Toxicological Sciences 58, 306-314 (2000)
Copyright © 2000 by the Society of Toxicology
In Vitro Toxicology |
Corneal Organ Culture Model for Assessing Epithelial Responses to Surfactants
The Schepens Eye Research Institute and Department of Ophthalmology, Harvard Medical School, 20 Staniford Street, Boston, Massachusetts 02114
Received June 8, 2000; accepted August 21, 2000
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The main goal of the present study was to investigate the response of cultured bovine corneas to the application of irritant substances and its potential use for predicting ocular irritancy in humans. We hypothesized that chemicals causing eye irritation may induce disruption of epithelial tight junctions and trigger cell stress responses modulated via transcription factors such as AP-1 and NF-
B. A simple air-lifted corneal organ culture system was used as an ex vivo model for ocular irritancy test. The effects of two surfactants, sodium dodecyl sulfate (SDS) and benzalkonium chloride (BAK), on corneal epithelial permeability and DNA-binding activity of AP-1 and NF-B were studied in cultured bovine corneas. Both SDS and BAK induced tight junction disruption and increased permeability of corneal epithelium assessed using surface biotinylation in a concentration- and time-dependent manner. An increase in DNA-binding activity measured using electrophoretic mobility shift assay was observed when cultured corneas were treated with surfactants at concentrations causing minimal to mild ocular irritation, indicating epithelial cell stress response. Furthermore, exposure of cultured corneas to SDS or BAK at concentrations causing severe ocular irritancy resulted in a decrease in DNA-binding activity of these transcription factors in epithelial cells. These results indicate that the combination of corneal organ culture and measurements of corneal epithelial permeability and DNA-binding activity of stress-response transcription factors following chemical exposure has the potential to be used as a mechanistically based alternative to in vivo animal testing.
Key Words: ex vivo toxicity test; corneal organ culture; Draize test; epithelial permeability; activator protein 1; nuclear transcription factor-B; cell-surface biotinylation; electrophoretic mobility shift assay.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Commercial products, whether drugs, cosmetics, home and automobile care products, or bulk chemicals, are subject to a variety of tests, usually performed on animals, to determine any potential adverse health effects. Among the routinely performed tests is one designed to evaluate the test material's potential to cause eye irritation or injury: the Draize test (Draize et al., 1944
). The Draize test and its numerous modifications have remained the accepted method for evaluating ocular toxicity for more than five decades (Symposium and Proceeding, 1996). It has, however, been subject to much criticism for its inhumane treatment to animals as well as the irreproducibility of the subjective scoring procedure. As such, there is a great demand for minimizing the use of animals in chemical toxicity tests and for developing a mechanistically based in vitro testing system (Symposium and Proceeding, 1996).
Eye irritation is a local and reversible response of normal living corneal and conjunctival cells to direct injury caused by contact with an irritant. Approximately 75% of the Draize test scores are derived from corneal effects (Symposium and Proceeding, 1996). The cornea consists of three cell types: stratified epithelium covering the anterior surface, keratocytes interspersed in the stroma, and an inner layer of endothelium. The corneal epithelium functions as a barrier that separates the internal ocular tissues from the external environment and is therefore vulnerable to chemical insult (Gipson and Sugrue, 1994). Thus, measurement of the epithelial response to toxicants would provide a reliable means for predicting irritation potential. As such, in vitro models using corneal epithelial cells are being developed with the goal of minimizing the use of animals (Symposium and Proceeding, 1996). Some of these models use cultured cells to monitor chemical-induced changes in cellular metabolism, viability, or membrane integrity, the parameters for assessing cytotoxicity (Braunstein et al., 1999; Kruszewski et al., 1997; Saarinen-Savolainen et al., 1998; Yang and Acosta, 1994). However, a comprehensive study revealed that the cytotoxicity end points did not correlate well with the in vivo data (Sina et al., 1995). Possible reasons for this may include the end points chosen and the different stress responses of cultured and in vivo cells (Symposium and Proceeding, 1996). A three-dimensional in vitro model of cultured human corneal epithelial cells (Kruszewski et al., 1997; Ward et al., 1997) and, more recently, human corneal equivalents constructed from cell lines (Griffith et al., 1999) have been developed, and their potential use in eye irritation tests suggested (Ferber, 1999). However, at present, it may not be practical to generate enough three-dimensional culture or human corneal equivalents for routine safety testing. Furthermore, there are some inherent differences between the corneas in vivo and these in vitro models. Such differences include the presence of an overlying mucin layer and the underlying apical layer of epithelium of the cornea, which form a highly impermeable line of defense against biological and chemical insults.
Corneal organ culture is an appropriate model for in vitro ocular safety testing. The use of ocular tissues is recognized as a valuable alternative to the Draize test because it closely resembles the in vivo animal test as compared to cell culture and other approaches (Sina et al., 1995). Bovine eyes, by-products of the meat industry, are most frequently used. Measurements of opacity and thickness of bovine corneas exposed to toxicants have been used to accurately screen out severe irritants and assign relative potencies to these irritants (Gautheron et al., 1992; Sina et al., 1995). Recently, a bovine cornea was successfully cultured for up to 3 weeks without significant stromal edema or keratocyte deterioration and with little loss of epithelial architecture (Foreman et al., 1996). In this system, the corneas are cultured in an air-lifted format, where endothelia are exposed to medium containing agar-collagen gel and epithelia that are exposed to the air but remain moist. Corneal organ culture, like cell culture, allows easy manipulation under well-defined conditions and is much more physiologically relevant than cell culture.
Apical cells of corneal epithelium are joined to one another by tight junction (TJ) complexes that form a primary barrier to toxicant penetration (Gipson and Sugrue, 1994; Goodenough, 1999; Sugrue and Zieske, 1997). Disruption of TJs is likely to be a relevant end point for toxicological evaluations of eye irritation potential. Many methods have been developed to evaluate the functional integrity of epithelial TJs, including measurements of transepithelial electrical resistance (Maurice and Singh, 1986). Transepithelial electrical resistance is indicative of TJ formation between epithelial cells and has been used to assess TJ integrity in cultured epithelial cells (Cereijido et al., 1993). Recently, a surface biotinylation method has been developed for assessing epithelial barrier integrity (Chen et al., 1997). This method allows visualization of the epithelial barrier and penetration of small molecules into cell layers, indicative of TJ disruption (Chen et al., 1997; Saitou et al., 1998). Therefore, it can be used to assess tissue level alteration of TJ-based barrier function in cultured corneas in response to toxicant challenge.
Exposure of epithelial cells to irritants would undoubtedly trigger cellular stress response via, at least in part, changes in gene expression associated with responses appropriate for the initiation of healing (Camhi et al., 1995). The coordinated cellular responses to stress are known to be controlled by a group of proteins collectively called transcription factors (Camhi et al., 1995). AP-1 (Karin et al., 1997) and NF-B (Janssen et al., 1995; Pinkus et al., 1996; Siebenlist et al., 1994) are two well-characterized stress-response transcription factors. They mediate the expression of many stress-response genes in a variety of cells and tissues. The transcription factor AP-1, formed by either jun-fos (members of the immediate early-response gene family) heterodimers or jun-jun homodimers, is characterized by its ability to alter gene expression upon its activation (Karin et al., 1997; Morgan and Curran, 1995). Alteration of AP-1 activity in response to chemical exposure has been reported in a variety of cells and considered indicative of cell stress (Guyton et al., 1996; Pinkus et al., 1996; Wesselborg et al., 1997). NF-B is a ubiquitous transactivator. Cells treated with a variety of chemical agents elicit NF-B–binding activity within minutes (Janssen et al., 1995; Pinkus et al., 1996). Thus, if toxicity manifested at tissue levels is preceded by altered expression of related genes, detection of the altered expression and/or activation of stress-response transcription factors like AP-1 and NF-B can serve as an early marker for subsequent deteriorative outcomes (Ramesh et al., 1999).
As a step toward the development of a biologically relevant in vitro ocular chemical toxicity test system, we adopted a corneal organ culture originally established for long-term in vitro study of corneal wound healing (Foreman et al., 1996). In this study, we treated cultured bovine corneas with two surfactants, sodium dodecyl sulfate (SDS) and benzalkonium chloride (BAK), and evaluated the effects of these toxicants on the corneal epithelia by assessing chemical-induced leakage of TJ and alteration of AP-1 and NF-B activity. We found that exposure of bovine corneas to surfactants that induce severe eye irritation resulted in breakdown of the epithelial barrier. Furthermore, AP-1 and NF-B DNA-binding activities of epithelial extracts were altered in response to surfactants in a chemical-, dose-, and time-dependent manner. These data support our belief that measurement of TJ disruption and transactivation of stress transcription factors would be applicable to the development of alternative ocular irritation tests using ex vivo corneas.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bovine eyes were obtained from a local abattoir, transported to the laboratory on ice in a moisture chamber, and processed for culture within 6 h of animal death. Minimum essential medium (MEM) and nonessential amino acid solution were purchased from GIBCO (Grand Island, NY). Agarose and ultra pure SDS were obtained from ICN Biomedicals (Aurora, OH). Rat tail tendon collagen sulfate was from Collaborative Biomedical (Medford, MA). Sulfo-NHS-LC-biotin was from Pierce Chemicals (Rockford, IL). Rhodamine avidin D was from Vector Laboratories (Burlingame, CA). Double-stranded oligonucleotides having AP-1 and NF-
B consensus binding sequences were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) or Promega (Madison, WI). BAK and other chemicals were purchased from Sigma (St. Louis, MO).
Organ culture.
The organ culture technique was adopted with modifications (Foreman et al., 1996). Corneal-scleral rims, with approximately 4 mm of the limbal conjunctiva present, were excised and rinsed in sterilized phosphate-buffered saline (PBS). The excised corneas were placed epithelial-side down into a sterile cup containing MEM. The endothelial corneal concavity was then filled with MEM containing 1% agarose and 1 mg/ml rat tail tendon collagen maintained at 42°C. This mixture was allowed to gel. The cornea, along with its supporting gel, was inverted and then transferred to a 35-mm dish. The culture medium (about 2 ml) was added dropwise to the surface of central cornea until the limbal conjunctiva was covered, leaving the epithelium exposed to the air (Fig. 1).
|
Treatment of cultures.
SDS and BAK solutions were prepared in MEM. After removal of culture medium, the surfactant-containing MEM (2 ml) was added dropwise onto the center surface of the cornea. At the end of incubation, the treated corneas were washed 3–5 times by dipping into PBS solution and then used for either cell-surface biotin labeling or epithelial nuclear extraction.
Surface biotinylation-tight junction permeability assay.
TJ permeability assay using surface biotinylation technique was performed as described by Chen et al. (1997) with modifications. Cultured corneas, chemical treated, or untreated control, were wetted with freshly made 1 mg/ml sulfo-NHS-LC-biotin in Hank's balanced salt solution containing 1 mM CaCl2 and 2 mM MgCl2. After 30 min incubation, the surface-labeled corneas were rinsed with PBS, embedded in OCT, frozen in liquid nitrogen, and sectioned on a cryostat at a thickness of 6 µm. The sections were fixed with ice-cold acetone and labeled with rhodamine-avidin D in PBS containing 1% BSA for 1 h. The slides were mounted and examined under a Nikon Eclipse E-800 fluorescence microscope equipped with a SPOT digital camera.
Electrophoretic mobility shift assay (EMSA).
Bovine corneal epithelial whole-cell extracts for EMSA were prepared using a protocol described by Abdulkadir et al. (1995). In brief, the corneas were treated with surfactants, washed, and further cultured for 15 min in normal medium at room temperature. The epithelial cells were then removed from cultured bovine corneas under a dissection microscope with a scalpel blade. Cells from each cornea were resuspended in 100 µl of 20 mM HEPES-KOH (pH 7.9), 1 mM DTT, 1 mM EDTA, 200 mM KCl, 20% glycerol, 0.1 mM PMSF, 0.1% NP40, and incubated with periodic agitation at 4°C for 1 h. The cells and large debris were pelleted by 10-min 14,000 rpm centrifugation in a microfuge at 4°C; the supernatants were removed, aliquoted, quick-frozen in liquid nitrogen, and stored at –80°C. Protein concentration was determined using Pierce Micro BCA Protein Assay Reagent kit (Pierce Chemical Co). This epithelial extract contains proteins solubilized from both the nucleus and the cytosol in the extract media and is suitable for EMSA to assess AP-1 and NF-B DNA-binding activities (Abdulkadir et al., 1995). AP-1 and NF-B activation were analyzed by EMSA. The sequences of the probes are listed in Table 1.
|
The oligonucleotides were labeled using T4 polynucleotide kinase and [
-32P] ATP. The labeled probes were purified from free nucleotides by ethanol precipitation (Sambrook et al., 1989). For detection of AP-1 and NF-B DNA-binding activities, a 10-µl volume containing 5 µg of extracted proteins was mixed with 2 µl of 5x gel shift binding buffer (20% glycerol, 5 mM MgCl2, 2.5 mM EDTA, 2.5 mM DTT, 250 mM NaCl, 50 mM Tris–HCl, pH 7.5, 0.25 mg/ml poly(dI-dC). This reaction incubated at room temperature for 5–10 min. To each reaction, 1 µl (50,000–200,000 cpm.) of 32P-labeled oligonucleotides was added. After a 20-min binding reaction at room temperature, samples were loaded on a 4% nondenaturing polyacrylamide gel and run in 1x TGE (25 mM Tris, 190 mM glycine, 1 mM EDTA). The gel was fixed 15 min with 7% acetic acid and 10% methanol, dried, and autoradiographed. Specificity of transcription factor binding was verified by competition analysis with 50-fold molar excess of unlabeled oligonucleotides. The relative amount of radiolabeling, determined by PhosphorImager analysis (Bio-Rad, Rochmont, CA), was the sum of all the shifted bands in each lane. The amounts varied between experiments, in part, because of differences in exposure times and in probe-labeling efficiency. Thus, they may not be compared between experiments. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bovine Corneal Culture and Chemical Treatment
The organ culture technique was adopted from Foreman et al. (1996) and originally developed for the study of corneal wound healing. Agarose gel prepared with phenol red–free MEM and collagen was used to support corneal architecture. This setting allowed direct application of tested chemicals, dissolved in culture media, to the surface of cultured corneas. To mimic in vivo exposure of the eye to the toxicants, the chemical incubation was limited to a short time (typically 2 min), as illustrated in Figure 1
.
Evaluation of the Paracellular Leakage of Cultured Bovine Corneas Induced by Surfactant Exposure
We examined the effects of test chemicals on epithelial barrier function of cultured bovine corneas using the tracer experiment procedure developed by Chen et al. (1997). In the control cornea, the outermost surface of epithelium was continually labeled with biotin, as indicated by a fluorescent line on the surface of the cornea (Fig. 2). Treatment of cultured cornea with SDS, an anionic surfactant, for 2 min resulted in disruption of linear surface labeling, as indicated by the presence of break points in the fluorescent line. At lower concentrations (0.1–0.3%), SDS caused a few break points of corneal surface labeling; no apparent labeling of underlying epithelial cells was observed. In 1% SDS-treated cornea, in addition to visible disruption of surface labeling, the penetration of biotin into the apical epithelial cell layer started to be seen as faint labeling of cells beneath the apical surface labeling. Incubation of 3% SDS for 2 min led to further breakdown of linear corneal surface labeling, and the penetration of biotin into the stroma can be seen as faint stromal labeling. Corneas exposed to 15% SDS lost most surface labeling and exhibited stromal biotinylation, indicating penetration of biotin through epithelial cell layers and the underlying basement membrane. Morphological alteration was also observed when corneas were exposed to higher SDS concentrations (3 and 15%, inserts, Fig. 2).
|
BAK, a cationic surfactant, is the most commonly used preservative in ophthalmic solutions, and its toxicity is well known. Treatment of bovine corneas for 2 min with 0.01–0.3% BAK extended the biotinylation line (Fig. 3
). Incubation of 1% BAK further increased the size of biotin labeling to about three times that of control. Exposure of bovine corneas to 3% BAK led to a great increase in the depth of biotin penetration, as indicated by width of surface labeling. Stromal penetration in BAK-treated corneas was minimal. Morphological alteration was also observed when corneas were exposed to 0.1% or higher BAK (inserts, Fig. 3).
|
The effects of surfactant exposure time on corneal epithelial barrier function were also examined (Fig. 4
). Similar to the higher concentration of SDS, a 5-min incubation of 1% SDS led to further disturbed biotin surface labeling and stromal labeling, indicating breakdown of TJ and penetration of biotin into the epithelial layer (left panels, Fig. 4). BAK, on the other hand, did not disrupt epithelium surface labeling, while leakage of TJ was evident as more epithelial cell layers were biotinylated in prolonged exposure of 1% BAK. Furthermore, the deeper layer of epithelium in 10-min BAK-exposed cornea exhibited a non-cell staining pattern, suggesting the destructive effects of BAK on epithelial cells (right panels, Fig. 4).
|
Alteration of AP-1 and NF-
B DNA-Binding Activity in the Corneal Epithelia Following Surfactant TreatmentTo determine whether AP-1 and NF-
B DNA-binding activity was altered in response to surfactant challenge of cultured corneas, we used EMSA with extracts from epithelial cells. We first assessed AP-1 DNA-binding specificity using 50-fold molar excess of unlabeled AP-1 (consensus and mutant) or unrelated probes to compete with labeled AP-1 probe (Fig. 5; see Table 1 for probe sequences). There were multiple shifted bands observed in untreated corneal epithelial cells, probably due to different heterodimers. Addition of unlabeled AP-1 competitor greatly decreased the intensity of all three bands; AP-1 mutant also reduced AP-1 DNA binding, but to a lesser extent. Addition of unlabeled, unrelated probes (NF-B and SP-1) did not affect AP-1 DNA binding significantly. Taken together, the competitive binding study suggested that these complexes are AP-1 specific.
|
Figure 6
showed alteration of AP-1 DNA-binding activity following surfactant exposure of cultured corneas. There was no difference in AP-1 DNA-binding activity between untreated controls and cornea treated with 0.1% SDS (Fig. 6A). A 1.5-fold increase in DNA-binding activity was noted when the cultured cornea was exposed to 0.3% SDS for 2 min. Further increase in SDS concentration led to a decrease in AP-1 DNA-binding activity; 1% SDS reduced AP-1 DNA-binding activity to 70%, and 3% SDS to 29.2% of the control. A similar pattern of AP-1 activation was observed when cultured corneas were exposed to BAK (Fig. 6B). Low concentration of BAK (0.01–0.3%) increased AP-1 binding activity in cultured corneas in a dose-dependent manner; close to a 3-fold increase in AP-1 DNA-binding activity was observed when 0.3% BAK was used. BAK at a concentration causing severe ocular irritation (1 and 3%) dramatically reduced AP-1 DNA-binding activity.
|
Changes in NF-
B DNA-binding activity in response to surfactant exposure was also assessed. NF-B DNA-binding specificity was first determined using 50-fold molar excess of labeled NF-B and its mutant with a G/C substitution in the NF-B DNA-binding motif (Fig. 7). Addition of unlabeled NF-B competitor blocked NF-B–DNA interaction, as the intensity of the major shifted band was barely detectable, whereas the same amount of unlabeled NF-B mutant reduced band intensity, suggesting that the observed complex was NF-B specific.
|
Untreated control corneas exhibited a high level of NF-
B DNA-binding activity (Fig. 8). Exposure to SDS reduced NF-B DNA binding in a concentration-dependent manner, with maximal reduction reached at 3% SDS (3.5-fold decrease) (Fig. 8A). The effect of BAK on NF-B activation in the cultured cornea was also concentration dependent; 0.01% BAK had no effect on NF-B activity, 0.1–0.3% BAK elicited approximately 1.4-fold activation of NF-B over control levels (Fig. 8B). NF-B DNA-binding activity of 1% BAK-treated cornea was similar to that of control, while 3% BAK reduced NF-B DNA-binding activity to 50% of the control.
|
Table 2
shows comparison of SDS- and BAK-induced alterations in epithelial permeability and AP-1 and NF-B-DNA-binding activity with concentration-dependent maximum average score of the Draize test. These results indicated that the degrees of the alterations induced by these detergents are in good agreement with the results of in vivo testing.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The aim of this work was to develop an ex vivo model to assess ocular irritancy potential of test chemicals using a simple corneal organ culture system. Using two surfactants, SDS and BAK, we investigated alterations of corneal epithelial permeability and DNA-binding activity of stress-response transcription factors AP-1 and NF-
B caused by these surfactants at the concentrations known to cause different levels of ocular irritation. We observed that the degree of surfactant–induced TJ disruption and the extent of small molecular penetration measured by surface biotinylation is correlated to in vivo irritancy measurements as determined by the Draize test (Table 2). Using DNA-binding activity of AP-1 and NF-B as indicators of cellular stress response, we demonstrated that surfactants at concentrations causing minimal to mild ocular irritation increased DNA-binding activity, indicating epithelial stress response. The surfactants at concentrations causing severe ocular irritancy induced a decrease in DNA-binding activity, indicating destruction of epithelium. Thus, corneal organ culture, coupled with measurements of chemical-induced corneal epithelial barrier disruption and transactivation of stress-related genes, may be used as a mechanistically based alternative to in vivo animal testing.
TJs provide a continuous seal around the apical aspect of adjoining epithelial cells, thereby preventing free passage of molecules into cell layers (Goodenough, 1999). We demonstrated the impermeable nature of the cornea using sulfo-NHS-LC-biotin that covalently cross-links to proteins. Sulfo-NHS-LC-biotin (
500 Da) should be impermeable if the TJ barrier is intact, whereas it would diffuse across paracellular spaces into cell layers when TJs are disrupted (Chen et al., 1997). The demonstration that sulfo-NHS-LC-biotin labels only the apical surface of normal cornea suggests that under organ culture condition the cornea possesses a fully functional barrier. As the stratified epithelial surface of the cornea is covered by a layer of highly glycosylated mucin (Watanabe et al., 1995), biotin may be cross-linked to cell surface-associated mucins as well as to the apical surface of epithelium. Thus, both apical epithelial TJs and overlying mucin may provide the cornea with defense against chemical insults.
Chemicals causing eye irritation are likely to affect corneal epithelial integrity and cause breakdown of barrier function. Consistent with this hypothesis, we observed that the surface-labeling pattern is altered in a concentration-dependent manner in corneas exposed to the ionic surfactants SDS and BAK. We chose a short exposure time (2 min) to mimic in vivo accidental exposure to ocular irritants, which are usually removed quickly by tears and blinking. Our data showed that the concentrations causing apparent disruption of TJs for both detergents were much higher than those causing lysis of epithelial monolayer (Grant et al., 1992) or three-dimensional (Kruszewski et al., 1997; Ward et al., 1997) cultures. Both the mucin layer and the TJ-bearing apical layer, which is relatively inactive in metabolism, may contribute to the high resistance of the cornea to these ionic detergents. These data indicate that the organ culture model is less likely to overpredict toxicity of test chemicals than is cell culture model. Furthermore, surfactant-induced alteration of surface labeling was consistent with irritancy potential of these chemicals as determined by the in vivo Draize test (Table 2). However, at the present time, the changes in biotin labeling cannot be quantitated. Interestingly, the effects of these two detergents on corneal surface biotinylation were different. SDS primarily caused disruption of surface labeling and the intensity of disruption was related to the concentration-dependent irritancy. At a concentration lower than what is known to be mildly irritating, SDS caused only a few break points of surface labeling; at a concentration known to cause severe irritancy, most surface labeling was lost and the epithelium and stroma were also labeled. Incubation of BAK, however, led to extended penetration of the biotin molecules into the epithelium and the depth of epithelial labeling was BAK concentration-dependent. Thus, measuring the depth of biotinylation can assess the ocular toxicity of BAK. Taken together, our results suggest that surface biotinylation of cultured cornea is an effective way to evaluate TJ permeability and, after a large number of chemicals with known ocular toxicity are tested, to grade the ocular toxicity of surfactant-based consumer products and other chemicals.
Disruption of TJs would allow chemicals to diffuse into deeper, innervated epithelial layers, causing irritation and epithelial-cell stress responses. Depending on the nature of the chemical insult, cells in organ culture, like those in vivo, respond to chemicals via changes in gene expression that are associated with the initiation of healing (Camhi et al., 1995). It is generally believed that most toxicologically relevant outcomes require not only differential gene expression, but also differential expression of multiple genes (Farr and Dunn, 1999). Transcriptional regulation is a primary strategy to control differential expression of functionally related genes by organisms in response to either physiological or environmental stress signals (Scheidereit, 1996). Activated transcription factors coordinately modulate target gene expression via binding to the specific DNA sequences located in the promoter or enhancer. Modulation of transcription factor activities by chemical stress is the underlying transcriptional mechanism responsible for differential expression of 2% of the human genome involved in the stress response (Camhi et al., 1995; Farr and Dunn, 1999). AP-1 (Karin et al., 1997) and NF-B (Janssen et al., 1995; Pinkus et al., 1996) are well-characterized stress-response transcription factors known to transactivate a large number of stress response genes and therefore fit the profile of indicators of stress-induced gene expression. Although the mechanisms for regulating AP-1 and NF-B activity in cells are very complex, we observed changes in AP-1 and NF-B activity in the cultured cornea in response to surfactant challenges; the concentration-dependent response of epithelial cells to these surfactants correlated to toxicity, as determined by the Draize test (Table 2). The molecular identity of proteins interacting with AP-1 and NF-B probes has not been assessed, as the available antibodies for supershifting to determine subunits forming AP-1 or NF-B may not cross-react with the proteins of bovine origin. There are two phases of epithelial response in surfactant challenge. The concentrations that are known to have minimal effects on surface biotinylation caused an increase in AP-1 and NF-B activities, consistent with the hypothesis that chemical challenges induce stress-response transcription-factor activation/expression. The surfactant concentration known to cause severe ocular irritation markedly reduced AP-1 and NF-B activity. We suggest that this reduction is due to decreased viability that affects transcription factor synthesis (AP-1) or activation (NF-B) either by surfactant influence directly and/or secondary to decreased number of viable cells. Nevertheless, our results demonstrate a causal relationship between alteration of AP-1 and NF-B activities and ocular toxicity, suggesting that AP-1 and NF-B binding activity may serve as specific and quantifiable toxicological end points for ocular irritation potential of test chemicals.
In order to estimate the predictive value of the data obtained with corneal organ culture, we evaluated different concentrations of two detergents known to cause a range of effects, from no response to severe irritation, in a concentration-dependent manner. These analyses demonstrate that corneal organ culture presents a suitable model for evaluating ocular irritation potential of the innocuous, mild, moderate, and severe irritants, and establish the feasibility of using this ex vivo system as an alternative to the in vivo Draize test. Application of the test substances to cultured bovine corneas for a short time (5 min or less) was sufficient to predict the degree of ocular irritation elicited by the substances, apparently without overpredicting or detecting false negatives. Using this approach, we have also evaluated three hair-care products in a double-blind manner and correctly distinguished one mildly irritating product (Draize score 14.2) from those causing moderate irritation (Draize scores 37.8 and 57.4), indicating the value of the system (Yu, et al. unpublished results). However, a precise fine-tuning of the scoring system is needed in order to quantitatively assess the potency of the ocular irritant response elicited by the test products. This will only be possible after more extensive intralaboratory and interlaboratory evaluation of the experimental model, aimed to optimize its performance.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by grants from the NIH/NEI (EY10869) and from the Johns Hopkins Center for Alternatives to Animal Testing/Gillette Program Project. XFL is a recipient of Alcon/ARVO 1998–1999 Postdoctoral Fellowship Award. The authors thank Dr. Sherry L. Ward, Gillette Med Evaluation Labs, for critical reading and comments on the manuscript and Dr. Hong-qing Ye for the artwork in designing Figure 1
. |
NOTES |
|---|
1 To whom correspondence should be addressed. Fax: (617) 912-0101. E-mail: fushinyu{at}vision.eri.harvard.edu.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abdulkadir, S. A., Krishna, S., Thanos, D., Maniatis, T., Strominger, J. L., and Ono, S. J. (1995). Functional roles of the transcription factor Oct-2A and the high mobility group protein I/Y in HLA-DRA gene expression. J. Exp. Med. 182, 487–500.
Braunstein, S. G., Deramaudt, T. G., Rosenblum, D. G., Dunn, M. G., and Abraham, N. G. (1999). Heme oxygenase-1 gene expression as a stress index to ocular irritation. Curr. Eye. Res. 19, 115–122.[Web of Science][Medline]
Camhi, S. L., Lee, P., and Choi, A. M. (1995). The oxidative stress response. New Horiz. 3, 170–182.[Medline]
Cereijido, M., Gonzalez-Mariscal, L., Contreras, R. G., Gallardo, J. M., Garcia-Villegas, R., and Valdes, J. (1993). The making of a tight junction. J. Cell Sci. 17(suppl.), 127–132.
Chen, Y., Merzdorf, C., Paul, D. L., and Goodenough, D. A. (1997). COOH terminus of occludin is required for tight junction barrier function in early Xenopus embryos. J. Cell Biol. 138, 891–899.
Draize, J. H., Woodard, G., and Ho, C. (1944). Method for the study of irritation and toxicity of substances applied topically to the skin and mucous membrane. J. Pharmacol. Exp. Ther. 82, 377–389.
ECETOC (1992). Eye irritation: Reference chemicals data Bank, ECETOC technical report no. 48. European Centre for Ecotoxicology and Toxicology of Chemicals (Brussels, Belgium).
Farr, S., and Dunn, R. T., II (1999). Concise review: Gene expression applied to toxicology. Toxicol. Sci. 50, 1–9.
Ferber, D. (1999). Tissue engineering. Growing human corneas in the lab. Science 286, 2051, 2053.
Foreman, D. M., Pancholi, S., Jarvis-Evans, J., McLeod, D., and Boulton, M. E. (1996). A simple organ culture model for assessing the effects of growth factors on corneal re-epithelialization. Exp. Eye Res. 62, 555–564.[Web of Science][Medline]
Gautheron, P., Dukic, M., Alix, D., and Sina, J. (1992). Bovine corneal opacity and permeability test: An in vitro assay of ocular irritancy. Fundam. Appl. Toxicol. 18, 442–449.[Web of Science][Medline]
Gipson, I., and Sugrue, S. (1994). Cell biology of the corneal epithelium. In Principles and Practice of Ophthalmology, (D. Albert and F. Jakobiec, eds.), pp. 3–16, W. B. Saunders, Philadelphia.
Goodenough, D. A. (1999). Plugging the leaks. Proc. Natl. Acad. Sci. U. S. A. 96, 319–321.
Grant, R. L., Yao, C., Gabaldon, D., and Acosta, D. (1992). Evaluation of surfactant cytotoxicity potential by primary cultures of ocular tissues: I. Characterization of rabbit corneal epithelial cells and initial injury and delayed toxicity studies. Toxicology 76, 153–176.[Web of Science][Medline]
Griffith, M., Osborne, R., Munger, R., Xiong, X., Doillon, C. J., Laycock, N. L., Hakim, M., Song, Y., and Watsky, M. A. (1999). Functional human corneal equivalents constructed from cell lines [see comments]. Science 286, 2169–2172.
Guyton, K. Z., Xu, Q., and Holbrook, N. J. (1996). Induction of the mammalian stress response gene GADD153 by oxidative stress: Role of AP-1 element. Biochem. J. 314, 547–554
Janssen, Y. M., Barchowsky, A., Treadwell, M., Driscoll, K. E., and Mossman, B. T. (1995). Asbestos induces nuclear factor kappa B (NF-kappa B) DNA-binding activity and NF-kappa B-dependent gene expression in tracheal epithelial cells. Proc. Natl. Acad. Sci. U. S. A. 92, 8458–8462.
Karin, M., Liu, Z., and Zandi, E. (1997). AP-1 function and regulation. Curr. Opin. Cell Biol. 9, 240–246. Review.[Web of Science][Medline]
Kruszewski, F. H., Walker, T. L., and DiPasquale, L. C. (1997). Evaluation of a human corneal epithelial cell line as an in vitro model for assessing ocular irritation. Fundam. Appl. Toxicol. 36, 130–140.[Web of Science][Medline]
Maurice, D., and Singh, T. (1986). A permeability test for acute corneal toxicity. Toxicol. Lett. 31, 125–130.[Web of Science][Medline]
Morgan, J. I., and Curran, T. (1995). Immediate-early genes: Ten years on. Trends Neurosci. 18, 66–67. Review.[Web of Science][Medline]
Pinkus, R., Weiner, L. M., and Daniel, V. (1996). Role of oxidants and antioxidants in the induction of AP-1, NF-kappaB, and glutathione S-transferase gene expression. J. Biol. Chem. 271, 13422–13429.
Ramesh, G. T., Manna, S. K., Aggarwal, B. B., and Jadhav, A. L. (1999). Lead activates nuclear transcription factor-kappaB, activator protein- 1, and amino-terminal c-Jun kinase in pheochromocytoma cells. Toxicol. Appl. Pharmacol. 155, 280–286.[Web of Science][Medline]
Saarinen-Savolainen, P., Jarvinen, T., Araki-Sasaki, K., Watanabe, H., and Urtti, A. (1998). Evaluation of cytotoxicity of various ophthalmic drugs, eye drop excipients and cyclodextrins in an immortalized human corneal epithelial cell line. Pharm. Res. 15, 1275–1280.[Web of Science][Medline]
Saitou, M., Fujimoto, K., Doi, Y., Itoh, M., Fujimoto, T., Furuse, M., Takano, H., Noda, T., and Tsukita, S. (1998). Occludin-deficient embryonic stem cells can differentiate into polarized epithelial cells bearing tight junctions. J. Cell Biol. 141, 397–408.
Sambrook, J., Fritsch, E., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Plainview, New York.
Scheidereit, C. (1996). Transcription factors: Important tools and targets for molecular medicine. J. Mol. Med. 74, 707–709.[Web of Science][Medline]
Siebenlist, U., Franzoso, G., and Brown, K. (1994). Structure, regulation, and function of NF-B. Annu. Rev. Cell Biol. 10, 405–455.[Web of Science]
Sina, J. F., Galer, D. M., Sussman, R. G., Gautheron, P. D., Sargent, E. V., Leong, B., Shah, P. V., Curren, R. D., and Miller, K. (1995). A collaborative evaluation of seven alternatives to the Draize eye irritation test using pharmaceutical intermediates. Fundam. Appl. Toxicol. 26, 20–31.[Web of Science][Medline]
Sugrue, S., and Zieske, J. (1997). ZO1 in corneal epithelium: Association to the zonula occludens and adherens junctions. Exp. Eye Res. 64, 11–20.[Web of Science][Medline]
Technical Committee on Alternatives to Animal Testing U.S.A. (1996). Replacing the Draize eye irritation test: Scientific background and research needs. J. Toxicol. Cutan. Ocul. Toxicol. 15, 211–234.
Ward, S. L., Walker, T. L., and Dimitrijevich, S. D. (1997). Evaluation of chemically induced toxicity using in vitro model of human corneal epithelium. Toxicol. in Vitro 11, 121–139.
Watanabe, H., Fabricant, M., Tisdale, A. S., Spurr-Michaud, S. J., Lindberg, K., and Gipson, I. K. (1995). Human corneal and conjunctival epithelia produce a mucin-like glycoprotein for the apical surface. Invest. Ophthalmol. Vis. Sci. 36, 337–344.
Wesselborg, S., Bauer, M. K. A., Vogt, M., Schmitz, M. L., and Schulze-Osthoff, K. (1997). Activation of transcription factor NF-kappaB and p38 mitogen-activated protein kinase is mediated by distinct and separate stress effector pathways. J. Biol. Chem. 272, 12422–12429.
Yang, W., and Acosta, D. (1994). Cytotoxicity potential of surfactant mixtures evaluated by primary cultures of rabbit corneal epithelial cells. Toxicol. Lett. 70, 309–318.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Yanai, J.-A. Ko, N. Morishige, T.-i. Chikama, H. Ichijima, and T. Nishida Disruption of Zonula Occludens-1 Localization in the Rabbit Corneal Epithelium by Contact Lens-Induced Hypoxia Invest. Ophthalmol. Vis. Sci., October 1, 2009; 50(10): 4605 - 4610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Yin, J. Lu, and F.-S. X. Yu Role of Small GTPase Rho in Regulating Corneal Epithelial Wound Healing Invest. Ophthalmol. Vis. Sci., March 1, 2008; 49(3): 900 - 909. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Guo, M. Satpathy, G. Wilson, and S. P. Srinivas Benzalkonium Chloride Induces Dephosphorylation of Myosin Light Chain in Cultured Corneal Epithelial Cells Invest. Ophthalmol. Vis. Sci., May 1, 2007; 48(5): 2001 - 2008. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-P. Xu, J. Yin, and F.-S. X. Yu Lysophosphatidic Acid Promoting Corneal Epithelial Wound Healing by Transactivation of Epidermal Growth Factor Receptor Invest. Ophthalmol. Vis. Sci., February 1, 2007; 48(2): 636 - 643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Okabe, H. Kimura, J. Okabe, A. Kato, H. Shimizu, T. Ueda, S. Shimada, and Y. Ogura Effect of Benzalkonium Chloride on Transscleral Drug Delivery Invest. Ophthalmol. Vis. Sci., February 1, 2005; 46(2): 703 - 708. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Suuronen, C. R. McLaughlin, P. K. Stys, M. Nakamura, R. Munger, and M. Griffith Functional Innervation in Tissue Engineered Models for In Vitro Study and Testing Purposes Toxicol. Sci., December 1, 2004; 82(2): 525 - 533. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Kenney, S. R. Atilano, N. Zorapapel, B. Holguin, R. N. Gaster, and A. V. Ljubimov Altered Expression of Aquaporins in Bullous Keratopathy and Fuchs' Dystrophy Corneas J. Histochem. Cytochem., October 1, 2004; 52(10): 1341 - 1350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-P. Xu, Y. Ding, J. Ling, Z. Dong, and F.-S. X. Yu Wound-Induced HB-EGF Ectodomain Shedding and EGFR Activation in Corneal Epithelial Cells Invest. Ophthalmol. Vis. Sci., March 1, 2004; 45(3): 813 - 820. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Perrot, H. Dutertre-Catella, C. Martin, P. Rat, and J.-M. Warnet Resazurin Metabolism Assay Is a New Sensitive Alternative Test in Isolated Pig Cornea Toxicol. Sci., March 1, 2003; 72(1): 122 - 129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Stepp, H. E. Gibson, P. H. Gala, D. D. Sta. Iglesia, A. Pajoohesh-Ganji, S. Pal-Ghosh, M. Brown, C. Aquino, A. M. Schwartz, O. Goldberger, et al. Defects in keratinocyte activation during wound healing in the syndecan-1-deficient mouse J. Cell Sci., January 12, 2002; 115(23): 4517 - 4531. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||