Molecular Vision 2007; 13:1181-1197 <http://www.molvis.org/molvis/v13/a129/> Received 15 January 2007 | Accepted 11 July 2007 | Published 17 July 2007

Download Reprint

Oligonucleotide microarray analysis of human lens epithelial cells: TGFβ regulated gene expression

L.J. Dawes,¹ R.M Elliott,² J.R. Reddan,³ Y.M.Wormstone,² I.M. Wormstone¹
 
 

¹School of Biological Sciences, University of East Anglia, Norwich, UK; ²Institute of Food Research, Norwich Research Park, Norwich, UK; ³Oakland University, Rochester, MI

Correspondence to: Dr I.M. Wormstone, School of Biological Sciences, University of East Anglia, Norwich, NR4 7TJ, UK; Phone: 44-1603-591319; FAX: 44-1603-592250; email: i.m.wormstone@uea.ac.uk

Abstract

Purpose: Transforming growth factor beta (TGFβ), a pro-fibrotic cytokine has been proposed a causative factor in the progression of lens pathologies including posterior capsule opacification (PCO), a condition that occurs after cataract surgery. This study employs oligonucleotide microarrays to provide a global profile of gene expression in FHL 124 cells, to identify changes in gene expression following treatment with TGFβ1 and TGFβ2, and to enable putative genes relating to TGFβ regulation and PCO to be identified.

Methods: Routinely cultured FHL 124 cells maintained in serum free Eagle's Minimum Essential Medium (EMEM) were treated with either TGFβ1 or TGFβ2 at 10 ng/ml for 24 h then total RNA extraction was carried out. Total RNA (16 μg) was used to analyze gene expression by spotted oligonucleotide microarray hybridization. The spotted oligonucleotide microarrays employed contained 13,971 oligonucleotide probes, each designed to be specific for an individual gene. Array images were analyzed using GenePix Pro 3.0, followed by raw data import into GeneSpring 7.0 where a cross gene error model (CGEM) filter was applied. Data was subjected to LoWess normalization prior to comparison of the different treatment groups. Quantitative real-time polymerase chain reaction (QRT-PCR) was used to validate the oligonucleotide microarray data, using a select number of genes exhibiting differential expression.

Results: A total of 301 genes were up-regulated by more than 1.5 fold in FHL 124 cells by both TGFβ1 and TGFβ2. Many of these up-regulated genes had biological functions relevant to lens epithelial cells including roles in contraction, transdifferentiation and as extracellular matrix (ECM) components. A total of 164 genes were down-regulated by more that 1.5 fold in FHL 124 cells by both TGFβ1 and TGFβ2. Many of these down-regulated genes have biological functions including roles in apoptosis, signaling, and as anti-oxidants. Following treatment with TGFβ1 and TGFβ2, QRT-PCR successfully validated the differential changes in gene expression detected by oligonucleotide microarrays.

Conclusions: TGFβ1 and TGFβ2 regulate the gene expression of genes that have important roles in human lens epithelial cell biology. Most importantly, TGFβ induces the gene expression of a number of fibrotic markers which may have a role in promoting the development of PCO such as transdifferentiation markers, contractile factors, and ECM components.

Introduction

The transforming growth factor beta (TGFβ) family of cytokines regulate fundamental aspects of cellular function, which include cell growth, differentiation, inflammation, and wound healing [1]. In addition, substantial evidence also implicates TGFβ in many human diseases [2] including fibrotic pathologies of the lens [3-5]. Investigations have specifically proposed that TGFβ is a causative factor in the progression of posterior capsule opacification (PCO). Posterior capsule opacification (PCO) or 'after cataract' is the most common complication of cataract surgery, which results from the resilient growth onto the posterior capsule of residual lens epithelial cells [6,7]. TGFβ2, the major TGFβ isoform in the eye, has been widely investigated with respect to PCO. For instance, experimental models such as human capsular bags [6] and most recently, a human lens epithelial cell line (FHL 124) [7] have been employed to investigate the effects of TGFβ2 and its role in PCO. In these systems, TGFβ2 promotes aberrant lens cell transdifferentiation. This gives rise to functional changes including matrix contraction and increased extracellular matrix (ECM) deposition, all of which are characteristics of PCO. However, TGFβ1 may also have a potential role to play in PCO formation because following surgery (e.g. cataract surgery), active levels of TGFβ1 can be elevated due to a breakdown of the blood-aqueous barrier [8].

A body of evidence has been obtained in recent years concerning TGFβ signal transduction pathways. The major intracellular signaling system identified for TGFβ is through translocation of Smad proteins. TGFβ initiates its response through a complex of high affinity cell surface receptors consisting of two 'type I' and two 'type II' transmembrane serine/threonine kinase receptors [9]. In the presence of TGFβ ligand, the receptor-activated Smads (R-Smads), Smad2 and 3, are phosphorylated directly by the TGFβ receptor I kinase and are bound to the common mediator Smad, Smad4. The Smad2/3-Smad4 complex is free to associate with transcriptional co-activators or co-repressors before translocating to the nucleus [1]. TGFβ can terminate the induction of its own target genes by the induction of Smad7, an inhibitory Smad that prevents R-Smad phosphorylation [10]. In recent years, the TGFβ/Smad signalling pathway has been implicated in fibrotic pathologies of the lens including PCO [4,7,11]. However, a number of studies have shown that TGFβ can activate Smad independent pathways [12].

In the present study, spotted oligonucleotide microarray technology is utilized to investigate global changes in gene expression in a human lens epithelial cell line (FHL 124) treated with TGFβ1 and TGFβ2. In particular, the study focuses on the effects of TGFβ on the gene expression profiles of selected functional groups that are relevant to the cellular function of lens epithelial cells. Additionally, this study employs quantitative real-time polymerase chain reaction (QRT-PCR) to validate the microarray results on a number of genes apparently exhibiting differential expression. The data in the present study provides an information resource relating to TGFβ gene regulation and identifies the range of genes expressed in FHL 124 cells. Furthermore, the current study provides valuable relevant information to aid in the understanding of traditional signaling pathways that regulate fibrotic events which give rise to PCO.

Methods

Cell culture and total RNA extraction

FHL 124 cells were seeded on either 60 mm dishes at 200,000 cells in 3 ml of 5% FCS-Eagle's Minimum Essential Medium (EMEM) and grown for five days (for microarray analysis) or 35 mm dishes at 30,000 cells in 400 μl of 5% FCS-EMEM and grown for three days (for QRT-PCR validation). The medium was replaced with nonsupplemented EMEM and cells were cultured for a further 24 h prior to treatment with 10 ng/μl TGFβ1 or TGFβ2 for 24 h. At the end of the 24 h experimental period, total RNA was isolated from the cells using an RNeasy® mini kit (Qiagen Ltd., Crawley, UK) according to the manufacturer's instructions. Total RNA yield was quantified using a NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Delaware) and RNA quality was checked using a RNA 600 Nano Assay Kit.

Oligonucleotide microarray analysis

For microarray analysis, total RNA was prepared from untreated (control), TGFβ1-, and TGFβ2-treated cells. RNA samples were prepared from four replicate experiments. Each experiment was performed on a separate day to obtain four true biological replicate samples for each treatment. The spotted oligonucleotide microarrays employed contained 13,971 oligonucleotide probes, each designed to be specific for an individual gene [13]. A reference design format was used for the array analysis. Cy5-labeled cDNA was produced by reverse transcription of each of the different test RNA samples using a well established protocol described by Dr. Joseph DeRisi. Cy3-labeled reference cDNA samples were prepared in the same way from a single large pooled sample of control FHL 124 cell RNA. The One Cy5-labeled test sample was hybridized to each array, together with Cy3-labeled reference cDNA, as previously described [13].

Following hybridization and washing, the arrays were scanned using an Agilent G2565BA microarray scanner system (Agilent Technologies UK Limited, South Queensferry, UK). Flawed features, identified by visual inspection of the array images, were flagged and the data from these features were removed prior to further analysis. The data from all arrays were analyzed using GeneSpring 7.2 (Agilent Technologies UK Limited, South Queensferry, UK). The data were normalized in two different ways according to the subsequent analytical approach to be employed. In both cases, the first step involved normalization of the datasets for each individual array using locally weighted (LoWess) linear regression [14]. For statistical evaluation, the data from all arrays were normalized on a gene by gene basis against the mean expression ratio across the four control samples. Alternatively, for subsequent analysis based on arbitrary cutoff values for apparent up- and down-regulation, the data for each of the four replicate experiments were normalized as separate sets on a gene-by-gene basis against the expression ratio for the corresponding gene in the appropriate control sample. This approach reduced the expression ratio for every gene in each of the control samples to 1 and set the expression ratio for each gene in each test sample to a value indicative of its expression relative to that in the corresponding control.

Statistical analysis was performed on the log expression ratios that were produced using the first normalization strategy utilizing Welch's one way analysis of variance with Benjamini and Hochberg multiple test correction. Tukey's test was used for post-hoc analysis.

The normalized data obtained using the second normalization strategy were analyzed to identify genes whose expression appeared to be consistently (i.e. in at least three of the four biological replicates) up- or down-regulated by an arbitrary cutoff of at least 1.5 fold. The cross gene error model (CGEM) within GeneSpring was used to identify genes for which the signal strength was consistently high enough to be considered high trust. This was performed by identifying the gene set for which the raw control (Cy3) signal intensity was greater than the average value of base/proportional error for the entire set of arrays, calculated using the CGEM model (based on deviation from 1) on at least 11 of the 12 of arrays [13].

Quantitative real-time polymerase chain reaction

Five hundred nanograms of total RNA from FHL 124 cells were reverse-transcribed in a 20 μl reaction mixture using Superscript II^TM (Invitrogen, Paisley, UK). The QRT-PCR was performed with an Opticon 2 DNA Engine (MJ Research Inc, Reno, NV) to validate the microarray results on a number of genes apparently exhibiting differential expression. Primer oligonucleotide sequences specific for the genes examined are shown in Table 1. Level of product was determined by SYBR® Green (Finnzymes, Espoo, Finland), which binds exclusively to double stranded DNA resulting in a fluorescence emission. Therefore, the product is proportional to fluorescence. A 50 μl reaction mixture was prepared for each cDNA sample containing: 50 ng cDNA; SYBR® green 2X; 2 μM forward and reverse primers (Invitrogen) and ddH₂O to total final volume. Serial dilutions of cDNA known to express the gene of interest were prepared to permit relative levels between test samples to be determined. QRT-PCR was performed using the following program: (step 1) initial denaturation at 94 °C for four min; (step 2) denaturation at 94 °C for 20 s; (step 3) annealing at 55 °C for 30 s, except for Smad7 and TGFβR3 which was at 65 °C and 50 °C for 30 s, respectively; (step 4) extension at 72 °C for 20 s; and (step 5) 'cut off' for 10 s to denature potential primer dimers (Table 1). The 'cut off' temperature cycle was then followed by a fluorescent dye measurement. Steps 2-5 were repeated for 35 cycles for all gene primers except BMP4 and Smad4, which were repeated for 25 cycles. Melting curve analysis was performed to determine the quality of the amplified gene product. The principle of the melting curve analysis is that a single peak on the melting curve indicates a 'clean' amplified gene product not contaminated with primer dimers and gives a single band of the correct base pair size when run on a 0.8% agarose gel. Contamination of the amplified gene product with primer dimers and non-specific products is indicated by multiple melting curve peaks which give multiple bands when run on a 0.8% agarose gel. All the gene products analyzed in this study produced single melting curve peaks and single bands of the correct base pair size when run on a 0.8% agarose gel.

Statistical analysis of quantitative real-time polymerase chain reaction results

A t-test analysis using Excel software (Microsoft, Redmand, WA) was performed to determine any statistical difference between experimental groups; significance was assessed using a p-value of less than or equal to 0.05.

Results

Genes up-regulated by both TGFβ1 and TGFβ2 in FHL 124 cells

The spotted oligonucleotide microarray employed in this investigation contained 13,971 oligonucleotide probes, each designed to be specific for an individual gene. Of these 13,971, a total of 6,290 genes passed the CGEM criteria set out above. When Welch's one way analysis of variance was employed together with Benjamini and Hochberg multiple test correction to compare the gene expression in the three different groups, a total of 33 genes were identified as being altered significantly (p<0.05) in one or more of the treatment groups. However, when Tukey's test was used for post hoc analysis, only one of the genes (smooth muscle actin) appeared to be altered by TGFβ (1 or 2) exposure. Since multiple test corrections applied to datasets such as these can lead to a significant number of type 2 errors (false negatives), we also employed an alternative approach based on arbitrary changes in apparent gene expression to look for more genes whose expression was altered by exposure to the TGFβ isoforms. A total of 301 genes were up-regulated by more than 1.5 fold in FHL 124 cells by both TGFβ1 and TGFβ2 in at least three out of four biological replicates. Of the 301 genes up-regulated by both TGFβ isoforms, 167 passed the cross gene error model (CGEM) filter. The CGEM uses a mathematical algorithm (available within the GeneSpring v7.0 software that was used for the data analysis) to model the variation in the Cy5 signal intensities of all the features on the arrays compared with Cy3 signal intensities and to determine signal intensities below which data may be considered as potentially unreliable. Based on the result of the model fitting for this data set, the CGEM filter used excluded those genes that had a control (Cy3 channel) signal of less than 130 (base/proportional error) in more than one out of eight samples (i.e. both TGFβ1- and 2-treated). Many of the genes up-regulated by both TGFβ isoforms that were either above (Table 2) or below (Table 3) the CGEM filter have biological functions relevant to lens epithelial cells including roles in contraction, transdifferentiation, and as ECM components.

Genes down-regulated by both TGFβ1 and TGFβ2 in FHL 124 cells

A total of 164 genes were down-regulated by more than 1.5 fold in FHL 124 cells by both TGFβ1 and TGFβ2 in at least three out of four samples. Of the 164 genes down-regulated by both TGFβ isoforms, 104 passed the CGEM filter. Many of the genes down-regulated by both TGFβ isoforms that were either above (Table 4) or below (Table 5) the CGEM filter have biological functions relevant to lens epithelial cells including roles in apoptosis, signaling, and as anti-oxidants.

Quantitative real-time polymerase chain reaction validation of oligonucleotide microarray

QRT-PCR was used to verify oligonucleotide microarray expression data of selected genes that showed up-regulation, down-regulation, or no change in expression following treatment with TGFβ1 and 2. The same FHL 124 RNA samples employed in the microarray analysis could not be used for QRT-PCR analysis due to low sample volumes. Therefore, different FHL 124 samples treated with exactly the same experimental conditions (either TGFβ1 or TGFβ2 (10 ng/ml) for 24 h) were used for QRT-PCR analysis. Use of separately generated samples enhanced the validation process as it provided independent confirmation and demonstrated that the results observed with the microarrays are reproducible. The up-regulation of αSMA, CTGF, Smad7, and MMP2 gene expression in FHL 124 cells following TGFβ treatment was confirmed by QRT-PCR analysis. This analysis showed there was a significant difference between TGFβ-treated and untreated samples (Figure 1 and Figure 2). Microarray analysis of bFGF and Smad4 gene expression showed no change following TGFβ treatment. QRT-PCR confirmed these results and no significant difference between TGFβ-treated and untreated samples was obtained (Figure 3). The down-regulation of TGFβR3 and BMP4 gene expression in FHL 124 cells following TGFβ treatment was confirmed by QRT-PCR analysis, where there was again a significant difference between TGFβ treated and untreated samples (Figure 4). Such a high success rate of identifying differentially expressed genes from the arrays provides a high degree of confidence in the array results even for genes that have not been independently validated by QRT-PCR.

Gene expression profiling of FHL 124 cells

The following data address groups of the genes of interest to the current study. This section covers both the baseline gene expression and the gene expression following TGFβ treatment in FHL 124 cells of selected receptors, growth factors, signaling components, matrix components, myosin light chains, and myosin regulatory enzymes. This section will enable those genes that may be of interest in PCO formation, particularly the processes of transdifferentiation and matrix contraction, to be considered.

Gene expression profile of growth factor receptors in FHL 124 cells

TGFβR2 and TGFβR3 were detected in FHL 124 cells (Figure 5A) with TGFβR2 giving a higher baseline signal value compared to TGFβR3. The baseline signal of TGFβR1 was undetected; however, this appears to be anomalous as QRT-PCR revealed its presence (data not shown). TGFβR2 gene expression remained unchanged following treatment with both TGFβ1 and TGFβ2; in contrast, TGFβR3 gene expression was down-regulated more than two-fold (Figure 5B). Of the three bone morphogenic protein (BMP) receptors detected in FHL 124 cells BMPR2 had the highest value for baseline signal expression (Figure 5A). The gene expression of BMPR1A remained unchanged following treatment with TGFβ1 and TGFβ2 (Figure 5B). BMPR1B and BMPR2 gene expression was up-regulated more than 1.5 fold following TGFβ1 treatment. Of the four fibroblast growth factor (FGF) receptors detected in FHL 124 cells, FGFR1 had the highest value for baseline signal expression (Figure 5A). The gene expression of the four FGF receptors remained unchanged following treatment with TGFβ1 and TGFβ2 (Figure 5B). Three epidermal growth factor (EGF) receptors family members, ERBB1, ERBB2, and ERBB3, were detected in FHL 124 cells with ERBB2 giving the highest baseline signal value (Figure 5A). Of the three EGF receptors expressed in FHL 124 cells, only ERBB3 was affected by TGFβ treatment where TGFβ2 down-regulated its gene expression by more than two-fold (Figure 5B).

Gene expression profile of growth factors in FHL 124 cells

TGFβ isoforms 1, 2, and 3 were detected in FHL 124 cells. However, the baseline signal of TGFβ3 was low with a value less than 130 (Figure 6A). TGFβ2 gene expression was up-regulated more than two-fold following treatment with TGFβ1 and TGFβ2. In contrast, the gene expressions of TGFβ1 and TGFβ3 were unchanged by TGFβ treatment (Figure 6B). Connective tissue growth factor (CTGF) was detected in FHL 124 cells with a high baseline signal value of more than 8000 (Figure 6A). However, CTGF was upregulated by more than two fold following treatment with TGFβ1 and TGFβ2 (Figure 6B). Out of 10 BMP isoforms on the microarray, four were not detected in FHL 124 cells (Figure 6A). Of the six BMP members detected BMPs 1, 4, and 5 gave relatively high baseline signal values of more than 2000. Treatment of FHL 124 cells with TGFβ1 and TGFβ2 up-regulated BMP1 gene expression by 1.5 fold and down-regulated BMP4 gene expression by more than 1.5 fold (Figure 6B). The gene expression of all other BMPs detected remained unchanged following TGFβ treatment. Of 19 FGF isoforms on the microarray, six were not detected in FHL 124 cells (Figure 6A). Of the 13 FGF's detected, seven FGF isoforms had low baseline signal values of less than 130. The gene expressions of FGF1, FGF4, and FGF5 were up-regulated more than 1.5 fold following treatment with TGFβ (Figure 6B). FGF7 gene expression was down-regulated more than two-fold following TGFβ2 treatment. The gene expressions of all remaining FGFs detected in FHL 124 cells were unchanged by TGFβ treatment. The growth factors, EGF and HB-EGF (heparin binding-EGF), were detected in FHL 124 cells (Figure 6A). The gene expression of HB-EGF was up-regulated more than two-fold following treatment with TGFβ1 and TGFβ2 (Figure 6B). In contrast, EGF gene expression was unchanged by TGFβ treatment.

Gene expression profile of signaling components in FHL 124 cells

All eight Smad family members present on the microarray were detected in FHL 124 cells and showed varying baseline signal values with Smad6 having the lowest value of 233 (Figure 7A). Both Smad6 and Smad7 gene expression was upregulated by 1.5 fold and 2 fold, respectively, following treatment with TGFβ1 and 2 (Figure 7B). The gene expression of all other Smads detected remained unchanged following TGFβ treatment. Both TNF receptor-associated protein (TRAP-1) and SMURF were detected in FHL 124 cells (Figure 7A). With respect to TGFβ treatment, TRAP-1 gene expression was unchanged. However, Smad ubiquitylation regulatory factor (SMURF) gene expression was upregulated 1.5 fold following treatment with TGFβ1 (Figure 7B).

Of 12 mitogen activated protein kinase (MAPK) family members present on the microarray, two were not detected in FHL 124 cells (Figure 7A). The MAPK members detected varying baseline signal values with MAPK1 and 4 having low signal values of less than 130. MAPK7 gene expression was upregulated 1.5 fold following treatment with both TGFβ isoforms, MAPK 13 was upregulated 1.5 fold by TGFβ1 only (Figure 7B). The gene expressions of all other MAPK members remain unchanged following TGFβ treatment.

Inhibitors of FGF signaling, SPRY1 (sprouty 1), SPRY2, and similar expression to FGFs (SEF), were detected in FHL 124 cells. However, the baseline signal of SEF was low with a value less than 130 (Figure 7A). Treatment with TGFβ affected the gene expression of SPRY1 and 2; SPRY1 gene expression was down-regulated 1.5 fold by TGFβ1, whereas SPRY2 gene expression was down-regulated 1.5 fold by both TGFβ isoforms (Figure 7B). The gene expression of SEF remained unchanged following TGFβ treatment.

Gene expression profile of matrix components in FHL 124 cells

Of 33 collagens present on the microarray, eight were not detected in FHL 124 cells (Figure 8A). The collagens show varying baseline signal values with collagen 1α2 (COL1A2) and collagen 4α2 (COL4A2) having high signal values over 8000. With respect to TGFβ treatment, the gene expression of nine collagens including collagens 1α2 and 4α2 were up-regulated by at least 1.5 fold (Figure 8B). The gene expression of collagens 4α5, 4α6, and 5α3 were downregulated by at least 1.5 fold following TGFβ treatment. The gene expression of all other detected collagens remained unchanged following TGFβ treatment. Fibronectin was detected in FHL 124 cells with a high baseline signal value of over 8000 (Figure 8A). Following TGFβ1 and 2 treatment, fibronectin gene expression was upregulated more than 2 fold (Figure 8B). Of 11 laminin family members present on the microarray, three were not detected in FHL 124 cells (Figure 8A). The laminin members showed varying baseline signal values with laminin α2 (LAMA2) having the lowest baseline signal of less than 130. Laminin α2 (LAMC2) gene expression was up-regulated more than two-fold following treatment with both TGFβ1 and 2 (Figure 8B). In addition, the gene expressions of laminin β3 (LAMB3) and laminin γ1 (LAMC1) were up-regulated more than 1.5 fold following treatment with TGFβ1 and TGFβ2, respectively. The gene expression of laminin α3 (LAMA3) was down-regulated at least 1.5 fold following treatment with both TGFβ isoforms. The gene expression of all other detected laminins remained unchanged following TGFβ treatment.

Gene expression profile of integrins in FHL 124 cells

Of 28 integrins present on the microarray, seven were not detected in FHL 124 cells (Figure 9A). The integrins showed varying baseline signal values with integrins α3, α5, and β1 having high baseline signal values over 8000. With respect to TGFβ1 and 2 treatments, the gene expressions of integrins α5, α11, αV, and β5 were up-regulated by at least 1.5 fold (Figure 9B). The gene expression of integrin α4 was upregulated by TGFβ2 only. In contrast, integrin β8 gene expression was down-regulated by more than 1.5 fold by both TGFβ isoforms. The gene expressions of the remaining 15 integrins were unchanged by TGFβ treatment.

Gene expression profile of myosin light chains and regulatory enzymes in FHL 124 cells

All seven members of the myosin light chain family were detected on the microarray. These showed varying baseline signal values with MLCB and MYL6 having high baseline signal values over 8000 whereas MYL3 and MYL4 had low baseline signal values of less than 130 (Figure 10A). Expression levels for each of these genes were unaffected by either TGFβ treatment (Figure 10B). With respect to myosin regulatory enzymes, MYLK (also known as MLCK; myosin light chain kinase) and myosin phosphotase 1 (MYPT1) were detected but MYPT2 was not (Figure 10A). Only the gene expression of MYLK (myosin light chain kinase) was upregulated (more than two-fold) following TGFβ treatment (Figure 10B).

Discussion

The microarray is an advanced molecular biological technology which allows for the study of gene expression on a global level. In the present study, microarrays were employed to investigate gene expression in the human lens epithelial cell line, FHL 124. A previous study that used Affymetrix gene microarrays identified a 99.5% homology between FHL 124 cells and native human lens epithelium with both preparations expressing lens epithelial cell phenotype markers such as FOXE3 [7]. The present study further investigates gene expression in FHL 124 cells by identifying a range of genes regulated by TGFβ1 and TGFβ2, which have been implicated in many lens pathologies [3,6,7,15]. Importantly, the present work provides valuable information regarding possible signaling pathways that regulate the fibrotic processes that promote the development of PCO [5]. TGFβ is a potent activator of transdifferentiation in a number of cell types throughout the body (reviewed in [16]). In this study, FHL 124 cells treated with both TGFβ1 and 2 showed an up-regulation in α smooth muscle actin (αSMA) gene expression, which is the major marker of transdifferentiation [17]. In addition, the gene expression of fibronectin, an ECM component and marker of transdifferentiation [16], was also up-regulated by TGFβ1 and 2. Interestingly, it has been proposed that TGFβ-induced fibronectin synthesis is important for the induction of αSMA expression and thus transdifferentiation [18]. In addition to the more traditional markers of transdifferentiation, TGFβ1 and 2 isoforms also up-regulated a number of novel transdifferentiation genes in FHL 124 cells, which are expressed by both fibroblasts and smooth muscle cells. These genes include fibroblast activating protein [16], tenascin C [19], smoothelin [20], and transgelin [21,22]. Altogether, from the array of transdifferentiation genes up-regulated by TGFβ1 and 2, it can be proposed that TGFβ induces FHL 124 cells to change phenotype from a lens epithelial cell to a myofibroblast. A previous study by Gordon-Thomson et al. [3] revealed TGFβ2 was 10 times more potent than TGFβ1 at inducing αSMA expression (transdifferentiation) in the rat lens epithelium. However, this study used TGFβ at concentrations ranging from 0.02-4.00 ng/ml. The present investigation employed 10 ng/ml each of TGFβ1 and 2 in order to promote significant changes in global gene expression rather than to detect comparative changes in select genes using lower concentrations of TGFβ. Therefore, with respect to FHL 124 cells, lower dose response concentrations of TGFβ1 and 2 isoforms would need to be applied in order to determine clear differences in isoform potency regarding TGFβ induced events.

Following cataract surgery, the laying down of ECM by lens epithelial cells is an important aspect of wound healing [23]. The present work reveals that along with fibronectin, the gene expression of various ECM components including collagen types I, IV, and XVIII are up-regulated by TGFβ in FHL 124 cells. Previous investigations have shown TGFβ to be a potent inducer of ECM expression and deposition in fibrotic conditions throughout the body [24]. The expression of collagen types I-IV has been detected in lens epithelial cells of human cataracts [25]. In addition, collagen XVIII has been detected in the lens capsule under normal conditions [26] and following trauma, its expression can increase playing a role in wound healing [27]. Collagen I expression is associated with PCO and, most importantly, the process of transdifferentiation where collagen I deposits have been observed to form around 'myofibroblast like' lens cells [28,29]. The novel ECM components, lysyl oxidase (LOX) and transglutaminase 2, are up-regulated in FHL 124 cells following treatment with TGFβ1 and 2. Both LOX and transglutaminase 2 have important functional roles in stabilizing the interaction between the cell and ECM by cross-linking ECM proteins such as fibronectin and collagen [30,31]. Importantly, transglutaminase 2 has been detected in cataractous lenses [32] and its up-regulation by TGFβ has been associated with the formation of ASC plaques [33].

In the present study the gene expression of αSMA, fibronectin, and collagen I was detected in non-stimulated FHL 124 cells. These proteins are normally undetectable in the native lens epithelium [6,20,34]. However, a recent study has shown αSMA to be expressed in lens epithelial cells of several species [35]. Moreover, αSMA expression has previously been detected in non-stimulated in vivo capsular bag cultures. As a result, it has been proposed that αSMA may be expressed in a vigorously growing or wounded system [7] thus, indicating the regulation of this gene through mechanical trauma. Such discrepancies in gene expression may arise due to FHL 124 cells being grown on a foreign matrix (i.e. plastic) rather than their native collagen IV matrix. As extracellular matrix interactions are important in regulating the gene expression of αSMA [18], it is a possibility that the collagen IV capsule under 'normal' (non-stimulated) conditions may regulate the expression of certain genes in lens epithelial cells. FHL 124 cells produce their own extracellular matrix (ECM) [7], which may explain why fibronectin and collagen I show abundant expression in non-stimulated FHL 124 cells. Altogether, this data suggests that the FHL 124 cell line does not strictly represent the normal native lens epithelium and perhaps is a better representation of a wounded system. Therefore, application of the FHL 124 cell line may be of great benefit to investigating cellular processes that give rise to PCO following cataract surgery.

The gene expression of many members of the integrin family of cell adhesion molecules were detected in FHL 124 cells in the present study. Integrins are composed of α and β chains and act as cell surface receptors for ECM components, the binding of integrin receptors to their ECM ligands enables cells to adhere to and migrate across the ECM [36]. The gene expression levels of α3, α5, and β1 integrins in FHL 124 cells were abundant under non-stimulated conditions. β1 integrin is the most widely expressed integrin on many different cell types [37]. Of all the integrin chains expressed in FHL 124 cells, β1 integrin is probably the most important as it can associate with 12 different α integrin chains [37] therefore, forming many different α-β1 dimer combinations that one or more ECM ligands can bind. Both TGFβ1 and TGFβ2 up-regulate the gene expressions of a number of integrins in FHL 124 cells including α5, αV, and β5 integrins. As a result α5-β1, αV-β1, and αV-β5 integrin combinations may dimerize and show increased expression in FHL 124 cells following TGFβ stimulation. αV-β1 integrin is the receptor for both fibronectin and vitronectin, whereas αV-β5 integrin is a receptor only for vitronectin [37]. The α5-β1 integrin binds solely to fibronectin and is recognized to be the main receptor for this ligand [38]. Data suggest α5-β1 integrin to be the most highly expressed integrin in FHL 124 cells following TGFβ treatment due to the α5 integrin chain possessing a high signal level of gene expression in non-stimulated conditions. Therefore, the overall potential for α5-β1 integrin to dimerize and subsequently bind to fibronectin is greatly enhanced by the addition of TGFβ1 or 2. Previous studies using FHL 124 cells have reported a redistribution of α5-β1 integrin can occur in response to TGFβ [39]. Most importantly, it has been reported that the inhibition of the fibronectin/fibronectin receptor interaction prevents TGFβ-induced αSMA expression [40] therefore, interaction between fibronectin and its integrin receptor can regulate the expression of αSMA (transdifferentiation) [40]. Furthermore, it has been proposed that fibronectin and its α5-β1 integrin form a putative contractile apparatus with αSMA [40,41].

In the current study TGFβ1 and 2 up-regulate a number of genes in FHL 124 cells that have functional roles in the process of contraction. Many studies regarding PCO suggest that active levels of TGFβ induce wrinkling of the posterior capsule as a result of matrix contraction [5]. In non-lenticular studies it has been proposed that αSMA is essential to matrix contraction [41-43]. In addition to αSMA, TGFβ1 and 2 up-regulate the gene expression of contractile factors α cardiac actin and α actin 1 in FHL 124 cells [44,45] along with more novel contractile genes such as smoothelin [20] and paladin [46]. Myosin light chain kinase (MLCK) and myosin phosphatase 1 (MYPT1) gene expressions were detected in FHL 124 cells, both of which have fundamental roles in the contraction of smooth muscle [47]. The activity of myosin, a fundamental component of the contractile apparatus, is regulated by the balance in expression and/or activity of MYPT1, which suppresses contraction and MLCK, which promotes contraction [47]. The present study reveals that TGFβ1 and 2 up-regulate MLCK gene expression in FHL 124 cells but do not effect MYPT1 gene expression. Therefore, the balance between MLCK and MYPT1 is likely to be disrupted following treatment with TGFβ, resulting in promotion of MLCK activity, which indicates contraction suggested by greater myosin activity.

FHL 124 cells express a range of growth factors including members of the TGFβ, BMP, FGF, and EGF family and their corresponding receptors, all of which have important roles in lens cell biology [48,49]. The gene expressions of TGFβ isoforms 1, 2, and 3 were detected in FHL 124 cells. However, of the three isoforms, TGFβ3 showed a very low level of gene expression. This is a similar result to a previous observation in the rat lens epithelium where TGFβ3 gene expression was negligible [3]. In the present investigation, both TGFβ isoforms up-regulated TGFβ2 expression in the FHL 124 cell line. A similar response has also been observed with TGFβ2 in human lens capsular bags [50] thus, ultimately increasing the overall pool of TGFβ2 available to lens epithelial cells. TGFβR2 and TGFβR3 are expressed in FHL 124 cells. However, surprisingly, TGFβR1 expression is undetected.This particular result appears to be anomalous as TGFβR1 is essential for signaling [1]. Therefore, FHL 124 cells would not respond to TGFβ without TGFβR1. Moreover, QRT-PCR revealed the presence of TGFβR1 in FHL 124 cells. In addition, TGFβR1 expression was also detected in primary lens cells and the native lens epithelium [15,51]. It is likely that TGFβR1 expression was undetected due to limiting experimental factors such as the oligonucleotide probe and/or hybridization inefficiency [52]. A surprising finding from the present study is that TGFβ1 and 2 down-regulate TGFβR3 expression in FHL 124 cells. TGFβR3 is important in TGFβ2 signaling [53,54] as it binds to TGFβ2 ligand with high affinity and presents it to TGFβR2 [55]. The importance of TGFβR3 in TGFβ2 signaling has been confirmed in investigations where cells lacking TGFβR3 do not respond to TGFβ2 [53,54]. In contrast, TGFβ isoforms 1 and 3 can bind TGFβR2 directly and do not require TGFβR3 [1]. Therefore, when lens epithelial cells are in the presence of high levels of both TGFβ1 and TGFβ2 (e.g. following cataract surgery), TGFβ2 signaling may be impaired due to low levels of TGFβR3 expression. Amongst the members of the FGF family of growth factors and receptors detected in FHL 124 cells, basic FGF (bFGF) and FGFR1 showed relatively high levels of gene expression. A previous investigation detected bFGF and FGFR1 gene expression in cultured human lens capsular bags and in capsular bags removed from donor eyes that had previously undergone cataract surgery [56]. Furthermore, following cataract surgery, the concentration of bFGF in the aqueous humor [57,58] and native lens epithelium can increase dramatically [56]. Recent studies suggest bFGF can exacerbate TGFβ-induced cellular events that give rise to lens pathologies [59,60]. Therefore, with respect to FHL 124 cells, bFGF may affect TGFβ-induced events such as transdifferentiation and matrix contraction. TGFβ up-regulates a number of growth factors that are implicated in fibrotic pathologies of the lens including PDGF [49], IL6 [23], and CTGF [5]. Of particular interest is CTGF (connective tissue growth factor), which has also been identified in capsular bags [61] and is upregulated in the lens epithelium of patients with ASC [15]. It has been proposed that CTGF may act as a downstream mediator of some TGFβ-promoted fibrotic effects, such as transdifferentiation and matrix contraction [24,62].

A vast array of signaling components are expressed in FHL 124 cells with many of these being regulated by TGFβ. Both TGFβ1 and 2 up-regulate the gene expression of LTBP2/3 (latent TGFβ binding protein 2/3) and thrombospondin-1 in FHL 124 cells, which have opposing effects regarding TGFβ signaling. LTBP2/3 bind to TGFβ isoforms making them incapable of binding to TGFβ receptors [63,64], whereas thrombospondin-1 activates TGFβ from its latent form by removing LTBP via proteolytic cleavage [65]. Therefore, the balance between latent and active TGFβ in FHL 124 cells is probably unchanged following TGFβ treatment. All members of the Smad family of signaling molecules and their associated components (i.e. TRAP1 and SMURF) are expressed in FHL 124 cells. This suggests an active signaling pathway is in place. Smad signaling is an integral component of TGFβ mediated events and several studies have evidence of active Smad signaling in the lens and in association with lens pathologies [4,7,11,66]. Smads 2, 3, and 4 show high levels of expression in FHL 124 cells and are essential to the TGFβ/Smad signaling pathway [1]. In a previous study employing FHL 124 cells, αSMA expression and matrix contraction were proposed to be a consequence of increased Smad4 transcriptional activity in response to TGFβ2 [7]. Smad7, the major inhibitory regulator of the TGFβ/Smad signaling pathway is expressed in FHL 124 cells and furthermore, is up-regulated following TGFβ1 and TGFβ2 treatment. This, therefore, indicates that Smad7 is involved in a negative feedback mechanism. TGFβ also down-regulates the gene expression of Ets-V1 (Ets variant gene 1) in FHL 124 cells, which activates Smad7 gene expression independently of TGFβ/Smad signaling [67]. Thus, following TGFβ treatment, the Smad7 negative feedback mechanism may dominate regulation of TGFβ/Smad signaling in FHL 124 cells.

The gene expression of members of the MAP kinase (MAPK) signaling family such as ERK1 (MAPK3), JNK (MAPK8), and p38 (MAPK14) were detected in FHL 124 cells. FGF and EGF signal via the MAPK pathway [68,69]. Furthermore, FGF signaling via the ERK/MAPK pathway is essential to lens cell proliferation and differentiation [70]. Thus, suggesting that an active bFGF and EGF pathway is present in FHL 124 cells. Importantly, ERK/JNK/p38 MAPK have all been implicated in matrix contraction [71,72]. In particular, the ERK/MAPK pathway promotes matrix contraction by its activation of MLCK [73]. TGFβ/Smad signaling is tightly controlled by MAP kinase signaling cascades [24]. ERK phosphorylation sites exist in the linker regions of Smad3 and Smad4, which enables the ERK/MAPK signaling cascade to modify Smad activity via multiple pathways [24]. Furthermore, TGFβ can signal independent of Smads via the ERK/JNK/p38 MAPK signaling pathways [12]. In non-lenticular systems the up-regulation of fibronectin and CTGF expression by TGFβ are Smad-independent processes requiring JNK/MAPK and ERK/MAPK [24]. Consequently, the precise role of the MAPK signaling pathway in TGFβ-induced events such as transdifferentiation and matrix contraction in FHL 124 cells needs further investigation.

The present study reveals a variety of transcription factors and signaling molecules that are regulated by TGFβ in FHL 124 cells. TGFβ up-regulates the gene expressions of many transcription factors in FHL 124 cells including the co-activators, JUN-B [74], P8 [75], and FOXO1A [76], which promote Smad transcriptional activity. In contrast, TGFβ also up-regulates SnoN expression in FHL 124 cells, a well characterized co-repressor of TGFβ/Smad signaling that directly binds R-Smads resulting in the disruption of the R-Smad-Smad4 complex [1]. This suggests that a negative feedback mechanism with respect to SnoN is promoted following TGFβ treatment of FHL 124 cells. TGFβ also down-regulates the gene expression of a number of transcription factors and signaling molecules in FHL 124 cells including Sprouty 2 and PNP1 (dual specificity phosphatase 6). Sprouty 2 is an intracellular antagonist of FGF signaling and an inhibitor of the FGF/MAPK pathway [68]. PNP1 is a negative regulator of the ERK/MAPK signaling pathway [77]. Therefore, the TGFβ-induced down-regulation of Sprouty 2 and PNP1 gene expression in FHL 124 cells may help promote bFGF signaling via the ERK/MAPK pathway, which may impact on TGFβ-induced events.

The present study successfully employed microarrays to identify a number of genes that are expressed in FHL 124 cells. TGFβ regulates the gene expression of a multitude of genes that have important functional roles in lens epithelial cell biology. Most notably, TGFβ induces the gene expressions of many transdifferentiation markers and contractile factors along with the genes that may regulate these processes. Lens epithelial cell transdifferentiation and matrix contraction are detrimental processes in PCO. Therefore, the mechanisms that induce these functional events merit further investigation. The current study reveals that TGFβ1 and 2 could potentially signal through the Smad signaling pathway and the MAPK pathway (Smad-independent) in FHL 124 cells. Therefore, the signaling pathways promoting TGFβ-induced transdifferentiation, matrix contraction, and other important TGFβ events need to be determined. Altogether, the information gained from this study aids the general understanding of the cellular changes induced by TGFβ, which give rise to PCO.

Acknowledgements

Funding: Cambridge Antibody Technology; BBSRC; NEI and the Humane Research Trust

References

1. Massague J. How cells read TGF-beta signals. Nat Rev Mol Cell Biol 2000; 1:169-78.

2. Blobe GC, Schiemann WP, Pepin MC, Beauchemin M, Moustakas A, Lodish HF, O'Connor-McCourt MD. Functional roles for the cytoplasmic domain of the type III transforming growth factor beta receptor in regulating transforming growth factor beta signaling. J Biol Chem 2001; 276:24627-37.

3. Gordon-Thomson C, de Iongh RU, Hales AM, Chamberlain CG, McAvoy JW. Differential cataractogenic potency of TGF-beta1, -beta2, and -beta3 and their expression in the postnatal rat eye. Invest Ophthalmol Vis Sci 1998; 39:1399-409.

4. Saika S, Miyamoto T, Ishida I, Shirai K, Ohnishi Y, Ooshima A, McAvoy JW. TGFbeta-Smad signalling in postoperative human lens epithelial cells. Br J Ophthalmol 2002; 86:1428-33.

5. Wormstone IM. Posterior capsule opacification: a cell biological perspective. Exp Eye Res 2002; 74:337-47.

6. Wormstone IM, Tamiya S, Anderson I, Duncan G. TGF-beta2-induced matrix modification and cell transdifferentiation in the human lens capsular bag. Invest Ophthalmol Vis Sci 2002; 43:2301-8.

7. Wormstone IM, Tamiya S, Eldred JA, Lazaridis K, Chantry A, Reddan JR, Anderson I, Duncan G. Characterisation of TGF-beta2 signalling and function in a human lens cell line. Exp Eye Res 2004; 78:705-14.

8. Ohta K, Yamagami S, Taylor AW, Streilein JW. IL-6 antagonizes TGF-beta and abolishes immune privilege in eyes with endotoxin-induced uveitis. Invest Ophthalmol Vis Sci 2000; 41:2591-9.

9. Massague J. TGF-beta signal transduction. Annu Rev Biochem 1998; 67:753-91.

10. von Gersdorff G, Susztak K, Rezvani F, Bitzer M, Liang D, Bottinger EP. Smad3 and Smad4 mediate transcriptional activation of the human Smad7 promoter by transforming growth factor beta. J Biol Chem 2000; 275:11320-6.

11. Saika S, Kono-Saika S, Ohnishi Y, Sato M, Muragaki Y, Ooshima A, Flanders KC, Yoo J, Anzano M, Liu CY, Kao WW, Roberts AB. Smad3 signaling is required for epithelial-mesenchymal transition of lens epithelium after injury. Am J Pathol 2004; 164:651-63.

12. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGF-beta family signalling. Nature 2003; 425:577-84.

13. Eady JJ, Wortley GM, Wormstone YM, Hughes JC, Astley SB, Foxall RJ, Doleman JF, Elliott RM. Variation in gene expression profiles of peripheral blood mononuclear cells from healthy volunteers. Physiol Genomics 2005; 22:402-11.

14. Listgarten J, Graham K, Damaraju S, Cass C, Mackey J, Zanke B. Clinically validated benchmarking of normalisation techniques for two-colour oligonucleotide spotted microarray slides. Appl Bioinformatics 2003; 2:219-28.

15. Lee EH, Joo CK. Role of transforming growth factor-beta in transdifferentiation and fibrosis of lens epithelial cells. Invest Ophthalmol Vis Sci 1999; 40:2025-32.

16. Kalluri R, Neilson EG. Epithelial-mesenchymal transition and its implications for fibrosis. J Clin Invest 2003; 112:1776-84.

17. Darby I, Skalli O, Gabbiani G. Alpha-smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing. Lab Invest 1990; 63:21-9.

18. Serini G, Bochaton-Piallat ML, Ropraz P, Geinoz A, Borsi L, Zardi L, Gabbiani G. The fibronectin domain ED-A is crucial for myofibroblastic phenotype induction by transforming growth factor-beta1. J Cell Biol 1998; 142:873-81.

19. Tamaoki M, Imanaka-Yoshida K, Yokoyama K, Nishioka T, Inada H, Hiroe M, Sakakura T, Yoshida T. Tenascin-C regulates recruitment of myofibroblasts during tissue repair after myocardial injury. Am J Pathol 2005; 167:71-80.

20. van der Loop FT, Schaart G, Timmer ED, Ramaekers FC, van Eys GJ. Smoothelin, a novel cytoskeletal protein specific for smooth muscle cells. J Cell Biol 1996; 134:401-11.

21. Feil S, Hofmann F, Feil R. SM22alpha modulates vascular smooth muscle cell phenotype during atherogenesis. Circ Res 2004; 94:863-5.

22. Lawson D, Harrison M, Shapland C. Fibroblast transgelin and smooth muscle SM22alpha are the same protein, the expression of which is down-regulated in many cell lines. Cell Motil Cytoskeleton 1997; 38:250-7.

23. Meacock WR, Spalton DJ, Stanford MR. Role of cytokines in the pathogenesis of posterior capsule opacification. Br J Ophthalmol 2000; 84:332-6.

24. Leask A, Abraham DJ. TGF-beta signaling and the fibrotic response. FASEB J 2004; 18:816-27.

25. Nishi O, Nishi K, Fujiwara T, Shirasawa E, Ohmoto Y. Effects of the cytokines on the proliferation of and collagen synthesis by human cataract lens epithelial cells. Br J Ophthalmol 1996; 80:63-8.

26. Halfter W, Dong S, Schurer B, Cole GJ. Collagen XVIII is a basement membrane heparan sulfate proteoglycan. J Biol Chem 1998; 273:25404-12.

27. Kato T, Chang JH, Azar DT. Expression of type XVIII collagen during healing of corneal incisions and keratectomy wounds. Invest Ophthalmol Vis Sci 2003; 44:78-85.

28. Hales AM, Chamberlain CG, McAvoy JW. Cataract induction in lenses cultured with transforming growth factor-beta. Invest Ophthalmol Vis Sci 1995; 36:1709-13.

29. Hatae T, Ishibashi T, Yoshitomi F, Shibata Y. Immunocytochemistry of types I-IV collagen in human anterior subcapsular cataracts. Graefes Arch Clin Exp Ophthalmol 1993; 231:586-90.

30. Maki JM, Sormunen R, Lippo S, Kaarteenaho-Wiik R, Soininen R, Myllyharju J. Lysyl oxidase is essential for normal development and function of the respiratory system and for the integrity of elastic and collagen fibers in various tissues. Am J Pathol 2005; 167:927-36.

31. Telci D, Griffin M. Tissue transglutaminase (TG2)--a wound response enzyme. Front Biosci 2006; 11:867-82.

32. Lorand L, Hsu LK, Siefring GE Jr, Rafferty NS. Lens transglutaminase and cataract formation. Proc Natl Acad Sci U S A 1981; 78:1356-60.

33. Welge-Lussen U, May CA, Lutjen-Drecoll E. Induction of tissue transglutaminase in the trabecular meshwork by TGF-beta1 and TGF-beta2. Invest Ophthalmol Vis Sci 2000; 41:2229-38.

34. Schmitt-Graff A, Pau H, Spahr R, Piper HM, Skalli O, Gabbiani G. Appearance of alpha-smooth muscle actin in human eye lens cells of anterior capsular cataract and in cultured bovine lens-forming cells. Differentiation 1990; 43:115-22.

35. Garcia CM, Kwon GP, Beebe DC. alpha-Smooth muscle actin is constitutively expressed in the lens epithelial cells of several species. Exp Eye Res 2006; 83:999-1001.

36. Schwartz MA. Integrin signaling revisited. Trends Cell Biol 2001; 11:466-70.

37. Elner SG, Elner VM. The integrin superfamily and the eye. Invest Ophthalmol Vis Sci 1996; 37:696-701.

38. Aota S, Yamada KM. Fibronectin and cell adhesion: specificity of integrin-ligand interaction. Adv Enzymol Relat Areas Mol Biol 1995; 70:1-21.

39. Marcantonio JM, Reddan JR. TGFbeta2 influences alpha5-beta1 integrin distribution in human lens cells. Exp Eye Res 2004; 79:437-42.

40. Jester JV, Huang J, Barry-Lane PA, Kao WW, Petroll WM, Cavanagh HD. Transforming growth factor(beta)-mediated corneal myofibroblast differentiation requires actin and fibronectin assembly. Invest Ophthalmol Vis Sci 1999; 40:1959-67.

41. Jester JV, Petroll WM, Barry PA, Cavanagh HD. Expression of alpha-smooth muscle (alpha-SM) actin during corneal stromal wound healing. Invest Ophthalmol Vis Sci 1995; 36:809-19.

42. Grinnell F. Fibroblasts, myofibroblasts, and wound contraction. J Cell Biol 1994; 124:401-4.

43. Hinz B, Gabbiani G, Chaponnier C. The NH2-terminal peptide of alpha-smooth muscle actin inhibits force generation by the myofibroblast in vitro and in vivo. J Cell Biol 2002; 157:657-63.

44. Gordon AM, Homsher E, Regnier M. Regulation of contraction in striated muscle. Physiol Rev 2000; 80:853-924.

45. Mogensen J, Klausen IC, Pedersen AK, Egeblad H, Bross P, Kruse TA, Gregersen N, Hansen PS, Baandrup U, Borglum AD. Alpha-cardiac actin is a novel disease gene in familial hypertrophic cardiomyopathy. J Clin Invest 1999; 103:R39-43.

46. Mykkanen OM, Gronholm M, Ronty M, Lalowski M, Salmikangas P, Suila H, Carpen O. Characterization of human palladin, a microfilament-associated protein. Mol Biol Cell 2001; 12:3060-73.

47. Parizi M, Howard EW, Tomasek JJ. Regulation of LPA-promoted myofibroblast contraction: role of Rho, myosin light chain kinase, and myosin light chain phosphatase. Exp Cell Res 2000; 254:210-20.

48. Belecky-Adams TL, Adler R, Beebe DC. Bone morphogenetic protein signaling and the initiation of lens fiber cell differentiation. Development 2002; 129:3795-802.

49. Duncan G, Wormstone IM. Physiology Of The Lens. In: Tasman W, Jaeger EA, editors. Duane's Clinical Ophthalmology. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 1-20.

50. Wormstone IM, Anderson IK, Eldred JA, Dawes LJ, Duncan G. Short-term exposure to transforming growth factor beta induces long-term fibrotic responses. Exp Eye Res 2006; 83:1238-45.

51. Weng J, Liang Q, Mohan RR, Li Q, Wilson SE. Hepatocyte growth factor, keratinocyte growth factor, and other growth factor-receptor systems in the lens. Invest Ophthalmol Vis Sci 1997; 38:1543-54.

52. Schuchhardt J, Beule D, Malik A, Wolski E, Eickhoff H, Lehrach H, Herzel H. Normalization strategies for cDNA microarrays. Nucleic Acids Res 2000; 28:E47.

53. Brown CB, Boyer AS, Runyan RB, Barnett JV. Requirement of type III TGF-beta receptor for endocardial cell transformation in the heart. Science 1999; 283:2080-2.

54. Sankar S, Mahooti-Brooks N, Centrella M, McCarthy TL, Madri JA. Expression of transforming growth factor type III receptor in vascular endothelial cells increases their responsiveness to transforming growth factor beta 2. J Biol Chem 1995; 270:13567-72.

55. Lopez-Casillas F, Wrana JL, Massague J. Betaglycan presents ligand to the TGF beta signaling receptor. Cell 1993; 73:1435-44.

56. Wormstone IM, Del Rio-Tsonis K, McMahon G, Tamiya S, Davies PD, Marcantonio JM, Duncan G. FGF: an autocrine regulator of human lens cell growth independent of added stimuli. Invest Ophthalmol Vis Sci 2001; 42:1305-11. Erratum in: Invest Ophthalmol Vis Sci 2001; 42:1690.

57. Shirai K, Saika S, Tanaka T, Okada Y, Flanders KC, Ooshima A, Ohnishi Y. A new model of anterior subcapsular cataract: involvement of TGFbeta/Smad signaling. Mol Vis 2006; 12:681-91 <http://www.molvis.org/molvis/v12/a76/>.

58. Tanaka T, Saika S, Ohnishi Y, Ooshima A, McAvoy JW, Liu CY, Azhar M, Doetschman T, Kao WW. Fibroblast growth factor 2: roles of regulation of lens cell proliferation and epithelial-mesenchymal transition in response to injury. Mol Vis 2004; 10:462-7 <http://www.molvis.org/molvis/v10/a59/>.

59. Cerra A, Mansfield KJ, Chamberlain CG. Exacerbation of TGF-beta-induced cataract by FGF-2 in cultured rat lenses. Mol Vis 2003; 9:689-700 <http://www.molvis.org/molvis/v9/a82/>.

60. Mansfield KJ, Cerra A, Chamberlain CG. FGF-2 counteracts loss of TGFbeta affected cells from rat lens explants: implications for PCO (after cataract). Mol Vis 2004; 10:521-32 <http://www.molvis.org/molvis/v10/a64/>.

61. Wunderlich K, Pech M, Eberle AN, Mihatsch M, Flammer J, Meyer P. Expression of connective tissue growth factor (CTGF) mRNA in plaques of human anterior subcapsular cataracts and membranes of posterior capsule opacification. Curr Eye Res 2000; 21:627-36.

62. Garrett Q, Khaw PT, Blalock TD, Schultz GS, Grotendorst GR, Daniels JT. Involvement of CTGF in TGF-beta1-stimulation of myofibroblast differentiation and collagen matrix contraction in the presence of mechanical stress. Invest Ophthalmol Vis Sci 2004; 45:1109-16.

63. Bashir MM, Han MD, Abrams WR, Tucker T, Ma RI, Gibson M, Ritty T, Mecham R, Rosenbloom J. Analysis of the human gene encoding latent transforming growth factor-beta-binding protein-2. Int J Biochem Cell Biol 1996; 28:531-42.

64. Saharinen J, Hyytiainen M, Taipale J, Keski-Oja J. Latent transforming growth factor-beta binding proteins (LTBPs)--structural extracellular matrix proteins for targeting TGF-beta action. Cytokine Growth Factor Rev 1999; 10:99-117.

65. Frazier WA. Thrombospondins. Curr Opin Cell Biol 1991; 3:792-9.

66. Saika S, Okada Y, Miyamoto T, Ohnishi Y, Ooshima A, McAvoy JW. Smad translocation and growth suppression in lens epithelial cells by endogenous TGFbeta2 during wound repair. Exp Eye Res 2001; 72:679-86.

67. Dowdy SC, Mariani A, Janknecht R. HER2/Neu- and TAK1-mediated up-regulation of the transforming growth factor beta inhibitor Smad7 via the ETS protein ER81. J Biol Chem 2003; 278:44377-84.

68. Bottcher RT, Niehrs C. Fibroblast growth factor signaling during early vertebrate development. Endocr Rev 2005; 26:63-77.

69. Maidment JM, Duncan G, Tamiya S, Collison DJ, Wang L, Wormstone IM. Regional differences in tyrosine kinase receptor signaling components determine differential growth patterns in the human lens. Invest Ophthalmol Vis Sci 2004; 45:1427-35.

70. Lovicu FJ, McAvoy JW. FGF-induced lens cell proliferation and differentiation is dependent on MAPK (ERK1/2) signalling. Development 2001; 128:5075-84.

71. Borbiev T, Birukova A, Liu F, Nurmukhambetova S, Gerthoffer WT, Garcia JG, Verin AD. p38 MAP kinase-dependent regulation of endothelial cell permeability. Am J Physiol Lung Cell Mol Physiol 2004; 287:L911-8.

72. Petrich BG, Wang Y. Stress-activated MAP kinases in cardiac remodeling and heart failure; new insights from transgenic studies. Trends Cardiovasc Med 2004; 14:50-5.

73. Klemke RL, Cai S, Giannini AL, Gallagher PJ, de Lanerolle P, Cheresh DA. Regulation of cell motility by mitogen-activated protein kinase. J Cell Biol 1997; 137:481-92.

74. Liberati NT, Datto MB, Frederick JP, Shen X, Wong C, Rougier-Chapman EM, Wang XF. Smads bind directly to the Jun family of AP-1 transcription factors. Proc Natl Acad Sci U S A 1999; 96:4844-9.

75. Garcia-Montero AC, Vasseur S, Giono LE, Canepa E, Moreno S, Dagorn JC, Iovanna JL. Transforming growth factor beta-1 enhances Smad transcriptional activity through activation of p8 gene expression. Biochem J 2001; 357:249-53.

76. Massague J, Seoane J, Wotton D. Smad transcription factors. Genes Dev 2005; 19:2783-810.

77. Tonks NK, Neel BG. Combinatorial control of the specificity of protein tyrosine phosphatases. Curr Opin Cell Biol 2001; 13:182-95.