Predictive performance of the Short Time Exposure test for identifying eye irritation potential of chemical mixtures

Toxicology in Vitro 29 (2015) 617–620

Contents lists available at ScienceDirect

Toxicology in Vitro journal homepage: www.elsevier.com/locate/toxinvit

Predictive performance of the Short Time Exposure test for identifying eye irritation potential of chemical mixtures Kazutoshi Saito ⇑, Masaaki Miyazawa, Yuko Nukada, Kyo Ei, Takayuki Abo, Hitoshi Sakaguchi Safety Science Research Laboratories, Kao Corporation, Tochigi, Japan

a r t i c l e

i n f o

Article history: Received 26 August 2014 Accepted 17 January 2015 Available online 11 February 2015 Keywords: STE Eye irritation Alternative method Chemical mixture SIRC cell line

a b s t r a c t The Short Time Exposure (STE) test is an in vitro eye irritation test based on the cytotoxicity in SIRC cells (rabbit corneal cell line) following a 5 min treatment of chemicals. This study evaluated the predictive performance of the STE test to identify the globally harmonized system (GHS) Not Classified category and other irritant categories (i.e., GHS Category 1 or 2) when used to test 40 chemical mixtures that included irritants. The STE test correctly identified 30 tested mixtures classified as GHS irritant categories and 5 out of 10 tested mixtures classified as GHS Not Classified. The sensitivity, specificity, positive predictivity, negative predictivity, and overall accuracy of the STE test were 100% (30/30), 50% (5/10), 86% (25/30), 100% (5/5), and 88% (35/40), respectively. These predictive performances were comparative to or greater than those in other in vitro eye irritation tests that have been accepted as test guideline by the Organisation for Economic Co-operation and Development. This suggests that the STE test has sufficient predictivity for identifying the eye irritation potential of chemical mixtures. Since no false negatives in this study were found, this indicates that the STE test is applicable as a part of the bottom-up approach. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction The rabbit Draize test has been used for many years to evaluate the eye irritation potential of chemicals (Draize et al., 1944). However, animal welfare concerns and EU regulatory policies prohibiting the testing of cosmetic ingredients in animals for a number of toxicological endpoints (Directive 2003/15/EC, 2003) have led to the development of a greater number of alternative eye irritation methods that use various cell lines and tissues (Balls et al., 1999; Ohno et al., 1999; Eskes et al., 2005). Among them, the Bovine Corneal Opacity and Permeability (BCOP) assay, Isolated Chicken Eye (ICE) test, and a fluorescein leakage (FL) test method have recently been adopted as the Organisation for Economic Co-operation and Development (OECD) test guidelines (TGs) for predicting eye irritation (OECD, 2009a,b, 2012). The Short Time Exposure (STE) test was submitted to the OECD in 2011 as a new alternative method and the draft TGs are currently under review. The STE test is a cytotoxicity-based alternative test that uses SIRC cells (rabbit cornea cell line) to identify chemicals that induce eye irritation (Irritant, ‘‘I’’) and chemicals ⇑ Corresponding author at: Safety Science Research Laboratories, Kao Corporation, 2606 Akabane, Ichikai-machi, Haga-gun, Tochigi 321-3497, Japan. Tel.: +81 285 68 7474; fax: +81 285 68 7452. E-mail address: [email protected] (K. Saito). http://dx.doi.org/10.1016/j.tiv.2015.01.008 0887-2333/Ó 2015 Elsevier Ltd. All rights reserved.

that do not inducing eye irritation (Not irritant, ‘‘NI’’) as well as to classify minimal, moderate, or severe eye irritation potency (Takahashi et al., 2008). Since the STE test uses cell viability as an endpoint after 5 min of chemical exposure, the procedure is simple and quick, and the evaluation cost is low. The STE test also has the advantage of being able to evaluate the eye irritation potential of water insoluble chemicals by using mineral oil as the test vehicle. Moreover, when the STE test was used to test 44 chemicals in order to predict the globally harmonized system (GHS) irritation categories, the results demonstrated that this test has a high accuracy (>90%) between the STE irritation categories (‘‘NI’’ and ‘‘I’’) and the two-rank GSH classification (‘‘Not Classified’’ and ‘‘Category 1 or 2’’) (Takahashi et al., 2009). To date, the STE test has been shown to have a good predictive performance when evaluating single substances. However, it is also important to assess the eye irritation potential of chemical mixtures (e.g., surfactant based mixtures) when conducting safety assessments, as new combinations of ingredients could potentiate the toxicity of one or more of the ingredients in a mixture (Bruner et al., 1998). If there is a significantly increased toxicity of one or more ingredients when delivered in a particular mixture, this raises the possibility that the new mixture may be poorly tolerated, or in the worst case, cause injury (Baker and Bruner, 1997). In the current study, we evaluated the utility of the STE test within the context of the GHS classification and assessed its ability to predict

618

K. Saito et al. / Toxicology in Vitro 29 (2015) 617–620

the eye irritation potential of 40 chemical mixtures, which included alcohol-based or surfactant-based mixtures. 2. Materials and methods 2.1. Materials The study analyzed 40 chemical mixtures, which included alcohol-based or surfactant-based mixtures. The GHS classifications of the 40 chemical mixtures were defined as ‘‘NI’’ (Not Classified, or not an irritant) or ‘‘I’’ (Irritant classified as Category 1 or 2) based on the historical databases. 2.2. STE test 2.2.1. Cell culture SIRC cells (CCL-60) were obtained from American Type Culture Collection (Manassas, VA, USA). SIRC cells were cultured in Eagle’s minimum essential medium (Sigma–Aldrich Co., St. Louis, MO, USA) containing 10% (v/v) fetal bovine serum, 2 mM L-glutamine, 50 units/ml penicillin, and 50 lg/ml streptomycin (Invitrogen Co., Carlsbad, CA, USA). Once the cells proliferated in the culture flask to confluence, they were dispersed with trypsin-ethylenediaminetetraacetic acid (EDTA) solution (Sigma–Aldrich Co.). The dispersed cells were spread into 96-well flat-bottomed plates (Corning Coster Co., Cambridge, MA, USA) at 3.0  103 cells/well. After incubation (37 °C, 5% CO2) for 5 days (or 6.0  103 cells/well for 4 days), the cells reached confluence. 2.2.2. STE test protocol The STE test was carried out using the procedure of Takahashi et al. (2008). Physiological saline (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) was used as the vehicle for all tested mixtures and as a control substance. The cells cultured in the 96-well plates were exposed to 200 ll of 5% solutions of test mixtures for 5 min. With the STE protocol, exposure to 5% and 0.05% solutions of the test mixtures is used to predict the three-rank GHS classifications (Category 1, 2, or Not Classified) of the mixtures, whereas the 5% exposure is sufficient for the ‘‘I’’ or ‘‘NI’’ predictions (Takahashi et al., 2008). Since the purpose of this study was to assess the performance of the STE test for predicting the ‘‘I’’ or ‘‘NI’’ mixtures, the test mixtures were only tested at 5%. After exposure, the cells were washed twice with Dulbecco’s phosphate buffered saline (without magnesium and calcium) [DPBS (-); Takara Bio Inc., Shiga, Japan], with 200 ll of methylthiazolydiphenyltetrazolium bromide (MTT, Sigma–Aldrich Co.) solution (0.5 mg MTT/ml of medium) then added. After a 2 h reaction time, MTT formazan was extracted with 0.04 N HCl–isopropanol (Kanto Chemical Co., Inc., Tokyo, Japan) over a 30 min period and the absorbance of the extract was measured at 570 nm with a plate reader (Corona Electric Co., Ltd., Ibaraki, Japan or BMG LABTECH GmbH, Offenburg, Germany). The ratio of absorbance (%) for cells treated with test mixtures to that of the control was represented as relative viability (triplicate determinations). The mean of three wells was calculated for one independent test. Three independent tests were conducted for each test mixture and the overall calculated mean of the three independent tests was used for estimation of the eye irritation. 2.2.3. STE test category classification of eye irritation The STE test (‘‘I’’ or ‘‘NI’’) determined the category classification of the eye irritation based on the relative cell viability. A tested mixture that had a relative viability of 70% or less at a 5% concentration was categorized as ‘‘I’’ while a tested mixture that had a relative viability greater than 70% at the same concentration was categorized as ‘‘NI’’ (Takahashi et al., 2008).

3. Results Table 1 shows the summary of the results for the 40 chemical mixtures. The STE test showed that 35 mixtures exhibited less than 70% cell viability while five (2 alcohol based mixtures and 3 cationic surfactant based mixtures) exhibited more than 70% cell viability. Five out of 10 tested mixtures that were classified as GHS Not Classified showed more than 70% cell viability and were classified as ‘‘NI’’ by the STE test. The STE test determined that the five other mixtures (3 cationic surfactant based mixtures (B, C, and E) and 2 surfactant based mixtures (O and P)) exhibited less than 70% viability (with the viability ranging from 2.2% in surfactant based mixture O to 53.9% in cationic surfactant based mixture E). On the other hand, the STE test found that 30 of the tested mixtures that were classified as GHS irritant categories (i.e., Category 1 or 2) exhibited less than 70% cell viability (with the viability ranging from 0% to 5.1%). Table 2 summarizes the predictive performance of the STE test for the chemical mixtures. The accuracy of the STE test was judged based on whether the tested mixtures that were classified as GHS

Table 1 Summary of the STE test results for 40 chemical mixtures. Tested mixture

GHS

STE

Category

Code

Viability (%)

Agricultural chemical

A

I

Alcohol based mixture

A B

NI NI

Anionic surfactant based mixture

A B C D E F G H I J K L

I I I I I I I I I I I I

Cationic surfactant based mixture

A B C D E F

NI NI NI NI NI NI

Sodium percarbonate based mixture

A B C

I I I

1.3 0.9 2.7

I I I

Surfactant based mixture

A B C D E F G H I J K L M N O P

I I I I I I I I I I I I I I NI NI

1.9 3.8 1.6 2.1 0 3.6 1.4 1.6 1.8 2.6 1.7 2.4 0.5 1.5 2.2 5.6

I I I I I I I I I I I I I I I I

5.1 76.8 75.6 0.5 0 0 4.2 0.6 1.7 0 0 2.2 0.8 0.6 0.3 88.7 46.5 35.7 79.1 53.9 95.6

Category I NI NI I I I I I I I I I I I I NI I I NI I NI

Code of mixtures was randomly allocated. ‘‘I’’ in GHS; Irritation categories (category 1 or 2), ‘‘NI’’ in GHS; Not-classified category. ‘‘I’’ in STE; Irritants, ‘‘NI’’ in STE; Not irritants (see Section 2.2.3).

K. Saito et al. / Toxicology in Vitro 29 (2015) 617–620 Table 2 Predictive performance of the STE test. n

GHS

I NI

STE

30 10

I

NI

30 5

0 5

Predictivity (%) Sensitivity Specificity Positive predictivity Negative predictivity Accuracy

100 50 86 100 88

Table 3 Comparison of the predictive performance of the STE test with those of OECD TG methods.

Number of tested mixtures Sensitivity Specificity Positive predictivity Negative predictivity Accuracy

STE

BCOPa

ICEb

40 100% (30/30) 50% (5/10) 86% (30/35) 100% (5/5) 88% (35/40)

92 100% (52/52) 20% (8/40) 62% (52/84) 100% (8/8) 65% (60/92)

80 95% 80% 63% 98% 84%

(20/21) (47/59) (20/32) (47/48) (67/80)

Based on the data referred in (a) OECD (2013a), (b) OECD (2013b). The values in bracket mean; (the number of correctly predicted mixtures/the total number of mixtures used for the calculation of each predictive value).

irritant or GHS Not Classified categories were correctly determined to be ‘‘I’’ or ‘‘NI’’, respectively. The STE test correctly predicted 30 of the tested mixtures classified as GHS irritant categories and five out of 10 classified as GHS Not Classified. The STE test overestimated the results of five mixtures classified as GHS Not Classified and predicted they were ‘‘I’’. Overall, the sensitivity, specificity, positive predictivity, negative predictivity and accuracy of the STE test were 100% (30/30), 50% (5/10), 86% (25/30), 100% (5/5) and 88% (35/40), respectively. Finally, we examined the predictive performance of the STE test and compared it with two OECD TG accepted alternative methods (BCOP, ICE) (Table 3). The sensitivity and negative predictivity results (100%) were comparable to those of the other tests, whereas the positive predictivity and overall accuracy (86% and 88%) were comparable or higher. On the other hand, the specificity (50%) of the STE test was higher or lower than that of the BCOP (20%) and the ICE (80%), respectively. 4. Discussion Our current study assessed the predictive performance of the STE test for evaluating the eye irritation potential of chemical mixtures within the context of the GHS classification. The STE test results for 40 chemical mixtures revealed that the STE exhibited a good predictive performance (88%) for evaluating the irritation potential (‘‘I’’ or ‘‘NI’’) and reported no false negatives. These data suggest that the STE test is a useful method for predicting the eye irritation potential (‘‘I’’ or ‘‘NI’’) of mixtures. Two approaches (top-down or bottom-up) have been used to identify the eye irritation potential of chemicals based on the GHS classification system. The top-down approach is used to distinguish severe eye irritants (i.e., GHS Category 1) from all other categories (i.e., GHS Category 2 or Not Classified). On the other hand, the bottom-up approach is used to distinguish GHS Not Classified substances from all other irritant categories (i.e., GHS Category 1 or 2). When using the bottom-up approach, a low false negative rate is required in order to avoid irritants being classified and mislabeled as non-irritants. In our current study, the STE test

619

reported no false negatives for any of the ‘‘I’’ or ‘‘NI’’ predictions for the tested chemical mixtures, suggesting that the STE test may be applicable when using a bottom-up approach to examine test chemical mixtures. Since animal tests for assessing the eye irritation potential of finished cosmetic products have been banned (Directive 2003/15/EC, 2003), the STE test would be beneficial for use by a cosmetic company, as most cosmetic ingredients and products have to be non-irritants or have a mild to moderate eye irritation potential. On the other hand, the STE test did report five false positives. Although the precise reasons for these overestimations remain unclear, high false positive rates are not a critical concern when using the bottom-up approach, as substances judged positive in the STE test would normally be tested with at least one other method. Thus, a bottom-up tiered approach has been proposed in which the STE test is combined with other tests (e.g., BCOP) for the purpose of predicting the three GHS categories (Category 1, 2 or Not Classified) for single chemicals (Hayashi et al., 2012a,b). Further research that assesses whether such a tiered approach would also be applicable for evaluating chemical mixtures will need to be undertaken in the future. The predictive performance of the STE test was compared with two other OECD TG accepted tests (Table 3). Although both the number and content of the tested chemical mixtures differed, our data demonstrated that the STE test results were comparable, or in some cases, exhibited a greater predictive performance when compared to the two other TG methods. Moreover, as previously discussed, the STE test specificity (50%) appeared not to be a critical factor, especially when taking into account the lower specificity found for the BCOP (20%). Overall, these results suggest that the STE test has a predictive performance that is sufficient for identifying the eye irritation potential of chemical mixtures. Considering that the STE test is a cell-based assay, which is simple, quick, and less expensive than other methods, and the fact that the assay did not show any false negatives when using the bottom-up approach, the STE test might be a suitable first test when using a bottom-up tiered approach to identify GHS Not Classified mixtures from mixtures classified as GHS irritant categories. Conflict of Interest The authors declare that there are no conflicts of interest. Transparency Document The Transparency document associated with this article can be found in the online version. References Baker, F.W., Bruner, L.H., 1997. In house strategies for the safety evaluation of cosmetic products: the contribution of alternative methods. Animal Alternatives, Welfare and Ethics. In: Proceedings of 2nd World Congress on Alternatives and Animal Use in the Life Sciences. World Congress. van Zutphen, L.F.M., Balls, M. Developments in Animal and Veterinary Sciences, vol. 27. pp. 567–4. Balls, M., Berg, N., Bruner, L.H., Curren, R.D., De Silva, O., Earl, L.K., Esdaile, D.J., Fentem, J.H., Liebsch, M., Ohno, Y., Prinsen, M.K., Spielmann, H., Worth, A.P., 1999. Eye irritation testing: the way forward. The report and recommendations of ECVAM workshop 34. Altern. Lab. Anim. 27, 53–77. Bruner, L.H., Evans, M.G., Mcpherson, J.P., Southee, J.A., Williamson, P.S., 1998. Investigation of ingredient interactions in cosmetic formulations using isolated bovine corneas. Toxicol. In Vitro 12, 669–690. Directive 2003/15/EC of the European Parliament and of the Council of 27 February 2003. Official Journal of the European Union, L66. pp. 26–35. Draize, I.H., Woodard, G., Calvery, H.O., 1944. Methods for the study of irritation and toxicity of substances applied topically to the skin and mucous membranes. J. Pharmacol. Exp. Ther. 82, 377–390. Eskes, C., Bessou, S., Bruner, L., Curren, R., Harbell, J., Jones, P., Kreiling, R., Liebsch, M., McNamee, P., Pape, W., Prinsen, M.K., Seidle, T., Vanparys, P., Worth, A., Zuang, V.,

620

K. Saito et al. / Toxicology in Vitro 29 (2015) 617–620

2005. Alternative (Non-Animal) methods for cosmetics testing: current status and future prospects, 3.3, eye irritation. Altern. Lab. Anim. 33, 47–81. Hayashi, K., Mori, T., Abo, T., Koike, M., Takahashi, Y., Sakaguchi, H., Nishiyama, N., 2012a. A tiered approach combining the short time exposure (STE) test and the bovine corneal opacity and permeability (BCOP) assay for predicting eye irritation potential of chemicals. J. Toxicol. Sci. 37, 269–280. Hayashi, K., Mori, T., Abo, T., Ooshima, K., Hayashi, T., Komano, T., Takahashi, Y., Sakaguchi, H., Takatsu, A., Nishiyama, N., 2012b. Two-stage bottom-up tiered approach combining several alternatives for identification of eye irritation potential of chemicals including insoluble or volatile substances. Toxicol. In Vitro 26, 1199–1208. OECD, 2009a. OECD Guideline for the Testing of Chemicals. Test No. 437: Bovine Corneal Opacity and Permeability Test Method for Identifying Ocular Corrosives and Severe Irritants. Paris. . OECD, 2009b. OECD Guideline for the Testing of Chemicals. Test No. 438: Isolated Chicken Eye Test Method for Identifying Ocular Corrosives and Severe Irritants. Paris. . OECD, 2012. OECD Guideline for the Testing of Chemicals. Test No. 460: Fluorescein Leakage Test Method for Identifying Ocular Corrosives and Severe Irritants. Paris. .

OECD, 2013a. Streamlined Summary Document Supporting OECD Test Guideline 437 on the Bovine Corneal Opacity and Permeability for Eye irritation/ Corrosion. Series on Testing an Assessment No. 189, OECD, Paris. OECD, 2013b. Streamlined Summary Document Supporting OECD Test Guideline 438 on the Isolated Chicken Eye for Eye irritation/Corrosion. Series on Testing an Assessment No. 188 (part 1 and part 2), OECD, Paris. Ohno, Y., Kaneko, T., Inoue, T., Morikawa, Y., Yoshida, T., Fujii, A., Matsuda, M., Ohno, T., Hayashi, M., Momma, J., Uchiyama, T., Chiba, K., Ikeda, N., Imanishi, Y., Itagaki, H., Kakishima, H., Kasai, Y., Kurishita, A., Kojima, H., Matsukawa, K., Nakamura, T., Ohkoshi, K., Okumura, H., Saijo, K., Sakamoto, K., Suzuki, T., Takano, K., Tutsumi, H., Tani, N., Usami, M., Watanabe, R., 1999. Interlaboratory validation of the in vitro eye irritation tests for cosmetic ingredients. (1) Overview of the validation study and Draize scores for the evaluation of the tests. Toxicol. In Vitro 13, 73–98. Takahashi, Y., Koike, M., Honda, H., Ito, Y., Sakaguchi, H., Suzuki, H., Nishiyama, N., 2008. Development of the short time exposure (STE) test: an in vitro eye irritation test using SIRC cells. Toxicol. In Vitro 22, 760–770. Takahashi, Y., Hayashi, T., Watanabe, S., Hayashi, K., Koike, M., Aisawa, N., Ebata, S., Sakaguchi, H., Nakamura, T., Kuwahara, H., Nishiyama, N., 2009. Interlaboratory study of short time exposure (STE) test for predicting eye irritation potential of chemicals and correspondence to globally harmonized system (GHS) classification. J. Toxicol. Sci. 34, 611–626.