Causes and countermeasure for blank absorbance increase in the ROS assay

Abstract

A reactive oxygen species (ROS) assay is an in chemico photoreactivity test listed in ICH S10 guideline and OECD Test Guideline No. 495. We currently utilize the ROS assay to assess the photosafety of cosmetic ingredients. We have recently confronted a problem that there was an absorbance increase of blank assessing superoxide anion generation after irradiation, whereas this did not occur in the negative control (sulisobenzone), leading to a dissatisfaction of the acceptance criteria. Therefore, we aimed to investigate the causes and find countermeasures. No significant effects of impurities and manufacturer differences of sodium phosphate and DMSO on blank absorbance increases were observed. In contrast, when Cu²⁺ was added to the buffer, the increase of blank absorbance after irradiation did not occur. We then confirmed the dose-response relationship and found that adding 0.1 μM of Cu²⁺ (corresponding to 6 ppb of Cu²⁺) was sufficient in suppressing the blank absorbance increase, suggesting the need of Cu²⁺ supplementation to the buffer. Finally, we confirmed that the ROS assay using the buffer supplemented with 0.1 μM of Cu²⁺ obtained stable test results by using 17 proficiency chemicals listed in TG 495. Our results suggest that the modified ROS assay protocol would be useful for obtaining stable test results.

INTRODUCTION

A reactive oxygen species (ROS) assay is an in chemico photoreactivity test listed in International Council for Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) S10 guideline (ICH, 2014) and the Organisation for Economic Cooperation and Development (OECD) Test Guideline No. 495 (OECD, 2019). The ROS assay is designed for the qualitative photoreactivity assessment of chemicals, assessing type I (an electron or hydrogen transfer, resulting in the formation of free radical species) and type II (an energy transfer from excited triplet photosensitizer to the oxygen) photochemical reactions in test chemicals exposed to simulated sunlight (Onoue and Tsuda, 2006). In the ROS assay, superoxide anion (SA) can be measured by the reduction of nitroblue tetrazolium, and singlet oxygen (SO) can be assessed using p-nitrosodimethylaniline bleaching by oxidized imidazole (OECD, 2019). The ROS assay is recommended to be used as a screening method for the initial photosafety evaluation because photoreactivity is described as a key characteristic of phototoxic and/or photoallergic chemicals, and the ROS assay has high sensitivity. We currently utilize the ROS assay to assess photosafety of cosmetic ingredients. We have recently confronted a problem that there is an absorbance increase of blank assessing SA generation after irradiation, whereas this did not occur in the negative control (sulisobenzone), leading to a dissatisfaction of the acceptance criteria (Table 1). In this study, therefore, we investigated the causes of the issues and countermeasures to obtain stable test results.

Table 1. ROS assay data of blank, positive control and negative control using buffer with and without Cu²⁺.

MATERIALS AND METHODS

Chemicals and reagents

Sodium phosphate

NaH₂PO₄ (Japanese pharmaceutical excipients), NaH₂PO₄▪2H₂O (Guaranteed reagent), Na₂HPO₄ (for pH standard solution and Japanese pharmaceutical excipients) and Na₂HPO₄▪12H₂O (Guaranteed reagent) were purchased from Fujifilm Wako Pure Chemical Corporation (Osaka, Japan). NaH₂PO₄▪H₂O (ACS reagent) and Na₂HPO₄▪H₂O (ACS reagent) were obtained from Sigma-Aldrich Co. LLC (St. Louis, MO, USA). A 20 mM sodium phosphate buffer (NaPB, pH7.4) was prepared using ultrapure water (Fujifilm Wako) as described in OECD TG 495 (OECD, 2019).

DMSO

Guaranteed reagent 99.0+% and for residual solvents analysis (99.95+%) were purchased from Fujifilm Wako. For spectroscopy > 99.7% was obtained from Kanto Chemical Co., Inc. (Tokyo, Japan). For HPLC (≥ 99.7%) was from Sigma-Aldrich.

Solvents and Ions

Methanol, ethanol, acetonitrile, acetone, KCl, NaCl, FeCl₂▪4H₂O, FeCl₃▪6H₂O, K₂CO₃, ZnCl₂, MgCl₂, MnCl₂▪4H₂O, CuSO₄ and H₂SO₄ were purchased from Fujifilm Wako. CuCl₂, AlCl₃▪6H₂O and CaCl₂ were obtained from Nacalai Tesque, Inc. (Kyoto, Japan). Ammonia solution (28%) was from Sigma-Aldrich.

Proficiency chemicals and plant extracts

Proficiency chemicals and plant extracts used in this study are listed in Table 2 and Table 3.

Table 2. ROS assay results. SD was calculated from three independent experiments.

Table 3. ROS assay results of plant extracts used in Nishida et al. (2015). SD was calculated from three independent experiments.

Irradiation conditions

The ROS assay was conducted using an Atlas Suntest CPS Plus (Atlas Material Technology LLC, Chicago, IL, USA) equipped with a Xenon arc lamp (1500 W) and cooling unit SR-P20FLE (Hitachi, Tokyo, Japan). A UV special filter was installed to adapt the spectrum of the artificial light source to that of natural day-light, and the Atlas Suntest CPS series had a high irradiance capability that met CIE85/1989 daylight simulation requirements. The irradiation was carried out at 25°C with an irradiance of ca. 2.0 mW/cm² as determined with a calibrated UVA detector Dr. Hönle #0037 (Dr. Hönle, München, Germany).

ROS assay

The ROS assay was conducted as described in OECD TG 495 (OECD, 2019) and the workflow diagram is shown in Fig. 1. Briefly, the test compounds were dissolved in DMSO or NaPB at 1 or 10 mM (final concentration is 20 or 200 μM) as a stock solution. Guaranteed reagent DMSO (99.0+%) from Wako was used, unless otherwise noted. For molecular weight unknown materials such as plant extracts, all substances were dissolved in DMSO at 2.5 mg/mL (final concentration is 50 μg/mL) as described by (Nishida et al., 2015); note that this protocol is not validated method. Each test compound, p-nitrosodimethylaniline (50 μM) and imidazole (50 μM) were dissolved in NaPB to determine singlet oxygen (SO) generation spectrophotometrically at 440 nm. Each test compound and nitroblue tetrazolium (50 μM) were dissolved in NaPB to measure superoxide anion (SA) generation spectrophotometrically at 560 nm. A plastic 96-well microplate (clear, untreated, flat-bottomed; Asahi Glass, Tokyo, Japan) containing reaction mixtures was fixed in the reaction container with a quartz cover (Ozawa Science, Aichi, Japan) and irradiated for 1 hr. Before and after exposure to artificial sunlight, absorbances at 440 and 560 nm were monitored using a SYNERGY H1 microplate reader (BioTek, Winooski, VT, USA). The photoreactivity was judged as follows: (i) positive for SO (ΔA_440nm × 10³): 25 or more; and/or SA (ΔA_560nm × 10³): 20 or more, or (ii) negative for SO (ΔA_440nm × 10³): less than 25, and SA (ΔA_560nm × 10³): less than 20.

Fig. 1

Workflow diagram for ROS assay.

RESULTS

Purities and manufacturer differences of sodium phosphate and DMSO

Nitroblue tetrazolium (50 μM) and DMSO (0.2%) were dissolved in NaPB to measure SA generation. No significant effects of impurities and manufacturer differences of sodium phosphate and DMSO on SA generation were observed (Fig. 2A). These results suggested that impurities of sodium phosphate and DMSO were not the cause.

Fig. 2

(A) Effect of manufacturer differences of DMSO on SA generation. Dashed line represents 20 (positive threshold of SA). (B) Effect of solvents on SA generation. Dashed line represents 20 (positive threshold of SA). (C) Dose-response relationships of CuCl₂, CuSO₄ or MnCl₂. (D) Effect of EDTA on Cu²⁺ supplementation. The error bars indicate standard deviations obtained from three independent experiments.

Solvents

Nitroblue tetrazolium (50 μM) and solvents (0.2%) such as NaPB, DMSO, methanol, ethanol, acetonitrile or acetone were dissolved in NaPB to measure SA generation. The effects of solvents on SA generation were investigated, and only when DMSO was used, the absorbance increase occurred (Fig. 2B). This suggested that DMSO itself was relevant to SA generation. Note that DMSO is supposed to be added to the reaction mixture in the ROS assay even when the test chemical is prepared in NaPB (Fig. 1).

Impurity ions in NaPB

To determine whether impurity ions in NaPB are related to blank absorbance increase, various kinds of ions were added to NaPB. Impurity ions were selected from Drinking Water Quality Standards in Japan (MHLW, 2015). Supplementation of 2 μM Cu²⁺ (CuSO₄, CuCl₂) or Mn²⁺ to NaPB suppressed SA generation, whereas K⁺, Na⁺, Ca²⁺, Mg²⁺, Al³⁺, Fe²⁺, Fe³⁺, Zn²⁺, NH₄⁺, Cl^-, CO₃^2- and SO₄^2- had no effect (Data not shown). We then investigated optimum concentrations by confirming the dose-response relationships (Fig. 2C). We found that adding 0.1 μM of Cu²⁺ (corresponding to 6 ppb of Cu²⁺) or 10 μM of Mn²⁺ (corresponding to 549 ppb of Mn²⁺) was sufficient in suppressing the blank absorbance increase.

The effect of EDTA on Cu²⁺ supplementation

The effect of adding chelating agent EDTA on Cu²⁺ was studied (Fig. 2D). The results showed that EDTA inhibited the effect of Cu²⁺ supplementation, suggesting that Cu²⁺ not in chelated form has the property of reducing SA generation and chelating agents tend to produce false-positive results.

Effect of Cu²⁺ on ROS generation

The effects of Cu²⁺ on ROS generation were investigated using 200 μM of ketoprofen, quinine (positive control), acridine or norfloxacin (Fig. 3A-D). The results revealed that Cu²⁺ inhibited SA generation in a dose-dependent manner by maximum of 90% (ketoprofen), 44% (quinine) and 65% (acridine) at a concentration of 10 μM, respectively, while no effect was observed for norfloxacin which is generally known as a cation chelator. These results suggested that Cu²⁺ has superoxide anion scavenging properties. In contrast to SA generation, Cu²⁺ slightly inhibited SO generation in a dose-dependent manner by a maximum of 21% (ketoprofen) and 13% (quinine) at a concentration of 10 μM, respectively, while no effects were observed for acridine and norfloxacin.

Fig. 3

Effects of Cu²⁺ on ROS generation using 200 μM of (A) ketoprofen, (B) quinine, (C) acridine or (D) norfloxacin. The error bars indicate standard deviations obtained from three independent experiments. Dashed line represents 0.1 μM (concentration sufficient in suppressing the blank absorbance increase).

Performance of modified protocol

Retesting was conducted by using 17 proficiency chemicals listed in TG 495 and plant extracts under the same conditions other than adding Cu²⁺ at a concentration of 0.1 μM to NaPB (Table 2 and Table 3). For the proficiency chemicals, all substances except ketoprofen were within an acceptable range, while only the SA value of ketoprofen was slightly below the criteria. For the plant extracts, all phototoxic materials were judged as positive and all non-phototoxic materials except lemongrass oil were predicted as negative. These results suggested that the modified ROS assay protocol would be useful for obtaining stable test results.

DISCUSSION

We have recently encountered the problem of an absorbance increase of blank assessing SA generation after irradiation. We investigated the causes and countermeasures in order to obtain stable test results. We first suspected that impure reagents, such as sodium phosphate or DMSO, might be the cause. However, impurities and manufacturer differences had no significant effects for sodium phosphate and DMSO on SA generation. We then confirmed the effects of solvents using NaPB, DMSO, methanol, ethanol, acetonitrile or acetone. Only when DMSO was used did absorbance increase, suggesting that DMSO itself was relevant to SA generation. Note that DMSO is supposed to be added to the reaction mixture in a ROS assay even when the test chemical is prepared in NaPB. Finally, we suspected that impurity ions in NaPB were related to blank absorbance increases, and various kinds of ions were added to NaPB. Impurity ions were selected from Drinking Water Quality Standards in Japan (MHLW, 2015), and we found that adding 0.1 μM of Cu²⁺ (corresponding to 6 ppb of Cu²⁺) or 10 μM of Mn²⁺ (corresponding to 549 ppb of Mn²⁺) was sufficient in suppressing the blank absorbance increase. According to the Drinking Water Quality Standards in Japan, the standard value of copper is 1.0 mg/L (= 1000 ppb) and that of manganese is 0.05 mg/L (= 50 ppb), suggesting that copper might be the main cause. During the validation study of in chemico skin sensitization: amino acid derivative reactivity assay (ADRA), dimerization of N-(2-(1-naphthyl)acetyl)-L-cysteine (NAC) was found at participating laboratories and 32 ppb of Cu²⁺ was detected in distilled water treated by glass labware (Fujita et al., 2019). They also found that Cu²⁺ has the property in promoting the oxidation of thiols and concluded that NAC oxidation is caused by the presence of minute quantities of Cu²⁺. In the study by Qiao et al. (2001), the analysis of DMSO of chemical purity grade showed that there are some ionic impurities such as Fe, Cu and Zn. However, accurate amounts of ions were unclear because the quantitation was conducted using boiling residues (vacuumed under 1.0 × 10⁴ Pa, b.p. > 60°C) and the limit of detection of the inductively coupled plasma atomic emission spectrometry (ICP-AES; P-4010) that they used is in the ppm order. These suggested that metal ion contamination was not uncommon when the ROS assay was developed.

Cu²⁺ is the active site of copper-zinc superoxide dismutase, which is an oxidoreductase enzyme responsible for the rapid two-step dismutation of the toxic superoxide radical to molecular oxygen (O₂) and hydrogen peroxide (H₂O₂) through the alternate reduction and oxidation of Cu²⁺. However, there are no or few reports that Cu²⁺ not buried in enzyme active site acts as a superoxide radical scavenger. Figure 4A illustrates possible pathways to suppressing blank absorbance increases by copper. Cu⁺ (CuCl) inhibited SA generation in a dose-dependent manner similar to Cu²⁺ (Fig. 4B) and EDTA inhibited the effect of Cu²⁺ supplementation (Fig. 2D), supporting our hypothesis that the suppression of SA generation is caused by the alternate reduction and oxidation of copper ions. Therefore, radical scavenging capacities were investigated using ketoprofen, quinine (positive control), acridine or norfloxacin, and different behaviors were observed between chelating and non-chelating agents (Fig. 3A-D). Norfloxacin is a broad-spectrum fluoroquinolone antibiotic with variable activity against gram-positive and gram-negative bacteria (Wishart et al., 2018), and is known as a copper chelator (Ruíz et al., 2007). Quinine is an alkaloid used to treat uncomplicated plasmodium falciparum malaria (Wishart et al., 2018) and is reported to have a copper-chelating property (Tsangaris and Kabanos, 1982). No significant effect of Cu²⁺ on SA generation was observed for norfloxacin, while a dose-response curve was obtained for quinine at concentrations above 1 μM of Cu²⁺. This might be due to differences of chelating potency and photostability. On the other hand, Cu²⁺ suppressed SA generation in a dose-dependent manner at a maximum of 90% (ketoprofen) and 65% (acridine). These results suggested that Cu²⁺ is a superoxide anion scavenger. As far as we know, this might be the first report that copper not buried in enzyme active sites has superoxide anion scavenging properties. In contrast to SA, inhibition of SO generation was slightly observed. Although Cu²⁺ is reported to be a singlet oxygen scavenger (Joshi, 1998), its scavenging capacity was confirmed at a mM order. Therefore, 0.1 μM of Cu²⁺ supplementation to the buffer is regarded as being within the acceptance range.

Fig. 4

(A) Possible pathways to suppressing blank absorbance increase by the alternate reduction and oxidation of copper. (B) Effect of CuCl on SA generation. The error bars indicate standard deviations obtained from three independent experiments.

The performance of a modified protocol using a buffer supplemented with 0.1 μM of Cu²⁺ was confirmed by using 17 proficiency chemicals listed in TG 495. Although the SA value of ketoprofen was slightly below the criteria, all substances except ketoprofen were within an acceptable range. In the photodegradation of ketoprofen, hydroxy radical, SO and SA were involved (Costanzo et al., 1989), and radical scavenging experiments and dissolved oxygen experiments revealed that the SA played a primary role in the photolytic process of ketoprofen in pure water (Wang et al., 2017). These suggested that the effect of superoxide anion scavenger has a great impact on SA values in chemicals such as ketoprofen that SA plays a significant role in the photodegradation process.

According to the study by (Onoue et al., 2013; Seto et al., 2013), almost all of the phototoxic chemicals showed above SO positive criteria and there were few phototoxic chemicals that only exhibited above SA positive criteria. This suggests that there might be less possibility of causing false negatives by Cu²⁺ supplementation. However, an excess of Cu²⁺ in the buffer might under-represent the ROS assay’s results. Thus, it is highly recommended that the buffer must be prepared using disposable labware, and Cu²⁺ level should be monitored, and the level should be adjusted to 0.1 μM if it is lower than that level. Further studies should be conducted to redetermine the lowest Cu²⁺ concentration that shows the minimum effect on ROS generation, and a ROS assay using vehicles other than DMSO should be considered in the future.

In conclusion, in the present study, the problem of blank absorbance increase assessing superoxide anion generation after irradiation was investigated. The solvent DMSO was found to be the main cause and 0.1 μM of Cu²⁺ supplementation to the buffer was determined as being able to suppress blank absorbance increases. The performance of the modified protocol was confirmed using 17 proficiency chemicals listed in TG 495 and our results show that the modified ROS assay protocol could be useful for obtaining stable test results.

ACKNOWLEDGMENT

We are grateful to Ms. Du Bin for technical assistance.

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Costanzo,  L.L.,  De Guidi,  G.,  Condorelli,  G.,  Cambria,  A. and  Fama,  M. (1989): Molecular mechanism of drug photosensitization--II. Photohemolysis sensitized by ketoprofen. Photochem. Photobiol., 50, 359-365.
•  Fujita,  M.,  Yamamoto,  Y.,  Watanabe,  S.,  Sugawara,  T.,  Wakabayashi,  K.,  Tahara,  Y.,  Horie,  N.,  Fujimoto,  K.,  Kusakari,  K.,  Kurokawa,  Y.,  Kawakami,  T.,  Kojima,  K.,  Kojima,  H.,  Ono,  A.,  Katsuoka,  Y.,  Tanabe,  H.,  Yokoyama,  H. and  Kasahara,  T. (2019): Cause of and countermeasures for oxidation of the cysteine-derived reagent used in the amino acid derivative reactivity assay. J. Appl. Toxicol., 39, 191-208.
•  Joshi,  P.C. (1998): Copper(II) as an efficient scavenger of singlet molecular oxygen. Indian J. Biochem. Biophys., 35, 208-215.
• International Council on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. (2014): ICH Guidance S10 on Photosafety Evaluation of Pharmaceuticals. Available at: https://www.ema.europa.eu/en/documents/regulatory-procedural-guideline/ich-guideline-s10-photosafety-evaluation-pharmaceuticals-step-5_en.pdf. Accessed October 1^st, 2021
•  Nishida,  H.,  Hirota,  M.,  Seto,  Y.,  Suzuki,  G.,  Kato,  M.,  Kitagaki,  M.,  Sugiyama,  M.,  Kouzuki,  H. and  Onoue,  S. (2015): Non-animal photosafety screening for complex cosmetic ingredients with photochemical and photobiochemical assessment tools. Regul. Toxicol. Pharmacol., 72, 578-585.
• OECD. (2019): OECD Guideline for Testing of Chemicals. Test No. 495: Ros (Reactive Oxygen Species) Assay for Photoreactivity. Available at: https://www.oecd-ilibrary.org/environment/tg-495-ros-reactive-oxygen-species-assay-for-photoreactivity_915e00ac-en. Accessed October 1^st, 2021
•  Onoue,  S. and  Tsuda,  Y. (2006): Analytical studies on the prediction of photosensitive/phototoxic potential of pharmaceutical substances. Pharm. Res., 23, 156-164.
•  Onoue,  S.,  Suzuki,  G.,  Kato,  M.,  Hirota,  M.,  Nishida,  H.,  Kitagaki,  M.,  Kouzuki,  H. and  Yamada,  S. (2013): Non-animal photosafety assessment approaches for cosmetics based on the photochemical and photobiochemical properties. Toxicol. In Vitro, 27, 2316-2324.
•  Qiao,  X.,  Chen,  S.,  Tan,  L.,  Zheng,  H.,  Ding,  Y. and  Ping,  Z. (2001): Investigation of formation of superoxide anion radical in DMSO by ESR: Part 1. Influence of Fe²⁺ and Cu²⁺. Magn. Reson. Chem., 39, 207-211.
•  Ruíz,  P.,  Ortiz,  R.,  Perelló,  L.,  Alzuet,  G.,  González-Álvarez,  M.,  Liu-González,  M. and  Sanz-Ruíz,  F. (2007): Synthesis, structure, and nuclease properties of several binary and ternary complexes of copper(II) with norfloxacin and 1,10 phenantroline. J. Inorg. Biochem., 101, 831-840.
•  Seto,  Y.,  Kato,  M.,  Yamada,  S. and  Onoue,  S. (2013): Development of micellar reactive oxygen species assay for photosafety evaluation of poorly water-soluble chemicals. Toxicol. In Vitro, 27, 1838-1846.
• The Ministry of Health. Labour and Welfare (MHLW), Diagram13 Drinking Water Quality Standards in Japan (April 2015~). Available at: https://www.mhlw.go.jp/english/policy/health/water_supply/dl/4a.pdf, Accessed October 1^st, 2021
•  Tsangaris,  J.M. and  Kabanos,  T.A. (1982): Circular dichroism of the complexes of quinine with copper(II), nickel(II), cobalt(II), chromium(III) and palladium(II) chlorides. Monatsh. Chem., 113, 1393-1398.
•  Wang,  Y.,  Deng,  W.,  Wang,  F.,  Su,  Y.,  Feng,  Y.,  Chen,  P.,  Ma,  J.,  Su,  H.,  Yao,  K.,  Liu,  Y.,  Lv,  W. and  Liu,  G. (2017): Study of the simulated sunlight photolysis mechanism of ketoprofen: the role of superoxide anion radicals, transformation byproducts, and ecotoxicity assessment. Environ. Sci. Process. Impacts, 19, 1176-1184.
•  Wishart,  D.S.,  Feunang,  Y.D.,  Guo,  A.C.,  Lo,  E.J.,  Marcu,  A.,  Grant,  J.R.,  Sajed,  T.,  Johnson,  D.,  Li,  C.,  Sayeeda,  Z.,  Assempour,  N.,  Iynkkaran,  I.,  Liu,  Y.,  Maciejewski,  A.,  Gale,  N.,  Wilson,  A.,  Chin,  L.,  Cummings,  R.,  Le,  D.,  Pon,  A.,  Knox,  C. and  Wilson,  M. (2018): DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res., 46 (D1), D1074-D1082.