Up-Regulation of Urokinase-Type Plasminogen Activator in Corneal Epithelial Cells Induced by Wounding

Abstract

purpose. To investigate the possible role of urokinase-type plasminogen activator (uPA) in corneal epithelial wound healing by examining its expression both in the rabbit corneal epithelium in situ and in rabbit corneal epithelial (RCE) cells in vitro.

methods. The rabbit cornea was subjected to mechanical wounding, and frozen sections of the tissue were subsequently prepared and subjected both to immunostaining with antibodies to uPA and to in situ zymography for the detection of PA activity. RCE cell monolayers were also subjected to scrape wounding, after which they were immunostained for uPA. The amounts of uPA protein in the culture medium and of uPA mRNA in cell lysates were also determined by enzyme-linked immunosorbent assay and by reverse transcription and real-time quantitative polymerase chain reaction analysis, respectively.

results. Immunostaining and in situ zymography of the wounded cornea revealed that uPA was restricted to the leading edge of the migrating corneal epithelium. In contrast, tissue-type PA was expressed throughout the corneal epithelium. Scraping of RCE cell monolayers induced the expression of uPA in the migrating cells at the wound edge. The amount of uPA in the culture medium of RCE cells increased with the number of scrape wounds applied. Wounding also induced a time-dependent increase in the abundance of uPA mRNA in the cell monolayers. The migration of RCE cells was inhibited by antibodies to uPA.

conclusions. Mechanical wounding induces up-regulation of uPA at both the protein and mRNA levels in corneal epithelial cells. uPA may thus contribute to epithelial cell migration during corneal epithelial wound healing.

The corneal epithelium serves as an important barrier between the eye and the external environment. Although it is repeatedly exposed to various types of insults, the corneal epithelium normally undergoes rapid healing mediated by a well-orchestrated series of cellular events that result in the complete resurfacing of defects. Various cytokines, growth factors, and extracellular matrix proteins modulate the healing of corneal epithelial wounds. ^{1 2 3 4} In vascularized tissues such as the skin, wound healing is initiated by the leakage of blood-derived components into the wound and the consequent formation of a temporary seal. The subsequent dissolution of the fibrin clot requires the generation of plasmin from plasminogen by plasminogen activator (PA)–mediated cleavage. Although the cornea is not vascularized, fibrin has been detected at the site of corneal wounds, ^{5 6} and both PA and plasmin activities have been detected in tear fluid. ^{7 8 9 10}  

Two different types of PA have been identified, tissue-type (tPA) and urokinase-type (uPA), the biological activities of which are similar. Both enzymes thus degrade fibrin clots in vivo. However, whereas tPA exhibits a high affinity for fibrin and primarily contributes to fibrinolysis in the circulation, uPA functions in cell adhesion, cell migration, and tissue remodeling. ^{11 12} During corneal wound healing, epithelial cells at the wound margin migrate into and resurface the wounded area. To elucidate the mechanism of epithelial cell migration during corneal wound healing, we have investigated the possible role of PA in this process. We thus previously showed that the extent of epithelial migration after wounding of the rabbit cornea in an organ culture system was correlated with the activity of uPA in the culture medium. ¹³ Furthermore, the addition of plasmin inhibitors to the organ culture system blocked epithelial migration. ¹³ uPA, but not tPA, activity has also been detected at the leading edge of the migrating corneal epithelium after a scrape injury. ^{14 15}  

To delineate further the importance of uPA in corneal epithelial migration during wound healing, we have now characterized its expression, at both the transcript and protein levels, in migrating corneal epithelial cells both in situ and in culture. We also examined the effect of antibodies to uPA on corneal epithelial cell migration in vitro. 

Materials and Methods

Animals

Male Japanese white rabbits (2.5-3.0 kg) were obtained from Japan Laboratory Animals (Tokyo, Japan) and were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Purification of uPA from Rabbit Urine

The high molecular weight (HMW) form of uPA (45 kDa) was purified from rabbit urine as described previously. ¹⁶ Protein concentration was determined by the method of Lowry et al. ¹⁷ with bovine serum albumin (BSA) as standard. The activity of uPA was measured by the fibrin plate method ¹⁸ with purified human HMW uPA as standard (JCR Pharmaceuticals, Kobe, Japan). SDS-polyacrylamide gel electrophoresis revealed that the purified rabbit HMW uPA migrated as a single band. Its specific activity was determined to be 130,000 IU per milligram of protein. 

Monoclonal Antibodies to Rabbit uPA

Monoclonal antibodies (mAbs) to rabbit uPA were generated by injection of female BALB/c mice with purified rabbit HMW uPA. Two of the antibodies (U-108 and U-109) were used for immunostaining, and two (U-109 and U-118) were used for quantitation of uPA by enzyme-linked immunosorbent assay (ELISA). The antibody U-120 was examined for its effects on corneal epithelial cell migration. For ELISA, U-109 was biotinylated with the use of N-hydroxysuccinimide biotin (Pierce, Rockford, IL). ¹⁹  

Wounding of the Rabbit Corneal Epithelium In Vivo

Rabbits were anesthetized by a subcutaneous injection of ketamine hydrochloride (30 mg/kg of body mass; Sankyo, Tokyo, Japan) and xylazine hydrochloride (2 mg/kg; Bayer Medical, Tokyo, Japan) and by topical instillation of 1 drop of oxybuprocaine (0.4%; Santen, Osaka, Japan). The corneal epithelium was then wounded with a trephine (diameter, 5.5 mm); the epithelium within the circle was removed with a small scalpel, leaving the basement membrane intact. Seven to 8 hours after injury, the animals were killed with an overdose of pentobarbital, and the injured corneas were removed for the preparation of cryostat sections. 

Immunohistofluorescence Microscopy

Corneal tissue in tissue-embedding medium (Shiraimatsu, Osaka, Japan) was frozen in liquid nitrogen. Cryosections (8 μm) were prepared, fixed for 15 minutes at room temperature with 4% paraformaldehyde, and permeabilized for 5 minutes with 1% Triton X-100. After two washes with phosphate-buffered saline (PBS), the sections were incubated for 30 minutes at room temperature with 1.5% horse serum in PBS to prevent nonspecific staining. The sections were then incubated for 45 minutes at room temperature with a mixture of the mAbs U-108 (10 μg/mL) and U-109 (10 μg/mL) to rabbit uPA, with a mAb to tPA (10 μg/mL; Cosmobio, Tokyo, Japan), or with normal mouse immunoglobulin G1 (IgG1; 10 μg/mL; Southern Biotechnology, Birmingham, AL) as a negative control. After two washes with PBS, the sections were incubated for 20 minutes at room temperature with fluorochrome–conjugated goat antibodies to mouse IgG (5 μg/mL; AlexaFluor 488; Molecular Probes, Eugene, OR). For double immunostaining to detect the epithelial basement membrane, the sections were subsequently washed twice with PBS, incubated for 45 minutes at room temperature with sheep antibodies to human laminin (3 μg/mL; Binding Site, Birmingham, UK), washed an additional two times with PBS, and incubated for 20 minutes with fluorochrome–conjugated donkey antibodies to sheep IgG (10 μg/mL; AlexaFluor 594; Molecular Probes). After a final four washes with PBS, coverslips were applied to each section in antifade mounting medium (Vectashield; Vector Laboratories, Burlingame, CA). The sections were observed with an epifluorescence microscope (BX-50; Olympus, Tokyo, Japan) and photographed. 

In Situ Zymography

To detect PA activity in sections of rabbit cornea, we performed in situ zymography with casein-agarose, as described previously. ²⁰ The casein overlay mixture (500 μL) consisted of 105 μL of PBS(+) (PBS supplemented with 0.9 mM CaCl₂ and 1 mM MgCl₂), 200 μL of 8% skim milk (Invitrogen/Gibco, Grand Island, NY) in PBS(+), 175 μL of 2.5% agar in PBS(+), and 20 μL rabbit plasminogen (1 mg/mL; Sigma-Aldrich, St. Louis, MO). Cryosections (8 μm) of rabbit cornea on glass slides were covered with 100 μL of the casein overlay mixture at 55°C. After the application of coverslips, the slides were placed in a humidified chamber for 20 hours at 25°C. Casein hydrolysis on the slides was detected by dark-field microscopy. To identify the sites of casein hydrolysis on the sections, we also photographed phase-contrast images of the corresponding dark-field areas. Control experiments were performed in the absence of plasminogen to determine whether the casein degradation was plasmin dependent. For determination of casein hydrolysis by uPA or tPA, experiments were performed with the addition of 1 mM amiloride (Sigma-Aldrich), a specific inhibitor of uPA, ²¹ to the overlay mixture. Goat polyclonal antibodies (IgG) to either human uPA or tPA (American Diagnostics, Greenwich, CT) were also included in the overlay mixture at final concentrations of 200 μg/mL for specific inhibition of uPA or tPA activity, respectively; normal goat IgG (Chemicon, Temecula, CA) was used as a negative control. 

Rabbit Corneal Epithelial Cell Culture

A single-cell suspension of rabbit corneal epithelial (RCE) cells was prepared as described previously. ^{22 23} The cells were plated in the designated culture dishes and cultured until they achieved confluence in growth medium (RCGM2; Kurabo, Osaka, Japan), prepared by the addition of bovine insulin (5 μg/mL), murine epidermal growth factor (10 ng/mL), hydrocortisone (0.5 ng/mL), gentamicin (50 μg/mL), amphotericin B (50 ng/mL), and bovine pituitary gland extract (0.4%, vol/vol) to RCBM2 basal medium containing 0.03 mM calcium. The medium was changed every 2 to 3 days. 

Wounding of Cell Monolayers and Immunofluorescence Staining

Confluent monolayers of RCE cells cultured in eight-well chamber slides (Laboratory-Tek II; Nalge Nunc, Rochester, NY) were wounded by scraping in a straight line with a plastic cell scraper (Corning, Corning, NY). The cells were washed three times with PBS and then either fixed immediately for 15 minutes on ice with 4% paraformaldehyde or first cultured for 20 hours in RCBM2 basal medium. The fixed cells were permeabilized with 1% Triton X-100 for 5 minutes, washed twice in PBS, and incubated for 30 minutes at room temperature with 1.5% horse serum in PBS. The cells were then incubated for 30 minutes at room temperature with a mixture of mAbs to rabbit uPA (U-108 and U-109), each at a concentration of 10 μg/mL, or with normal mouse IgG1 (10 μg/mL) as a negative control. After two washes with PBS, the cells were incubated for 30 minutes at room temperature with fluorescein isothiocyanate (FITC)–conjugated sheep antibodies to mouse IgG (10 μg/mL; Cappel, Costa Mesa, CA) and then washed an additional four times with PBS. Coverslips were finally applied to the sections, which were then observed with an epifluorescence microscope as described earlier. 

Multiple Wounding of Cell Monolayers for Quantitation of uPA

Multiple wounds were applied to cell monolayers as previously described. ²⁴ In brief, confluent RCE cell monolayers in six-well plates (Corning) were wounded by scraping in a straight line from one side of a well to the opposite side with a plastic hair comb (width, 1.4 cm). One, two, or four such scrape wounds (equally spaced) were applied by rotating the plate. After a wash with Hanks’ balanced salt solution, the cells were cultured for 24 hours in 3 mL of RCBM2 basal medium. For quantitation of uPA by ELISA, portions (100 μL) of the culture medium were collected, mixed with an equal volume of BSA (40 mg/mL) in 0.1% Tween 20, and stored at −30°C. After the cells were washed with Hanks’ balanced salt solution, they were harvested with a cell scraper into 0.5 mL of a solution containing 1% Triton X-100 and 0.4% Nonidet P-40. The protein concentration of the cell lysates was determined with the bicinchoninic acid (BCA) protein assay reagent (Pierce). 

Quantitation of uPA by ELISA

Purified rabbit HMW uPA or test samples diluted with PBS containing 0.05% Tween 20 (PBS-T) were added to the wells of a 96-well microtiter plate that had been coated consecutively with mAb U-118 to uPA and with a blocking reagent (Block Ace; Dainippon, Osaka, Japan). After incubation for 2 hours at 25°C, the wells were washed with PBS-T, incubated for an additional 2 hours at 25°C with biotinylated mAb U-109 to uPA, washed again, and then incubated for 1 hour at 25°C with streptavidin–horseradish peroxidase conjugate (Prozyme, San Leandro, CA). The wells were washed and then 3,3′,5,5′-tetramethylbenzidine (Moss, Inc., Pasadena, MD) was added to each. After incubation of the plate for 30 minutes at 25°C, the reaction was stopped by the addition of 1 M sulfuric acid, and absorbance was measured at 450 nm. The concentration of uPA in samples was determined by reference to a standard curve generated with the use of purified rabbit HMW uPA. 

Cloning of Rabbit uPA cDNA

We isolated a full-length rabbit uPA cDNA (the sequence of which has been deposited in GenBank under the accession no. AB087224; http://www.ncbi.nlm.nih.gov/Genbank; provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) by rapid amplification of cDNA ends (RACE). ²⁵ The 5′ and 3′ ends of the cDNA were amplified with the use of a cDNA amplification kit (SMART RACE; Clontech, Palo Alto, CA) and total RNA extracted from RCE cells. The 5′ RACE and 3′ RACE primers (5′-GGTCTTCTCTGGTGGTCTGGGTTCCTGC-3′ and 5′-GCAGGAACCCAGACCACCAGAGAAGACC-3′, respectively) were designed on the basis of available partial rabbit uPA cDNA sequences (GenBank accession nos. AF097647 and AF069711). After we had completed cloning of the rabbit uPA cDNA, the full-length cDNA sequence was also released by Sugiki et al. (GenBank accession no. AY029517). ²⁶ Both cDNA sequences are identical, although our cDNA contains an additional 8 bp at the 5′ end. 

RT and Real-Time Quantitative PCR

For determination of the abundance of uPA mRNA, we performed reverse transcription (RT) and real-time quantitative polymerase chain reaction (PCR) analysis with the a sequence detection system and Taq probes (Prism 7900HT and TaqMan; Applied Biosystems Inc. [ABI], Foster City, CA). The rabbit uPA–specific primers and TaqMan probe were designed with the use of a computer (Primer Express software; Applied Biosystems) on the basis of the sequence of the full-length rabbit uPA cDNA obtained as described earlier. They were as follows: 5′-CCA TCC CGG TCC ATA CAG ACT-3′ (forward primer), 5′-TCA CAG CTT GTG CCA AAA TTG-3′ (reverse primer), and 5′-TTG CCT GCC CCC ATG GAA TGC-3′ labeled at the 5′ end with 6-carboxyfluorescein and at the 3′ end with 6-carboxytetramethyl rhodamine (TaqMan probe). Primers and a probe (TaqMan) for 18S ribosomal RNA were obtained from Applied Biosystems for use as an endogenous control. 

Confluent monolayers of RCE cells were wounded four times with a hair comb as described earlier, washed twice with RCBM2 medium, and further cultured in the same medium. Nonwounded monolayers were washed and cultured as a control. Total RNA was isolated from the cells at various time points with the use of extraction reagent (Isogen; Nippon Gene, Tokyo, Japan). First-strand cDNA was synthesized from 1 μg of the total RNA with random primers and murine leukemia virus reverse transcriptase (Applied Biosystems). PCR was then performed with 5% of the resultant cDNA in a final volume of 50 μL containing 900 nM of each primer, 250 nM probe (TaqMan), and 0.5× PCR reagent mix (TaqMan Universal PCR Master Mix; Applied Biosystems). The amplification protocol comprised incubations at 50°C for 2 minutes and 95°C for 10 minutes followed by 40 cycles of incubation at 95°C for 15 seconds and 60°C for 1 minute. The PCR cycle number (C_T) at which fluorescence emission reached a threshold value above the baseline emission was used to quantitate the original amount of uPA mRNA, which was normalized by the amount of 18S ribosomal RNA. ²⁷ The validity of the use of 18S ribosomal RNA as an endogenous calibration standard was verified by showing that the amplification efficiencies for uPA mRNA and 18S ribosomal RNA were similar (data not shown). 

Migration Assay

To examine the effect of antibodies to uPA on RCE cell migration, we performed wound assays as described previously. ²⁸ In brief, confluent monolayers of RCE cells in 48-well plates were wounded with a cell scraper, washed with PBS, and then incubated for 20 hours in the presence of mAb U-120 to uPA or of normal mouse IgG (Zymed, South San Francisco, CA) as a negative control, each at a concentration of 100 μg/mL. The distance from the wound margin to the leading edge of the migrating RCE cells was determined. 

Statistical Analysis

ELISA data for uPA abundance are presented as means ± SEM of triplicates and were analyzed by the Dunnett multiple comparison test. RCE cell migration data are presented as the mean ± SEM of four replicates and were analyzed by the Tukey multiple comparison test. P < 0.05 was considered statistically significant. 

Results

With the use of immunofluorescence analysis, we first examined whether uPA or tPA is expressed in actively migrating corneal epithelial cells in vivo after surgical removal of a portion of the corneal epithelium in rabbits. Seven to 8 hours after the epithelium was removed, the remaining epithelial cells were apparent migrating as a sheet over the denuded surface (Figs. 1A 1D 1G) . To confirm the presence of a basement membrane in the wounded area, we examined the localization of laminin. Specific fluorescence for laminin was detected in the expected position for the basement membrane regardless of whether the epithelium was intact, removed, or resurfacing (Figs. 1C 1F 1I) . Staining of the cornea with antibodies to uPA revealed the presence of specific immunoreactivity at the leading edge of the migrating epithelial cells but not in the nonwounded portions of the epithelium (Fig. 1B) . In contrast, specific but weak fluorescence for tPA was observed associated with all epithelial cells regardless of whether they were migrating or stationary (Fig. 1E) . Almost no fluorescence was detected in corneal sections stained with normal mouse IgG1 as a control (Fig. 1H) . Although corneal keratocytes beneath the site of epithelial injury were lost, presumably as a result of apoptosis, ²⁹ those remaining were also weakly stained with antibodies to tPA (Fig. 1E) but not with antibodies to uPA (Fig. 1B) . We also examined cryosections of normal unwounded rabbit cornea. Virtually no fluorescence was detected with antibodies to uPA or with normal mouse IgG1. In contrast, a low level of tPA immunoreactivity was observed associated with both epithelial cells and keratocytes (data not shown). These results thus showed that uPA, but not tPA, was upregulated at the leading edge of the migrating epithelium during corneal wound healing, suggesting that uPA may play a role in the migration of corneal epithelial cells. 

To examine further the possible roles of uPA and tPA in corneal epithelial migration, we performed in situ casein zymography with cryosections of wounded rabbit cornea. In the absence of added plasminogen, hydrolysis of casein was not detected in the overlay mixture applied to the sections of wounded cornea (Figs. 2A 2B) . However, in the presence of exogenous plasminogen in the overlay mixture, hydrolysis of casein was apparent at the leading edge of the migrating epithelial cells (Figs. 2C 2D) ; it was not observed in the nonwounded regions of the epithelium. The observed casein hydrolysis was thus attributable to the conversion of plasminogen to plasmin by endogenous PA present at the leading edge of the migrating corneal epithelium. The casein hydrolysis associated with epithelial migration was completely inhibited by amiloride (Figs. 2E 2F) , a specific inhibitor of uPA. ²¹ It was also inhibited by goat antibodies to uPA (Figs. 2I 2J) , but not by either goat antibodies to tPA (Figs. 2K 2L) or normal goat IgG (Figs. 2G 2H) . These results thus demonstrated that the casein hydrolysis apparent at the leading edge of the migrating epithelial cells in the wounded rabbit cornea was attributable to the production of uPA. 

We next investigated whether uPA is expressed by actively migrating corneal epithelial cells in vitro. Confluent monolayers of RCE cells were wounded with a cell scraper and were then fixed either immediately or after further incubation for 20 hours. The fixed cells were subjected to immunofluorescence staining with antibodies to uPA. Although only a low level of uPA expression was detected in a small proportion of cells either at the wound edge or in the nonwounded regions of the monolayer immediately after injury (Figs. 3A 3C) , a marked increase in the abundance of uPA was apparent in the actively migrating epithelial cells at the wound edge 20 hours after injury (Figs. 3B 3D) . 

The effect of mechanical wounding on the production of uPA by confluent monolayers of RCE cells was further investigated by subjecting the cells to one, two, or four scrapings with a comb (Fig. 4A) and then measuring the amount of uPA in the culture medium after 24 hours by ELISA. The medium of nonwounded monolayers cultured for 24 hours contained approximately 200 ng of uPA per milligram of total cell protein. The amount of uPA in the medium increased as the number of scrapings increased (Fig. 4B) , with the increase in the production of uPA being statistically significant even after only one scraping. 

Next, we examined the effect of mechanical wounding on the abundance of uPA mRNA in cultured RCE cells by RT and real-time PCR analysis. The amount of uPA mRNA in RCE cells was increased as early as 30 minutes after multiple scraping of confluent monolayers and was further increased at 1 hour, compared with the amount in cells subjected to control manipulations (Fig. 5) . 

Finally, to assess the role of uPA in the migration of corneal epithelial cells, we examined the effect of antibodies to uPA on RCE cell migration in the in vitro wounding assay. The mAb U-120 to uPA significantly inhibited the migration of RCE cells, whereas control IgG had no effect (Fig. 6) . 

Discussion

With the use of immunohistofluorescence analysis, we have shown that uPA is expressed at the leading edge of the migrating epithelium after wounding of the rabbit cornea. In contrast, tPA was constitutively expressed at a low level throughout the migrating and nonmigrating portions of the corneal epithelium. Furthermore, in situ casein zymography revealed that casein hydrolysis was restricted to the leading edge of the migrating corneal epithelium and that this proteolytic activity was inhibited by antibodies to uPA and by a specific uPA inhibitor (amiloride), but not by antibodies to tPA. We further showed by immunostaining of primary cultured RCE cells that the expression of uPA was up-regulated in the migrating cells at the edge of a scrape wound; this up-regulation was also apparent from an increase in the release of uPA into the culture medium and an increase in the abundance of uPA mRNA in the cells. Our present results thus demonstrate that uPA, but not tPA, is induced in the migrating epithelial cells during corneal epithelial wound healing, and they are consistent with our previous observations with the rabbit cornea in organ culture. ¹³ In the latter study, we showed by fibrin zymography that PA activity released from a small block of rabbit cornea into the culture medium is attributable to uPA (not tPA). Other investigators have also described the localization of uPA activity at the leading edge of the migrating corneal epithelium after corneal wounding. ^{14 15} Moreover, we showed that the migration of corneal epithelial cells in vitro was inhibited by antibodies to uPA. Together, these observations indicate that uPA, but not tPA, is up-regulated in corneal epithelial cells by wounding and may play an important role in epithelial cell migration during corneal wound healing. A role for uPA in the migration of keratinocytes, vascular smooth muscle cells, and vascular endothelial cells has also been suggested. ^{30 31 32 33 34 35}  

Fibronectin is also an important player in epithelial wound healing in both the cornea and skin. ^{1 5 36 37 38 39} It thus provides a provisional matrix to support the attachment and migration of epithelial cells over the denuded surface of the cornea during the early stages of wound healing. It appears in the region of the stroma affected by the wound and is expressed by all the migrating epithelial cells in the injured cornea. ^{5 36 37} Furthermore, the appearance and disappearance of fibronectin are coordinated with corresponding changes in the expression of integrins during corneal epithelial wound healing. ³⁷ We have shown that the fibronectin-integrin system is important for the attachment of corneal epithelial cells to the underlying extracellular matrix ⁴⁰ ; the addition of a peptide (GRGDSP) derived from the cell-binding domain of fibronectin thus inhibits the attachment of corneal epithelial cells to a fibronectin matrix in vitro. ⁴⁰ These observations suggest to us that fibronectin degradation may also be important during corneal epithelial migration. The ability of epithelial cells to migrate depends on cycles of attachment to the matrix at their leading edge and detachment at their rear. Degradation of fibronectin may thus contribute to cell detachment. Our present results have revealed that the expression of uPA is up-regulated specifically at the leading edge of the migrating epithelium in the wounded rabbit cornea. The consequent generation of plasmin in this region of the wound may therefore result in the fibronectin degradation required for corneal epithelial migration. 

Recent studies have implicated uPA in various biological processes including cell adhesion, cell migration, and tissue remodeling. ^{11 12} In addition to its enzymatic activity, uPA triggers intracellular signaling systems essential for cell migration through its binding to specific cell surface receptors. ^{41 42 43 44 45} These receptors have been detected in various cell types including keratinocytes, ^{30 31 32} vascular smooth muscle cells, ³³ and vascular endothelial cells. ³⁵ Our preliminary results also indicate that wounding increases the abundance of uPA receptor mRNA in cultured RCE cells (Watanabe M, et al., unpublished data, 2002). The action of uPA in corneal epithelial wound healing may thus be mediated in part by the uPA receptor expressed on the surface of epithelial cells. 

 Supported in part by a Grant-in-Aid for Scientific Research A4207070 from the Japan Society for the Promotion of Science (TN).

 Submitted for publication December 13, 2002; revised March 10, 2003; accepted March 20, 2003.

 Disclosure: M. Watanabe, Kowa Co. Ltd. (E); W. Yano, Kowa Co. Ltd. (E); S. Kondo, Kowa Co. Ltd. (E); Y. Hattori, Kowa Co. Ltd. (E); N. Yamada, None; R. Yanai, None; T. Nishida, None

 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.

 

Figure 1.

 

Immunohistofluorescence staining of uPA and tPA in the wounded rabbit corneal epithelium. Sections of rabbit cornea isolated 7 to 8 hours after epithelial wounding were subjected to immunostaining with mouse antibodies to uPA (B) or to tPA (E) or with normal mouse IgG1 as a negative control (H). The epithelial basement membrane was also detected by staining of the respective sections with antibodies to laminin (C, F, I). Phase-contrast micrographs of the same fields are shown in (A), (D), and (G), respectively. (A–H, arrowheads) Leading edge of the migrating epithelium. Bars, 100 μm.

Figure 2.

 

Detection of uPA activity at the leading edge of the migrating epithelium in the wounded rabbit cornea by in situ zymography. Sections of rabbit cornea isolated 7 to 8 hours after wounding of the corneal epithelium were overlaid with a casein-agarose mixture, without (A, B) or with (C–L) added plasminogen. The overlay mixture also contained amiloride (E, F), normal goat IgG (G, H), or goat antibodies to uPA (I, J) or tPA (K, L). The same fields were observed by phase-contrast microscopy (left) and dark-field microscopy (right). Arrows: leading edge of the migrating corneal epithelium. Bars, 300 μm.

Figure 3.

 

Immunofluorescence staining for uPA in wounded monolayers of RCE cells. Confluent monolayers of RCE cells were wounded with a cell scraper, and the cells were fixed and stained with antibodies to uPA immediately (A) or 20 hours later (B). The same respective fields were also observed by phase-contrast microscopy (C, D). Bars, 50 μm.

Figure 4.

 

Relation between the number of scrape wounds and the production of uPA by confluent monolayers of RCE cells. (A) Cell monolayers in six-well plates were left unwounded (a) or were wounded by scraping with a comb once (b), twice (c), or four times (d) and photographed. (B) Cell monolayers were treated as in (A), and, after 24 hours, the amount of uPA in the culture medium was measured by ELISA and normalized by the protein content of the remaining cells. Data are means ± SEM and are representative of three independent experiments. ***P < 0.001 versus nonwounded monolayers.

Figure 5.

 

Time course of the abundance of uPA mRNA in RCE cell monolayers after scrape wounding. Cell monolayers were subjected either to four scrapings with a comb (▪) or to control manipulations (□), and the amount of uPA mRNA in the cells was determined at the indicated times thereafter by RT and real-time quantitative PCR. Data are representative of four independent experiments.

Figure 6.

 

Effect of antibodies to uPA on RCE cell migration. Confluent monolayers of RCE cells were wounded with a cell scraper and then cultured for 20 hours in the presence of an mAb to uPA or normal mouse IgG (control), each at a concentration of 100 μg/mL. The cells were photographed and the migration distance was determined. Data are means ± SEM and are representative of four independent experiments. **P < 0.01, ***P < 0.001.

The authors thank Ayako Kadowaki for technical assistance.