Epidermal growth factor induces p11 gene and protein expression and down-regulates calcium ionophore-induced arachidonic acid release in human epithelial cells.

p11, a member of the S-100 family of proteins, is the cellular ligand of annexin II and also interacts with the C-terminal region of cytosolic phospholipase A(2) (cPLA(2)), inhibiting cPLA(2) activity and arachidonic acid (AA) release. It has been reported that epidermal growth factor (EGF) induces cPLA(2) activation or cPLA(2) expression and subsequent AA release. It was of interest to study the effect of EGF on p11 production and on AA release in human epithelial cells (HeLa). EGF (20 ng/ml) treatment of HeLa cells increased the cellular p11 protein and the steady-state levels of p11 mRNA in a time- and dose-dependent manner but did not affect cPLA(2) protein expression over a 4-48-h incubation time. Transient transfection experiments of a reporter gene construct containing 1498 bp of the 5'-flanking region of p11 promoter demonstrated that EGF induced p11 gene expression at the transcriptional level. EGF caused a rapid phosphorylation of p44/42 and p38 kinases with a maximum level at 10 min. AG 1478 (EGF receptor tyrosine kinase inhibitor), PD 98059 (ERK1/2 inhibitor), and SB 203580 (p38 inhibitor) significantly inhibited EGF-induced p11 expression. EGF-induced AA release was significantly suppressed by AG 1478, PD 98059, SB 203580, and methyl arachidonyl fluorophosphate (a specific cPLA(2) inhibitor). Methyl arachidonyl fluorophosphate (50 microm) also significantly inhibited EGF-induced p11 expression, demonstrating that the activation of cPLA(2) may have a role in the EGF-induced p11 expression. Immunoprecipitation experiments showed that EGF induced increased p11 binding to cPLA(2) in a time- and dose-dependent manner. EGF treatment for 30 min increased -induced AA release, whereas EGF treatment for 24 h inhibited -induced AA release. These results suggest that EGF treatment increased p11 bound to cPLA(2) may lead to the late suppression of AA release induced by EGF.

The S-100 protein family is a multigenic family of low molecular mass (9 -11-kDa) calcium-binding proteins (1). S-100A10, known as p11 or calpactin I light chain, is a distinct member of the S-100 family, because its two EF-hands carry mutations that limit its ability to bind calcium (2,3). p11 is a natural ligand of annexin II, forming an annexin II 2 -p11 2 heterotetramer (AIIt) 1 (2,4,5), which may have a variety of functions in different cell types. AIIt regulates exocytosis and endocytosis by reorganization of F-actin. It also serves as a receptor to interact with tissue plasminogen activator, plasminogen, and procathepsin B (6,7). The C-terminal lysine residues of the p11 subunit of the AIIt bind plasminogen and participate in the stimulation of tissue plasminogen activator-dependent plasminogen activation. A recombinant p11 subunit of AIIt stimulates tissue plasminogen activatordependent plasminogen activation (8,9). These data suggest that p11, acting as a regulatory protein, modulates the activity of the annexin II subunit and stimulates protease activity by AIIt. p11 also interacts with the C terminus of cytosolic phospholipase A 2 (cPLA 2 ) and inhibits its activity, resulting in reduced arachidonic acid (AA) release (10). Antisense inhibition of p11 mRNA results in enhanced cPLA 2 activity and increased AA release, whereas p11 overexpression reduces cPLA 2 activity and AA release (11). Dexamethasone is known to reduce cPLA 2 activity, and recent studies suggest that this effect may be mediated by up-regulation of p11 (11). Kim et al. (12) have reported that annexin I and annexin II 2 -p11 2 , but not annexin II alone, inhibits cPLA 2 activity. Akiba et al. (13,14) have reported that transforming growth factor-␣ inhibited A23187induced AA release through increased p11 binding to cPLA 2 . These data suggest that p11 may play a role in inflammation by regulation of cPLA 2 activity and AA release. cPLA 2 , an 85-kDa enzyme, which has high selectivity to hydrolyze phospholipids containing AA esterified in the sn-2 position, has been implicated in receptor-mediated eicosanoid production and intracellular signal transduction processes (15)(16)(17)(18). cPLA 2 knock-out mice show markedly reduced production of prostaglandins, leukotrienes, and platelet-activating factor in peritoneal macrophages and in bone marrow-derived mast cells (19 -22). cPLA 2 is activated by physiologically relevant concentrations of calcium and by phosphorylation via mitogenactivated protein kinase ERK and p38 pathways (23)(24)(25)(26). Maximum activation of cPLA 2 also requires increased intracellular Ca 2ϩ concentrations, which induces translocation of cPLA 2 to cellular membranes. A variety of stimuli such as thrombin, platelet-derived growth factor, and fibroblast growth factors have been found to activate cPLA 2 phosphorylation through the MAP kinase ERK pathway (27). EGF has been reported to activate a rapid cPLA 2 phosphorylation mediated by the MAP kinase ERK. EGF is also reported to regulate cPLA 2 gene expression (28 -30). In this paper, we investigated the effect of EGF on p11 production, cPLA 2 activation, and AA release. We found that EGF induces p11 expression, which might contribute to the late suppression of calcium ionophore-induced AA release by EGF through increased p11 binding to cPLA 2 .
Cell Culture and Preparation of 3 H-Labeled Cells-HeLa cells (American Type Culture Collection, Manassas, VA) were grown in DMEM with 10% fetal bovine serum (FBS) at 37°C in 5% CO 2 in 175-cm 2 tissue culture flasks or six-well plates. Cells were seeded in six-well plates in DMEM with 10% fetal calf serum until 60% confluence and then incubated in the presence or absence of [ 3 H]AA (1 Ci/ml) for 16 h. The labeled cells were washed three times with DMEM without serum. The cells were incubated in 2 ml of DMEM with 10% fetal calf serum containing EGF at a final concentration for the following experiment. All experiments were performed when cells were 80 -90% confluent.
Experimental Design-For the time course experiments, after replacing culture medium at the same time, the cells were treated with or without 20 ng/ml EGF at 4, 12, 24, or 48 h prior to harvest at the 48-h time point. For the dose-response experiments, cells were treated with or without 0.2, 2, or 20 ng/ml EGF and harvested at 12 h. All inhibitors were preincubated for 2 h before treatment with or without EGF. Inhibitors were maintained for the incubation period.
Cellular Arachidonic Acid Release-The [ 3 H]AA-labeled HeLa cells were stimulated with EGF for 30 min or 24 h, washed, and placed in 2 ml of medium in the presence or absence of A23187 (10 Ϫ6 M) for 30 min. The 3 H-labeled HeLa cells were preincubated with MAFP (50 M) (cPLA 2 inhibitor), tyrphostin AG 1478 (10 M) (EGF receptor tyrosine kinase inhibitor), PD 98059 (25 M) (ERK inhibitor), or SB 203580 (10 M) (p38 pathway inhibitor) for 2 h, and the medium was replaced with fresh medium containing EGF for 30 min. The supernatants were collected, centrifuged at 750 ϫ g for 10 min, and counted in a scintillation counter (Beckman, Fullerton, CA) for quantification of AA release.
Immunoblot of p11 and cPLA 2 -HeLa cells grown in six-well plates were treated with or without EGF for the indicated time and dose. Cells were washed three times with cold PBS and lysed in 100 l of homogenization buffer: 50 mM Hepes (pH 8.0), 1 mM EDTA, 1 mM EGTA, 100 M leupeptin, 1 mM dithiothreitol, 10 mM phenylmethylsulfonyl fluoride, 0.5 mM soybean trypsin inhibitor, 15 mM aprotinin, and 0.25% Triton X-100. Cells were then sonicated three times for 15 s and centrifuged at 14,000 rpm for 10 min. Total protein was measured using the BCA method. Ten micrograms of crude cell lysate protein were separated on 16% Tris-glycine gels (Novex, San Diego, CA) for p11 and 8% Tris-glycine gels for cPLA 2 using Tris-glycine SDS running buffer. The separated proteins were electrophoretically transferred onto a nitrocellulose membrane (Novex), which was then blocked with 5% nonfat dry milk with 0.1% Tween 20 for 1 h. p11 or cPLA 2 protein expression was detected by immunoblotting with a 1:2000 dilution of mouse-antihuman annexin II light chain monoclonal antibody (anti-p11 antibody) or sheep anti-human cPLA 2 and a 1:5000 dilution of horseradish peroxidase-conjugated donkey anti-mouse IgG or goat anti-sheep IgG as the second antibody. The blot was developed using the ECL Western blotting detection system and exposed to Eastman Kodak Co. MR radiographic film.
Ribonuclease Protection Assay (RPA)-HeLa cells grown in T-175 cm 2 flasks were treated with and without EGF (0.2, 2, and 20 ng/ml) for 4 -48 h. The cells were washed three times with cold PBS and then FIG. 1. EGF induces an increase in p11 protein levels. Cells treated with or without EGF were lysed, and 10 g of crude cell lysates were studied by Western blot analysis using anti-annexin II light chain monoclonal antibody. A, time course (4 -48 h) effect of EGF (20 ng/ml) on HeLa cells showing increased p11 protein levels, with peak levels at 12 h. EGF treatment did not effect cPLA 2 protein expression. The result shown is representative of three separate experiments. B, densitometry measurements from three time course experiments of EGF (20 ng/ml) and control cells demonstrating increased p11 protein levels (data presented as mean Ϯ S.E., *, p Ͻ 0.05 compared with control value). C, EGF (0.2, 2, and 20 ng/ml) treatment for 12 h induces a dose-dependent increase in p11 protein level. The result shown is representative of three separate experiments. D, densitometry measurements from three dose-response (0 -20 ng/ml) experiments demonstrating a dose-related increase in p11 protein levels (data presented as mean Ϯ S.E., p Ͻ 0.001 for dose-related effect by single factor analysis of variance). collected by centrifugation for 10 min at 1000 ϫ g at 4°C. Total cellular RNA was extracted using the RNAqueous kit from Ambion (Austin, TX) according to the manufacturer's instructions. RNA was quantified using 260-nm optical density. To construct the probe for p11 mRNA, a 319-bp product of p11 cDNA was amplified by PCR using the following sets of sense and antisense primers: 5Ј primer, 5Ј-ACCACACCAAAAT-GCCATCT-3Ј (corresponding to bases 61-80 of the human p11 cDNA sequence; GenBank TM accession number M81457); 3Ј primer, 5Ј-CT-GCTCATTTCTGCCTACTT-3Ј (corresponding to bases 361-379 of the p11 cDNA sequence). The product was cloned into the pGEM-T Easy vector (Promega, Madison, WI). Orientation of the insert was determined by DNA sequencing. The p11 cRNA probe and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Ambion, Austin, TX) were radiolabeled using an in vitro transcription kit (Ambion) with SP6 polymerase with [␣-32 P]UTP (800 Ci (29.6 TBq)/mmol) (PerkinElmer Life Sciences). A p11 sense riboprobe was radiolabeled using in vitro transcription with T7 polymerase. An RPA assay kit (RPAII; Ambion) was used to quantitate target mRNA. Ten micrograms of total RNA (for GAPDH) or 50 g (for p11) were mixed with 10,000 cpm (for GAPDH) or 20,000 cpm (for p11) of ␣-32 P-labeled riboprobe. The mixture was hybridized at 45°C overnight and digested by the addition of a 1:100 dilution of RNase A/T1 at 37°C for 60 min. Digestion was terminated by the addition of the RNase inactivation and precipitation mixture. The protected fragments were separated on 6% polyacrylamide 8 M urea gels (Novex) and visualized by autoradiography.
Transient Transfection Assay-HeLa cells were maintained at 37°C under 5% CO 2 in DMEM with 10% FBS. The day before transfection, 5 ϫ 10 5 cells were seeded into six-well plates. 1.8 g of p11 promoter constructs were co-transfected with 0.1 g of the pCMV/␤-galactosidase construct (CLONTECH, Palo Alto, CA) into serum-free cells using 6 l of LipofectAMINE reagent (Invitrogen) in each well. After 5 h of transfection, cells were incubated in fresh DMEM medium with 10% FBS for 36 h. The transfected cells were stimulated with EGF (20 ng/ml) for the indicated times and lysed. pCAT and ␤-galactosidase activity was measured with pCAT and ␤-galactosidase enzyme-linked immunosorbent assay kits (Roche Molecular Biochemicals). The pCAT activity was normalized to ␤-galactosidase activity to represent relative p11 promoter activity.
Immunoblot Analysis of Phosphorylation of p44/42, p38, and cPLA 2 -HeLa cells cultured in six-well plates were treated with or without EGF for 5 min, 10 min, 30 min, 1 h, or 2 h. The cells were washed three times with cold PBS, and the medium was aspirated completely. 100 l of 1ϫ protein loading buffer was added into each well and immediately scraped and collected into microcentrifuge tubes. After sonication for 20 s, the samples were boiled for 10 min and cooled on ice. 20 l of each sample was separated on a 4 -20% Tris-glycine gel, electrophoretically transferred onto a nitrocellulose membrane, and blocked with 5% nonfat milk. The blots were then probed with a 1:2000 dilution of anti-p44/42, anti-p38 or phosphospecific anti-p44/42, anti-p38, or anti-phosphoserine-cPLA 2 antibody for 2 h. After three washes, the blots were probed with a 1:2000 dilution of second antibody and developed by using the ECL Western blotting detection system.
Immunoprecipitation of p11 and cPLA 2 Protein-HeLa cells grown in 175-cm 2 tissue culture flasks were washed three times with cold PBS

FIG. 2. EGF treatment increases p11 steady state mRNA levels.
Cells were incubated in the presence or absence of EGF at the indicated time and dose, followed by extraction of total RNA. Ten and fifty micrograms of total cellular RNA were hybridized to GAPDH and p11-specific radiolabeled cRNA, respectively, and subjected to RPA. The protected fragments of p11 (319 bp) and GAPDH were visualized by autoradiography. A, ribonuclease protection assay. Left panel, total cellular RNA was incubated with a p11 antisense or sense riboprobe. and lysed in 0.5 ml of homogenization buffer as described above. 500 g of crude cell lysate was added to 200 l of immunoprecipitation buffer (150 mM NaCl, 50 mM Tris-HCl, 0.05% Nonidet P-40) containing 10 l of sheep anti-human cPLA 2 antibody and incubated at 4°C on a rotor at medium speed for 4 h. Thirty microliters of protein G-agarose (Pierce) beads were then added to each sample, and the mixture was incubated at 4°C for 2 h, followed by centrifugation in a microcentrifuge at 2000 rpm for 5 min at 4°C. The supernatants were aspirated, and the pellet was washed four times with immunoprecipitation buffer. The pellet was then suspended in 20 l of protein loading buffer and boiled for 10 min before electrophoresis on 16% Tris-glycine gels for p11 and 8% Trisglycine gels for cPLA 2 . The separated proteins were electrophoretically transferred onto a nitrocellulose membrane blocked with 5% nonfat milk and then probed with a 1:2000 dilution of anti-p11 antibody or 1:500 dilution of anti-phosphoserine antibody for 2 h. After three washings, the blots were incubated with a 1:2000 dilution of second antibody and developed by using the ECL Western blotting detection system.
Transfection of HeLa Cells with iRNAs-iRNAs were prepared by IDT (Coralville, IA) and targeted the coding region 4 -24, relative to the start codon of the cPLA 2 gene. Single-stranded RNAs were annealed by incubating a 20 M concentration of each strand in annealing buffer (100 mM potassium acetate, 30 mM HEPES buffer at pH 7.4, 2 mM magnesium acetate) for 2 min at 90°C. The iRNA sequences used in this study are 5Ј-GUAAGGAUCUAUAAAUGATT-3Ј and 5Ј-UCAUUU-AUAGAUCCUUACTT-3Ј.
HeLa cells were grown as described above and were transfected with 10 nM iRNA duplexes using LipofectAMINE reagent (Invitrogen). Four hours after transfection, media were changed, and cells were incubated with or without [ 3 H]AA in DMEM with 10% of FBS for 12 h, and an AA release assay was performed as described above. Unlabeled cells were incubated with EGF (20 ng/ml) or with media. After 12 h of incubation with or without EGF, cPLA 2 and p11 proteins were detected using immunoblotting as described above.
Quantification of Autoradiographs-An Amersham Biosciences 301 computing densitometer was used to digitize images. The optical density of bands was analyzed with background subtraction using Image-Quant software (Amersham Biosciences).
Statistical Analysis-The dose-related effects were analyzed with one-way analysis of variance, and comparisons were performed using two-tailed unpaired Student's t tests. Values of p Ͻ 0.05 were considered statistically significant.

EGF Increases the Level of the p11 Protein in Human
Epithelial Cells-The effect of EGF on p11 expression was studied. HeLa cells were treated with or without EGF, and cellular p11 protein levels were assessed by Western blot analysis. Fig. 1, A and B, demonstrates that p11 is constitutively expressed in untreated HeLa cells. EGF (20 ng/ml) increased p11 protein expression over a 4 -24-h period, with a maximum effect observed at 12 h. However, over the incubation period, EGF had no effect on cPLA 2 protein expression. Treatment of cells with 0.2, 2, or 20 ng/ml EGF for 12 h resulted in a dose-dependent increase in p11 protein levels ( Fig. 1, C and D).
The Effect of EGF on p11 Steady-state mRNA Levels in HeLa Cells-To determine whether the observed increase in p11 protein levels correlates with p11 mRNA expression, HeLa cells were incubated with or without EGF (20 ng/ml) for 4 -48 h or EGF (0.2, 2, or 20 ng/ml) for 12 h. Total RNA was isolated, and the steady-state p11 mRNA expression was studied by an RPA. The specificity of the p11 antisense riboprobe is presented in Fig. 2A. The antisense riboprobe produced a protected band at ϳ300 bases. A sense riboprobe produces no protected band. Cellular RNA incubated with a GAPDH riboprobe is presented in the right panel of Fig. 2A. As shown in Fig. 2, B and C, EGF treatment stimulated an increase in p11 steady-state mRNA levels over 4 -12 h, compared with the untreated control cells. Cells treated with 0.2, 2, or 20 ng/ml EGF for 12 h demonstrated a dose-related increase in the steady state p11 mRNA levels in HeLa cells as shown in Fig. 2, D and E. GAPDH mRNA levels are presented as an internal control for equivalent RNA loading and normalization of p11 mRNA expression.
EGF Induces p11 Expression at the Transcriptional Level in HeLa Cells-To further investigate whether the observed increase in steady-state p11 mRNA levels reflects an EGF-induced increase in p11 gene transcription, a reporter gene construct containing the 1498-bp sequence of the 5Ј-flanking region of the p11 promoter in the pCAT vector was transfected into HeLa cells. The results are shown in Fig. 3. EGF (20 ng/ml) treatment of HeLa cells induced increased p11 gene transcriptional expression over 2-8 h. ␤-Galactosidase activity was used to correct for transfection efficiency. Thus, EGF induces p11 expression, at least in part, at the transcriptional level. EGF Induces p44/42 and p38 Phosphorylation-EGF exerts its effect by binding and activating a specific 170-kDa tyrosine kinase receptor and triggers a subsequent MAP kinase signal FIG. 4. EGF induces p44/42 and p38 phosphorylation. HeLa cells were grown in six-well plates in DMEM with 10% FBS until 80% confluent and treated with EGF for the indicated times. Cells were washed three times with cold PBS and lysed in 100 l of SDS loading buffer. Cell lysates were sonicated for 15 s, boiled for 5 min, and cooled on ice. Equal amounts of cell lysate were subjected to electrophoresis 4 -20% Tris-glycine gels. The Western blot was performed by immunoblotting with rabbit anti-human p44/42, p38, and specific anti-phospho-p44/42 and anti-phospho-p38. A, EGF induced p44/42 phosphorylation with a maximum phosphorylation at 5 and 10 min. The total amount of p44/42 protein was unchanged. B, EGF induced a rapid p38 phosphorylation starting at 5 min and with a maximum phosphorylation at 10 min. EGF did not affect the total amount of p38 protein.
FIG. 3. EGF regulates p11 gene expression at the transcriptional level. HeLa cells were transfected with the reporter gene construct containing 1498 bp of the p11 5Ј promoter region and with the pCMV/␤-gal plasmid as a transfection control. After 36-h transfection, the cells were treated with or without EGF for the indicated time. Cells were lysed, and pCAT and ␤-galactosidase activity were assayed. The relative p11 promoter activity was determined by the ratio of pCAT and ␤-galactosidase activities. EGF induced a significantly increased p11 reporter gene activity compared with control (data presented as mean Ϯ S.E., n ϭ 12-15; *, p Ͻ 0.05 for EGF treated for 2 h compared with control; **, p Ͻ 0.001 for EGF treated for 4 h compared with control). transduction cascade. As shown in Fig. 4A, EGF stimulated the phosphorylation of p44/42 in a time-dependent manner with a maximum activation at 5 and 10 min. To examine whether EGF also stimulated p38 phosphorylation, the same cell lysate was used to immunoblot with phosphospecific anti-pp38 antibody. EGF-induced p38 phosphorylation was also transient, starting at 5 min and with a maximum at 10 min (Fig. 4B).
The Effect of AG 1478, PD 98059, and SB 203580 on EGFinduced p11 Expression-To examine whether EGF-induced p11 expression is mediated via the activation of EGF receptor and mitogen-activated protein kinase pathway, the effect of tyrphostin AG 1478, an EGF receptor tyrosine kinase inhibitor, on EGF-induced p11 expression was studied. As shown in Fig.  5, A and B, AG 1478 significantly inhibited EGF-induced p11 expression, whereas AG 1478 alone did not affect p11 expression. These data demonstrated that EGF-induced p11 expression is mediated via the activation of the EGF receptor. To further explore whether EGF-activated MAP kinase affects p11 expression, the ERK inhibitor, PD 98059, and p38 inhibitor, SB 203580, were used to study the effect of EGF on p11 expression.
After pretreatment with PD 98059 or SB 203580 for 2 h, HeLa cells were incubated with EGF for 12 h. As shown in Fig. 5, C-F, PD 98059 and SB 203580 significantly inhibited EGFinduced p11 expression. Cells treated with inhibitors alone exhibited no effect. These data suggest that both the ERK pathway and the p38 pathway are involved in the induction of p11 by EGF.
The Effect of EGF on cPLA 2 Phosphorylation-Phosphorylation of cPLA 2 was studied by Western blot and immunoprecipitation. HeLa cells were treated with EGF (20 ng/ml) for 5 min, 10 min, 30 min, 1 h, and 2 h, and the crude cell lysates were immunoblotted with phosphoserine-specific anti-cPLA 2 (Ser-505). As shown in Fig. 6A, EGF induced cPLA 2 phosphorylation over a 2-h incubation period. These results were further confirmed by immunoprecipitation, using anti-cPLA 2 antibody to immunoprecipitate cPLA 2 , and then immunoblotting with antiphosphoserine antibody. The results are shown in Fig. 6B. EGF induced maximum cPLA 2 phosphorylation at 10 -30 min. To determine whether EGF-induced phosphorylation of cPLA 2 is mediated via activation of one or both of the above mentioned MAP kinase pathways, cells were preincubated with PD 98059 FIG. 5. Inhibitors of the EGF receptor-MAP kinase pathway block the effect of EGF on p11 protein production. HeLa cells grown in DMEM with 10% FBS were preincubated with or without the different inhibitors for 2 h and then treated with or without 20 ng/ml EGF for 12 h. Ten micrograms of crude cell lysate was separated on a 16% Tris-glycine gel. p11 protein expression was performed by immunoblotting with mouse anti-human annexin II light chain antibody. A, Western blot showing that tyrphostin AG 1478 significantly inhibited EGF-induced p11 expression; AG 1478 alone had a little or no effect on p11 expression. B, densitometry measurements from three experiments demonstrating that AG 1478 pretreatment significantly inhibited EGF-induced p11 protein production (data presented as mean Ϯ S.E.; *, p Ͻ 0.05 for EGF-treated cells versus EGF plus AG 1478-treated cells). C, Western blot showing that PD 98059 significantly inhibited EGF-induced p11 production. D, densitometry measurements from eight experiments demonstrating that PD 98059 inhibited EGF-induced p11 protein production in HeLa cells (data presented as mean Ϯ S.E.; *, p Ͻ 0.05 for EGF-treated cells versus EGF plus PD 98059-treated cells). E, Western blot showing that SB 203580 inhibited EGF-induced p11 production, whereas SB 203580 had little or no effect on p11 production. F, densitometry measurement from five different experiments demonstrating that SB 203580 suppressed EGF induced p11 production (data presented as mean Ϯ S.E.; *, p Ͻ 0.05 for EGF-treated cells versus EGF plus SB 203580-treated cells). (25 M) or SB 203580 (10 M) for 2 h followed by treatment with EGF (20 ng/ml). Fig. 6, C and D, demonstrates that EGF treatment induces cPLA 2 phosphorylation and that pretreatment with PD 98059 partially inhibits that phosphorylation. Similarly, Fig. 6, E and F, demonstrate that pretreatment with the p38 pathway inhibitor, SB 203580, also partially inhibits EGF-induced phosphorylation of cPLA 2 , suggesting that EGF may signal through both pathways to induce phosphorylation of cPLA 2 .
EGF Induces AA Release through the Activation of EGF Receptor and the MAP Kinase Pathway-EGF initiates intracellular signaling by binding to the EGF receptor. To explore whether HeLa cells express the EGF receptor and whether EGF stimulation altered the amount of EGF receptor, HeLa cells were treated with or without EGF (20 ng/ml) for 12 h. Total cell lysates were immunoblotted with anti-EGF receptor antibody. As shown in Fig. 7A, HeLa cells express the EGF receptor, and EGF treatment did not affect EGF receptor expression. To demonstrate that EGF-induced AA release is mediated through EGF receptor activation, the effect of tyrphostin AG 1478 on AA release was studied. The results in Fig. 7B show that AG 1478 (10 M) significantly inhibited EGF-induced AA release and did not affect basal release of AA. To confirm that EGF-induced MAP kinase activity directly mediated AA release, PD 98059, a p44/42 kinase inhibitor, and SB 203580, an inhibitor of the p38 pathway, were used to study EGFstimulated AA release. As shown in Fig. 7C, PD 98059 inhibited EGF-induced AA release, whereas PD 98059 alone did not affect unstimulated AA release. In addition, pretreatment with SB 203580 partially inhibited EGF-induced AA release, suggesting activation of the p38 pathway as well (Fig. 7D). Pretreatment with both inhibitors blocked EGF-induced AA release (Fig. 7E). Thus, these data demonstrate that EGF mediated the activation of the EGF receptor and MAP kinases, resulting in the cellular release of AA. cPLA 2 -mediated EGF-induced AA Release-EGF stimulated the rapid phosphorylation of cPLA 2 and subsequent AA release. To confirm that EGF-induced AA release in HeLa cells is mediated by cPLA 2 , the effect of MAFP, a cPLA 2 inhibitor, on AA release was studied. As shown in Fig. 8, EGF-induced AA release was significantly inhibited by MAFP (50 M). These data suggest that EGF-induced AA release is mediated by cPLA 2 .
MAFP Inhibits EGF-induced p11 Expression in HeLa Cells-EGF activates cPLA 2 and induces AA release. Therefore, it was interest to investigate the effect of activation of cPLA 2 on EGF-induced p11 production. The effect of MAFP on EGFinduced p11 protein expression was studied. The HeLa cells were preincubated with MAFP (50 M) for 2 h and stimulated with EGF (20 ng/ml) for 12 h. The total cell lysates were then analyzed by Western blot. The data are shown in Fig. 9. MAFP pretreatment inhibited EGF-induced p11 expression, whereas MAFP alone had little effect on p11 expression. These data demonstrated that the activation of cPLA 2 may mediate the induction of p11 expression by EGF.

FIG. 6. EGF induces cPLA 2 phosphorylation. A, Western blot
showing that EGF induces a rapid phosphorylation of cPLA 2 . HeLa cells grown in six-well plate were treated with or without EGF (20 ng/ml) for the indicated time and washed three times with cold PBS. The cells were lysed in 100 l of 1ϫ SDS loading buffer. Equal amounts of cell lysate were separated to 8% Tris-glycine gels. The Western blot was performed, and the membrane was probed with rabbit anti-phosphoserine cPLA 2 antibody. Total cPLA 2 protein was unchanged. B, immunoprecipitation experiment demonstrating that EGF induced an increased cPLA 2 phosphorylation and did not affect the total amount of cPLA 2 . HeLa cells were grown in T-175 cm 2 flasks and treated with or without EGF (20 ng/ml) for the indicated time. After washing with cold PBS, the cells were lysed in the lysate buffer described under "Experimental Procedures." 200 l (500 g) of cell lysates were incubated with anti-cPLA 2 antibody for 4 h and then with 20 l of protein G-agarose for 2 h. After repeated washing with immunoprecipitation buffer, 20 l of protein loading buffer was added to suspend the agarose beads, and the beads were boiled for 5 min and then subjected to electrophoretic separation on an 8% Tris-glycine gel. Precipitated cPLA 2 was immunoblotted with anti-cPLA 2 antibody or anti-phosphoserine antibody. C, Western blot demonstrating the effect of PD 98059 on EGF-induced phosphorylation of cPLA 2 . Cells were treated with or without PD 98059 (25 M) for 2 h followed by treatment with or without EGF (20 ng/ml). D, densitometry measurements from four experiments demonstrating that PD 98059 inhibited EGF-induced cPLA 2 phosphorylation in HeLa cells (data presented as mean Ϯ S.E.; *, p Ͻ 0.05 for EGF-treated cells versus EGF plus PD 98059-treated cells). E, Western blot showing that SB 203580 inhibited EGF-induced cPLA 2 phosphorylation. Cells were treated with or without SB 203580 (10 M) for 2 h followed by treatment with or without EGF (20 ng/ml). F, densitometry measurement from five different experiments demonstrating that SB 203580 inhibited EGF-induced cPLA 2 phosphorylation (data presented as mean Ϯ S.E.; *, p Ͻ 0.05 for EGF-treated cells versus EGF plus SB 203580-treated cells).

Effect of cPLA 2 -inhibitory RNAs on Arachidonic Acid Release and p11
Production in Response to EGF-Anti-cPLA 2 iRNAs were utilized in order to further assess the role of cPLA 2 in EGF-induced arachidonic acid release and in p11 production. Cells treated with iRNAs exhibited reduced EGF-induced arachidonate release compared with control cells treated with LipofectAMINE alone (Fig. 10A). Western blot of cell lysates from cells treated with iRNAs also demonstrated a reduction in cellular cPLA 2 protein (Fig. 10B). Finally, treatment of cells with iRNAs resulted in a diminished p11 production in response to EGF compared with cells treated with the transfection reagent alone (Fig. 10C).
EGF Increases Native p11 Bound to cPLA 2 -To study the effect of EGF-induced p11 expression on cPLA 2 , immunoprecipitation complex from HeLa cells was studied. As shown in Fig. 11, A and B, treatment with 20 ng/ml EGF for 4 -24 h induced an increase in p11 co-immunoprecipitated with cPLA 2 in a time-dependent manner with a maximum effect at 12-24 h. There also was a dose-dependent effect of EGF treatment on co-immunoprecipitated p11-cPLA 2 (Fig. 11, C and D). These an EGF receptor tyrosine kinase inhibitor, for 2 h. The cells were washed and stimulated with or without EGF (20 ng/ml) for 30 min. An aliquot of medium was used to measure AA release by scintillation counter. Data are presented as mean Ϯ S.E., n ϭ 6, results demonstrate that EGF treatment results not only in an increase of cellular p11 expression but also in an increase of p11 bound to cPLA 2 .
The Effect of EGF on the Cellular Arachidonic Acid Release-To study whether increased binding between p11 and cPLA 2 could affect AA release, [ 3 H]AA-labeled HeLa cells were treated with or without EGF (20 ng/ml) for 30 min or 24 h. After EGF treatment, medium was changed, cells were treated with or without A23187 for 30 min, and medium was harvested for scintillation counting. The results shown in Fig. 12, A and B, indicate that EGF treatment for 30 min increased both basal and A23187-induced AA release. Interestingly, EGF treatment for 24 h diminished A23187-induced AA release. These data suggest that the inhibitory effect of EGF on AA release at 24 h may be due to EGF-induced p11 expression, which may inhibit cPLA 2 activity and reduce AA release. DISCUSSION p11, or calpatin light chain, is a unique member of the S-100 family of calcium-binding proteins. Although it shares sequence homology with the S100 family, it does not have the ability to bind Ca 2ϩ due to amino acid deletions and substitutions in the two EF-hand motifs. p11 is present in a variety of cells separately or as a heterotetramer with annexin II (AIIt). The expression of p11 and annexin II is not always coordinated, and the ratio of p11 to annexin II varies with different cell types. Munz et al. (31) reported that wound-derived growth factors (transforming growth factor-␤ 1 , EGF, and keratinocyte growth factor) differentially regulate p11 and annexin II expression in cultured keratinocytes during skin injury and modify the ratio between p11 and annexin II. Annexin II tetramer also can act as a surface protein receptor for plasminogen and t-PA.
Studies in cPLA 2 -deficient mice demonstrate that cPLA 2 plays an essential role in eicosanoid production and allergic response (19). It was important to investigate the negative regulation of cPLA 2 in physiological conditions. Annexin I, or lipocortin I, has been reported to suppress cPLA 2 activity not only in vitro but also in cultured cells. Thus, annexin I may function as an endogenous negative regulator of cPLA 2 . p11 interacts with and inhibits cPLA 2 activity and subsequent AA release in vitro and in vivo (10). Therefore, p11 may also function as a negative regulator of cPLA 2 . p11 has been reported to be regulated in different cell types. Nerve growth factor increases p11 mRNA expression in rat pheochromocytoma (PC12) cells (32). Retinoic acid reduces p11 protein levels by a post-translational mechanism in BEAS-2B cells (33). Transforming growth factor-␣ or the combination of transforming growth factor-␣ and interleukin-1␤ induces an increase in p11 protein expression in rat gastric epithelial cells (14). Nitric oxide induces p11 expression through a cGMP-dependent pathway in epithelial cells (34). In this study, we show that EGF induces p11 production in a time-and dose-dependent manner in human epithelial cells (HeLa cells). The effect was correlated with increased steady-state levels of p11 mRNA in response to EGF in a time-and dose-related manner. This effect of EGF on p11 expression was regulated at least in part at the transcriptional level. EGF has been reported to induce cPLA 2 expression at the transcriptional level (28,29). However, in our experiments, over an incubation up to 48 h, EGF did not appear to affect cPLA 2 expression, indicating that EGF induced cPLA 2 expression may be cell type-dependent.
The binding of EGF to the EGF receptor on the cell surface FIG. 9. MAFP inhibition of EGF-induced p11 protein expression. HeLa cells grown in T-75 cm 2 flasks were pretreated with MAFP (50 M) for 2 h, and EGF (20 ng/ml) was added for an additional 12 h. Cells were lysed, 10 g of crude cell lysate was subjected to 16% Tris-glycine gel, and Western blot analysis was performed. A, Western blot demonstrating that MAFP inhibited EGF-induced p11 production. B, densitometry measurement from three different experiments demonstrating that MAFP significantly inhibited EGF-induced p11 protein expression, whereas MAFP alone had no or little effect on p11 expression in HeLa cells. Data are presented as mean Ϯ S.E. *, p Ͻ 0.05 for EGF-treated cells versus EGF plus MAFP-treated cells.
FIG. 10. Effect of cPLA 2 inhibitory RNAs on arachidonate release, cPLA 2 , and p11 in response to EGF stimulation. A, arachidonate release from cells treated with cPLA 2 iRNAs. Cells were transfected with cPLA 2 iRNA (10 nM) or treated with LipofectAMINE (control) as described under "Experimental Procedures." Then cells were labeled with [ 3 H]AA (1 Ci/ml) for 16 h. Medium was changed, cells were incubated with or without EGF (20 ng/ml) for 30 min, and AA release was performed as described under "Experimental Procedures." n ϭ 3; *, p Ͻ 0.05 as compared with control cells. B, Western blot from cells treated with cPLA 2 iRNAs. Cells were transfected with cPLA 2 iRNA (10 nM) or treated with LipofectAMINE as described under "Experimental Procedures." Sixteen hours after transfection, cells were treated for 12 h with EGF (20 ng/ml) and processed as described under "Experimental Procedures." The immunoblot was developed using anti-cPLA 2 antibody. The blot shown is representative of three experiments, each with similar results. C, Western blot from cells treated with cPLA 2 iRNAs. Cells were transfected with cPLA 2 iRNA (10 nM) or treated with LipofectAMINE as described under "Experimental Procedures." Sixteen hours after transfection, cells were treated for 12 h with EGF (20 ng/ml) and processed as described under "Experimental Procedures." The immunoblot was developed using anti-p11 antibody. The blot shown is representative of three experiments, each with similar results. triggers receptor trans-autophosphorylation and subsequent activation of the Ras/Raf/mitogen-activated protein kinase cascade. Several experiments were done to explore the signal pathway involved in the induction of EGF-induced p11 production. First, we observed that EGF induces rapid p44/42 and p38 phosphorylation starting at 5 min and with a maximum activation at 10 min. Second, the EGF receptor tyrosine kinase inhibitor, AG 1478, significantly inhibited EGF-induced p11 expression, indicating the involvement of the activation of EGF receptor. Third, PD 98059 and SB 203580 inhibited EGFinduced p11 production, demonstrating that both MAP kinase ERK and p38 participated in this event. Thus, the addition of EGF activates the EGF receptor tyrosine kinase and subsequent MAP kinase ERK and p38 pathway, resulting in the induction of p11.
Previous work has shown that EGF stimulates AA release through the phosphorylation of cPLA 2 . In our studies, we observed similar results with two sets of experiments. First both Western blot and immunoprecipitation experiments showed that EGF treatment resulted in phosphorylation of cPLA 2 , FIG. 11. EGF increases p11 bound to cPLA 2 in a time-dependent and dose-dependent manner. A, immunoprecipitation experiments showing a time-dependent increase of the p11-cPLA 2 complex. After HeLa cells were treated with EGF (20 ng/ml) for 4, 12, or 24 h, cell lysate (500 g) was incubated with anti-human cPLA 2 antibody for 4 h followed by incubation with protein G-agarose beads for an additional 2 h. The beads were collected by centrifugation at 500 ϫ g, washed five times with immunoprecipitation buffer, and separated on 16% Trisglycine gel. The coprecipitated p11 protein was then detected by immunoblotting with mouse anti-human annexin II light chain monoclonal antibody. The blot presented is representative of three different experiments. B, densitometry measurement from three experiments (data presented as mean Ϯ S.E.; *, p Ͻ 0.05 compared with control value). C, EGF induces increased p11 bound to cPLA 2 in a dose-dependent manner. Cells were incubated with or without EGF (0.2, 2, or 20 ng/ml) for 12 h. Cell lysates were incubated with anti-cPLA 2 antibody for 4 h followed by incubation with protein G-agarose beads for 2 h. The coprecipitated p11 was detected by immunoblotting with mouse antihuman annexin II light chain monoclonal antibody. The blot is representative of three different experiments. D, densitometry measurement from three dose-response experiments. (Data are presented as mean Ϯ S.E., p Ͻ 0.001 for dose-related effect by single-factor analysis of variance).

FIG. 12. Effect of EGF on A23187-induced arachidonic acid release in HeLa cells.
A, EGF increases A23187-induced AA release at 30 min. [ 3 H]AA (1 Ci/ml)-labeled HeLa cells were incubated with or without EGF (20 ng/ml) for 30 min, the medium was changed, and the cells were then treated with or without 10 Ϫ6 M A23187 for 30 min. The supernatants were collected and centrifuged. AA release was measured by a scintillation counter. Data are presented as AA release from one (n ϭ 6) of three experiments with similar results. *, p Ͻ 0.001 for A23187-induced AA release compared with non-A23187-stimulated cells treated with or without EGF. B, EGF inhibits A23187-induced AA release at 24 h. [ 3 H]AA (1 Ci/ml)-labeled HeLa cells were incubated with or without EGF (20 ng/ml) for 24 h. The medium was changed, cells were treated with or without 10 Ϫ6 M A23187 for 30 min, and the supernatants were collected and centrifuged. AA release into the medium was measured by a scintillation counter. Data are presented as mean Ϯ S.E. (n ϭ 6). [ 3 H]AA release from one of three separate experiments each with similar results. *, p Ͻ 0.001 for A23187-induced AA compared with control value. starting at 5 min and with a maximum activation at 10 -30 min. Second, cells pretreated with MAFP significantly inhibited EGF-induced AA release, suggesting that the activation of cPLA 2 is responsible for EGF-induced AA release. In addition, AG 1478, PD 98059, and SB 203580 significantly inhibited EGF-induced AA release, indicating that the activation of cPLA 2 by EGF is through the activation of EGF receptor and subsequent MAP kinase pathways. To further study whether these events of the activation of cPLA 2 and AA release could influence p11 production, we showed that MAFP pretreatment significantly suppressed EGF-induced p11 expression. These data suggest that the activation of cPLA 2 may play a role in the induction of p11 by EGF. We postulate that EGF induced p11 expression through the activation of EGF receptor, MAP kinase, and cPLA 2 . Whether EGF-induced AA or AA metabolites contribute to the p11 expression by EGF requires further study.
To better understand the temporal influence of EGF-induced p11 expression on cPLA 2 activity and the cellular AA release, we demonstrated that EGF increased p11 co-immunoprecipitated with cPLA 2 in a time-and dose-dependent manner. Further, the treatment of HeLa cells with EGF for 30 min increased A23187-induced AA release, whereas treatment with EGF for 24 h significantly diminished A23187-induced AA release. This observation led us to postulate that an inhibitory effect of EGF on A23187-induced AA release at 24 h might be due to an EGF-induced increase in p11, which binds to and inhibits cPLA 2 activity and subsequent AA release. Recently, it has been reported that RGM1 cells treated with transforming growth factor-␣ resulted in an inhibitory effect of A23187induced AA release at 3-24 h, which correlated with increased p11 binding to cPLA 2 (13,14). Our results were similar to those reported by Akiba et al. (13,14).
In conclusion, we have demonstrated that EGF induces p11 production through EGF activation of the EGF receptor tyrosine kinase and activation of p44/42, p38, and cPLA 2 .