HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 47, 45662-45669, November 22, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/47/45662    most recent M205935200v1 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 Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Shah, M. Articles by Sehgal, P. B. Search for Related Content PubMed PubMed Citation Articles by Shah, M. Articles by Sehgal, P. B. Social Bookmarking                      What's this?

Interactions of STAT3 with Caveolin-1 and Heat Shock Protein 90 in Plasma Membrane Raft and Cytosolic Complexes

PRESERVATION OF CYTOKINE SIGNALING DURING FEVER^*

Mehul Shah, Kirit Patel, Victor A. Fried, and Pravin B. Sehgal^§¶

From the  Department of Cell Biology and Anatomy and ^§ Department of Medicine, New York Medical College, Valhalla, New York 10595

Received for publication, June 14, 2002, and in revised form, August 7, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Interleukin-6 (IL-6) initiates STAT3 signaling in plasma membrane rafts with the subsequent transit of Tyr-phosphorylated STAT3 (PY-STAT3) through the cytoplasmic compartment to the nucleus in association with accessory proteins. We initially identified caveolin-1 (cav-1) as a candidate STAT3-associated accessory protein due to its co-localization with STAT3 and PY-STAT3 in flotation raft fractions, and heat shock protein 90 (HSP90) due to its inclusion in cytosolic STAT3-containing 200-400-kDa complexes. Subsequent immunomagnetic bead pullout assays showed that STAT3, PY-STAT3, cav-1, and HSP90 interacted in plasma membrane and cytoplasmic complexes derived from uninduced and stimulated Hep3B cells. This was a general property of STAT3 in that these interactions were also observed in alveolar epithelial type II-like cells, lung fibroblasts, and pulmonary arterial endothelial cells. Exposure of Hep3B cells to the raft disrupter methyl--cyclodextrin for 1-10 min followed by IL-6 stimulation for 15 min preferentially inhibited the appearance of PY-STAT3 in the cav-1-enriched sedimentable cytoplasmic fraction, suggesting that these complexes may represent a trafficking intermediate immediately downstream from the raft. Because IL-6 is known to function in the body in the context of fever, the possibility that HSP90 may help preserve IL-6-induced STAT3 signaling at elevated temperature was investigated. Geldanamycin, an HSP90 inhibitor, markedly inhibited IL-6-stimulated STAT3 signaling in Hep3B hepatocytes cultured overnight at 39.5 °C as evaluated by DNA-shift assays, trafficking of PY-STAT3 to the nucleus, cross-precipitation of HSP90 by anti-STAT3 polyclonal antibody, and reporter/luciferase construct experiments. Taken together, the data show that IL-6/raft/STAT3 signaling is a chaperoned pathway that involves cav-1 and HSP90 as accessory proteins and suggest a mechanism for the preservation of this signaling during fever.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Fever ("calor") is a common response of the body to infection and injury (reviewed in Refs. 1 and 2). Interleukin-6 (IL-6)¹ is a major systemic mediator of this "acute-phase" response that includes stimulation of the liver to synthesize and secrete a large number of protective plasma proteins such as anti-proteinases, clotting factors, complement factors, and scavenger proteins that help to limit the site of injury or infection (1, 2). Furthermore, IL-6 is itself a centrally acting pyrogen (3-5). How is the major signaling pathway used by IL-6 to enhance acute-phase plasma protein gene expression in hepatocytes, the JAK-STAT3 pathway (Janus kinase-signal transducer and activator of transcription 3) (6-11), maintained during an increase in body temperature? We report the identification of caveolin-1 (cav-1) and heat shock protein 90 (HSP90) as STAT3-associated proteins at the level of plasma membrane rafts and in high molecular mass cytoplasmic complexes, and we explore their role in cytokine signaling at normal and elevated temperatures.

In previous cell fractionation studies of human hepatoma Hep3B cells, approximately 10% of the cellular STAT3 and STAT1 pools were observed to be associated with a detergent-resistant plasma membrane raft fraction which included the plasma membrane marker 5'-nucleotidase (5'-ND), the integral raft proteins caveolin-1 (cav-1) and flotillin-1, the IL-6-receptor chain gp130, the interferon (IFN)--receptor chain , and the chaperone glucose-regulated protein 58 (GRP58) (12, 13). Upon activation by cytokine or orthovanadate Tyr-phosphorylated (PY) STAT3 and STAT1 were also found in association with the detergent-resistant membrane raft fraction (12, 13). The departure of PY-STATs from the membrane fraction into the cytoplasm was accompanied by acquisition of DNA binding competence (11, 14). The raft disrupter methyl--cyclodextrin (MCD) (15, 16) markedly inhibited IL-6 and IFN- signaling (12).

When characterized in the cytoplasm by gel filtration chromatography, STAT3, a protein of monomer size 91 kDa, was largely in the form of soluble 200-400 kDa and sedimentable 1-2 MDa "statosome" complexes (11, 14).² In the cytoplasm, IL-6-activated PY-STAT3 was also found in a broad distribution of sizes from 200-400 kDa to 1-2 MDa (14). Moreover, GRP58 was also identified as a STAT3-associated accessory protein in the cytosolic 200-400-kDa complexes (13, 14).

In this report we provide evidence showing that IL-6/raft/STAT3 signaling is a chaperoned pathway that involves cav-1 and HSP90 as accessory proteins. Furthermore, the association between HSP90 and STAT3 appears to be part of a biological mechanism for the preservation of this signaling pathway in liver cells at elevated temperature.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell Lines-- The parental human hepatoma Hep3B cell line and a derivative cell line ("Line 1") have been described earlier (14, 17-20). Line 1 Hep3B cells were used in the experiments reported here. Human lung alveolar epithelial type II-like cell line A549 (ATII-like) and human lung fibroblast strain CCD Lu-11 (LF) were a gift from Dr. Anuradha Ray, University of Pittsburgh School of Medicine, Pittsburgh. ATII-like and LF cells were grown in 100-mm plastic Petri dishes in a manner similar to growth of Hep3B cells. Cultures of primary bovine pulmonary arterial endothelial cells (BPAEC) in T-75 flasks were a gift from Dr. Susan Olson, New York Medical College, and were used directly.

Cytokine Treatment-- Hep3B cells were grown to confluence in 100-mm Petri dishes at 37 °C as described earlier (14, 17-20). For IL-6 or IFN- treatment, cultures were washed twice with phosphate-buffered saline, replenished with 5 ml of serum-free medium for 1-2 h, and then exposed to the recombinant cytokine (10 ng/ml; R & D Systems Inc., Minneapolis, MN) for the indicated times. In some experiments confluent Hep3B cultures were incubated at 39.5 °C overnight (12-14 h), and cytokine stimulation was also carried out at 39.5 °C.

Cell Fractionation-- Cells were harvested by scraping into ice-cold phosphate-buffered saline (PBS), washed twice with ice-cold PBS, resuspended in ~0.8 ml of extract lysis buffer (ELB; 10 mM Hepes, pH 7.9, 10 mM NaCl, 3 mM MgCl₂, 1 mM dithiothreitol, 0.4 mM phenylmethylsulfonyl fluoride, and 0.1 mM sodium orthovanadate) per cell pellet derived from five 100-mm cultures, and fractionated into cytoplasmic and nuclear fractions by gentle cell breakage in a loose-fitting Dounce homogenizer as described earlier (12-14, 17-19). All cell fractionation and gradient centrifugation steps were carried out at 4 °C. Nuclei were removed from the cell homogenate by two rounds of low speed centrifugation at 1000 rpm for 3 min in an IEC Centra GP8R centrifuge. The post-nuclear cytoplasm was further fractionated by centrifugation at 15,000 × g in an Eppendorf centrifuge for 15 min to yield the "P15 membrane fraction" (P15, pellet fraction of post-nuclear cytoplasm centrifuged at 15,000 × g for 15 min) which contained mitochondria, endoplasmic reticulum, and plasma membrane (13). This P15 membrane pellet was washed once with 0.5 ml of ELB, resedimented, and then resuspended in 50 µl of ELB. The supernatant from the initial 15,000 × g 15-min sedimentation was further fractionated into pellet P100 (P100, pellet fraction of post-membrane cytoplasm centrifuged at 100,000 × g for 60 min) and cytosol S100 fractions (S100, supernatant of post-membrane cytoplasm centrifuged at 100,000 × g for 1 h) in a TL-100 Beckman ultracentrifuge. After removal of the S100 cytosol from the centrifuge tube, the walls were wiped carefully, and then the P100 pellet was resuspended in 50 µl of ELB. Nuclear extracts were prepared in high salt buffer containing 0.6 M NaCl as described before (18, 19).

Enzymatic assays for 5'-nucleotidase (5'-ND, a marker for plasma membrane) were carried out using kits purchased from Sigma. Additional Western blot markers included caveolin-1, glucose-regulated protein 78 (GRP78/BiP), and nuclear pore-associated protein 62 (Nup62) (Organelle Marker kit; Transduction Laboratories, Lexington, KY).

Western Blot Analyses of Proteins-- Western blotting was carried out using 7.5, 10, or 12.5% SDS-polyacrylamide gels under reducing denaturing conditions in accordance with procedures and protocols provided by Transduction Laboratories, Lexington, KY, and the ECL detection kit (Amersham Biosciences) (14, 17-19). For quantitation purposes, extensive previous calibration controls (14, 17-19) have shown that detection of STAT1 and STAT3 proteins and their Tyr-phosphorylated derivatives was linear with the amount of respective protein in this assay. Routinely, multiple exposures of each blot were obtained so as to ensure that each of the signals was within the linear range; and in order to reduce quantitation errors, sample volumes in different fractions were adjusted to provide signals of equivalent strength each within the linear range of the assay. Western blot signals were quantitated using a Hoefer Scientific GS-300 scanning densitometer.

DNA Gel-shift Assays-- STAT-specific DNA binding activity was assayed using the m67 mutant serum-inducible element (SIE) oligonucleotide derived from the c-fos promoter (purchased from Santa Cruz Biotechnology, Santa Cruz, CA) as described earlier (14, 18). Briefly, this oligonucleotide yields the typical pattern of SIE-oligonucleotide binding factor (SIF)-A, -B, and -C complexes in gel-shift assays corresponding to the STAT3 homodimer, STAT1/3 heterodimer, and STAT1 homodimer, respectively (6-14). The SIE oligonucleotide was ³²P-end-labeled at its 5' end using T4 polynucleotide kinase and used in electrophoretic shift assays as described earlier (14, 18). Typically the binding buffer (15-µl reaction volume; 30 min of incubation at room temperature) contained 40 mM KCl, 1 mM MgCl₂, 0.1 mM EDTA, 20 mM Hepes, pH 7.9, 4% (v/v) Ficoll, 7% Me₂SO, and 1 µg per reaction poly(dI·dC)·poly(dI·dC) (Amersham Biosciences). DNA-protein complexes were separated by electrophoresis through 5% PAGE (acrylamide/bisacrylamide ratio 30:0.8) in 0.2× TBE (0.02 M Tris, pH 8.3, 0.02 M boric acid, 0.25 mM EDTA) at 300 V at 4 °C for ~2 h and dried for autoradiography. Antibody inhibition assays were performed as described earlier (14, 18). Routinely, multiple exposures of each autoradiogram were obtained on Kodak XAR5 film so as to be within the linear range of exposure of the film. The signal intensities were quantitated using a Hoefer Scientific GS-300 scanning densitometer. The identification of the respective SIF-A, -B, and -C bands as STAT3 homodimer, STAT3/1 heterodimer, and STAT1 homodimer in our experiments was confirmed using respective antibody supershift assays (see Refs. 12-14 and additional data).

IL-6-responsive Reporter/Luciferase Construct Assays-- Transfections into Hep3B cell cultures in 6-well plates were carried out using the LipofectAMINE reagent (Polyfect, Qiagen, Valencia, CA) and the manufacturer's protocol. Briefly, a reporter/luciferase construct (p950M4) containing four copies of the STAT3-binding DNA element from the human angiotensinogen promoter (5' CGTTTCTGGGAACCT 3') cloned into pBL₂Luc (Promega Biotech, Madison, WI) (a gift from Dr. Ashok Kumar, Department of Pathology, New York Medical College) was transfected into Hep3B cells (250-500 ng/well) together with the constitutive -galactosidase expression plasmid pCH110 (50 ng/well) and appropriate amounts of pBR322 or Bluescript vector to give 1 µg of transfected DNA per well. After incubation at 37 °C for 10-12 h, the cultures were either kept at 37 °C or incubated overnight at 39.5 °C. IL-6 (10 ng/ml) induction was carried out for 6 h at the respective temperatures. The cells were harvested in 300 µl of lysis buffer (Promega Biotech), and 50-µl aliquots of the clarified extracts were used to assay luciferase and -galactosidase activity using respective assay kits/reagents from Promega Biotech and Roche Diagnostics and the manufacturer's protocols.

Antibody Reagents-- Murine monoclonal antibodies (mAbs) to STAT1, STAT3, and HSP90 were purchased from Transduction Laboratories (Lexington, KY), and rabbit polyclonal antibodies (pAbs) to STAT3, STAT5b, and caveolin-1, as well as a murine mAb to PY-STAT3 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-PY-STAT1 and anti-PY-STAT3 pAbs were purchased from New England Biolabs (Beverly, MA). Rabbit antiserum to recombinant human glucose-regulated protein 58 (GRP58/ER-60/ERp57) was a gift from Drs. Mohammed Bourdi and Lance Pohl (National Institutes of Health, Bethesda).

Immunopanning Using Protein A Magnetic Beads-- Protein A magnetic beads were purchased from New England Biolabs (Beverly, MA). Prior to use in immunopanning experiments, the beads were blocked with 5% non-fat dry milk in PBS for 1 h and then washed three times with PBS. Aliquots of the P15 membrane fraction (150 µl), the P100 pellet (150 µl), or the S100 cytosol (300 µl) were adjusted to 0.05% Triton X-100 in 25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, and 1 mM dithiothreitol ("binding buffer") and incubated overnight at 4 °C with respective rabbit immunoglobulin (nonimmune rabbit serum (NRS) or anti-caveolin-1 pAb or anti-STAT3 pAb) and then for 1 h at 4 °C with the pre-blocked protein A magnetic beads. The magnetic beads were washed 5 times with binding buffer containing 0.05% Triton X-100 and then twice with binding buffer adjusted to 0.5% Triton X-100. Immunopanned protein complexes were analyzed by SDS-PAGE and Western blotting under reducing denaturing conditions (14, 19).

Electron Microscopy-- All electron microscopy reagents were purchased from Electron Microscopy Sciences (Fort Washington, PA) and used in accordance with the manufacturer's protocols. Briefly, pellets of respective cellular fractions sedimented to the bottom of Eppendorf tubes were fixed in 2% paraformaldehyde, 2.5% glutaraldehyde in PBS (Karnovsky's fixative) and then post-fixed with 1% osmium tetroxide in PBS. The fixed pellets were embedded in LR White Hard resin, thin sectioned using a Sorvall Porter-Blum MT2-8 Ultra-microtome, and visualized using an Hitachi H-7000 electron microscope.

Protein Identification by Mass Spectroscopy-- Coomassie Blue-stained protein bands were excised from SDS-PAGE gels using a scalpel and digested in situ with trypsin (Promega, Madison, WI) as described by Rosenfeld et al. (21) except that the detergent Tween 20 was absent from the digestion buffers. The digest containing the gel bits was extracted in 50% acetonitrile, 0.1% trifluoroacetic acid, concentrated using a SpeedVac to less than 10 µl, and desalted using a Zip-Tip (Millipore, Bedford, MA). An aliquot of the digest was mixed with an equal volume of matrix material (-cyano-4-hydroxycinnamic acid (Sigma) as a saturated solution in 50% acetonitrile, 0.05% trifluoroacetic acid), and ion masses were determined by matrix-assisted laser desorption ionization-time of flight mass spectroscopy (Kratos KOMPACT MALDI III). Data bases were searched at 0.08% mass accuracy with the MS-Fit algorithm (University of California at San Francisco web site) to identify digested protein.

Additional Reagents-- MCD and filipin III were purchased from Sigma. Geldanamycin (GA) was a gift from the Drug Synthesis & Chemistry Branch, Developmental Therapeutics Program, Division of Cancer Treatment and Diagnosis, NCI, National Institutes of Health.

Statistical Analyses-- These were carried out using the SPSS 10.0 software package for Windows (SPSS Inc., Chicago, IL) and the Student's t test (for paired samples, two-tailed).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Identification of Candidate STAT3-associated Accessory Proteins-- STAT proteins associate with different subcellular fractions in Hep3B cells (11-14). In the experiment illustrated in Fig. 1, Hep3B cells were fractionated using hypotonic swelling, Dounce homogenization, and differential centrifugation methods into the plasma membrane-enriched P15 fraction, the post-membrane cytoplasmic P100 pellet and S100 cytosol fractions, and the nuclear extract (NE). Fig. 1 shows that the P15 fraction was highly enriched in the plasma membrane marker 5'-ND and contained the endoplasmic reticulum (ER) marker BiP, whereas the P100 cytoplasmic pellet was depleted in both 5'-ND as well as BiP. The nuclear marker Nup62 was detected only in the nuclear extract. These marker distributions were consistent with electron microscopic characterization of the P15 and P100 fractions, which confirmed that the former fraction contained cytoplasmic membranes, mitochondria, and ER, and, most important, the P100 pellet appeared to be largely membrane-free (data not shown). Fig. 1 further shows that, as expected, the bulk of STAT1, STAT3, and STAT5b in Hep3B cells was present in the S100 cytosol (see quantitation in legend to Fig. 1). However ~10% of each of the cellular STAT1 and STAT3 pools were observed in the P15 and P100 pellet fractions (see quantitation in legend to Fig. 1). Upon IL-6 treatment for 30 min, ~15% of cellular STAT3, 5% of STAT1, 37% of PY-STAT3, and 56% of PY-STAT1 translocated to the nucleus (Fig. 1). Cytoplasmic PY-STAT3 was associated with both the P100 pellet fraction and the S100 cytosol (also see Fig. 6, below).

View larger version (53K): [in this window] [in a new window]   Fig. 1.   Characterization of Hep3B cell fractions. Control and IL-6-treated (30 min) Hep3B cells were fractionated as described under "Materials and Methods" into a P15 membrane fraction, the post-membrane cytoplasmic P100 pellet, and S100 cytosol and the NE fraction. Aliquots (5 µl each) from each fraction were used to measure total protein by the micro-Bradford method and 5'-ND activity. Enzyme activity data are represented as the fraction in each compartment as a percentage of the sum of that in the P15, P100, S100, and NE compartments. Aliquots corresponding to 5% of the P15, 10% each of the P100 and NE compartments, and 2.5% of the S100 compartment of the IL-6-treated cells and a volume of the corresponding fractions from untreated cells containing matching amounts of total protein (27 µg in P15, 42 µg in P100, 44 µg in S100, and 20 µg in NE) were Western-blotted, probed using different antibodies, and quantitated by densitometry. The compartmental distribution expressed as % of total in the cell for STAT3 in uninduced cells was 11, 7, 82, and <1% in the P15, P100, S100, and NE fractions, respectively, whereas it was 10, 12, 63, and 15% in IL-6-treated cells. The compartmental distribution for STAT1 in uninduced cells was 11, 9, 79, and 1% in the P15, P100, S100, and NE fractions, respectively, whereas it was 8, 12, 75, and 5% in IL-6-treated cells. Additionally, thin section electron microscopy was carried out on the P15 and P100 pellets. The P15 fraction contained cellular membranes and cytoplasmic organelles, whereas the P100 fraction appeared to be membrane-free (data not shown).

The co-purification of STAT proteins in detergent-resistant plasma membrane flotation raft fractions, which also contained cav-1 (12), raised the possibility that cav-1 may be a candidate STAT3-interacting protein. Among additional candidate STAT3-interacting proteins, we previously identified GRP58 in S100 cytosolic 200-400-kDa complexes using an antibody-subtracted differential protein display approach (13, 14). We now report that polypeptide A4 enumerated in Table I in Ref. 14 was HSP90 as identified using mass spectroscopic methods. Additional candidate STAT3-associated proteins in the 200-400-kDa complexes included proteins typically found on the cytosolic face of cellular membranes (adaptin subunits and the protein Ire1P).³ The association between STAT3 and adaptin subunits is consistent with a recent report (22) showing the co-localization of STAT3 with -adaptin-containing AP-2 complexes at the level of the plasma membrane and in cytoplasmic structures. In the present studies we have focused on an evaluation of the interactions between and among STAT3, cav-1, and HSP90.

Association of STAT3 with Cav-1 and HSP90 in Detergent-resistant Membrane and Cytoplasmic Complexes-- We investigated whether the candidate accessory proteins cav-1 and HSP90 were present together with STAT3 in the same detergent-resistant physical unit using cross-immunomagnetic bead pullout assays in the presence of Triton X-100 (final concentration, 0.5%) in plasma membrane and cytoplasmic complexes. Fig. 2A shows the distribution of cav-1 species in the different Hep3B subcellular fractions. It is known from previous reports (23, 24) that in SDS-PAGE the 21-kDa cav-1 protein located in the plasma membrane can appear as high molecular mass species because oligomers of palmitoylated cav-1 can resist complete dissociation. Fig. 2A shows that the P15 membrane fraction from Hep3B cells contained cav-1 not only as the 21-kDa species (double arrowhead) but also as a 63-kDa and a higher molecular mass species (single arrowheads). In comparison, analyses of the S100 cytosol showed more of the 21-kDa species together with a higher molecular mass species at ~100 kDa. Remarkably, the membrane-free P100 pellet was highly enriched in the 21-kDa cav-1 species. The sedimentable P100 cav-1-enriched cytoplasmic fraction from Hep3B cells appears similar to the sedimentable supramolecular cav-1-containing non-membranous proteolipid complexes (large of size 40 × 26 nm by negative stain electron microscopy, and small of size 26 × 20 nm) isolated by Palade and colleagues (25) from the cytosol of rat pulmonary endothelial cells.

View larger version (48K): [in this window] [in a new window]   Fig. 2.   Association between STAT3 and cav-1 in Hep3B hepatocytes. A, cav-1 in Hep3B cell fractions. Duplicate groups of Hep3B cultures were fractionated into P15, P100, S100, and NE fractions, and aliquots were assayed for cav-1 species by SDS-PAGE under reducing denaturing conditions (12.5% polyacrylamide) and Western blotting (20% of each of the P15 (10 µl out of 50 µl) and P100 (10 µl out of 50 µl) fractions, and 10% of each of the S100 (50 µl out of 500 µl) and NE (10 µl out of 100 µl) fractions). Double arrowhead represents cav-1 monomer, and single arrowheads point to modified/oligomeric cav-1. B, cross-immunopanning (IP) of STAT3 by anti-cav-1 pAb. Aliquots of the P15 (150 µl), P100 (150 µl), and S100 (300 µl) fractions derived from uninduced Hep3B cells were adjusted to 0.05% Triton X-100 and incubated overnight with non-immune rabbit serum (NRS), anti-cav-1 pAb, or anti-STAT3 pAb as indicated, and then immunopanned using protein-A magnetic beads followed by washing in Triton X-100-containing buffers (5 times in 0.05% and then twice in 0.5%). Immunopanned proteins were analyzed by SDS-PAGE and Western blotting using anti-STAT3 mAb.

Fig. 2B shows that anti-cav-1 pAb cross-panned STAT3 from detergent-treated P15, P100, and S100 fractions. Thus, cav-1 and STAT3 were associated with the same detergent-resistant physical unit not only in the membrane fraction but also in the cytoplasmic P100 and S100 fractions. The detergent-resistant cross-immunoprecipitation of STAT3 by anti-cav-1 pAb from membrane and cytoplasmic fractions was not restricted to Hep3B hepatocytes. We confirmed these observations in uninduced alveolar type II-like lung epithelial cells (A549 line) (ATII-like), in human lung fibroblasts (CCD Lu11 strain) (LF), and bovine pulmonary endothelial cells (BPAEC) (Fig. 3).

View larger version (51K): [in this window] [in a new window]   Fig. 3.   Association between STAT3 and cav-1 in lung-derived cell types. Respective P15, P100, S100, and NE cellular fractions derived from ATII-like epithelial, LF, and BPAEC cells were immunopanned (IP) using NRS or anti-cav-1 as indicated, and the blots were probed using anti-STAT3 mAb. Both blots were also re-probed using anti-HSP90 mAb with similar results; data from blot B are shown.

Does activated PY-STAT3 also associate with cav-1-containing membrane raft and cytoplasmic complexes? Fig. 4A shows that anti-cav-1 pAb cross-immunopanned PY-STAT3 from the isolated plasma membrane raft fraction derived from IL-6-treated Hep3B cells. Fig. 4B shows that PY-STAT3 in the P100 pellet and Fig. 4C shows that PY-STAT3 in the S100 fractions of vanadate- or IL-6-stimulated Hep3B cells were also cross-immunopanned by anti-cav-1 pAb. Thus, PY-STAT3 was associated with cav-1 in complexes in the membrane raft and in the cytoplasm.

View larger version (41K): [in this window] [in a new window]   Fig. 4.   Association between PY-STAT3 and cav-1 in plasma membrane rafts and in the cytoplasmic P100 and S100 fractions. A, the P15 membrane fraction derived from Hep3B cells treated with IL-6 for 30 min (IL-6) and uninduced controls ()(4 cultures per group) were resuspended in solubilizing buffer containing 0.05% Triton X-100, adjusted to 60% (w/v) sucrose, and subjected to equilibrium flotation through a discontinuous sucrose gradient (60:45:13% w/v) in an SW50.1 rotor for at 35,000 rpm for 16 h as described under "Materials and Methods." The flotation raft fraction at the 13/30% interface was obtained, washed once in ELB, and resuspended in buffer containing 0.05% Triton X-100. Raft fractions from uninduced and IL-6-induced cells were subjected to immunopanning (IP) using anti-cav-1 pAb or non-immune rabbit serum (NRS) and Western blotted using anti-PY-STAT3 mAb. B, aliquots (150 µl) of the P100 pellet fractions derived from control Hep3B cells () or those treated with orthovanadate (0.3 mM; a general activator of STATs) for 4 h (VO4) (5 cultures per group) were subjected to immunopanning using anti-cav-1 pAb or NRS and Western-blotted using anti-PY-STAT3 mAb. C, aliquots (300 µl) of the S100 cytosol fraction derived from control Hep3B cells () or cells treated with orthovanadate for 4 h (VO4), IL-6 for 30 min (IL-6), or IFN- for 30 min (IFN-) (4 cultures per group) were immunopanned using anti-cav-1 or NRS as indicated, and the Western blots were probed using anti-PY-STAT3 mAb. Similar results were obtained using anti-STAT3 pAb as the probe (data not shown). Additionally, PY-STAT1 was also detected in the immunoprecipitates derived from vanadate or IFN--treated cells (data not shown).

Fig. 5 illustrates a composite of experiments that confirm that HSP90 was also cross-precipitated by anti-STAT3 pAb from the P15 membrane and the cytoplasmic P100 and S100 fractions. (We did not distinguish between HSP90 and - in the present analyses because although the anti-HSP90 mAb used for developing the Western blots was raised against a peptide sequence of the  species, it does not have the requisite  versus  specificity).⁴ HSP90 was also included with STAT3 in complexes precipitated with anti-cav-1 pAb (Fig. 5). Moreover, anti-GRP58 pAb also cross-immunopanned HSP90 together with STAT3 from S100 cytosolic complexes confirming our previous observation that GRP58 was a STAT3-associated accessory protein (13, 14). STAT1 was also included in these cross-immunopanned complexes from the P15, P100, and S100 fractions (Fig. 5). Additionally, HSP90 was also included together with STAT3 in cross-immunoprecipitations by anti-cav-1 pAb of the P15 membrane fractions from human LF and primary BPAEC (Fig. 3B). Taken together, our magnetic bead immunopanning data indicate that STAT3, cav-1, and HSP90 interact, directly or indirectly, within detergent-resistant physical units in the P15 membrane and cytoplasmic P100 and S100 compartments.

View larger version (50K): [in this window] [in a new window]   Fig. 5.   Associations between and among STAT3, HSP90, cav-1, STAT1, and GRP58. Aliquots of the P15 (150 µl), P100 (150 µl), and S100 (300 µl) fractions from uninduced Hep3B cells were immunopanned (IP) using NRS, anti-cav-1 pAb, anti-STAT3 pAb, or anti-GRP58 as indicated, and the Western blots were probed sequentially using anti-STAT3 mAb, anti-HSP90 mAb, and anti-STAT1 mAb.

Trafficking of Activated STATs from Plasma Membrane Rafts to Cytoplasmic Complexes-- The compounds MCD and filipin III extract cholesterol from cav-1-containing cholesterol-rich plasma membrane rafts disrupting raft integrity and function (15, 16). We have shown previously that exposure of Hep3B cells to MCD for 15 min before IL-6 or IFN- stimulation markedly inhibited cytokine signaling to the cytoplasmic P100 and S100 fractions and the nuclear compartments as assayed using DNA-binding assays for activated STAT3 and STAT1. These data suggested that almost all of IL-6 and IFN- signaling was initiated at the level of cholesterol-rich plasma membrane rafts (12). We have investigated whether progressively shorter exposure to MCD might provide kinetic trafficking data concerning the movement of activated STATs from the P15 membrane raft fraction to the cytoplasmic P100/S100 fractions and then to the nucleus. Fig. 6A shows an experiment in which Hep3B cells were treated with MCD for 10, 5, or 1 min before IL-6, and the cells were harvested 15 min after the beginning of IL-6 treatment and STAT-specific DNA binding activity assayed in the P15, P100, S100, and NE fractions. Fig. 6B shows Western blot analyses for PY-STAT3 of the respective fractions shown in Fig. 6A. The data in Fig. 6A show that a brief treatment of cells with MCD interrupted the downstream flow of DNA-binding competent STAT3 from the P15 plasma membrane fraction. In the 1-min MCD-treated groups there was increased accumulation of STAT3 DNA binding activity in the P15 fraction accompanied by a marked inhibition in the downstream P100 and S100 fractions, with little decrease in the nuclear extract. Similar data were obtained using IFN- except that the inhibition of PY-STAT1 DNA binding activity in cells treated with MCD at 1 was selectively marked in the P100 fraction with minimal effects in the S100 and NE fractions (data not shown).

View larger version (88K): [in this window] [in a new window]   Fig. 6.   Effect of a brief exposure to MCD on IL-6-induced STAT3-specific DNA shift activity (A) and PY-STAT3 content (B) in different Hep3B subcellular fractions. A, effect of treating Hep3B cells with MCD before cytokine on DNA binding activity in different cellular compartments. Groups of Hep3B cultures (4 each) were treated with MCD (12.5 mM) for 10, 5, or 1 min before addition of IL-6 for another 15 min. The P15, P100, S100, and NE fractions were prepared, and aliquots (10 µl for P15 and P100, 20 µl for S100, and 2 µl for NE) assayed for DNA shift activity. The autoradiograms were quantitated by densitometry. The compartmental distribution as % of total in the cell of SIF-A DNA shift activity in Hep3B cells treated only with IL-6 was ~1, 16, 35, and 48% in the P15, P100, S100, and NE fractions, respectively, in this and additional experiments. These respective values were used at 100% to express the effect of MCD pretreatment on DNA shift activity in each compartment as depicted in the figure. B, Western blot analyses for PY-STAT3 in fractions illustrated in A. Aliquots of the respective P15 (10 µl), P100 (10 µl), S100 (50 µl), and NE (20 µl) fractions from A were Western-blotted and probed with anti-PY-STAT3 pAb. The data were quantitated by densitometry. The compartmental distribution as % of total in the cell of PY-STAT3 in cells treated with IL-6 alone for 15 min was 13, 18, 50, and 19% in the P15, P100, S100, and NE fractions, respectively. These respective values were used at 100% to express the effect of MCD pretreatment on PY-STAT3 in each compartment as depicted in the figure.

Fig. 6B shows that a brief treatment with MCD selectively and markedly decreased the content of Western blottable PY-STAT3 in the P100 cytoplasmic pellet fraction. Additionally, PY-STAT3 in the P15 fraction of MCD-free IL-6-treated cells was DNA binding deficient, consistent with the previous suggestion (12) that this pool of raft-associated PY-STAT3 represented molecules very recently Tyr-phosphorylated before the dimerization step and acquisition of DNA binding competence. However, the PY-STAT3 retained in the P15 membrane fraction in the 1-min MCD group was DNA binding competent suggesting that MCD had inhibited/slowed the departure from the raft of PY-STAT3 which was already DNA binding competent at that location.

Treatment of Hep3B cells with filipin III for 15 min before IL-6 or IFN- also provided trafficking data similar to the effect of a brief exposure to MCD (Fig. 7). The inhibitory effect of filipin III on IFN- signaling (Fig. 7, right side) confirms the prior observations of Taniguchi and colleagues (26). Moreover, in this experiment there was also increased accumulation of SIF-A PY-STAT3 and SIF-C PY-STAT1 DNA binding activity in the P15 membrane fraction of IL-6- or IFN--treated cells, respectively, accompanied by an inhibition in the P100 and S100 fractions and a lesser effect in the nuclear compartment.

View larger version (42K): [in this window] [in a new window]   Fig. 7.   Effect of filipin III on cytokine signaling in Hep3B cells. Groups of Hep3B cultures (4 each) were treated with filipin III (5 µg/ml) for 15 min before addition of IL-6 or IFN- for another 15 min. The P15, P100, S100, and NE fractions were prepared, and aliquots (10 µl for P15 and P100 and 20 µl for S100 and 5 µl for NE) were assayed for STAT-specific DNA shift activity. The autoradiograms were quantitated by densitometry, and the effect of filipin III pretreatment on DNA shift activity in each compartment is expressed as % of that in control cells treated only with cytokine in the same manner as in Fig. 6.

The data in Figs. 6 and 7 taken together suggest that raft integrity is required for the efficient departure of activated STATs from the plasma membrane. Moreover, the data in Fig. 6B showing the selectively marked inhibition of PY-STAT3 levels in the P100 fraction are consistent with a model in which this compartment may be located immediately downstream of the site of action of MCD, i.e. immediately downstream of the raft. This possibility is consistent with a model, proposed previously by Palade and colleagues (25), in endothelial cells that the cav-1-enriched sedimentable cytosolic complexes represent membrane-free complexes departing the rafts/caveolae.

Function of HSP90 in IL-6-Raft-STAT3 Signaling-- The inhibitor GA binds in the ATP-binding pocket in HSP90 and inhibits the physical and functional interactions of this chaperone with its target proteins (27). We have used GA to probe the function of HSP90 in STAT3 signaling. The design of these experiments involved maintaining confluent Hep3B cultures at 37 °C or exposing them to 39.5 °C for 12-16 h to raise their levels of HSP90 (a temperature stress equivalent to a fever of 103°F), followed by an evaluation of the effects of a 15-min pretreatment with GA (typically at 20 µM) on various IL-6-induced STAT3-mediated responses at the two temperatures. The Western blot data in Fig. 8A confirm the up-regulation of HSP90 in Hep3B cells at the higher temperature as illustrated in this instance in the P15 membrane and S100 cytosol fractions. Figs. 8B and 9A illustrate the effect of GA on the trafficking of IL-6-induced PY-STAT3 to the nucleus at the two temperatures. Whereas GA had a minimal effect at 37 °C, there was a clear reduction in trafficking of PY-STAT3 to the nucleus at 39.5 °C in the presence of the HSP90 inhibitor. Figs. 8C and Fig. 9B illustrate the effect of GA on IL-6-induced STAT3-specific DNA binding activity in the nuclear extracts. DNA binding activity was observed upon IL-6 treatment at both 37 and 39.5 °C, with a modest reduction at the higher temperature (to 28% that at 37 °C). Whereas GA had a minimal inhibitory effect on STAT3 DNA binding activity at 37 °C, the inhibition in cells induced with IL-6 at 39.5 °C was particularly marked.

View larger version (58K): [in this window] [in a new window]   Fig. 8.   Effect of GA on IL-6-induced PY-STAT3 and DNA shift activity in the nuclear compartment at 37 and 39.5 °C. A, HSP90 levels assayed by Western blotting in the P15 and S100 fractions from three different groups of Hep3B cells at 37 °C or after overnight (14 h) exposure 39.5 °C. Aliquots of the respective P15 (225 µg each) and S100 (45 µg each) were Western-blotted and probed using an anti-HSP90 mAb reactive with both the HSP90 and - species. B, GA inhibits PY-STAT3 nuclear trafficking at 39.5 °C. Hep3B cultures (3 per group) kept at 37 °C or exposed to 39.5 °C overnight were pretreated with GA (20 µM) for 15 min and then stimulated with IL-6 for 30 min. The protein concentrations in the NE fractions were determined using the micro-Bradford assay, and aliquots of the NE fraction (150 µg of protein per sample) were Western-blotted and probed for PY-STAT3. This experiment was replicated three times, and quantitative densitometric data are summarized in Fig. 9A. C, STAT3-specific DNA binding activity was assayed using the m67 SIE oligonucleotide in protein-matched (20 µg) volumes of the NE fractions from IL-6 and GA-treated cells as described in B. Three replications of this experiment were quantitated densitometrically, and the numerical data are summarized in Fig. 9B. Asterisk points to a constitutive DNA shift activity.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Effect of GA on IL-6-induced PY-STAT3 and DNA shift activity in the nuclear compartment at 37 and 39.5 °C. A, nuclear trafficking of PY-STAT3. The nuclear content of PY-STAT3 assayed by Western blotting in the experiment shown in Fig. 8B and two additional replications (n = 3) was quantitated densitometrically (mean ± S.E.). Panels a and b show the data normalized to the IL-6-treated value at 37 °C, whereas panel b' shows data normalized to the IL-6-treated value at 39.5 °C. The statistically significant (p < 0.05) paired two-way comparisons using a two-tailed Student's t test were p = 0.014 for the comparison between IL-6 alone at the two temperatures and p = 0.04 for the effect of GA at 39.5 °C. B, nuclear DNA shift activity. The nuclear content of DNA-binding competent PY-STAT3 (SIF-A) assayed in an oligonucleotide shift assay in the experiment shown in Fig. 8C and two additional replications (n = 3) was quantitated densitometrically (mean ± S.E.). Panels a and b show the data normalized to the IL-6-treated value at 37 °C, whereas panel b' shows data normalized to the IL-6-treated value at 39.5 °C. The statistically significant (p < 0.05) paired two-way comparisons using a two-tailed Student's t test were p = 0.002 for the comparison between IL-6 alone at the two temperatures and p = 0.001 for the effect of GA at 39.5 °C.

Immunopanning and Western blotting analyses data confirmed that, as expected, GA disrupted the interaction between HSP90 and STAT3 in cells at both 37 and 39 °C in that anti-STAT3 pAb failed to cross-immunoselect HSP90 from the S100 cytosol of cells treated with GA (Fig. 10A, upper panel). Furthermore, Fig. 10A, lower panel, confirms that treatment of Hep3B with GA at 39.5 °C but not at 37 °C inhibited the levels of IL-6-induced PY-STAT3 that could be immunoprecipitated with anti-STAT3 pAb from the S100 cytosol. Fig. 10B illustrates direct Western blot analyses without prior immunoprecipitation of PY-STAT3 in the S100 cytoplasmic fractions in cells treated with IL-6 and GA at the two temperatures. The data confirm that GA markedly inhibited PY-STAT3 activation by IL-6 at 39.5 °C in the cytoplasmic compartment with a lesser effect at 37 °C.

View larger version (54K): [in this window] [in a new window]   Fig. 10.   GA disrupts the interaction between STAT3 and HSP90 at both 37 and 39.5 °C. A, aliquots of the S100 fraction from the 37 and 39.5 °C portions of experiments similar to that shown in Fig. 8B (260 µg protein per sample) were immunopanned (IP) using anti-STAT3 pAb and the inclusion of HSP90 and PY-STAT3 in the immunoprecipitates assayed by Western blotting. B, aliquots (protein-matched in each set) of the S100 (45 µg) fractions illustrated in A above were directly Western-blotted and probed for PY-STAT3.

To directly test whether HSP90 was involved in the function of an IL-6-inducible STAT3-responsive liver-specific gene promoter at normal and/or elevated temperatures, we evaluated the effect of GA on the IL-6-induced up-regulation of a reporter/luciferase construct containing four copies of the STAT3-binding element from the human angiotensinogen promoter at 37 and 39.5 °C. Fig. 11 summarizes the pooled data from four such experiments. The data show that IL-6 inducibility of this reporter construct was maintained at 39.5 °C at a level approximately half that at 37 °C. Whereas GA had an inhibitory effect on IL-6-inducibility of this reporter construct at both temperatures (37 °C, p = 0.001; 39.5 °C, p < 0.001), this inhibitor dramatically reduced reporter construct inducibility at 39.5 °C compared with that at 37 °C (10-fold compared with 4-fold respectively; p = 0.002). These data show that from a functional standpoint HSP90 not only contributes toward assisting/chaperoning IL-6 signaling at 37 °C but has a major role in maintaining this signaling at elevated temperature.

View larger version (11K): [in this window] [in a new window]   Fig. 11.   GA inhibits activation of an STAT3-DNA binding-element/luciferase reporter construct by IL-6. Hep3B cell cultures in 6-well multiwell plates were transfected with p950M4 reporter/luciferase plasmid (250-500 ng per well) together with a constitutive expression vector for -galactosidase (pCH110) (50 ng per well) as described under "Materials and Methods." The transfected cells were incubated at 37 °C for 10-12 h and then half the cultures shifted to 39.5 °C for another 10-12 h. Duplicate wells were pretreated with GA for 15 min and then with IL-6 for 6 h at the respective temperature, and the cell extracts were prepared, and luciferase and -galactosidase activities were assayed. Within each experiment, the luciferase data were normalized for -galactosidase activity in each extract. Figure illustrates pooled data (mean ± S.E.) from four independent experiments (n = 4; with each experimental group in duplicate) expressed in terms of the luciferase activity in cells treated with IL-6 alone at 37 °C, as 100% in panels a and b, and that in cells treated with IL-6 alone at 39.5 °C in panel b'. The statistically significant (p < 0.05) paired two-way comparisons using the two-tailed Student's t test were p = 0.002 for the comparison between IL-6 alone at the two temperatures, p = 0.001 for the effect of GA at 37 °C (4-fold reduction), p < 0.001 for the effect of GA at 39.5 °C (10-fold reduction), and p = 0.002 for a comparison of the difference in the effect of GA at the two temperatures (4- versus 10-fold reduction).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this report we identify cav-1 and HSP90 as STAT3-associated accessory proteins at the level of the plasma membrane raft and in cytoplasmic complexes, and we provide evidence for a functional role for HSP90 as a chaperone in STAT3 signaling. The associations between and among STAT3, cav-1, and HSP90 were observed in the membrane raft and cytoplasmic fractions of human Hep3B hepatocytes, human lung alveolar type II-like cells (A549 cell line), a human lung fibroblast cell strain, and BPAEC. Thus these associations were observed in different cell types and may represent a general feature of STAT3 signaling.

There is now increasing evidence that cytokine signaling is initiated at the level of the plasma membrane in specialized cav-1-containing detergent-resistant microdomains (12, 13, 26, 28-32). The Janus kinases JAK1, JAK2, and Tyk2, the STAT species STAT1, STAT3, and their PY-activated versions, and the cytokine receptor chains gp130, IFNAR1, IFNAR2, IFN-R, IFN-R, and IL-2R have all been found in cav-1-containing plasma membrane rafts in different cell types (12, 13, 26, 28-32). Functionally, we reported earlier that disruption of raft integrity by MCD markedly inhibited IL-6 and IFN- signaling (12), and Taniguchi and colleagues (26) reported that filipin III could markedly and reversibly inhibit IFN- signaling (also see Ref. 32). However, the mechanisms by which PY-STATs depart the plasma membrane raft to enter the cytoplasm remain largely unexplored.

The observation that a brief exposure to the raft disrupters MCD or filipin III resulted in accumulation of DNA binding competent STAT3 and STAT1 in the P15 membrane fraction of IL-6 or IFN--treated Hep3B cells is consistent with a role for raft integrity in the departure of activated STATs from the plasma membrane. The Western blotting data showing a marked selective depletion of PY-STAT3 by brief MCD pretreatment in the cytoplasmic P100 sedimentable cav-1-enriched fraction of IL-6-treated cells is consistent with these P100 complexes representing a trafficking intermediate immediately downstream of the membrane raft. A similar model has been previously proposed by Palade and colleagues (25) for the departure of cav-1-containing membrane-free supramolecular complexes from plasma membrane rafts/caveolae in bovine pulmonary endothelial cells.

With respect to STAT proteins in the cytoplasm, we reported several years ago that STAT3, STAT1, and STAT5b existed in Hep3B and rat liver cytoplasm in high molecular mass complexes of size 200-400 kDa ("statosome I") and 1-2 MDa ("statosome II") by gel filtration in association with accessory proteins (14). Upon IL-6 induction, PY-STAT3 also appeared in complexes with a broad size distribution from 200-400 kDa to 1-2 MDa (14). Independently, Lackmann et al. (33) reported observing inactive STAT1 in HeLa cell cytosol in complexes of size 150-300 kDa, and the inclusion of IFN--induced PY-STAT1 in complexes of the same size. Also independently, Yeung et al. (34) reported the association of PY-STAT3 and PY-STAT5a and -b in the cytosol of colony-stimulating factor-1-induced BAC1.2F5 macrophage line in complexes of size 150-600 kDa and >2 MDa. We have identified previously GRP58 as an accessory protein associated with STAT3 in the 200-400-kDa cytosolic complexes (13, 14). Subsequently GRP58, a thiol-dependent protein-disulfide isomerase and a chaperone, was also found in association with STAT3 and PY-STAT3 in purified plasma membrane raft fractions (13). The function of GRP58 in STAT3 signaling remains largely unexplored.

In the present study we have identified cav-1 and HSP90 as STAT3-associated accessory proteins at the level of the plasma membrane and in the cytoplasm. A precedent for an interaction of cav-1 and HSP90 with the same cellular signaling molecule has been established previously in the case of endothelial cell nitric-oxide synthase (reviewed in Refs. 35-37). In the case of STAT3, the function of cav-1 appears to involve raft/cytoplasmic trafficking. The function of HSP90 in IL-6-raft-STAT3 signaling appears to include preservation of signaling at elevated temperature.

IL-6, which is itself a central pyrogen, functions in the body in the context of fever (1-5). During acute infections and injury, IL-6 stimulates the synthesis and secretion of protective "positive acute-phase" plasma proteins by the liver (1-5) in large part by activating STAT3 (6-11). Thus, during the acute-phase response, the hepatocyte, held at a high temperature in the body core during the accompanying fever, needs to preserve its ability to respond to IL-6. Our data show that HSP90 contributes to the continued maintenance of IL-6-induced STAT3- dependent responses at 39.5 °C (corresponding to a fever of 103 °C). We suggest that the protective chaperone effect of HSP90 on STAT3 signaling is likely to be a general phenomenon and to include responses to other cytokines in other cell types during fever.

A further insight of physiological significance is the observation that IL-6 itself through activation of STAT3 and CAAT enhancer-binding factor  up-regulates HSP90 gene expression (38, 39). Thus, IL-6 itself turns on a cellular mechanism that protects its signaling pathway from the deleterious effect of elevated body temperature.

	ACKNOWLEDGEMENTS

We thank Dr. Joseph D. Etlinger for numerous insightful discussions including suggestions for experiments at elevated body temperature (39.5 °C), Ann-Marie Snow for assistance with electron microscopy, and Joanne Abrahams for help with image analyses. We thank Dr. Ashok Kumar for providing the p950M4 4XSTAT3element/luciferase reporter construct, Dr. Susan Olson for BPAEC cells, and Dr. Anuradha Ray for ATII-like and LF cells.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Research Grant CA-82647.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. Cell Biology & Anatomy, Rm. 201, Basic Sciences Bldg., New York Medical College, Valhalla, NY 10595. Tel.: 914-594-4196; Fax: 914-594-4825; E-mail: pravin_sehgal@nymc.edu.

Published, JBC Papers in Press, September 13, 2002, DOI 10.1074/jbc.M205935200

² K. Patel, V. Kumar, and P. B. Sehgal, unpublished data.

³ P. B. Sehgal and V. A. Fried, unpublished data.

⁴ Correspondence from BD Biosciences.

	ABBREVIATIONS

The abbreviations used are: IL, interleukin; 5'-ND, 5'-nucleotidase; mAb, monoclonal antibody; ATII-like cells, alveolar epithelial type II-like cell line A549; BiP, glucose-regulated protein 78 GRP78/BiP; BPAEC, bovine pulmonary arterial endothelial cells; cav-1, caveolin-1; ELB, extract lysis buffer; GA, geldanamycin; GRP58, glucose-regulated protein 58 (also called ER-60 and ERp57); HSP90, heat shock protein 90; IFN, interferon; JAK, Janus kinase; LF, lung fibroblast strain CCD Lu-11; MCD, methyl--cyclodextrin; NE, nuclear extract; NRS, nonimmune rabbit serum; Nup62, nuclear pore-associated protein 62; PBS, phosphate-buffered saline; SIE, m67 mutant of serum-inducible element from c-fos promoter; SIF, SIE oligonucleotide-binding factor; STAT, signal transducer and activator of transcription protein family; pAb, polyclonal antibody; ER, endoplasmic reticulum; PY, Tyr-phosphorylated.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES