Molecular Vision 2006; 12:55-64 <http://www.molvis.org/molvis/v12/a6/> Received 25 July 2005 | Accepted 23 January 2006 | Published 23 January 2006

Download Reprint

Microarray and protein analysis of human pterygium

Molykutty John-Aryankalayil,¹ Nicholas Dushku,² Cynthia J. Jaworski,¹ Constance A. Cox,¹ Gregory Schultz,³ Janine A. Smith,¹ Keri E. Ramsey,⁴ Dietrich A. Stephan,⁴ Kenn A. Freedman,⁵ Ted W. Reid,⁵ Deborah A. Carper¹
 
 

¹National Eye Institute, National Institutes of Health, Bethesda, MD; ²Kaiser Permanente Medical Center, Sacramento, CA; ³Department of Obstetrics and Gynecology, University of Florida, Gainesville, FL; ⁴Translational Genomics Research Institute, Phoenix, AZ; ⁵Department of Ophthalmology and Visual Sciences, Texas Tech University Health Science Center, Lubbock, TX

Correspondence to: Dr. Deborah Carper, National Eye Institute, National Institutes of Health, 31 Center Drive, MSC 2510, Building 31, Room 6A03, Bethesda, MD, 20892; Phone: (301) 496-2234; FAX: (301) 496-9970; email: carperd@nei.nih.gov

Abstract

Purpose: Pterygium is a sunlight-related, ocular-surface lesion that can obscure vision. In order to identify specific genes that may play a role in pterygium pathogenesis, we analyzed the global gene expression profile of pterygium in relation to autologous conjunctiva.

Methods: Oligonucleotide microarray hybridization was used to compare the gene expression profile between human whole pterygium and autologous conjunctiva. Selected genes were further characterized by RT-PCR, western blot, and immunohistochemistry, and comparisons were made with limbal and corneal tissues.

Results: Thirty-four genes exhibited a 2 fold or greater difference in expression between human whole pterygium and autologous conjunctiva. Twenty-nine transcripts were increased and five transcripts were decreased in pterygium. Fibronectin, macrophage-inflammatory protein-4 (MIP-4), and lipocalin 2 (oncogene 24p3; NGAL) were increased 9, 5, and 2.4 fold, respectively, while Per1 and Ephrin-A1 were decreased 2 fold in pterygium. Western blots showed that fibronectin and MIP-4 were increased in pterygium compared to limbus, cornea, and conjunctiva. Immunohistochemical analysis showed fibronectin in the stroma; lipocalin 2 in the basal epithelial cells, basement membrane, and extracellular stroma; and MIP-4 in all areas of the pterygium.

Conclusions: These data show both novel and previously identified extracellular-matrix-related, proinflammatory, angiogenic, fibrogenic, and oncogenic genes expressed in human pterygium. Comparisons of selected genes with limbal and corneal tissues gave results similar to comparisons between pterygium and normal conjunctiva. The increased expression of lipocalin 2, which activates matrix metalloproteinases (MMP), is consistent with our previous findings that MMP-9 and other MMPs are highly expressed in pterygium basal epithelium.

Introduction

Pterygium is a sunlight-related, ocular-surface disease that can obscure vision [1,2]. It occurs most frequently in populations located near the equator and in laborers who work outdoors or in specific factory environments [3]. There is a higher prevalence of pterygium in farmers, watermen, postal workers, sawmill workers, and welders. A cost of illness study in Australia showed that pterygium occurred in 12% of Australian men over 60 years of age [4], and a population-based survey conducted in rural southern China demonstrated a pterygium prevalence of 33% in subjects aged 50 years or older [5]. In the United States, pterygium is common in the southern and western states and in rural populations throughout the United States. In addition, watermen who work on the Chesapeake Bay in Maryland have a higher incidence of pterygium. For example, Taylor et al. [6] studied a group of watermen and found 140 pterygia and 642 pinguecula (a fibroblastic growth associated with pterygium pathogenesis) in 838 watermen.

Pterygium is a wing-shaped, epithelial-covered fibrovascular lesion that originates from the limbus, more often on the nasal than temporal side [1]. Visual impairment can result from astigmatism induced by the lesion even before involvement of the central cornea; progression of the lesion with migration centrally into the visual axis results in vision loss [7,8].

The standard treatment for pterygium is surgical removal, which has a recurrence rate as high as 61-82% [8,9]. If the excision is combined with adjunct treatments, such as conjunctival autograft, amniotic membrane grafts, beta-radiation, or mitomycin C, recurrence can be reduced to 2-31%, depending on factors such as age, geographic location, occupation, pterygium morphology, and surgeon experience [8,10-12].

Numerous theories have been postulated for the pathogenesis of pterygia, including degeneration, ultraviolet irradiation, oxidative stress, immunological and anti-apoptotic mechanisms, and viral infections [13-19]. Epidemiological data strongly links sunlight exposure to the occurrence of pterygium, however, its etiology has not been fully explained.

To address this, a number of studies have focused on identifying proteins expressed in pterygium. Pro-inflammatory cytokines and angiogenic and fibrogenic growth factors have been reported in pterygium [1]. Several groups, including ours, have demonstrated the expression of matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) in the epithelium overlaying the pterygium, in altered limbal basal epithelial cells and in cultured pterygium fibroblasts [20-22]. These findings demonstrate the dynamic and angiogenic nature of this lesion. Other genes, such as p53, have been reported to be overexpressed in pterygium [13,23,24], although lower levels and monoallelic deletions of p53 have also been reported [25,26].

In addition to studies on pterygium-associated proteins, a cDNA microarray study showed differential gene expression in cultured pterygial fibroblasts compared to cultured normal conjunctival fibroblasts [27]. However, a global gene expression profile of fresh, surgically removed whole pterygium, which includes all component cell types, has not been reported. In the present, more extensive, oligonucleotide microarray study, we identified gene transcripts that are over- or underexpressed in pterygium when compared to normal conjunctiva, limbal, and corneal tissues. We report a number of new genes associated with pterygium, including lipocalin 2 and macrophage inflammatory protein-4 (MIP-4), and show that these proteins have different distributions within the pterygium tissue.

Methods

The use of pterygium specimens was approved by the Institutional Review Board of Kaiser Permanente Medical Center, Sacramento, CA and specimens were handled in accordance with the Declaration of Helsinki. Whole pterygium, which includes both epithelial cells and stromal fibroblasts, and normal residual superior limbal conjunctiva graft tissues were surgically removed and, depending on the subsequent analysis, were immediately frozen, fixed in formalin, or placed in a solution of RNAlater (Ambion, Austin, TX). Corneal and limbal tissues were dissected from donor eyes (NDRI, Philadelpha, PA) and prepared in the same manner.

RNA isolation

Total RNA was extracted from individual specimens with TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA). All RNA samples were treated with 20 U RNase-free DNase 1 (Boehringer Mannheim Biochemicals/Roche Molecular Biochemicals, Indianapolis, IN) to eliminate any contaminating DNA. The concentration of the RNA was measured with a UV spectrophotometer (Shimadzu, Columbia, MD). The quality of the RNA was evaluated by running RNA samples on 2% E-gels (Invitrogen Life Technologies).

Microarray analysis

Total RNA (6 μg) from each of three whole pterygia (two primary and one recurrent) and three autologous conjunctiva was converted to cDNA using the SuperScript Choice System (Invitrogen Life Technologies) and a T7-(dT₂₄) primer (GENESET Corp., San Diego, CA). cDNA was purified using a phenol-chloroform-isoamyl alcohol extraction. The cDNA was used for the in vitro synthesis of biotin-labeled cRNA using the BioArray RNA transcript labeling kit (Enzo Diagnostics, Farmingdale, NY). cRNA was purified using RNeasy mini kits (Qiagen, Valencia, CA) and fragmented using a fragmentation buffer (200 mM Tris-acetate, pH 8.2, 500 mM potassium acetate, and 150 mM magnesium acetate). Labeled cRNA (10 μg) was hybridized for 16 h to Affymetrix human HG_U95Av2 chips, representing over 10,000 genes. Duplicate chips were used for each of the six individual samples. The GeneChips were washed and stained according to the manufacturer's recommendations (Affymetrix, Santa Clara, CA). Each chip was scanned using a confocal laser scanner after an initial staining with streptavidin phycoerythrin, and after a second antibody staining used to amplify the signal.

Data analysis was performed using Affymetrix Microarray Suite 5.0 software to generate an absolute analysis for each chip. The gene signal intensities were analyzed using GeneSpring software (Silicon Genetics, Redwood City, CA). A Welsh t-test was performed to find genes with a p<0.05. This gene list was then filtered to find genes with a 2 fold change between the composite data from pterygium and conjunctival tissue. Genes were clustered according to information available at the National Center for Biotechnology Information (NCBI), National Library of Medicine, NIH.

RT-PCR

Specific gene products were generated with the One Step RT-PCR system (Invitrogen Life Technologies). Primer sets, listed in Table 1, were designed from published human DNA sequences obtained from the NCBI. The linear range for each reaction was determined by varying the number of cycles and RNA concentrations. Subsequently, four different concentrations of total RNA template (Figure 1A) and three separate experimental sets of RNA were used for each gene. Control reactions with heat-inactivated reverse transcriptase were performed to confirm that the RNA was devoid of DNA contamination, and RT-PCR products were sequenced to verify the target gene (CEQ 2000, Beckman Coulter, Fullerton, CA). RT-PCR products were analyzed on 2% E-gels (Invitrogen Life Technologies). The gels were photographed (Syngene, Synoptics Ltd., Cambridge, UK), and the densities of the bands were determined using ImageQuant for Windows NT (Version 5.0) from Molecular Dynamics (Sunnyvale, CA). Mean values and standard deviations were calculated from three separate sets of primary pterygia and normal conjunctiva.

Western blot analysis

Proteins were extracted from primary whole pterygia and from specific regions of the cornea and limbus from donor eyes (NDRI) using T-PER (Pierce, Rockford, IL). Protein concentrations were determined (Bio-Rad Protein Assay, Bio-RAD Laboratories, Hercules, CA). Equal amounts of protein were loaded onto NuPage 3-8% Tris acetate gels for fibronectin and 12% Bis-tris gels for MIP-4 (Invitrogen Life Technologies). Proteins were electrotransferred to nitrocellulose membranes (Bio-Rad), and the blots were then incubated for 1 h in PBS/Tween containing 5% (w/v) nonfat milk powder. The blots were incubated overnight at 4 °C with primary antibodies; monoclonal antibodies for fibronectin (1:4,000; Accurate Chemical & Scientific Corporation, Westbury, NY) and monoclonal antibody for MIP-4 (1:500; Cell Sciences, Canton, MA). After overnight incubation, the blots were washed in PBS/Tween three times for 15 min and then incubated with a 1:60,000 dilution of peroxidase-labeled mouse antibody (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MD) at 37 °C for 1 h. The proteins were visualized using Western Lightning Chemiluminescence Reagent^TM (PerkinElmer Life Sciences, Inc., Boston, MA). Recombinant fibronectin (Sigma, St. Louis, MO) and MIP-4 (Chemicon, Temecula, CA) were used as standards.

Immunohistochemistry

All samples were fixed in 10% neutral buffered formalin and embedded in paraffin. Immunohistochemical staining for fibronectin, lipocalin 2, MIP-4, and lipocalin-type prostaglandin D synthase (L-PGDS) were performed as follows; four μm-thick sections were deparaffinized, treated with 3% hydrogen peroxide to block endogenous peroxidase activity and then treated with 10 mM citrate buffer, (pH 6.0) for 10 min at 100 °C, followed by cooling at room temperature for 20 min. Slides were then incubated with primary antibodies against fibronectin, (1:300, Chemicon), lipocalin 2 (1:250, R&D Systems, Inc., Minneapolis, MN), MIP-4 (1:100, Cell Sciences, Canton, MA), or lipocalin-type prostaglandin D synthase (1:250 Cayman Chemical, Ann Arbor, MI). Tissue sections were incubated with appropriate primary antibody overnight at 4 °C. After washing unbound primary antibody, sections were treated with commercial biotinylated secondary anti-immunoglobulin, followed by avidin coupled to biotinylated horseradish peroxidase, at room temperature, according to the manufacturer's instructions (LSAB2kit, DAKO Corporation, Carpinteria, CA). Immunohistochemical reactions were developed with diaminobenzidine as the chromogenic peroxidase substrate. Sections were counterstained with hematoxylin. No staining was observed in tissues without primary antibody.

Results

Microarray analysis

Microarray analysis was performed on three separate whole pterygia (two primary and one recurrent) and three superior conjunctiva samples that had been surgically removed from the same patients. Thirty-four expressed genes, common to both primary and recurrent pterygium, demonstrated a 2 fold or greater difference between ptergium and autologous conjunctiva (Table 2). Twenty-nine genes were overexpressed and five genes were underexpressed in pterygium.

Upregulated genes included proteins associated with cell adhesion and migration, extracellular matrix, inflammation, angiogenesis, and oncogenesis. Fibronectin, a glycoprotein involved in cell adhesion and migration, was upregulated 9 fold at the mRNA level in pterygium compared to conjunctiva. MIP-4, a chemokine and T cell chemoattractant, was upregulated 5 fold. Two members of the lipocalin family, retinol binding protein 1 and lipocalin 2 (oncogene 24p3; NGAL), were increased by 3.6 and 2.4 fold, respectively.

Differentially expressed genes were clustered into functional categories (Table 3). The category having the greatest number of overexpressed genes is a cluster of cell adhesion and extracellular matrix proteins comprised of fibronectin, collagen type III, versican, extracellular matrix protein 1, von Willebrand factor, collagen type VI, CD31 antigen, and cartilage oligomeric matrix protein.

Confirmation by RT-PCR

Gene expression in three primary whole pterygia and the autologous conjunctiva were evaluated by RT-PCR. As with the microarray data, fibronectin, lipocalin, and MIP-4 gene transcripts were more abundant and ephrin-A1 and Per1 were less abundant in pterygia compared to conjunctiva (Figure 1B, Table 4). The levels of expression for these genes were similar with the two methods (Table 4) with the exception of MIP-4, which potentially could be far greater than 5 fold. Some variability among samples was observed for Per1 and CLIC2. Statistical analysis indicated that Per1 gene transcripts were significantly decreased in pterygium compared to conjunctiva, while CLIC2 gene transcript levels were not significantly different between the two tissues. In contrast to the microarray data, RT-PCR showed no difference in carbonic anhydrase gene expression between pterygium and normal conjunctiva.

UV-damaged limbal stem cells are thought to give rise to pterygia based on shared cytokeratin and vimentin characteristics, although this continues to be an area of active investigation. To expand our comparative study to include putative pterygium progenitor cells, we examined limbal epithelium and nonepithelial cells plus other areas of the cornea. Three pterygia and dissected tissues from three separate donor eyes were compared (Figure 1C). Fibronectin mRNA levels were higher in pterygium compared to limbal epithelium, central cornea epithelium, limbal nonepithelium, central cornea nonepithelium, and normal conjunctiva. Per1 mRNA levels were lower in pterygium compared to limbal epithelium, central cornea nonepithelium, and normal conjunctiva. Per1 mRNA was not detected in central cornea epithelium or in limbal nonepithelium.

Western blot analysis

Antibodies to fibronectin and MIP-4 were used to examine the level of protein expression in primary whole pterygium, cornea, and limbus. Western blot analysis showed that fibronectin (Figure 2B) and MIP-4 (Figure 2C) were increased in pterygium samples compared to limbal epithelium, central cornea epithelium, limbal nonepithelium, central cornea nonepithelium, and normal conjunctiva (Figure 2). The protein expression profiles for fibronectin and MIP-4 were consistent with the RT-PCR results (Figure 1).

Immunohistochemistry

Immunohistochemical analysis showed specific, increased staining of fibronectin, lipocalin 2, and MIP-4 in primary pterygium relative to cornea (Figure 3). Abundant fibronectin staining was noted along the entire length of the pterygium stroma and in some stromal fibroblasts (Figure 3B). Lipocalin 2 immunoreactivity was localized most intensely to basal epithelial cells, basement membrane, and to some stromal extracellular matrix (Figure 3D). Staining was also observed in normal corneal basal epithelial cells (Figure 3C). MIP-4 was widely distributed in the stroma and the overlying epithelial cell layers with expression in both the mature and immature epithelial cells (Figure 3F). Since the existence of lipocalin-type prostaglandin D synthase (L-PGDS) in a number of ocular tissues and in some tumors has previously been reported, it was included in the imunohistochemical study. The protein was differentially localized to basal epithelial cells (Figure 3H).

Discussion

These results have led to the identification of a set of differentially expressed genes in pterygium that may be important in the pathogenesis of this disease. Significant, and in several cases, substantial differences in expression were observed between whole pterygium (which is comprised of epithelial cells and fibroblasts) and normal conjunctiva, and other corneal and limbal regions. The list of genes (Table 2) provides novel targets for future investigations.

Our study confirms that pterygium is associated with oncogenic, angiogenic, fibrogenic, and inflammatory factors and, further, provides evidence of underrepresented genes that may contribute to pathogenesis.

A number of cell-adhesion and extracellular-matrix-related genes were increased in pterygium, including fibronectin, versican, extracellular matrix protein 1, cartilage oligomeric matrix protein, and collagen III and VI. The presence of fibronectin and collagen III is consistent with the immunohistochemical findings of Naib-Majani et al. [28]. Also, there is fair agreement on fibronectin localization. We found strong antibody staining of fibronectin in the stroma, as did Naib-Majani et al. [28], although they also reported staining in the basal epithelial cells.

Damage to the cornea by debridement or incision leads to a rapid accumulation of fibronectin. This adhesive glycoprotein functions as a temporary matrix for corneal epithelial cell attachment and migration [29-32]. After corneal resurfacing, fibronectin disappears. It seems plausible that the accumulation of fibronectin in the fibrovascular stroma of the pterygium serves the same function, providing a substratum for pterygial cell migration.

Fibronectin has also been associated with malignant transformation and resistance to apoptosis of tumors and may play a similar role in pterygium. Early signs of malignant transformation include the production of fibronectin by the peritumoral stroma [33]. Accumulation of fibronectin has been noted in a number of malignancies including breast, lung, and pancreatic cancers [34-36]. Han et al. [35] showed that addition of fibronectin to cultured human lung carcinoma cells increased proliferation and substantially decreased apoptosis, reflected by the reduced amount of DNA fragmentation in cells cultured in serum-free medium. Pterygium is considered a benign neoplastic lesion with characteristics of transformed cells including loss of heterozygosity and microsatellite instability [26,37,38]. In addition, pterygium displays reduced apoptosis with increased levels of the apoptosis-inhibiting protein bcl-2 [16]. It is possible that the increased level of fibronectin in pterygium facilitates cell proliferation with reduced apoptosis.

Lipocalins are extracellular proteins that bind and transport a variety of low-molecular-weight molecules, including iron and lipophilic ligands such as retinoids, steroids, pheromones, and odorants [39-41]. Because of the diversity of ligands transported, lipocalins can have an impact upon a wide range of cellular functions. Lipocalins may play a biophysical role in some tissues. For example, it has been suggested that human tear lipocalin (TL), which accounts for about 15-33% of the tear proteins, increases the solubility of lipids in the tear film and reduces evaporation of water from the corneal surface [42-45]. Some lipocalins also serve as biochemical markers of inflammatory diseases, cancer, and lipid disorders [46].

Our microarray data indicates that retinol-binding protein 1 and lipocalin 2 are associated with pterygium. Another lipocalin family member, L-PGDS, a PGD₂-producing enzyme and retinoid transporter, has previously been reported in a number of ocular tissues and in some tumors [47-51]. L-PGDS represents the most abundant species in not only an unnormalized NEIBank mouse rpe/choroid cDNA library (NbLib0059) but in a human corneal endothelial cDNA library as well [52]. We therefore examined the immunohistochemical distribution of this lipocalin and found it to be differentially localized to pterygial basal epithelial cells.

Immunostaining of lipocalin 2 (oncogene 24p3) was increased in pterygium and localized to the basal pterygial epithelial cells, basement membrane, and some areas of the extracellular stroma. It is known to be highly expressed in colonic epithelium in areas of inflammation, both in nonmalignant epithelium (diverticulitis, inflammatory bowel disease, and appendicitis) and in premalignant and malignant neoplastic lesions of the colon [53]. It is also associated with hepatic, pancreatic, breast, and ovarian cancers [54-57].

Lipocalin 2 is also known as neutrophil gelatinase-associated lipocalin (NGAL) [58,59], due to its binding and activation of latent matrix metalloproteinases, in particular MMP-8 and MMP-9 [39,56,59,60]. As such, it has been proposed to play a role in tissue remodeling and metastasis. It is noteworthy that NGAL is highly expressed in the pterygium basal epithelial cells, considering that we previously observed strong immunostaining of a number of matrix metalloproteinases, including MMP-9, in the same cellular location [21]. The colocalization suggests that NGAL may augment the MMP-driven dissolution of Bowman's layer and invasion by limbal basal epithelial cells.

The chemokine MIP-4, also called pulmonary and activation-regulated chemokine (PARC), dendritic cell derived C-C chemokine (DCCK) 1, and alternative macrophage activation associated C-C chemokine (AMAC) [61], was highly expressed in pterygium, providing further evidence of the proinflammatory nature of the lesion [1]. MIP-4 functions as a chemoattractant for CD4⁺ and CD8⁺ T lymphocyte subpopulations (indicating humoral and cell-mediated regulation), with no effect on activated T lymphocytes, granulocytes, or monocytes [61]. It has been reported in a number of tissues, including alveolar macrophages and dendritic cells, which are potent antigen presenting cells [62,63]. Langerhans' cells, which belong to the dendritic mononuclear cell family, and macrophages have been identified in pterygium [64] and could be the source of secreted MIP-4. The strong staining of this protein in the fairly homogeneous epithelial layer, however, seems to suggest that these cells may also secrete MIP-4.

Blood vessels are a major component of the pterygium stroma. Our microarray data indicated a 2 fold increase in CD31 antigen, which is involved in intercellular adhesion of endothelial cells [65,66]. We also observed a 2.5 fold increase in von Willebrand factor in pterygium compared to conjunctiva. Our findings are in general agreement with Naib-Majani et al. [28], who detected this protein in pterygium blood vessel walls. von Willebrand factor is essential for platelet adhesion, and it binds directly to collagen III [67]. Collagen III is also overexpressed in pterygium.

Disruption of circadian rhythm has been associated with malignancy. Per1 is a core component of the circadian system and was underexpressed in pterygium compared to conjunctiva, limbal epithelium, and central cornea nonepithelium. Expression of Per1 is significantly decreased in endometrial carcinoma and in breast cancer [68,69]. Matsuo et al. [70] showed that the circadian clock controls the expression of cell cycle-related genes, and it has been suggested that the decrease in Per1 deregulates the cell cycle, favoring proliferation [68-70].

Our microarray data confirmed a number of genes that had previously been reported in pterygium. In addition, this global gene expression approach identified many new genes that were either over- or underrepresented in pterygium. In particular, several cell adhesion/extracellular matrix proteins and proteins involved in inflammation and malignant transformation had increased expression levels. The finding that lipocalin 2 (NGAL) is overexpressed in pterygium is consistent with the functional activation of MMPs by NGAL. MMPs are matrix-degrading enzymes that have been implicated in the pathogenesis of pterygium. Inhibition or silencing of NGAL might inactivate this pathway and the progression of pterygium. Other areas that could be addressed include further evaluation of these genes in recurrent pterygium and identification of genes that are singularly representative of primary or recurrent pterygium or of selected morphological areas such as the epithelial cell leading edge.

Acknowledgements

This research was supported in part by the Intramural Research Program of the NIH, National Eye Institute.

References

1. Di Girolamo N, Chui J, Coroneo MT, Wakefield D. Pathogenesis of pterygia: role of cytokines, growth factors, and matrix metalloproteinases. Prog Retin Eye Res 2004; 23:195-228.

2. Taylor HR, West S, Munoz B, Rosenthal FS, Bressler SB, Bressler NM. The long-term effects of visible light on the eye. Arch Ophthalmol 1992; 110:99-104.

3. Wong TY, Foster PJ, Johnson GJ, Seah SK, Tan DT. The prevalence and risk factors for pterygium in an adult Chinese population in Singapore: the Tanjong Pagar survey. Am J Ophthalmol 2001; 131:176-83.

4. Wlodarczyk J, Whyte P, Cockrum P, Taylor H. Pterygium in Australia: a cost of illness study. Clin Experiment Ophthalmol 2001; 29:370-5.

5. Wu K, He M, Xu J, Li S. Pterygium in aged population in Doumen County, China. Yan Ke Xue Bao 2002; 18:181-4.

6. Taylor HR, West SK, Rosenthal FS, Munoz B, Newland HS, Emmett EA. Corneal changes associated with chronic UV irradiation. Arch Ophthalmol 1989; 107:1481-4.

7. Lin A, Stern G. Correlation between pterygium size and induced corneal astigmatism. Cornea 1998; 17:28-30.

8. Hirst LW. The treatment of pterygium. Surv Ophthalmol 2003; 48:145-80.

9. Tan DT, Chee SP, Dear KB, Lim AS. Effect of pterygium morphology on pterygium recurrence in a controlled trial comparing conjunctival autografting with bare sclera excision. Arch Ophthalmol 1997; 115:1235-40. Erratum in: Arch Ophthalmol 1998; 116:552.

10. Lewallen S. A randomized trial of conjunctival autografting for pterygium in the tropics. Ophthalmology 1989; 96:1612-4.

11. Ma DH, See LC, Hwang YS, Wang SF. Comparison of amniotic membrane graft alone or combined with intraoperative mitomycin C to prevent recurrence after excision of recurrent pterygia. Cornea 2005; 24:141-50.

12. Ti SE, Chee SP, Dear KB, Tan DT. Analysis of variation in success rates in conjunctival autografting for primary and recurrent pterygium. Br J Ophthalmol 2000; 84:385-9.

13. Dushku N, Reid TW. Immunohistochemical evidence that human pterygia originate from an invasion of vimentin-expressing altered limbal epithelial basal cells. Curr Eye Res 1994; 13:473-81.

14. Nolan TM, DiGirolamo N, Sachdev NH, Hampartzoumian T, Coroneo MT, Wakefield D. The role of ultraviolet irradiation and heparin-binding epidermal growth factor-like growth factor in the pathogenesis of pterygium. Am J Pathol 2003; 162:567-74.

15. Tsai YY, Cheng YW, Lee H, Tsai FJ, Tseng SH, Lin CL, Chang KC. Oxidative DNA damage in pterygium. Mol Vis 2005; 11:71-5 <http://www.molvis.org/molvis/v11/a8/>.

16. Tan DT, Tang WY, Liu YP, Goh HS, Smith DR. Apoptosis and apoptosis related gene expression in normal conjunctiva and pterygium. Br J Ophthalmol 2000; 84:212-6.

17. Dushku N, Hatcher SL, Albert DM, Reid TW. p53 expression and relation to human papillomavirus infection in pingueculae, pterygia, and limbal tumors. Arch Ophthalmol 1999; 117:1593-9.

18. Gallagher MJ, Giannoudis A, Herrington CS, Hiscott P. Human papillomavirus in pterygium. Br J Ophthalmol 2001; 85:782-4.

19. Chen KH, Hsu WM, Cheng CC, Li YS. Lack of human papillomavirus in pterygium of Chinese patients from Taiwan. Br J Ophthalmol 2003; 87:1046-8.

20. Di Girolamo N, Wakefield D, Coroneo MT. Differential expression of matrix metalloproteinases and their tissue inhibitors at the advancing pterygium head. Invest Ophthalmol Vis Sci 2000; 41:4142-9.

21. Dushku N, John MK, Schultz GS, Reid TW. Pterygia pathogenesis: corneal invasion by matrix metalloproteinase expressing altered limbal epithelial basal cells. Arch Ophthalmol 2001; 119:695-706. Erratum in: Arch Ophthalmol 2002; 120:234-7.

22. Li DQ, Lee SB, Gunja-Smith Z, Liu Y, Solomon A, Meller D, Tseng SC. Overexpression of collagenase (MMP-1) and stromelysin (MMP-3) by pterygium head fibroblasts. Arch Ophthalmol 2001; 119:71-80.

23. Dushku N, Reid TW. P53 expression in altered limbal basal cells of pingueculae, pterygia, and limbal tumors. Curr Eye Res 1997; 16:1179-92.

24. Tan DT, Lim AS, Goh HS, Smith DR. Abnormal expression of the p53 tumor suppressor gene in the conjunctiva of patients with pterygium. Am J Ophthalmol 1997; 123:404-5.

25. Onur C, Orhan D, Orhan M, Dizbay Sak S, Tulunay O, Irkec M. Expression of p53 protein in pterygium. Eur J Ophthalmol 1998; 8:157-61.

26. Reisman D, McFadden JW, Lu G. Loss of heterozygosity and p53 expression in Pterygium. Cancer Lett 2004; 206:77-83.

27. Solomon A, Grueterich M, Li DQ, Meller D, Lee SB, Tseng SC. Overexpression of Insulin-like growth factor-binding protein-2 in pterygium body fibroblasts. Invest Ophthalmol Vis Sci 2003; 44:573-80.

28. Naib-Majani W, Eltohami I, Wernert N, Watts W, Tschesche H, Pleyer U, Breipohl W. Distribution of extracellular matrix proteins in pterygia: an immunohistochemical study. Graefes Arch Clin Exp Ophthalmol 2004; 242:332-8.

29. Nishida T, Nakagawa S, Awata T, Ohashi Y, Watanabe K, Manabe R. Fibronectin promotes epithelial migration of cultured rabbit cornea in situ. J Cell Biol 1983; 97:1653-7.

30. Watanabe K, Nakagawa S, Nishida T. Chemotactic and haptotactic activities of fibronectin for cultured rabbit corneal epithelial cells. Invest Ophthalmol Vis Sci 1988; 29:572-7.

31. Nakamura M, Nishida T. Synergistic effects of hyaluronan and fibronectin on epithelial migration in rabbit cornea in vitro. Cornea 1999; 18:686-92.

32. Tanaka T, Furutani S, Nakamura M, Nishida T. Changes in extracellular matrix components after excimer laser photoablation in rat cornea. Jpn J Ophthalmol 1999; 43:348-54.

33. Labat-Robert J. Fibronectin in malignancy. Semin Cancer Biol 2002; 12:187-95.

34. Jiang Y, Harlocker SL, Molesh DA, Dillon DC, Stolk JA, Houghton RL, Repasky EA, Badaro R, Reed SG, Xu J. Discovery of differentially expressed genes in human breast cancer using subtracted cDNA libraries and cDNA microarrays. Oncogene 2002; 21:2270-82.

35. Han S, Sidell N, Roser-Page S, Roman J. Fibronectin stimulates human lung carcinoma cell growth by inducing cyclooxygenase-2 (COX-2) expression. Int J Cancer 2004; 111:322-31.

36. Gress TM, Muller-Pillasch F, Lerch MM, Friess H, Buchler M, Adler G. Expression and in-situ localization of genes coding for extracellular matrix proteins and extracellular matrix degrading proteases in pancreatic cancer. Int J Cancer 1995; 62:407-13.

37. Detorakis ET, Sourvinos G, Tsamparlakis J, Spandidos DA. Evaluation of loss of heterozygosity and microsatellite instability in human pterygium: clinical correlations. Br J Ophthalmol 1998; 82:1324-8.

38. Spandidos DA, Sourvinos G, Kiaris H, Tsamparlakis J. Microsatellite instability and loss of heterozygosity in human pterygia. Br J Ophthalmol 1997; 81:493-6.

39. Coles M, Diercks T, Muehlenweg B, Bartsch S, Zolzer V, Tschesche H, Kessler H. The solution structure and dynamics of human neutrophil gelatinase-associated lipocalin. J Mol Biol 1999; 289:139-57.

40. Yang J, Mori K, Li JY, Barasch J. Iron, lipocalin, and kidney epithelia. Am J Physiol Renal Physiol 2003; 285:F9-18.

41. Schlehuber S, Skerra A. Lipocalins in drug discovery: from natural ligand-binding proteins to "anticalins". Drug Discov Today 2005; 10:23-33.

42. Delaire A, Lassagne H, Gachon AM. New members of the lipocalin family in human tear fluid. Exp Eye Res 1992; 55:645-7.

43. Gachon AM, Richard J, Dastugue B. Human tears: normal protein pattern and individual protein determinations in adults. Curr Eye Res 1982-83; 2:301-8.

44. Gachon AM, Lacazette E. Tear lipocalin and the eye's front line of defence. Br J Ophthalmol 1998; 82:453-5.

45. Glasgow BJ, Abduragimov AR, Gasymov OK, Yusifov TN. Tear lipocalin: structure, function and molecular mechanisms of action. Adv Exp Med Biol 2002; 506:555-65.

46. Xu S, Venge P. Lipocalins as biochemical markers of disease. Biochim Biophys Acta 2000; 1482:298-307.

47. Beuckmann CT, Gordon WC, Kanaoka Y, Eguchi N, Marcheselli VL, Gerashchenko DY, Urade Y, Hayaishi O, Bazan NG. Lipocalin-type prostaglandin D synthase (beta-trace) is located in pigment epithelial cells of rat retina and accumulates within interphotoreceptor matrix. J Neurosci 1996; 16:6119-24.

48. Gerashchenko DY, Beuckmann CT, Marcheselli VL, Gordon WC, Kanaoka Y, Eguchi N, Urade Y, Hayaishi O, Bazan NG. Localization of lipocalin-type prostaglandin D synthase (beta-trace) in iris, ciliary body, and eye fluids. Invest Ophthalmol Vis Sci 1998; 39:198-203.

49. Berryhill BL, Beales MP, Hassell JR. Production of prostaglandin D synthase as a keratan sulfate proteoglycan by cultured bovine keratocytes. Invest Ophthalmol Vis Sci 2001; 42:1201-7.

50. Su B, Guan M, Zhao R, Lu Y. Expression of prostaglandin D synthase in ovarian cancer. Clin Chem Lab Med 2001; 39:1198-203.

51. Yamashima T, Sakuda K, Tohma Y, Yamashita J, Oda H, Irikura D, Eguchi N, Beuckmann CT, Kanaoka Y, Urade Y, Hayaishi O. Prostaglandin D synthase (beta-trace) in human arachnoid and meningioma cells: roles as a cell marker or in cerebrospinal fluid absorption, tumorigenesis, and calcification process. J Neurosci 1997; 17:2376-82.

52. Sakai R, Kinouchi T, Kawamoto S, Dana MR, Hamamoto T, Tsuru T, Okubo K, Yamagami S. Construction of human corneal endothelial cDNA library and identification of novel active genes. Invest Ophthalmol Vis Sci 2002; 43:1749-56.

53. Nielsen BS, Borregaard N, Bundgaard JR, Timshel S, Sehested M, Kjeldsen L. Induction of NGAL synthesis in epithelial cells of human colorectal neoplasia and inflammatory bowel diseases. Gut 1996; 38:414-20.

54. Bratt T. Lipocalins and cancer. Biochim Biophys Acta 2000; 1482:318-26.

55. Furutani M, Arii S, Mizumoto M, Kato M, Imamura M. Identification of a neutrophil gelatinase-associated lipocalin mRNA in human pancreatic cancers using a modified signal sequence trap method. Cancer Lett 1998; 122:209-14.

56. Yan L, Borregaard N, Kjeldsen L, Moses MA. The high molecular weight urinary matrix metalloproteinase (MMP) activity is a complex of gelatinase B/MMP-9 and neutrophil gelatinase-associated lipocalin (NGAL). Modulation of MMP-9 activity by NGAL. J Biol Chem 2001; 276:37258-65.

57. Bartsch S, Tschesche H. Cloning and expression of human neutrophil lipocalin cDNA derived from bone marrow and ovarian cancer cells. FEBS Lett 1995; 357:255-9.

58. Hraba-Renevey S, Turler H, Kress M, Salomon C, Weil R. SV40-induced expression of mouse gene 24p3 involves a post-transcriptional mechanism. Oncogene 1989; 4:601-8.

59. Kjeldsen L, Johnsen AH, Sengelov H, Borregaard N. Isolation and primary structure of NGAL, a novel protein associated with human neutrophil gelatinase. J Biol Chem 1993; 268:10425-32.

60. Tschesche H, Zolzer V, Triebel S, Bartsch S. The human neutrophil lipocalin supports the allosteric activation of matrix metalloproteinases. Eur J Biochem 2001; 268:1918-28.

61. Guan P, Burghes AH, Cunningham A, Lira P, Brissette WH, Neote K, McColl SR. Genomic organization and biological characterization of the novel human CC chemokine DC-CK-1/PARC/MIP-4/SCYA18. Genomics 1999; 56:296-302.

62. Adema GJ, Hartgers F, Verstraten R, de Vries E, Marland G, Menon S, Foster J, Xu Y, Nooyen P, McClanahan T, Bacon KB, Figdor CG. A dendritic-cell-derived C-C chemokine that preferentially attracts naive T cells. Nature 1997; 387:713-7.

63. Hieshima K, Imai T, Baba M, Shoudai K, Ishizuka K, Nakagawa T, Tsuruta J, Takeya M, Sakaki Y, Takatsuki K, Miura R, Opdenakker G, Van Damme J, Yoshie O, Nomiyama H. A novel human CC chemokine PARC that is most homologous to macrophage-inflammatory protein-1 alpha/LD78 alpha and chemotactic for T lymphocytes, but not for monocytes. J Immunol 1997; 159:1140-9.

64. Ioachim-Velogianni E, Tsironi E, Agnantis N, Datseris G, Psilas K. HLA-DR antigen expression in pterygium epithelial cells and lymphocyte subpopulations: an immunohistochemistry study. Ger J Ophthalmol 1995; 4:123-9.

65. Albelda SM, Oliver PD, Romer LH, Buck CA. EndoCAM: a novel endothelial cell-cell adhesion molecule. J Cell Biol 1990; 110:1227-37.

66. Newman PJ, Berndt MC, Gorski J, White GC 2nd, Lyman S, Paddock C, Muller WA. PECAM-1 (CD31) cloning and relation to adhesion molecules of the immunoglobulin gene superfamily. Science 1990; 247:1219-22.

67. Savage B, Sixma JJ, Ruggeri ZM. Functional self-association of von Willebrand factor during platelet adhesion under flow. Proc Natl Acad Sci U S A 2002; 99:425-30.

68. Chen ST, Choo KB, Hou MF, Yeh KT, Kuo SJ, Chang JG. Deregulated expression of the PER1, PER2 and PER3 genes in breast cancers. Carcinogenesis 2005; 26:1241-6.

69. Yeh KT, Yang MY, Liu TC, Chen JC, Chan WL, Lin SF, Chang JG. Abnormal expression of period 1 (PER1) in endometrial carcinoma. J Pathol 2005; 206:111-20.

70. Matsuo T, Yamaguchi S, Mitsui S, Emi A, Shimoda F, Okamura H. Control mechanism of the circadian clock for timing of cell division in vivo. Science 2003; 302:255-9.

Typographical corrections