Local lymph node assay (LLNA): Comparison of different protocols by testing skin-sensitizing epoxy resin system components

Regulatory Toxicology and Pharmacology 52 (2008) 290–298

Contents lists available at ScienceDirect

Regulatory Toxicology and Pharmacology journal homepage: www.elsevier.com/locate/yrtph

Local lymph node assay (LLNA): Comparison of different protocols by testing skin-sensitizing epoxy resin system components Armin O. Gamer a, Eberhard Nies b, Hans-Werner Vohr c,* a

BASF SE, Experimental Toxicology and Ecology, 67056 Ludwigshafen, Germany Institute for Occupational Safety and Health of the German Social Accident Insurance (BGIA), 53757 Sankt Augustin, Germany c Bayer HealthCare AG, Toxicology, Aprather Weg, 42096 Wuppertal, Germany b

a r t i c l e

i n f o

Article history: Received 13 May 2008 Available online 13 September 2008 Keywords: Local lymph node assay variants Quantification of lymph node stimulation Vehicle effect Potency ranking of sensitizers Epoxy resin system components

a b s t r a c t Thirteen epoxy resin system components were tested in the LLNA with regard to their sensitizing potency. Lymph node stimulation was quantified not only by measuring the incorporation of [3H]-thymidine into the ear lymph nodes but also the counts of cells recovered from these organs. Equivalent figures were obtained with both endpoints used for the evaluation of lymph node cell proliferation if the reference stimulation indices were adjusted. When dissolved in acetone, all test substances showed skin-sensitizing potential, mainly on the boundary between ‘‘strong” and ‘‘moderate” according to common potency evaluation schemes. Replacing acetone with acetone/olive oil (4:1) as a vehicle for four selected test items, resulted in considerably lower estimated concentrations for sensitization induction. The challenges in comparing the results obtained by different LLNA variations are discussed. Ó 2008 Elsevier Inc. All rights reserved.

1. Introduction The principle of the murine local lymph node assay (LLNA), which detects the sensitizing properties of test substances by measuring lymphocyte proliferation in the auricular lymph nodes, was published in 1989 (Kimber and Weisenberger, 1989). A report on the first collaborative validation study was released in 1991 (Basketter et al., 1991). In these initial studies the stimulation of the lymph nodes, i.e. cell proliferation, was measured by [3H]-thymidine incorporation. In 1999 the LLNA was accepted by the Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM, 1999; Dean et al., 2001) and in Europe by the European Centre for the Validation of Alternative Methods (ECVAM; Balls and Hellsten, 2000) as a valid alternative to guinea pig assays, although the need for further modifications was also noted. The method had already been incorporated into Skin Sensitisation Guideline No. 406 of the OECD (1992) and by the EPA in 1998 (OPPTS 870.2600, Skin Sensitization). However, it took another 10 years until the LLNA was accepted as a stand-alone test by the OECD Test Guideline 429 (OECD, 2002), and a revised EPA guideline was published in 2003 (OPPTS 870.2600, Skin Sensitization: EPA 712-C-03-197, March 2003). So far, some of the recommendations contained in the two guidelines or the ICCVAM report have been implemented rather * Corresponding author. Fax: +49 202 364137. E-mail addresses: [email protected] (A.O. Gamer), [email protected] (E. Nies), [email protected] (H.-W. Vohr). 0273-2300/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.yrtph.2008.08.018

inconsistently by different researchers, but have become increasingly prominent as a result of growing experience (Basketter et al., 1994; Loveless et al., 1996; Montelius et al., 1998). One of these points is the concern about a possible confounding impact of irritant properties of test materials or other non-specific activation of immune competent cells which may cause non-specific cell proliferation in the draining lymph node and thus – in the worst case – lead to false positive or ‘‘unexpected” results (Montelius et al., 1994, 1998; Basketter et al., 1994, 1998; Loveless et al., 1996; Vohr et al., 2000; Vohr and Ahr, 2005; Ulrich et al., 2001a; Kreiling et al., 2008). Limitations of all three tests available in regulatory toxicology for predicting sensitizing properties of compounds, i.e. the guinea-pig maximization test, the occluded patch test of Buehler and the LLNA, were reviewed by Basketter and Kimber (2007) in large part. A number of new endpoints and modifications of the LLNA have been suggested as a means of discriminating effectively between the irritant and sensitizing properties of test items (Homey et al., 1998; Ulrich et al., 2001a; Suda et al., 2001; Takeyoshi et al., 2001; Yamashita et al., 2005; Gerberick et al., 1996, 2002). In this context, several authors have introduced a flow cytometric evaluation of activated cells in the ear draining lymph nodes (Takeyoshi et al., 2001; Yamashita et al., 2005; Gerberick et al., 1996, 2002). Another important matter is the demand for the avoidance of aqueous formulations. They should not be used in the LLNA because they drip off the ears immediately after application. OECD Guideline 429 recommends, in order of preference, acetone/olive oil (4:1 v/v, ‘‘AOO”), dimethyl formamide, methyl ethyl ketone,

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298

propyleneglycol and dimethyl sulfoxide as suitable vehicles, adding that ‘‘others may be used if sufficient scientific rationale is provided”. An IPCS/WHO workshop concluded that the LLNA had never been validated for the use of aqueous formulations or mixtures, and the recommendation was to have the existing data re-evaluated in this respect by ICCVAM (van Loveren et al., 2008). Last but not least, the standard Local Lymph Node Assay has been challenged for the in vivo use of radioactive labeling of the draining lymph nodes. Many laboratories wanted to circumvent this cost-intensive method, which may also show high individual variations of [3H]-thymidine incorporation into the lymph node cells. On the other hand modifications of the LLNA described by different authors, which avoid the use of radioactive in vivo labeling, have been criticized by others as being not sensitive enough (Gerberick et al., 1992; van Och et al., 2000; Piccotti et al., 2006). However, it is obvious that for endpoints other than thymidine incorporation, their individual specific cut-off value or threshold (SI or EC value) has also to be determined based on the variance and the maximum stimulation index achievable with the method used (Vohr et al., 2000; Suda et al., 2001; Ehling et al., 2005a,b; Sakaguchi et al., in press). Due to their importance in occupational settings, the Institute for Occupational Safety and Health of the German Social Accident Insurance (BGIA) decided to investigate a set of epoxy resin system components by means of LLNA under GLP conditions, with the aim of ranking these substances according to their sensitizing potency. Epoxy resins are used for a variety of purposes, e.g. in the construction, metalworking, turbine, automotive and paint industries. Although the hardened, finished polymers are considered to be virtually inert, the handling of uncured resin system components is causing a growing number of cases of allergic contact eczema (Geier et al., 2003, 2004; Tavakoli, 2003). Epoxy resin systems are made up of an epoxy resin and a (mostly aminic) curing agent or hardener. In order to process epoxy resins of high viscosity, socalled reactive thinners are added, such as monoglycidyl ethers of phenols or glycidyl ethers based on mono- or difunctional aliphatic or cycloaliphatic alcohols. Unfortunately, the resins themselves as well as reactive thinners and curing components have a sensitizing effect. Hence, there is a need to substitute highly sensitizing epoxy resin system components with substances with lower sensitizing potency (Kalberlah, 2007). The European project ‘‘EPOXYCODE” (Terwoert and Spee, 2005) has suggested a ranking based predominantly on physico-chemical data and qualitative hazard categories. This approach necessarily neglects some important physiological principles. Studies of exposed workers have revealed that different components show different incidences of allergic contact dermatitis (e.g. Geier et al., 2004). However, they are not able to determine whether these differences are based on the varying sensitizing potency of the components or are affected by exposure. BGIA has urged the performing laboratories to employ different LLNA protocol variations when testing epoxy resin system components. Here we are presenting the main results of this project with a focus on the comparison of radioactive vs. non-radioactive methods. But we will also address the matter of the incorporation of skin irritation in the evaluation of the data. In addition, the influence of the vehicle on the results has been studied for selected test compounds. 2. Material and methods 2.1. Animals CBA/CaOlaHsd female mice were used throughout all studies described herein. Six to ten weeks old animals were purchased

291

from Harlan Winkelmann GmbH, Borchen, Germany. The individually housed animals were identified by cage cards. The animals were housed in Makrolon type I cages in fully airconditioned rooms in which a central air-conditioning system ensured a temperature in the range of 20–24 °C and a relative humidity in the range of 30–70%. The illumination period was as follows: 12 h light (6.00 a.m.–6.00 p.m.) and 12 h darkness (6.00 p.m.– 6.00 a.m.). Tap water and diet was provided ad libitum. The experiments were conducted in accordance with German animal welfare legislation in an AAALAC certified laboratory. 2.2. Chemicals Acetone pro analysi from Merck or Riedel de Haen, Germany was used as vehicle. Methyl-[3H]–thymidine TRA 120 was purchased from GE Healthcare, Germany. The tested epoxy resin system components (Table 1) were provided by the purchasers in the same quality as in ready-to-use preparations. In order to facilitate concentration selection, a pretest was carried out with each chemical using 3 mice and a standard concentration of 5%. Ear and lymph node weights were measured as indicators of ear skin irritation. Based on the results of the pretests the high concentration for each chemical was selected and two lower concentrations were tested spaced by a factor of 3 each. If an indication of ear skin irritation was observed, lower concentrations were used for the main studies. 2.3. Protocol Each component was tested in a separate study, which was carried out in accordance with OECD Guideline 429 under GLP conditions. In addition to the endpoints described in this guideline, lymph node cell counts as well as ear weight and thickness were determined. These latter parameters served as indicators for acute ear reactions which could lead to non-specific activation of draining lymph node cells (Homey et al., 1998; Vohr et al., 2000; Ulrich et al., 2001a,b). For each study, groups of 6 female CBA/Ca mice were treated with different concentrations of the test substances in acetone or with acetone alone (vehicle control). Because single animal evaluation was performed, six animals per test group were used, in order to ensure at least 5 valid results per test group to achieve sufficient statistical power. In additional studies acetone/olive oil (4:1) was used in order to compare the findings to those obtained with acetone. The studies used 3 test groups and 1 control group, each. Twenty five microliter per ear of the respective test substance preparation was applied to the dorsum of both ears for three consecutive days. The control group was treated solely with 25 lL per ear of the vehicle. The correctness of the concentrations and the stability of the test substance preparations were routinely checked by GC analyses in acetonitrile/deionized water mixtures (1:1 v/v). Three days after the last application, the mice were injected intravenously (i.v.) with 20 lCi of [3H]-thymidine in 250 lL of sterile saline into a tail vein. About 5 h after the [3H]-thymidine injection, the mice were sacrificed and the auricular lymph nodes on both sides were removed and the weight of each animal’s pooled lymph nodes was determined. Lymph node response was evaluated by measuring the cellular content and [3H]-thymidine incorporation into lymph node cells (indicators of cell proliferation). To this end, single cell suspensions were prepared from the pooled lymph nodes of each animal as soon as possible after dissection by carefully passing the lymph nodes through an iron mesh (mesh size 200 lm) into 6 mL of phosphate-buffered physiological saline. For determination of cell counts, an aliquot of each suspension was further diluted with CasyÒton in a ratio 1:500. The cell count was

292

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298

Table 1 Epoxy resin system components tested in the project CA Index-name

Synonym(s) used in the text

CAS No./supplier

1,6-Hexanediamine, C,C,C-trimethyl1,3-Propanediamine, N1-(3-aminopropyl) – (98%) 1,2-Cyclohexanediamine Cyclohexanemethanamine, 5-amino-1,3,3-trimethyl1,3-Benzenedimethanamine Ethanol, 2-[(2-aminoethyl)amino] – (99%) Oxirane, 2,20 -[1,6-hexanediylbis(oxymethylene)]bisOxirane, 2,20 -[[2-ethyl-2-[(2-oxiranylmethoxy)methyl]-1,3propanediyl]bis(oxymethylene)]bisOxirane, 2-[(C12-14-alkyloxy)methyl]-derivs. Oxirane, 2-[[4-(1,1-dimethylethyl)phenoxy]methyl] – (98%) Formaldehyde, polymer with 2-(chloromethyl)oxirane and phenol (molecular weight < 700) Oxirane, 2,20 -[(1-methylethylidene)bis(4,1-phenyleneoxymethylene)]bis(distilled) Phenol, 4,40 -(1-methylethylidene)bis-, polymer with 2-(chloromethyl)oxirane (molecular weight < 700)

Trimethyl-1,6-hexanediamine Dipropylenetriamine 1,2-Diaminocyclohexane Isophorone diamine m-Xylylenediamine Aminoethylethanolamine; AEEA 1,6-Hexanediol diglycidyl ether Trimethylolpropane triglycidyl ether C12/C14 Alkyl glycidyl ether p-tert-Butylphenyl glycidyl ether Bisphenol F diglycidyl ether

25620-58-0/MC-Bauchemie, Bottrop, Germany 56-18-8/Sigma-Aldrich Chemie GmbH, Germany 694-83-7/MC-Bauchemie, Bottrop, Germany 2855-13-2/MC-Bauchemie, Bottrop, Germany 1477-55-0/MC-Bauchemie, Bottrop, Germany 111-41-1/Sigma-Aldrich Chemie GmbH, Germany 16096-31-4/MC-Bauchemie, Bottrop, Germany 3454-29-3/MC-Bauchemie, Bottrop, Germany

Bisphenol A diglycidyl ether (distilled) Bisphenol A diglycidyl ether (technical grade)

determined using a CasyÒ-Counter. The remaining cell suspensions were washed twice with phosphate buffered saline (PBS) and precipitated with 5% trichloroacetic acid (TCA). Each precipitate was transferred to scintillation fluid and incorporation of [3H]-thymidine into the cells was measured in a b-scintillation counter. Moreover, the ear thickness was measured in each animal and thereafter a defined area with a diameter of 0.8 cm was punched out of the apical part of each ear. For each animal the weight of the pooled punches was determined. These measurements were performed to obtain an indication of possible acute skin reaction due to irritation. Since the test protocol basically followed the current OECD guideline, these determinations were done on day 6, and not on day 4 as proposed in the original publications describing additional ear thickness measurements to quantify ear skin irritation (Homey et al., 1998; Vohr et al., 2000; Ehling et al., 2005a,b). Nonetheless the ear thickness data from day 6 were expected to facilitate the interpretation of borderline lymph node responses. 2.4. Evaluation of results The parameters used to characterize the response were: lymph node cell counts, [3H]-thymidine incorporation into the lymph node cells and to a certain extent also lymph node weights. Because lymph node responses may be caused not only by induction of test substance specific immune responses but also by irritation of the ear skin by the test substance, the weight of ear punches taken from the area of test substance application and ear thickness were determined as parameters for inflammatory ear swelling, serving as an indicator of the irritant action of the test substance.1 A test substance was considered to display skin-sensitizing potential if increases of lymph node cell counts greater than 1.5-fold (stimulation index (SI) 1.5; Ehling et al., 2005b) and/or increases of [3H]-thymidine incorporation above 3-fold (SI3) of the concurrent control value were observed in the absence of significant ear skin irritation. Mean values exceeding these cut-off SIs are considered

1 Although, according to the OECD guideline, evaluation of LLNA results is based on the measurement of [3H]-thymidine incorporation into the lymph node cells in the first place, it is stated that ‘‘other endpoints for assessment of proliferation may be employed provided there is justification and appropriate scientific support”. Lymph node cell count is a more direct measurement of cell proliferation than determination of DNA synthesis and therefore considered an appropriate parameter for evaluation of cell proliferation in the assay. Although this modification is not yet acknowledged as fully validated (Basketter et al., 2008), clear evidence on its validity is present (e.g. Ehling et al., 2005a and b).

68609-97-2/MC-Bauchemie, Bottrop, Germany 3101-60-8/Sigma-Aldrich Chemie GmbH, Germany 9003-36-5/Hexion Specialty Chemicals GmbH, Duisburg, Germany 1675-54-3/Hexion Specialty Chemicals GmbH, Duisburg, Germany 25068-38-6/Hexion Specialty Chemicals GmbH, Duisburg, Germany

to indicate biologically relevant signs of lymph node cell proliferation. The skin-sensitizing potency of each test substance was characterized by the estimated concentration (EC) above which the cutoff level of the stimulation index for the respective endpoint was exceeded (skin sensitization threshold for lymph node cell count: SI = 1.5, and for or [3H]-thymidine incorporation: SI = 3). The EC values were estimated from the concentration response curves. They were calculated by linear regression between the data points directly below and above the SI, if possible, or using the two nearest points below or above the SI, with or without logarithmic transformation of the concentration. In addition, the evaluation was based on the following considerations:  If a test substance does not show a statistically significant and/or biologically relevant increase in cell counts or [3H]-thymidine incorporation as compared to the vehicle control in the presence of statistically significant and/or biologically relevant increased ear weights, ear thickness and/or lymph node weights as indication of skin irritation, it is considered not to be a sensitizer.  If at least one concentration tested causes a concentrationdependent statistically significant and/or biologically relevant increase in cell counts or [3H]-thymidine incorporation without being accompanied by a statistically significant and/or biologically relevant increase in ear weight, ear thickness or pronounced lymph node weights, the test substance is considered to be a sensitizer. Apart from these points, evaluation was performed according to the scheme published by Ehling et al. (2005b). The categorization of the relative skin sensitizing potency was performed according to Kimber et al. (2003) and ECETOC (2003). This system classifies the sensitization potency of a chemical as extreme (EC3 < 0.1), strong (EC3 P 0.1 to < 1), moderate (EC3 P 1 to < 10), and weak (EC3 P 10 to 6 100). For reasons explained in the ‘‘Discussion” section, we replaced the EC3 values by EC1.5 values for categorization purposes if the proliferation rate of the draining lymph node was determined by cell counting instead of [3H]-thymidine incorporation. 2.5. Statistical analyses Mean values and standard deviations of the measured parameters were calculated for the test and control groups from the individual values. The stimulation indices of cell counts, [3H]-thymidine

293

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298

incorporation, lymph node weights, ear weights and ear thickness measurements were calculated as the ratio of the test group mean values for these parameters divided by those of the concurrent vehicle control group. For cell counts, [3H]-thymidine incorporation, lymph node weights, ear weights and ear thickness, the WILCOXON test was applied.

to produce a three-fold or 1.5-fold increase in stimulation index, which were calculated using logarithmically transformed concentrations are generally somewhat lower than those calculated with linear data. However, in terms of biological variation, the differences are negligible. The EC values give rise to the following overall ranking of potency (from high to low):

3. Results As can be seen from Tables 2 and 3, all substances tested showed skin-sensitizing potential in the LLNA. The EC values, i.e. substance concentrations estimated from the dose response curve

1. Bisphenol A diglycidyl ether (technical grade and distilled) > 1,2-Diaminocyclohexane, m-xylylenediamine, C12/C14 alkyl glycidyl ether, p-tert-butylphenyl glycidyl ether and bisphenol F diglycidyl ether

Table 2 Study mean values of stimulation indices for cell count, [3H]-thymidine incorporation, lymph node weight, ear weights and ear thickness Vehicle/Test substance

Concentration (%)

LNW

LNCC

DPM

EW

ET

AOO/hexyl cinnamic aldehyde (HCA)

2.5 5 10 3 10 30 5 10 25 1 3 10 0.3 1 3 0.1 0.3 1 0.3 1 3 0.3 1 3 3 10 30 0.3 1 3 1 3 10 0.3 1 3 0.1 0.3 1 0.1 0.3 1 0.3 1 3 0.3 1 3

1.13 1.31** 1.64** 1.47* 1.98** 2.46** 1.29* 1.51** 1.78** 1.40** 1.78** 2.44** 1.07 1.28* 2.07** 0.97 1.05 1.63** 0.91 1.17* 2.13** 1.13 1.73** 3.39** 1.18* 1.16* 1.74** 0.94 1.18* 1.71** 1.17* 1.49** 2.14** 1.13 1.37* 2.37** 1.05 0.99 1.68** 0.94 1.23* 1.57** 1.88** 2.17** 2.95** 1.69** 2.27** 3.07**

1.13 1.30 1.83** 1.75* 2.36** 2.98** 1.50** 1.68** 2.13** 1.63** 2.18** 3.42** 1.16* 1.50* 2.50** 0.97 1.06 1.98** 1.01 1.38* 2.94** 1.37* 2.70** 5.73** 1.22* 1.30* 2.23** 0.92 1.21 2.01** 1.27* 1.95** 2.87** 1.24 1.73** 3.11** 1.04 1.14 2.25** 0.94 1.32* 1.80** 2.66** 3.85** 4.97** 2.22** 3.24** 4.29**

1.12 1.19 2.84** 4.56** 6.63** 9.86** 3.05** 4.78** 8.46** 2.15** 4.00** 8.86** 2.16** 3.17** 12.4** 1.19 1.81* 8.39** 1.17 2.68 20.1** 1.92* 9.09** 44.2** 2.00** 1.72* 6.60** 0.94 1.67* 4.65** 2.06* 6.52** 12.0** 2.35** 4.16** 22.7** 1.36* 1.68** 14.2** 1.02 2.37** 3.49** 10.5** 19.9** 39. 9** 5.58** 16.1** 28.1**

1.07 1.04 1.13* 1.00 1.03 1.22** 1.12** 1.36** 1.78** 1.04 1.06* 1.19** 1.01 1.02 1.07* 0.98 1.00 1.02 1.02 1.04 1.07* 1.02 1.06 1.11** 1.03 1.07* 1.67** 1.02 1.11** 1.06 1.03 1.03 1.11** 1.01 1.02 1.12** 1.01 1.01 0.99 1.04 1.06 1.09* 1.06** 1.07** 1.10** 1.13** 1.09* 1.14**

1.04 1.03 1.04 1.02 1.05* 1.15** 1.04 1.19** 1.43** 1.01 1.01 1.06** 1.02 1.06** 1.08** 1.01 1.04 1.03 1.03 1.05** 1.03* 1.02 1.01 1.02 1.00 1.02 1.16** 0.97 0.97 0.94 1.01 1.03 1.11** 1.02 1.02 1.06* 1.05* 1.04* 1.05* 0.99 1.00 1.02 1.04 1.05* 1.05* 1.02 1.00 1.05

Acetone/HCA

Pluronic L92/sodium dodecyl sulfate

Acetone/trimethyl-1,6-hexanediamine

Acetone/dipropylenetriamine

Acetone/1,2-diaminocyclohexane

Acetone/isophorone diamine

Acetone/m-xylylenediamine

Acetone/aminoethylethanolamine

Acetone/1,6-hexanediol diglycidyl ether

Acetone/trimethylolpropane triglycidyl ether

Acetone/C12/C14 alkyl glycidyl ether

Acetone/p-tert-butylphenyl glycidyl ether

Acetone/bisphenol F diglycidyl ether

Acetone/bisphenol A diglycidyl ether (distilled)

Acetone/bisphenol A diglycidyl ether (technical grade)

AOO, Acetone/olive oil (4:1). LNW, Lymph node weight. LNCC, Lymph node cell count. DPM, Decays per minute of [3H]-thymidine incorporated into the lymph node cells. EW, Ear weight. ET, Ear thickness. The lymph node assays were performed as described in the M&M section. Each mean is based on 6 single animals treated with the test substances. Stimulation indices (SI, ratio of test group result over control group result) were calculated for both protocols by setting the absolute numbers obtained with vehicle treated animals to 1.0. * Wilcoxon test of the absolute values significant p < 0.05%. ** Wilcoxon test of the absolute values significant p < 0.01%.

294

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298

Table 3 Tested concentrations and Estimated Concentrations (EC) of the skin-sensitizing threshold of epoxy resin components Substance name

Concentrations tested (w/w%)

Bisphenol A diglycidyl ether (distilled) Bisphenol A diglycidyl ether (technical grade) p-tert-Butylphenyl glycidyl ether 1,2-Diaminocyclohexane m-Xylylenediamine C12/14 Alkyl glycidyl ether Bisphenol F diglycidyl ether Dipropylenetriamine Isophorone diamine a-Hexyl cinnamic aldehyde/Acetone Trimethylolpropane triglycidyl ether 1,6-Hexanediol diglycidyl ether Trimethyl-1,6-hexanediamine a-Hexyl cinnamic aldehyde/AOO Aminoethylethanolamine

3, 1, 0.3 3, 1, 0.3 1, 0.3, 0.1 1, 0.3, 0.1 3, 1, 0.3 3, 1, 0.3 1, 0.3, 0.1 3, 1, 0.3 3, 1, 0.3 30, 10, 3 10, 3, 1 3, 1, 0.3 10, 3, 1 10, 5, 2.5 30, 10, 3

Estimated concentrations (EC%) 3

1.5

3lg

1.5lg

Class

– 0.1 0.4 0.4 0.4 0.6 0.7 0.9 1.0 – 1.4 1.9 1.9 10.5 15.2

– – 0.5 0.6 0.4 0.7 0.6 1.0 1.2 0.1 1.7 1.7 0.5 6.9 14.4

0.2 0.2 0.3 0.4 0.4 0.5 0.6 0.8 1.0 1.2 1.3 1.6 1.7 10.7 13.3

0.1 0.1 0.4 0.5 0.3 0.6 0.5 1.0 1.1 1.8 1.4 1.5 0.8 6.5 12.7

Strong Strong Strong Strong Strong Strong Strong Strong Strong Moderate Moderate Moderate Moderate Weak Weak

The lymph node assays were performed as described in the Materials and methods section. EC 3 and EC 1.5 values (estimated concentrations of the substance required to produce a 3-fold (1.5-fold) stimulation of draining lymph node cell proliferation compared with concurrent control) were calculated for both protocols by linear regression, and EClg was estimated by linear regression using a log transformation of the concentration. The test substances are listed in order of decreasing skin sensitizing potential using EC3lg and annotated with a potency class according to Kimber et al. (2003) based on this parameter. –, No meaningful calculation possible.

> Trimethyl-1,6-hexanediamine, dipropylenetriamine and isophorone diamine > 1,6-Hexanediol diglycidyl ether and trimethylolpropane triglycidyl ether > Aminoethylethanolamine. The differences in potencies are not very pronounced among the three groups of substances ranked between the most potent and least potent. Although many chemicals displayed ECs at the border of potency classes moderate and strong, the assignment of the chemicals matched well, independent of being derived from 3H-thymidine incorporation or cell count measurement. Only in three out of the 15 studies divergences were present: for isophorone diamine, which displays an EC3 of 1%, cell count would have ‘‘underclassified” the chemical as moderate (EC1.5 = 1.1%) instead of strong and trimethyl-1,6-hexanediamine as well as a-hexyl cinnamaldehyde in AOO would have been ‘‘overclassified” by cell count as strong instead of moderate or moderate instead of weak, respectively (see Table 3). From our additional data obtained for four of these substances (see above, underlined), which were applied in acetone and acetone/olive oil (AOO), respectively, in different laboratories and at different times, it can be assumed, however, that acetone as a vehicle might have contributed to a prediction of higher potencies and a narrower spacing of results. Table 4 illustrates the considerable differences caused by using another vehicle. While the LLNA results with acetone showed an acceptable interlaboratory reproducibility for the independently retested substances C12/C14 alkyl glycidyl ether, 1,2-diaminocy-

clohexane, bisphenol F diglycidyl ether and bisphenol A diglycidyl ether (technical grade), stimulation indices were clearly reduced when switching from acetone to AOO. Positive results in acetone for three of the four test substances became negative when the ‘‘standard” vehicle AOO was used at the same concentrations. As can be seen from Table 3, even for the positive standard HCA, the use of AOO as vehicle results in a much higher EC value compared to pure acetone as the vehicle. EC values for the standards determined in AOO were in the expected range as indicated in the Draft ICCVAM Performance Standards (ICCVAM, 2007). Non-specific activation by acute skin inflammation based on ear swelling and ear weights was not observed at any concentration applied here for the compounds retested in AOO, i.e. 1,2-diaminocyclohexane, C12/C14 alkyl glycidyl ether, bisphenol F diglycidyl ether and bisphenol A diglycidyl ether (technical grade). Further, Table 3 neatly indicates that equivalent results are obtained with both endpoints used for the evaluation of lymph node cell proliferation, namely lymph node cell counts and [3H]-thymidine incorporation into lymph node cells, if the reference stimulation index (SI = 1.5 instead of 3) is adjusted for the cell counts. The relationship between cell counts and [3H]-thymidine incorporation on an individual animal and study basis is presented in Fig. 1. 4. Discussion From a quantitative point of view, most of the compounds tested displayed skin-sensitizing potency at the boundary between ‘‘strong” and ‘‘moderate” according to a potency evaluation scheme presently under discussion (ECETOC, 2003; Kimber et al., 2003;

Table 4 Comparison of cell count stimulation indices of four epoxy resin components tested in two different vehicles Concentration tested

1,2-Diaminocyclohexane

C12/C14 Alkyl glycidyl ether

Bisphenol F diglycidyl ether

Bisphenol A diglycidyl ether (technical grade)

Acetone

AOO

Acetone

AOO

Acetone

AOO

Acetone

AOO

0.3% 1% 3% EC value Class#

0.96 1.60* 2.70* 0.89 Strong

1.02 0.85 1.29* nd Negative

0.82 1.27 1.91* 1.27 Moderate

0.84 0.87 0.97 nd Negative

1.11 1.28 1.61* 2.33 Moderate

1.07 1.26 1.43* (5.90) Moderate

1.22 1.47* 1.78* 1,19 Moderate

1.25 1.26 1.33 nd Negative

The lymph node assays were performed as described in the M&M section (without [3H]-thymidine treatment). The stimulation indices were calculated as mentioned there. The corresponding estimated concentrations (EC1.5 in (%)) based on these indices were determined by linear regression. nd, Not determined. * Statistically significantly different to vehicle treated animals with at least p < 0.05. # Classification of potency in accordance to ECETOC (2003).

295

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298 α-Hexyl cinnamic aldehyde, techn. 85% (AOO) 1,6-Hexanediol diglycidyl ether C12/C14 Alkyl glycidyl ether Trimethyl-1,6-hexanediamine Trimethylolpropane triglycidyl ether Dipropylenetriamine p-tert-Butylphenyl glycidyl ether Bisphenol A diglycidyl ether (distilled)

α-Hexyl cinnamic aldehyde, techn. 85% (Acetone) m-Xylylenediamine 1,2-Diaminocyclohexane Sodium dodecyl sulfate Isophorone diamine Aminoethylethanolamine Bisphenol F diglycidyl ether Bisphenol A diglycidyl ether (technical grade)

³H-Thymidine [DPM]

1E+05

1E+04

1E+03

1E+02 1E+06

1E+07

1E+08

Cell Counts/ Lymph Node Pair 3

Fig. 1. Comparison of lymph node cell count and [ H]-thymidine incorporation into lymph node cells of individual animals. The lymph node assays were performed as described in the Materials and methods section. Individual cell counts per LN pair are compared to corresponding data from radioactive labeling over all groups and studies are shown. Note the log-scale of the both axis of the diagram. The line represents a potency function fitted with Excel (DPM = 2  1013  CC2.2178; R2 = 0.826).

Schneider and Akkan, 2004). Bisphenol A diglycidyl ether (distilled and technical grade) appeared to possess a clearly higher and aminoethylethanolamine a clearly lower potency. This picture does not mirror a ranking list of sensitizers recently obtained by medical examinations of exposed workers in Germany (Geier et al., 2003, 2004), which generally proved the sensitizing potential of all chemicals tested but showed a more differentiated scale of potency. It is beyond the scope of this paper to discuss this discrepancy in detail, but it should be mentioned that, among other reasons, quantity and quality of exposure to single components in the occupational environment is an important factor. 4.1. Radioactive vs. non-radioactive variants From the very beginning, several modifications of the LLNA which introduced non-radioactive endpoints to the standard protocol were published. Some authors simply switched from radioactive labeling of lymph node cells with [3H]-thymidine to nonradioactive labeling with the mutagenic compound 5-bromo-20 deoxyuridine (BrdU) (Suda et al., 2001; Takeyoshi et al., 2001; Piccotti et al., 2006). It was a matter of considerable dispute, however, whether the latter method was similarly sensitive as the radioactive protocol and able to detect all skin sensitizers. Apart from BrdU incorporation, some authors have used ex vivo restimulation by T cell mitogen, cytokine expression (Suda et al., 2001; Ulrich et al., 2001b) or ATP determination as endpoints (Yamashita et al., 2005). In addition, others have sought to quantify draining lymph node cells using flow cytometry (Sikorski et al., 1996; Ulrich et al., 1998; Suda et al., 2001; Gerberick et al., 2002; Takeyoshi et al., 2001; Yamashita et al., 2005). However, only one of these alternative endpoints, lymph node cell count, has yet been evaluated thoroughly in the context of intra- (Vohr et al., 2000) and inter-laboratory trials (Ehling et al., 2005a,b). This subsequently performed validation study includes a comparison between [3H]-thymidine incorporation and cell counts to determine stimulation indices after application of inter-

nationally accepted standard chemicals. In addition, the acute irritant properties of the standards tested were determined by measuring ear thickness before and ear weight after the animals were sacrificed. The results confirmed similar sensitivity between both methods (as had already been observed by Gerberick et al., 1992; Ikarashi et al., 1993; Suda et al., 2001; Yamano et al., 2005), good correlation between participating laboratories, and identical classification of the compounds as compared to other methods or medical experience with humans. Here we are presenting additional data directly comparing both endpoints under ‘‘standard protocol” conditions as described in the OECD 429 Guideline. Normally, the radioactive method results in relatively high stimulation indices (SI) as well as high individual variances if measured on a single animal basis. In the light of this observation, the threshold for classifying a substance as a sensitizer had to be set to an increase in the stimulation index (i.e. ratio of radioactivity incorporated into the auricular lymph nodes of a group of treated mice to the radioactivity incorporated into the corresponding lymph nodes of a group of vehicle control mice) by a factor of 3. It is obvious that for each new endpoint the specific threshold value has to be determined on the basis of the individual variances and the maximum SI that can be expected with the method. The difference in response range of [3H]-thymidine measurement as compared to cell count can be readily observed in Fig. 1, in which the radioactivity measurements span 2 orders of magnitude, whereas cell count values span only 1. On the other hand, the individual variation is lower with cell counts as opposed to radioactive labeling. In our study, the mean coefficient of variation over the vehicle control groups was 40.1% (range 20.5–79.9%) for 3H-incorporation and 21.5% (range 11.1–30.9%) for cell counting. This shows that the used stimulation index (i.e. in this case, the ratio of cell number in the auricular lymph nodes of a group of treated mice to the cell number of the corresponding lymph nodes of a group of vehicle treated control mice) of 1.5 as the cut-off SI for cell count is appropriate (data available in the BGIA project report; Gamer and Landsiedel, 2006).

296

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298

Our results obtained by testing two classes of chemicals (epoxides and amines) under strict GLP conditions verify the above-mentioned findings from other authors proving that stimulation indices based on cell counts are as sensitive as those grounded on radioactive labeling provided that an appropriate cut-off stimulation index is established. Accordingly, the corresponding EC values are in a similar range regardless of the protocol used (cf. Table 3). It has to be emphasized that the reference stimulation index depends on two parameters, (i) the individual (single animal) variation of the endpoint measured, and (ii) the maximum stimulation index observable for the relevant endpoint of evaluation. In most substances, the EC3 values based on [3H]-thymidine incorporation closely match the EC1.5 values based on cell counts. The mean quotient in all studies of EC3 versus EC1.5 is 1.2, independent of concentration transformation, indicating the derivation of slightly higher EC1.5 values as compared to EC3 values. Considerably larger differences are observed in calculations using SI values exclusively beyond the respective reference SI. Thus, the mean ratio decreases to 0.9 or 1 if the only studies compared are those for which the EC is estimated with a data pair below and another measuring point above the reference SI. This shows that EC3/[3H]-thymidine incorporation and EC1.5/cell count are equivalent when the precision of the test method is taken into account. Besides avoiding radioactivity, determining cell counts as opposed thymidine incorporation has the accessory advantage that cell suspensions are readily amenable to further analysis with various methods (flow cytometry, chemiluminescence response and immunofluorescence) in order to gain insight into mechanistic events (Ikarashi et al., 1993; Vohr et al., 1994; Gerberick et al., 1996; Ulrich et al., 2001b; Yamashita et al., 2005). 4.2. Assessing the impact of irritation Ear skin irritation may modify the lymph node reaction and lead to false positive or ‘‘unexpected” results in the LLNA. For instance, the strong skin irritant sodium dodecyl sulfate (SDS) causes a positive outcome in the LLNA in 5% or higher concentrations in aqueous Pluronic L92 (a polymeric surfactant) as the vehicle (Table 2). Such findings have also been discussed by Basketter et al. (1998), who presented criteria for identification of false positive reactions. These criteria rely mainly on structure-activity considerations and on establishing a relationship between skin irritation potency and the lymph node reaction. In our opinion it would be more appropriate to integrate a measurement of irritation into the test method (Vohr et al., 2000). This might be ear thickness at a suitable time after the application period (day 4?) or ear weight at the end of the study. It has to be determined further which parameter is most adequate and how to systematically use the derived information in the evaluation of the LLNA. By comparing the specific immune reaction induced by the test item in the draining lymph nodes (LN cell counts/LN weights) to the immediate non-specific acute skin reaction (ear swelling/ear weight), it is possible to distinguish the irritant potential from the sensitizing potential of the compound tested. This second modification (measurement of ear swelling/ear weight) after treatment was shown to yield very useful additional information, preventing not only the overlooking of ‘‘false positives”, but also the faulty categorization of compounds with respect to their skin-sensitizing potency (Homey et al., 1998; Blotz et al., 2000; Ulrich et al., 2001a; Ehling et al., 2005a). Such modifications are also urged in the Note for Guidance SWP/2145/00 of the CPMP (2001) and OECD Guideline 429. Because this has already been discussed extensively in other publications, it is not the main emphasis of this paper. In our study, only aminoethylethanolamine (AEEA) exhibited clear signs of irritation of the ear skin at a concentration of 30%.

It had been reported to be skin-sensitizing in a guinea pig maximization test after intradermal induction with 5%, epidermal induction with 50% in distilled water, and challenge with 25% in distilled water, showing a response in 40% of the animals (Leung and Auletta, 1997). According to the categorization proposed in the ECETOC Technical Report No. 87, this makes it a weak sensitizer. On the other hand, an unpublished report (CTL, 2001) about an LLNA with 2.5%, 5%, 10% and 20% AEEA in acetone/olive oil (4:1) presents stimulation indices (SIs) of 2.2, 2.8, 6.3 and 14.7, resulting in an EC3 value of 5.3%, i.e. ‘‘moderate”. This is a clearly lower EC3 than that observed in the present study (13–15% depending on endpoint evaluated and method of calculation), but the AEEA study described in the unpublished report, which investigated a number of ethyleneamines, showed the lowest vehicle control value of [3H]-thymidine incorporation (range 197–386 dpm/lymph node). If the SIs had been calculated using the upper end of the vehicle control values, an EC3 of about 10% would have resulted. In conclusion, AEEA is a weak skin sensitizer and skin irritant. 4.3. Vehicle effect Another item that should be discussed in more detail is the impact of the solvent. The vehicle suggested first and foremost in the guideline and thus most commonly used in the LLNA is acetone/olive oil (AOO). Olive oil, however, is a complex mixture and rather poorly defined from a chemical point of view. The stability of a reactive test substance in olive oil is therefore not easy to predict and is difficult to monitor. Therefore, we originally chose acetone as a suitable solvent in the context of this study. As a consequence of unexpected results obtained with some of the epoxy resin system components, we investigated the possible influence of vehicles on the calculation of potency (Table 4). It was specifically the considerable sensitizing potency of C12/C14 alkyl glycidyl ether resulting from an LLNA using pure acetone as a vehicle that prompted us to repeat the experiment with AOO. The outcome of the acetone-based murine assay is neither supported by structure-activity considerations nor by human-medical experience. As expected, three of the four tested chemicals showed a much lower sensitizing activity when applied in AOO instead of pure acetone, even were not recognized as skin sensitizers when applied in the same concentration range. In this connection it has to be mentioned that for AOO higher test concentrations would have been chosen as a consequence of results of the pre-testing of irritant properties in the relevant vehicles. Hence, it could be expected that the four selected compounds – and probably all the other epoxy resin system components studied here – would turn out to be positive in AOO if an appropriate concentration range is employed in the LLNA. Nonetheless, since the test substances would have to be applied in higher concentrations, this would lead to higher EC values, which could result in divergent categorization and possibly different rankings of the test substances. It was also shown by others (Ikarashi et al., 1993; Montelius et al., 1994; Lea et al., 1999; Warbrick et al., 1999), that the choice of vehicle can result in different dose–effect relationships (for review, see McGarry, 2007). Given sufficient stability in the matrix, vehicle-dependent shifts would not change the sensitizing potential of a chemical substance, but may cause different classifications of a test compound with regard to its sensitizing potency, depending on the applied classification scheme. On the one hand, there are schemes that distinguish only a few categories (like the one proposed by the committee of experts on the transport of dangerous goods and on the Globally Harmonized System (GHS) of the classification and labeling of chemicals via OECD (http://www.unece. org/trans/danger/publi/ghs/ghs_rev02/English/03e_part3.pdf)) and, on the other hand, more highly differentiated systems based on more classes (as proposed by ECETOC, for example).

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298

Thus, the vehicle effect is not at all negligible and it may in certain cases severely hamper the potency ranking of sensitizers when LLNA data obtained from different laboratories are pooled. Furthermore, as potent sensitizers tend to be chemically reactive substances, interactions of test compounds with their vehicles have to be carefully investigated. Although our routinely applied GC analyses, which used acetonitrile/water as solvent, did not indicate any instability of the examined epoxy resin system components in acetone, subsequent IR and GC studies with the exclusion of water proved the quick formation of Schiff bases as a result of a reaction of the aminic test substances with acetone (data not shown). It remains unclear to what extent those adducts are hydrolyzed in a physiological environment and how they influence the potency estimates for the affected amines. 5. Conclusion All the tested epoxy resin system components showed skin-sensitizing potential when dissolved in acetone. Most of the test substances displayed skin-sensitizing potency on the boundary between strong and moderate according to potency evaluation schemes presently under discussion (ECETOC, 2003; Schneider and Akkan, 2004). Only two test substances seem to possess a clearly higher and one substance a clearly lower potency. The data provide evidence that equivalent results are obtained with both endpoints used for the evaluation of lymph node cell proliferation, i.e. lymph node cell counts and [3H]-thymidine incorporation into lymph node cells, if an adjusted reference stimulation index (1.5 instead of 3) is used for the cell counts, taking into account the quantitative differences of the concentration response curves. This corroborates the conclusion of Ehling et al. (2005a,b) that the lymph node cell count is an alternative and equally sensitive endpoint in the LLNA, with the advantage that no radioactivity has to be used for testing. When interpreting LLNA results, one has first of all to realize that the LLNA measures at best the induction of a local immune response, and not a challenge like guinea pig assays do. Second, a classification of chemicals according to their potency solely based on the relevant EC value without taking non-specific activation or vehicle effects into account could result in misleading classification. Therefore, risk assessment has to consider the fact that categorization is more reliable for the ranking of chemicals tested under comparable conditions within a single laboratory than when results between labs are compared, especially if different vehicles are used. All in all, several modifications of the radioactive protocol of the LLNA have been published so far, but only few of these have been examined in an international trial. Reasonable refinement of the standard protocol of the LLNA should be accepted in the future, as was the case for guinea pig assays previously, especially if such modifications open the door to obtaining valuable additional information from the scientific point of view and lead to a more reliable quantification of risk for regulatory purposes. Acknowledgments This study was financially supported by the German Social Accident Insurance (DGUV) and Verband der Chemischen Industrie (VCI, the German Association of the Chemical Industry). References Balls, M., Hellsten, E., 2000. Statement on the validity of the local lymph node assay for skin sensitisation testing. ECVAM Joint Research Centre, European Commission, Ispra, Italy. ATLA 28, 366–367. Basketter, D.A., Scholes, E.W., Kimber, I., Botham, P.A., Hilton, J., Miller, K., Robbins, M.C., Harrison, P.T.C., Waite, S.J., 1991. Interlaboratory evaluation of the local

297

lymph node assay with 25 chemicals and comparison with guinea pig test data. Toxicol. Methods 1, 30–43. Basketter, D.A., Scholes, E.W., Kimber, I., 1994. The performance of the local lymph node assay with chemicals identified as contact allergens in the human maximization test. Food Chem. Toxicol. 32, 543–547. Basketter, D.A., Gerberick, G.F., Kimber, I., 1998. Strategies for identifying false positive responses in predictive skin sensitization tests. Food Chem. Toxicol. 36, 327–333. Basketter, DA., Kimber, I., 2007. Information derived from sensitization test methods: test sensitivity, false positives and false negatives. Contact Dermatitis 56, 1–4. Basketter, D., Cockshott, A., Corsini, E., Gerberick, G.F., Idehara, K., Kimber, I., van Loveren, H., Takeyoshi, M., Matheson, J., Mehling, A., Omori, T., Rovida, C., Sozu, T., Casati, S., 2008. An evaluation of performance standards and non-radioactive endpoints for the local lymph node assay. ATLA 36, 1–5. Blotz, A., Michel, J., Moysan, A., Blümel, J., Dubertret, L., Ahr, H.J., Vohr, H.W., 2000. Analyses of cutaneous fluoroquinolones photoreactivity using the integrated model for the differentiation of skin reactions. J. Photochem. Photobiol. 58, 46– 53. CTL (Central Toxicology Laboratory), 2001. Report CTL/L/8918, unpublished report for the Dow Chemical Company. Dean, J.H., Twerdok, L.E., Tice, R.R., Sailstad, D.M., Hattan, D.G., Stokes, W.S., 2001. ICCVAM evaluation of the murine local lymph node assay. II. Conclusions and recommendations of an independent scientific peer review panel. Regul. Toxicol. Pharmacol. 34, 258–273. ECETOC (European Centre for Ecotoxicology and Toxicology of Chemicals), 2003. Contact sensitisation: classification according to potency. Technical Report No. 87, Brussels. Ehling, G., Hecht, M., Heusener, A., Huesler, J., Gamer, A.O., van Loveren, H., Maurer, Th., Riecke, K., Ullmann, L., Ulrich, P., Vandebriel, R., Vohr, H.W., 2005a. An European inter-laboratory validation of alternative endpoints of the murine local lymph node assay 1st round. Toxicology 212, 60–68. Ehling, G., Hecht, M., Heusener, A., Huesler, J., Gamer, A.O., van Loveren, H., Maurer, Th., Riecke, K., Ullmann, L., Ulrich, P., Vandebriel, R., Vohr, H.W., 2005b. An European inter-laboratory validation of alternative endpoints of the murine local lymph node assay 2nd round. Toxicology 212, 69–79. EPA (Environmental Protection Agency), 2003. Skin sensitization – Revised Final Health Effects Test Guideline. Federal Register, Vol. 68, No. 58, 14635–14636. Gamer, A.O., Landsiedel, R., 2006. Project report BGIA Project FP251, Sensitizing potency of epoxy resin components in the murine local lymph node assay (LLNA). Unpublished report. Geier, J., Lessmann, H., Hillen, U., Jappe, U., Dickel, H., Koch, P., Frosch, P.J., Schnuch, A., Uter, W., 2004. An attempt to improve diagnostics of contact allergy due to epoxy resin systems, EPOX 2002. First results of the multicentre study. Contact Dermatitis 51, 263–272. Geier, J., Uter, W., Lessmann, H., Hillen, U., Goergens, U., Kersting, K., Fuchs, T., Schnuch, A., 2003. Kontaktallergien gegen Epoxidharze – ein unterdiagnostiziertes Problem. Allergol. J. 12, 323–328 (in German). Gerberick, G.F., House, R.V., Fletcher, E.R., Ryan, C.A., 1992. Examination of the local lymph node assay for use in contact sensitization risk assessment. Fundam. Appl. Toxicol. 19, 438–445. Gerberick, G.F., Cruse, L.W., Ryan, C.A., Sikorsky, E.E., Miller, C.M., Ridder, G.M., 1996. Differential expression of T cell activation markers (CD62L, CD44, CD45RB, CD25) in draining lymph nodes of mice following exposure to allergens and irritants. Toxicologist 30, 95. Gerberick, G.F., Cruse, L.W., Ryan, C.A., Hulette, B.C., Chaney, J.G., Skinner, R.A., Dearman, R.J., Kimber, I., 2002. Use of a B cell marker (B220) to discriminate between allergens and irritants in the local lymph node assay. Toxicol. Sci. 68, 420–428. Homey, B., von Schilling, C., Blümel, J., Schuppe, H.C., Ruzicka, T., Ahr, H.J., Lehmann, P., Vohr, H.W., 1998. An integrated model for the differentiation of chemicalinduced allergic and irritant skin reactions. Toxicol. Appl. Pharmacol. 153, 83– 94. ICCVAM (Interagency Coordinating Committee on the Validation of Alternative Methods), 1999. The murine local lymph node assay: a test for assessing the allergic contact dermatitis potential of chemicals/compounds. The results of an independent peer review evaluation coordinated by the ICCVAM. NIH Publication No. 99-4494. Available from . ICCVAM (Interagency Coordinating Committee on the Validation of Alternative Methods), 2007. Draft ICCVAM performance standards for the murine local lymph node assay: methods for assessing lymphocyte proliferation. Available from . Ikarashi, Y., Tsuchiya, T., Nakamura, A., 1993. A sensitive mouse lymph node assay with two application phases for detection of contact allergens. Arch. Toxicol. 67, 629–636. Kalberlah, F., 2007. Vergleichende Bewertung von Epoxidharzen – Teil A: Entwicklung eines Rankingsystems für Epoxidharze. Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), Forschung Projekt F 2062, Dortmund/Berlin/Dresden (in German). Kimber, I., Weisenberger, C., 1989. A murine local lymph node assay for the identification of contact allergens. Assay development and result of an initial validation study. Arch. Toxicol. 63, 274–282. Kimber, I., Basketter, D.A., Butler, M., Gamer, A., Garrigue, J.L., Gerberick, G.F., Newsome, C., Steiling, W., Vohr, H.W., 2003. Classification of contact allergens according to potency: proposals. Food Chem. Toxicol. 41, 1799–1809.

298

A.O. Gamer et al. / Regulatory Toxicology and Pharmacology 52 (2008) 290–298

Kreiling, R., Hollnagel, H.M., Hareng, L., Eigler, D., Lee, M.S., Griem, P., Dreeßen, B., Kleber, M., Albrecht, A., Garcia, C., Wendel, A., 2008. Comparison of the skinsensitizing potential of unsaturated compounds as assessed by the murine local lymph node assay (LLNA) and the guinea pig maximization test (GPMT). Food Chem. Toxicol. 46, 1896–1904. Lea, L.J., Warbrick, E.V., Dearman, R.J., Kimber, I., Basketter, D.A., 1999. The impact of vehicle on assessment of relative skin sensitization potency of 1,4dihydroquinone in the local lymph node assay. Am. J. Contact Dermatitis 10, 213–218. Leung, H.W., Auletta, C.S., 1997. Evaluation of skin sensitization and cross-reaction of nine alkyleneamines in the guinea pig maximization test. J. Toxicol. Cutan. Ocul. Toxicol. 16, 189–195. Loveless, S.E., Ladics, G.S., Gerberick, G.F., Ryan, C.A., Basketter, D.A., Scholes, E.W., House, R.V., Dearman, R.J., Kimber, I., 1996. Further evaluation of the local lymph node assay in the final phase of an international collaborative trial. Toxicology 108, 141–152. McGarry, H.F., 2007. The murine local lymph node assay: Regulatory and potency considerations under REACH. Toxicology 238, 71–89. Montelius, J., Wahlkvist, H., Boman, A., 1994. Experience with the murine local lymph node assay: inability to discriminate between allergens and irritants. Acta Derm. Venereol. 74, 22–27. Montelius, J., Wahlkvist, H., Boman, A., Wahlberg, J.E., 1998. Murine local lymph node assay for predictive testing of allergenicity: two irritants caused significant proliferation. Acta Derm. Venereol. 78, 433–437. OECD (Organisation for Economic Co-operation and Development), 1992. OECD guideline for the testing of chemicals No. 406 (Skin sensitisation). Adopted 17 July 1992. OECD (Organisation for Economic Co-operation and Development), 1992. OECD guideline for the testing of chemicals No. 429 (Skin sensitisation: local lymph node assay). Adopted 24 April 2002. Piccotti, J.R., Knight, S.A., Gillhouse, K., Lagattuta, M.S., Bleavins, M.R.J., 2006. Evaluation of an ex vivo murine local lymph node assay: multiple endpoint comparison. Appl. Toxicol. 26, 333–340. Sakaguchi, H., Ashikaga, T., Miyazawa, M. Kosaka, N., Ito, Y., Yoneyama, K., Sono, S., Itagaki, H., Toyoda, H., Suzuki, H., in press. The relationship between CD86/CD54 expression and THP-1 cell viability in an in vitro skin sensitization test – human cell line activation test (h-CLAT). Cell Biol. Toxicol. Schneider, K., Akkan, Z., 2004. Quantitative relationship between the local lymph node assay and human skin sensitization assays. Regul.Toxicol. Pharmacol. 39, 245–255. Sikorski, E.E., Gerberick, G.F., Ryan, C.A., Miller, C.M., Ridder, G.M., 1996. Phenotypic analysis of lymphocyte subpopulations in lymph nodes draining the ear following exposure to contact allergens and irritants. Fundam. Appl. Toxicol. 34, 25–35. Suda, A., Yamashita, M., Tabei, M., Taguchi, K., Vohr, H.W., Tsutsui, N., Suzuki, R., Kikuchi, K., Sakaguchi, K., Mochizuki, K., Nakamura, K., 2001. Local lymph node assay with non-radioactive endpoints. J. Toxicol. Sci. 27, 205–218.

Takeyoshi, M., Yamasaki, K., Yakabe, Y., Takatsuki, M., Kimber, I., 2001. Development of non-radio isotopic endpoint of murine local lymph node assay based on 5bromo-20 -deoxyuridine (BrdU) incorporation. Toxicol. Lett. 119, 203–208. Tavakoli, S.M., 2003. An assessment of skin sensitisation by the use of epoxy resin in the construction industry. Research report 079. HSE Book. Terwoert, J., Spee, T., 2005. Management of allergic skin disease by epoxys in the construction industry; development of an international ‘‘code of practice”. Paper presented on the 6th International Scientific Conference of the International Occupational Hygiene Association (IOHA), 19–23 September 2005 Pilanesberg, South Africa. Ulrich, P., Homey, B., Vohr, H.W., 1998. A modified murine local lymph node assay for the differentiation of contact photoallergy from phototoxicity by analysis of cytokine expression in skin-draining lymph node cells. Toxicology 125, 149–168. Ulrich, P., Streich, J., Suter, W., 2001a. Validation of alternative endpoints in the murine local lymph node assay for the identification of contact allergic potential: primary ear skin irritation and ear-draining lymph node hyperplasia induced by topical chemicals. Arch. Toxicol. 74, 733–744. Ulrich, P., Grenet, O., Bluemel, J., Vohr, H.-W., Wiemann, C., Grundler, O., Suter, W., 2001b, 2002. Expression profiles during murine contact allergy: T helper 2 cytokines are expressed irrespective of the type of contact allergen. Arch. Toxicol. 75, 470–479 (Erratum in: Arch. Toxicol. 76, 62). van Loveren, H., Cockshott, A., Gebel, T., Gundert-Remy, U., de Jong, W.H., Matheson, J., McGarry, H., Musset, L., Selgrade, M.J.K., Vickers, C., 2008. Skin sensitization in chemical risk assessment: Report of a WHO/IPCS international workshop focusing on dose–response assessment. Regul. Toxicol. Pharmacol. 50, 155–199. van Och, F.M., Slob, W., de Jong, W.H., Vandebriel, R.J., van Loveren, H., 2000. A quantitative method for assessing the sensitizing potency of low molecular weight chemicals using a local lymph node assay: employment of a regression method that includes determination of the uncertainty margins. Toxicology 146, 49–59. Vohr, H.W., Homey, B., Schuppe, H., Kind, P., 1994. Photoreactions detected in a modified local lymph node assay in the mouse. Photoderm. Photoimmunol. Photomed. 10, 57–64. Vohr, H.W., Blümel, J., Blotz, A., Homey, B., Ahr, H.J., 2000. An intra-laboratory validation of IMDS: Discrimination between (photo)allergic and (photo)irritant skin reactions in mice. Arch. Toxicol. 73, 501–509. Vohr, H.W., Ahr, H.J., 2005. Sensitivity of the LLNA. Arch. Toxicol. 79, 721–728. Warbrick, E.V., Dearman, R.J., Basketter, D.A., Kimber, I., 1999. Influence of application vehicle on skin sensitization to ethylchloroisothiazolinone/ methylisothiazolinone: an analysis using the local lymph node assay. Contact Dermatitis 41, 325–329. Yamano, T., Shimizu, M., Noda, T., 2005. Quantitative comparison of the results obtained by the multiple-dose guinea pig maximization test and the nonradioactive murine local lymph-node assay for various biocides. Toxicology 211, 165–175. Yamashita, K., Idehara, K., Fukuda, N., Yamagishi, G., Kawada, N., 2005. Development of a modified local lymph node assay using ATP measurement as an endpoint. AATEX 11, 136–144.