Upregulation of Phospholipase Cγ1 Activity during EGF-Induced Proliferation of Corneal Epithelial Cells: Effect of Phosphoinositide-3 Kinase

Abstract

purpose. Previously, the authors showed that epidermal growth factor (EGF) stimulates phospholipase Cγ1 (PLCγ1) and phosphoinositide-3 kinase (PI3K) activities in confluent rabbit corneal epithelial cells (RCECs). The purpose of this study was to investigate whether PLCγ1 activity is upregulated during EGF-induced proliferation of RCECs and to determine whether there is any cross-talk between PLCγ1 and PI3K in these cells.

methods. Simian virus (SV)-40–immortalized RCECs were cultured in the presence and absence of EGF and other agents. At prescribed time intervals, the cultures were terminated and the cells counted. PLCγ1 activity in intact cells was assessed by measuring the production of[ ³H]IP₃ in[ ³H]myoinositol-labeled cells. The in vitro enzyme activity was assayed using immunoprecipitated PLCγ1 and[ ³H]PI(4,5)P₂ as substrate.[ ³H]IP₃, the product of PLCγ1, was analyzed by anion-exchange chromatography. The changes in protein content and level of phosphorylation of PLCγ1 were determined by Western immunoblot analysis, with the appropriate antibodies.

results. Addition of EGF (50 ng/ml) caused a time-dependent increase in proliferation of RCECs. The effect of EGF peaked at approximately 36 hours. Under the same experimental conditions, EGF stimulated PLCγ1 activity with a time course similar to that of cell proliferation. Data from Western immunoblot analysis revealed that the EGF-stimulated PLCγ1 activity was due to increased synthesis of the enzyme. Furthermore, during cell proliferation, tyrosine phosphorylation of PLCγ1 increased in a time-dependent manner that corresponded closely with the expression of PLCγ1. EGF exerted its effects both on cell proliferation and PLCγ1 activation in a dose-dependent manner. Treatment of the cells with U-73122, a PLC inhibitor, or myr-GLYRKAMRLRY, a myristoylated PLCγ1 inhibitor peptide, caused attenuation of both the EGF-stimulated cell proliferation and PLCγ1 activity. Treatment of the cells with the PI3K inhibitors, wortmannin or LY294002, caused inhibition of both EGF-stimulated cell proliferation and PLCγ1 activation. Addition of PI(3,4,5)P₃ to the in vitro PLCγ1 assay mixture stimulated the enzyme activity in a dose-dependent manner.

conclusions. The data suggest a positive correlation between EGF-stimulated PLCγ1 activation and cell proliferation in RCECs. The EGF-stimulated PLCγ1 activity was mirrored by increased synthesis and tyrosine phosphorylation of the enzyme. The data also show that PLCγ1 activation and cell proliferation were inhibited by PI3K inhibitors, suggesting a role for PI3K in EGF-stimulated proliferation of corneal epithelial cells.

Stimulation of a variety of cell surface receptors by the appropriate ligand results in activation of phospholipase C (PLC) and phosphoinositide 3-kinase (PI3K) in several cell types. ^{1 2 3 4} PLC catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP₂) to generate two intracellular second-messenger molecules, diacylglycerol (DAG) and 1,4,5-trisphosphate (IP₃). ^{1 2} IP₃ binds to intracellular receptors to initiate calcium release from intracellular stores. DAG, on the other hand, functions as an allosteric activator of protein kinase C (PKC), which is involved in cell proliferation and differentiation. To date, 10 mammalian PLC isozymes have been identified and can be divided into three types: PLCβ, PLCγ, and PLCδ. ² PLCβ subtypes are regulated by heterotrimeric guanosine triphosphate (GTP)-binding proteins of the G_q subfamily. PLCγ isoforms are activated by both receptor and nonreceptor tyrosine kinases, but the regulation of the PLCδ family of enzymes is less clear. PI3Ks phosphorylate the hydroxyl group at the 3-position of inositol in phosphoinositides to generate PI(3)P, PI(3,4)P₂ and PI(3,4,5)P₃. ^{3 4} These phosphoinositides are not hydrolyzed by known PLCs and therefore do not serve to generate additional intracellular second-messenger molecules. 

Recently, however, isoforms of phospholipase D, which specifically hydrolyze PI(3,4)P₂ and PI(3,4,5)P₃ to phosphatidic acid and the corresponding inositol phosphates, have been reported. ⁵ PI(3,4)P₂ and PI(3,4,5)P₃ are present only in trace amounts in normal resting cells, but their levels increase several fold in response to many types of stimuli. ³ Several studies using genetics, PI3K inhibitors, and PI3K overexpression have implicated the PI3Ks in the regulation of cell growth and differentiation, cell survival, cytoskeletal reorganization, and membrane trafficking. PI3Ks have been classified into three types on the basis of their primary structure, regulation and in vitro substrate specificity. ⁴ Type I PI3Ks are heterodimers consisting of p85 regulatory and p110 catalytic subunits. They can phosphorylate PI, PI(4)P, and PI(4,5)P₂, and are activated by receptor and nonreceptor tyrosine kinases and G-protein–linked receptors. Type II PI3Ks have substrate specificity restricted to PI and PI(4)P, and their mode of activation remains obscure. Type III PI3Ks are homologues of Saccharomyces cerevisiae Vps34p, which phosphorylates PI exclusively and is activated by a serine/threonine (Ser/Thr) kinase. 

It is known that interaction of epidermal growth factor (EGF) with its receptor is followed by receptor dimerization and phosphorylation of its tyrosine residues, which provides docking sites for several SH₂-containing proteins, including PLCγ1 and PI3K. ^{2 6} The recruitment of PLCγ1 and PI3K to EGF receptors results in their tyrosine phosphorylation and activation. Regeneration of corneal epithelium after in vivo injury begins with the migration of epithelial cells to the wounded area. This is followed by mitosis and cell proliferation to repair the defect. We and others have shown that topical application of EGF to the injured cornea stimulates cell proliferation and enhances wound repair. ^{7 8} The same outcome results from the addition of EGF to corneal epithelial cells in culture. ^{9 10}  

The exact biochemical events leading from EGF-receptor interaction and culminating in enhanced cell proliferation are not well understood. Previously, we reported that addition of EGF to quiescent corneal epithelial cells causes activation of PLCγ1 and PI3K. ^{11 12} Recently, we showed that activation of PI3K by EGF, both in vivo and in vitro, correlates positively with cell proliferation and wound repair in rabbit corneal epithelium. ^{8 10} The objective of the present study was to investigate whether there is any correlation between PLCγ1 activation and corneal epithelial cell proliferation induced by EGF. In addition, we examined the effects of PI3K on PLCγ1 activation in the EGF-treated epithelial cells. For this work, we used simian virus (SV)-40–immortalized rabbit corneal epithelial cells (RCECs), which are capable of growing for many passages without any alteration in their morphology or biochemical characteristics. ^{13 14}  

Materials and Methods

Dulbecco’s modified Eagle’s medium (DMEM), insulin, and gentamicin were purchased from Gibco (Grand Island, NY); human recombinant EGF, leupeptin, aprotinin, protein A (immobilized on Sepharose CL- 4B), and phenylmethylsulfonyl fluoride (PMSF) from Sigma (St. Louis, MO); PLCγ1 and -β1 polyclonal antibodies from Santa Cruz Biotechnology (Santa Cruz, CA); PI(4,5)P₂, PI(3)P, PI(3,4)P₂, and PI(3,4,5)P₃ from Biomol (Plymouth Meeting, PA); phosphatidylserine (PS) and phosphatidylethanolamine (PE) from Avanti Polar Lipids (Birmingham, AL); and wortmannin, LY294002, U-73122, and U-73343 from Calbiochem (La Jolla, CA).[ ³H]PIP₂ (specific radioactivity 10 Ci/mmol) was purchased from DuPont NEN (Boston, MA) and [³H]thymidine (specific radioactivity 64 Ci/mmol) from ICN Radiochemicals (Irvine, CA). 

Cell Culture

The SV-40–transformed RCECs were thawed and suspended in complete medium (DMEM/F-12 containing 40 μg/ml gentamicin, 5 μg/ml insulin, and 10% fetal bovine serum [FBS]) and cultured in a humidified atmosphere of 95% air-5% CO₂. The cultures were maintained by changing the medium every other day until the cells became confluent. To initiate subculture, the confluent cells were washed with Ca²⁺-Mg²⁺-free warm phosphate-buffered saline (PBS) and treated with 0.1% trypsin-0.02% EDTA for 5 minutes at 37°C. Next, complete medium was added and the cell suspension centrifuged at 300g for 5 minutes. The pelleted cells were resuspended in complete medium and then seeded in a 25-cm² culture flask at a density of 2 × 10⁴ cells/cm². The cultures were maintained by changing the medium on alternate days. 

Cell Proliferation Assay

Cell proliferation was assessed by counting (model Z1, Coulter, Hialeah, FL). The cells were washed with warm Ca²⁺- and Mg²⁺-free PBS, trypsinized, and suspended in 2 ml complete medium. A portion (100μ l) of the cell suspension was used for cell counting. Triplicate counts were taken for each data point. The incorporation of[ ³H]thymidine into DNA was measured as described previously. ¹⁰ Briefly, subconfluent (∼60%) cultures were incubated for 36 hours with different concentrations of EGF in DMEM containing 2 μCi/ml [³H]thymidine (specific radioactivity, 2 Ci/mmol). The culture medium was then removed and the cells washed in 10% trichloroacetic acid (TCA). The precipitated DNA was dissolved in 1% SDS-0.3 N NaOH and counted in a scintillation counter. 

In Vitro Assay of PLCγ1 and PLCβ1

Subconfluent (∼60%) cultures were serum starved for 24 hours before treatment with EGF or other agents. At appropriate times, the cultures were terminated and the cells scraped and homogenized in 20 mM Tris-HCl buffer (pH 7.0) containing 5 mM MgCl₂, 5 mM EDTA, 1 mM EGTA, 1 mM PMSF, 2 mM Na₃VO₄, 10 μg/ml leupeptin, and 50 μg/ml aprotinin. The homogenate was centrifuged at 600g for 10 minutes, and the resultant supernatant quantified for protein concentration by the method of Lowry et al. ¹⁵ Supernatants (cell lysates) containing equal amounts of protein were treated with PLCγ1 or PLCβ1 antibody, and the immunoprecipitated protein was used to assay PLC activity as described previously, with minor modifications. ¹⁶ Briefly, the reaction mixture contained 20 mM Tris-HCl buffer (pH 7.0), 0.1 M NaCl, 2 mM CaCl₂, 1 mM EGTA, 1 mM EDTA, 0.1% sodium cholate, 50 μM [³H]PIP₂ (30,000 disintegrations per minute [dpm]), and the immunoprecipitated enzyme protein in a final volume of 125 μl. The substrate was prepared by mixing chloroform solutions of[ ³H]PIP₂-unlabeled PIP₂, PE, and PS at a molar ratio of 1:2:2, respectively. After evaporation of the solvent under N₂, the lipids were suspended by sonication in sodium cholate–containing reaction buffer. The reaction was initiated by addition of the enzyme protein, incubated for 30 minutes at 37°C, and terminated by addition of 0.5 ml chloroform-methanol-concentrated HCl (50:50:1, by volume). Next, 150 μl of 1 M HCl containing 5 mM EGTA was added and the reaction mixture thoroughly mixed and centrifuged. An aliquot (400 μl) of the upper aqueous phase was removed and counted in a scintillation counter. When the effects of PI(3)P, PI(3,4)P₂, and PI(3,4,5)P₃ on the activity of PLCγ1 or PLCβ1 were to be determined, the immunoprecipitated enzymes were treated for 15 minutes with different concentrations of these lipids before their addition to the reaction mixture. 

Western Immunoblot Analysis

The protein content and level of tyrosine phosphorylation of PLCγ1 were determined by Western immunoblot analysis, as described previously. ¹¹ Briefly, RCECs lysates containing equal amounts of protein from EGF-treated and untreated cells were immunoprecipitated using anti-PLCγ1 antibody. The precipitates were boiled in Laemmli’s buffer for 5 minutes, separated by 10% SDS-PAGE, and the proteins transferred to nitrocellulose membranes. To determine total PLCγ1 content, the membranes were successively blotted with anti-PLCγ1 primary antibody and anti-rabbit horseradish peroxidase (HRP)-conjugated goat secondary antibody. To determine the level of PLCγ1 tyrosine phosphorylation, the blots were stripped and reprobed successively with anti-phosphotyrosine primary antibody and anti-mouse HRP-conjugated secondary antibody. The protein bands were visualized using the enhanced chemiluminescence (ECL) Western detection system (Amersham Pharmacia Biotech, Parsippany, NJ). 

Synthesis of Inhibitor Peptides for PLCγ1

The myristoylated PLCγ1 inhibitor peptide, myr-GLYRKMRLRY (myr-PCI(Y)), and its weaker analogue, myr-GLFRKMRLRF (myr-PC I(F)), were synthesized (Pioneer; PerSeptive Biosystems, Framingham, MA) at 0.1 mM, using the 9-fluorenylmethoxycarbonyl (Fmoc) solid phase chemistry. The myristoyl group was added as the free acid. The myristic acid was treated as an amino acid using an extended coupling procedure. Peptide purity was determined by HPLC on a C-18 reversed-phase column and the molecular weight determined by mass spectrometry. 

Statistical Analysis of the Data

Each experiment consisted of incubations that, when pooled, yielded three independent samples for each data point. All experiments were performed at least twice, with the results expressed as mean ± SEM. Statistical analysis was performed with Student’s t-test for nonpaired data. P ≤ 0.05 was considered significant. 

Results

Effect of EGF on PLCγ1 Activity and Tyrosine Phosphorylation during Cell Proliferation

To determine whether cell counting can provide an accurate measure for cell proliferation, the data from cell counting were compared with data from [³H]thymidine incorporation into DNA. As shown in Table 1 , there was a close correlation between the cell numbers and the[ ³H]thymidine incorporation into DNA in RCECs treated with different concentrations of EGF. Therefore, in further experiments, we routinely used cell counting to estimate proliferation in RCECs. To investigate whether EGF exerts an effect on PLCγ1 activity during cell proliferation, the serum-starved RCECs were cultured in the presence and absence of EGF, followed by cell counting and determination of PLCγ1 activity. As shown in Figure 1A , in the absence of EGF, there was a gradual increase in cell number that reached a plateau at 36 hours. Addition of EGF (50 ng/ml) exerted a large, time-dependent increase in cell proliferation, compared with the untreated cells. When analyzed for PLCγ1 activity, there was a time-dependent increase in enzyme activity in the untreated cells (Fig. 1B) . Treatment with EGF caused a large increase in PLCγ1 activity, compared with the untreated cells. 

Western blot data revealed a time-dependent increase in PLCγ1 synthesis in the untreated cells (Fig. 1C) . PLCγ1 protein levels were further increased at all time points after treatment with EGF. Reprobing the PLCγ1 immunoblot with phosphotyrosine antibody showed a time-dependent increase in tyrosine phosphorylation of PLCγ1 in EGF-treated and untreated cells (Fig. 1D) . There appeared to be a close correlation between PLCγ1 phosphorylation and the level of PLCγ1 expression. As shown in Figure 2 , EGF caused a dose-dependent increase in cell proliferation and PLCγ1 activity. The effects were observed with as low as 5 ng/ml EGF and were maximal at 50 ng/ml EGF, followed by a slight decrease. 

Effect of U-73122 on EGF-Stimulated PLCγ1 Activity and Cell Proliferation

To further examine the relationship between EGF-stimulated PLCγ1 activity and cell proliferation, RCECs were cultured in the presence and absence of EGF and different concentrations of the PLC inhibitor U-73122 and its inactive analogue U-73343. The cultures were terminated and PLCγ1 immunoprecipitated and assayed for enzyme activity in vitro. U-73122 caused a dose-dependent decrease in cell number in EGF-treated RCECs (Fig. 3A) . Maximal inhibition of EGF-stimulated cell proliferation was observed at 2 μM of the inhibitor. When examined for its effect on PLCγ1 activity, U-73122 caused a dose-dependent decrease in PLCγ1 activity in EGF-treated and untreated cells (Fig. 3B) . The stimulatory effect of EGF on PLCγ1 activity was maximally inhibited in the presence of 2μ M U-73122. U-73343, the inactive analogue, failed to exert any significant inhibitory effect on cell proliferation or PLCγ1 activity. U-73122 exerted no effect on EGF-induced expression or tyrosine phosphorylation of PLCγ1 in RCECs during proliferation (Figs. 3C 3D) . 

Effect of Myristoylated PLC Inhibitor Peptide on IP₃ Formation and Cell Proliferation

myr-PCI(Y), the myristoylated PLCγ1 inhibitor peptide, was used to further confirm the involvement of PLCγ1 in EGF-stimulated cell proliferation in RCECs. The cells were cultured in the presence of different concentrations of the peptide before counting and determination of PLCγ1 activity in vitro. myr-PCI(Y) inhibited the EGF-induced cell proliferation in a dose-dependent manner (Fig. 4A) . The EGF effect was reduced by approximately 40% in the presence of 10 μM myr-PCI(Y). Unlike U-73122, the peptide did not inhibit the PLCγ1 activity in the untreated cells. However, it inhibited the EGF-induced activation of PLCγ1 in a dose-dependent manner (Fig. 4B) . At 10 μM myr-PCI(Y), the EGF effect was inhibited by almost 50%. myr-PCI(F), a PCI peptide analogue, was considerably less potent in inhibiting the EGF-induced PLCγ1 activation and cell proliferation. 

Effect of PI3K Inhibitors on PLCγ1 Activity and Cell Proliferation

We used wortmannin and LY294002 as inhibitors of PI3K to determine whether PLCγ1 lies downstream of PI3K in mediating the effect of EGF on proliferation of corneal epithelial cells. Treatment with wortmannin caused a dose-dependent decrease in EGF-stimulated proliferation of RCECs (Fig. 5A) . Wortmannin also inhibited the EGF-induced activation of PLCγ1in a dose-dependent manner (Fig. 5B) . LY294002 also caused a dose-dependent decrease in EGF-stimulated cell proliferation and PLCγ1 activity in RCECs (Fig. 6) . 

Effect of PI3K Products on the Activity of PLCγ1

We used PI(3)P, PI(3,4)P₂, and PI(3,4,5)P₃ to investigate whether PLCγ1 could be a direct target of the PI3K products. PLCγ1 was immunoprecipitated from cultured RCECs and assayed for its activity in the presence and absence of 3-phosphoinositides. As shown in Figure 7 , PI(3,4,5)P₃ increased the activation of PLCγ1 in a dose-dependent manner up to 100 μM, followed by a slight tapering off. The maximal stimulation, observed at 100 μM, then decreased at higher concentrations of PI(3,4,5)P₃. PI(3)P and PI(3,4)P₃ did not have any significant stimulatory effect on PLCγ1 activity. Similarly, PI(3,4,5)P₃ had no effect on the activity of PLCβ1 immunoprecipitated from RCECs (data not shown). 

Discussion

Wound repair in corneal epithelium is highly complex and involves cell migration, proliferation and differentiation, and secretion of extracellular matrix proteins. It has been reported that EGF treatment, both in vitro and in vivo, accelerates the wound-healing process by promoting migration and proliferation of the corneal epithelial cells. ^{7 8} The biochemical events that relay signals from the EGF receptor to the nucleus, resulting in increased DNA synthesis, are not clear. It has been reported that addition of EGF to quiescent corneal epithelial cells causes stimulation of PLCγ1, possibly because of its translocation to, and phosphorylation by, the EGF receptor. ¹¹  

The data from the present study show that serum-starved RCECs can proliferate, albeit at a low rate, in the absence of any exogenously added EGF (Fig. 1A) . The observed proliferative response is most likely due to autocrine secretion of growth-promoting substances by cells in the culture medium. When treated with EGF, the cells exhibited a marked time-dependent increase in cell number. An important finding of the present study is that, under conditions when RCECs were undergoing active proliferation, there was a concomitant increase in PLCγ1 activity both in the EGF-treated and untreated cells (Fig. 1B) . Western blot analysis revealed that the increased PLCγ1 activity was due to increased synthesis and tyrosine phosphorylation of the enzyme (Figs. 1C 1D) . The effect of EGF on PLC was found to be specific for the PLCγ1 isoform. Both the activity and expression of PLCβ1 remained unchanged during cell proliferation (data not shown). 

These data provide evidence that PLCγ1 is involved in a signaling pathway that mediates the EGF’s effect on DNA synthesis in RCECs. The findings are in accord with previously published reports in which microinjection of anti-PLCγ1 antibody blocked serum- and Ras-stimulated DNA synthesis. ¹⁷ In other studies, microinjection of PLCγ1 SH₂ domains into Madin-Darby canine kidney (MDCK) epithelial cells or fibroblasts blocked the platelet-derived growth factor (PDGF)–induced DNA synthesis. ^{18 19}  

Additional evidence to support the suggestion that PLCγ1 is involved in corneal epithelial cell proliferation comes from the use of the PLC inhibitor U-73122. This inhibitor has been widely used in a number of studies examining the role of PLC in intracellular signaling mechanisms. ^{20 21} In the present study, by using U-73122, we were able to significantly inhibit the EGF-stimulated PLCγ1 activation, which resulted in a corresponding decrease in EGF-induced proliferation in RCECs (Fig. 3) . The inhibitory effect of U-73122 was not due to the inhibition of PLCγ1 synthesis or its phosphorylation. It has been reported that PLCγ1 possesses, adjacent to its SH₂ and SH₃ motifs, a PLC inhibitor (PCI) region that strongly suppresses its catalytic activity. ²² It has been suggested that stimulation of the cells with growth factors probably dissociates the PCI region of PLCγ1 from the catalytic region, resulting in the activation of the enzyme. Myristoylation of PCI facilitates its entry into Swiss 3T3 cells causing inhibition of cell growth and phosphoinositide hydrolysis. ²³ We used myr-PCI(Y) to further examine the involvement of PLCγ1 in RCEC proliferation. When added to RCECs, myr-PCI(Y) suppressed the EGF-induced PLCγ1 activation and cell proliferation (Fig. 4) . The inactive structural analogue myr-PCI(F) did not show any consistent inhibitory effect on PLCγ1 activity and cell proliferation, even when used at high concentrations. Taken together, these data provide convincing evidence that PLCγ1 is upregulated during EGF-induced proliferation of RCECs. 

Activation of EGF receptor is followed by autophosphorylation of its tyrosine residues, which serve as high-affinity binding sites for SH₂-containing proteins, including PLCγ1. Recruitment of PLCγ1 to the receptor results in tyrosine phosphorylation and stimulation of its enzyme activity. ^{2 24} Tyrosine phosphorylation of PLCγ1 has been reported to be critical in transducing mitogenic signals from growth factor receptors to the interior of the cell. In particular, substitution of Phe for Tyr783 completely blocks the activation of PLCγ1 by PDGF in NIH 3T3 cells. ²⁵ In addition to being activated by tyrosine phosphorylation, PLCγ1 has been shown to be stimulated by lipid-derived second messengers, such as arachidonic acid, phosphatidic acid, and PI(3,4,5)P₃. ²⁴ The activity of PLCγ1 has been reported to be greatly enhanced by PIP₃, whereas that of the β- or δ-PLC isozymes remains unchanged. ²⁶ Furthermore, overexpression of PI3K in COS-7 cells causes increased activation of PLCγ1 that is reduced to the normal level when the cells are treated with PI3K inhibitors. In other studies, the SH₂ and PH domains of PLCγ1 have been found to potently bind PIP₃ generated by PI3K. ^{19 27}  

To investigate whether EGF-induced PLCγ1 activation in RCECs is dependent on concomitant activation of PI3K, PLCγ1 activity was assayed in cells treated with the PI3K inhibitors wortmannin and LY294002. The results showed that both wortmannin and LY294002 dose-dependently inhibited EGF-stimulated PLCγ1 activation and cell proliferation (Figs. 5 6) . Because the cell number remained essentially unchanged with increasing concentrations of wortmannin or LY294002, it was unlikely that the inhibitory effects of these compounds on EGF-induced responses was due to increased cell death. When wortmannin or LY294002 was added directly to the PLCγ1 assay mixture, there was no inhibition of the enzyme activity (data not shown), which suggests that these compounds inhibit PLCγ1 by selectively blocking PI3K activity. Further evidence that PLCγ1 lies downstream of PI3K is provided by the experiments in which PLCγ1 was assayed in the presence of 3-phosphoinositides. The enzyme activity was markedly increased by PI(3,4,5)P₃ ≫ PI(3,4)P₂ and PI(3)P was without effect (Fig. 7) . 

Taken together, the data provide support for the hypothesis that EGF-induced proliferation of RCECs involves both increased PLCγ1 synthesis and enzyme activation by PI(3,4,5)P₃, a product of PI3K. One likely mechanism for PLCγ1 activation by PI(3,4,5)P₃ may involve recruitment of the enzyme by PI(3,4,5)P₃ to the plasma membrane adjacent to its substrate, PIP₂. The SH₂, SH₃, and PH domains of PLCγ1 can specifically interact with inositol phospholipids. After PLCγ1 recruitment to the plasma membrane, the enzyme could be activated directly by PI(3,4,5)P₃. 

In conclusion, the current data demonstrate a close correlation between PLCγ1 activation and cell proliferation in EGF-treated and untreated RCECs. PLCγ1 activation involved both increased PLCγ1 synthesis and tyrosine phosphorylation. Furthermore, the activity of PLCγ1 was found to be modulated by PI3K, suggesting that PLCγ1 lies downstream of PI3K. However, our studies did not provide information regarding the exact biochemical events that follow PLCγ1 activation and culminate in increased DNA synthesis during cell proliferation. One second-messenger molecule generated by PIP₂ hydrolysis is DAG, which is a natural activator of PKC. There are several studies suggesting that PKC can increase DNA synthesis by activating the Raf/mitogen-activated kinase kinase (MEK)/extracellular signal-regulated (ERK) pathway. ^{28 29} Therefore, it is possible that the mitogen-activated protein (MAP) kinase cascade also plays a similar key role in EGF-stimulated wound repair in corneal epithelium. 

 Supported by Grant EY05738 from the National Institutes of Health.

 Submitted for publication July 6, 2000; revised December 13, 2000 and February 8, 2001; accepted February 23, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Effect of EGF on Proliferation of RCECs Measured by[ ³H]Thymidine Incorporation into DNA and by Cell Counting

EGF (ng/ml)	[3H]Thymidine Incorporated (dpm × 10−5)	% Control	Cell Number (10−5)	% Control
0	1.52 ± 0.04	—	0.87 ± 0.05	—
10	2.30 ± 0.05	151	1.57 ± 0.07	180
20	3.50 ± 0.04	230	2.30 ± 0.10	264
50	4.25 ± 0.03	279	2.75 ± 0.15	316
100	3.75 ± 0.05	246	2.35 ± 0.08	270

 Serum-starved, subconfluent (∼60%) cultures were incubated with different concentrations of EGF for 36 hours in DMEM containing 2μ Ci/ml [³H]thymidine (specific radioactivity 2 Ci/mmol). The culture medium was removed and cells washed with 10% TCA. The precipitated DNA was dissolved in 1% SDS-0.3 N NaOH and counted for radioactivity. For determination of cell numbers, the cells were trypsinized and counted. Data are mean ± SEM of two experiments, each conducted on quadruplicate cultures.

Figure 1.

 

Time-course effect of EGF on PLCγ1 activity during RCEC proliferation. Serum-starved cells were cultured for different time intervals in the absence and presence of EGF (50 ng/ml). (A) Cells were trypsinized and counted. (B) Cells were lysed and the lysate immuno-precipitated with PLCγ1 antibody. The immunoprecipitate was collected and assayed for PLC activity, using[ ³H]PIP₂ as substrate. The enzyme activity is expressed as[ ³H]IP₃ produced per milligram protein in RCECs. (C) Western blot analysis of PLCγ1 immunoprecipitates recovered from cell lysates. (D) Western blot stripped and reprobed with anti-phosphotyrosine antibody. Data are mean ± SEM of three separate experiments with at least duplicate cultures for each data point. *Statistically significant (P < 0.05) increase in cell number or[ ³H]IP₃, compared with the corresponding control.

Figure 2.

 

Dose-response effect of EGF on cell proliferation and PLCγ1 activity in RCECs. Serum-starved cells were cultured for 36 hours in the absence and presence of different concentrations of EGF. (A) Cells were trypsinized and counted. (B) Cells were lysed and the lysate immunoprecipitated with PLCγ1 antibody. The immunoprecipitate was collected and assayed for PLC activity using[ ³H]PIP₂ as substrate. Data are mean ± SEM of three independent experiments with triplicate cultures for each data point. *Statistically significant (P < 0.05) increase in cell number, compared with the control.

Figure 3.

 

Effect of U-73122 on EGF-stimulated cell proliferation and PLCγ1 activity in RCECs. Serum-starved cells were cultured for 36 hours in the presence and absence of EGF (50 ng/ml) and different concentrations of U-73122. The inactive analogue U-73343 was used as a negative control. (A) Cells were trypsinized and counted. (B) Cells were lysed and immunoprecipitated using PLCγ1 antibody. The resultant immunoprecipitate was assayed for PLC activity using [³H]PIP₂ as the substrate. (C) Western blot of the immunoprecipitated PLCγ1. (D) PLCγ1 immunoblot stripped and reprobed with anti-phosphotyrosine antibody. Data are mean ± SEM from three independent experiments with three to four cultures for each data point. *Statistically significant (P < 0.05) decrease in cell number or PLCγ1 activity, compared with the corresponding cultures not treated with U-73122.

Figure 4.

 

Effect of myristoylated PCI peptides on EGF-stimulated cell proliferation and PLCγ1 activity in RCECs. Serum-starved cells were cultured for 36 hours in the absence and presence of EGF (50 ng/ml) and different concentrations of myr-PCI(Y). myr-PCI(F) was used as a negative control. (A) Cells were trypsinized and counted. (B) Cells were lysed and immunoprecipitated using PLCγ1 antibody. The resultant immunoprecipitate was assayed for PLC activity using [³H]PIP₂ as substrate. Data represent mean ± SEM of three independent experiments with at least duplicate cultures for each data point.* Statistically significant (P < 0.05) decrease in cell number or [³H]IP₃, compared with the corresponding cultures not treated with myr-PCI(Y).

Figure 5.

 

Effect of wortmannin on EGF-stimulated cell proliferation and PLCγ1 activity in RCECs. Serum-starved cells were cultured for 36 hours with EGF (50 ng/ml) and different concentrations of wortmannin. (A) Cells were trypsinized and counted. (B) Cells were lysed and immunoprecipitated using PLCγ1 antibody. The resultant immunoprecipitate was assayed for PLC activity using[ ³H]PIP₂ as substrate. Data are mean ± SEM of three to four experiments conducted with at least duplicate cultures for each data point. *Statistically significant (P < 0.05) decrease in cell number or[ ³H]IP₃, compared with the corresponding cultures not treated with wortmannin.

Figure 6.

 

Effect of LY294002 on EGF-stimulated cell proliferation and PLCγ1 activity in RCECs. All experimental conditions were the same as described in Figure 5 , except that LY294002 was used as a PI3K inhibitor. Data are mean ± SEM from three to four experiments, each with triplicate cultures. *Statistically significant (P < 0.05) decrease in cell number or[ ³H]IP₃, compared with the corresponding cultures not treated with LY294002.

Figure 7.

 

Effects of PI(3)P, PI(3,4)P₂, and PI(3,4,5)P₃ on PLCγ1 activity. RCECs were lysed and immunoprecipitated with PLCγ1 antibody. The immunoprecipitate was collected and incubated with 3-phosphoinositides for 15 minutes at room temperature. The immunoprecipitate was then assayed for PLC activity using[ ³H]PIP₂ as a substrate. Data are mean ± SEM of three to four experiments, each conducted in triplicate.* Statistically significant (P < 0.05) increase in[ ³H]IP₃, compared with the cultures not treated with 3-phosphoinositides.

The authors thank Kaoru Araki-Sasaki for providing SV-40–transformed rabbit corneal epithelial cells and Rhea Markowitz for critical reading of the manuscript.