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

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Murthy, S. Articles by Field, F. J. Search for Related Content PubMed PubMed Citation Articles by Murthy, S. Articles by Field, F. J. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 13, 9222-9229, March 31, 2000

Tumor Necrosis Factor- and Interleukin-1 Inhibit Apolipoprotein B Secretion in CaCo-2 Cells via the Epidermal Growth Factor Receptor Signaling Pathway^*

Shubha Murthy, Satya N. Mathur, and F. Jeffrey Field

From the Department of Internal Medicine and Veterans Affairs, University of Iowa, Iowa City, Iowa 52242

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In inflammatory conditions of the gut, cytokines are released into the mucosa and submucosa propagating and sustaining the inflammatory response. In CaCo-2 cells, we have shown that various inflammatory cytokines interfere with the secretion of lipids, an effect that is likely caused by the release of a ligand to the epidermal growth factor (EGF) receptor. In the present study, the role of the EGF receptor signaling pathway and the effects of the cytokines tumor necrosis factor- (TNF-) and and interleukin 1 (IL-1) on triacylglycerol-rich lipoprotein secretion were investigated. CaCo-2 cells were incubated with oleic acid to enhance triacylglycerol-rich lipoprotein secretion. TNF- and IL-1 significantly decreased the basolateral secretion of apolipoprotein B (apoB) mass, with IL-1 being more potent. Tyrphostin, an inhibitor of the EGF receptor intrinsic tryosine kinase, prevented or markedly attenuated the decrease in apoB secretion by TNF- or IL-1. Both cytokines increased the phosphorylation of the EGF receptor by 30 min. Moreover, phosphotyrosine immunoblots of the EGF receptor demonstrated an increase in tyrosine residues phosphorylated by 0.5 and 6.5 h. At both these time points, TNF- and IL-1 also decreased the binding of EGF to its cell surface receptor. At 6.5 h, activation of the EGF receptor was sustained. In contrast, the early activation of the receptor was only transient as receptor phosphorylation and binding of EGF to its receptor returned to basal levels by 2 h. Preventing ligand binding to the EGF receptor by a receptor-blocking antibody attenuated receptor activation observed after 6.5 h. This did not occur at 0.5 h, suggesting that early activation of the EGF receptor was non-ligand-mediated. Similarly, apoB secretion was inhibited by an early non-ligand-mediated process; whereas at the later time, inhibition of apoB secretion was ligand-mediated. Thus, the inflammatory cytokines TNF- and IL-1 interfere with the secretion of triacylglycerol-rich lipoproteins by both early and delayed signaling events mediated by the EGF receptor signaling pathway.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Inflammatory conditions of the small intestine, such as gluten-sensitive enteropathy or Crohn's disease, can result in mucosal damage leading to malabsorption of nutrients (1). The localized release of inflammatory cytokines into the mucosa and submucosa likely mediates and perpetuates the inflammatory response (2). Lymphocytes, monocyte/macrophages, and mast cells that infiltrate the mucosa are the major source of these inflammatory peptides (3, 4). Recent evidence, however, demonstrates that intestinal epithelial cells also synthesize and secrete a number of inflammatory cytokines (5-10). Moreover, because they possess receptors for several cytokines (8, 11-13), it is likely that enterocytes, by interacting directly with inflammatory cytokines, participate in and contribute to the various pathophysiological derangements observed in inflammatory conditions of the gut. Evidence in support of such a notion was demonstrated by the ability of inflammatory cytokines to up-regulate the synthesis of acute phase proteins (13) and complement factors (12) in cultured human intestinal cells.

Cytokines, however, are not the only mediators of inflammation which are released into the mucosa under conditions of inflammation. Several other bioactive molecules such as growth factors, prostaglandins, and reactive oxygen species are secreted by intestinal epithelial cells (14). Together with cytokines, these factors act coordinately to regulate the extent of mucosal injury as well as mediate tissue repair. Growth factors EGF¹ and TGF-, ligands to the transmembrane EGF receptor present on intestinal epithelial cells, have been shown to play a significant role in the restitution of mucosal damage in the gut (15, 16). Ligand-mediated activation of the EGF receptor triggers a cascade of events resulting in enhanced cell migration and proliferation that serve to repair the denuded surface of the mucosa. Moreover, by mediating the increased production of mucopolysaccharides, prostaglandins, and extracellular matrix components, the EGF receptor likely plays an important role in protecting mucosal surfaces from further injury (17). By modulating transport processes such as ion exchange (18-20) and glucose absorption (19-21) the receptor might also play a significant role in regulating intestinal cell function during inflammation. The role of the EGF receptor in decreasing the absorptive function of enterocytes in inflammation, however, has not been investigated. Cytokines that are considered proinflammatory, under certain conditions, may also suppress inflammation and promote wound healing and repair (22, 23). Similar to the EGF receptor, they have been shown to mediate cellular functions such as proliferation, differentiation, deposition of extracellular matrix, and cell motility. Thus, together with the EGF receptor signaling pathway, cytokines likely modulate intestinal epithelial cell function during inflammation.

In a previous study, we demonstrated that certain inflammatory cytokines interfered with normal intestinal lipoprotein synthesis and secretion (24). In the present study, we addressed whether cytokines inhibit the secretion of lipoproteins by activating the EGF receptor signaling pathway. The effects of two inflammatory cytokines, TNF- and IL-1, on EGF receptor activation and triacylglycerol-rich lipoprotein secretion were studied in a cultured human intestinal cell line, CaCo-2. The results demonstrate that TNF- and IL-1 inhibit the secretion of triacylglycerol-rich lipoproteins by both a rapid and delayed activation of the EGF receptor. This occurs by non-ligand- and ligand-mediated mechanisms, respectively.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Recombinant human TNF- and IL-1 were purchased from R & D Systems (Minneapolis). Carrier-free EGF was purchased from Becton Dickinson (Bedford, MA). Horseradish peroxidase substrate, SuperSignal West Femto maximum sensitivity substrate kit, and IODO-GEN were purchased from Pierce. Rabbit polyclonal antibody to human EGF receptor was from Upstate Biotechnology Inc. (Lake Placid, NY). Mouse monoclonal anti-phosphotyrosine antibody, mouse monoclonal anti-EGF receptor-blocking antibody (mAb 528), goat anti-mouse IgG conjugated to horseradish peroxidase, and protein A+G-agarose were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-human apoB polyclonal antibody and tyrphostin AG1478 were obtained from Calbiochem. Mouse monoclonal antibody to human apoB and rabbit anti-human apoB polyclonal antibody conjugated to horseradish peroxidase were bought from Biodesign (Kennebunkeport, ME). Recombinant protein A-Sepharose was purchased from Repligen (Cambridge, MA). Oleic acid, BSA, and Glycerol Phosphate Oxidase Trinder Kit were purchased from Sigma. A TMB microwell peroxidase substrate system containing 3,3',5,5'-tetramethylbenzidine and hydrogen peroxide was purchased from Kirkegaard and Perry (Gaithersburg, MD). Nunc 96-well immunoplates were obtained from PGC Scientific (Gaithersburg, MD). CellTiter 96 was from Promega (Madison, WI). ³²P_i (6,000 Ci/mmol) was purchased from NEN Life Science Products. Carrier-free ¹²⁵I (100 mCi/ml) was purchased from ICN Biomedicals Inc. (Costa Mesa, CA).

Cell Culture-- CaCo-2 cells were cultured on T-75 flasks (Corning Glassworks, Corning, NY) in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) with 4.5 g/liter glucose and supplemented with 10% fetal bovine serum (Summit Biotechnology, Fort Collins, CO), 2 mM glutamine, 100 units/ml penicillin, 100 µg/ml streptomycin, and 50 µg/ml gentamicin. Once the flasks reached 80% confluence, the cells were split and seeded at a density of 0.2 × 10⁵ cells/well onto polycarbonate micropore membranes (0.4-µm pore size, 6.5-mm diameter) inserted into transwells (Costar, Cambridge, MA). For experiments in which triacylglycerol mass, EGF receptor phosphorylation, and cell surface ¹²⁵I-EGF binding were estimated, cells were subcultured in 24-mm diameter transwells. Cells were fed every other day and were used 14 days after seeding.

On the day of the experiment, cells were washed with media, and cytokines were added to the lower chambers in serum-free Hanks' balanced saline solution and 1 M HEPES (HBSS) or M199 and 1 M HEPES (M199) containing 0.1% BSA. Control cells received medium containing 0.1% BSA alone. All cells received 250 µM oleic acid and 62.5 µM BSA in the apical chamber. Incubations were carried out for 18 h or less at 37 °C in an atmosphere of 95% compressed air and 5% CO₂.

Cell Viability/Proliferation-- Cell viability and proliferation were assessed by measuring the activity of mitochondrial dehydrogenase using the CellTiter 96 assay kit as described previously (24). This assay is based on the mitochondrial conversion of a tetrazolium salt into a blue formazan product that is released into the medium. After an overnight incubation with the treatments, the release of the colored formazan dye into the medium was measured spectrophotometrically. Compared with control cells, the relative absorbance of the dye released from cells incubated with TNF- or IL-1 was 0.97 ± 0.09 or 0.96 ± 0.13, respectively.

Estimation of ApoB Mass-- ApoB mass in cells and basal media was determined by sandwich enzyme-linked immunosorbent assay as described previously (24). The presence of the treatments in the media did not interfere with the estimation of apoB mass by the enzyme-linked immunosorbent assay.

EGF Receptor Phosphorylation-- Cells were incubated for 18 h with 500 µCi of ³²P_i/well in phosphate-free Dulbecco's modified Eagle's medium. Treatments were added to the basal wells in the continued presence of labeled inorganic phosphate. After incubation, cells were rinsed in ice-cold phosphate-buffered saline and scraped and lysed in 1 ml of radioimmune precipitation buffer containing 1 mM phenylmethanesulfonyl fluoride, 21 µM leupeptin, 2 mM benzamidine, 30 µl/ml aprotinin, 1 mg/ml soybean trypsin inhibitor, 2 mM sodium orthovanadate, 20 µM sodium pyrophosphate, and 20 µM sodium fluoride. The cell lysates were precleared by shaking for 1 h at 4 °C with protein A-Sepharose followed by a quick high speed centrifugation. EGF receptor was immunoprecipitated from the precleared supernatants by incubating for 18 h with 1 µg/ml rabbit anti-human EGF receptor antibody. The antigen-antibody complexes were precipitated by incubating with protein A+G-agarose for 1 h at room temperature followed by a brief high speed centrifugation. The immunoprecipitates were washed extensively with phosphate-buffered saline and the EGF receptor protein dissociated from the antibody-antigen complex with 30 µl of 2 × Laemmli sample buffer and 15 µl of 0.2 M glycine buffer (pH 2). The protein was resolved by SDS-PAGE on 8% porous gels as described previously (25). Gels were fixed with 7% acetic acid and 5% methanol, dried, and exposed to x-ray film for 8 h. The incorporation of labeled inorganic phosphate into the EGF receptor was estimated by scanning the gels on Ambis 4000 biological image analyzer (Scanalytics, Billerica, MA).

Binding of ¹²⁵-EGF to Cell Surfaces-- EGF was iodinated using IODO-GEN reagent as prescribed by Pierce. After treatment with TNF- or IL-1, binding of ¹²⁵I-EGF to cell surface EGF receptors was estimated as described previously (26). Cells were washed with M199 and incubated for 2 h at 4 °C with 0-1000 ng/ml iodinated EGF (0.0003 µCi/ng). From previous experiments it was found that the binding of 100 ng/ml radiolabeled EGF to cell surface EGF receptors plateaus after 2 h of incubation. EGF was diluted in M199 containing 0.1% BSA and added to the lower wells. M199 was added to the upper chambers. After extensive washing with ice-cold M199 containing 0.1% BSA followed by several rinses with M199 alone, cells were scraped in 1 ml of radioimmune precipitation buffer and counted in a gamma counter. Nonspecific binding was estimated by incubating cells with 5 µg/ml cold EGF in the presence of 50 ng/ml ¹²⁵I-EGF and did not exceed 5% of the total binding of labeled EGF. The specificity of the binding of EGF to its cell surface receptor in CaCo-2 cells was determined by Bishop and Wen (26), who demonstrated that the binding of iodinated EGF to cell surfaces is abolished in the presence of the EGF receptor-blocking antibody, mAb 528. Furthermore, when cells were incubated at 4 °C with labeled EGF and then exposed to bis(sulfosuccinimidyl)suberate (Pierce) to cross-link the ligand to its receptor, more than 80% of the EGF bound to cell surfaces was recovered in a band corresponding to the EGF receptor (data not shown). In experiments in which cells were incubated with the cytokines in the presence of mAb 528, the blocking antibody bound to cell surfaces was removed prior to estimation of cell surface binding of labeled EGF. This was accomplished by extensively washing the cells with 100 mM sodium chloride and 500 mM glycine, pH 3, followed by several rinses with M199. Subsequent binding of labeled EGF to control cells incubated with or without mAb 528 was similar, indicating therefore that the stringent wash protocol effectively removed the bound monoclonal antibody from cell surfaces.

Western Blotting of the EGF Receptor-- After incubation with the treatments, EGF receptor was immunoprecipitated from precleared cell lysates as described above. The receptor protein was dissociated from the antigen-antibody complex, separated by SDS-PAGE, and electroblotted onto polyvinylidene difluoride membranes at 15 V for 18 h. The membranes were blocked for 1 h at 37 °C in phosphate-buffered saline (10 mM sodium phosphate, 100 mM sodium chloride, pH 7.4) containing 5% non-fat dry milk, 5% normal goat serum, and 0.1% Tween 20 (blocking buffer). The membranes were then incubated for 1 h with mouse monoclonal anti-phosphotyrosine antibody diluted 20,000-fold in the blocking buffer. After washing with phosphate-buffered saline containing 0.1% Tween 20, the membranes were incubated for 1 h at room temperature with goat anti-mouse IgG-horseradish peroxidase diluted 50,000-fold in blocking buffer. After extensive washing, the membranes were incubated with horseradish peroxidase chemiluminescent substrates, wrapped in Saran Wrap, and then exposed to x-ray film. Band densities were scanned on Hewlett-Packard ScanJet IIcx/T scanner, Hewlett Packard (Greely, CO) and quantitated with the computer-assisted program, Sigma Gel, Jandel Scientific (San Rafael, CA).

Chemical Analyses-- Total protein content in cells was determined by the method of Lowry et al. (27). Triacylglycerol mass in cells was measured using the Glycerol Phosphate Oxidase Trinder Kit as described previously (28).

Statistical analyses of data were performed by analysis of variance, Tukey's t test, Student's t test, the least squares method of determining the best fitting straight line, and small sample t tests for parallelism and common intercepts (29).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

ApoB and Triacylglycerol Secretion-- In a previous study, we demonstrated that TNF- and IL-1 decreased the basolateral secretion of apoB by CaCo-2 cells in the absence of fatty acids, experimental conditions that do not promote the secretion of triacylglycerol-rich lipoproteins (24). To address the regulation of triacylglycerol-rich lipoprotein secretion by TNF- and IL-1, CaCo-2 cells were incubated for 18 h with increasing concentrations of TNF- or IL-1 and 250 µM oleic acid. Oleic acid, at this concentration, has been demonstrated in CaCo-2 cells to stimulate the secretion of lipoproteins enriched in triacylglycerols (30). After the incubation, the mass of apoB within cells and that secreted into the basolateral medium was estimated. Both TNF- and IL-1 decreased the secretion of apoB with IL-1 being the more potent cytokine (Table I). Compared with control cells, IL-1, at concentrations of 0.001 and 0.1 ng/ml, decreased apoB secretion by 30 and 60%, respectively. Higher concentrations of IL-1 did not decrease apoB secretion further. In contrast, compared with the effects of IL-1, TNF- decreased apoB secretion to a similar degree but required much higher concentrations, 10 and 100 ng/ml, respectively. Neither TNF- nor IL-1 altered the amount of apoB mass within cells.

View this table: [in this window] [in a new window]   Table I Effect of TNF- and IL-1 on the basolateral secretion of apoB in CaCo-2 cells CaCo-2 cells were incubated for 18 h with increasing concentrations of TNF- or IL-1 dissolved in HBSS containing 0.1% BSA. Control cells received HBSS plus BSA alone. Cytokines were added to the lower well. The upper well contained 250 µM oleic acid complexed to 62.5 µM BSA. The accumulation of apoB in the lower wells and in cells was estimated by sandwich enzyme-linked immunosorbent assay as described under "Experimental Procedures." Results are expressed as mean ± S.E. of six wells/treatment.

A decrease in apoB secretion by cells incubated with either cytokine suggested that TNF- and IL-1 caused a decrease in the number of lipoprotein particles being secreted. To address the effect of the cytokines on the amount of triacylglycerols carried per lipoprotein particle, cells were incubated for 18 h with oleic acid and 100 ng/ml TNF- or 10 ng/ml IL-1. The amount of triacylglycerols within cells and that secreted into the basolateral medium was then estimated. Compared with control cells, the secretion of triacylglycerols by cells incubated with TNF- was decreased modestly (23.61 ± 2.39 versus 19.48 ± 1.09 µg/mg protein, n = 6, p < 0.05). In cells incubated with IL-1, less triacylglycerol was secreted compared with control cells or cells incubated with TNF- (23.61 ± 2.39 versus 16.04 ± 0.03 µg/mg protein, p < 0.01, versus TNF- p < 0.05). Neither TNF- nor IL-1 altered the amount of triacylglycerols within cells (control cells, 58.71 ± 1.7; TNF- cells, 61.53 ± 2.37; IL-1 cells, 62.48 ± 2.98 µg/mg; n = 6). Thus TNF- and IL-1 decrease the secretion of triacylglycerol-rich lipoproteins.

Activation of the EGF Receptor-- We have demonstrated previously that ligand-mediated activation of the EGF receptor decreases the secretion of lipids and apoB in CaCo-2 cells (31). Because TNF- and IL-1 have been observed to modulate the function of the EGF receptor in human fibroblasts (32), we addressed the possibility that TNF- and IL-1 interfered with triacylglycerol-rich lipoprotein secretion in CaCo-2 cells by activating the EGF receptor. Intrinsic tyrosine kinase activity of the EGF receptor is essential for EGF receptor-mediated signaling events (33). To investigate the role of the EGF receptor signaling pathway in apoB secretion, EGF receptor tyrosine kinase activity was inhibited by tyrphostin AG1478 (34). CaCo-2 cells were incubated for 18 h with oleic acid and increasing concentrations of TNF- or IL-1 in the presence or absence of tyrphostin AG1478. The amount of apoB secreted was then estimated. As shown in Fig. 1A, tyrphostin completely prevented the decrease in apoB secretion observed in cells incubated with TNF- alone. In cells incubated with IL-1, tyrphostin significantly attenuated the decrease in apoB secretion. Moreover, tyrphostin completely blocked the decrease in apoB secretion observed in cells incubated with EGF, a ligand of the EGF receptor which activates receptor tyrosine kinase. Tyrphostin itself had no effect on the secretion of apoB. These results suggest that the cytokines decreased apoB secretion by activating the EGF receptor signaling pathway. This was addressed further by incubating cells for 18 h with 100 ng/ml TNF-, 10 ng/ml IL-1, or both. In addition, some cells were also incubated with 100 ng/ml EGF alone or together with either TNF- or IL-1. This concentration of EGF causes saturation of cell surface binding (see Fig. 3A) and likely maximal stimulation of the EGF receptor signaling pathway. At the end of the incubation, the amount of apoB secreted was estimated. The results are shown in Fig. 1B. ApoB secretion by cells incubated with EGF was decreased dramatically and was 2-fold less than the amount secreted by cells incubated with either cytokine alone. When cells were incubated with both cytokines together, the decrease in apoB secretion was additive and was similar to the inhibition induced by EGF alone. It is likely, therefore, that together TNF- and IL-1 act additively to activate the EGF receptor signaling pathway leading to an inhibition in apoB secretion. Moreover, compared with the effects of EGF alone, addition of either TNF- or IL-1 to cells incubated with EGF did not decrease the secretion of apoB further, suggesting that the cytokines were acting through the same pathway as EGF in inhibiting apoB secretion.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Effect of TNF- or IL-1 on the basolateral secretion of apoB in the presence or absence of tyrphostin AG1478. Panel A, CaCo-2 cells were incubated for 18 h with increasing concentrations of TNF-, IL-1, or 1.25 ng/ml EGF in the presence (closed bars) or absence (open bars) of 1 µM tyrphostin. Tyrphostin dissolved in 0.003% dimethyl sulfoxide (DMSO) was added to the lower wells containing HBSS, 0.1% BSA, and either TNF-, IL-1, or EGF. Control wells received medium containing BSA and dimethyl sulfoxide alone. 250 µM oleic acid bound to 62.5 µM BSA was added to the upper chamber. Accumulation of apoB in the basal medium was estimated by sandwich enzyme-linked immunosorbent assay. Results from one of three representative experiments are shown as the mean ± S.E. of four wells/treatment. *p < 0.01 compared with control cells; **p < 0.01 compared with corresponding treatment without tyrphostin. Panel B, cells were incubated for 18 h with 100 ng/ml TNF-, 10 ng/ml IL-1, or both. Some cells were also incubated with 100 ng/ml EGF alone or together with either TNF- or IL-1. The cytokines were dissolved in HBSS containing 0.1% BSA and were added to the lower wells. 250 µM oleic acid bound to 62.5 µM BSA was added to the upper chamber. Control cells received BSA alone. ApoB secreted into the basal wells was estimated. Results from one of two representative experiments are shown as the mean ± S.E. of six wells/treatment. *p < 0.01 compared with BSA alone.

Because the EGF receptor is a substrate for its intrinsic tyrosine kinase (33), the above results would suggest that activation of the EGF receptor tyrosine kinase by TNF- or IL-1 should result in receptor autophosphorylation. To address this, cells were prelabeled with ³²P_i. They were then incubated for 5-60 min with 100 ng/ml TNF- or 10 ng/ml IL-1. The incorporation of inorganic phosphate into the EGF receptor was estimated after immunoprecipitation of the receptor and SDS-PAGE separation (Fig. 2, A and B). Both cytokines caused a rapid, early increase in phosphorylation of the receptor which returned to base line by 15 min. At 30 min, however, there was a marked increase in the incorporation of labeled phosphate into the EGF receptor. This was also transient, returning to basal levels by 60 min. EGF, a potent ligand for the EGF receptor, caused a marked increase in the phosphorylation of its receptor at 30 min. Cells were next incubated with TNF-, IL-1, or EGF in the presence or absence of tyrphostin (Fig. 2C). Tyrphostin completely prevented the increase in EGF receptor phosphorylation caused by EGF, providing evidence of its ability to inhibit the activity of the receptor intrinsic tyrosine kinase in CaCo-2 cells. Moreover, tyrphostin significantly attenuated the effects of TNF- and IL-1 on EGF receptor phosphorylation, suggesting that in addition to tyrosine residues, threonine and/or serine residues were likely being phosphorylated as well.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Effect of TNF- or IL-1 on EGF receptor phosphorylation. CaCo-2 cells were incubated for 18 h in phosphate-free Dulbecco's modified Eagle's medium containing 500 µCi/well 32Pi added to the upper wells. Cells were then incubated for up to 1 h with 100 ng/ml TNF-, 10 ng/ml IL-1, or 300 ng/ml EGF in the continued presence of labeled inorganic phosphate. The treatments were added to the basal wells. Apical wells contained 250 µM oleic acid and 62.5 µM BSA. At the indicated times, cells were lysed, and the EGF receptor was immunoprecipitated and resolved by SDS-PAGE. The gels were dried and exposed to Kodak X-Omat x-ray film for 8 h. Panel A, representative autoradiogram of a gel demonstrating the time-dependent increase in phosphorylation of the EGF receptor after incubation with either TNF-, IL-1, or EGF. Panel B, the gels were counted on Ambis 4000 plate scanner, and the counts incorporated into the EGF receptor are represented as the mean ± S.E. cpm/well, n = 3/treatment. , TNF-; , IL-1. Panel C, cells were prelabeled for 18 h with 500 µCi/well 32Pi. This was followed by incubation for 1 h with 1 µM tyrphostin or 0.003% dimethyl sulfoxide (DMSO) alone in the continued presence of labeled Pi. 100 ng/ml TNF-, 10 ng/ml IL-1, or 300 ng/ml EGF was then added to the incubation medium. Control cells received phosphate-free Dulbecco's modified Eagle's medium containing 0.1% BSA and 0.003% dimethyl sulfoxide. The treatments were added to the lower wells. 250 µM oleic acid bound to 62.5 µM BSA was added to the upper wells at the same time as the cytokines. Cells were lysed and processed as described above. A representative autoradiogram is shown.

Cell Surface Binding of EGF-- After activation of the EGF receptor, its affinity for its ligand and/or cell surface receptor number is down-regulated, resulting in a decrease in cell surface binding of EGF (33). To address whether TNF- or IL-1 altered cell surface binding of EGF, cells were incubated with TNF- or IL-1 for 30 min, a time of maximal receptor phosphorylation. Cell surface binding of EGF was then estimated. The results of this experiment are shown in Fig. 3A. In cells incubated with TNF-, IL-1, or BSA alone, saturation of binding occurred at 50 ng/ml EGF. Compared with control cells, however, cells incubated with TNF- or IL-1 bound significantly less EGF at all concentrations examined. Scatchard plot analyses demonstrated a single class of high affinity EGF binding sites in cells incubated with either cytokine (Fig. 3B). Compared with control cells, the number of EGF receptors on cells incubated with TNF- or IL-1 decreased significantly from 228 ± 23 to 136 ± 9 and 84 ± 11 fmol/well, p < 0.01, respectively. Moreover, the affinity of binding was also decreased in cells incubated with the cytokines as evidenced by the increase in K_D, the dissociation constant of binding, from 0.17 ± 0.03 fM in control cells, to 0.27 ± 0.03 and 0.43 ± 0.11 fM in cells incubated with TNF- and IL-1, respectively (p < 0.01). Thus, TNF- and IL-1 decreased cell surface binding of EGF by decreasing both receptor number and affinity.

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Effect of TNF- or IL-1 on cell surface binding of EGF. Panel A, cells were incubated for 30 min with either 100 ng/ml TNF- () or 10 ng/ml IL-1 (). The cytokines were added to the lower wells in HBSS containing 0.1% BSA. Control cells () received medium containing BSA alone. 250 µM oleic acid complexed to 62.5 µM BSA was added to the upper wells. At the end of the incubation, cells were washed thoroughly with ice-cold M199, and the basolateral medium was replaced with M199 containing 0.1% BSA and increasing concentrations of 125I-EGF (0.1-1000 ng/ml, 0.0003 µCi/ng EGF). M199 was added to the upper wells. After 2 h of incubation at 4 °C, cells were washed thoroughly to remove unbound labeled EGF, lysed, and counted by gamma counting to estimate cell-associated labeled EGF. Nonspecific binding of labeled EGF to cell surfaces was estimated by incubating cells with 5 µg/ml cold EGF in the presence of 50 ng/ml labeled EGF. Results are expressed as the mean ± S.E. of bound EGF detected/mg of protein. Panel B, Scatchard plot analysis of the results presented in panel A. Symbols are the same as in panel A. Panel C, cells were incubated for increasing times under the same conditions as described for panel A. At the indicated times, cells were washed and incubated for 2 h at 4 °C with 50 ng/ml 125I-EGF as described above. Cells were washed, lysed, and counted. Results represent the mean ± S.E. of fmol of EGF bound/mg of protein, n = 4-12/treatment/time point. All data points in cells incubated with TNF- or IL-1 are significantly different from the control, p < 0.01. Symbols are the same as in panel A. Panel D, cells were incubated for 30 min or 6.5 h with 100 ng/ml TNF- or 10 ng/ml IL-1 in the presence of 250 µM oleic acid and 62.5 µM BSA as described above. Control cells were incubated with BSA alone. Cells were lysed, and the EGF receptor was immunoprecipitated and separated by SDS-PAGE. The proteins were electroblotted onto polyvinylidene difluoride membranes and probed with anti-phosphotyrosine antibody followed by anti-mouse IgG horseradish peroxidase as indicated under "Experimental Procedures." A representative autoradiogram of the immunoblots at each time point is shown.

It was postulated in human fibroblasts that rapid phosphorylation of the EGF receptor and modulation of its function by TNF- and IL-1 occur by a non-ligand-mediated mechanism (32). In CaCo-2 cells, however, IL-6, another inflammatory cytokine, inhibited the secretion of apoB by activating the EGF receptor by a ligand-mediated process (31). We next investigated a possible mechanism by which TNF- and IL-1 activated the EGF receptor and whether this activation played a role in regulating lipoprotein secretion by the cytokines. Cells were incubated for up to 24 h with TNF- or IL-1. At various time points, cells were harvested, and the cell surface binding of saturating amounts of labeled EGF was estimated. The results are shown in Fig. 3C. Compared with control cells, both TNF- and IL-1 caused a rapid and marked decrease in the binding of EGF to cell surfaces. The decrease in binding was apparent by 7.5 min and reached a maximal effect by 30 min. The rapid decrease in binding, however, was transient. After 2 h of incubation, EGF binding approached 80-85% of that observed in control cells. In cells incubated for 6.5 h or longer with the cytokines, however, EGF binding decreased again and remained suppressed throughout the entire 18-h incubation. Thus, in cells incubated with TNF- or IL-1, the EGF receptor is activated first by a rapid but transient mechanism and then by a second mechanism that is more delayed but sustained.

Because EGF receptor activation is associated with an increase in receptor tyrosine kinase activity, one would expect TNF- and IL-1 to increase EGF receptor tyrosine phosphorylation. To examine this, cells were incubated for 30 min or 6.5 h with TNF- or IL-1. The EGF receptor was then immunoprecipitated and separated by SDS-PAGE. After transfer to a filter, receptor tyrosine phosphorylation was estimated by immunoblotting with an anti-phosphotyrosine antibody. At both time points, compared with control cells, TNF-a and IL-1 increased the level of EGF receptor tyrosine phosphorylation (Fig. 3D). These results substantiate that TNF- and IL-1 activate the EGF receptor at both an early and late time point. The results also confirm that the increase in phosphorylation of the EGF receptor observed after 30 min of incubation with the cytokines (Fig. 2) is caused, at least in part, by an increase in phosphorylation of tyrosine residues.

Non-ligand versus Ligand-mediated Activation of the EGF Receptor-- The initial transient and later sustained activation of the EGF receptor by TNF- and IL-1 suggest that the cytokines were activating the receptor by two separate mechanisms. We investigated the possibility that these two mechanisms of receptor activation involved a ligand- versus a non-ligand-mediated process. CaCo-2 cells were incubated for 0.5, 6.5, or 18 h with either TNF- or IL-1 in the presence or absence of mAb 528, a specific blocking monoclonal antibody to the ligand binding domain of the EGF receptor (35). Because the antibody possesses no receptor agonist activity, it prevents ligand binding without itself activating the receptor. The antibody was added at a concentration of 0.5 µg/ml, which completely prevents binding of saturating amounts of EGF to cell surfaces (results not shown). After the incubation, cells were washed thoroughly to remove cytokines and the blocking antibody. Cell surface binding of EGF was then estimated. As shown in Fig. 4, the initial transient decrease in EGF binding to cells incubated for 30 min with either TNF- or IL-1 was not altered by the receptor blocking antibody. In contrast, after 6.5 h of incubation with either cytokine, mAb 528 significantly attenuated and by 18 h completely prevented the decrease in binding of EGF to CaCo-2 cells. These results indicate that TNF- and IL-1 cause a rapid activation of the EGF receptor by a non-ligand-mediated mechanism, whereas the more prolonged and gradual delay in activation is caused by ligand binding to the receptor.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Ligand-mediated and non-ligand-mediated activation of the EGF receptor by TNF- and IL-1. Cells were incubated for 30 min, 6.5 h, or 18 h with 100 ng/ml TNF-, 10 ng/ml IL-1, or BSA alone in the presence (closed bars) or absence (open bars) of 0.5 µg/ml mAb 528. The treatments were provided as described for Fig. 1. At the indicated times, cells were washed extensively with 100 mM sodium chloride and 500 mM glycine, pH 3, followed by M199. Cell surface binding of 50 ng/ml 125I-EGF was estimated as described for Fig. 3C. The results are shown as the mean ± S.E. of fmol of EGF bound/mg of cell protein. n = 8/treatment. *p < 0.01 compared with cells incubated with BSA alone; **p < 0.01 compared with corresponding treatment without mAb 528.

The results suggest that TNF- and IL-1 inhibit the secretion of apoB by activating the EGF receptor. During the first few hours of incubation, the EGF receptor is activated by non-ligand-mediated events (Figs. 3C and 4). After 5 h of incubation, however, the activation of the EGF receptor occurs by ligand binding and remains sustained throughout the incubation period. Whether the decrease in apoB secretion is mediated by the early and/or delayed mechanism of receptor activation was addressed next. CaCo-2 cells were incubated for two consecutive 4.5 h-periods followed by 9 h with TNF- or IL-1. MAb 528 was added to some of the cells to prevent ligand binding. Another set of cells was incubated for 18 h under similar conditions. After the incubations, the secretion of apoB mass was estimated. The results are shown in Fig. 5. Compared with control cells, TNF- and IL-1 decreased the secretion of apoB during each incubation period. The decrease in apoB secretion was modest yet significant at the end of the first 4.5 h and increased in magnitude during the following two incubation periods. The EGF receptor blocking antibody, which had no effect on the secretion of apoB itself, did not prevent the decrease in apoB secretion caused by TNF- or IL-1 during the first 4.5 h of incubation. During the second 4.5-h and the following 9-h incubation periods, however, mAb 528 significantly attenuated the decrease in apoB secretion by cells incubated with TNF- or IL-1. The blocking antibody was more effective in preventing the decrease in apoB secretion during the last 9 h of incubation than during the preceding 4.5 h. mAb 528 attenuated the inhibition of apoB secretion observed in cells incubated for 18 h with TNF-, IL-1, or EGF.

View larger version (34K): [in this window] [in a new window]   Fig. 5.   Effect of mAb 528 on cytokine mediated inhibition of apoB secretion. Cells were incubated with 100 ng/ml TNF- or 10 ng/ml IL-1 or BSA alone in the presence (closed bars) or absence (open bars) of 0.5 µg/ml mAb 528. The cytokines were added to the lower wells, and 250 µM oleic acid and 62.5 µM BSA were added to the upper wells. Cells were incubated for two consecutive 4.5-h periods followed by 9 h. Another set of cells was incubated for 18 h with either cytokine or 1.25 ng/ml EGF. At the end of each time point, the basal medium was collected and apoB mass estimated by sandwich enzyme-linked immunosorbent assay. The results are expressed as ng of apoB secreted/well, n = 6/treatment/time point. *p < 0.01 compared with cells incubated with BSA alone; **p < 0.01 compared with corresponding treatment without mAb.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The EGF receptor is a 170-kDa transmembrane glycoprotein comprised of an extracellular ligand binding domain, a single transmembrane hydrophobic region, and a highly conserved catalytic domain consisting of tyrosine kinase (33). Binding to the receptor by one of its ligands initiates receptor dimerization, activation of the intrinsic tyrosine kinase, and autophosphorylation of the receptor. Stimulation of receptor tyrosine kinase is essential for the transduction of signaling via the EGF receptor. We have demonstrated previously that IL-6, a potent inflammatory cytokine, inhibits apoB secretion from CaCo-2 cells by releasing a ligand to the EGF receptor (31). In this previous study, however, activation of the EGF receptor was not addressed directly. The results from the present study now clearly demonstrate that in intestinal epithelial cells, inflammatory cytokines interfere with lipoprotein secretion by activating the EGF receptor. By inhibiting EGF receptor tyrosine kinase activity with tyrphostin, we found that triacylglycerol-rich lipoprotein secretion was not altered by TNF-, and the effects of IL- were markedly attenuated. Both cytokines activated the EGF receptor by two distinct mechanisms. One required ligand binding to the receptor, and the other did not. Non-ligand-mediated activation of the EGF receptor by TNF- and IL- occurred early and was transient. In contrast, ligand-mediated activation of the EGF receptor occurred later and was more long lasting. The mechanism for inhibition of apoB secretion by TNF- or IL-1 reflected exactly the mechanisms for the activation of the EGF receptor by the cytokines. Moreover, neither TNF- nor IL-1 caused a further decrease in apoB secretion by cells incubated with EGF, suggesting that both cytokines were acting through the same pathway as EGF to inhibit apoB secretion. The results, therefore, clearly demonstrate that the inflammatory cytokines, TNF- and IL-1, inhibit the secretion of triacylglycerol-rich lipoproteins by the EGF receptor signaling pathway.

The EGF receptor is an allosteric protein that can be modulated by several agents that are not ligands to the receptor (36). Growth factors, stimulators of protein kinase C, and cytokines that bind to their own specific cell surface receptors alter the function of the EGF receptor within minutes of incubation. For instance, in human fibroblasts, TNF- and IL-1 were demonstrated to induce a rapid but transient increase in phosphorylation of the EGF receptor and a decrease in cell surface binding of EGF (32). This rapid modulation of the EGF receptor is believed to occur by a process that does not involve ligand binding to the receptor. Instead, it is postulated to involve phosphorylation of serine and/or threonine residues on the cytoplasmic domain of the EGF receptor by various intracellular kinases, such as protein kinase C, mitogen-activated protein kinase and calcium-calmodulin protein kinase (37). Phosphorylation at these residues has been shown to regulate receptor tyrosine kinase activity, phosphorylation of tyrosine residues on the receptor, and signal transduction through the receptor. Thus, in human fibroblasts, it was observed that TNF- and IL-1 phosphorylated the EGF receptor on threonine and serine residues (32). In contrast, in the present study, TNF- and IL-1 increased the phosphorylation of the EGF receptor on its tyrosine residues. However, because tyrphostin could not completely prevent the increase in phosphorylation of the EGF receptor in CaCo-2 cells incubated with TNF- or IL-1, it is likely that threonine and/or serine residues on the receptor were being phosphorylated as well. TNF- and IL-1 have been demonstrated to activate various cytoplasmic kinases such as mitogen-activated protein kinase (38, 39). It is possible, therefore, that through the action of one or more such kinases, CaCo-2 cell EGF receptor tyrosine phosphorylation and signal transduction were being regulated by a non-ligand-mediated process by TNF- and IL-1. In human fibroblasts, the rapid phosphorylation of the EGF receptor by TNF- or IL-1 was found to be independent of the activity of protein kinase C (32). Moreover, we also found that stimulating protein kinase C activity in CaCo-2 cells had no effect on apoB secretion (40). Thus, we suspect that protein kinase C has little or no role in the initial transient activation of the EGF receptor and inhibition in apoB secretion by TNF- or IL-1. We are currently examining whether TNF- or IL-1 phosphorylates other sites on the EGF receptor molecule and, if so, what kinases are involved. This line of investigation should provide insight on probable mechanisms by which the EGF receptor is activated in a ligand-independent manner.

Hydrolytic products of sphingomyelin, sphingosine, and ceramide serve as intracellular second messengers involved in cell growth and differentiation (41). They also modulate the activity of the EGF receptor (42, 43). TNF- and IL-1 have been shown to cause the hydrolysis of sphingomyelin in various cells (44). In CaCo-2 cells, TNF- and IL- cause rapid hydrolysis of sphingomyelin within minutes of incubation.² We have shown previously that incubation of CaCo-2 cells with sphingosine and analogs of ceramide results in a decrease in apoB secretion (40). It is possible, therefore, that early activation of the EGF receptor by TNF- and IL-1 is mediated by products of sphingomyelin hydrolysis. In fact, in A431 cells, sphingosine and ceramide increase EGF receptor phosphorylation within minutes (42, 43), similar to the rapid phosphorylation of the receptor we observed in CaCo-2 cells soon after adding TNF- and IL-1. Although most of the receptor phosphorylation in A431 cells was on a unique threonine residue, sphingosine also caused phosphorylation of tyrosine residue 1173 (42). In vitro, sphingosine activates the EGF receptor intrinsic tyrosine kinase, and it is postulated to do the same in intact cells (37). Moreover, sphingosine inhibits phosphorylation of the EGF receptor on threonine residue 654, which has been shown to decrease receptor tyrosine kinase activity (37). Thus, it is very possible that TNF- and IL- increase EGF receptor tyrosine phosphorylation by causing the release of sphingosine through hydrolysis of sphingomyelin. Not all of the reported observations on EGF receptor activation by sphingoid bases, however, are consistent with our present findings. For example, in contrast to the decrease in EGF binding to CaCo-2 cells incubated with TNF- or IL-1, in A431 cells, sphingosine increased the affinity of the ligand for the receptor and EGF receptor number (42). In Chinese hamster ovary cells, however, another sphingolipid, ganglioside G_M3, did not alter EGF binding (37). These results suggest that different species of sphingolipids may exert different effects on the EGF receptor, and furthermore, the effects may be dependent upon the cell type used. Whether sphingolipids play a role in the initial transient non-ligand-mediated phosphorylation of the EGF receptor in CaCo-2 cells incubated with TNF- or IL-1 is under investigation.

In contrast to the rapid and transient activation of the EGF receptor by TNF- or IL-, the later activation, which occurred after 6.5 h of incubation with the cytokines, was more long lasting and required ligand binding to the receptor. In a previous study, we found that IL-6 decreased apoB secretion by causing the release of EGF or an EGF-like molecule (31). In data not shown, however, we found that a neutralizing antibody to EGF did not prevent the inhibitory effects of TNF- or IL- on apoB secretion, suggesting that EGF was likely not the putative ligand. Other studies have demonstrated that TNF- induces the release of TGF- and have speculated that the growth-stimulatory effects of the cytokines are mediated by this ligand to the EGF receptor (45-47). Moreover, we ourselves have shown that TGF- decreases apoB secretion from CaCo-2 cells (31). It is possible, therefore, that TGF- is the ligand that mediates the inhibitory effects of the cytokines on apoB secretion.

Using fetal intestinal explants, Levy et al. (48, 49) demonstrated that EGF increased the secretion of apoB48-containing chylomicrons but inhibited the secretion of apoB100 in very low density lipoproteins. The conditions employed in this study, however, differed considerably from those used in our study. Fetal explants were incubated for 48 h with EGF at concentrations greater than 25 ng/ml. It is unlikely that the ligand or ligands induced by TNF- or IL- in CaCo-2 cells would have approached such levels. Moreover, in that study, contrary to the well recognized mitogenic effects of EGF in intestine, EGF inhibited protein synthesis in the fetal explants. In our study, cell proliferation was not altered after 18 h of incubation with TNF- or IL-1. This was not unexpected because it is known that DNA synthesis is delayed after EGF receptor activation (36). In contrast to what was observed in the intestine, in primary cultures of rat hepatocytes, Blake et al. (50) demonstrated that EGF decreased the secretion of apoB. Taken together, these studies and our present results strongly suggest that EGF receptor activation alters lipoprotein transport from both the intestine and liver.

The EGF receptor has a major role in repairing damaged mucosal surfaces after inflammatory injury (17). The results from this study demonstrate that the inflammatory cytokines TNF- and IL- activate the EGF receptor of CaCo-2 cells and, in so doing, cause a decrease in the transport of triacylglycerol-rich lipoproteins. It would make good sense that in the presence of inflammation and release of cytokines, small intestinal epithelial cells would divert their cell machinery and metabolism from nutrient transport to that of growth, restitution, and repair.

	ACKNOWLEDGEMENT

We thank Dr. Jheem Medh (University of Iowa, Iowa City) for providing assistance in iodinating EGF.

	FOOTNOTES

^* This work was supported by the Veterans Affairs and National Institutes of Health Grants HL49264 and 56032.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. of Internal Medicine and Veterans Affairs, University of Iowa, 200 Hawkins Dr., Iowa City, IA 52242. Tel.: 319-335-7073; Fax: 319-356-7893; E-mail: shubha-murthy@uiowa.edu.

² S. Murthy, S. N. Mathur, and F. J. Field, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: EGF, epidermal growth factor; TGF, transforming growth factor; TNF, tumor necrosis factor; IL, interleukin; mAb, monoclonal antibody; BSA, bovine serum albumin; PAGE, polyacrylamide gel electrophoresis; G_M3, N-acetylneuraminylgalactosylceramide.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. Chen, H. Xu, J. Dong, J. Li, and F. K. Ghishan Tumor necrosis factor-alpha impairs intestinal phosphate absorption in colitis Am J Physiol Gastrointest Liver Physiol, April 1, 2009; 296(4): G775 - G781. [Abstract] [Full Text] [PDF]

				S. Hevi and S. L. Chuck Ferritins Can Regulate the Secretion of Apolipoprotein B J. Biol. Chem., August 22, 2003; 278(34): 31924 - 31929. [Abstract] [Full Text] [PDF]

				R. M. Hobbs and F. M. Watt Regulation of Interleukin-1{alpha} Expression by Integrins and Epidermal Growth Factor Receptor in Keratinocytes from a Mouse Model of Inflammatory Skin Disease J. Biol. Chem., May 23, 2003; 278(22): 19798 - 19807. [Abstract] [Full Text] [PDF]

				K. Singh, O. A. Batuman, H. O. Akman, M. H. Kedees, V. Vakil, and M. M. Hussain Differential, Tissue-specific, Transcriptional Regulation of Apolipoprotein B Secretion by Transforming Growth Factor beta J. Biol. Chem., October 11, 2002; 277(42): 39515 - 39524. [Abstract] [Full Text] [PDF]

				R. Hemi, K. Paz, N. Wertheim, A. Karasik, Y. Zick, and H. Kanety Transactivation of ErbB2 and ErbB3 by Tumor Necrosis Factor-alpha and Anisomycin Leads to Impaired Insulin Signaling through Serine/Threonine Phosphorylation of IRS Proteins J. Biol. Chem., March 8, 2002; 277(11): 8961 - 8969. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Murthy, S. Articles by Field, F. J. Search for Related Content PubMed PubMed Citation Articles by Murthy, S. Articles by Field, F. J. Social Bookmarking                      What's this?