July 2011
Volume 52, Issue 8
Free
Cornea  |   July 2011
Activation of MAPK Signaling Pathway and NF-κB Activation in Pterygium and Ipsilateral Pterygium-Free Conjunctival Specimens
Author Affiliations & Notes
  • Jenice Torres
    From the IOBA, University of Valladolid, Valladolid, Spain;
  • Amalia Enríquez-de-Salamanca
    From the IOBA, University of Valladolid, Valladolid, Spain;
    CIBER-BBN Valladolid, Valladolid, Spain;
  • Itziar Fernández
    From the IOBA, University of Valladolid, Valladolid, Spain;
    CIBER-BBN Valladolid, Valladolid, Spain;
  • Maria Teresa Rodríguez-Ares
    University of Santiago de Compostela, Compostela, Spain;
  • Maria J. Quadrado
    IBILI, University of Coimbra, Coimbra, Portugal;
  • Joaquim Murta
    IBILI, University of Coimbra, Coimbra, Portugal;
  • José M. Benítez del Castillo
    Ramón Castroviejo Institute, Complutense University, Madrid, Spain; and
  • Michael E. Stern
    Allergan Inc., Irvine, California.
  • Margarita Calonge
    From the IOBA, University of Valladolid, Valladolid, Spain;
    CIBER-BBN Valladolid, Valladolid, Spain;
  • Corresponding author: Amalia Enríquez-de-Salamanca, IOBA, Universidad de Valladolid, Campus Miguel Delibes, Paseo de Belén, 17, Valladolid E-47011, Spain; [email protected]
  • Footnotes
    2  These authors contributed equally to the work presented here and should therefore be regarded as equivalent authors.
Investigative Ophthalmology & Visual Science July 2011, Vol.52, 5842-5852. doi:https://doi.org/10.1167/iovs.10-6673
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jenice Torres, Amalia Enríquez-de-Salamanca, Itziar Fernández, Maria Teresa Rodríguez-Ares, Maria J. Quadrado, Joaquim Murta, José M. Benítez del Castillo, Michael E. Stern, Margarita Calonge; Activation of MAPK Signaling Pathway and NF-κB Activation in Pterygium and Ipsilateral Pterygium-Free Conjunctival Specimens. Invest. Ophthalmol. Vis. Sci. 2011;52(8):5842-5852. https://doi.org/10.1167/iovs.10-6673.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To evaluate mitogen-activated protein kinases (MAPKs) and nuclear factor-κB (NF-κB) signaling pathways in pterygium and pterygium-free conjunctivas.

Methods.: Primary pterygia (n = 21), ipsilateral superior-temporal bulbar conjunctivas (n = 8), and healthy conjunctival (n = 5) biopsy specimens were analyzed. Total and phosphorylated (phospho) levels of extracellular-regulated 1/2 (ERK1/2), p38, and c-jun N-terminal (JNK) MAPKs and NF-κB inhibitor-alpha (IκΒ-α) were analyzed by immunobead-based assay. Tissue phospho-, total protein, and activation values determined by phospho/total ratios were compared. Correlation among those values and clinical parameters were determined. Average-linkage hierarchical cluster analysis identified patients with similar protein activation values. The k-nearest neighbor classifier predicted the origin of specimens based on protein levels.

Results.: Pterygium samples had significantly lower total JNK and IκΒ-α levels than did healthy conjunctivas. Decreased total JNK and IκΒ-α and increased phospho-IκΒ-α levels and phospho/total ratio of JNK and IκΒ-α were present in ipsilateral conjunctivas compared with healthy conjunctivas. Protein levels were correlated among them in pterygium, ipsilateral, and healthy conjunctivas and with sun exposure, pterygium grade, and pterygium measurements. Cluster analysis of activation values and ratios in pterygium and ipsilateral-conjunctiva revealed different groups of patients with similar values. Prediction accuracy was 70% to 80% for the classifiers phospho- and total protein levels and phospho/total ratio.

Conclusions.: Pterygium and pterygium-free ipsilateral conjunctivas had alterations in MAPK and NF-κB pathways not present in healthy conjunctivas. The high prediction accuracy based on phospho- and total protein levels and phospho/total ratio of ERK1/2, p38, JNK, and IκB-α suggests these molecules as potential biomarkers of inflammation in pterygia.

Pterygium is a proliferative and invasive disease of the human ocular surface particularly prevalent in sun-exposed persons. 1,2 It appears as a fleshy fibrovascular growth involving the interpalpebral area, typically triangular in shape, and consists of a cap, head, and body. The lesion is characterized by centripetal growth of a leading edge of altered limbal epithelial cells, followed by squamous cell metaplasia with goblet cell hyperplasia. Activated stromal fibroblasts, neovascularization, and extracellular matrix remodeling are also present. 3 Pterygium has inflammatory and infiltrative processes that are composed of neutrophils, 4 mast cells, 5,6 and lymphocytes. 5,7 It can also present a vascular reaction 5,8 that is likely to be exacerbated by excessive cytokines, 9 11 growth factors, and growth factor receptor–associated signaling pathways. 11 16 Gene expression studies in pterygium specimens 17,18 have revealed that pathways associated with increases in extracellular matrix, structural and mitotic proteins, and proteins involved in tissue invasion are significantly affected. 
However, the mechanism of pterygium formation is not yet fully understood, and its origin remains controversial. One of the most widely accepted theories involves ultraviolet radiation (UV) in the pathogenesis, 3,19 21 and this has been confirmed with relevant data implicating UV exposure as a probable trigger for this lesion. 1,2 On UV-Β exposure, the expression of several key effector molecules, such as cytokines, growth factors, and matrix metalloproteinases (MMPs), that are likely to participate in the inflammatory, proliferative, and remodeling phases of this disease, is increased. 3,5 Other studies regarding UV-B effects on pterygium and pterygial epithelial cells in culture 4,9,10,22 have shown that the UV-B–mediated induction of cytokines, growth factors, and MMP-1 involves the phosphorylation of epidermal growth factor receptor and activation of mitogen-activated protein kinase (MAPK) pathways. 
Activation of epidermal growth factor receptor and MAPK initiates the downstream induction of transcription factors, such as nuclear factor-κB (NF-κB). In the classical NF-κB signaling pathway, on activation, NF-κB inhibitor proteins (IκB) that are bound to NF-κB in the cytoplasm are phosphorylated. This phosphorylation triggers their polyubiquitination and their subsequent degradation by the 26S proteasome, releasing free NF-κB dimers that are then translocated to the nucleus. NF-κB is an essential regulator for the expression of numerous genes involved in the function and development of the immune system and in inflammatory and acute responses. 13,23,24 Additionally, NF-κB activation mediates some cellular responses to UV damage and hyperosmotic stress. 25,26 For cultured pterygium cells, the MAPK pathway is activated by UV-B exposure, 22 adding further support to growing evidence of an inflammatory component in pterygia. However, to our knowledge, MAPK and NF-κB activation levels in pterygium specimens have not been correlated with clinical parameters. Furthermore, ipsilateral pterygium-free conjunctiva from pterygium-affected patients is routinely used in pterygium surgery. Although it is assumed that the transplanted tissue is composed of normal conjunctiva, no studies document the expression of inflammatory signals in it. 
The aim of this study was to evaluate signs of inflammation present in both pterygium and ipsilateral pterygium-free conjunctiva specimens. We measured the activation of the MAPK signaling pathway and NF-κB in isolated tissues and sought to determine whether there was any correlation between the levels of expression of those molecules and clinical parameters. 
Patients, Materials, and Methods
Patients and Tissues
This study was approved by the IOBA Ethics Committee of the University of Valladolid and by institutional review boards at each clinical center involved (Hospital Xeral, University of Santiago de Compostela, Spain; IBILI, University of Coimbra, Portugal; and Castroviejo Institute, Complutense University, Madrid, Spain). The protocol followed the tenets of the Declaration of Helsinki, and written informed consent was obtained from each patient. 
The inclusion criteria consisted of the following: patients of at least 18 years of age, presence of primary pterygium exclusively in the nasal interpalpebral area, absence of redness or any other sign of inflammation other that in the area affected by the pterygium, and duration of at least 2 years. Exclusion criteria were major systemic diseases (diabetes mellitus, collagen vascular diseases), ocular surface disease other than pterygium or intraocular diseases, glaucoma or any previous ocular surgery, known trauma to the affected eye, contact lens wear, or chronic use of any topical medications. If patients were using a short-term topical drug, a washout period of at least 30 days was required before pterygium excision. Eyes with only temporal or with both nasal and temporal pterygia were excluded. Primary pterygia invading the cornea <2 mm, including the head plus the avascular fibrotic edge, were also excluded. 
A comprehensive medical and ocular history was obtained before surgery. All patients were asked about their country of origin if other than Spain or Portugal and whether they had spent >50% of their lives there. Additionally, they were asked whether they worked mainly (more than the 50% of the time) outdoors and whether they usually wore sunglasses during their routine, both at work and for casual wear. 
As described by Tan et al., 27 pterygia were graded according to the following classification: grade 1, atrophic with episcleral vessels under the body of the pterygium that were not obscured and clearly distinguished; grade 2, intermediate and including all other pterygia not falling into grade 1 or 3; grade 3, fleshy with episcleral vessels totally obscured. 
After routine standard ophthalmologic sterile preparation and draping, the size of the pterygium was measured in millimeters with a surgical compass immediately before excision. The head and cap of the pterygium entering the cornea was measured on the vertical meridian at the limbus and on the horizontal meridian from the limbus to the leading edge. The body from the limbus to the end of the tail was also measured at the widest vertical meridian and horizontally at 3 or 9 o'clock. 
Pterygia were removed by a standard surgical technique only when the ocular surface was free of inflammation other than in the area of pterygium. The upper half (including head and body) was sent to us, and the lower half was sent to the pathology laboratory of each referring institution as legally mandated. Subsequent autologous limbal-conjunctival autograft transplantation was performed to cover the bare sclera. This conjunctiva was always excised from the ipsilateral superior-temporal quadrant (approximately 6 × 6 mm). Additionally, we attempted to obtain a 2 × 2-mm piece from the contiguous nasal edge of the ipsilateral pterygium-free conjunctiva as a separate sample. Thirteen of these biopsy specimens were smaller than required, and the protein content was insufficient for assay. All valid biopsy specimens came from the same clinical center and were excised by the same surgeon. 
For control tissue, superior-temporal conjunctival biopsy specimens (n = 5) were obtained from healthy donors undergoing cataract surgery. These pterygium-free donors were selected based on the same inclusion/exclusion criteria described, and they had no other previous ocular disease. These biopsy specimens were from the same region as the pterygium-free conjunctivas in pterygium patients, obtained for autologous limbal-conjunctival autograft transplantation. 
Thus, for this study we analyzed the following tissues: the upper half of the head and body of extirpated primary nasal pterygia (n = 21); control ipsilateral superior-temporal bulbar pterygium-free conjunctivas (n = 8) from diseased eyes but not macroscopically involved in the lesion; and healthy conjunctivas from healthy donors (n = 5) as absolute normal controls. To biopsy the healthy conjunctiva from the fellow eye of pterygium patients was not permitted by the Ethics Committee. Even if it were permitted, it would not have represented a genuinely healthy control because pterygium is often bilateral. Thus, complete normality could not have been assumed in the contralateral eyes of these patients. 
In all cases, after the tissues were surgically removed, they were washed with balanced saline solution, and kept at −20°C until sent to our laboratory whereupon they were stored at −80°C until used. 
Determination of Phosphorylated and Total Levels of Extracellular-Regulated 1/2 (ERK 1/2), p38, and c-jun N-Terminal MAPKs, and NF-κB Inhibitor IκB-α
Activation of MAPK and NF-κB pathways in samples was determined as the ratio between the levels of phosphorylated (phospho-) and total ERK1/2, p38, and JNK kinases and those of NF-κB inhibitor IκB-α, for the MAPK and NF-κB pathway, respectively. 
Molecule phospho- and total levels were analyzed by multiplex immunobead-based assay (Cell Signaling assay, Upstate Beadlyte, Dundee, UK) based on biological testing technology (Luminex xMAP; Luminex Corp., Austin, TX), according to the manufacturer's protocol. Briefly, tissues were homogenized in 300 μL lysis buffer B from the kit supplemented with protease inhibitors (Complete Mini Protease Inhibitor Cocktail, Roche Applied Science, Mannheim, Germany) and phosphatase inhibitor (1 mM sodium orthovanadate; Sigma-Aldrich Quimica SA Madrid, Spain). Homogenates were gently agitated for 15 minutes at room temperature and then centrifuged at 6000 rpm for 4 minutes Supernatants were collected and stored at −80°C until assayed. Total protein concentration in supernatants was determined by the BCA protein assay (Pierce, Rockford, IL), and 2 μg of total protein (25 μL) of each sample was pipetted into assay plate wells. To this was added a 25 μL mixture of beads coupled to phospho-ERK1/2–, phospho-p38–, phospho-JNK–, and phospho-IκB-α–specific capture antibodies (Upstate Beadlyte, Dundee, UK). A separate assay was run with a 25 μL mixture of beads coupled to total ERK1/2-, total p38-, total JNK-, and total IκB-α–specific capture antibodies (Upstate Beadlyte). After overnight incubation under agitation at 4°C in the dark, the beads were washed and mixed with biotinylated specific reporter antibodies for 1 hour at room temperature, followed by washing and subsequent incubation with streptavidin-phycoerythrin, for 30 minutes. The beads were finally washed and resuspended in 100 μL assay buffer, and the amounts of phospho- and total JNK, ERK1/2, and p38 kinases, and IκB-α (determined as the median fluorescence intensity in arbitrary units for each molecule in each sample) were measured (Luminex 100 IS 2.3 system; Luminex, Austin, TX). Samples were analyzed in duplicate. 
Statistical Analysis
Comparisons and Correlations.
Data were expressed as mean ± SEM. Statistical analyses were performed by a licensed statistician (I.F.) using the Statistical Package for the Social Sciences software (SPSS 15.0 for Windows; SPSS Inc., Chicago, IL) and R software. 28 Levels of phosphoproteins and total proteins, as well as phospho/total protein ratios for each were compared using the Kruskal-Wallis test with Dunn's multiple comparison posttest. Spearman Rho coefficient was used to determine the correlation among different protein amounts. Correlations between inflammatory molecule levels and clinical parameters were performed by the Mann-Whitney U test or Kruskal-Wallis ANOVA with Dunn's multiple comparison posttest. P ≤ 0.05 was considered the limit for statistical significance. 
Hierarchical Clustering.
The average-linkage agglomerative hierarchical cluster technique 29 was used to identify samples with similar protein activation levels. This method is based on the classification of m objects in m − 1 steps. In each consecutive step, the two most similar objects (or clusters) are merged. 28 The clusters to be joined are derived from the dissimilarity matrix, representing the dissimilarities between each pair of clusters. The two least dissimilar pairs of clusters are merged, and the dissimilarity matrix is recalculated for the new situation. 29 The result of such classification is visualized as a dendrogram. Dissimilarities between two clusters were defined as the average of all dissimilarities calculated between any object in both clusters. 30  
The weighted and unweighted average linkage, single and complete linkage, and centroid methods were evaluated in this study. In the centroid method, clusters are merged based on the minimal squared Euclidean distance between the centroids. In the dendrograms, the dissimilarity of the objects (system) was represented by the height at which the branches were connected. In the zone maps drawn, systems were ranked according to decreasing dissimilarities in the trees. Package stats in R software was used for the clustering analysis. 30  
k-Nearest Neighbor Classifier.
The k-nearest neighbor classifier was used to design a system able to accurately predict the origin of a potential new specimen based on its protein levels. It is a simple method that has a single parameter, the number of nearest neighbors, to be predefined given that the distance metric is Euclidean. An object is classified by a majority vote of its neighbors, with the object assigned to the class most common among its k-nearest neighbors. 31  
Phospho- and total protein values and phospho/total protein ratio for each type of sample were plotted, and prediction regions for new ocular specimens were statistically generated with the k-nearest neighbor classifier and represented. Package class in R software 31 was used to fit the classifier model. 
Results
Clinical Examination and Patient Characteristics
Specimens of resected whole primary pterygia were derived from 20 women and one man whose mean age was 45.4 years (SD, 14.9 years; range, 21–76 years; Table 1). All specimens used were nasal primary pterygia. Most of them (12 of 21) were grade 2, 5 of 21 were grade 3, and 4 of 21 were grade 1. Many of the patients (23.8%) lived >50% of their lives in tropical or equatorial zones. Most (66.7%) had worked mainly outdoors, and only 33.3% had worn sunglasses. Healthy conjunctival biopsy specimens were derived from four women and one man (mean age, 62.3 years; SD, 4.3 years; range, 58–70 years) who underwent cataract surgery. 
Table 1.
 
Baseline Characteristics of Patients
Table 1.
 
Baseline Characteristics of Patients
Patient Age (y) Sex Country of Origin Present City, Country of Residence Time Lived in Equatorial or Tropical Zones (%) Outdoor or Indoor Work Regular Use of Sunglasses Pterygium Grade*/Eye Ipsilateral Conjunctiva Sample Available
1 31 F Equator Madrid, Spain >50 Indoor No 2/R Yes
2 35 M Spain Galicia, Spain <50 Outdoor No 2/R Yes
3 67 F Spain Galicia, Spain <50 Outdoor No 2/L Yes
4 43 F Spain Galicia, Spain <50 Outdoor Yes 1/L Yes
5 59 F Spain Galicia, Spain <50 Outdoor No 2/L Yes
6 76 F Spain Galicia, Spain <50 Outdoor No 3/R Yes
7 41 F Spain Galicia, Spain <50 Outdoor Yes 1/L Yes
8 44 F Spain Galicia, Spain <50 Indoor Yes 1/L Yes
9 37 F Colombia Madrid, Spain >50 Indoor No 1/L No
10 36 F Spain Galicia, Spain <50 Outdoor Yes 2/L No
11 35 F Spain Galicia, Spain <50 Outdoor No 3/R No
12 53 F Spain Galicia, Spain <50 Outdoor No 2/R No
13 21 F Brazil Galicia, Spain >50 Outdoor Yes 2/R No
14 41 F Spain Galicia, Spain <50 Indoor Yes 2/R No
15 28 F Peru Madrid, Spain >50 Indoor No 3/R No
16 49 F Spain Galicia, Spain <50 Outdoor No 2/L No
17 59 F Portugal Coimbra, Portugal <50 Outdoor No 2/R No
18 46 F Portugal Coimbra, Portugal <50 Indoor Yes 3/R No
19 51 F Portugal Coimbra, Portugal <50 Outdoor No 2/R No
20 28 F Brazil Aveiro, Portugal >50 Indoor No 3/R No
21 73 F Portugal Porto, Portugal <50 Outdoor No 2/L No
MAPKs and IκB-α Analysis in Pterygium, Ipsilateral Pterygium-Free Specimens, and Healthy Conjunctiva
Total and phosphorylated forms of ERK1/2, p38, JNK MAPKs, and IκB-α were detected in all three types of samples. Both phospho- and total ERK1/2 were present in the highest levels, especially in ipsilateral pterygium-free conjunctiva and pterygium specimens (Figs. 1A, 1B). Although higher than the level in healthy conjunctiva, the differences were not significant. The ipsilateral pterygium-free conjunctiva had significantly higher levels of phospho-IκB-α than the pterygium (Fig. 1A). The pterygium and ipsilateral pterygium-free conjunctiva had significantly lower total JNK and total IκB-α levels than the healthy conjunctiva (Fig. 1B). 
Figure 1.
 
MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A) and total (t) (B) levels of ERK1/2, P38, and JNK MAPKs and of IκB-α were determined in pterygium, ipsilateral pterygium-free conjunctiva (ipsilateral conj), and healthy conjunctival (healthy conj) tissue lysates by a multiplex immunobead-based array. (C) Activation levels for each molecule were calculated as the ratio between corresponding phosphorylated (phospho) and total protein levels for each molecule. *P ≤ 0.05; statistically significant compared with healthy conjunctiva specimens. MFI, median fluorescence intensity.
Figure 1.
 
MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A) and total (t) (B) levels of ERK1/2, P38, and JNK MAPKs and of IκB-α were determined in pterygium, ipsilateral pterygium-free conjunctiva (ipsilateral conj), and healthy conjunctival (healthy conj) tissue lysates by a multiplex immunobead-based array. (C) Activation levels for each molecule were calculated as the ratio between corresponding phosphorylated (phospho) and total protein levels for each molecule. *P ≤ 0.05; statistically significant compared with healthy conjunctiva specimens. MFI, median fluorescence intensity.
The phospho/total protein ratios were calculated as indexes of activation for each molecule. The ipsilateral pterygium-free conjunctiva had significantly increased phospho/total protein ratios of both JNK and IκB-α compared with healthy conjunctiva (Fig. 1C). The activation ratio of JNK in pterygium specimens versus healthy conjunctivas was elevated but did not attain statistical significance (P =0.08). 
Correlation between Protein Levels
In healthy conjunctivas there was a significantly positive correlation between total ERK1/2 and total IκB-α (Fig. 2A). In ipsilateral pterygium-free conjunctivas, there were significant correlations between the levels of total ERK1/2 and total IκB-α, total p38 and total JNK, total p38 and total IκB-α, and phospho/total p38 and phospho/total ERK1/2 ratios (Fig. 2B). For pterygium specimens, statistically significant correlations existed for phospho-JNK and phospho-ERK1/2 and for phospho-JNK and phospho-p38 (Fig. 2C). Total p38 levels were significantly correlated with total ERK1/2, total JNK, and total IκB-α levels. Total JNK was significantly correlated with total ERK1/2, total IκB-α, and total p38 levels. The JNK phospho/total protein ratio was correlated with the phospho/total protein ratio of ERK1/2 and total p38. 
Figure 2.
 
Correlation studies among protein levels. Correlations among phosphorylated (p), total (t), and phosphorylated/total (p/t) protein ratio levels in (A) healthy conjunctiva, (B) pterygium, and (C) ipsilateral pterygium-free conjunctiva specimens were calculated. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. MFI, median fluorescence intensity.
Figure 2.
 
Correlation studies among protein levels. Correlations among phosphorylated (p), total (t), and phosphorylated/total (p/t) protein ratio levels in (A) healthy conjunctiva, (B) pterygium, and (C) ipsilateral pterygium-free conjunctiva specimens were calculated. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. MFI, median fluorescence intensity.
Association of MAPKs and IκB-α Phospho- and Total Protein Levels and Phospho/Total Ratio with Clinical Parameters
We analyzed the correlation between phospho- and total protein levels and protein activation values and patient clinical parameters such as age, sex, birthplace, current city of residence, and whether they had worked mainly outdoors and had worn sunglasses. We also looked for correlations between protein levels and pterygium parameters such as pterygium grade and pterygium head and body size (horizontal and vertical). Several statistically significant correlations were present. 
Sun Exposure.
Patients who were born or lived in the tropics or near an equatorial zone for >50% of their lives showed significantly decreased phospho-ERK1/2, phospho-JNK, total p38, and total JNK levels (Figs. 3A, 3B). They also had significantly increased phospho/total p38 ratio (Fig. 3C). 
Figure 3.
 
Effect of sun exposure and pterygium grade on MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A), total (t) (B), and phospho/total (p/t) ratio (C) levels of MAPKs and IκBα in pterygium samples from patients who have lived >50% of their lives in equatorial zones were compared with those from patients who had lived <50% of their lives in equatorial zones. Differences in levels of phosphorylated (p) (D), total (t) (E), and phosphorylated/total ratio (p/t) (F) levels of molecules in dependence of pterygium grade were also statistically compared. *P ≤ 0.05.
Figure 3.
 
Effect of sun exposure and pterygium grade on MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A), total (t) (B), and phospho/total (p/t) ratio (C) levels of MAPKs and IκBα in pterygium samples from patients who have lived >50% of their lives in equatorial zones were compared with those from patients who had lived <50% of their lives in equatorial zones. Differences in levels of phosphorylated (p) (D), total (t) (E), and phosphorylated/total ratio (p/t) (F) levels of molecules in dependence of pterygium grade were also statistically compared. *P ≤ 0.05.
Pterygium Grade.
Total ERK1/2 levels were significantly lower in grade 3 pterygia compared with grades 1 and 2 (Fig. 3E). Phospho/total protein ERK1/2 ratio was significantly increased in grade 3 pterygia compared with grade 1 pterygia (Fig. 3F). 
Pterygium Measurements.
The horizontal measurements of pterygia heads were significantly correlated with phospho-ERK1/2 and phospho-JNK levels and phospho/total protein ERK1/2 ratios (Fig. 4A). Vertical measurements of the head were significantly correlated with phospho-ERK1/2, phospho-JNK, total IκB-α, total p38, and total JNK (Fig. 4B). The horizontal measurements of pterygium bodies were significantly correlated with phospho/total ERK1/2 ratio (Fig. 4C). 
Figure 4.
 
Correlation studies between MAPKs and IκB-α levels and pterygium measurements. ERK1/2, p38, JNK MAPKs, and IκB-α levels (expressed as median fluorescence intensity in arbitrary units) in pterygium specimens were correlated with pterygium (A) head horizontal, (B) head vertical, and (C) body horizontal measurements. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. p, phosphorylated; t, total; p/t, phosphorylated/total ratio.
Figure 4.
 
Correlation studies between MAPKs and IκB-α levels and pterygium measurements. ERK1/2, p38, JNK MAPKs, and IκB-α levels (expressed as median fluorescence intensity in arbitrary units) in pterygium specimens were correlated with pterygium (A) head horizontal, (B) head vertical, and (C) body horizontal measurements. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. p, phosphorylated; t, total; p/t, phosphorylated/total ratio.
Hierarchical Clustering
To determine whether similarities existed between patients and the ratios of activation of the four proteins, we performed a cluster analysis of activation ratio values in pterygium samples (n = 21). There were two groups of patients who displayed similar activation value patterns. One group consisted of four patients (patients 3, 6, 9, and 17) and was characterized by a high JNK activation value. The other group was composed of the remaining 17 patients (Fig. 5A). 
Figure 5.
 
Hierarchical clustering analysis. To identify patients with similar protein activation values, hierarchical clustering analysis was performed. (A) For activation (phosphorylated/total protein ratio) values in pterygium samples (n = 21), patients 3, 6, 9, and 17 were clustered in group 1 (G1) with higher activation of JNK. All other patient samples fell into group 2 (G2). (B) In ipsilateral pterygium-free conjunctiva samples (n = 8), patients 3 and 5 clustered into group 1 (G1) with elevated ERK1/2 activation. Patients 1, 4, and 7 clustered into group 2 (G2) with elevated JNK activation. Patients 2, 6, and 8 clustered into group 3 (G3) with low JNK and p38 ratios. (C) Cluster analysis was performed on the ratio between activation values in the pterygium sample and the ipsilateral pterygium-free conjunctiva sample in the eight patients for whom both samples were available. Based on the pterygium activation/ipsilateral pterygium-free activation ratios, group 1 (G1) consisted of six patients (1–5 and 7). Patient 6 (P6) had a high JNK ratio with a moderately high ERK1/2 ratio. Patient 8 (P8) had a very high ERK1/2 ratio along with moderate p38 and JNK ratios.
Figure 5.
 
Hierarchical clustering analysis. To identify patients with similar protein activation values, hierarchical clustering analysis was performed. (A) For activation (phosphorylated/total protein ratio) values in pterygium samples (n = 21), patients 3, 6, 9, and 17 were clustered in group 1 (G1) with higher activation of JNK. All other patient samples fell into group 2 (G2). (B) In ipsilateral pterygium-free conjunctiva samples (n = 8), patients 3 and 5 clustered into group 1 (G1) with elevated ERK1/2 activation. Patients 1, 4, and 7 clustered into group 2 (G2) with elevated JNK activation. Patients 2, 6, and 8 clustered into group 3 (G3) with low JNK and p38 ratios. (C) Cluster analysis was performed on the ratio between activation values in the pterygium sample and the ipsilateral pterygium-free conjunctiva sample in the eight patients for whom both samples were available. Based on the pterygium activation/ipsilateral pterygium-free activation ratios, group 1 (G1) consisted of six patients (1–5 and 7). Patient 6 (P6) had a high JNK ratio with a moderately high ERK1/2 ratio. Patient 8 (P8) had a very high ERK1/2 ratio along with moderate p38 and JNK ratios.
Cluster analysis of activation values in the ipsilateral pterygium-free conjunctiva specimens (n = 8) revealed three groups (Fig. 5B). Group 1 contained two patients (patients 3 and 5) characterized by a high ERK1/2 activation value, group 2 was composed of three patients (patients 1, 4, and 7) characterized by a high JNK activation value, and group 3 was composed of three patients (patients 2, 6, and 8) characterized by a low JNK and p38 ratios. 
A third cluster analysis was performed for ratio values between activation value in pterygium and ipsilateral pterygium-free samples in the eight patients for whom both samples were available (Fig. 5C). This clustering revealed only one group composed of six patients (patients 1–5 and 7). Two patients (patients 6 and 8) did not group with any others. Patient 6 was characterized by a high JNK ratio and a moderately high ERK1/2 ratio, and patient 8 was characterized by a very high ERK1/2 ratio and moderate p38 and JNK ratios. 
Predictive Model by the k-Nearest Neighbor Classifier
The k-nearest neighbor classifier algorithm was designed to predict the putative origin (i.e., pterygium, ipsilateral pterygium-free conjunctiva, or healthy conjunctiva) of a new sample based on the phospho- and total protein values and the phospho/total ratios of the four molecules studied. The accuracy of the plotted prediction region was 70.6% for phospho-molecule values, 76.5% for total molecule values, and 70.6% for molecule activation values (Fig. 6). 
Figure 6.
 
k-Nearest neighbor classifier predictive model for sample type. Diagrams of sample type predictive regions calculated by the k-nearest neighbor classifier algorithm were plotted (▲, pterygium; ■, ipsilateral pterygium-free conjunctiva; ●, healthy conjunctiva) for (A) phosphorylated (p), (B) total (t) protein, and (C) phosphorylated/total (p/t) protein ratio values. Prediction regions for potential specimens were statistically generated and represented by color code (white, healthy conjunctiva; gray, ipsilateral pterygium-free conjunctiva; dark gray, pterygium).
Figure 6.
 
k-Nearest neighbor classifier predictive model for sample type. Diagrams of sample type predictive regions calculated by the k-nearest neighbor classifier algorithm were plotted (▲, pterygium; ■, ipsilateral pterygium-free conjunctiva; ●, healthy conjunctiva) for (A) phosphorylated (p), (B) total (t) protein, and (C) phosphorylated/total (p/t) protein ratio values. Prediction regions for potential specimens were statistically generated and represented by color code (white, healthy conjunctiva; gray, ipsilateral pterygium-free conjunctiva; dark gray, pterygium).
Discussion
In this article we provide evidence of the MAPK signaling pathway and NF-κB activation in primary pterygium specimens. For the first time to our knowledge, we also identify changes in these pathways in ipsilateral pterygium-free conjunctiva showing no signs of pterygium growth. Previous studies have shown some evidence of MAPK pathway activation in pterygium, 9,22,32 but much less attention has been paid to ipsilateral pterygium-free conjunctiva. Ipsilateral conjunctiva is routinely used in pterygium surgery, and it has been assumed to be composed of normal tissue. Our study shows that this is might not be the case because it also shows an inflammatory component. In addition, in this study we provide correlation studies between pterygium-related parameters and clinical variables and a predictive model of tissue sample origin based on its protein levels. 
We found high expression levels of phospho- and total ERK1/2 in ipsilateral pterygium-free conjunctiva and in the pterygium itself. These high levels are in agreement with previous studies. 22,32 Elevated levels of ERK1/2, p38, and JNK kinases can be induced by physiological doses of UV-B radiation in ex vivo pterygium as well as in cultured pterygium epithelial cells. 22 In our study, though not quite reaching statistical significance, both ipsilateral pterygium-free conjunctiva and pterygium had higher levels of phospho-ERK1/2 and total ERK1/2 than normal conjunctiva. MAPK signaling pathways play an important role in inflammatory responses. They coordinate and regulate diverse cellular activities, including gene expression, mitosis, metabolism, mobility, survival, apoptosis, and differentiation. 33,34 MAPKs can be activated by wide variety of different stimuli, but, in general, ERK1/2 is preferentially activated in response to growth factors and phorbol esters, whereas the JNK and p38 kinases are more responsive to stress stimuli such as osmotic shock, ionizing radiation, and cytokine stimulation. 33  
ERK1/2 is a molecule that acts in a signaling cascade that regulates various cellular processes such as proliferation, differentiation, and cell cycle progression in response to a variety of extracellular signals. It seems to specifically mediate the induction of MMP-1 by UV-B in pterygium biopsy specimens and culture pterygium epithelial cells. 22,32 All ERK, JNK, and p38 MAPK-specific inhibitors blocked UV-B–induced secretion of IL-6 and IL-8 and vascular endothelial growth factor by cultured epithelial pterygial cells. 9 Of these, only blocking the ERK pathway specifically decreased MMP-1 production. 22  
JNK activation was significantly greater in ipsilateral pterygium-free conjunctiva than in control healthy ones. Although not statistically significant, activated JNK was also greater in the pterygia samples than the healthy control ones. With greater sample size, it is possible that the difference could be significant. JNK is a stress-activated protein kinase that increases in response to cytokines, UV irradiation, growth factor deprivation, and agents that interfere with DNA and protein synthesis. These types of kinases are involved in cytokine production and other aspects of inflammation and apoptosis. 35  
We have also included in our study the NF-κB pathway because it mediates some cellular responses to UV damage. 3,23,24 Our analysis demonstrated that ipsilateral pterygium-free conjunctivas had significantly increased levels of phospho-IκB-α compared with pterygium and increased phospho/total protein ratios of IκB-α compared with healthy conjunctivas. NF-κB is an essential regulator for the expression of numerous genes involved in the function and development of the immune system and in inflammatory and acute responses. Moreover, NF-κB is a key mediator of IL-1β and p53 expression 36 that potentially plays important roles in pterygium development. 37,38 NF-κB is also suppressed by retinoic acid, which also inhibits IL-6 and IL-8 in UV-irradiated pterygium epithelial cells. 
Another important point is that the ipsilateral pterygium-free conjunctiva showed molecular signs of inflammation. This conjunctiva is universally used as a graft to cover the bare sclera after pterygium removal and has always been considered healthy. Our findings suggest that this might not be a good choice. It would be interesting to find out the origin of this inflammatory component. UV-related damage can be ruled out because this area is not directly exposed to the sun. One explanation could be that this tissue receives stimulatory signals from the adjacent pterygium or from the tears. In fact Di Girolamo et al. 10 have shown that two inflammatory cytokines, IL-6 and IL-8, were abundantly expressed by the pterygium epithelium. Further, the expression of both IL-6 and IL-8 mRNA and proteins was induced in UVB-irradiated pterygium epithelial cells in a time- and dose-dependent manner. They also found that IL-6 and IL-8 were significantly elevated in UVB-treated pterygia compared with nonirradiated pterygia. 9,10 Zho et al. 39 have found that human α-defensins and S100 calcium-binding proteins A8 and A9 are elevated in the tear fluid of patients with pterygium. Our finding could reflect a genetic predisposition that is generally expressed in all locations of the conjunctiva. Such persons receiving UV exposure could be more likely to develop the disease in the sun-exposed areas. This idea is supported by our findings in normal conjunctiva from the same area, which is clearly different from the pterygium conjunctiva. Further studies are warranted to elucidate these data. 
For the correlation studies, we used not only the molecule activation values (phospho/total ratio) but also the two values needed to calculate them (phospho and total). Although the activation ratio values are more often used in correlation studies, it is interesting to determine whether there is upregulation or downregulation of the total or the phospho proteins. For example, the activation ratio can be the same, but total and activated levels could be different. 
The correlations in the pterygium sample, the pterygium-free conjunctiva sample, and the healthy conjunctivas were not the same. This is relevant because it clearly proves that each tissue behaves differently, indicating that the cells responded differently in the three different tissues. MAPK expression and activation are related to cell survival or cell death in response to stress. These associations increase our understanding of pathway activations that may be positively or negatively correlated in pterygium disease. In turn, the correlations add to our knowledge regarding the cell mechanisms involved in pathogenesis, and they further support a role for inflammation in pterygium. 
An increased incidence of pterygium occurs in the tropics and in an equatorial zone between 30° north and 30° south latitudes, 19 where the rays of the sun are perpendicular to the surface of the earth. We found that patients who were born or lived in the tropics or the equatorial zone >50% of their lives had significantly decreased phospho-ERK1/2, phospho-JNK, total p38, and total JNK levels. This same group also had significantly increased activation of p38. These findings add further support to the implication of UV light in the inflammation component of pterygium. p38, which is involved in the immune responses to pathogens in various cells, can be activated by various physical and chemical stresses, such as oxidative stress, hypo-osmolarity, and UV irradiation. 33,34  
We also found that the larger the pterygium head and the more advanced the pterygium grade were, the more activated was the level of ERK1/2. Ours results add support to an important role of this molecule in pterygium disease, described in previous studies by DiGirolamo et al. 9,22,32  
We used hierarchical cluster analysis to determine whether similarities exist between patients with respect to the activation levels of the four molecules analyzed. Particularly interesting was the cluster analysis for the ratio of pterygium activation to ipsilateral pterygium-free conjunctiva activation in the eight patients in whom both samples were analyzed. We did this to determine whether all patients maintained a similar ratio between activation of molecules in pterygium and ipsilateral pterygium-free conjunctivas. Six of the patients formed a homogeneous group, whereas two patients had dissimilar ratios compared with them. There was nothing in the clinical history of these two patients that explained why they were different. Thus, other factors that were not recorded by us may affect the activation ratios of the pterygium and ipsilateral pterygium-free conjunctivas. 
Based on the phosphorylated, total, and activated levels of ERK1/2, p38, and JNK kinases and of IκB-α, the k-nearest neighbor classifier algorithm was used to predict the origin—i.e., pterygium, ipsilateral pterygium-free conjunctiva, or healthy conjunctiva—of the samples. The accuracy of these predictions was very high (70.6%–76.5%), suggesting that these four molecules could be a potential valid marker of inflammation in pterygium samples. 
In summary, we have shown four points. First, there were significant differences in the phosphorylated and total levels of some of the molecules in the MAPK and NF-κB pathways. These differences were clearly associated with the tissue origin, e.g., normal conjunctiva, ipsilateral conjunctiva, or pterygium. Second, the phosphorylated, total, and ratio levels of ERK1/2, p38, JNK, and IκBα molecules were correlated among themselves. This correlation was different in each type of tissue, again indicating the different behavior of each of the tissues. They were also correlated with clinical parameters. Third, cluster analysis of activation values and ratios in pterygium and ipsilateral-conjunctiva showed that different groups of patients with similar values were found. Fourth, a prediction model with 70.6% to 76.5% accuracy was designed based not only on the levels of the activation value (phosphorylated/total ratio) for the four molecules studied but also on the actual levels of both the phosphorylated and the total forms. 
Although we have found several significant alterations in MAPK and NF-κB pathways in pterygia and have developed a very accurate type of tissue prediction model, we are also aware that our study has some limitations. For instance, the patient population was relatively small. In addition, by chance, the sex distribution of our study population and our control population was unbalanced. We do not think this affected our results because there are no reports of a sex effect in pterygium after treatment. Nevertheless, our results obviously refer to the population of this study rather than to the general population. Another limitation is that the healthy control conjunctiva used in this study was nonautologous. Further studies with more specimens and a more balanced sex distribution are needed to confirm our results. Although it would have been be interesting to determine which cells produced each of these proteins, the small sample size restricted the possibility of performing concomitant immunohistochemical studies. The localization of ERK1/2 in pterygium tissue has been described 32 ; therefore, we elected to study our samples by multiplex analysis based on biological testing technology (Luminex xMAP). This is a well-established technique that has not, until now, been applied to the analysis of pterygia. It is reliable, highly sensitive, less time consuming, and it requires only a small amount of sample. 
In conclusion, we analyzed changes in MAPK and NF-κB pathways in pterygium and, for the first time, in the ipsilateral pterygium-free conjunctiva. The changes we documented are consistent with an important role for an inflammatory component in pterygium pathogenesis. This role is further supported by the high prediction values obtained with the k-nearest neighbor classifier algorithm that analyzed phospho-, total, and activation values of ERK1/2, p38, and JNK MAPKs and IκB-α levels. Our data also suggest that inhibitors to MAPKs or NF-κB might provide useful therapeutic intervention in pterygium, but further studies are needed to prove such a hypothesis. 
Footnotes
 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, April 2008 and May 2010.
Footnotes
 Supported by Red Temática de Investigación Cooperativa CO3/13, Instituto de Salud Carlos III, Ministerio de Sanidad, Spain, CIBER-BBN, Valladolid, Spain, and Centro en Red. de Medicina Regenerativa y Terapia Celular, Spain.
Footnotes
 Disclosure: J. Torres, None; A. Enríquez-de-Salamanca, None; I. Fernández, None; M.T. Rodríguez-Ares, None; M.J. Quadrado, None; J. Murta, None; J.M. Benítez-del-Castillo, None; M.E. Stern, Allergan, Inc. (E); M. Calonge, Allergan, Inc. (C)
The authors thank Carmen García-Vázquez for excellence technical assistance. 
References
Coroneo MT . Pterygium as an early indicator of ultraviolet insolation: a hypothesis. Br J Ophthalmol. 1993;77:734–739. [CrossRef] [PubMed]
Threlfall TJ English DR . Sun exposure and pterygium of the eye: a dose-response curve. Am J Ophthalmol. 1999;128:280–287. [CrossRef] [PubMed]
Chui J Di GN Wakefield D Coroneo MT . The pathogenesis of pterygium: current concepts and their therapeutic implications. Ocul Surf. 2008;6:24–43. [CrossRef] [PubMed]
Di Girolamo N McCluskey P Lloyd A Coroneo MT Wakefield D . Expression of MMPs and TIMPs in human pterygia and cultured pterygium epithelial cells. Invest Ophthalmol Vis Sci. 2000;41:671–679. [PubMed]
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. [CrossRef] [PubMed]
Butrus SI Ashraf MF Laby DM Rabinowitz AI Tabbara SO Hidayat AA . Increased numbers of mast cells in pterygia. Am J Ophthalmol. 1995;119:236–237. [CrossRef] [PubMed]
Pinkerton OD Hokama Y Shigemura LA . Immunologic basis for the pathogenesis of pterygium. Am J Ophthalmol. 1984;98:225–228. [CrossRef] [PubMed]
Seifert P Sekundo W . Capillaries in the epithelium of pterygium. Br J Ophthalmol. 1998;82:77–81. [CrossRef] [PubMed]
Di Girolamo N Wakefield D Coroneo MT . UVB-mediated induction of cytokines and growth factors in pterygium epithelial cells involves cell surface receptors and intracellular signaling. Invest Ophthalmol Vis Sci. 2006;47:2430–2437. [CrossRef] [PubMed]
Di Girolamo N Kumar RK Coroneo MT Wakefield D . UVB-mediated induction of interleukin-6 and −8 in pterygia and cultured human pterygium epithelial cells. Invest Ophthalmol Vis Sci. 2002;43:3430–3437. [PubMed]
Kria L Ohira A Amemiya T . Immunohistochemical localization of basic fibroblast growth factor, platelet derived growth factor, transforming growth factor-beta and tumor necrosis factor-alpha in the pterygium. Acta Histochem. 1996;98:195–201. [CrossRef] [PubMed]
Lee DH Cho HJ Kim JT Choi JS Joo CK . Expression of vascular endothelial growth factor and inducible nitric oxide synthase in pterygia. Cornea. 2001;20:738–742. [CrossRef] [PubMed]
Nakagami T Watanabe I Murakami A Okisaka S Ebihara N . Expression of stem cell factor in pterygium. Jpn J Ophthalmol. 2000;44:193–197. [CrossRef] [PubMed]
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;44:573–580. [CrossRef] [PubMed]
van Setten G Aspiotis M Blalock TD Grotendorst G Schultz G . Connective tissue growth factor in pterygium: simultaneous presence with vascular endothelial growth factor—possible contributing factor to conjunctival scarring. Graefes Arch Clin Exp Ophthalmol. 2003;241:135–139. [CrossRef] [PubMed]
Jin J Guan M Sima J . Decreased pigment epithelium-derived factor and increased vascular endothelial growth factor levels in pterygia. Cornea. 2003;22:473–477. [CrossRef] [PubMed]
Tong L Chew J Yang H Ang LP Tan DT Beuerman RW . Distinct gene subsets in pterygia formation and recurrence: dissecting complex biological phenomenon using genome wide expression data. BMC Med Genomics. 2009;2:14. [CrossRef] [PubMed]
John-Aryankalayil M Dushku N Jaworski CJ . Microarray and protein analysis of human pterygium. Mol Vis. 2006;12:55–64. [PubMed]
McCarty CA Fu CL Taylor HR . Epidemiology of pterygium in Victoria, Australia. Br J Ophthalmol. 2000;84:289–292. [CrossRef] [PubMed]
Moran DJ Hollows FC . Pterygium and ultraviolet radiation: a positive correlation. Br J Ophthalmol. 1984;68:343–346. [CrossRef] [PubMed]
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;131:176–183. [CrossRef] [PubMed]
Di Girolamo N Coroneo MT Wakefield D . UVB-elicited induction of MMP-1 expression in human ocular surface epithelial cells is mediated through the ERK1/2 MAPK-dependent pathway. Invest Ophthalmol Vis Sci. 2003;44:4705–4714. [CrossRef] [PubMed]
Abeyama K Eng W Jester JV . A role for NF-κB-dependent gene transactivation in sunburn. J Clin Invest. 2000;105:1751–1759. [CrossRef] [PubMed]
Liu Z Xie Y Zhang M . Overexpression of type I growth factor receptors in pterygium. Chin Med J (Engl). 2002;115:418–421. [PubMed]
Nemeth ZH Deitch EA Szabo C Hasko G . Hyperosmotic stress induces nuclear factor-κB activation and interleukin-8 production in human intestinal epithelial cells. Am J Pathol. 2002;161:987–996. [CrossRef] [PubMed]
Matsuo H Tamura M Kabashima N . Prednisolone inhibits hyperosmolarity-induced expression of MCP-1 via NF-κB in peritoneal mesothelial cells. Kidney Int. 2006;69:736–746. [CrossRef] [PubMed]
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;115:1235–1240. [CrossRef] [PubMed]
R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2008.
Kaufman L Rousseeuw P . Finding Groups in Data: An Introduction to Cluster Analysis. New York: Wiley; 1990.
Massart D Kaufman L . The Interpretation of Analytical Chemical Data by the Use of Cluster Analysis. New York: Interscience; 1983.
Cover T Hart P . Nearest neigbour pattern classification. IEEE Trans Inf Theory IT. 1967;13:21–27. [CrossRef]
Di Girolamo N Coroneo M Wakefield D . Epidermal growth factor receptor signaling is partially responsible for the increased matrix metalloproteinase-1 expression in ocular epithelial cells after UVB radiation. Am J Pathol. 2005;167:489–503. [CrossRef] [PubMed]
Pearson G Robinson F Beers GT . Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocr Rev. 2001;22:153–183. [PubMed]
Roux PP Blenis J . ERK and p38 MAPK-activated protein kinases: a family of protein kinases with diverse biological functions. Microbiol Mol Biol Rev. 2004;68:320–344. [CrossRef] [PubMed]
Xia Z Dickens M Raingeaud J Davis RJ Greenberg ME . Opposing effects of ERK and JNK-p38 MAP kinases on apoptosis. Science. 1995;270:1326–1331. [CrossRef] [PubMed]
Ghosh S Hayden MS . New regulators of NF-κB in inflammation. Nat Rev Immunol. 2008;8:837–848. [CrossRef] [PubMed]
Wen Z Liu Z . [The abnormal expression of interleukin-1 family in pterygium]. Yan Ke Xue Bao. 2003;19:133–136. [PubMed]
Tsai YY Chang KC Lin CL . p53 Expression in pterygium by immunohistochemical analysis: a series report of 127 cases and review of the literature. Cornea. 2005;24:583–586. [CrossRef] [PubMed]
Zhou L Beuerman RW Ang LP . Elevation of human alpha-defensins and S100 calcium-binding proteins A8 and A9 in tear fluid of patients with pterygium. Invest Ophthalmol Vis Sci. 2009;50:2077–2086. [CrossRef] [PubMed]
Figure 1.
 
MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A) and total (t) (B) levels of ERK1/2, P38, and JNK MAPKs and of IκB-α were determined in pterygium, ipsilateral pterygium-free conjunctiva (ipsilateral conj), and healthy conjunctival (healthy conj) tissue lysates by a multiplex immunobead-based array. (C) Activation levels for each molecule were calculated as the ratio between corresponding phosphorylated (phospho) and total protein levels for each molecule. *P ≤ 0.05; statistically significant compared with healthy conjunctiva specimens. MFI, median fluorescence intensity.
Figure 1.
 
MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A) and total (t) (B) levels of ERK1/2, P38, and JNK MAPKs and of IκB-α were determined in pterygium, ipsilateral pterygium-free conjunctiva (ipsilateral conj), and healthy conjunctival (healthy conj) tissue lysates by a multiplex immunobead-based array. (C) Activation levels for each molecule were calculated as the ratio between corresponding phosphorylated (phospho) and total protein levels for each molecule. *P ≤ 0.05; statistically significant compared with healthy conjunctiva specimens. MFI, median fluorescence intensity.
Figure 2.
 
Correlation studies among protein levels. Correlations among phosphorylated (p), total (t), and phosphorylated/total (p/t) protein ratio levels in (A) healthy conjunctiva, (B) pterygium, and (C) ipsilateral pterygium-free conjunctiva specimens were calculated. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. MFI, median fluorescence intensity.
Figure 2.
 
Correlation studies among protein levels. Correlations among phosphorylated (p), total (t), and phosphorylated/total (p/t) protein ratio levels in (A) healthy conjunctiva, (B) pterygium, and (C) ipsilateral pterygium-free conjunctiva specimens were calculated. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. MFI, median fluorescence intensity.
Figure 3.
 
Effect of sun exposure and pterygium grade on MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A), total (t) (B), and phospho/total (p/t) ratio (C) levels of MAPKs and IκBα in pterygium samples from patients who have lived >50% of their lives in equatorial zones were compared with those from patients who had lived <50% of their lives in equatorial zones. Differences in levels of phosphorylated (p) (D), total (t) (E), and phosphorylated/total ratio (p/t) (F) levels of molecules in dependence of pterygium grade were also statistically compared. *P ≤ 0.05.
Figure 3.
 
Effect of sun exposure and pterygium grade on MAPKs and IκB-α levels in pterygium, ipsilateral pterygium-free conjunctiva, and healthy conjunctiva. Phosphorylated (p) (A), total (t) (B), and phospho/total (p/t) ratio (C) levels of MAPKs and IκBα in pterygium samples from patients who have lived >50% of their lives in equatorial zones were compared with those from patients who had lived <50% of their lives in equatorial zones. Differences in levels of phosphorylated (p) (D), total (t) (E), and phosphorylated/total ratio (p/t) (F) levels of molecules in dependence of pterygium grade were also statistically compared. *P ≤ 0.05.
Figure 4.
 
Correlation studies between MAPKs and IκB-α levels and pterygium measurements. ERK1/2, p38, JNK MAPKs, and IκB-α levels (expressed as median fluorescence intensity in arbitrary units) in pterygium specimens were correlated with pterygium (A) head horizontal, (B) head vertical, and (C) body horizontal measurements. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. p, phosphorylated; t, total; p/t, phosphorylated/total ratio.
Figure 4.
 
Correlation studies between MAPKs and IκB-α levels and pterygium measurements. ERK1/2, p38, JNK MAPKs, and IκB-α levels (expressed as median fluorescence intensity in arbitrary units) in pterygium specimens were correlated with pterygium (A) head horizontal, (B) head vertical, and (C) body horizontal measurements. Spearman rank correlation test was used, and the strength of correlation between variables was determined by the Spearman rho correlation coefficient. Only significant correlations are shown. p, phosphorylated; t, total; p/t, phosphorylated/total ratio.
Figure 5.
 
Hierarchical clustering analysis. To identify patients with similar protein activation values, hierarchical clustering analysis was performed. (A) For activation (phosphorylated/total protein ratio) values in pterygium samples (n = 21), patients 3, 6, 9, and 17 were clustered in group 1 (G1) with higher activation of JNK. All other patient samples fell into group 2 (G2). (B) In ipsilateral pterygium-free conjunctiva samples (n = 8), patients 3 and 5 clustered into group 1 (G1) with elevated ERK1/2 activation. Patients 1, 4, and 7 clustered into group 2 (G2) with elevated JNK activation. Patients 2, 6, and 8 clustered into group 3 (G3) with low JNK and p38 ratios. (C) Cluster analysis was performed on the ratio between activation values in the pterygium sample and the ipsilateral pterygium-free conjunctiva sample in the eight patients for whom both samples were available. Based on the pterygium activation/ipsilateral pterygium-free activation ratios, group 1 (G1) consisted of six patients (1–5 and 7). Patient 6 (P6) had a high JNK ratio with a moderately high ERK1/2 ratio. Patient 8 (P8) had a very high ERK1/2 ratio along with moderate p38 and JNK ratios.
Figure 5.
 
Hierarchical clustering analysis. To identify patients with similar protein activation values, hierarchical clustering analysis was performed. (A) For activation (phosphorylated/total protein ratio) values in pterygium samples (n = 21), patients 3, 6, 9, and 17 were clustered in group 1 (G1) with higher activation of JNK. All other patient samples fell into group 2 (G2). (B) In ipsilateral pterygium-free conjunctiva samples (n = 8), patients 3 and 5 clustered into group 1 (G1) with elevated ERK1/2 activation. Patients 1, 4, and 7 clustered into group 2 (G2) with elevated JNK activation. Patients 2, 6, and 8 clustered into group 3 (G3) with low JNK and p38 ratios. (C) Cluster analysis was performed on the ratio between activation values in the pterygium sample and the ipsilateral pterygium-free conjunctiva sample in the eight patients for whom both samples were available. Based on the pterygium activation/ipsilateral pterygium-free activation ratios, group 1 (G1) consisted of six patients (1–5 and 7). Patient 6 (P6) had a high JNK ratio with a moderately high ERK1/2 ratio. Patient 8 (P8) had a very high ERK1/2 ratio along with moderate p38 and JNK ratios.
Figure 6.
 
k-Nearest neighbor classifier predictive model for sample type. Diagrams of sample type predictive regions calculated by the k-nearest neighbor classifier algorithm were plotted (▲, pterygium; ■, ipsilateral pterygium-free conjunctiva; ●, healthy conjunctiva) for (A) phosphorylated (p), (B) total (t) protein, and (C) phosphorylated/total (p/t) protein ratio values. Prediction regions for potential specimens were statistically generated and represented by color code (white, healthy conjunctiva; gray, ipsilateral pterygium-free conjunctiva; dark gray, pterygium).
Figure 6.
 
k-Nearest neighbor classifier predictive model for sample type. Diagrams of sample type predictive regions calculated by the k-nearest neighbor classifier algorithm were plotted (▲, pterygium; ■, ipsilateral pterygium-free conjunctiva; ●, healthy conjunctiva) for (A) phosphorylated (p), (B) total (t) protein, and (C) phosphorylated/total (p/t) protein ratio values. Prediction regions for potential specimens were statistically generated and represented by color code (white, healthy conjunctiva; gray, ipsilateral pterygium-free conjunctiva; dark gray, pterygium).
Table 1.
 
Baseline Characteristics of Patients
Table 1.
 
Baseline Characteristics of Patients
Patient Age (y) Sex Country of Origin Present City, Country of Residence Time Lived in Equatorial or Tropical Zones (%) Outdoor or Indoor Work Regular Use of Sunglasses Pterygium Grade*/Eye Ipsilateral Conjunctiva Sample Available
1 31 F Equator Madrid, Spain >50 Indoor No 2/R Yes
2 35 M Spain Galicia, Spain <50 Outdoor No 2/R Yes
3 67 F Spain Galicia, Spain <50 Outdoor No 2/L Yes
4 43 F Spain Galicia, Spain <50 Outdoor Yes 1/L Yes
5 59 F Spain Galicia, Spain <50 Outdoor No 2/L Yes
6 76 F Spain Galicia, Spain <50 Outdoor No 3/R Yes
7 41 F Spain Galicia, Spain <50 Outdoor Yes 1/L Yes
8 44 F Spain Galicia, Spain <50 Indoor Yes 1/L Yes
9 37 F Colombia Madrid, Spain >50 Indoor No 1/L No
10 36 F Spain Galicia, Spain <50 Outdoor Yes 2/L No
11 35 F Spain Galicia, Spain <50 Outdoor No 3/R No
12 53 F Spain Galicia, Spain <50 Outdoor No 2/R No
13 21 F Brazil Galicia, Spain >50 Outdoor Yes 2/R No
14 41 F Spain Galicia, Spain <50 Indoor Yes 2/R No
15 28 F Peru Madrid, Spain >50 Indoor No 3/R No
16 49 F Spain Galicia, Spain <50 Outdoor No 2/L No
17 59 F Portugal Coimbra, Portugal <50 Outdoor No 2/R No
18 46 F Portugal Coimbra, Portugal <50 Indoor Yes 3/R No
19 51 F Portugal Coimbra, Portugal <50 Outdoor No 2/R No
20 28 F Brazil Aveiro, Portugal >50 Indoor No 3/R No
21 73 F Portugal Porto, Portugal <50 Outdoor No 2/L No
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×