagenase A (1.0 mg/ml) for 45–60 min at 37˚C. Endo- is regulated by phosphorylation at Tyr14 (21–23), or how Cav-1 thelial cells were purified using an anti–PECAM-1 mAb magnetic bead 14 (BD Pharmingen, San Diego, CA) separation technique and allowed to signaling by Tyr phosphorylation influences TLR4 activation by grow for 3–4 d in growth medium. The purity of endothelial cells reached LPS. In the current study, we explored the role of Cav-1 phosphor- .95%. For monolayer cultures, the cells were plated on fibronectin-coated ylation at Tyr14 in activating the LPS–TLR4–NF-kB–signaling dishes in endothelial cell growth medium 2 supplemented with the Bullet pathway in endothelial cells. We found that Cav-1 Tyr14 phosphor- Kit (BioWhittaker, Walkersville, MD) and 10% FBS, incubated (37˚C) under a humidified atmosphere of 5% CO2–95% air, and used at passages ylation was essential for TLR4 signaling and, hence, mediated the three to five. inflammatory response in endothelial cells. These results raise the The HEK-293 cell line stably expressing TLR4 and MD2 in combination possibility that blockade of Cav-1 phosphorylation at Tyr14 in lung with CD14 (293-hTLR4A-MD2-CD14; referred to as “HEK-TLR4” in this vascular endothelial cells would be beneficial in preventing lung article) was purchased from InvivoGen (San Diego, CA). HEK-TLR4 cells inflammatory injury during sepsis. were maintained in high-glucose DMEM (Cellgro, Manassas, VA) supplemented with penicillin, streptomycin, and 10% FBS. Methods and Materials Treatment with LPS and Src inhibitor Reagents and chemicals LPS was diluted with the appropriate basal culture media without sup- plements and added to the cells, which were preincubated in serum-deprived LPS (Escherichia coli 0127:B8; purity . 99%) and PP2 (purity $ 98%) media (without LPS-binding protein) for 2 h. Confluent MLMVEC mon- were obtained from Sigma-Aldrich. MyD88 siRNA (mouse [m]), TRIF olayers were incubated with Src inhibitor PP2 (15 mM) in HBSS for siRNA (m), CD14 siRNA (m), and control siRNA, anti–a-tubulin, anti– 15 min at 37˚C, followed by two washes with HBSS. IRAK-1, anti-TLR4, and anti–IkB-a mAbs were obtained from Santa Cruz Biotechnology. Cav-1 and b-actin mAbs were purchased from BD Bio- sciences. Anti-MyD88 mAb and anti–pY418-Src polyclonal Ab were ob- Cav-1 cDNA transfection tained from Abcam and Cell Signaling, respectively. Lipofectamine 2000 HEK-TLR4 cells and cav-12/2 MLMVECs were transfected with pcDNA6 was purchased from Invitrogen. plasmid vector containing cDNAs of Myc-tagged WT Cav-1 or nonphosphorylatable Cav-1 Y14F mutant (Y14F–Cav-1). HEK-TLR4 cells Mice and lung inflammatory injury were transfected using a mixture of 4 mg cDNA and 10 ml Lipofectamine 2/2 Wild-type (WT) B6/129SJ (cav-1+/+) and cav-12/2 mice were purchased 2000 (Invitrogen) in each well of a six-well plate. Cav-1 MLMVECs from The Jackson Laboratory (Bar Harbor, ME). Mice were housed in were transfected using a Nucleofector Kit (Lonza, Cologne, Germany), microisolator cages under specific pathogen–free conditions, fed with according to the manufacturer’s protocol. Successful expression of exog- autoclaved food, and used in experiments at 8–12 wk of age. All studies enous Cav-1 was confirmed by Western blot analysis of cell lysates. using mice were approved by the University of Illinois Institutional Animal Care and Use Committee. Mice were anesthetized using either inhaled siRNA transfection of endothelial cells isoflurane or ketamine injected i.p. (60 mg/kg). Acute lung injury was MyD88 siRNA, TRIF siRNA, or CD14 siRNA (Santa Cruz Biotechnology), induced by i.p. injection of LPS (10 mg/kg). at a concentration of 50 nM, was added to 50–70% confluent MLMVECs to Cell culture deplete the respective protein using the protocol provided by the manufacturer. Successful depletion of MyD88, TRIF, and CD14 was confirmed Mouse lung microvascular endothelial cells (MLMVECs) were isolated by Western blot analysis. All experiments were performed at 48 h post- from 3–5-d-old WT and cav-12/2 mouse pups, as previously described transfection.

FIGURE 1. LPS activation of Src induces phosphorylation of Cav-1 at Tyr14.(A) LPS induced phosphorylation of Cav-1 and Src in a time-dependent manner. Serum-starved MLMVECs were exposed to LPS (1 mg/ml) for the indicated times. After stimulation, cells were harvested for Western blotting. Rep- resentative Western blots for phosphorylated Cav-1 [p (Y14) Cav-1] and Src [p(Y418) Src](left panel). Pro- tein quantification by densitometry (n =3)(right panel). The density of proteins in each untreated control group was used as a standard (1 arbitrary unit) to compare the relative density of the other groups. (B) LPS dose dependently induced phosphorylation of Cav-1 and Src. MLMVECs were stimulated with LPS (0.1–2 mg/ml) for 30 min and harvested for Western blotting. Representative Western blots for phosphorylated Cav-1 [p(Y14) Cav-1] and Src [p(Y418) Src](left panel). Protein quantification by densitometry (n =3) (right panel). The density of proteins in each untreated control group was used as a standard (1 arbitrary unit) to compare the relative density of the other groups. (C) Effect of PP2 on LPS-induced Cav-1 phosphorylation. Representative Western blots for phosphorylated Cav-1 [p(Y14) Cav-1] (left panel). Protein quantification by densitometry (n =3)(right panel). The density of proteins in the vehicle (DMSO)-treated group was used as a standard (1 arbitrary unit) to compare the relative density of the other groups. *p , 0.05 versus untreated groups (A, B) or DMSO-treated group (C), †p , 0.05 versus corresponding DMSO-treated groups. The Journal of Immunology 6193

Liposome preparation and in vivo gene delivery perimental groups. The Student t test was performed for paired samples.

2/2 Parameter changes between different groups over time were evaluated by Rescue studies were made in mouse lungs from cav-1 mice by a two-way ANOVA with repeated measures. Differences between survival liposome-mediated plasmid DNA transfection. Myc-tagged WT Cav-1 and curves were determined by the Mantel–Cox test. Data are expressed mutant Y14F–Cav-1 cDNAs in pcDNA6 plasmid vector were used for as mean 6 SEM where applicable. Differences were considered significant Cav-1–repletion studies. The liposome was prepared as previously de- at p , 0.05. scribed and injected i.v. in mice (29, 30). Briefly, the liposome consisting of dimethyldioctadecylammonium bromide and cholesterol (1:1 molar ratio) was dried using a Rotavaporator (Brinkmann, Westbury, NY) and Results dissolved in 5% glucose. Each animal received 200 ml of the cDNA–li- LPS induces Cav-1 phosphorylation at Tyr14 via Src activation posome complexes containing 50 ml plasmid cDNA (2 mg/kg) and 150 ml in endothelial cells liposomes. Successful and equal expression of exogenous Cav-1 was confirmed by Western blot of lung homogenates and immunostaining of We observed that LPS treatment induced phosphorylation of Cav-1 lung tissue. All mice appeared normal from the time of injection until at Tyr14 in a time-dependent manner in MLMVECs (Fig. 1A). sacrifice in every experiment. Densitometric analysis showed that the relative intensity of phos- Immunohistochemistry Paraformaldehyde-fixed 5-mm-thick lung sections were immunostained with anti–VE-cadherin or Cav-1 Abs using a standard protocol. Lung sections were visualized with a LSM510 confocal microscope (Carl Zeiss). Western blot analysis and immunoprecipitation Cells were lysed in RIPA buffer, and lung samples were homogenized in buffer containing 1.5% Triton X-100, 0.5% deoxycholic acid, 0.1% SDS, phosphatase inhibitor I and II (1:100), and protease inhibitor mixture. Protein concentration was determined with a bicinchoninic acid kit (Thermo Scientific). For immunoprecipitation, cells were lysed in buffer containing 50 mM Tris-HCl (pH 8), 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 20 mM NaF, 1 mM PMSF, 1 mM Na3VO4, and protease inhibitor mixture. Samples were precleared for 1 h at 4˚C using 1 mg control IgG (normal rabbit IgG) together with protein A/G PLUS-agarose; subsequently, the supernatants were incubated overnight at 4˚C with primary Ab, followed by the addition of 25 ml protein A/G PLUS-agarose and further incubation at 4˚C for 2 h. Equal amounts of protein from the homogenates were electrophoresed on SDS-PAGE gels (10–12%) and subsequently trans- ferred to 0.22-mm nitrocellulose membranes. The membranes were blocked with 5% nonfat milk and probed with the appropriate Abs. Protein bands were detected using the ECL SuperSignal reagent (Pierce, Rockford, IL). Relative band densities of the various proteins were measured from scanned films using National Institutes of Health ImageJ Software. NF-kB activity assay NF-kB–induced secreted embryonic alkaline phosphatase (SEAP) activity was assessed using a HEK-Blue Detection kit (InvivoGen), according to the manufacturer’s protocols. HEK-TLR4 cells were transiently cotrans- fected with pcDNA6 plasmid vector containing cDNAs of Cav-1 (WT or Y14F) and NF-kB–inducible SEAP reporter system (pNiFty2-SEAP). Twenty-four hours after transfection, cells were detached, resuspended in PBS, and counted. A cell suspension of 4 3 105 cells/ml was added to each well of a 96-well plate. The plates were incubated at 37˚C in a 5% CO2 incubator for 16 h after 1 mg/ml LPS was added to each well. Twenty microliters of supernatant was added to 180 ml HEK-Blue Detection so- lution (InvivoGen) in a 96-well plate and incubated for 1–2 h for color development. SEAP activity was assessed by reading the OD at 655 nm with a microplate reader. Each experiment was performed in triplicates. Cytokine generation The levels of TNF-a and IL-6 in the supernatants of the experimental media were determined using commercial ELISA kits (BioLegend, San Diego, CA), according to the manufacturer’s instructions. Each value represents the mean of triplicate determinations. Survival studies Cav-12/2 mice transfected with the pcDNA6 plasmid vector containing WT Cav-1 cDNA or Y14F–Cav-1 cDNA were injected i.p. with LPS at a dose of 20 mg/kg in 0.9% saline (12 mice/group), monitored every 4 h, and sacrificed FIGURE 2. Src activation and Cav-1 phosphorylation induced by LPS when moribund or after 96 h, when the observations were terminated (31). are MyD88 and TRIF independent and CD14 dependent. MLMVECs were Lung edema measurement transfected with scrambled, MyD88, TRIF, or CD14 siRNA. Lysates of Wet/dry lung weight ratio was used as an index of lung edema. At the end of cells, which were treated as indicated, were analyzed by Western blotting the experiment, lungs were weighed, dried, and reweighed. For determi- analysis at 15, 30, and 60 min after LPS (1 mg/ml) stimulation. (A) Effects nation of dry lung weight, lung tissue was dried in an oven to a constant of depletion of MyD88 with a specific siRNA on phosphorylation of Src weight (60˚C for 72 h). and Cav-1. (B) Effects of depletion of TRIF with a specific siRNA on phosphorylation of Src and Cav-1. (C) Effects of depletion of both MyD88 Statistical analysis and TRIF with specific siRNAs on phosphorylation of Src and Cav-1. (D) One-way ANOVA and the Student’s Newman–Keuls test for post hoc Effects of depletion of CD14 with a specific siRNA on phosphorylation of comparisons were used to determine differences between control and ex- Src and Cav-1. All blots are representative of three separate experiments. 6194 CAVEOLIN-1 PHOSPHORYLATION ENHANCES TLR4 SIGNALING phorylation increased 4–5-fold after LPS stimulation. Cav-1 phos- To further elucidate the mechanism of Src activation and Cav-1 phorylation increased within 15 min, peaked at 90 min, and phosphorylation in response to LPS stimulation, we investigated remained high for up to 2 h (Fig. 1A). Coincident with the increase the role of the MyD88- and TRIF-dependent pathways. We found in p(Y14)–Cav-1, we also observed the activation of c-Src in en- that Src activation and Cav-1 phosphorylation were MyD88 and dothelial cells following LPS stimulation, as measured by p(Y418)- TRIF independent, because knockdown of MyD88 (Fig. 2A), Src phosphoimmunoblot (Fig. 1A). We exposed MLMVECs to TRIF (Fig. 2B), or both (Fig. 2C) with specific siRNAs did not different concentrations of LPS (0.1–2 mg/ml) and found that LPS alter Src phosphorylation on Y418 and Cav-1 phosphorylation at increased phosphorylation of Src and Cav-1 in a dose-dependent Tyr14. However, knockdown of CD14 (a component of the LPS manner (Fig. 1B). LPS-induced Cav-1 phosphorylation at Tyr14 was binding complex on the endothelial cell surface) markedly de- abolished by Src family kinase inhibitor PP2 (Fig. 1C). creased phosphorylation of Src and Cav-1 (Fig. 2D), indicating

FIGURE 3. LPS-induced Cav-1 Tyr14 phosphorylation mediates NF-kB activation. HEK-TLR4 cells were transiently transfected with Myc-tagged WT and phosphorylation-defective Y14F–Cav-1 mutant. At 24 h posttransfection, cells were exposed to LPS (1 mg/ml) for the indicated times (A, B). (A)Cav-1 phosphorylation enhanced IkB-a degradation following LPS stimulation in HEK-TLR4 cells. Representative Western blots for IkB-a (left panel). IkB-a quantification by densitometry (n =3)(right panel). The density of proteins in the untreated control group was used as a standard (1 arbitrary unit) to compare the relative density of the other groups. (B) Cav-1 phosphorylation enhanced NF-kB–induced SEAP activity following LPS stimulation in HEK-TLR4 cells. Cells were transfected with pNiFty2-SEAP. NF-kB–inducible SEAP reporter activity was determined at 16 h after LPS stimulation. Quantifications represent means of three independent measurements with three replicates each time. (C) Expression of TLR4 and exogenous Cav-1 in cav-12/2 MLMVECs. MLMVECs isolated from cav-12/2 mice were transfected with WT or Y14F Myc–Cav-1 cDNA (C–F). (D) Cav-1 phosphorylation enhanced IkB-a degradation following LPS stimulation in MLMVECs. Representative Western blots for IkB-a (left panel). IkB-a quantification by densitometry (n =3)(right panel). The density of proteins in untreated control (WT) group was used as a standard (1 arbitrary unit) to compare the relative density of the other groups. (E and F) Cav-1 phosphorylation facilitated TNF-a (E) and IL-6 (F) production following LPS stimulation in MLMVECs. Transfected MLMVECs were treated with LPS (1 mg/ml) for 24 h. The levels of TNF-a and IL-6 in the cell culture medium were measured by ELISA. Data are representative of three independent experiments. *p , 0.05 versus untreated control group (A) or WT control group (B), †p , 0.05 versus corresponding WT group. ND, Not detected. The Journal of Immunology 6195 that endothelial cell CD14 was required for Src activation and Cav-1 phosphorylation following LPS stimulation. Cav-1 phosphorylation at Tyr14 signals LPS-induced NF-kB activation To address the role of Tyr14 Cav-1 phosphorylation in regulating the LPS/TLR4-signaling pathway, we first used the well- characterized HEK-TLR4 cells expressing human TLR4 at high levels. HEK-TLR4 cells express low levels of endogenous Cav-1 (Fig. 3A). When the HEK-TLR4 stable cell line was transiently transfected with vectors containing WT Cav-1 or Y14F–Cav-1, we observed equal expression of the exogenous Cav-1 proteins (WT or Y14F mutant). IkB-a, the isoform of inhibitor of transcription factor NF-kB family, undergoes tyrosine phosphorylation and degradation following LPS stimulation and, hence, permits release and nuclear translocation of NF-kB, which controls expression of a number of proinflammatory genes (9, 10). We found that, FIGURE 5. Phosphorylation of Cav-1 promotes association of MyD88 with TLR4. Cav-12/2 MLMVECs grown to 90–95% confluence were in response to LPS, Cav-1 phosphorylation at Tyr14 increased, k a transiently transfected with WT or Y14F–Cav-1 cDNA in pcDNA6 plas- concurrently with a decrease in I B- protein expression in HEK- mid vector. At 24 h posttransfection, cells were stimulated with LPS (1 mg/ TLR4 cells transfected with WT Cav-1, whereas no phosphory- ml) for 1 h. Cell lysates were immunoprecipitated with Abs against TLR4 lation of Cav-1 was observed as the result of Y14F mutation, and or MyD88 and immunoblotted with anti-MyD88 or anti-TLR4 Ab. Total there was less IkB-a protein degradation in Y14F–Cav-1–trans- TLR4 or MyD88 served as a loading control. fected cells (Fig. 3A). We also measured NF-kB activity in HEK- TLR4 cells transfected with WT or Y14F–Cav-1 cDNA in We next determined the role of Tyr14 Cav-1 phosphorylation in pcDNA6 plasmid vector using the reporter plasmid pNiFty2- regulating TLR4 signaling using MLMVECs from cav-12/2 mice, SEAP, which contains an engineered ELAM promoter combin- which were transfected with pcDNA6 plasmid vector containing k ing five NF- B sites (GGGGACTTTCC) with the proximal WT Cav-1 or Y14F–Cav-1 cDNA, with equivalent protein k ELAM promoter. As seen in Fig. 3B, NF- B activity (reported by expressions of Cav-1 and TLR4 (Fig. 3C). LPS challenge caused SEAP activity) was increased by 10-fold after LPS treatment for IkB-a degradation in both WT endothelial cells (Supplemental k 2 2 16 h in WT Cav-1–transfected cells, whereas NF- B activation Fig. 1A) and WT Cav-1-transfected cav-1 / endothelial cells was markedly reduced in cells transfected with Y14F–Cav-1 transfected with WT Cav-1 cDNA (Fig. 3D). IkB-a peak degra- cDNA. dation occurred at 60 min, and the protein expression returned to normal levels at 120 min. However, much less degradation of IkB-a protein was found in the Cav-1–null cells (Supplemental Fig. 1A) and cav-12/2 cells transfected with Y14F–Cav-1 cDNA (Fig. 3D) compared with WT endothelial cells and cav-12/2 cells transfected with WT Cav-1 cDNA (Fig. 3D). These results further show that Cav-1 Tyr14 phosphorylation regulates the activation of the TLR4–NF-kB–signaling pathway in response to LPS.

FIGURE 4. Phosphorylation of Cav-1 promotes association of Cav-1 with TLR4. (A) LPS induced an association of Cav-1 with TLR4 in MLMVECs from Cav-1+/+ mice. MLMVECs were treated with LPS (1 mg/ml) for 1 h. Cell lysates from MLMVECs were immunoprecipitated with Ab against Cav-1 or TLR4. The levels of Cav-1 and TLR4 were measured by immunoblotting. Total Cav-1 or TLR4 served as a loading control. (B) Phosphorylation of Cav-1 promoted the association of Cav-1 with TLR4. Cav-12/2 MLMVECs, grown to 90–95% conﬂuence, were transiently transfected with WT or Y14F Cav-1 cDNA in pcDNA6 plasmid FIGURE 6. Cav-1 phosphorylation induces IRAK-1 uncoupling from vector. At 24 h posttransfection, cells were exposed to LPS (1 mg/ml) for 1 membranes. (A) Decreased membrane-bound IRAK-1 after LPS stimula- h and then harvested for coimmunoprecipitation using Abs against Cav-1 tion in cav-1+/+ MLMVECs. (B) Changes in membrane-bound IRAK-1 in or TLR4. The levels of Cav-1 and TLR4 were measured by immuno- cav-12/2 MLMVECs transfected with WT or Y14F–Cav-1 cDNA after blotting. Total Cav-1 or TLR4 served as a loading control. LPS treatment for 30 or 60 min. 6196 CAVEOLIN-1 PHOSPHORYLATION ENHANCES TLR4 SIGNALING

We next examined the role of Tyr14 Cav-1 phosphorylation in Ab after LPS treatment (Fig. 4B). However, LPS-induced asso- the production of cytokines TNF-a and IL-6 in mouse lung en- ciation of TLR4 with Cav-1 was abrogated in cav-12/2 dothelial cells. TNF-a and IL-6 were undetectable at basal con- MLMVECs transfected with Y14F–Cav-1 cDNA, indicating that dition in cav-12/2 cells transfected with cDNA of WT or phosphorylation of Cav-1 at Tyr14 facilitated Cav-1–TLR4 inter- Y14F– Cav-1 cDNA. After LPS stimulation, the levels of TNF-a actions. We also immunoprecipitated TLR4 protein and exam- and IL-6 were elevated in WT MLMVECs (Supplemental Fig. 1B, ined whether Cav-1 association with TLR4 was altered by the 1C) and cav-12/2 MLMVECs transfected with WT Cav-1 cDNA phosphorylation-defective mutant in response to LPS. LPS treat- (Fig. 3E, 3F), whereas Cav-1–null MLMVECs (Supplemental Fig. ment increased the binding of Cav-1 to TLR4 in cav-12/2 cells 1B, 1C) and cav-12/2 MLMVECs transfected with Y14F–Cav-1 transfected with WT Cav-1 cDNA, whereas no increase in Cav-1– cDNA (Fig. 3E, 3F) showed much lower production of TNF-a and TLR4 interactions were found in cav-12/2 cells transfected with IL-6, supporting our implication that Tyr14 Cav-1 phosphorylation Y14F–Cav-1 cDNA (Fig. 4B). mediates LPS-induced proinflammatory cytokine release. Next, we examined the effects of Tyr14 phosphorylation of Cav- 1 on the downstream TLR4-signaling pathway. Using the Ab Tyr14 phosphorylation of Cav-1 induces the association of against TLR4, we observed increased coprecipitation of the TLR4 and MyD88 TLR4-binding partner MyD88 in cav-12/2 MLMVECs trans- To delineate the mechanism by which Tyr14 Cav-1 phosphorylation fected with WT Cav-1 cDNA after LPS treatment, whereas LPS mediates NF-kB activation, we examined the effects of Tyr14 did not increase MyD88 interactions with TLR4 in cav-12/2 cells phosphorylation of Cav-1 on the interaction of TLR4 and Cav-1 in transfected with Y14F–Cav-1 cDNA (Fig. 5). In contrast, upon endothelial cells. In line with a previous finding (24), TLR4 was immunoprecipitating MyD88, we found increased binding of Cav- detected in immunoprecipitates using anti–Cav-1 Ab, and the 1 and TLR4 in the immunoprecipitates of cav-12/2 cells trans- association of TLR4 with Cav-1 was increased greatly by LPS fected with WT Cav-1 cDNA compared with those transfected treatment (Fig. 4A). We precipitated cell lysates with anti-TLR4 with Y14F–Cav-1 cDNA (Fig. 5). These findings indicate the Ab and also observed increased precipitation of Cav-1 by LPS crucial role for Tyr14 phosphorylation of Cav-1 in regulating the treatment (Fig. 4A). In cav-12/2 MLMVECs transfected with WT association of MyD88 and TLR4 and, thereby, in TLR4 signaling Cav-1 cDNA, more TLR4 protein was precipitated by anti–Cav-1 through the MyD88-dependent pathway.

FIGURE 7. Phosphorylation of Cav-1 increases mortality and cytokine production in mice challenged with LPS. Cav-12/2 mice were injected i.v. with liposomes containing Myc-tagged WT or Y14F–Cav-1 cDNA. After 48 h, mice were challenged with LPS (10 mg/kg, i.p.). (A) Western blots show an increase in Cav-1 phosphorylation in cav-1+/+ and WT Cav-1– transfected lung homogenates induced by LPS, the absence of Cav-1 phosphorylation in Cav-1 null lungs, and exogenous expression of WT and Y14F–Cav-1 in cav-12/2 lungs. Note that LPS induced phosphorylation of reconstituted WT Cav-1 in the mouse lung. (B) IkB-a degradation following LPS stimulation. (C) Decreased mortality in mice transfected with Y14F– Cav-1 cDNA. Cav-12/2 mice were transfected with WT or Y14F–Cav-1 cDNA. At 48 h posttransfection, mice were injected with LPS (20 mg/kg, i.p.) and housed under normal conditions (n = 12 mice/group). (D) Cav-1 Y14F mutation reduces lung wet/dry ratio after LPS (10 mg/kg, i.p.) challenge (n = 6/each group). Sera from mice were collected at 6 h after PBS alone or LPS (10 mg/kg, i.p.) challenge, and the levels of TNF-a (E) and IL-6 (F) were measured by ELISA (n = 3/group). *p , 0.05 versus control group (without LPS), †p , 0.05 versus corresponding WT group. The Journal of Immunology 6197

Phosphorylation of Cav-1 at Tyr14 induces IRAK-1 dissociation and TNF-a (Supplemental Fig. 3C) and IL-6 (Supplemental Fig. from Cav-1–IRAK-1 complex in endothelial cells 3D) production, whereas these parameters in cav-12/2 lungs de- IRAK-1 was shown to interact with Cav-1 in immune cell plasma creased dramatically. membrane (32); thus, we investigated the possible role for Cav-1 Tyr14 phosphorylation following LPS stimulation in mediating Discussion Cav-1–IRAK-1 interactions and, thereby, TLR4 signaling in en- In this study, we identified the requisite and sufficient role for Cav-1 14 dothelial cells. In cav-1+/+ MLMVECs, 5 min following LPS phosphorylation at Tyr in activating TLR4 signaling and the stimulation, Cav-1 in the membrane fraction was phosphorylated at production of proinflammatory cytokines in endothelial cells. The Tyr14. IRAK-1 was found in immunoprecipitated protein complex bacterial endotoxin LPS stimulated Cav-1 phosphorylation at 14 with anti–Cav-1 Ab at the basal condition. Although IRAK-1 pro- Tyr via CD14-dependent Src activation in endothelial cells. The tein expression was not altered in total lysates, membrane-bound phosphorylated Cav-1, in turn, facilitated the interaction of TLR4 k IRAK-1 was dissociated from Cav-1 30 min after LPS treatment. and MyD88, resulting in NF- B activation and the release of a Western blotting of whole-cell lysates confirmed that the level of TNF- and IL-6 in response to LPS stimulation. To our knowl- IkB-a was decreased at 30 min following LPS stimulation (Fig. 6A), edge, our results show for the first time the key role for LPS- 14 suggesting that IRAK-1 dissociation from Cav-1 elicited IkBdeg- induced phosphorylation of Cav-1 at Tyr in endothelial cells radation. A similar result was mirrored in cav-12/2 MLMVECs in activating the TLR4–MyD88–signaling pathway. Thus, this 14 transfected with WT Cav-1 cDNA. In cav-12/2 cells expressing function of Tyr Cav-1 phosphorylation in regulating TLR4 sig- exogenous WT Cav-1 cDNA, the reduction in membrane-associated naling may be particularly important in the host-defense function IRAK-1 was also accompanied by a decrease in IkB-a (Fig. 6B). By of endothelial cells, because the endothelium expresses the highest contrast, in cav-12/2 cells expressing exogenous Y14F–Cav-1 amount of Cav-1 (33). 14 cDNA, the level of membrane-bound IRAK-1 remained the same In addition to the role of Tyr phosphorylation of Cav-1 in 14 and IkB-a degradation was decreased after LPS treatment (Fig. 6B). regulating TLR4 signaling described in the current study, the Tyr Thus, Cav-1 phosphorylation at Tyr14 causes dissociation of IRAK-1 Cav-1 phospho-switch is crucial in regulating caveolae assembly 14 from Cav-1 in the plasma membrane, resulting in enhancement of in endothelial cells through Tyr -dependent homodimerization of TLR4 signaling. Cav-1 (34), caveolae-mediated endocytosis and transcytosis (35), and Tyr14 Cav-1–dependent eNOS association with Cav-1 (36). Thus, Cav-1 phosphorylation at Tyr14 may be a fundamental Expression of Cav-1 phosphorylation-defective Y14F mutant mechanism for regulating endothelial homeostasis. We observed in vivo protects mice following LPS challenge that LPS-induced Cav-1 phosphorylation at Tyr14 required Src 2 2 We transfected cav-1 / mice with Myc-tagged WT Cav-1 cDNA activation, as well as expression of the LPS receptor CD14 in or Myc-tagged Y14F–Cav-1 cDNA in pcDNA6 plasmid vector. endothelial cells. The mechanism responsible for Src activation is 2 2 Cav-1 / mice transfected with Myc-tagged WT Cav-1 or Y14–Cav-1 not clear; however, because reactive oxygen species (ROS) are cDNA showed equivalent protein expression of exogenous Cav-1 known to be generated downstream of LPS binding to CD14 (37) (Fig. 7A). Immunostaining of cav-12/2 lung sections transfected with WT or Y14F–Cav-1 cDNA with anti–VE cadherin and Cav-1 Abs showed that exogenous Cav-1 was expressed in pulmonary vascular endothelial cells (Supplemental Fig. 2). We did not detect bands corresponding to phosphorylated Cav-1 in Cav-1–null mice or cav-12/2 mice transfected with Myc-tagged Y14F–Cav-1 cDNA. We observed a similar pattern of degradation of IkB-a (lowest protein amount at 1 h after LPS challenge) in cav-1+/+ mice (Fig. 7B, Supplemental Fig. 3A) and cav-12/2 mice transfected with WT Cav-1 cDNA (Fig. 7B). However, no obvious IkB-a degradation was detected in Cav-1–null lungs (Supplemental Fig. 3A) or cav-12/2 lungs transfected with Y14F–Cav-1 cDNA (Fig. 7B). To determine whether phosphorylation-defective mutant of Cav-1 (Y14F) has a possible protective role against LPS challenge, we performed mortality studies with LPS using a dose shown to be lethal in mice (31). The survival of cav-12/2 mice transfected with WT Cav-1 cDNA was 35% after LPS administration (Fig. 7C). By contrast, the survival of cav-12/2 mice transfected with Y14F–Cav-1 cDNA was 75% after receiving the same LPS dose. We also measured lung wet/dry weight ratio to quantify pulmonary edema. No significant differences were observed in wet/dry ratios of cav-12/2 lungs transfected with WT and Y14F–Cav-1 cDNAs under basal conditions (Fig. 7D). However, at 6 h after LPS challenge, WT Cav-1 lungs had a significant increase in wet/dry ratio, whereas the ratio in Y14F–Cav-1 lungs did not increase significantly. The release of TNF-a and IL-6 was undetectable in both groups of mice without LPS treatment, whereas LPS challenge markedly increased TNF- a and IL-6 production in WT Cav-1 mice; however, in Y14F–Cav- 1 mice, the production of TNF-a (Fig. 7E) and IL-6 (Fig. 7F) was 2/2 greatly reduced. Consistently, after LPS challenge, cav-1 lungs FIGURE 8. Model of Cav-1 phosphorylation-dependent LPS-signaling had significant increases in wet/dry ratio (Supplemental Fig. 3B) pathway in mediating lung inflammation in endothelial cells. 6198 CAVEOLIN-1 PHOSPHORYLATION ENHANCES TLR4 SIGNALING and ROS signaling activates Src (30), the binding of LPS to the Therefore, our data showed Cav-1 phosphorylation in endothelial LPS receptor CD14 on endothelial cell surface and generation of cells to be a critical event for enhancing the innate immune re- ROS may be crucial in activating Src. Another related mechanism sponse, which determines survival during septic shock. Although of Src activation may involve the association that is known to the approach for cationic liposome-based gene delivery was occur between Src and CD14 within plasmalemmal lipid rafts shown to be selective in inducing transient gene transfer in lungs (38), which may facilitate Src activation in proximity to Cav-1. (41), we cannot exclude the potential role of Cav-1 in macro- Our results support the hypothesis that Cav-1 phosphorylation at phage, liver, or other vital organs (heart, kidney) in the regulation Tyr14 in response to LPS stimulation is essential for the activation of systemic inflammation and survival after LPS-induced sepsis. of TLR4–MyD88 signaling. We demonstrated that TLR4–MyD88 In summary, LPS engagement of TLR4 causes CD14-dependent signaling induced NF-kB activation in HEK-TLR4 cells trans- Src activation, which phosphorylates Cav-1 at Tyr14. The phos- fected with WT Cav-1 cDNA following LPS stimulation. This phorylation of Cav-1 stabilized the LPS–TLR4–MyD88 complex response was blocked in cells transfected with expression of and activated the downstream-signaling pathway, resulting in IkB-a Y14F–Cav-1 cDNA. Moreover, exogenous expression of WT degradation and subsequent NF-kBactivationandproinflammatory Cav-1 in cav-12/2 mouse lung endothelial cells induced IkB-a cytokine release (Fig. 8). Cav-1 phosphorylation also caused dis- degradation and production of cytokines IL-6 and TNF-a after sociation of IRAK-1 from Cav-1 in plasma membranes, further LPS stimulation. Importantly, transfection of endothelial cells and contributing to activation of TLR4 signaling. Mutation of tyrosine lung vascular endothelia in vivo with Y14F–Cav-1 cDNA blocked residue Y14 in mice transfected with Cav-1–Y14F cDNA impaired these responses. Compared with mice expressing WT Cav-1, the ability of Cav-1 to modulate complex interactions, inhibiting Y14F–Cav-1 mice demonstrated reduced inflammation upon LPS NF-kB activation in response to LPS and reducing immune re- challenge, as well as reduced mortality, cytokine release, and lung sponse as manifested by reduced cytokine production, less lung edema formation. Thus, Cav-1 phosphorylation at Tyr14 is re- edema formation, and higher survival of mice after LPS challenge. quired for activation of LPS–TLR4–MyD88 signaling and medi- Therefore, our work provides evidence that specific inhibition of ation of the endothelial inflammatory response to LPS. Tyr14 Cav-1 phosphorylation in endothelial cells may be an effec- We also investigated the mechanism by which Cav-1 phosphor- tive therapeutic strategy for treating sepsis-related lung inflamma- ylation regulates TLR4 signaling. We observed that binding of Tyr14 tory injury. Cav-1 to TLR4 was required to activate TLR4 signaling. Studies 739 747 showed that the motif FIQSRWCIF in the C-terminal intra- Acknowledgments cellular domain of TLR4 in mouse peritoneal macrophages can bind We thank Maricela Castellon (Departments of Anesthesiology and Pharma- to Cav-1 (24). We also observed an association of Tyr14 Cav-1 and cology, University of Illinois College of Medicine) for technical assistance. TLR4 in both endothelial cells and HEK-TLR4 cells. Because we showed that Tyr14 phosphorylation is required for the interaction with TLR4 and activation of TLR4–MyD88 signaling, the inter- Disclosures action of Tyr14 Cav-1 and TLR4 may be directly responsible for the The authors have no financial conflict of interest. activation of TLR4 signaling. Phosphorylation of Cav-1 at Tyr14 was also demonstrated to confer binding to H2-containing scaffolds References and to phosphotyrosine-binding domain-containing adaptor pro- 1. Martin, G. S., D. M. Mannino, S. Eaton, and M. Moss. 2003. The epidemiology teins, such as GRB7 (39); thus, an altered conformation induced by of sepsis in the United States from 1979 through 2000. N. Engl. J. Med. 348: Tyr14 of Cav-1 may favor a similar binding and activation of TLR4. 1546–1554. 2. Bannerman, D. D., and S. E. Goldblum. 2003. Mechanisms of bacterial IRAK4/IRAK1 are recruited to the TLR4 receptor complex lipopolysaccharide-induced endothelial apoptosis. Am. J. Physiol. Lung Cell. following LPS stimulation. IRAK1 is then phosphorylated by Mol. Physiol. 284: L899–L914. IRAK4, dissociated from the receptor complex, and subsequently 3. Bernard, G. R., A. Artigas, K. L. Brigham, J. Carlet, K. Falke, L. Hudson, b M. Lamy, J. R. Legall, A. Morris, and R. Spragg. 1994. The American-European interacts with TRAF6 and a TGF- –activated kinase 1–TAB1– Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, TAB2 kinase complex, leading to the activation of NF-kB (13– and clinical trial coordination. Am. J. Respir. Crit. Care Med. 149: 818–824. 15). In this study, Cav-1 phosphorylation in response to LPS 4. Dauphinee, S. M., and A. Karsan. 2006. Lipopolysaccharide signaling in endothelial cells. Lab. Invest. 86: 9–22. stimulation caused the dissociation of IRAK-1 from the plasma 5. Lien, E., T. K. Means, H. Heine, A. Yoshimura, S. Kusumoto, K. Fukase, membranes and enhanced IkB degradation, consistent with find- M. J. Fenton, M. Oikawa, N. Qureshi, B. Monks, et al. 2000. Toll-like receptor 14 4 imparts ligand-specific recognition of bacterial lipopolysaccharide. J. Clin. ings in monocytes (32). Thus, Tyr Cav-1 phosphorylation also Invest. 105: 497–504. activated LPS–TLR4 signaling through increased IRAK-1 ex- 6. Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. Van Huffel, X. Du, D. Birdwell, pression in the cytosol. E. Alejos, M. Silva, C. Galanos, et al. 1998. Defective LPS signaling in C3H/HeJ 2/2 and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282: 2085–2088. Increased eNOS activity observed in cav-1 mice accounted 7. Kerfoot, S. M., and P. Kubes. 2005. Local coordination verses systemic dis- for the inhibition of LPS-induced NF-kB activation (26, 27). regulation: complexities in leukocyte recruitment revealed by local and systemic eNOS-derived NO generation can impair the activity of IRAK4, activation of TLR4 in vivo. J. Leukoc. Biol. 77: 862–867. 8. Uhlig, S., F. Brasch, L. Wollin, H. Fehrenbach, J. Richter, and A. Wendel. 1995. a TLR4 downstream-signaling element, through tyrosine nitration Functional and fine structural changes in isolated rat lungs challenged with and, thereby, induce decreased NF-kB activation and inflamma- endotoxin ex vivo and in vitro. Am. J. Pathol. 146: 1235–1247. 9. Akira, S., and K. Takeda. 2004. Toll-like receptor signalling. Nat. Rev. Immunol. tory lung injury in response to LPS challenge (27). In a recent 4: 499–511. study, we demonstrated that Src-dependent phosphorylation of 10. O’Neill, L. A., and A. G. Bowie. 2007. The family of five: TIR-domain-containing Cav-1 at Tyr14 in endothelial cells promoted the binding of eNOS adaptors in Toll-like receptor signalling. Nat. Rev. Immunol. 7: 353–364. 11. Hoebe, K., X. Du, P. Georgel, E. Janssen, K. Tabeta, S. O. Kim, J. Goode, P. Lin, and Cav-1 and reduced eNOS activity (40). Thus, it is unlikely N. Mann, S. Mudd, et al. 2003. Identification of Lps2 as a key transducer of that Tyr14 Cav-1 phosphorylation mediated LPS–TLR4 signaling MyD88-independent TIR signalling. Nature 424: 743–748. though activation of eNOS activity and release of NO in lung 12. Kawai, T., O. Takeuchi, T. Fujita, J. Inoue, P. F. Mu¨hlradt, S. Sato, K. Hoshino, and S. Akira. 2001. Lipopolysaccharide stimulates the MyD88-independent pathway vascular endothelial cells. and results in activation of IFN-regulatory factor 3 and the expression of a subset of In the current study, exogenous expression of Y14F mutant lipopolysaccharide-inducible genes. J. Immunol. 167: 5887–5894. 2/2 13. Covert, M. W., T. H. Leung, J. E. Gaston, and D. Baltimore. 2005. Achieving nonphosphorylatable Cav-1 in cav-1 mouse pulmonary endo- stability of lipopolysaccharide-induced NF-kappaB activation. Science 309: thelium caused less sensitivity to septic shock induced by LPS. 1854–1857. The Journal of Immunology 6199

14. Cao, Z., J. Xiong, M. Takeuchi, T. Kurama, and D. V. Goeddel. 1996. TRAF6 is cells induces caveolae-mediated pulmonary vascular hyperpermeability. Circ. a signal transducer for interleukin-1. Nature 383: 443–446. Res. 102: e120–e131. 15. Medzhitov, R., P. Preston-Hurlburt, E. Kopp, A. Stadlen, C. Chen, S. Ghosh, and 30. Sun, Y., G. Hu, X. Zhang, and R. D. Minshall. 2009. Phosphorylation of C. A. Janeway, Jr. 1998. MyD88 is an adaptor protein in the hToll/IL-1 receptor caveolin-1 regulates oxidant-induced pulmonary vascular permeability via par- family signaling pathways. Mol. Cell 2: 253–258. acellular and transcellular pathways. Circ. Res. 105: 676–685. 16. Muzio, M., J. Ni, P. Feng, and V. M. Dixit. 1997. IRAK (Pelle) family member 31. Wang, Y. L., A. B. Malik, Y. Sun, S. Hu, A. B. Reynolds, R. D. Minshall, and IRAK-2 and MyD88 as proximal mediators of IL-1 signaling. Science 278: G. Hu. 2011. Innate immune function of the adherens junction protein p120- 1612–1615. catenin in endothelial response to endotoxin. J. Immunol. 186: 3180–3187. 17. Rothberg, K. G., J. E. Heuser, W. C. Donzell, Y. S. Ying, J. R. Glenney, and 32. Ohnuma, K., T. Yamochi, M. Uchiyama, K. Nishibashi, S. Iwata, O. Hosono, R. G. Anderson. 1992. Caveolin, a protein component of caveolae membrane H. Kawasaki, H. Tanaka, N. H. Dang, and C. Morimoto. 2005. CD26 mediates coats. Cell 68: 673–682. dissociation of Tollip and IRAK-1 from caveolin-1 and induces upregulation of 18. Liu, P., M. Rudick, and R. G. Anderson. 2002. Multiple functions of caveolin-1. CD86 on antigen-presenting cells. Mol. Cell. Biol. 25: 7743–7757. J. Biol. Chem. 277: 41295–41298. 33. Frank, P. G. 2010. Endothelial caveolae and caveolin-1 as key regulators of 19. Okamoto, T., A. Schlegel, P. E. Scherer, and M. P. Lisanti. 1998. Caveolins, atherosclerosis. Am. J. Pathol. 177: 544–546. a family of scaffolding proteins for organizing “preassembled signaling com- 34. Orlichenko, L., B. Huang, E. Krueger, and M. A. McNiven. 2006. Epithelial plexes” at the plasma membrane. J. Biol. Chem. 273: 5419–5422. growth factor-induced phosphorylation of caveolin 1 at tyrosine 14 stimulates 20. Smart, E. J., G. A. Graf, M. A. McNiven, W. C. Sessa, J. A. Engelman, caveolae formation in epithelial cells. J. Biol. Chem. 281: 4570–4579. P. E. Scherer, T. Okamoto, and M. P. Lisanti. 1999. Caveolins, liquid-ordered 35. del Pozo, M. A., N. Balasubramanian, N. B. Alderson, W. B. Kiosses, domains, and signal transduction. Mol. Cell. Biol. 19: 7289–7304. A. Grande-Garcıá, R. G. Anderson, and M. A. Schwartz. 2005. Phospho- 21. Bucci, M., J. P. Gratton, R. D. Rudic, L. Acevedo, F. Roviezzo, G. Cirino, and caveolin-1 mediates integrin-regulated membrane domain internalization. Nat. W. C. Sessa. 2000. In vivo delivery of the caveolin-1 scaffolding domain inhibits Cell Biol. 7: 901–908. nitric oxide synthesis and reduces inflammation. Nat. Med. 6: 1362–1367. 36. Parat, M. O., B. Anand-Apte, and P. L. Fox. 2003. Differential caveolin-1 po- 22. Santizo, R. A., H. L. Xu, E. Galea, S. Muyskens, V. L. Baughman, and larization in endothelial cells during migration in two and three dimensions. Mol. D. A. Pelligrino. 2002. Combined endothelial nitric oxide synthase upregulation Biol. Cell 14: 3156–3168. and caveolin-1 downregulation decrease leukocyte adhesion in pial venules of 37. Park, H. S., H. Y. Jung, E. Y. Park, J. Kim, W. J. Lee, and Y. S. Bae. 2004. ovariectomized female rats. Stroke 33: 613–616. Cutting edge: direct interaction of TLR4 with NAD(P)H oxidase 4 isozyme is 23. Feng, H., L. Guo, Z. Song, H. Gao, D. Wang, W. Fu, J. Han, Z. Li, B. Huang, and essential for lipopolysaccharide-induced production of reactive oxygen species X. A. Li. 2010. Caveolin-1 protects against sepsis by modulating inflammatory and activation of NF-kappa B. J. Immunol. 173: 3589–3593. response, alleviating bacterial burden, and suppressing thymocyte apoptosis. J. 38. Solomon, K. R., E. A. Kurt-Jones, R. A. Saladino, A. M. Stack, I. F. Dunn, Biol. Chem. 285: 25154–25160. M. Ferretti, D. Golenbock, G. R. Fleisher, and R. W. Finberg. 1998. Hetero- 24. Wang, X. M., H. P. Kim, K. Nakahira, S. W. Ryter, and A. M. Choi. 2009. The trimeric G proteins physically associated with the lipopolysaccharide receptor heme oxygenase-1/carbon monoxide pathway suppresses TLR4 signaling by CD14 modulate both in vivo and in vitro responses to lipopolysaccharide. J. Clin. regulating the interaction of TLR4 with caveolin-1. J. Immunol. 182: 3809–3818. Invest. 102: 2019–2027. 25. Wang, X. M., H. P. Kim, R. Song, and A. M. Choi. 2006. Caveolin-1 confers 39. Lee, H., D. Volonte, F. Galbiati, P. Iyengar, D. M. Lublin, D. B. Bregman, antiinflammatory effects in murine macrophages via the MKK3/p38 MAPK M. T. Wilson, R. Campos-Gonzalez, B. Bouzahzah, R. G. Pestell, et al. 2000. pathway. Am. J. Respir. Cell Mol. Biol. 34: 434–442. Constitutive and growth factor-regulated phosphorylation of caveolin-1 occurs at 26. Garrean, S., X. P. Gao, V. Brovkovych, J. Shimizu, Y. Y. Zhao, S. M. Vogel, and the same site (Tyr-14) in vivo: identification of a c-Src/Cav-1/Grb7 signaling A. B. Malik. 2006. Caveolin-1 regulates NF-kappaB activation and lung inflammatory cassette. Mol. Endocrinol. 14: 1750–1775. response to sepsis induced by lipopolysaccharide. J. Immunol. 177: 4853–4860. 40. Chen, Z., F. R. Bakhshi, A. N. Shajahan, T. Sharma, M. Mao, A. Trane, 27. Mirza, M. K., J. Yuan, X. P. Gao, S. Garrean, V. Brovkovych, A. B. Malik, P. Bernatchez, G. P. van Nieuw Amerongen, M. G. Bonini, R. A. Skidgel, et al. C. Tiruppathi, and Y. Y. Zhao. 2010. Caveolin-1 deficiency dampens Toll-like 2012. Nitric oxide-dependent Src activation and resultant caveolin-1 phosphor- receptor 4 signaling through eNOS activation. Am. J. Pathol. 176: 2344–2351. ylation promote eNOS/caveolin-1 binding and eNOS inhibition. Mol. Biol. Cell 28. Hu, G., R. D. Ye, M. C. Dinauer, A. B. Malik, and R. D. Minshall. 2008. 23: 1388–1398. Neutrophil caveolin-1 expression contributes to mechanism of lung inflamma- 41. Zhou, M. Y., S. K. Lo, M. Bergenfeldt, C. Tiruppathi, A. Jaffe, N. Xu, and tion and injury. Am. J. Physiol. Lung Cell. Mol. Physiol. 294: L178–L186. A. B. Malik. 1998. In vivo expression of neutrophil inhibitory factor via gene 29. Hu, G., S. M. Vogel, D. E. Schwartz, A. B. Malik, and R. D. Minshall. 2008. transfer prevents lipopolysaccharide-induced lung neutrophil infiltration and Intercellular adhesion molecule-1-dependent neutrophil adhesion to endothelial injury by a b2 integrin-dependent mechanism. J. Clin. Invest. 101: 2427–2437.