Skip to main content
NCBI home page
Toxicological Research logoLink to Toxicological Research
. 2022 Jun 4;38(4):531–544. doi: 10.1007/s43188-022-00130-8

Pre-validation study of spectrophotometric direct peptide reactivity assay (Spectro-DPRA) as a modified in chemico skin sensitization test method

Jung-Ah Seo 1, Sun-A Cho 2, Chang Eon Park 3, Dong Hyuk Seo 4, Myungsuk Choi 5, Susun An 2, Bae-Hwan Kim 1,✉
PMCID: PMC9532475  PMID: 36277359

Abstract

Skin sensitization is induced when certain chemicals bind to skin proteins. Direct peptide reactivity assay (DPRA) has been adopted by the OECD as an alternative method to evaluate skin sensitization by assessing a substance's reaction to two model peptides. A modified spectrophotometric method, Spectro-DPRA, can evaluate skin sensitization, in a high throughput fashion, to obviate some limitations of DPRA. Pre-validation studies for Spectro-DPRA were conducted to determine transferability and proficiency, within- and between-laboratory reproducibility, and predictive ability based on GLP principles at three laboratories (AP, KTR, and KCL). All laboratories confirmed high (> 90%) concordance for evaluating the sensitivity induced by ten chemical substances. The concordance among the three tests performed by each laboratory was 90% for AP, 100% for KTR, and 100% for KCL. The mean accuracy of the laboratories was 93.3% [compared to the standard operating procedure (SOP)]. The reproducibility among the three laboratories was as high as 86.7%; the accuracy was 86.7% for AP, 100% for KTR, and 86.7% for KCL (compared to the SOP). An additional 54 substances were assessed in 3 separate labs to verify the prediction rate. Based on the result, 29 out of 33 substances were classified as sensitizers, and 19 out of 21 identified as non-sensitizers; the corresponding sensitivity, specificity, and accuracy values were 87.9%, 90.5%, and 88.9%, respectively. These findings indicate that the Spectro-DPRA can address the molecular initiating event with improved predictability and reproducibility, while saving time and cost compared to DPRA or ADRA.

Keywords: DPRA, In chemico, Pre-validation, Skin sensitization, Spectro-DPRA

Introduction

Skin sensitization is a type of allergic reaction caused by exposure to certain chemical ingredients; it is becoming more common due to industrial development and environmental pollution [1, 2]. Animal tests for risk assessment of cosmetic raw materials and final products are being prohibited worldwide [3–5]. Thus, non-animal alternative platforms for detecting skin sensitization are urgently needed [6, 7].

The chemical and biological mechanisms involved in skin sensitization are summarized in the adverse outcome pathway (AOP) [8–11]. The first key event is the bindings of sensitizing substances to proteins in the skin, in which a cysteine or lysine residues of a protein bind to chemicals through haptenization reaction [12, 13]. Based on this mechanism, a direct peptide reactivity assay (DPRA) was developed and adopted in the Organization for Economic Co-operation and Development (OECD) test guideline 442C [11]. This assay determines the reactivity between chemicals and peptides by defining a peptide depletion (compared to a vehicle control) using high-performance liquid chromatography (HPLC) and ultraviolet (UV) detection [7]. However, this method has some disadvantages—it requires high-cost equipment with analysis technology and can be time-consuming and expensive for evaluating many samples.

Spectrophotometric-DPRA (Spectro-DPRA) was developed by the Amorepacific Corporation R&D Center in Korea to overcome these defects by simultaneously measuring many samples in a 96-well plate [14–16]. Spectro-DPRA evaluates the peptide depletion rate through the ratio of color development by adding the coloring reagents 5,5'-dithiobis-2-nitrobenzoic acid (DTNB) and fluorescamine to the remaining peptide solution. Specifically, cysteine peptide depletion can be quantitatively determined by measuring the absorbance of 5-mercapto-2-nitrobenzoic acid (TNB). TNB is generated when the thiol group of the remaining cysteine peptide (after reaction with the test substance) reacts with the disulfide bond of DTNB. In turn, lysine peptide depletion can be quantitatively determined by measuring the fluorescence intensity of the fluorophore produced when the free amine group in the lysine peptide reacts with fluorescamine [16, 17]. Overall, Spectro-DPRA assesses the amount of peptide used in the reaction by measuring the amount of reduced TNB or fluorophore compared to those before the reaction with the peptide. Nonetheless, for the Spectro-DPRA method to be recognized as a new test for assessing chemical sensitization, it must be validated through good laboratory practice (GLP) test.

Here, three laboratories jointly to performed pre-validation studies using Spectro-DPRA within- and between-laboratory tests to confirm whether Spectro-DPRA can be adopted as an international test guideline. Moreover, verification studies were conducted with additional substances to evaluate the predictive ability of the technique, by analyzing sensitivity, specificity, and accuracy.

Materials and methods

Test chemicals

Cysteine (Ac-RFAACAA-COOH, purity ≥ 98%, Mw = 750.35 g/mol) and lysine (Ac-RFAAKAA-COOH, purity ≥ 98%, Mw = 775.43 g/mol) peptides were custom-made by Peptron Co. (Daejeon, Korea). These peptide sequences were obtained from a previous report [18]. All substances were selected based on the data of the Local Lymph Node Assay (LLNA) [19] and the OECD TG 442C [11]. Ten substances were used in possibility tests for transferability, proficiency, and within-laboratory validation tests, and 15 substances were used for the between-laboratory validation tests. Additional 54 test substances (33 identified sensitizers and 21 identified non-sensitizers) were also selected, referring to the OECD validation report: Amino Acid Derivative Reactivity Assay (ADRA). Table 1 summarizes the properties of the test substance [20]. The chemical used for the test was diluted to 40 mM in isopropanol and prepared in stock. If the test substance was not dissolved in isopropanol, it was dissolved in distilled water (D.W.). The positive control chemicals (Cysteine: DNCB, Lysine: Glutaric dialdehyde) and the negative control chemical (Glycerol) were prepared at the same concentration as the chemical stock. Peptide solution and test substance were prepared and used on the test day. Peptide depletion was measured based on the standard curves of cysteine and lysine.

Table 1.

List of tested chemicals used in this study

NO Test chemical Chemical information In vivo prediction (LLNA) DPRA prediction Used
CAS NO. MW Purity (%) Physical state EC3 (%) Result
1 2,4-Dinitrochlorobenzene 97-00-7 202.55 99 Solid 0.0003 [21] Extreme Positive Pilot test for transferability and proficiency
2 Oxazolone 15646-46-5 217.22 90 Solid 0.003 [21] Extreme Positive
3 Benzylideneacetone 122-57-6 146.19 99 Solid 3.7 [21] Moderate Positive
4 Farnesal 19317-11-4 220.35 85 Liquid 12 [21] Weak Positive
5 Imidazolidinyl urea 39236-46-9 388.29 100 Solid 24 [21] Weak Positive
6 1-Butanol 7136-3 74.12 99 Liquid ND Non-sensitizer Negative
7 6-Methylcoumarin 92-48-8 160.17 99 Solid ND Non-sensitizer Negative Within and between laboratory validation
8 Lactic Acid 50-21-5 90.08 88 Liquid ND Non-sensitizer Negative
9 4-Methoxyacetophenone 100-06-1 150.17 99 Solid ND Non-sensitizer Negative
10 Glycerol 56-81-5 92.09 99 Liquid ND Non-sensitizer Negative
11 2-Mercaptobenzothiazole 149-30-4 167.25 97 Solid 1.7 [22] Moderate Positive Between laboratory validation
12 Nickelsulfate 10101-97-0 262.85 98 Solid 4.8 [23] Moderate Positive
13 Cinnamyl alcohol 104-54-1 134.18 98 Liquid 21 [23] Weak Negative
14 R( +)-Limonene 5989-27-5 136.23 97 Liquid 69 [24] Weak Positive
15 Eugenol 97-53-0 164.2 98 Liquid 13 [25] Weak Positive
16 Diphenylcyclopropenone 886-38-4 206.24 98 Solid 0.003 [22] Extreme Positive Additional chemicals for skin sensitization potentials
17 Benzoyl peroxide 94-36-0 242.23 75 Solid 0.004 [21] Strong Positive
18 p-Benzoquinone 106-51-4 108.09 99.5 Solid 0.0099 [21] Strong Positive
19 Glutaraldehyde solution 111-30-8 100.12 70 Liquid 0.1 [21] Strong Positive
20 Lauryl gallate 1166-52-5 338.44 99 Solid 0.3 [21] Strong Positive
21 Propyl gallate 121-79-9 212.2 98 Solid 0.32 [21] Strong Positive
22 Trimelliticanhydride 552-30-7 192.13 97 Solid 0.22 [25] Strong Positive
23 Metol 55-55-0 172.19 98 Solid 0.8 [21] Strong Positive
24 2-Hydroxyethyl acrylate 818-61-1 116.12 96 Liquid 1.4 [21] Moderate Positive
25 Glyoxal solution 107-22-2 58.04 40 Liquid 1.4 [21] Moderate Positive
26 2-Methyl-3(2H)-isothiazolone 2682-20-4 115.15 95 Solid 1.9 [21] Moderate Positive
27 1,2-Benzisothiazol-3(2H)-one 2634-33-5 151.19 97 Solid 2.3 [21] Moderate Positive
28 Methyl 2-nonynoate 111-80-8 168.23 99 Liquid 2.5 [21] Moderate Positive
29 trans-Cinnamaldehyde 14371-10-9 132.16 99 Liquid 3 [21] Moderate Positive
30 Phenylacetaldehyde 122-78-1 120.15 90 Liquid 3 [21] Moderate Positive
31 2,4-Heptadienal 4313-03-5 110.15 90 Liquid 4 [21] Moderate Positive
32 Squaric acid 2892-51-5 114.06 99 Solid 4.3 [21] Moderate Positive
33 trans-2-Hexen-1-al 6728-26-3 98.14 98 Liquid 5.5 [21] Moderate Positive
34 Resorcinol 108–46-3 110.11 99 Solid 5.5 [21] Moderate Negative
35 Diethyl maleate 141–05-9 172.18 97 Liquid 5.8 [21] Moderate Positive
36 2-Phenylpropionaldehyde 93-53-8 134.18 98 Liquid 6.3 [21] Moderate Positive
37 Palmitoyl chloride 112–67-4 274.87 98 Liquid 8.8 [21] Moderate Positive
38 Nonanoyl chloride 764–85-2 176.68 96 Liquid 1.8 [21] Moderate Positive
39 α-Hexylcinnamaldehyde 101–86-0 216.32 85 Liquid 11 [21] Weak Negative
40 α-Amylcinnamaldehyde 122–40-7 202.29 97 Liquid 11 [21] Weak Negative
41 2,3-Butanedione 431–03-8 86.09 97 Liquid 11 [21] Weak Positive
42 Oxalic acid 144-62-7 90.03 99 Solid 15 [21] Weak Negative
43 Benzyl benzoate 120-51-4 212.24 99 Liquid 17 [21] Weak Negative
44 4-Allylanisole 140-67-0 148.2 98 Liquid 18 [21] Weak Positive
45 2,2,6,6-Tetramethyl-3,5-heptanedione 1118-71-4 184.28 98 Liquid 27 [21] Weak Negative
46 Ethylene glycol dimethacrylate 97-90-5 198.22 98 Liquid 28 [22] Weak Positive
47 Ethyl acrylate 140-88-5 100.12 99 Liquid 28 [21] Weak Positive
48 Hydroxycitronellal 107-75-5 172.26 95 Liquid 20 [25] Weak Positive
49 Hexane 110-54-3 86.18 95 Liquid ND Non-sensitizer Negative
50 Diethyl phthalate 84-66-2 222.24 99.5 Liquid ND Non-sensitizer Negative
51 Octanoic acid 124-07-2 144.21 99 Liquid ND Non-sensitizer Negative
52 4-Hydroxybenzoic acid 99-96-7 138.12 99 Solid ND Non-sensitizer Negative
53 Methyl salicylate 119-36-8 152.15 99 Liquid ND Non-sensitizer Negative
54 Chlorobenzene 108-90-7 112.56 99.9 Liquid ND Non-sensitizer Negative
55 1-Bromobutane 109-65-9 137.02 99 Liquid ND Non-sensitizer Negative
56 2-Acetylcyclohexanone 874-23-7 140.18 97 Liquid ND Non-sensitizer Positive
57 Ethyl benzoylacetate 94-02-0 192.21 95 Liquid ND Non-sensitizer Negative
58 Isopropanol 67-63-0 60.1 99.9 Liquid ND Non-sensitizer Negative
59 Propylene glycol 57-55-6 76.09 99.5 Liquid ND Non-sensitizer Negative
60 Sulfanilamide 63-74-1 172.2 99 Solid ND Non-sensitizer Negative
61 Isopropyl myristate 110-27-0 270.45 98 Liquid 44 [25] Non-sensitizer Negative
62 Methyl paraben 99-76-3 152.15 99 Solid ND Non-sensitizer Negative
63 Nonanoic acid 112-05-0 158.24 97 Liquid 21 [21] Non-sensitizer Negative
64 Propylparaben 94-13-3 180.2 99 Solid ND Non-sensitizer Negative
65 Salicylic acid 69-72-7 138.12 99 Solid ND Non-sensitizer ND
66 Sulfanilic acid 121-57-3 173.19 99 Solid ND Non-sensitizer Negative
67 Vanillin 121-33-5 152.15 99 Solid ND Non-sensitizer Negative
68 Coumarin 91-64-5 146.14 99 Solid ND Non-sensitizer Negative
69 Vinylidenechloride 75-35-4 96.94 99 Liquid ND Non-sensitizer Negative
Open in a new tab

CAS NO., chemical abstracts service number; ND, not detected

Measurement of cysteine peptide depletion

Cysteine peptide stock was dissolved in DMSO (Sigma-Aldrich, St. Louis, USA) at 10 mM and stored in a deep-freezer at -80 °C. The cysteine peptide stock was made to 400 µM concentration by diluting in 100 mM phosphate buffer (w/ 1 mM EDTA, pH 8), and the chemical stock was made to 4 mM by diluting tenfold with 100 mM phosphate buffer (w/ 1 mM EDTA, pH 8) on the test day. Each 90 µL/well of peptide and chemical was added to a 96-well plate (Falcon, Cat. NO 353072) and left for 24 h at room temperature (22 ± 2℃) for reaction. After the reaction, 20 µL/well of 5 mM 5,5'-dithiobis-2-nitrobenzoic acid (DTNB; Dojindo laboratories, Kumamoto, Japan) solution diluted in 100 mM sodium phosphate buffer (w/ 1 mM EDTA, pH 8) was added, and the reaction was performed for 3 min. The absorbance before adding the DTNB solution and after the reaction were measured using a spectrophotometer at a wavelength of 412 nm.

Measurement of lysine peptide depletion

Lysine peptide stock was dissolved at 10 mM in distilled water (DW) and stored in a deep-freezer at − 80 °C. On the day of the reactivity, the lysine peptide stock was made to 200 µM by diluting in 100 mM phosphate buffer (w/ 1 mM EDTA, pH 10), and the chemical stock was made to 4 mM by diluting tenfold with isopropanol. Each 110 µL/well of peptide and chemical was added to a 96-well polypropylene plate (Falcon, Cat. no 351190, NY, USA) and left for 24 h at room temperature (22 ± 2 °C) for reaction. After the reaction, 180 µL of the reacted solution was transferred to a light-proof black clear-bottom 96-well plate (Greiner Bio-One, Cat. no 655090, Frickenhausen, Germany), and 20 µL/well of 10 mM fluorescamine (Sigma-Aldrich) diluted in DMSO was added for 3 min. The fluorescence values before and after the addition of the fluorescamine solution were measured with a fluorometer (SpectraMax M2e, Molecular Devices, CA, USA) at a wavelength of 390 nm (excitation) and 465 nm (emission).

Data analysis and acceptance criteria

Figure 1 shows the peptide depletion ratio after the reaction of the peptide and chemicals for 24 h [11, 26]. It was calculated as follows: the values before the reaction were corrected and calculated since the chemicals that can affect absorbance may change the results of the peptide depletion analysis. In other words, the intrinsic color of the test substance may affect the absorbance or fluorescence value. Therefore, before calculating the color change due to the detection reagent treatment, the value was calculated and reflected by considering the color difference depending on the presence or absence of peptide and sample treatment (Fig. 1). Values are derived in the same way for positive and negative controls. Negative values were set to “0”, and values over 100 were set to “100”, as in DPRA or ADRA [11]. The calculated peptide depletion rate was presented as mean ± SD of the value obtained in 3 replications. When the cysteine peptide was depleted more than 10%, and/or when the lysine peptide was depleted more than 30%, it was identified as a sensitizer based on the cut-off criterion verified in previous studies [11, 15]. In other words, the test substance is judged sensitizer even if one of the two peptides showed positive. Further, the final determination of the results through the data analysis was set as follows; (1) When the test is repeated twice, if the repeated test result (positive/negative) is the same as the first test result for each peptide, the test is terminated. If any of the results are not consistent, a third test is to be conducted and reconfirmed. (2) In the third test, the final determination is made as a sensitizer or non-sensitizer based on the two matched results out of three.

Fig. 1.

Fig. 1

Open in a new tab

Spectrophotometric direct peptide reactivity assay to calculate the depletion ratio. The change of spectrophotometric value (SV) was calculated by measurement of optical density (OD) or relative fluorescence unit (RFU) just before and after adding the detection reagent. SVCD and SVC are the spectrophotometric values before and after the addition of the detection reagent to the peptide-only and solvent-only control, respectively. SVDD and SVD are the spectrophotometric values before and after adding the detection reagent to reaction solutions between peptide and chemical, respectively. SVBD and SVB are the spectrophotometric values before and after adding the detection reagent to the chemical-only control, respectively

The following acceptance criteria should be met for a run to be considered valid: (a) the peptide standard calibration curve should have an r2 > 0.99, (b) the mean percent cysteine depletion ranges of DNCB (the positive control for cysteine) should be between 80 and 100; the mean percent lysine depletion ranges of Glutaric dialdehyde (the positive control for lysine) should be between 98 to 100; and the mean percent depletion range of glycerol and the negative control for cysteine and lysine should be between 0 to 8 and between 0 to 20, respectively and (c) The maximum standard deviations of positive and negative controls and chemicals should be ≤ 15.3% for cysteine responses and ≤ 13.5% for lysine responses in the 3 replicates. If one or more of these criteria is not met, the run should be repeated.

Pilot study for transferability and proficiency tests

Ten candidate chemicals (Table 1) were used to evaluate the possibility of transferability and proficiency. The test method developed by the lead lab, Amorepacific R&D Center (AP), was transferred to GLP-certified laboratories, Korea Testing & Research Institute (KTR), and Korea Conformity Laboratories (KCL) through on-site training. KTR and KCL have no experience performing Spectro-DPRA and DPRA assay, so AP guided them with video meetings and email communication after training. Transferability and proficiency were confirmed if 8 out of 10 substances matched, following the standard operating procedure (SOP) developed based on the OECD TG 442C [11]. This test was conducted in a non-blinded manner.

Evaluation of reproducibility within and between laboratories

Ten substances used in the within-laboratory reproducibility evaluation were coded and tested three times by each laboratory (Table 1). It was determined as a success if 8 of the 10 substances matched. In the between laboratory reproducibility evaluation, 15 coded substances were used, and it was considered a success if 12 of the 15 substances matched when three laboratory results were compared. These tests were conducted in a blinded manner.

Skin sensitization potential

Skin sensitization potential of Spectro-DPRA was evaluated using 54 test substances (33 sensitizers, 21 non-sensitizers) (Table 1) which were selected by mainly referring to the OECD guideline reports assigned to 3 separate laboratories. As shown in Table 1, KCL performed the tests for chemical numbers 16–31, KTR for numbers 32 to 50, and AP for numbers 51 to 69. These tests were conducted in a non-blinded manner. The accuracy, sensitivity, specificity, false positive, and false negative prediction values were generated by comparisons with the in vivo prediction (LLNA) results.

Results

Pilot study for transferability and proficiency

A pilot study to evaluate transferability and proficiency of Spectro-DPRA using 10 substances was conducted with a non-blind method at three laboratories. The standard curve of the peptide from the three laboratories was confirmed as “r2 = 0.99 or higher”. AP and KTR verified 100% transferability concordance with the SOP. For KCL, 9 substances were consistent with the SOP except for imidazolidinyl urea, which showed a false negative (data not shown).

To set the acceptance criteria for proficiency substances for further testing, the ranges for each peptide's depletion rate for 10 proficiency substances were established as mean ± 2SD values, which were obtained from the experiments repeated 6 times at each laboratory (AP, KTR, and KCL) during the pre-validation test of Spectro-DPRA (Table 2). A high depletion for 4 substances out of 5 that are identified as sensitizers according to LLNA data was found in cysteine peptides. Two sensitizers (oxazolone and imidazolidinyl urea) showed more than 30% cut-off depletion in the reaction with lysine peptide. In the case of imidazolidinyl urea, the lysine peptide depletion was shown to play an essential role in the positive reaction, although it showed negative depletion with cysteine peptide. Non-sensitizer substances showed less than the cut-off (cys dep. > 10% and/or lys dep. > 30%) in all peptides. The method is in line with the principles of OECD TG 442C in adopting the least number of substances to meet the proposed acceptance range to be considered as passing the proficient level (i.e., 8 out of 10) (Appendix II, annex 2, OECD TG 442C, 2021) [11].

Table 2.

Measuring ranges for proficiency substances

NO Proficiency substances Measuring ranges Prediction (cys > 10 and/or lys > 30) LLNA Result
1 2,4-Dinitrochlorobenzene 85.9 ≤ Cys del. ≤ 100 0 ≤ Lys del. ≤ 42.5 S S
2 Oxazolone 41.5 ≤ Cys del. ≤ 80 95.2 ≤ Lys del. ≤ 100 S S
3 Benzylideneacetone 69.2 ≤ Cys del. ≤ 97 0 ≤ Lys del. ≤ 26.5 S S
4 Farnesal 15 ≤ Cys del. ≤ 80.9 1.1 ≤ Lys del. ≤ 33.7 S S
5 Imidazolidinyl urea 0 ≤ Cys del. ≤ 1.8 30 ≤ Lys del. ≤ 72.8 S S
6 1-Butanol 0 ≤ Cys del. ≤ 5.1 0 ≤ Lys del. ≤ 14.1 NS NS
7 6-Methylcoumarin 0 ≤ Cys del. ≤ 6.8 0 ≤ Lys del. ≤ 18.7 NS NS
8 Lactic Acid 0 ≤ Cys del. ≤ 8.9 0 ≤ Lys del. ≤ 22.5 NS NS
9 4-Methoxyacetophenone 0 ≤ Cys del. ≤ 8.9 0 ≤ Lys del. ≤ 15.8 NS NS
10 Glycerol 0 ≤ Cys del. ≤ 8.9 0 ≤ Lys del. ≤ 15.2 NS NS
Open in a new tab

Sensitization Classification, S: Sensitizer, NS: Non-Sensitizer

Within- and between-laboratory validation

Reproducibility within the laboratories was determined through the depletion rate calculation for the 10 assessed substances in a blinded manner; three repeated trials were conducted at each institution (AP, KCL, and KTR) (Table 3). In AP, nine out of ten substances were matched except for imidazolidinyl urea, which matched two times out of three. For KTR, three repetitions for 10 substances were all matched. For KCL, all substances matched in three repetitions, but the result of imidazolidinyl urea showed a false negative. The mean accuracy of three laboratories (AP 90%, KTR 100%, and KCL 90%) was 93.3% compared to the LLNA data.

Table 3.

The results of within-laboratory validation study

NO Test substance AP KTR KCL LLNA Result
1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd
Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult Cys dep Lys dep Re -sult
1 2,4-Dinitro chlorobenzene 99.1 4.0 S 99.3 18.2 S 99.1 22.2 S 99.6 7.2 S 92.9 6.8 S 85.2 5.3 S 99.7 20.5 S 98.9 8.3 S 97.8 5.9 S S
2 Oxazolone 70.8 99.5 S 55.1 99.4 S 38.0 99.6 S 63.6 99.5 S 60.8 99.7 S 62.5 99.6 S 58.4 99.5 S 66.1 99.4 S 51.1 99.3 S S
3 Benzylidene acetone 90.0 7.3 S 88.9 4.2 S 86.8 0.0 S 72.5 7.1 S 73.2 0.0 S 69.4 0.0 S 89.3 9.4 S 86.9 9.4 S 91.3 9.1 S S
4 Farnesal 71.3 12.4 S 39.0 11.8 S 47.0 22.4 S 23.1 9.2 S 21.8 2.5 S 14.7 6.8 S 12.8 8.8 S 23.9 8.8 S 65.2 4.4 S S
5 Imidazolidinyl urea 0.0 40.1 S1) 0.0 28.3 NS1) 0.0 34.9 S 0.0 68.5 S 0.0 63.2 S 0.0 53.3 S 0.0 22.8 NS2) 0.0 9.3 NS2) 0.0 10.7 NS2) S
6 1-Butanol 0.0 2.3 NS 0.3 2.7 NS 0.8 7.3 NS 0.0 0.0 NS 1.3 4.3 NS 0.0 0.0 NS 0.0 3.8 NS 0.0 0.0 NS 0.0 0.0 NS NS
7 6-Methylcoumarin 2.2 4.8 NS 0.5 1.0 NS 0.1 5.1 NS 0.0 0.0 NS 0.0 0.0 NS 0.0 0.0 NS 0.6 8.5 NS 0.0 0.0 NS 3.2 0.0 NS NS
8 Lactic Acid 0.5 6.3 NS 0.8 4.0 NS 1.4 15.0 NS 0.4 7.4 NS 0.0 1.6 NS 4.7 1.2 NS 0.5 7.1 NS 2.7 7.1 NS 4.0 0.0 NS NS
9 4-Methoxy acetophenone 3.4 9.2 NS 3.4 10.1 NS 1.0 20.7 NS 8.7 0.0 NS 6.1 0.0 NS 11.8 0.0 NS 4.5 0.0 NS 2.9 0.0 NS 3.9 1.4 NS NS
10 Glycerol 0.0 2.9 NS 1.4 8.1 NS 0.0 5.6 NS 2.1 3.8 NS 0.4 0.0 NS 5.1 0.0 NS 3.7 5.4 NS 0.4 5.4 NS 5.1 3.3 NS NS
Open in a new tab

Sensitization classification, S sensitizer, NS non-sensitizer

Regarding the between-laboratory reproducibility, tests were conducted on 15 substances (including 10 proficiency substances) by three laboratories using a blinded method. Two substances (nickel sulfate and cinnamyl alcohol) showed false negatives in AP and KCL, and the other 13 substances showed consistent results. (86.7%) (Table 4).

Table 4.

The results of between-laboratory validation study

NO Test substance AP KTR KCL Between-lab agreement LLNA Result
Cys dep Lys dep Result Cys dep Lys dep Result Cys dep Lys dep Result
Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD
1 2,4-Dinitrochlorobenzene 99.6 0.4 3.7 6.3 S 97.9 1.3 12.3 6.9 S 92.2 5.8 14.3 13.6 S Yes S
2 Oxazolone 53.5 2.6 99.9 0.1 S 71.0 2.1 98.9 0.5 S 56.8 1.8 99.2 0.1 S Yes S
3 2-Mercaptobenzothiazole 23.0 3.3 0.0 0.5 S 54.2 1.6 21.9 3.5 S 56.2 5.3 3.9 6.8 S Yes S
4 Benzylideneacetone 86.9 1.0 0.0 0.0 S 77.0 0.6 21.0 2.5 S 87.2 1.0 0.0 0.0 S Yes S
5 Nickelsulfate 1.6 3.6 0.0 0.0 NS 27.3 4.1 9.0 3.4 S 7.3 7.2 8.9 9.8 NS No S
6 Farnesal 30.5 11.6 1.7 3.0 S 28.3 16.3 26.0 3.2 S 24.4 8.0 7.1 12.3 S Yes S
7 Cinnamyl alcohol 2.6 3.0 0.0 0.0 NS 58.1 1.0 14.1 6.0 S 0.0 0.0 0.0 0.0 NS No S
8 Imidazolidinyl urea 0.0 0.0 30.3 4.0 S 0.0 0.0 47.1 7.6 S 0.0 0.0 31.7 18.1 S Yes S
9 R( +)-Limonene 17.8 1.5 0.0 0.0 S 14.0 13.2 4.2 7.4 S 18.2 5.6 0.0 0.0 S Yes S
10 Eugenol 16.5 2.2 0.0 0.0 S 34.8 3.2 21.1 12.5 S 18.5 2.3 1.4 2.1 S Yes S
11 1-Butanol 7.2 2.5 0.0 0.0 NS 1.4 1.3 0.0 0.0 NS 3.7 4.3 1.6 2.8 NS Yes NS
12 6-Methylcoumarin 3.8 3.5 0.0 0.0 NS 0.0 0.0 0.9 1.6 NS 0.0 0.0 0.0 0.0 NS Yes NS
13 Lactic Acid 6.5 0.9 0.0 0.0 NS 1.1 1.4 2.5 3.5 NS 0.2 0.4 1.9 3.3 NS Yes NS
14 4-Methoxyacetophenone 4.8 1.0 1.7 2.9 NS 0.0 0.0 2.6 4.5 NS 0.3 0.6 0.0 0.0 NS Yes NS
15 Glycerol 9.7 5.8 0.3 0.5 NS 4.0 2.3 0.0 0.0 NS 0.1 0.2 0.0 0.0 NS Yes NS
Accuracy (%) 86.7 100 86.7
Reproducibility (%) 86.7
Open in a new tab

Sensitization classification, S sensitizer, NS non-sensitizer

Comparison of prediction rate with DPRA and ADRA

The predictive rates of Spectro-DPRA in the 54 substances (33 sensitizers, 21 non-sensitizers) were compared with the results of other in chemico skin sensitization evaluations such as DPRA and ADRA based on LLNA data (Table 5). In the Spectro-DPRA, 1,2-benzisothiazol-3(2H)-one, squaric acid, benzyl benzoate, and nonanoyl chloride out of 33 sensitizers were identified as non-sensitizers (false negative), and sulfanilamide and sulfanilic acid among 21 non-sensitizers were identified as sensitizers (false positive).

Table 5.

Evaluation of 54 chemicals for skin sensitization using different in chemico methods

NO Test chemical Cys dep Lys dep Spetro-DPRA prediction (cys > 10 or lys > 30) DPRA ADRA LLNA Result
Cys Lys Result NAC NAL Result
1 Diphenylcyclopropenone 93.9 19.7 S 98.8 [8] − 0.7 [8] S 99.8 [6] 99.6 [6] S S
2 Benzoyl peroxide 98.8 68.3 S 100.0 [8] 81.3 [8] S 100.0 [6] 100.0 [6] S S
3 p-Benzoquinone 99.9 95.4 S 99.0 [8] 91.0 [8] S 98.9 [6] 97.5 [6] S S
4 Glutaraldehyde solution 99.5 98.9 S 30.2 [8] 85.4 [8] S 49.2 [6] 96.2 [6] S S
5 Lauryl gallate 99.8 75.5 S 90.9 [8] 8.7 [8] S 99.6 [6] 33.8 [6] S S
6 Propyl gallate 99.7 80.6 S 59.9 [8] 26.6 [8] S 93.7 [6] 75.2 [6] S S
7 Trimellitic anhydride 0.0 91.3 S − 1.1 [8] 41.7 [8] S 1.0 [6] 1.6 [6] S S
8 Metol 95.4 97.5 S 100.0 [8] 44.7 [8] S 97.5 [6] 60.8 [6] S S
9 2-Hydroxyethyl acrylate 100.0 38.3 S 92.6 [8] 88.9 [8] S 100.0 [6] 81.5 [6] S S
10 Glyoxal solution 54.0 61.3 S 56.5 [8] 67.8 [8] S 33.3 [6] 21.3 [6] S S
11 2-Methyl-3(2H)-isothiazolone 33.1 1.0 S 97.9 [8] − 5.6 [8] S 100.0 [6] − 0.2 [6] S S
12 1,2-Benzisothiazol-3(2H)-one 0.0 11.9 NS 97.7 [8] 9.7 [8] S 100.0 [6] 0.1 [6] S S
13 Methyl 2-nonynoate 41.3 1.4 S 100.0 [8] 3.2 [8] S 99.6 [6] 10.2 [6] S S
14 trans-Cinnamaldehyde 26.9 15.4 S 70.6 [8] 43.2 [8] S 95.5 [6] 81.3 [6] S S
15 Phenylacetaldehyde 95.7 14.5 S 90.7 [8] 22.6 [8] S 99.8 [6] 99.2 [6] S S
16 2,4-Hepatadienal 98.0 93.0 S 97.3 [8] 23.9 [8] S 100.0 [6] 98.4 [6] S S
17 Squaricacid 4.2 21.7 NS 46.9 [8] 4.8 [8] S − 2.0 [6] 1.5 [6] NS S
18 trans-2-Hexen-1-al 34.8 9.0 S 97.9 [8] 3.6 [8] S 96.7 [6] 97.6 [6] S S
19 Resorcinol 74.5 0.4 S 1.6 [8] -0.8 [8] NS 10.0 [6] 64.1 [6] S S
20 Diethyl maleate 100.0 0.0 S 100.0 [8] 85.5 [8] S 98.9 [6] 18.7 [6] S S
21 2-Phenylpropionaldehyde 74.1 0.0 S 48.2 [8] 21.2 [8] S 70.6 [6] 8.5 [6] S S
22 Palmitoyl chloride 41.2 6.0 S 25.5 [8] 26.6 [8] S 17.1 [6] 100.0 [6] S S
23 α-Hexylcinnamaldehyde 14.5 0.7 S − 0.3 [8] − 1.6 [8] NS 0.1 [6] 1.8 [6] NS S
24 α-Amylcinnamaldehyde 100.0 5.6 S 0.6 [8] 3.9 [8] NS 2.2 [6] 6.0 [6] NS S
25 2,3-Butanedione 79.7 90.2 S 79 [8] 27.0 [8] S 100.0 [6] 73.1 [6] S S
26 Oxalic acid 24.2 7.5 S 0.9 [8] -0.9 [8] NS -4.3 [6] 4.5 [6] NS S
27 Benzyl benzoate 0.0 1.2 NS 0.2 [8] 3.0 [8] S 3.3 [6] 3.0 [6] S S
28 4-Allylanisole 16.7 1.8 S 20.6 [8] − 0.8 [8] S 66.0 [6] 10.9 [6] S S
29 Nonanoyl chloride 8.8 7.5 NS 18.2 [8] − 6.3 [8] NS 8.1 [6] 100.0 [6] S S
30 2,2,6,6-Tetramethyl-3,5-heptanedione 46.2 4.2 S 1.4 [8] 0.0 [8] NS 8.1 [6] 7.6 [6] S S
31 Ethylene glycol dimethacrylate 94.9 1.1 S 87.3 [8] 12.4 [8] S 100.0 [6] 24.3 [6] S S
32 Ethyl acrylate 99.2 6.3 S 96.4 [8] 5.0 [8] S 100.0 [6] 96.7 [6] S S
33 Hydroxycitronellal 67.2 2.9 S 17.5 [8] 6.5 [8] S 14.8 [6] 21.9 [6] S S
34 Hexane 0.0 14.0 NS − 0.4 [8] − 5.1 [8] NS 3.0 [6] 5.8 [6] NS NS
35 Diethyl phthalate 0.0 0.0 NS 0.8 [8] -0.7 [8] NS 4.8 [6] 6.8 [6] NS NS
36 Octanoic acid 1.9 0.0 NS − 1.0 [8] 0.9 [8] NS 1.8 [6] 5.0 [6] NS NS
37 4-Hydroxybenzoic acid 0.7 10.6 NS − 0.3 [8] 2.2 [8] NS 0.5 [6] 2.2 [6] NS NS
38 Methyl salicylate 1.6 0.0 NS 0.3 [8] 1.6 [8] NS − 0.7 [6] 9.1 [6] NS NS
39 Chlorobenzene 1.5 7.8 NS 0.4 [8] 1.3 [8] NS 2.1 [6] 4.6 [6] NS NS
40 1-Bromobutane 1.9 0.0 NS 13.8 [8] 1.2 [8] S 6.0 [6] 2.9 [6] NS NS
41 2-Acetylcyclohexanone 9.0 13.3 NS 18.2 [8] 12.5 [8] S 7.2 [6] 14.6 [6] S NS
42 Ethyl benzoylacetate 3.1 10.1 NS 2.3 [8] 1.9 [8] NS 3.6 [6] 5.4 [6] NS NS
43 Isopropanol 1.1 8.5 NS − 10.0 [8] 0.5 [8] NS − 0.2 [6] − 5.2 [6] NS NS
44 Propylene glycol 0.7 0.4 NS − 0.9 [8] 0.6 [8] NS − 5.0 [6] − 1.2 [6] NS NS
45 Sulfanilamide 6.2 93.8 S − 1.3 [8] 0.8 [8] NS − 4.4 [6] − 2.3 [6] NS NS
46 Isopropyl myristate 0.4 8.2 NS 0.8 [8] − 4.0 [8] NS − 0.9 [6] − 0.9 [6] NS NS
47 Methyl paraben 2.2 1.1 NS 3.6 [8] -0.4 [8] NS 2.6 [6] 3.7 [6] NS NS
48 Nonanoic acid 1.5 1.5 NS -3.7 [8] -9.6 [8] NS 0.9 [6] 3.8 [6] NS NS
49 Propyl paraben 0.0 4.8 NS 8.2 [8] -0.2 [8] NS -3.3 [6] 1.0 [6] NS NS
50 Salicylic acid 0.9 0.3 NS 3.5 [8] - S -2.3 [6] 1.5 [6] NS NS
51 Sulfanilic acid 2.5 80.3 S – – ND – – ND NS
52 Vanillin 2.4 17.4 NS 3.2 [8] -6.6 [8] NS − 0.2 [6] 62.7 [6] S NS
53 Coumarin 8.6 0.6 NS 1.0 [8] − 14.9 [8] NS 4.9 [6] 0.7 [6] NS NS
54 Vinylidene chloride 0.5 0.4 NS 2.4 [8] − 4.3 [8] NS 4.9 [6] 0.7 [6] NS NS
Sensitivity(%) 87.9 81.8 87.9
Specificity(%) 90.5 85.0 90.0
Accuracy(%) 88.9 83.0 88.7
Open in a new tab

Sensitization classification, S sensitizer, NS non-sensitizer, ND not detected

As a result, sensitivity, specificity, and accuracy were confirmed as 87.9%, 90.5%, and 88.9%, respectively. In DPRA and ADRA, the accuracy values were 83.0% and 88.7%, respectively, and the prediction rate was similar or lower than that of Spectro-DPRA.

Within- and between-laboratory reproducibility

Kappa statistics was conducted to evaluate the within- and between-laboratory reproducibility consistency at three laboratories (Table 6). Fleiss’ kappa is a measure of reliability used to determine the level of agreement between two or more raters when the assessment method is measured on a categorical scale [27, 28]. Kappa values ranging from 0.21–0.40, 0.41–0.60, 0.61–0.80, and 0.81–1.00 were classified as fair, moderate, good, and very good, respectively [29]. Kappa statistics for three reproducibility evaluations within laboratories were 0.722 for AP and KCL and 1.000 for KTR, indicating good and very good levels of agreement, respectively, which were statistically significant (p < 0.001). Based on the results, the reproducibility of each laboratory was confirmed. Between-laboratory validation Kappa statistics was 0.630, indicating a good agreement, and it was statistically significant (p < 0.001).

Table 6.

Within- and Between- Laboratory Reproducibility

Within laboratory No. of samples No. of trials Fleiss' Kappa 95% CI Z-score p-value
AP 10 3 0.722 (0.364, 1.080) 3.956  < 0.001
KTR 10 3 1.000 (0.642, 1.358) 5.477  < 0.001
KCL 10 3 0.722 (0.364, 1.080) 3.956  < 0.001
Between laboratory No. of samples No. of trials Fleiss' Kappa 95% CI Z-score p-value
15 3 0.630 (0.337, 0.922) 4.224  < 0.001
Open in a new tab

Discussion

A skin sensitization test is essential for evaluating the safety of chemicals used in cosmetics and pharmaceuticals industries [30]. For skin sensitization evaluation, the guinea pig maximization (GPMT) and Buehler occluded patch tests (Buehler test) are commonly used animal-based tests [31–33]. Recently, the local lymph node assay (LLNA) in mice was developed to replace such methods [18, 34, 35]. However, absolute non-animal methods such as in chemico (e.g., DPRA) [19] or in vitro (e.g., ARE-Nrf2 luciferase test, human cell line activation test) [36] assays have gathered increasing attention in recent years because of the regulatory environment [37–39].

DPRA evaluates protein reactivity as an initial molecular initiating event in the AOP of skin sensitization. The technique assesses the extent of covalent binding of sensitizing substances to skin proteins, responsible for inducing skin sensitization [17, 40, 41]. However, DPRA has a few drawbacks. It requires HPLC-specific data analysis methods [42]; thus, the results may vary between laboratories depending on the analytical conditions [19, 43], and it may be time-consuming when evaluating many samples. Spectro-DPRA is an improved method that was designed to cover these issues. It is designed to be repeated three times to improve signal/noise ratio and ensure the certainty of the results. In most cases, it tends to be completed with two repetitions due to the concurrence of repeated test results. Although Spectro-DPRA has standardized cut-off criteria for depletion assessment of cysteine and lysine peptides [11], additional validation studies are required for international acceptance.

Here, pre-validation studies were conducted according to the standard operating procedure using an optimized version of the Spectro-DPRA test. First, a pilot study for transferability and proficiency was performed. Three participating laboratories confirmed more than 80% accuracy. Notably, Imidazolidinyl urea showed inconsistent results. According to a previous study, the release capacity of IL-8, an inflammatory cytokine, was increased when the vehicle of Imidazolidinyl urea was DMSO. It was also found to affect the activation of p38 mitogen-activated protein kinases (MAPKs), which play an essential role in regulating inflammatory substances among the systems involved in intracellular signaling [44]. In this study, the inconsistency was improved when the vehicle was changed from saline to DMSO, and this aspect needs to be considered in further studies since vehicle change may affect the results. Further, approximate ranges for each peptide's depletion rate for 10 substances were established in this study for further validation (Table 2). A sensitizer is determined if one of the two peptides react with it producing colorimetric change. Hence, determining the features of the peptides that influence the positive determination of various substances is critical.

The agreement rate among three experiments conducted by each institution was 90% for AP, 100% for KTR, and 100% for KCL; 90% agreement was determined when compared with the criteria presented in the SOP. Moreover, between-reproducibility among the three laboratories revealed 86.7% concordance, and the degree of agreement with the SOP was confirmed as 86.7% for AP, 100% for KTR, and 86.7% for KCL. Two chemicals incorrectly predicted as non-sensitizers, nickel sulfate, and cinnamyl alcohol can be discussed as follows: Sensitizers containing metal cations such as nickel sulfate and cobalt chloride cannot be applied to Spectro-DPRA because they are known to react with proteins by different mechanisms other than covalent bonding. However, two substances (nickelsulfate and cinnamyl alcohol) showed false negatives. This finding may be attributed to the fact that they potentially react with skin proteins through distinctive mechanisms [45, 46].

Additionally, 54 substances were used to evaluate the potential of Spectro-DPRA for regulatory acceptance (using OECD test guidelines; compared to DPRA). Twenty-nine of 33 sensitizing substances were identified as sensitizers, and 19 of 21 non-sensitizing substances as non-sensitizers, confirming a sensitivity, specificity, and accuracy of 87.9%, 90.5%, and 88.9%, respectively. For some substances that showed false negative (α-hexylcinnamaldehyde, α-amylcinnamaldehyde, and oxalic acid) or false positive (2-acetylcyclohexanone) in DPRA and ADRA, Spectro-DPRA showed a consistent result with LLNA data. Overall, for 54 substances, the prediction model combining the classification cut-offs of the two peptide methods (cysteine and lysine) showed a high degree of sensitivity, specificity, PPV, NPV, and accuracy for skin sensitization reactions as compared with DPRA and ADRA (Table 5). Spectro-DPRA is an easy, rapid, and high-throughput screening method for the prediction of the skin sensitization potential of chemicals.

Spectro-DPRA has a few limitations. When analyzing a substance containing a thiol or amine group, false positives may result, and additional experiments are required for color interference. Nevertheless, based on the results from this study, Spectro-DPRA presented adequate predictive capacity compared with the existing techniques (DPRA and ADRA), providing a viable strategy to save time and cost. There have been many attempts to interpret the results through integrated testing strategies (ITS) [47], which combine several alternative tests to increase predictive capacity in safety evaluation, while avoiding animal-based methods. Especially defined approaches for skin sensitization in the 2021 OECD guideline 497 are being adopted to replace animal testing fully [48]. Likewise, the Spectro-DPRA test may be a competitive approach that can enter the ITS pipeline for skin sensitization testing. This technique may become a novel, internationally recognized method with further systematic validation.

Author contributions

All authors contributed to the study design and progress, together. Material preparation, study progress and data production were performed by S-AC, CEP, DHS, and SA. Data collection and analysis were performed by J-AS, MC, and B-HK. The first draft of the manuscript was written by J-AS and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This research was supported by grants (19182MFDS495 and 21182MFDS339) from the Ministry of Food and Drug Safety, Republic of Korea.

Declarations

Conflict of interests

The authors have no relevant financial or non-financial interests to disclose.

Ethical approval

This article does not contain any studies with human participants performed by any of the authors.

References

  • 1.Adam J, Pichler WJ, Yerly D. Delayed drug hypersensitivity: models of T-cell stimulation. Br J Clin Pharmacol. 2011;71:701–707. doi: 10.1111/j.1365-2125.2010.03764.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Kimber I, Basketter DA, Gerberick GF, Ryan CA, Dearman RJ. Chemical allergy: translating biology into hazard characterization. Toxicol Sci. 2011;120(Suppl 1):S238–S268. doi: 10.1093/toxsci/kfq346. [DOI] [PubMed] [Google Scholar]
  • 3.Adler S, Basketter D, Creton S, et al. Alternative (non-animal) methods for cosmetics testing: current status and future prospects-2010. Arch Toxicol. 2011;85:367–485. doi: 10.1007/s00204-011-0693-2. [DOI] [PubMed] [Google Scholar]
  • 4.EU Commission (2013) Communication from the Commission to the European Parliament and the Council. Brussels. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52013DC0135&from=fr
  • 5.Vogel R. Alternatives to the use of animals in safety testing as required by the EU-Cosmetics Directive 2009. Altex. 2009;26:223–226. doi: 10.14573/altex.2009.3.223. [DOI] [PubMed] [Google Scholar]
  • 6.Yamamoto Y, Tahara H, Usami R, et al. A novel in chemico method to detect skin sensitizers in highly diluted reaction conditions. J Appl Toxicol. 2015;35:1348–1360. doi: 10.1002/jat.3139. [DOI] [PubMed] [Google Scholar]
  • 7.Gerberick F, Aleksic M, Basketter D, et al. Chemical reactivity measurement and the predicitve identification of skin sensitisers. The report and recommendations of ECVAM Workshop 64. Altern Lab Anim. 2008;36:215–242. doi: 10.1177/026119290803600210. [DOI] [PubMed] [Google Scholar]
  • 8.Fujita M, Yamamoto Y, Tahara H, Kasahara T, Jimbo Y, Hioki T. Development of a prediction method for skin sensitization using novel cysteine and lysine derivatives. J Pharmacol Toxicol Methods. 2014;70:94–105. doi: 10.1016/j.vascn.2014.06.001. [DOI] [PubMed] [Google Scholar]
  • 9.Maxwell G, MacKay C, Cubberley R, et al. Applying the skin sensitisation adverse outcome pathway (AOP) to quantitative risk assessment. Toxicol Vitro. 2014;28:8–12. doi: 10.1016/j.tiv.2013.10.013. [DOI] [PubMed] [Google Scholar]
  • 10.Landesmann B, Mennecozzi M, Berggren E, Whelan M. Adverse outcome pathway-based screening strategies for an animal-free safety assessment of chemicals. Lab Anim. 2013;41:461–471. doi: 10.1177/026119291304100609. [DOI] [PubMed] [Google Scholar]
  • 11.OECD. OECD Guidelines for the Testing of Chemicals . Key event-based test guideline for in chemico skin senditisation assays addressing the adverse outcome pathway key event on covalent binding to proteins. Paris: Organization for Economic Co-Operation and Development; 2021. [Google Scholar]
  • 12.Ahlfors SR, Sterner O, Hansson C. Reactivity of contact allergenic haptens to amino acid residues in a model carrier peptide, and characterization of formed peptide-hapten adducts. Skin Pharmacol Appl Skin Physiol. 2013;16:59–68. doi: 10.1159/000068288. [DOI] [PubMed] [Google Scholar]
  • 13.Lalko JF, Kimber I, Gerberick GF, Foertsch LM, Api AM, Dearman RJ. The direct peptide reactivity assay: selectivity of chemical respiratory allergens. Toxicol Sci. 2012;129:421–431. doi: 10.1093/toxsci/kfs205. [DOI] [PubMed] [Google Scholar]
  • 14.Cho SA, Jeong YH, Kim JH, et al. Method for detecting the reactivity of chemicals towards peptides as an alternative test method for assessing skin sensitization potential. Toxicol Lett. 2014;225:185–191. doi: 10.1016/j.toxlet.2013.12.007. [DOI] [PubMed] [Google Scholar]
  • 15.Cho SA, An S, Park JH. High-throughput screening (HTS)-based spectrophotometric direct peptide reactivity assay (Spectro-DPRA) to predict human skin sensitization potential. Toxicol Lett. 2019;314:27–36. doi: 10.1016/j.toxlet.2019.07.014. [DOI] [PubMed] [Google Scholar]
  • 16.Jeong YH, An S, Shin K, Lee TR. Peptide reactivity assay using spectrophotometric method for high-throughput screening of skin sensitization potential of chemical haptens. Toxicol Vitro. 2013;27:264–271. doi: 10.1016/j.tiv.2012.08.032. [DOI] [PubMed] [Google Scholar]
  • 17.Udenfriend S, Stein S, Böhlen P, Dairman W, Leimgruber W, Weigele M. Fluorescamine: a reagent for assay of amino acids, peptides, proteins, and primary amines in the picomole range. Science. 1972;178:871–872. doi: 10.1126/science.178.4063.871. [DOI] [PubMed] [Google Scholar]
  • 18.Gerberick GF, Vassallo JD, Bailey RE, Chaney JG, Morrall SW, Lepoittevin JP. Development of a peptide reactivity assay for screening contact allergens. Toxicol Sci. 2004;81:332–343. doi: 10.1093/toxsci/kfh213. [DOI] [PubMed] [Google Scholar]
  • 19.OECD . OECD guidelines for the testing of chemicals, section 4. No. 429. Skin sensitization: local lymph node assay. Paris: Organization for Economic Co-Operation and Development; 2010. [Google Scholar]
  • 20.Gerberick GF, Vassallo JD, Foertsch LM, Price BB, Chaney JG, Lepoittevin JP. Quantification of chemical peptide reactivity for screening contact allergens: a classification tree model approach. Toxicol Sci. 2007;97:417–427. doi: 10.1093/toxsci/kfm064. [DOI] [PubMed] [Google Scholar]
  • 21.Wanibuchi S, Yamamoto Y, Sato A, Kasahara T, Fujita M. The amino acid derivative reactivity assay with fluorescence detection and its application to multi-constituent substances. J Toxicol Sci. 2019;44:821–832. doi: 10.2131/jts.44.821. [DOI] [PubMed] [Google Scholar]
  • 22.Hirota M, Ashikaga T, Kouzuki H. Development of an artificial neural network model for risk assessment of skin sensitization using human cell line activation test, direct peptide reactivity assay, KeratinoSens™ and in silico structure alert parameter and in silico structure alert parameter. J Appl Toxicol. 2018;38:514–526. doi: 10.1002/jat.3558. [DOI] [PubMed] [Google Scholar]
  • 23.Lambrechts N, Vanheel H, Nelissen I, et al. Assessment of chemical skin-sensitizing potency by an in vitro assay based on human dendritic cells. Toxicol Sci. 2010;116:122–129. doi: 10.1093/toxsci/kfq108. [DOI] [PubMed] [Google Scholar]
  • 24.Gerberick GF, Ryan CA, Kern PS, et al. Compilation of historical local lymph node data for evaluation of skin sensitization alternative methods. Dermatitis. 2005;16:157–202. doi: 10.1097/01206501-200512000-00002. [DOI] [PubMed] [Google Scholar]
  • 25.Loveless SE, Api AM, Crevel RW, et al. Potency values from the local lymph node assay: application to classification, labelling and risk assessment. Regul Toxicol Pharmacol. 2010;56:54–66. doi: 10.1016/j.yrtph.2009.08.016. [DOI] [PubMed] [Google Scholar]
  • 26.Cho SA, Choi M, Park SR, An S, Park JH. Application of Spectro-DPRA, KeratinoSens™ and h-CLAT to estimation of the skin sensitization potential of cosmetics ingredients. J Appl Toxicol. 2020;40:300–312. doi: 10.1002/jat.3904. [DOI] [PubMed] [Google Scholar]
  • 27.Fleiss JL. Measuring nominal scale agreement among many raters. Psychol Bull. 1971;76:378–382. doi: 10.1037/h0031619. [DOI] [Google Scholar]
  • 28.Fleiss JL, Levin B, Paik MC. Statistical methods for rates and proportions. 3. Hoboken: Wiley; 2003. [Google Scholar]
  • 29.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–174. doi: 10.2307/2529310. [DOI] [PubMed] [Google Scholar]
  • 30.Daniel AB, Strickland J, Allen D, et al. International regulatory requirements for skin sensitization testing. Regul Toxicol Pharmaco. 2018;95:52–65. doi: 10.1016/j.yrtph.2018.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Buehler EV. Delayed contact hypersensitivity in the guinea pig. Arch Dermatol. 1965;91:171–177. doi: 10.1001/archderm.1965.01600080079017. [DOI] [PubMed] [Google Scholar]
  • 32.Magnusson B. Identification of contact sensitizers by animal assay. Contact Dermat. 1980;6:46–50. doi: 10.1111/j.1600-0536.1980.tb03894.x. [DOI] [PubMed] [Google Scholar]
  • 33.OECD . OECD guidelines for the testing of chemicals, section 4 No. 406. Skin sensitisatioon Guinea pig maximisation test and buehler test. Paris: Organization for Economic Co-Operation and Development; 2021. [Google Scholar]
  • 34.Enoch SJ, Roberts DW. Predicting skin sensitization potency for Michael acceptors in the LLNA using quantum mechanics calculations. Chem Res Toxicol. 2013;26:767–774. doi: 10.1021/tx4000655. [DOI] [PubMed] [Google Scholar]
  • 35.Kimber I, Basketter DA, Berthold K, et al. Skin sensitization testing in potency and risk assessment. Toxicol Sci. 2001;59:198–208. doi: 10.1093/toxsci/59.2.198. [DOI] [PubMed] [Google Scholar]
  • 36.OECD . OECD guidelines for the testing of chemicals, section 4 No. 442D. In vitro skin sensitisation: ARE-Nrf2 luciferase test method. Paris: Organization for Economic Co-Operation and Development; 2018. [Google Scholar]
  • 37.OECD . OECD guidelines for the testing of chemicals, section 4 No. 442E. In vitro skin sensitisation assay addressing the key event on activation of dendritic cells on the adverse outcome pathway for skin sensitisation. Paris: Organization for Economic Co-Operation and Development; 2018. [Google Scholar]
  • 38.Basketter DA, Kimber I. Updating the skin sensitization in vitro data assessment paradigm in 2009. J Appl Toxicol. 2009;29:545–550. doi: 10.1002/jat.1443. [DOI] [PubMed] [Google Scholar]
  • 39.Emter R, Ellis G, Natsch A. Performance of a novel keratinocyte-based reporter cell line to screen skin sensitizers in vitro. Toxicol Appl Pharmacol. 2010;245:281–290. doi: 10.1016/j.taap.2010.03.009. [DOI] [PubMed] [Google Scholar]
  • 40.Reisinger K, Hoffmann S, Alépée N, et al. Systematic evaluation of non-animal test methods for skin sensitisation safety assessment. Toxicol Vitro. 2015;29:259–270. doi: 10.1016/j.tiv.2014.10.018. [DOI] [PubMed] [Google Scholar]
  • 41.Roberts DW. Is a combination of assays really needed for non-animal prediction of skin sensitization potential? Performance of the GARD™ (Genomic Allergen Rapid Detection) assay in comparison with OECD guideline assays alone and in combination. Regul Toxicol Pharmacol. 2018;98:155–160. doi: 10.1016/j.yrtph.2018.07.014. [DOI] [PubMed] [Google Scholar]
  • 42.Zhang F, Erskine T, Klapacz J, Settivari R, Marty S. A highly sensitive and selective high pressure liquid chromatography with tandem mass spectrometry (HPLC/MS-MS) method for the direct peptide reactivity assay (DPRA) J Pharmacol Toxicol Methods. 2018;94:1–15. doi: 10.1016/j.vascn.2018.07.004. [DOI] [PubMed] [Google Scholar]
  • 43.Gerberick GF, Ryan CA, Kern PS, et al. A chemical dataset for evaluation of alternative approaches to skin-sensitization testing. Contact Dermat. 2004;50:274–288. doi: 10.1111/j.0105-1873.2004.00290.x. [DOI] [PubMed] [Google Scholar]
  • 44.Mitjans M, Galbiati V, Lucchi L, et al. Use of IL-8 release and p38 MAPK activation in THP-1 cells to identify allergens and to assess their potency in vitro. Toxicol In vitro. 2010;24:1803–1809. doi: 10.1016/j.tiv.2010.06.001. [DOI] [PubMed] [Google Scholar]
  • 45.Basketter DA. Skin sensitization to cinnamic alcohol: The role of skin metabolism. Acta Derm Venereol. 1992;72:264–265. doi: 10.2340/0001555572264265. [DOI] [PubMed] [Google Scholar]
  • 46.Elahi EN, Wright Z, Hinselwood D, Hotchkiss SA, Basketter DA, Pease CK. Protein binding and metabolism influence the relative skin sensitization potential of cinnamic compounds. Chem Res Toxicol. 2004;17:301–310. doi: 10.1021/tx0341456. [DOI] [PubMed] [Google Scholar]
  • 47.Strickland J, Zang Q, Kleinstreuer N, et al. Integrated decision strategies for skin sensitization hazard. J Appl Toxicol. 2016;36:1150–1162. doi: 10.1002/jat.3281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.OECD . OECD guidelines for the testing of chemicals, section 4 No. 497. Guideline on defined approaches for skin sensitisation. Paris: Organization for Economic Co-Operation and Development OECD (OECD); 2021. [Google Scholar]

Articles from Toxicological Research are provided here courtesy of Korean Society of Toxicology

RESOURCES