Expression of Phosphatidylinositol 3–Kinase during EGF-Stimulated Wound Repair in Rabbit Corneal Epithelium

Abstract

purpose. To investigate the effect of epidermal growth factor (EGF) on the induction of phosphatidylinositol 3-kinase (PI 3–kinase) gene expression during rabbit corneal epithelial wound repair.

methods. Epithelial wounds (6 mm in size) were created in rabbit corneas and EGF (2 μg) applied every 8 hours to one eye, and the other eye served as a control. The wound repair was monitored by staining the tissue with fluorescein followed by photography. The wound area was quantified with a computer program. At different time intervals, the rabbits were killed and the corneal epithelium used for estimation of PI 3–kinase activity, western blot analysis, or reverse transcription–polymerase chain reaction (RT–PCR). For in situ hybridization, the whole corneas were sectioned and the sections processed with PI 3–kinase mRNA probes.

results. In the untreated eye, the epithelial wound progressively healed in a time-dependent manner, with 75% of the wound closed at 48 hours post wounding. Application of EGF to the corneal epithelium further stimulated wound repair at all time intervals, and the wound was completely closed at 48 hours. Analysis of PI 3–kinase showed a time-dependent increase in its enzyme activity that was maximally increased at 36 hours, the time when the wound was nearly closed. Western blot analysis revealed increased amounts of PI 3–kinase protein during the course of wound repair. Analysis of RT–PCR products from epithelial tissues, taken at different times during wound repair, showed increased PI 3–kinase expression that was maximum at 48 hours post wounding. A visible increase in PI 3–kinase gene expression was also detected by in situ hybridization during the course of the wound repair. This expression was increased maximally by EGF at 48 hours post wounding.

conclusions. The results indicate a temporal correlation between increased activation and expression of PI 3–kinase and the epithelial wound repair. Topical application of EGF further stimulates the activity and expression of PI 3–kinase. It is suggested that PI 3–kinase and its products may play a role in EGF-induced cell proliferation during corneal epithelial wound repair.

Phosphatidylinositol 3–kinase (PI 3–kinase) is a heterodimer composed of a 110-kDa catalytic subunit and an 85-kDa regulatory subunit, which are tightly associated. ¹ The activity of PI 3–kinase has been shown to increase in response to a number of hormonal and growth factor stimuli in several cell types. ² Once activated, PI 3–kinase phosphorylates phosphoinositides at the D3 position of the inositol ring to generate phosphatidylinositol 3-phosphate, phosphatidylinositol 3,4-bisphosphate, and phosphatidylinositol 3,4,5-trisphosphate, which are poor substrates for any known phospholipase C (PLC). ³ Although the physiological role of PI 3–kinase and its products has not been established, several studies have implicated PI 3–kinase in many cellular responses, including cell growth and transformation, enhanced cell motility, T-cell signaling, membrane ruffling, and vesicular trafficking. ⁴ Several studies have also shown that the lipid products of PI 3–kinase activate protein kinase C. ⁵ Recently, generation of PI 3,4,5-P₃ in HepG₂ cells has been reported to increase IP₃ production and intracellular calcium release, suggesting a role for PI 3–kinase in PLCγ-mediated calcium signaling. ⁶  

Corneal epithelial wound repair involves reorganization, migration, and proliferation of the epithelial cells. ⁷ It has been reported that epidermal growth factor (EGF) facilitates corneal epithelial wound repair by increasing migration and mitosis of the epithelial cells, in both in vivo and in vitro model systems. ^{8 9} Although the exact mechanism underlying the mitogenic effect of EGF is not clear, it is known that the first step in EGF signaling involves interaction with and activation of its receptor containing intrinsic tyrosine kinase activity. Once activated, the receptor phosphorylates several cellular proteins including the 85-kDa regulatory subunit of PI 3–kinase. We have previously shown that treatment of cultured corneal epithelial cells with EGF causes activation of PI 3–kinase. ¹⁰ More recently, using monolayers of cultured corneal epithelial cells, we found a correlation between the EGF-stimulated PI 3–kinase activity and the EGF-stimulated wound repair. ¹¹ The objective of the present study was to investigate whether EGF stimulates PI 3–kinase activity in rabbit corneal epithelium, in vivo, and whether a similar correlation exists between PI 3–kinase activation and EGF-stimulated wound repair in this tissue. We have also examined changes in mRNA, encoding for PI 3–kinase, in rabbit corneal epithelium at different times after wounding. 

Materials and Methods

Animal and Corneal Epithelial Wound Model

Albino rabbits weighing 4 to 5 kg were used, and all procedures conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The animals were anesthetized by intramuscular injection of a mixture of ketamine and xylazine. Corneal wounds were made in each eye by placing a 6-mm filter paper disc soaked in n-haptanol centrally on the corneal epithelium for 1 minute, followed by thorough irrigation with sterile saline and gentle removal of the epithelium. Two micrograms of EGF (0.1 mg/ml in 1% Carbapol gel) was topically applied to one eye every 8 hours, whereas the other eye received the Carbapol gel alone and served as a control. To monitor the time course of wound repair, the corneas were stained with 1% sodium fluorescein and photographed under filtered green light. The wound area was digitized with an optical scanner (ScanJet 3C; Hewlett–Packard, Hopkins, MN) and quantified with a computer program, SigmaScan (Jandel Scientific, San Rafael, CA). At different time intervals, the animals were killed with an overdose of pentobarbital, and either the epithelium was scraped off and collected for PI 3–kinase activity assay, western blot analysis, and RT–PCR, or the whole corneas were excised and immediately frozen in liquid nitrogen for sectioning. 

Immunoprecipitation and Assay of PI 3–Kinase

The procedures for immunoprecipitation and assay of PI 3–kinase were as described previously. ¹¹ Briefly, the scraped epithelium was immediately lysed in 200 μl of lysis buffer that contained 140 mM NaCl, 20 mM Tris–HCl (pH 7.4), 1 mM CaCl₂, 1 mM MgCl₂, 1% Nonidet P-40, 1 mM phenylmethyl sulfonyl, 50 μM leupeptin, 10 μg/ml aprotinin, and 200 μM sodium vanadate. After brief centrifugation, same amount of total protein (∼1 mg) was taken out of each supernatant and diluted in 0.5 ml of the same buffer. Ten microgram of anti-PI 3–kinase polyclonal antibody, conjugated to protein A agarose, was added to each sample and the tubes incubated overnight at 4°C with gentle rotation. After centrifugation, the lysis buffer was removed and the pellets successively washed in each of the following buffers: phosphate buffered saline (PBS) containing 1% Nonidet P-40, and 200 μM sodium vanadate; 100 mM Tris–HCl (pH 7.4), 5 mM LiCl, and 200 μM sodium vanadate; or 10 mM Tris–HCl (pH 7.4), 150 mM NaCl, 5 mM EDTA, and 200 μM sodium vanadate. The enzyme activity was assayed directly on the beads by adding 20 μg phosphatidylinositol, 50 μM[ ³²P]ATP (2 × 10⁶ dpm) and 20 mM MgCl₂ in 10 mM Tris–HCl (pH 7.4) to each sample. After incubation at 32°C for 20 minutes, the reaction was terminated by the addition of 6 N HCl. The enzyme product was extracted by the method of Bligh and Dyer ¹² and chromatographed on oxalate-impregnated t.l.c. plate with a solvent system consisting of chloroform/acetone/methanol/acetic acid/water (80:30:26:24:14, by volume). The radioactivity in the 3-phosphoinositide lipid spot was determined by liquid scintillation counting. The PI 3–kinase activity in wounded epithelium was expressed as the percentage of increase over the basal activity recovered in unwounded corneal epithelium. 

Western Blot Analysis

Corneal epithelium, scraped at different times during wound repair, was solubilized in 1% sodium dodecyl sulfate (SDS) buffer containing 10% glycerol, 100 μg/ml phenylmethyl sulfonyl, 60 μg/ml aprotinin, and 5% β-mercaptoethanol. After centrifugation at 15,000g for 20 minutes, the protein concentration of the resulting supernatant was measured by the method of Lowry et al. ¹³ with bovine serum albumin as standard. Each sample, containing 10 μg of total protein, was then analyzed on 7.5% SDS–polyacrylamide gel electrophoresis mini-gel and transferred to nitrocellulose membrane. After incubation with a blocking solution containing 3% bovine serum albumin in Tris-buffered saline (TBS) for 60 minutes, the paper was incubated with anti–PI 3–kinase antibody overnight. The unbound antibody was removed by washing the membrane in TBS containing 0.05% Tween-20. The antibody bound to the nitrocellulose membrane was detected by alkaline phosphatase–conjugated goat anti-mouse IgG and the chromogenic substrate 5-bromo-4-chloro-3-indoyl phosphate and nitro blue tetrazolium. 

Extraction of Total RNA

Total RNA was extracted from the scraped epithelial tissues by using RNeasy kit from Qiagen (QIAGEN, Chatsworth, CA). Briefly, 2 to 4 mg of corneal epithelium was mixed with 350 μl guanidinium isothiocyanate containing lysis buffer and 500 mg treated sand, ¹⁴ and the mixture vigorously vortexed for 2 minutes. The resultant homogenate was transferred to an RNase-free microfuge tube and then passed through a 23-gauge needle 10 times to shear DNA. Next, an equivalent volume of 70% ethanol was added to the homogenate, and the mixture applied to an RNeasy mini-column. The column was centrifuged and washed twice with the wash buffer. The RNA was eluted with 50 to 80 μl of DEPC-treated water. Purity of RNA was confirmed with 260/280 OD spectrophotometer readings, which ranged from 1.8 to 2.0. 

Reverse Transcription-Polymerase Chain Reaction (RT-PCR)

The RNA was reversely transcribed to generate cDNAs using oligo(dT) as primers. Samples containing 1 μg of RNA per time point were purified by incubating with DNase I and then heat-denatured for 10 minutes at 68°C. After chilling the mixture on ice, the RNA was incubated at 37°C for 1 hour in a total volume of 20 μl that contained 1.25 mM each dNTP, 0.5 μg oligo (dT), 20 units of RNAsin, 1 unit of M-MLV reverse transcriptase, 10 mM DTT, and 50 mM Tris-HCl pH 8.3. The reaction was terminated by heating the mixture at 68°C for 10 minutes The reverse transcriptase reaction was brought to 100 μl by adding TE buffer. PCR reactions were carried out using twenty bp oligonucleotide primers. The primers were designed to correspond to the highly conserved regions of PI 3–kinase (p85) cDNA sequences in different species. The upstream primer sequence was 5′-AGGAGCGGTACAGCAAAGAA. The downstream primer sequence was 5′-GCCGAACACCTTTCTGAGTC. PCR was carried out on a Perkin Elmer GeneAmp 2400 thermocycler. The PCR reaction mixture in a final volume of 50μ l contained 1 μl of reverse transcription product, 20 mM Tris-HCl pH 8.3, 50 mM KCl, 2 mM MgCl₂, 0.2 mM each of dNTP, 1 μM each of both the primers, and 5 units of TaqE polymerase. The reaction mixture was subjected to the following program: First cycle, 94°C for 4 minutes, 55°C for 30 seconds, and 72°C for 45 seconds Subsequent 28 cycles, 92°C for 30 seconds, 55°C for 30 seconds, and 72°C for 45 seconds Last cycle, 92°C for 30 seconds, 55°C for 30 seconds, and 72°C for 10 minutes An aliquot was subjected to agarose gel electrophoresis and the amplification products were visualized by ethidium bromide staining. The housekeeping gene, GAPDH, served as control to ensure that equal amounts of RNA were analyzed from each sample. The sequence of upstream primer for GAPDH was 5′-CGGGAAGCTGGTCATCAACGG, and the downstream primer was 5′-TTTCTCCAGGCGGCAGGTCAG. 

Cloning and Sequencing of cDNA Fragment of PI 3–Kinase

The RT-PCR product corresponding to approximately 260 bp was excised from the agarose gel and purified using the Promega Wizard Purification Kit. The purified fragment was ligated with pGEM-T vector (Promega) and cloned into E. coli strain JM109 according to the manufacturer’s instructions. The plasmid DNA with insert was isolated from the transformed E. Coli and purified using the Promega Wizard Mini-preparation System. The cDNA insert was released using restriction endonuclease BstZI and identified on 2% agarose gel. The nucleotide sequence of the DNA insert was determined by the dideoxynucleotide chain-termination reaction and automated sequencing. The primers used in the dye terminator reaction were T7 and SP6 promoters. 

In Situ Hybridization

The plasmid pGEM DNA, with 260 bp PI 3–kinase insert, was linearized with ApaI and used as a template for the synthesis of an antisense riboprobe using T7 polymerase, or was linearized with PstI for the synthesis of a sense riboprobe using SP6 polymerase. In vitro transcription was performed in the presence of digoxigenin-11-uridine triphosphate (DIG-UTP) to produce DIG-UTP-labeled single strand antisense or sense RNA probes using the DIG RNA Labeling Kit (Boehringer Mannheim) according to the manufacturer’s protocol. The riboprobes were purified using Bio-spin 6 columns. The size and integrity of the riboprobes were checked using agarose gel electrophoresis and a dot blot analysis. 

Corneas from the wounded and unwounded eyes were sandwiched between two slices of rabbit muscles, and sectioned at 10 μm thickness on a cryostat. The sections were immediately fixed in 4% paraformaldehyde in PBS for 20 minutes, dried through a graded ethanol series, and stored at 4°C until use. After briefly hydrating in ethanol/PBS, the sections were incubated in 1 μg/ml proteinase K in 10 mM Tris.HCl, 5 mM EDTA and 5% SDS at 37°C for 10 minutes Next, the sections were washed twice for 5 minutes in PBS and re-fixed in 4% paraformaldehyde in PBS for 10 minutes This was followed by washing twice for 5 minutes in PBS and autoclaved water. The sections were then acetylated by incubating the slides in freshly prepared 0.25% acetic anhydride in 0.1 M triethanolamine for 10 minutes The slides were washed twice in PBS and once in saline 5 minutes each. Before air drying, the slides were briefly dehydrated in graded series of ethanol. Hybridization was performed in a buffer containing 50% formamide, 5x SSC (0.3 M NaCl and 0.3 M sodium citrate), 2% blocking reagent (Boehringer Mannheim), 1% N-lauroylsarcosine, and 0.02% SDS. Fifty microliters per section of the hybridization buffer containing 0.2 ng/μl of DIG-labeled probe (antisense or sense RNA) was applied to the sections, which were covered by glass coverslips. The slides were placed in moisture slides mailers which were sealed with parafilm and incubated overnight at 42°C in a hybridization oven. The next day the sections were washed as follows: 1x SSC at RT for 5 minutes, 1x SSC at RT for 15 minutes, 0.5x SSC at RT for 30 minutes, and 1x SSC at 37°C for 30 minutes To detect DIG-labeled riboprobes, the slides were briefly washed in buffer 1 (0.1 M maleic acid, pH 7.5 and 0.15 M NaCl), incubated in blocking buffer (buffer 1 containing 0.05% Triton X-100, 5% rabbit serum) for 30 minutes, and then washed in buffer 1 for 5 minutes Next, the slides were incubated with alkaline phosphatase-labeled anti-digoxigenin antibody in buffer 1 containing 0.05% Triton-100 and 1% serum for 3 hr. The slides were washed twice in buffer 1 containing 0.2 mg/ml levamisole for 15 minutes The color was developed by incubating the slides with nitroblue tetrazolium salt (NBT, 0.34 mg/ml) and 5-bromo-4-chloro-3-indolyl phosphate (BICP, 0.18 mg/ml) which were dissolved in 100 mM Tris.HCl pH 9.5, 100 mM NaCl and 50 mM Mg Cl₂. The color development was examined under microscope and stopped by placing the slides in 10 mM Tris.HCl pH 8.0/1 mM EDTA and washing them in distilled water. 

Results

Effect of EGF on Corneal Epithelial Wound Repair

Twelve hours after wounding, a significant decrease was detected in the wound area in the untreated corneal epithelium (Figs. 1 and 2) . This process of wound repair continued and by 48 hours, 75% of the wound was already closed. The wound had completely repaired within 72 hours post wounding (data not shown). EGF, applied topically to the wounded epithelium, further stimulated wound repair at all time points compared with the untreated control epithelium. At 48 hours, the wound was completely closed. 

Effect of EGF on PI 3–Kinase Activity during Corneal Epithelial Wound Repair

To examine PI 3–kinase activity during the course of wound repair, the epithelial tissues were collected at different time intervals and analyzed for PI 3–kinase activity. Unwounded corneal epithelium showed very little PI 3–kinase activity (data not shown). As shown in Figure 3 , the PI 3–kinase activity increased as early as 12 hours after wounding. The increase in PI 3–kinase activity peaked at 36 hours and then slightly declined thereafter. EGF treatment resulted in a large stimulation of PI 3–kinase activity; a peak was reached at 36 hours, and the enzyme activity was markedly reduced by 48 hours, a time when the wound was completely closed. 

RT–PCR, Sequence Analysis, and Expression of PI 3–Kinase mRNA in Healing Corneal Epithelium

To determine whether PI 3–kinase mRNA levels were altered during the course of epithelial wound repair, paired primers were used to conduct RT–PCR on total RNA isolated from corneal epithelium. The PCR product was examined on agarose gel by ethidium bromide staining. As shown in Figure 4 , RT–PCR on RNA isolated from rabbit corneal epithelium resulted in a PCR product of the expected size (270 bp). To confirm the identity of PI 3–kinase mRNA, the PCR product was cloned into pGEM-T vector, and the inserted cDNA fragment sequenced by the dideoxynucleotide chain–termination reaction. As seen in Figure 5 , the sequence analysis of the cDNA fragment from rabbit corneal epithelium revealed a high degree of homology to bovine (91%), human (89%), rat (88%), and mouse (87%) PI 3–kinase p85α subunits. A time course analysis of the expression of PI 3–kinase mRNA was performed using RT–PCR. An enhanced expression of PI 3–kinase mRNA was detectable at 24 hours post wounding (Fig. 6) . This increase in expression peaked at 48 hours and then gradually decreased to the basal value at day 7 after wounding. 

Western Blot Analysis of PI 3–Kinase during Wound Repair

To determine whether PI 3–kinase protein levels increased, coinciding with the increased mRNA levels, after wounding, a time-course analysis of the expression of PI 3–kinase protein was performed using western blot analysis. As seen in Figure 7 , the anti–PI 3–kinase antibody reacted with a single band of 85 kDa. There was an increase in the level of this protein in epithelium harvested at 24 hours after wounding. This increase peaked at 48 hours but returned to the basal value at day 5 post wounding. 

Expression and Localization of PI 3–Kinase mRNA during Corneal Epithelial Wound Repair Determined by In Situ Hybridization

To correlate PI 3–kinase expression with the process of wound repair, the presence of PI 3–kinase mRNA was detected by in situ hybridization in corneal epithelium treated with and without EGF. As shown in Figure 8 , there was a very low expression of PI 3–kinase mRNA in uninjured corneal epithelium. As early as 6 hours after wounding, there was a slight increase in PI 3–kinase mRNA. This increase peaked at 48 hours and then declined to reach basal level at about day 5 after wounding. Figure 8 also shows PI 3–kinase expression after the corneal epithelium was treated with EGF. The growth factor caused a further increase in PI 3–kinase expression at all time intervals as indicated by the intensity of the band. This stimulated increase in mRNA lasted longer than 48 hours before it returned to the basal value at day 5 after wounding. 

Discussion

In our previous work, we demonstrated that EGF stimulates PI 3–kinase activity in cultured corneal epithelial cells and that this effect is correlated with EGF-stimulated wound repair. ¹¹ In the present studies, we have extended these findings to rabbit corneal epithelium, in vivo. To minimize precorneal elimination and maximize continuous supply of EGF, we used semi-solid Carbopol gel to topically deliver EGF to the wounded eye. Our results show that in the untreated eye, the wounded epithelium progressively healed on its own, and the wound was approximately 75% closed over a period of 48 hours. This was not unexpected because the cornea is continuously bathed in tear film that can provide EGF to the wounded epithelium. The potential sources of EGF in tears include the conjunctiva and lacrimal gland, ^{15 16} and it has been reported that EGF concentration in tears increases after wounding of the corneal epithelium. ¹⁷ The data presented in this study show that exogenous application of EGF to the wounded corneal epithelium significantly reduced the time for wound repair, compared with the control epithelium not treated with EGF. The corneal wound was completely healed at 48 hours, a time significantly shorter than that needed by the untreated cornea. Our data on time course of wound repair are consistent with studies in which EGF was also reported to stimulate epithelial wound repair in primate and rabbit corneas. ^{8 18 19 20}  

PI 3–kinase is a lipid kinase that generates 3-phosphorylated phosphoinositides in cells stimulated with a variety of agonists, including EGF and platelet-derived growth factor. ² Although the physiological significance of these lipids remains elusive, increasing evidence suggests that they may function as intracellular second-messenger molecules in several cellular processes including cell growth and proliferation, cell differentiation, protection from apoptosis, actin–cytoskeleton rearrangement, membrane ruffling, and vesicle transport. ⁴ The data presented herein demonstrate that there is a marked increase in PI 3–kinase activity in the regenerating corneal epithelium. This increase in PI 3–kinase activity appeared to correlate with the time course of wound repair, except at 48 hours when the enzyme activity started to decline although the wound had not completely repaired yet. As mentioned above, after the creation of corneal wounds there is increased production of EGF in the tear film. Therefore, the observed increase in PI 3–kinase activity during reepithelialization of the untreated cornea is probably due to endogenous EGF present in the vicinity of the wound. Additional topical application of EGF to the wounded corneal epithelium further enhanced the PI 3–kinase activity. This increase in PI 3–kinase activity was peaked at 36 hours, a time when the epithelial wound had nearly healed. At 48 hours, when the epithelial wound was completely healed, the enzyme activity had significantly declined from its peak value. After this, the enzyme activity gradually decreased to the baseline over the course of the next several days (data not shown). These data could suggest that even though the wound appears to be closed, the cells probably are still undergoing the repair process. Therefore, wound closure may not necessarily indicate the end of wound healing. Using a cell culture model for corneal epithelial wound repair, we have previously shown that both EGF-induced PI 3–kinase activation and wound repair were inhibited when the cells were treated with wortmannin, a PI 3–kinase inhibitor, ¹¹ suggesting a causal relationship between the two EGF-induced effects. The data obtained from rabbit corneal epithelium, in vivo, also suggest a possible correlation between PI 3–kinase activation and corneal epithelial wound repair. There are several reports implicating PI 3–kinase in the regulation of cell growth and proliferation by receptor tyrosine kinases, ²¹ nonreceptor tyrosine kinases, ²² and cytokine receptors. ²³ More recently, PI 3–kinase was shown to induce migration in several epithelial cell lines, suggesting that PI 3–kinase may have a role in wound healing and tissue repair. ²⁴ In corneal endothelial cells, cell proliferation induced by fibroblast growth factor was inhibited by wortmannin and LY294002 (a PI 3–kinase inhibitor), again suggesting that PI 3–kinase might be involved in cell proliferation in corneal endothelial cells. ²⁵  

PI 3–kinase is composed of a catalytic 110-kDa protein (p110) and a regulatory 85-kDa protein (p85). ¹ The regulatory subunit contains two proline-rich motifs, two Src homology-2 (SH2) domains, and a domain that binds the catalytic subunit. ²⁶ When stimulated by growth factors, many receptors with tyrosine kinase activity can interact with the SH2 domain in the p85 subunit and result in recruitment and activation of PI 3–kinase. ²⁷ To date, five regulatory subunits of PI 3–kinase have been cloned, two 85-kDa proteins (p85α, p85β), two 55-kDa proteins (p55α, p55γ), and one 50-kDa protein (p50α). ²⁸ In the present work, we have demonstrated by RT–PCR and cloning experiments that PI 3–kinase mRNA is expressed in rabbit corneal epithelium. We used primers specific for the highly conserved region of p85 PI 3–kinase cDNA. The amplified 270-bp cDNA fragment was found to be highly homologous (>87%) to the published p85 PI 3–kinase cDNA sequences in bovines, humans, rats, and the mouse. ^{26 29 30} We found that expression of PI 3–kinase mRNA was low in the unwounded corneal epithelium. However, at 24 hours post wounding, the expression of PI 3–kinase mRNA was significantly increased to peak at 48 hours and then declined to the basal level as the wound repair was being completed. When analyzed by western blot analysis, we found increases in PI 3–kinase protein content that corresponded to the changes in PI 3–kinase mRNA during the course of wound repair. These findings strongly suggest that induction of PI 3–kinase mRNA is important in the process of corneal epithelial wound repair. Direct evidence for a possible role of PI 3–kinase in wound repair comes from our in situ hybridization studies. There was very little expression of PI 3–kinase mRNA in the unwounded epithelium; however, when the epithelium was wounded, we could detect a significant increase in PI 3–kinase expression as early as 12 hours after wounding. This expression continued to increase during the course of the wound repair and peaked at 48 hours post wounding. An important finding from these studies is that in tissues treated with EGF, there was a marked increase, at all times, in the expression of PI 3–kinase mRNA, which correlated with the time course of wound repair. It should be noted, however, that the expression remains elevated at points (>48 hours) when the wound has already been closed. This probably indicates that the repair process continues even after the wound has completely closed. It is known that the earliest response to wound trauma is cell migration from the periphery, followed by proliferation. At the time of complete wound closure, only a few layers of cells cover the wound area, and more cells need to be grown and stratified until the epithelium reaches the normal thickness (about seven layers). 

Although the work presented in this article demonstrates that PI 3–kinase might play a role in corneal epithelial wound repair, the mechanism or mechanisms by which PI 3–kinase exerts its effect remain to be determined. The calcium-independent protein kinase C isoforms were shown to be activated in vitro by PI 3,4-P₂ and PI 3,4,5-P₃. ⁵ Similarly, in some cell types, PI 3–kinase was required for activation of mitogen-activated protein (MAP) kinases, ⁵ the enzymes frequently implicated in cell proliferation. It has been reported that the lipid products of PI 3–kinase can associate with SH2 domains of phospholipase Cγ and modify its activity. ⁶ It is possible, therefore, that multiple signal transduction enzymes, including protein kinase C, phospholipase Cγ, and/or MAP kinase, might mediate the effect of PI 3–kinase on cell proliferation during corneal epithelial wound repair. 

In summary, we have provided evidence that PI 3–kinase is expressed in rabbit corneal epithelium. After wounding of the corneal epithelium, there is increased activation and expression of this enzyme, which correlates temporally with the process of wound repair. Topical application of EGF further stimulates the expression of PI 3–kinase with corresponding increase in wound repair. It can be concluded from these data that PI 3–kinase and its products may play a role in EGF-induced cell proliferation during corneal epithelial wound repair. The signaling pathway or pathways leading from activation of PI 3–kinase in corneal epithelium remain to be investigated. 

 Supported by Grant EY 05738 (RAA) from the National Institutes of Health, Bethesda, Maryland.

 Submitted for publication March 30, 1999; revised June 10, 1999; accepted June 28, 1999.

 Commercial relationships policy: N.

 

Figure 1.

 

Representative photographs of rabbit corneal epithelium undergoing wound repair with or without treatment with EGF. Wounds of uniform size were created in rabbit corneal epithelia using n-heptanol–soaked filter paper discs. After thoroughly washing the cornea with sterile saline, 2 μg EGF was topically applied to one eye every 8 hours. The other eye served as a control. To monitor the wound repair, the corneas were stained with fluorescein and photographed under filtered green light.

Figure 2.

 

Time course of the effect of EGF on corneal epithelial wound repair. Rabbit corneal epithelia were wounded and treated with EGF as in Figure 1 . From the photographs taken at different time intervals, the wound areas were digitized with an optical scanner and quantified with a computer program, SigmaScan. Values are mean ± SEM of three experiments with two rabbits used for each data point.

Figure 3.

 

Effect of EGF on PI 3–kinase activity during corneal epithelial wound repair. The epithelial tissues were scraped at different time intervals from corneas with or without treatment with EGF. The scraped tissues were processed for immunoprecipitation with anti–PI 3–kinase antibody and assayed for PI 3–kinase activity. Values are mean ± SEM from three experiments with two rabbits for each data point.

Figure 4.

 

Ethidium bromide–stained gel showing the expression of mRNA for PI 3–kinase. One microgram of RNA extracted from corneal epithelium, corneal epithelial cell line, and liver was reverse-transcribed, and the cDNA obtained was amplified by PCR using primers specific for PI 3–kinase. Lanes 1 and 2: Rabbit corneal epithelium harvested from two different eyes. Lane 3: Immortalized rabbit corneal epithelial cells. Lane 4: Liver. M, 2,000–bp DNA ladder.

Figure 5.

 

Sequence homology between the cloned p85 PI 3–kinase cDNA segment from rabbit corneal epithelium and the corresponding published sequence for bovine, human, rat, and mouse p85 PI 3–kinase. The letters indicate nonidentity between the cDNA sequences.

Figure 6.

 

Upper panel: RT–PCR analysis of RNA isolated from unwounded (0 time) rabbit corneal epithelium and epithelium harvested at different times post wounding. Arrow indicates the PI 3–kinase PCR product. The PCR products were visualized on agarose gels stained with ethidium bromide. C, cloned PI 3–kinase segment. M, 2,000–bp DNA ladder. Lower panel: Amplification of GAPDH generated approximately equal amounts of RT–PCR product, confirming that equal amounts of RNA were used from epithelial tissues for RT–PCR analysis. We obtained similar data in three independent experiments.

Figure 7.

 

Western blot of unwounded (0 time) rabbit corneal epithelium and corneal epithelium harvested at different times post wounding. The blot was reacted with anti–PI 3–kinase (p85) antibody as described in the Materials and Methods section. Molecular masses determined from standard proteins are shown on the left. The results are from a representative experiment repeated three times.

Figure 8.

 

Effect of EGF on expression of PI 3–kinase mRNA in corneal epithelium during wound repair. In situ hybridization of PI 3–kinase mRNA in rabbit corneal epithelium during the course of wound repair is shown. Antisense PI 3–kinase cRNA (digoxigenin-labeled) was hybridized with corneal cryostat sections obtained from wounded and unwounded (0 time) corneas treated with or without EGF for different time intervals. The top right panel represents cryosection from unwounded cornea probed with antisense PI 3–kinase cRNA. The top left panel is from unwounded cornea probed with sense PI 3–kinase cRNA. The cells seen above the corneal epithelial layer are from muscle tissue used to sandwich the cornea before cryosectioning. The leading edge of wounded epithelium can be seen as a narrower region in the sections at day 0.25. These results were typical of those in two independent experiments.