Research article
Expanding the applicability of the amino acid derivative reactivity assay: Determining a weight for preparation of test chemical solutions that yield a predictive capacity identical to the conventional method using molar concentration and demonstrating the capacity to detect sensitizers in liquid mixtures

https://doi.org/10.1016/j.vascn.2019.01.001Get rights and content

Abstract

Introduction

The amino acid derivative reactivity assay (ADRA) is a novel in chemico alternative to animal testing for assessment of skin sensitization potential. The conventional ADRA protocol stipulates that test chemical solutions should be prepared to a specific molar concentration, allowing only for use of test chemicals with known molecular weights. Since many potential test substances are prepared by weight concentration or contain multiple unknown chemicals, this study was conducted to verify if it is possible to accurately assess the sensitization potential of test chemical solutions prepared at a specific weight concentration.

Methods

(1) Test chemical solutions for 82 chemicals were prepared at four different weight concentrations. Results were evaluated for agreement with in vivo results. (2) A liquid mixture comprising ten different non-sensitizers was prepared at 1 mg/mL. Ten different sensitizers of varying sensitization potencies were added individually to this mixture. The resulting pseudobinary mixtures were tested to confirm that the sensitizers could be detected.

Results

(1) The accuracies for test chemical solutions prepared at 0.5 and 0.2 mg/mL were 87.8% and 86.6%, respectively, which were roughly equivalent to the accuracy of 86.6% achieved with a solution prepared at the conventional molar concentration of 1 mM. In contrast, the accuracies for solutions prepared at 0.1 and 0.05 mg/mL were 82.9% and 74.4%, respectively, both of which were lower than that obtained with the conventional method. (2) Sensitizers added to the liquid mixture at 0.5 mg/mL were all correctly detected.

Discussion

Preparing test chemical solutions at a weight concentration of 0.5 mg/mL decreased false negatives and increased false positives while improving prediction accuracy, which suggests that the sensitization potential of mixtures can also be assessed with this method.

Introduction

In recent years, a number of alternative methods to the use of laboratory animals in testing for the sensitization potential of chemical substances have been developed, six of which have been adopted and issued as OECD test guidelines.

Of these, alternative methods for testing skin sensitization potential that address elements of the adverse outcome pathway (AOP) as endpoints include the Direct Peptide Reactivity Assay (OECD TG442C, 2015), which addresses the Molecular Initiating Event of covalent binding with proteins (haptenization); the ARE-Nrf2 Luciferase KeratinoSensTM and ARE-Nrf2 Luciferase LuSens Tests (OECD TG442D, 2018a), which address the Key Event of keratinocyte activation; the Human Cell Line Activation Test, U937 Cell Line Activation Test, and IL-8 Luc Assay (OECD TG442E, 2018b), which address the Key Event of dendritic cell activation; and the murine Local Lymph Node Assay (OECD429, 2010), which addresses the Key Event of activation and proliferation of antigen-specific T-cells.

The sensitization mechanism comprising the Molecular Initiating Event and other Key Events mentioned above is a complex one, that should be assessed using a combination of test methods under an Integrated Approach to Testing and Assessment (IATA) (OECD, Series on Testing & Assessment No. 255, 2016). In fact, Urbisch et al. reported in 2016 that predictions for the skin sensitization potential of test chemicals made using an IATA comprising DPRA, KeratinoSens, and h-CLAT were even more accurate than those made using LLNA when compared with human data.

In developing ADRA, we addressed the Molecular Initiating Event of covalent binding with proteins by synthesizing two nucleophilic reagents from heptapeptides. We introduced naphthalene rings to the N-termini of both cysteine and lysine, thereby synthesizing N-(2-(1-naphthyl)acetyl)-l-cysteine (NAC) and α-N-(2-(1-naphthyl)acetyl)-l-lysine (NAL) (Fujita et al., 2014).

ADRA affords two significant improvements over DPRA: First, since it can be performed using test chemical solutions at just 1% of the concentration required by DPRA, there is virtually no precipitation of the test chemical in the test chemical solution. Second, since UV is measured at a wavelength of 281 nm, there is virtually no co-elution of the test chemical and the nucleophilic reagents during HPLC analysis. Moreover, the addition of EDTA to the NAC solution provides increased stability and prevents oxidative dimerization of cysteine derivatives (Fujita et al., 2019).

As described above, ADRA ameliorates a number of limitations that affected DPRA, but there are some issues that still require improvement. Both DPRA and ADRA require that test chemical solutions and the peptide or NAC and NAL reaction solutions be prepared with their molar concentrations at a specific ratio to each other. Thus, test chemical solutions must be prepared to a specific molar concentration.

Nevertheless, the reaction solutions used in ADRA are prepared so that the molar concentration of NAC or NAL to that of the test chemical solution is at a ratio of 1:50, and this significant excess of test chemical relative to the nucleophilic reagent means we do not anticipate that small variations in this specific ratio would have a considerable impact on reactivity.

Also, since it was necessary for the test chemical solutions used in DPRA to be prepared at the very high molar concentration of 100 mM, preparing test chemical solutions according to weight concentration often resulted in molar concentrations significantly higher than 100 mM, which we think increases the potential that the test chemical would precipitate in the reaction solution or that the peaks of the peptides would co-elute with the peaks of the test chemical during HPLC analysis. Since, however, the test chemical solutions used in ADRA are prepared at a molar concentration of 1 mM, which is just 1% of that used in DPRA, the potential for the test chemical precipitating in the reaction solution even at higher than specified concentrations remains low. Moreover, since HPLC analysis is performed at the relatively long detection wavelength of 281 nm, the potential for the peaks of NAC or NAL to co-elute with the peaks of the test chemical also remains low.

Based on the above, we sought to develop an alternative to animal testing that would be useful in assessing the sensitization potential of chemicals of unknown molecular weight by determining an optimal weight concentration for test chemical solutions. In addition, we prepared ten pseudobinary mixtures comprising 10 different non-sensitizers and one sensitizer each to verify whether or not this method was capable of assessing the sensitization potential of liquid mixtures.

Section snippets

Test chemicals

The 82 test chemicals and their CAS numbers, suppliers, and the solvents used to prepare their test chemical solutions are summarized in Table 1. These chemicals were used to prepare test chemical solutions per different weight/volume concentrations, which were tested to determine an optimal weight concentration. Additionally, the following ten non-sensitizers were selected from these 82 test chemicals, for use in studying the assessment of mixtures: diethyl phthalate, 4-hydroxybenzoic acid,

Results

The ADRA test method specifies to prepare test chemical solutions at a molar concentration of 1 mM. In this study, 82 test chemicals used in development of the DPRA and ADRA test methods were evaluated to verify whether test chemical solutions prepared by weight concentrations (w/v) could be used to accurately predict sensitization potential and thereby establish a means for testing chemicals of unknown molecular weight (Fujita et al., 2014; Gerberick et al., 2007; Yamamoto et al., 2015). We

Discussion

Both DPRA (OECD TG 442C, 2015) and ADRA (Fujita et al., 2014; Yamamoto et al., 2015) use HPLC analysis to measure unreacted levels of peptides or amino acid derivatives, which are used as nucleophilic reagents. Each of these tests require that molar concentrations of the test chemicals and the nucleophilic reagents are at a specific ratio in the reaction solution, which means that the molecular weight of the test chemicals must be known.

The ability to prepare test chemical solutions based on

Acknowledgments

We thank Ms. Maiko Takasaki for technical assistance with HPLC operation.

References (12)

There are more references available in the full text version of this article.

Cited by (22)

  • A newly developed means of HPLC-fluorescence analysis for predicting the skin sensitization potential of multi-constituent substances using ADRA

    2019, Toxicology in Vitro
    Citation Excerpt :

    Finally, we prepared mono-constituent solutions of some of the major components of the three plant extracts as well as multi-constituent model extract solutions containing these major components to determine whether or not the predictions obtained from the substances were concordant with predictions from the plant extracts themselves. Yamamoto et al. (2019) has reported the testing of multi-constituent substances comprising 10 non-sensitizers and a single sensitizer, and because we used these to verify that ADRA-FL demonstrates sensitivity equivalent to that of ADRA-UV, we compared the results of measurements made using HPLC-UV with those made using HPLC-FL for two such multi-constituent substances: one containing p-benzoquinone, which co-eluted and could not be predicted correctly using HPLC-UV, and one containing perillaldehyde, which was predicted correctly using HPLC-UV. As can be seen in Fig. 3C, we found that when a test sample (positive control 1) containing p-benzoquinone was measured using ultraviolet detection, there were a significant number of peaks not only from degradation products, oxidation products, and these reactants of p-benzoquinone but from reaction products of the nucleophilic reagents with p-benzoquinone, which made it impossible to avoid co-elution with unreacted nucleophilic reagents.

  • Precipitation of test chemicals in reaction solutions used in the amino acid derivative reactivity assay and the direct peptide reactivity assay

    2019, Journal of Pharmacological and Toxicological Methods
    Citation Excerpt :

    In this study, we review results for 82 test chemicals that were tested using both DPRA and ADRA as well as compare whether or not precipitation was found in the reaction solution for either test in order to demonstrate ADRA's utility as a means of predicting the skin sensitization potential of test chemicals that are poorly soluble in water. The 82 test chemicals consisting of 53 sensitizers and 29 non-sensitizers that were used to establish the DPRA (Gerberick et al. 2007) and ADRA (Fujita et al., 2014, Yamamoto et al., 2015, Fujita et al., 2019b, Yamamoto et al., 2019) were also used in this study. Suppliers were identical to those reported by Fujita et al., 2019b) and Yamamoto et al. (2019).

View all citing articles on Scopus
View full text