Promotion of Corneal Epithelial Wound Healing In Vitro and In Vivo by Annexin A5

Abstract

purpose. To investigate the effect of annexin A5, a calcium-dependent phospholipid-binding protein, on corneal epithelial wound healing.

methods. The effect of annexin A5 on migration of rabbit corneal epithelial (RCE) cells in vitro was examined in scrape-wounded cell monolayers. The effect of annexin A5 on the release of urokinase-type plasminogen activator (uPA) from cultured RCE cells was determined by zymography, fluorogenic assay of PA activity, and enzyme-linked immunosorbent assay. The proliferation of RCE cells was assessed by measurement of [³H]thymidine incorporation. The effect of annexin A5 on corneal wound closure in rabbits was investigated after removal of the corneal epithelium, either by exposure to iodine vapor or surgically. Eye drops containing annexin A5 were instilled into one eye and vehicle into the other. The area of the epithelial defect was measured at various times after wounding, and the healing rate was calculated by linear regression analysis.

results. Annexin A5 significantly promoted the migration of RCE cells in a wounded monolayer. However, annexin A5 had no effect on RCE cell proliferation. Annexin A5 also increased the release of uPA both from wounded RCE cell monolayers and from nonwounded semiconfluent RCE cells. In both models of corneal wound closure, the healing rate was significantly increased by instillation of eye drops containing annexin A5 compared with that apparent in the eyes that received vehicle.

conclusions. Annexin A5 promoted corneal epithelial wound healing both in vitro and in vivo. Upregulation of uPA release from corneal epithelial cells may contribute to this effect of annexin A5.

Annexins are a family of proteins that bind to phospholipids in a calcium-dependent manner. ^{1 2 3 4} The structure of annexins is characterized by a conserved carboxyl-terminal core, formed by either four or eight homologous repeats of ∼70 amino acids and a diverse amino-terminal domain. Annexins are present in a wide variety of species, with 13 distinct types (A1–A13) having been identified in mammals. Annexin A5, which has a molecular mass of 34 kDa, was first identified as an anticoagulant protein in human placenta and termed calphobindin-I. ⁵ Annexin A5 binds preferentially to negatively charged phospholipids, especially phosphatidylserine, in a calcium-dependent manner. ⁶  

Various biological actions of annexin A5, including its anticoagulant effect, ⁵ inhibition of both phospholipase A₂ (see Refs. ⁷ , ⁸ ) and protein kinase C, ^{9 10} and calcium channel activity in cell membranes, ^{11 12 13 14} have been described. Examination of the tissue distribution of annexin A5 in the rat revealed it to be present in the cornea as well as in other tissues, including lung, kidney, and testis. ¹⁵ We have also detected annexin A5 in tear fluid of humans and rabbits (Watanabe M, et al., unpublished data, 2005). However, the physiological functions of this protein remain unclear. We have shown that human annexin A5 (calphobindin-I) promotes both the migration of normal human skin keratinocytes in a cell-wounding assay in vitro, ¹⁶ as well as the re-epithelialization of dermal wounds in rats, ¹⁶ suggesting that annexin A5 may promote wound healing in the skin. 

The cornea is directly exposed to the external environment. The corneal epithelium thus functions as a barrier to maintain the integrity of the ocular surface, with corneal injury triggering the immediate re-epithelialization of the wound. The first step in corneal epithelial wound healing is coverage of the damaged area mediated by migration of the surrounding epithelial cells. These cells then undergo proliferation and differentiation to restore the normal epithelium. ¹⁷ The migration of corneal epithelial cells during wound repair involves detachment of the cell rear from the extracellular matrix (ECM) and adhesion of the cell front to the matrix. ^{17 18 19} Urokinase-type plasminogen activator (uPA) has been implicated in this cycle of cell detachment and attachment during corneal epithelial cell migration. ^{20 21} This protein is thought to promote the migration of various cell types by binding to its cell surface receptor and activating an intracellular signaling pathway in a manner independent of its proteolytic activity. ^{22 23 24 25 26} The presence of uPA at the leading edge of the migrating corneal epithelium has been demonstrated by in situ zymography. ^{20 21} Actively migrating corneal epithelial cells in vitro also secrete uPA into the culture medium, ²⁷ and the addition of a proteinase inhibitor was found to reduce the extent of corneal epithelial migration in an organ culture system of the cornea. ²⁷ We recently showed that mechanical wounding induces upregulation of uPA expression, without affecting the expression of tissue-type plasminogen activator (tPA), in corneal epithelial cells, and that antibodies to uPA inhibits the migration of these cells. ²⁸ These results thus support the notion that uPA participates in corneal epithelial wound healing. 

We have now investigated the effects of human annexin A5 on corneal epithelial wound healing, both in an in vitro assay of rabbit corneal epithelial (RCE) cells as well as in two in vivo rabbit models. We also examined whether annexin A5 stimulates the release of uPA from wounded RCE cell monolayers. 

Materials and Methods

Animals

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

Primary RCE Cell Culture

RCE cells were isolated as described previously. ^{29 30} The cells were cultured in rabbit corneal basal medium (RCBM; Kurabo, Osaka, Japan) supplemented with insulin (5 μg/mL), murine epidermal growth factor (EGF, 10 ng/mL), hydrocortisone (500 ng/mL), 0.4% (vol/vol) bovine pituitary extract, gentamicin (50 μg/mL), and amphotericin B (50 ng/mL). The Ca²⁺ concentration of RCBM is 0.03 mM. The cells were maintained in 24-well (6.4 × 10³ cells per well) or 48-well (2.4 × 10⁴ cells per well) culture plates for the designated periods. 

Assay of RCE Cell Migration In Vitro

The effect of annexin A5 on RCE cell migration was determined in an in vitro wound assay as described previously. ^{16 28} In brief, confluent monolayers of RCE cells in 48-well plates were wounded with a cell scraper, washed with phosphate-buffered saline (PBS), and cultured for 24 hours in RCBM containing various concentrations of recombinant human annexin A5 (expressed in and purified from yeast; Chemo-Sero-Therapeutic Research Institute, Kumamoto, Japan) in the absence or presence of a mouse monoclonal antibody to human annexin A5 (A180). ⁵ For reference, the wound area was marked with two or three dots on the external surface of the bottom of the plate with a black marker pen. The area marked by the reference dots was examined with the use of a phase-contrast microscope (IX70; Olympus, Tokyo, Japan) and photographed, both immediately and 24 hours after wounding, to determine the extent of RCE cell migration into the wound area. The distance from the wound’s margin to the leading edge of the migrating RCE cells was measured. The culture supernatants were also collected and stored at −80°C for subsequent determination of the concentration of uPA with an enzyme-linked immunosorbent assay (ELISA) based on two monoclonal antibodies to rabbit uPA, as described previously. ²⁸  

Cell Proliferation Assay

RCE cells were cultured with supplemented RCBM in 48-well plates for 2 days, after which the culture medium was changed to RCBM containing various concentrations of human annexin A5 or murine EGF (Kurabo) at 10 ng/mL. After incubation of the cells for 20 hours, [³H]thymidine (3.7 kBq; GE Healthcare, Little Chalfont, UK) was added to each well, and the cells were cultured for an additional 3 hours. The cells were then washed twice with PBS, treated with 10% trichloroacetic acid for 20 minutes at room temperature, washed with PBS, dissolved in 200 μL of 0.4 M NaOH containing 0.1% sodium dodecyl sulfate (SDS), and transferred to 5 mL of scintillation cocktail (Hionic-Fluor; PerkinElmer, Meriden, CT). The cell-associated radioactivity was measured with a liquid scintillation analyzer (Tri-Cab 2700TR; PerkinElmer). 

Analysis of PA Activity

RCE cells were cultured in 24-well plates for 20 hours, after which the culture medium was further supplemented with various concentrations of human annexin A5. The cells were cultured for 48 hours, and the culture supernatants were then collected and stored at −80°C until zymographic analysis or spectrofluorometric measurement of plasminogen activator (PA) activity. 

For fibrin-agarose zymographic analysis of PA activity, ^{27 31 32} the culture supernatants were fractionated by SDS-polyacrylamide gel electrophoresis on a 10% gel. Purified human uPA (55 kDa; Kowa, Nagoya, Japan) also was subjected to electrophoresis on the same gel for reference. The gel was washed with 2.5% Triton X-100 for 1 hour, placed on an opaque fibrin-agarose gel plate containing plasminogen (prepared as described previously ³² ), and incubated at 37°C overnight. PA activity was visualized as transparent bands of fibrinolysis on the fibrin-agarose gel plate, which was photographed against a black background to better reveal the bands. 

PA activity in the culture supernatants was also measured by spectrofluorometry with a plasmin-specific fluorogenic peptide substrate, t-butyloxycarbonyl-l-valyl-l-leucyl-l-lysine-4-methylcoumaryl-7-amide (Boc-Val-Leu-Lys-MCA; Peptide Institute, Osaka, Japan), ³³ in the presence of plasminogen. Purified human uPA (Kowa) was used as a standard. Portions (100 μL) of the diluted culture supernatants or human uPA in 0.05 M Tris-HCl (pH 7.4) containing 0.1 M NaCl were mixed with 250 μL of plasminogen (0.01 U/mL in 0.05 M Tris-HCl [pH 7.4] containing 0.1 M NaCl; Sigma-Aldrich, St. Louis, MO) and incubated for 3 minutes at 37°C. The reaction was then initiated by the addition of 100 μL of peptide substrate (0.5 mM in 0.05 M Tris-HCl [pH 7.4] containing 0.1 M NaCl). After incubation for 1 hour at 37°C, the reaction was stopped by the addition of 50 μL of 50% acetic acid. Fluorescence (excitation at 380 nm, emission at 460 nm) was then measured with a spectrofluorometer (model F-4010; Hitachi, Tokyo, Japan). Plasminogen-dependent proteolytic activity (PA activity) was calculated by subtraction of plasminogen-independent activity, which was measured in the absence of plasminogen, from the total activity. 

In Vivo Rabbit Models of Corneal Wound Closure

We used two types of rabbit corneal wound closure model: the iodine model and the surgical model. For the iodine model, rabbits were anesthetized with a subcutaneous injection of ketamine hydrochloride (40 mg/kg of body mass; Sankyo, Tokyo, Japan) and xylazine hydrochloride (5 mg/kg; Bayer Medical, Tokyo, Japan), as well as with oxybuprocaine eye drops. The corneal epithelium was wounded by exposure to iodine vapor as described. ³⁴ A glass tube (inner diameter, 5.5 mm) containing iodine crystals was thus placed on the central portion of each cornea for 3.5 minutes, and each eye was then washed with 50 mL of physiological saline. We confirmed that the basement membrane of the de-epithelialized cornea remained intact after such treatment by immunostaining of laminin, as previously described ²⁸ (data not shown). We applied 50 μL of human annexin A5 (0, 10, 30, or 100 μg/mL in PBS) to one eye and 50 μL of PBS to the other (control) eye of each animal every hour between 4 and 10 hours and between 17 and 25 hours after wounding. The corneal epithelial defects were stained with fluorescein and photographed at 10, 17, 20, 23, and 26 hours after wounding; the camera (Nikon, Tokyo, Japan) was equipped with a lens with a focal length of 120 mm (Medical Nikkor 120 mm; Nikon) and a yellow filter (SC50; Fuji Photograph Film, Tokyo, Japan) over the lens as well as with a cobalt blue excitation filter (Wratten 47B; Eastman-Kodak, Rochester, NY) fitted over the photoflash. The area of each epithelial defect was measured (Photoshop 5.5 software; Adobe Systems, Mountain View, CA) and then plotted against time. Corneal epithelial wound healing is characterized by a lag phase, a linear phase, and a final slow phase. ^{34 35 36 37 38 39} Our preliminary experiments revealed that the healing rate was linear between 10 and 26 hours after wounding in the iodine model (the R ² value for the regression line between these times was >0.970). We therefore calculated the rate of wound healing for each eye by linear regression analysis of the data collected at 10, 17, 20, 23, and 26 hours after wounding. The healing rate was expressed in square millimeters per hour. As a means of minimizing potential bias, the nature of the administered eye drops (human annexin A5 or vehicle) and the identity of the photographs were masked until after the analysis was completed. 

For the surgical model of wound closure, rabbits were anesthetized as described, and the corneal epithelium, but not the basement membrane, was abraded with a dulled scalpel blade over an area demarcated with a 5.5-mm-diameter circular trephine in the center of both eyes. ⁴⁰ Again, the basement membrane of the de-epithelialized cornea was confirmed to be intact by immunostaining of laminin (data not shown). Eye drops (50 μL) containing human annexin A5 (100 or 300 μg/mL) were applied to one eye, and the same volume of vehicle (20 mM borate buffer [pH 7.5] containing 2.3% [wt/vol] glycerin) was applied to the contralateral eye 2, 4, 6, and 8 hours after wounding. The corneal epithelial defects were stained with fluorescein and photographed immediately as well as 4, 12, 21, and 24 hours after wounding. The area of each epithelial defect was measured and plotted against time, as described for the iodine model. Our preliminary experiments showed that the healing rate was linear between 4 and 24 hours after wounding in the surgical model (the R ² of the regression line between 4 and 24 hours was >0.970). We therefore calculated the wound-healing rate by linear regression analysis of the data collected at 4, 12, 21, and 24 hours after wounding. Again, the nature of the eye drops (human annexin A5 or vehicle) and the identity of the photographs were masked until the analysis was completed. 

Statistical Analysis

Data are presented as the mean ± SEM. The statistical significance of the effects of annexin A5 on RCE cell migration and proliferation as well as on PA activity and uPA concentration was analyzed by the Dunnett multiple comparison test. The effect of antibodies to human annexin A5 on RCE cell migration was evaluated by Student’s t-test. Differences in the in vivo models of corneal wound closure were assessed by the paired Student’s t-test. P <0.05 was considered statistically significant. 

Results

Effects of Annexin A5 on the Migration and Proliferation of RCE Cells

We first investigated the effect of human annexin A5 on corneal epithelial cell migration with an in vitro model of wounding based on a monolayer of RCE cells (Fig. 1) . Annexin A5 increased the extent of RCE cell migration in this assay in a concentration-dependent manner, with the effect achieving statistical significance at a concentration of 100 μg/mL. Furthermore, the stimulatory effect of annexin A5 on RCE cell migration was abolished by the simultaneous addition of a monoclonal antibody to human annexin A5 (Fig. 2A) . We next examined whether annexin A5 promotes the proliferation of RCE cells. Although murine EGF (positive control) induced a marked increase in the incorporation of [³H]thymidine into RCE cells, annexin A5 had no effect on RCE cell proliferation at concentrations up to 1000 μg/mL (Fig. 2B) . 

Effect of Annexin A5 on uPA Release from RCE Cells

With the use of fibrin-agarose zymography, we examined the effect of annexin A5 on the amount of PA activity released into the culture supernatants of semiconfluent RCE cells. A single fibrinolytic band was detected with all culture supernatants (Fig. 3A) . This band corresponded to a molecular size of ∼48 kDa, whereas a human uPA standard yielded a fibrinolytic band at 52 kDa. These results suggested that the PA activity in the culture supernatants of RCE cells was attributable to uPA, consistent with previous observations. ²⁷ Culture of the cells with annexin A5 increased the intensity of the fibrinolytic band in a concentration-dependent manner, with the maximum effect apparent at a concentration of 30 μg/mL (Fig. 3A) . This effect of annexin A5 on PA activity was confirmed by measurement of such activity in the culture supernatants of RCE cells with a fluorogenic peptide substrate (Fig. 3B) . Furthermore, an ELISA for rabbit uPA revealed that annexin A5 increased the amount of uPA released from scrape-wounded monolayers of RCE cells; this effect was also concentration dependent and was statistically significant at a concentration of 100 μg/mL (Fig. 3C) . 

Effect of Annexin A5 on Rabbit Corneal Epithelial Wound Closure

We examined the effect of annexin A5 on corneal epithelial wound healing in vivo with two rabbit models. In rabbits subjected to removal of the corneal epithelium with iodine vapor, instillation of eye drops containing human annexin A5 at concentrations of 30 or 100 μg/mL every hour between 4 and 10 hours and between 17 and 25 hours after wounding in one eye resulted in a significant increase in the rate of epithelial wound healing compared with that apparent in the contralateral eye treated with PBS alone (Table 1 , Fig. 4 ). In the surgical model of corneal epithelial wound closure, instillation of eye drops containing human annexin A5 (300 μg/mL) four times within 8 hours after wounding also significantly increased the rate of corneal epithelial wound healing (Table 2) . 

Discussion

We have shown that human annexin A5 promotes corneal epithelial cell migration in vitro, and that this effect is accompanied by an increase in the release of uPA from the cells. Furthermore, instillation of eye drops containing human annexin A5 stimulated the closure of corneal epithelial wounds in two rabbit models in vivo. 

The migration and proliferation of corneal epithelial cells are important steps in the healing of corneal epithelial wounds. ¹⁷ In the current study, human annexin A5 promoted the migration of RCE cells in vitro but did not affect the proliferation of these cells. These results are similar to those previously described for normal human skin keratinocytes. ¹⁶ The effect of annexin A5 on corneal epithelial cells thus differs from that of growth factors (e.g., EGF), which stimulate both the migration and proliferation of these cells. ^{39 41 42 43} Human annexin A5 had no effect on the proliferation of other cell types, including human umbilical vein endothelial cells (data not shown). These results suggest that the migration and proliferation of corneal epithelial cells are regulated differentially and that annexin A5 does not act as a mitogen for these cells but instead selectively promotes cell migration. 

Coordination between cell adhesion to and detachment from the ECM is essential during cell migration. ^{18 19} Migrating cells extend lamellipodia at the leading edge, and integrins expressed on the surface of the lamellipodia then bind to ECM proteins such as fibronectin. Subsequent forward movement of the cell body is achieved by disruption of the integrin-ECM interaction at the rear of the cell and retraction of the trailing edge. ^{18 19} The expression of uPA is upregulated in migrating cells, including corneal epithelial cells and skin keratinocytes, ^{28 44} and the proteolytic activity that localizes to the cell surface as a result of the interaction of uPA with its receptor has been thought to contribute to cell detachment through degradation of the ECM. ^{20 21} However, given that an aminoterminal, receptor-binding fragment (ATF) of uPA that lacks enzymatic activity also promotes the migration of various cell types, ^{22 23 24 25} the binding of uPA to its receptor is also thought to regulate cell migration in a manner independent of its proteolytic activity through activation of an intracellular signaling pathway. ^{26 45 46} To investigate the mechanism by which annexin A5 stimulates the migration of corneal epithelial cells, we thus examined its effect on the release of uPA from RCE cells in culture. We found that annexin A5 increased not only the amount of PA activity but also the amount of uPA protein released by RCE cells into the culture medium, similar to our previous observations with normal human skin keratinocytes. ¹⁶ Upregulation of uPA release from RCE cells by annexin A5 may thus underlie, at least in part, the stimulatory effect of annexin A5 on RCE cell migration. 

Annexin A5 is localized predominantly in the cytosol of cells. Despite its lack of a hydrophobic signal sequence at its amino terminus, ⁵ however, annexin A5 has also been detected in extracellular fluids such as plasma, saliva, milk, and amniotic fluid. ⁴⁷ Although alternative modes of secretion have been proposed for other annexins, ^{48 49 50} extracellular annexin A5 may derive from injured or disrupted cells ⁵¹ and has been suggested to play a role in wound healing. 

We now show that instillation of eye drops containing human annexin A5 increased the rate of corneal epithelial wound closure in rabbits. These results are consistent with our previous observation that the local administration of annexin A5 promotes dermal wound healing. ¹⁶ Exogenous annexin A5 binds to negatively charged phospholipids, such as phosphatidylserine externalized on the surface of early apoptotic cells, in a calcium-dependent manner. ⁵² However, in addition to apoptotic cells, phosphatidylserine is externalized at the surface of nonapoptotic cells such as activated platelets, ⁵³ developing B cells in bone marrow and peripheral lymphoid organs ^{54 55} and activated RAW264.7 macrophages. ⁵⁶ We previously showed that ¹²⁵I-labeled human annexin A5 binds to normal human keratinocytes ¹⁶ and RCE cells (Watanabe M, et al., IOVS 2003;44:ARVO E-Abstract 3823). Cells at the leading edge of the migrating epithelium express uPA as they migrate to cover the denuded surface of the cornea during the early stage of corneal wound healing. ^{20 21 28} Injury-induced activation of cells at the leading edge of the migrating corneal epithelium may result in the externalization of phosphatidylserine at the cell surface. If so, annexin A5, either derived from the adjacent disrupted cells or provided externally, may be expected to bind to the surface of these cells and thereby to increase the release of uPA. Elucidation of the mechanism by which annexin A5 promotes the release of uPA from, and the migration of, RCE cells will require identification of the cellular molecules with which annexin V interacts and the intracellular signaling pathway that it activates. 

 Submitted for publication May 13, 2005; revised October 21 and December 17, 2005; accepted February 27, 2006.

 Disclosure: M. Watanabe, Kowa Company Ltd. (E); S. Kondo, Kowa Company Ltd. (E); K. Mizuno, Kowa Company Ltd. (E); W. Yano, Kowa Company Ltd. (E); H. Nakao, Kowa Company Ltd. (E); Y. Hattori, Kowa Company Ltd. (E); K. Kimura, 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.

 

Stimulatory effect of annexin A5 on RCE cell migration in a wounded monolayer. Confluent monolayers of RCE cells in a 48-well plate were wounded with a cell scraper. The cells were photographed immediately after wounding (A, C) or 24 hours after wounding and subsequent culture without (B) or with (D) human annexin A5 (100 μg/mL). Photographs in (A) and (B) and those in (C) and (D) are of the same wells. Black dots were marked on the external surface of the plate for reference. The dashed lines in (B) and (D) indicate the initial edge of the wound. Scale bar, 200 μm. (E) Confluent monolayers of RCE cells were wounded and then cultured for 24 hours in the presence of the indicated concentrations of human annexin A5. The extent of cell migration was determined and expressed as a percentage of the value for cells cultured in the absence of annexin A5 (control). Data are the mean ± SEM of four or five determinations. *P < 0.05 versus control.

Figure 2.

 

Specificity of the effect of annexin A5 on RCE cell migration in vitro. (A) Inhibition of the stimulatory effect of annexin A5 on RCE cell migration by a specific antibody. Confluent monolayers of RCE cells were wounded and then cultured for 24 hours in the absence or presence of human annexin A5 (10 μg/mL) or either a monoclonal antibody to human annexin A5 (A180) or control mouse immunoglobulin G (mIgG) at 100 μg/mL, as indicated. The extent of cell migration was determined and expressed as a percentage of the value for cells incubated without additions. Data are the mean ± SEM of six determinations. **P < 0.01 for the indicated comparisons; NS, not significant. (B) Lack of effect of annexin A5 on RCE cell proliferation. Semiconfluent RCE cells in a 48-well plate were cultured in the presence of the indicated concentrations of human annexin A5 or with murine EGF (10 ng/mL). After 20 hours, the cells were incubated in the additional presence of [³H]thymidine for 3 hours. The radioactivity incorporated into cellular DNA was then determined. Data are the mean ± SEM of five determinations. ***P < 0.001 versus control (no annexin A5 or EGF).

Figure 3.

 

Stimulatory effect of annexin A5 on uPA release from RCE cells. (A) Zymographic analysis of PA activity in culture supernatants of RCE cells. RCE cells were cultured for 48 hours in the presence of the indicated concentrations of human annexin A5, and the culture supernatants (as well as purified human uPA as a positive control) were subsequently subjected to SDS-polyacrylamide gel electrophoresis and zymography with a fibrin-agarose gel plate. (B) Assay of PA activity in culture supernatants of RCE cells with a fluorogenic assay. RCE cells were treated as in (A), after which the culture supernatants were assayed for PA activity with a fluorogenic peptide substrate. Data are expressed as a percentage of the activity apparent for cells incubated without annexin A5 (control) and are the mean ± SEM of three or four determinations. **P < 0.01, ***P < 0.001 versus control. (C) Effect of annexin A5 on uPA release by wounded RCE cells in culture. Confluent monolayers of RCE cells in 48-well plates were wounded with a cell scraper and then cultured in the presence of the indicated concentrations of human annexin A5 for 24 hours. The culture supernatants were collected and subsequently assayed for uPA with an ELISA. Data are the mean ± SEM of four or five determinations. *P < 0.05 versus control (no annexin A5).

Table 1.

 

Effect of Instillation of Eye Drops Containing Human Annexin A5 on the Wound-Healing Rate in the Iodine Model of Corneal Epithelial Wound Closure in Rabbits

Annexin A5 (μg/mL)	n	Healing Rate (mm2/hour)			Significance
				Treated Eye		Control Eye
0	8	0.896 ± 0.024	0.907 ± 0.036	NS
10	9	0.912 ± 0.034	0.890 ± 0.029	NS
30	11	0.970 ± 0.032	0.905 ± 0.025	P < 0.05
100	10	0.976 ± 0.030	0.897 ± 0.034	P < 0.01

 Data are the mean ± SEM for the indicated number of eyes. NS, not significant.

Figure 4.

 

Effect of annexin A5 on wound healing in the iodine model of corneal wounding. The corneas of a rabbit treated with eye drops containing either human annexin A5 (100 μg/mL) or PBS vehicle alone were stained with fluorescein and photographed 10 and 26 hours after exposure to iodine vapor.

Table 2.

 

Effect of Instillation of Eye Drops Containing Human Annexin A5 on the Wound-Healing Rate in the Surgical Model of Corneal Epithelial Wound Closure in Rabbits

Annexin A5 (μg/mL)	n	Healing Rate (mm2/hour)			Significance
				Treated Eye		Control Eye
100	10	0.903 ± 0.024	0.862 ± 0.018	NS
300	11	0.878 ± 0.029	0.807 ± 0.029	P < 0.01

 Data are the mean ± SEM for the indicated number of eyes. NS, not significant.

The authors thank Ayako Kadowaki for technical assistance.