Keratinocyte ATP Release Assay for Testing Skin-Irritating Potentials of Structurally Diverse Chemicals  Norikatsu Mizumoto, Mark E. Mummert, David Shalhevet, Akira Takashima  Journal of Investigative Dermatology  Volume 121, Issue 5, Pages 1066-1072 (November 2003) DOI: 10.1046/j.1523-1747.2003.12558.x Copyright © 2003 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 1 Chemical structures and in vivo skin-irritating potentials of the test compounds. Corrosive chemicals are indicated with asterisks. Animal test results are shown in parentheses, whereas the human 4-h patch test results are shown in brackets. In the Draize test, the samples causing rapid (<4 h) and irreversible changes penetrating into dermis are classified as corrosive chemicals, whereas those causing erythema and edema after prolonged exposure (e.g., 24 h) are considered as irritant chemicals, which are further classified based on the primary irritation index (PII) values. The European Union classifies the corrosive chemicals as “R34” and the irritant chemicals with PII values above 2 as “R38.” The United Nations classification, which is required for international transportation of packaged corrosive chemicals, is based on the minimal exposure periods necessary for causing skin damage: Group I (3 min), Group II (1 h), and Group III (4 h) (Bagley et al, 1995;Barratt et al, 1998). N/C, nonclassified. Journal of Investigative Dermatology 2003 121, 1066-1072DOI: (10.1046/j.1523-1747.2003.12558.x) Copyright © 2003 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 2 In vitro test results with the Pam 212 keratinocyte-based ATP release assay and the liposome disruption assay. Pam 212 keratinocytes were incubated for 10 min at room temperature with each chemical at the indicated concentrations. The supernatants were then examined for ATP concentrations by the standard luciferin–luciferase assay (means±SD, n=3, *p<0.05 and **p<0.01 compared to the negative control treated with PBS alone) (red lines). Liposomes containing fluorescein diphosphate were incubated for 10 min with each chemical at the indicated concentrations. The samples were then examined for fluorescence signals after activation with ALP (blue lines). None of the compounds directly affected the enzymatic activities of luciferase or ALP in the concentration ranges shown in this figure. Data shown are representative of at least two independent experiments. Journal of Investigative Dermatology 2003 121, 1066-1072DOI: (10.1046/j.1523-1747.2003.12558.x) Copyright © 2003 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 3 ATP release from human keratinocytes. The human keratinocyte line A431 (A) or normal human keratinocyte cultures (B) were exposed for 10 min to each of the three irritants at the indicated concentrations and the supernatants were examined for ATP concentrations (means±SD, n=3, **p<0.01 compared to the negative control with PBS alone). Data shown are representative of two independent experiments. Journal of Investigative Dermatology 2003 121, 1066-1072DOI: (10.1046/j.1523-1747.2003.12558.x) Copyright © 2003 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 4 ATP release following exposure to structurally similar contact allergens. DNFB and DNBS, which have similar chemical structures (A), were examined for the potential to induce ATP release (B). Pam 212 keratinocytes were exposed for 10 min to DNFB (closed circles) or DNBS (open circles) at the indicated concentrations in terms of the molarity and the supernatants were examined for ATP concentrations (means±SD, n=3, **p<0.01 compared to the negative control exposed to PBS alone). Data shown are representative of two independent experiments. Journal of Investigative Dermatology 2003 121, 1066-1072DOI: (10.1046/j.1523-1747.2003.12558.x) Copyright © 2003 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 5 In vivo test results with a mouse model of irritant dermatitis. The test compounds, except for those that are classified as corrosive chemicals, were examined for their in vivo potentials to produce irritant dermatitis in our mouse model. BALB/c mice (five mice/panel) received topical application of each chemical at the indicated concentration and were then examined for ear swelling responses after 20 min and 4, 24, 48, and 72 h (means±SEM). Statistically significant differences are indicated with asterisks (*p<0.05 and **p<0.01). Data are representative of two independent experiments. Journal of Investigative Dermatology 2003 121, 1066-1072DOI: (10.1046/j.1523-1747.2003.12558.x) Copyright © 2003 The Society for Investigative Dermatology, Inc Terms and Conditions