Galectins-3 and -7, but not Galectin-1, Play a Role in Re-epithelialization of Wounds*

Disorders of wound healing characterized by impaired or delayed re-epithelialization are a serious medical problem. These conditions affect many tissues, are painful, and are difficult to treat. In this study, using cornea as a model, we demonstrate for the first time the importance of carbohydrate-binding proteins galectins-3 and -7 in re-epithelialization of wounds. In two different models of corneal wound healing, re-epithelialization of wounds was significantly slower in galectin-3-deficient (gal3−/−) mice compared with wild-type (gal3+/+) mice. In contrast, there was no difference in corneal epithelial wound closure rates between galectin-1-deficient and wild-type mice. Quantitation of the bromodeoxyuridine-labeled cells in gal3+/+ and gal3−/− corneas revealed that corneal epithelial cell proliferation rate is not perturbed in gal3−/− corneas. Exogenous galectin-3 accelerated re-epithelialization of wounds in gal3+/+ mice but, surprisingly, not in the gal3−/− mice. Gene expression analysis using cDNA microarrays revealed that healing corneas of gal3−/− mice contain markedly reduced levels of galectin-7 compared with those of gal3+/+ mice. More importantly, unlike galectin-3, galectin-7 accelerated re-epithelialization of wounds in both gal3−/− and gal3+/+ mice. In corresponding experiments, recombinant galectin-1 did not stimulate the corneal epithelial wound closure rate. The extent of acceleration of re-epithelialization of wounds with both galectin-3 and galectin-7 was greater than that observed in most of the published studies using growth factors. These findings have broad implications for developing novel therapeutic strategies for treating nonhealing wounds.

In a number of organ systems including cornea, skin, and gastrointestinal tract, disorders of wound healing characterized by impaired or delayed re-epithelialization and associated persistent epithelial defects constitute a serious medical problem (1)(2)(3)(4). Persistent epithelial defects of the cornea may progress through the anterior stroma, resulting in stromal ulcer-ation and, in the worst cases, perforation of the tissue with significant visual loss (3,4). Delayed re-epithelialization and persistent epithelial defects are also characteristic of chronic wounds in the elderly, decubitus ulcers, and venous statis ulcer of the skin, conditions that together affect millions of individuals worldwide (1,2). In most cases, failure of re-epithelialization is not attributed to inadequate cell proliferation but the result of a reduced potential of the epithelium to migrate across the wound bed (5)(6)(7). Cell migration is a complex biological process that entails sequential adhesion and release from the substrate, a process in which cell-matrix interactions play a key role (8,9). Whereas the regulation of cell-matrix interactions is a poorly understood process, it is known that in many instances, cell attachment to the matrix is mediated by recognition between extracellular matrix (ECM) 1 molecules and transmembrane integrin receptors. Recent studies have provided evidence that members of the galectin class of ␤-galactoside-binding proteins (10 -13), in particular galectin-1 and galectin-3, also have the potential to mediate cell-matrix interactions by a novel mechanism (14 -17). Both lectins are expressed in inflammatory cells and in epithelia and fibroblasts of various tissues (10 -13). They are found on the cell surface within ECM and in the cytoplasm of cells and are thought to influence cell-matrix adhesion by binding to the ECM and cell surface-glycosylated counter receptors (e.g. certain isoforms of laminin, fibronectin, vitronectin, and integrins). In addition, galectin-3 is found in the nucleus of the cells and may influence cell-matrix interactions indirectly by influencing the expression of well known cell adhesion molecules (e.g. ␣ 6 ␤ 1 and ␣ 4 ␤ 7 integrins) (16,17) and cytokines (e.g. IL-1) (18). Although all members of the galectin family bind to ␤-galactoside residues, each galectin has unique fine specificity for more complex oligosaccharides; therefore, different members of the galectin family may bind distinct glycoconjugate receptors, resulting in specific downstream effects. Here we demonstrate that galectin-3 but not galectin-1 plays a role in re-epithelialization of corneal wounds and that another member of the galectin family, galectin-7, is also involved in the wound healing process.

EXPERIMENTAL PROCEDURES
Wound Healing Experiments-All animal treatments in this study conformed to the Association for Research in Vision and Ophthalmology Resolution on the Use of Animals in Vision Research and the recommendations of the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Gal3 Ϫ/Ϫ mice were generated by targeted interruption of the galectin-3 gene as described previously (19). Gal3 Ϫ/Ϫ and gal3 ϩ/ϩ mice were obtained by interbreeding gal3 ϩ/Ϫ mice and carried as separate lineages. Galectin-1-deficient mice originally prepared by Poirier and Robertson (20) were made available by Drs. Douglas N. Cooper, (University of California, San Francisco, CA) and Carrie Miceli (University of California, Los Angeles, CA). Both gal1 Ϫ/Ϫ and gal3 Ϫ/Ϫ mice are fertile and viable and do not exhibit any overt defects. To produce corneal wounds, mice were anesthetized by an intramuscular injection of 1.25% avertin (0.2 ml/10 g body weight) (Aldrich). Proparacaine eye drops (Alcain, Alcon Labs, Inc., Fort Worth, TX) were applied to the cornea as a topical anesthetic. We produced 2-mm corneal wounds on the right eye of each animal by: (i) transepithelial excimer laser ablations (2-mm optical zone; 42-44-micron ablation depth, phototherapeutic keratectomy mode) using Summit Apex Plus Excimer Laser (Waltham, MA); (ii) placing alkali (0.5 N NaOH) soaked filter discs on the surface of the cornea for 2 min; or (iii) abrasion using Algerbrush (Lago Vista, TX). The left eye served as a control. Following surgery, all of the animals received buprenorphine (intramuscular, 0.2 ml of 0.3 mg/ml Buprenex, Reckitt and Colman Pharmaceuticals, Inc., Richmond, VA) as a pain killer. Antibiotic ointment (Vetropolycin, Pharmaderm, Melville, NY) was applied, and the corneas were allowed to partially heal in vivo for 16 -18 h. For healing in vitro, the eyes were excised out following surgery and were pinned down on paraplast wax in 24-well plates using 1 cm-long tips of 20-gauge needles (one eye/well). The eyes were then incubated in serum-free medium (0.5 ml/well) in a tissue culture incubator for 20 -22 h (21). At the end of the healing period, wound areas were visualized by staining with methylene blue and were quantitated using Sigma Scan software (21). An unpaired two-group Student's t test was used to test for significance of the data. For immunohistochemical analyses of normal and healing corneas, the eyes were fixed in buffered formalin for 2 h and were processed for the preparation of paraffin sections and immunolocalization of galectins using monoclonal rat anti-human-galectin-3 (undiluted hybridoma fluid; mAb M3/38, American Type Culture Collection, Manassas, VA) (22), polyclonal rabbit anti-rat galectin-1 (1:100 dilution, provided by Dr. D. N. Cooper), and polyclonal goat anti-human galectin-7 (1:250 dilution) (23). For bromodeoxyuridine (BrdUrd) labeling and detection 2 h before the mice were sacrificed, BrdUrd (12 mg/ml in PBS, Sigma) was injected subcutaneously in the dorsal subcapular region (100 g/g body weight). The eyes were removed, fixed in buffered formalin for 2 h, and were processed for preparation of paraffin sections and immunolocalization of BrdUrd using an anti-BrdUrd monoclonal antibody (1:100, BD Biosciences) (24). The number of BrdUrd-labeled cells was counted in normal and healing corneas of both gal3 ϩ/ϩ and gal3 Ϫ/Ϫ mice.
Experiments to Test the Therapeutic Potential of Galectins for Acceleration of Re-epithelialization of Corneal Wounds-Recombinant fulllength human galectin-3 and galectin-7 were produced in Escherichia coli and purified as described previously (23,25). Recombinant fulllength galectin-1 was provided by Dr. D. N. Cooper (26). Alkali-burn wounds (2-mm-diameter) were produced on both eyes of anesthetized animals. The left eyes of animals served as controls and were incubated in the media alone. Right eyes were incubated in media containing various test reagents including: (i) galectin-1, -3, or -7 (5-20 g/ml), (ii) a galectin plus 0.1 M ␤-lactose, or (iii) a galectin plus 0.1 M sucrose. At the end of the healing period, remaining wound areas were stained and quantitated. Each group contained a minimum of four eyes. All experiments were performed at least twice.
Gene Expression Analysis Using cDNA Microarrays-Transepithelial excimer laser ablations (2-mm diameter) were produced on the right eye of 30 gal3 ϩ/ϩ and 30 gal3 Ϫ/Ϫ mice. Corneas were allowed to partially heal in vivo for 16 -18 h. At the end of the healing period, animals were sacrificed, and the corneas were excised and processed for gene expression analysis using SMART cDNA technology (Clontech Laboratories, Inc., Palo Alto, CA). Total RNA was isolated using the reagents provided in the Atlas Pure Total RNA-labeling system kit (yield: 3.5 and 2.6 g for gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas, respectively; A 260 /A 280 , 1.48 and 1.37 for gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas, respectively; 28 S/18 S, 1.8 for both preparations) and was used for cDNA probe synthesis as described in the instruction manual for SMART cDNA Probe Synthesis for Atlas (Clontech). The amplified cDNAs (500 ng) were radiolabeled using Klenow enzyme and [ 33 P]␣ATP and were then hybridized to cDNA microarrays (mouse 1.2-1 broad spectrum, ϳ1200 genes, Clontech). Following hybridization, the membranes were exposed to a phosphorimaging screen. Results were analyzed by Atlas Image 2.0 software (Clontech). The data were verified by semiquantitative RT-PCR, Western blot analysis, and immunohistochemical localization studies.
Semiquantitative RT-PCR was performed using total RNA preparations of the corneas and gene-specific custom primers purchased from Clontech and other reagents from the Advantage 2 PCR kit (Clontech).

Galectin-3 Is Expressed in Normal and Healing Corneal Epithelium at Sites of Cell-Matrix and Cell-Cell Adhesion-Cor-
neal epithelium is a prototype-stratified squamous epithelium. In mouse, it constitutes 20 -25% of total corneal thickness and is composed of 5-6 layers of cells (Fig. 1Ai, arrow). Posterior to the epithelial basement membrane is corneal stroma, which in mouse represents 70 -80% of the total corneal thickness (Fig.  1Ai). Abrasion wounds (Fig. 1Aii) as well as alkali-burn wounds (data not shown) removed the epithelium, leaving the corneal stroma intact. In contrast, excimer laser treatment, which is commonly used for correction of myopia, removed the epithelium as well as the anterior corneal stroma (Fig. 1Aiii).
To determine the expression pattern of galectin-3 in normal and healing corneas, corneas with 2-mm excimer laser ablations (n ϭ 9) and abrasion wounds (n ϭ 7) were allowed to partially heal in vivo and were then processed for immunostaining with anti-galectin-3, mAb M3/38. In normal mouse corneas, galectin-3 was expressed in basal and middle cell layers of all of the 16 corneas analyzed (Fig. 1Bi). Here, the galectin-3 immunoreactivity was largely detected at sites of cell-cell and cell-matrix attachment, and in many cases, the immunoreactivity was very much pronounced at the sites of FIG. 1. Galectin-3 is expressed in epithelium of normal and healing corneas. Corneas with 2-mm abrasion or excimer laser wounds were allowed to partially heal in vivo and were then analyzed for galectin-3 immunoreactivity in paraffin sections. A, hematoxylinand eosin-stained normal corneas (i) and corneas immediately after abrasion (ii) and excimer laser (iii) injury. B, immunohistochemical staining of normal and healing corneas after excimer laser injury. Compared with the normal nonmigrating epithelia (i), markedly greater galectin-3 immunoreactivity was detected in the migrating epithelia of healing corneas (ii). No galectin-3 immunoreactivity was detected in normal (iii) or healing (iv) gal3 Ϫ/Ϫ corneas. No galectin-1 immunoreactivity was detected in corneal epithelium of normal (v) and healing (vi) corneas. Brown color indicates positive immunostaining. WE, wound edge; LE, leading edge of migrating epithelium; arrows, epithelium; arrowheads, leukocytes/stromal cells.
cell-matrix interactions. In healing corneas, the leading edge of migrating epithelium stained more intensely with mAb M3/38 compared with the normal epithelium of the contralateral eye of all of the nine corneas with excimer laser wounds (Fig. 1Bii) and all of the seven corneas with abrasion wounds (data not shown) (staining pattern indistinguishable from Fig. 1Bii). Also, in healing epithelium, immunostaining was more intense at sites of cell-matrix attachment. Whereas stromal cells of normal corneas did not react with anti-galectin-3 mAb, cells in the anterior stroma under healing corneas expressed galectin-3, especially in the region under migrating epithelium (Fig.  1Bii, arrowhead). Corneas allowed to heal in organ culture showed similar results with respect to galectin-3 immunoreactivity in corneal epithelium (n ϭ 2, excimer laser wounds) (data not shown). However, anterior stroma of corneas, which were allowed to heal in vitro, lacked cells expressing galectin-3, suggesting that galectin-3-positive cells seen in the stroma of corneas allowed to heal in vivo are most probably lymphocytes and not keratocytes. As expected, no galectin-3 immunoreactivity was detected in normal (Fig. 1Biii) and healing (Fig.  1Biv) corneas of gal3 Ϫ/Ϫ mice. Also, no galectin-1 immunoreactivity was detected in the epithelia of normal (Fig. 1Bv) and healing (Fig. 1Bvi) corneas of the wild-type mice.
Re-epithelialization of Corneal Wounds Is Perturbed in gal3 Ϫ/Ϫ Mice-To determine whether re-epithelialization of corneal wounds is impaired in galectin-3-deficient mice, we used two different models of corneal wound healing. Corneas with excimer laser ablations or alkali-burn wounds were allowed to partially heal in vivo or in vitro for up to 22 h. At the end of the healing period, remaining wound areas were quantitated and compared among different groups. Regardless of whether the corneas were injured by excimer laser or by alkali treatment and whether the corneas were allowed to heal in vivo or in vitro, corneal epithelial wound closure rate (expressed as mm 2 /h) was significantly slower in gal3 Ϫ/Ϫ mice compared with gal3 ϩ/ϩ mice (Fig. 2, A-E). In contrast, no differences were found in the wound closure rates between galectin-1 ϩ/ϩ and galectin-1 Ϫ/Ϫ groups (Fig. 2F). Wound closure rates expressed as mm 2 /h in corneas from different groups are indicated in Fig.  2 legend.
The Rate of Corneal Epithelial Cell Proliferation Is Not Perturbed in gal3 Ϫ/Ϫ Mice-To determine whether delayed reepithelialization of corneal wounds in gal3 Ϫ/Ϫ mice was because of a deficiency in the rate of corneal epithelial cell proliferation, normal and healing gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas were labeled with BrdUrd to identify cells undergoing DNA synthesis. It is well established that in response to corneal epithelial injury, proliferation of cells distal to the wound edge is increased, whereas that at the migratory front is decreased (28,29). Consistent with this finding, BrdUrd-labeled cells in both gal3 ϩ/ϩ and gal3 Ϫ/Ϫ healing corneas were seen largely in the peripheral epithelium (Fig. 3A). There was no significant difference in the number BrdUrd-labeled cells between gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas (Fig. 3B), suggesting that the rate of corneal epithelial cell proliferation is not perturbed in gal3 Ϫ/Ϫ mice.
Exogenous Addition of Galectin-3 Stimulates Corneal Epithelial Wound Closure in gal3 ϩ/ϩ Mice but Not in gal3 Ϫ/Ϫ Mice-Having demonstrated that the corneal epithelial wound closure rate is perturbed in gal3 Ϫ/Ϫ mice, it was of interest to determine whether exogenous galectin-3 would stimulate re-epithelialization of corneal wounds in organ culture. In this study, the corneas of gal3 ϩ/ϩ and gal3 Ϫ/Ϫ mice with alkali-burn wounds were incubated in serum-free media in the presence and absence of varying amounts of recombinant galectin-3. After a 20 -22-h healing period, the remaining wound areas were quantified. The exogenous galectin-3 had no influence on the rate of re-epithelialization of corneal wounds in gal3 Ϫ/Ϫ mice (Fig. 4A), but it stimulated the rate of wound closure in a concentration-dependent manner in gal3 ϩ/ϩ mice (Fig. 4B). Overall, the extent of acceleration of re-epithelialization of wounds was 43 and 71% in the presence of 10 and 20 g/ml galectin-3, respectively. The stimulatory effect of galectin-3 on corneal epithelial wound closure was specifically inhibited by a competing sugar, ␤-lactose, but not by an irrelevant disaccharide, sucrose (Fig. 4C). In corresponding experiments, recombinant galectin-1 did not stimulate the corneal epithelial wound closure rate (Fig. 4D).
Healing gal3 Ϫ/Ϫ Corneas Express Reduced Levels of Galectin-7-In an attempt to understand the reason that re-epithelialization of corneal wounds is perturbed in gal3 Ϫ/Ϫ mice and why gal3 Ϫ/Ϫ corneas are not responsive to exogenous galectin-3, we compared gene expression patterns of healing gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas using cDNA microarrays. The expression of 16 genes was up-regulated, and that of 13 genes was downregulated over 5-fold in the healing gal3 Ϫ/Ϫ corneas compared with healing gal3 ϩ/ϩ corneas. A list of all differentially expressed genes can be viewed on our web site: www.neec.com/ MicroarrayData.html. Of the 29 genes exhibiting altered expression pattern in healing gal3 Ϫ/Ϫ corneas, galectin-7, another galactose-binding protein, appeared to be the most relevant to the current study and was selected for further investigation. Gene expression analysis revealed that healing corneas of gal3 Ϫ/Ϫ mice contained ϳ12-fold less gene transcripts for galectin-7 (Fig. 5A)  wild-type mice. That healing corneas of gal3 Ϫ/Ϫ mice contain reduced levels of galectin-7 transcripts was further confirmed by semiquantitative RT-PCR (Fig. 5B). Western blot analysis using detergent extracts of healing gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas (Fig. 5C) and immunohistochemical studies using paraffin sections of the corneas (Fig. 5D) revealed that healing corneas of gal3 Ϫ/Ϫ mice contain markedly reduced levels of galectin-7 protein compared with those of the wild-type littermates. Also, gal3 Ϫ/Ϫ mouse embryonic fibroblasts (MEF) grown in cell culture expressed reduced levels of galectin-7 compared with gal3 ϩ/ϩ MEF cultures (Fig. 5E). Both semiquantitative RT-PCR and Western blot analyses were performed at least twice on each sample with reproducible results.
Galectin-7 Stimulates Re-epithelialization of Wounds in gal3 Ϫ/Ϫ Corneas-In this study, the corneas of gal3 ϩ/ϩ and gal3 Ϫ/Ϫ mice with alkali-burn wounds were incubated in serum-free media in the presence and absence of varying amounts of recombinant galectin-7. After a 20 -22-h healing period, remaining wound areas were quantified. Unlike galectin-3 (Fig. 4A), galectin-7 accelerated the re-epithelialization of wounds in gal3 Ϫ/Ϫ corneas (Fig. 6A). The stimulatory effect of galectin-7 on corneal epithelial wound closure was also specifically inhibited by a competing sugar, ␤-lactose, but not by sucrose (Fig. 6A). Galectin-7 also accelerated the re-epithelialization of wounds in gal3 ϩ/ϩ corneas (Fig. 6B). DISCUSSION In this study, we demonstrate that galectin-3 plays a role in the re-epithelialization of corneal wounds. First, immunohistochemical studies revealed that galectin-3 is located in high density at sites of corneal epithelial cell-matrix adhesion, an ideal location for influencing cell-matrix interactions and cell migration. Second, in two different models of corneal wound healing, re-epithelialization of corneal wounds was significantly slower in the gal3 Ϫ/Ϫ mice compared with that in wildtype mice. In contrast, there was no difference in the rate of re-epithelialization of corneal wounds between galectin-1-deficient and wild-type mice. Third, the exogenous recombinant galectin-3 stimulated re-epithelialization of corneal wounds in gal3 ϩ/ϩ mice in a concentration-dependent manner. In contrast, recombinant galectin-1 had no influence. We further demonstrated that the stimulatory effect of galectin-3 on the rate of corneal epithelial wound closure can be almost completely abrogated by a competing disaccharide, ␤-lactose, but not by an irrelevant disaccharide, sucrose. This finding suggested that the carbohydrate recognition domain was directly involved in the beneficial effect of the exogenous lectin on the wound closure.
Regarding the mechanism by which galectin-3 may influence re-epithelialization of corneal wounds, the lectin is thought to FIG. 5. Epithelia of healing gal3 ؊/؊ corneas express reduced levels of galectin-7. A, tabulated results of microarray data of selected genes. GAPDH, RPS29, and ODC are housekeeping genes. Arrows indicate fold up or down in healing corneas of gal3 Ϫ/Ϫ mice compared with gal3 ϩ/ϩ mice. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RPS29, ribosomal protein S29; ODC, ornithine decarboxylase; NC, no change. B, confirmation of microarray data by semiquantitative RT-PCR. Lanes 1-4, represent amplified products collected at 23, 28, 33, and 38 cycles, respectively, for GAPDH, RPS29, and ODC and amplified products collected at 32, 34, 36, and 38 cycles, respectively, for galectin-7. C, Western blot analysis showing that healing corneas of gal3 Ϫ/Ϫ mice express reduced levels of galectin-7 protein. radioimmune precipitation buffer extracts of whole corneas representing 20 g of protein were loaded in each lane. After immunostaining with antibodies to galectin-7, the blot was stripped and subsequently reprobed with anti-galectin-3 and antitubulin. Western blot analysis was performed at least twice on each sample with reproducible results. D, immunohistochemical staining of paraffin sections using anti-galectin-7. (i and ii), corneas stained with hematoxylin; (iii and iv), corneas stained with a anti-galectin-7 polyclonal antibody. In this study, 42 tissue sections each of the gal3 ϩ/ϩ and gal3 Ϫ/Ϫ -healing corneas were analyzed (10 -12 sections/cornea; 4 corneas/group). The immunoreactivity was graded as intense (ϩϩϩ), moderate (ϩϩ), weak (ϩ), and negative (Ϫ). Significantly reduced galectin-7 immunoreactivity was detected in epithelium of healing gal3 Ϫ/Ϫ mice corneas compared with that in the epithelium of the wild-type mice (gal3 ϩ/ϩ , ϩϩϩ 36/42, ϩϩ 5/42; ϩ or Ϫ 1/42; gal3 Ϫ/Ϫ , ϩϩϩ 3/42, ϩϩ 26/42, ϩ or Ϫ 13/42). E, compared with gal3 ϩ/ϩ MEF, gal3 Ϫ/Ϫ MEF expressed reduced levels of galectin-7 (see legends to panels B and C for experimental details). mediate cell-cell and cell-matrix interactions by binding to complementary glycoconjugates containing polylactosamine chains found in many ECM and cell surface molecules, such as certain isoforms of fibronectin, laminin, and integrins (10 -15). However, our findings that exogenous galectin-3 does not accelerate the re-epithelialization of wounds in gal3 Ϫ/Ϫ mice suggest that intracellular galectin-3 contributes significantly to the process of wound healing, most probably, by influencing the expression of specific cell surface and/or ECM receptors, which in turn influence cell-matrix interactions and cell migration. The precedence for this notion is suggested by published studies in which galectin-3 was stably overexpressed in breast carcinoma cell lines, and it was shown that cells overexpressing the lectin expressed elevated levels of ␣ 4 ␤ 7 and ␣ 6 ␤ 1 integrins and exhibited enhanced adhesion to various ECM molecules including laminin, fibronectin, and vitronectin compared with parental cell lines expressing little or no galectin-3 (16,17). In another study (30), colon cancer carcinoma cell lines transfected with galectin-3 expressed the elevated levels of a specific mucin, MUC2, a major ligand of the lectin itself (31). The fact that the stimulatory effect of exogenous galectin-3 on the rate of re-epithelialization of wounds in gal3 ϩ/ϩ mice is lactose-inhibitable raises an intriguing possibility that intracellular galectin-3 may in fact regulate the glycosylation of the proteins, which serve as cell surface or ECM receptors of the lectin itself. That intracellular galectin-3 has the potential to act on nuclear matrix to influence complex biological processes is suggested by findings that under certain conditions the lectin can be found to be associated in the nucleus with ribonucleoprotein complexes and can act as a pre-mRNA-splicing factor (32). Also, Wang et al. (33) have demonstrated that in prostate adenocarcinoma cells, galectin-3 is associated with nuclear matrix and binds with single-stranded DNA and with RNA.
An analysis of gene expression patterns of corneas of healing gal3 ϩ/ϩ and gal3 Ϫ/Ϫ mice using mouse cDNA microarrays revealed that the healing corneas of gal3 Ϫ/Ϫ mice expressed markedly reduced levels of galectin-7 compared with those of wild-type mice. Galectin-7, first reported in 1994 (12), is not as well characterized as galectins-1 and -3. To the best of our knowledge, transgenic mice deficient in galectin-7 have not been reported. Unlike galectins-1 and -3, galectin-7 exhibits a remarkable degree of tissue specificity. In adult animals, its expression is restricted to epithelia that are stratified or are destined to become stratified (34,35). The protein is thought to be involved in cell-matrix and cell-cell interactions and in apoptosis (36 -38). In general, an inverse correlation exists between galectin-7 expression and keratinocyte proliferation, and galectin-7 expression is abrogated in SV40-transformed keratinocytes as well as in cell lines derived from epidermal tumors. Our findings that exogenous galectin-3 does not stimulate reepithelialization of wounds in gal3 Ϫ/Ϫ corneas and that the healing gal3 Ϫ/Ϫ corneas contain reduced levels of galectin-7 lead us to speculate that galectin-3 may influence the re-epithelialization of wounds, at least in part, by modulating galectin-7. Indeed, we found that unlike galectin-3, galectin-7 accelerated the re-epithelialization of wounds in gal3 Ϫ/Ϫ corneas in a lactose-inhibitable manner. Also, MEF of gal3 Ϫ/Ϫ mice expressed reduced levels of galectin-7. Regardless of the mechanisms involved, our findings that both galectin-3 and galectin-7 stimulate the re-epithelialization of corneal wounds have broad implications for developing novel therapeutic strategies for the treatment of nonhealing wounds. At present, the treatment of persistent epithelial defects of the cornea is a major clinical problem. Moreover, the need continues for effective treatment of chronic wounds in the elderly, decubitus ulcers, and venous stasis ulcers of the skin. A number of growth factors (epidermal growth factor, transforming growth factor-␣, fibroblast growth factor, keratinocyte growth factor, and hepatocyte growth factor) known to stimulate cell proliferation have been tested for usefulness in corneal as well as cutaneous epithelial wound healing with overall disappointing results (1, 2, 39 -42). The extent of acceleration of re-epithelialization of wounds was far less in most of these studies using growth factors (39,40) than that observed with galectins in the current study. Also, the epithelium of corneas treated with growth factors such as epidermal growth factor is hyperplastic (42)(43)(44), a clearly undesirable condition. In this respect, the clinical potential of galectin-3 and galectin-7 may be more attractive than that of growth factors because the lectins have not been shown to induce cell mitosis in epithelial cells. That galectin-3 does not induce cell mitosis is further suggested by our findings that the corneal epithelial cell proliferation rate is not perturbed in gal3 Ϫ/Ϫ corneas. Over the last decade, the potential of excimer laser keratectomy to modify the corneal profile for correction of myopia has been realized. Thousands of such procedures are performed each week, providing myopic individuals freedom from eye glasses and contact lenses. In view of the fact that 25-30% of the adult population worldwide is myopic, it has been esti- FIG. 6. A, exogenous galectin-7 stimulates re-epithelialization of corneal wounds in gal3 Ϫ/Ϫ mice. Gal3 Ϫ/Ϫ corneas with 2-mm alkali-burn wounds were allowed to heal in organ culture in serum-free media in the presence and absence of recombinant galectin-7 and saccharides for 20 -22 h. At the end of the healing period, wound areas were quantified. A value of 1.0 was assigned to the healing rate of control corneas incubated in media alone. The value of corneas incubated in media containing galectin-7 and saccharides is expressed as change in healing rate with respect to the control corneas. Galectin-7 stimulated corneal epithelial wound closure of gal3 Ϫ/Ϫ corneas. ␤-Lactose but not sucrose inhibited stimulatory effect of galectin-7 on wound closure rate. *, p Ͻ 0.05 compared with media and gal7 ϩ Lac groups. B, galectin-7 stimulated corneal epithelial wound closure rate in both gal3 Ϫ/Ϫ and gal3 ϩ/ ϩmice. Actual healing rates expressed as mm 2 /h are indicated. p Ͻ 0.05 among (i) gal7 and respective media groups, (ii) media groups of gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas, and (iii) gal7 groups of gal3 ϩ/ϩ and gal3 Ϫ/Ϫ corneas. Mean Ϯ S.E. of two of more experiments are reported. Lac, lactose; Suc, sucrose. mated that nearly half a million such procedures will be performed in the U. S. alone in a given year (45). In some cases following excimer laser surgery, there is a delay in epithelial healing. Such a delay is highly undesirable, because it puts the cornea at risk of developing postoperative haze, infectious keratitis, and ulceration. Again, galectin-based drugs may help promote re-epithelialization of wounds in such cases.