Oecd/Ocde 442E: Key Event-Based Test Guideline

OECD/OCDE 442E
Adopted:
25 June 2018
KEY EVENT-BASED TEST GUIDELINE
IN VITRO SKIN SENSITISATION ASSAYS ADDRESSING THE KEY

EVENT ON ACTIVATION OF DENDRITIC CELLS ON THE
ADVERSE OUTCOME PATHWAY FOR SKIN SENSITISATION
GENERAL INTRODUCTION
Activation of dendritic cells Key Event based Test Guideline

1. A skin sensitiser refers to a substance that will lead to an allergic response following skin contact as
defined by the United Nations Globally Harmonized System of Classification and Labelling of Chemicals
(UN GHS) (1). There is general agreement on the key biological events underlying skin sensitisation. The
current knowledge of the chemical and biological mechanisms associated with skin sensitisation has been
summarised as an Adverse Outcome Pathway (AOP) (2), starting with the molecular initiating event
through intermediate events to the adverse effect, namely allergic contact dermatitis. In this instance, the
molecular initiating event (i.e. the first key event) is the covalent binding of electrophilic substances to
nucleophilic centres in skin proteins. The second key event in this AOP takes place in the keratinocytes
and includes inflammatory responses as well as changes in gene expression associated with specific cell
signalling pathways such as the antioxidant/electrophile response element (ARE)-dependent pathways. The
third key event is the activation of dendritic cells (DC), typically assessed by expression of specific cell
surface markers, chemokines and cytokines. The fourth key event is T-cell activation and proliferation,
which is indirectly assessed in the murine Local Lymph Node Assay (LLNA) (3).
2. This Test Guideline (TG) describes in vitro assays that address mechanisms described under the Key
Event on activation of dendritic cells of the AOP for skin sensitisation (2). The TG comprises test methods
to be used for supporting the discrimination between skin sensitisers and non-sensitisers in accordance
with the UN GHS (1).
The test methods described in this TG are:
- Human Cell Line Activation test (h-CLAT)
- U937 cell line activation Test (U-SENS™)
- Interleukin-8 Reporter Gene Assay (IL-8 Luc assay)
© OECD, (2018)
You are free to use this material subject to the terms and conditions available at http://www.oecd.org/termsandconditions/.
In accordance with the decision of the Council on a delegation of authority to amend Annex I of the decision of the council on the Mutual
Acceptance of Data in the assessment of chemicals [C(2018)49], this Guideline was amended by the OECD’s Joint Meeting of the Chemicals
Committee and the Working Party on Chemicals, Pesticides and Biotechnology by written procedure on 25 June 2018.
2│ 442E OECD/OCDE
3. The test methods included in this Test Guideline may differ in relation to the procedure used to
generate the data and the readouts measured but can be used indiscriminately to address countries’
requirements for test results on the Key Event on activation of dendritic cells of the AOP for skin
sensitisation while benefiting from the Mutual Acceptance of Data.
Background and principles of the test methods included in the Key Event based
Test Guideline
4. The assessment of skin sensitisation has typically involved the use of laboratory animals. The classical
methods that use guinea-pigs, the Guinea Pig Maximisation Test (GPMT) of Magnusson and Kligman, and
the Buehler Test (TG 406) (4), assess both the induction and elicitation phases of skin sensitisation. The
murine tests, the LLNA (TG 429) (3) and its two non-radioactive modifications, LLNA: DA (TG 442 A)
(5) and LLNA: BrdU-ELISA (TG 442 B) (6), all assess the induction response exclusively, and have also
gained acceptance, since they provide an advantage over the guinea pig tests in terms of animal welfare
together with an objective measurement of the induction phase of skin sensitisation.
5. Recently mechanistically-based in chemico and in vitro test methods addressing the first key event
(OECD TG 442C; Direct Peptide Reactivity Assay (7)), and second key event (OECD TG 442D; ARE-
Nrf2 Luciferase Test Method (8)) of the skin sensitisation AOP have been adopted for contributing to the
evaluation of the skin sensitisation hazard potential of chemicals.
6. Test methods described in this TG either quantify the change in the expression of cell surface marker(s)
associated with the process of activation of monocytes and DC following exposure to sensitisers (e.g.
CD54, CD86) or the changes in IL-8 expression, a cytokine associated with the activation of DC. Skin
sensitisers have been reported to induce the expression of cell membrane markers such as CD40, CD54,
CD80, CD83, and CD86 in addition to induction of proinflammatory cytokines, such as IL-1β and TNF-α,
and several chemokines including IL-8 (CXCL8) and CCL3 (9) (10) (11) (12), associated with DC
activation (2).
7. However, as DC activation represents only one key event of the skin sensitisation AOP (2) (13),
information generated with test methods measuring markers of DC activation alone may not be sufficient
to conclude on the presence or absence of skin sensitisation potential of chemicals. Therefore data
generated with the test methods described in this Test Guideline are proposed to support the discrimination
between skin sensitisers (i.e. UN GHS Category 1) and non-sensitisers when used within Integrated
Approaches to Testing and Assessment (IATA), together with other relevant complementary information,
e.g. derived from in vitro assays addressing other key events of the skin sensitisation AOP as well as non-
testing methods, including read-across from chemical analogues (13). Examples of the use of data
generated with these methods within Defined Approaches, i.e. approaches standardised both in relation to
the set of information sources used and in the procedure applied to the data to derive predictions, have been
published (13) and can be employed as useful elements within IATA.
8. The test methods described in this Test Guideline cannot be used on their own, neither to sub-categorise
skin sensitisers into subcategories 1A and 1B as defined by UN GHS (1), for authorities implementing
these two optional subcategories, nor to predict potency for safety assessment decisions. However,
depending on the regulatory framework, positive results generated with these methods may be used on
their own to classify a chemical into UN GHS category 1.
© OECD 2018
OECD/OCDE 442E │3
9. The term "test chemical" is used in this Test Guideline to refer to what is being tested1 and is not related
to the applicability of the test methods to the testing of mono-constituent substances, multi-constituent
substances and/or mixtures. Limited information is currently available on the applicability of the test
methods to multi-constituent substances/mixtures (14) (15). The test methods are nevertheless technically
applicable to the testing of multi-constituent substances and mixtures. When considering testing of
mixtures, difficult-to-test chemicals (e.g. unstable), or test chemicals not clearly within the
applicability domain described in this Guideline, upfront consideration should be given to whether
the results of such testing will yield results that are meaningful scientifically. Moreover, when
testing multi-constituent substances or mixtures, consideration should be given to possible interference of
cytotoxic constituents with the observed responses.
LITERATURE
1. United Nations UN (2015), Globally Harmonized System of Classification and Labelling of

Chemicals (GHS). Sixth revised edition. New York & Geneva: United Nations Publications. ISBN:
978-92-1-117006-1.
Available at: [https://www.unece.org/trans/danger/publi/ghs/ghs_rev06/06files_e.html].
2. OECD (2012), The Adverse Outcome Pathway for Skin Sensitisation Initiated by Covalent Binding to
Proteins. Part 1: Scientific Evidence. Series on Testing and Assessment No. 168. Available at:
http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=ENV/JM/MONO(2012)10/
PART1&docLanguage=En
3. OECD (2010), The Local Lymph Node Assay. OECD Guideline for the Testing of Chemicals No.
429, Organisation for Economic Cooperation and Development, Paris. Available at:
http://www.oecd.org/env/testguidelines
4. OECD (1992), Skin Sensitisation. OECD Guideline for the Testing of Chemicals No. 406,
Organisation for Economic Cooperation and Development, Paris. Available at:
5. OECD (2010), Skin Sensitisation: Local Lymph Node Assay: DA. OECD Guideline for the Testing of
Chemicals No. 442A, Organisation for Economic Cooperation and Development, Paris. Available at:
6. OECD (2010), Skin sensitisation: Local Lymph Node Assay: BrdU-ELISA. OECD Guideline for the
Testing of Chemicals No. 442B, Organisation for Economic Cooperation and Development, Paris.
Available at: http://www.oecd.org/env/testguidelines
7. OECD (2015), OECD Guideline for the Testing of Chemicals No. 442C: In Chemico Skin
Sensitisation: Direct Peptide Reactivity Assay (DPRA). Paris, France: Organisation for Economic
Cooperation and Development. Available at: http://www.oecd.org/env/testguidelines
1
In June 2013, the Joint Meeting agreed that where possible, a more consistent use of the
term "test chemical" describing what is being tested should be applied in new and updated Test
Guidelines.
© OECD 2018
4│ 442E OECD/OCDE
8. OECD (2015), OECD Guideline for the Testing of Chemicals No. 442D: In Vitro Skin Sensitisation:
ARE-Nrf2 Luciferase Test Method. Paris, France: Organisation for Economic Cooperation and
Development. Available at: http://www.oecd.org/env/testguidelines
9. Steinman RM. 1991. The dendritic cell system and its role in immunogenicity. Annu Rev Immunol
9:271-96.
10. Caux C, Vanbervliet B, Massacrier C, Azuma M, Okumura K, Lanier LL, and Banchereau J. 1994.
B70/B7-2 is identical to CD86 and is the major functional ligand for CD28 expressed on human
dendritic cells. J Exp Med 180:1841-7.
11. Aiba S, Terunuma A, Manome H, and Tagami H. 1997. Dendritic cells differently respond to haptens
and irritants by their production of cytokines and expression of co-stimulatory molecules. Eur J
Immunol 27:3031-8.
12. Aiba S, Manome H, Nakagawa S, Mollah ZU, Mizuashi M, Ohtani T, Yoshino Y, and Tagami. H.
2003. p38 mitogen-activated protein kinase and extracellular signal-regulated kinases play distinct
roles in the activation of dendritic cells by two representative haptens, NiCl2 and DNCB. J Invest
Dermatol 120:390-8.
13. OECD (2016). Series on Testing & Assessment No. 256: Guidance Document On The Reporting Of
Defined Approaches And Individual Information Sources To Be Used Within Integrated Approaches
To Testing And Assessment (IATA) For Skin Sensitisation, Annex 1 and Annex 2.
ENV/JM/HA(2016)29. Organisation for Economic Cooperation and Development, Paris. Available at:
[https://community.oecd.org/community/iatass].
14. Ashikaga T, Sakaguchi H, Sono S, Kosaka N, Ishikawa M, Nukada Y, Miyazawa M, Ito Y,
NishiyamaN, Itagaki H. (2010), A comparative evaluation of in vitro skin sensitisation tests: the
human cell-line activation test (h-CLAT) versus the local lymph node assay (LLNA). Altern. Lab.
Anim. 38, 275-284.
15. Piroird, C., Ovigne, J.M., Rousset, F., Martinozzi-Teissier, S., Gomes, C., Cotovio, J., Alépée, N.
(2015). The Myeloid U937 Skin Sensitization Test (U-SENS) addresses the activation of dendritic cell
event in the adverse outcome pathway for skin sensitization. Toxicol. In Vitro 29, 901-916.
© OECD 2018
OECD/OCDE 442E │5
ANNEX I: IN VITRO SKIN SENSITISATION: HUMAN CELL LINE

ACTIVATION TEST (H-CLAT)
INITIAL CONSIDERATIONS AND LIMITATIONS
1. The h-CLAT method quantifies changes in the expression of cell surface markers associated with
the process of activation of monocytes and dendritic cells (DC) (i.e. CD86 and CD54), in the human
monocytic leukaemia cell line THP-1, following exposure to sensitisers (1) (2). The measured expression
levels of CD86 and CD54 cell surface markers are then used for supporting the discrimination between
skin sensitisers and non-sensitisers.
2. The h-CLAT method has been evaluated in a European Union Reference Laboratory for
Alternatives to Animal Testing (EURL ECVAM)-coordinated validation study and subsequent
independent peer review by the EURL ECVAM Scientific Advisory Committee (ESAC). Considering all
available evidence and input from regulators and stakeholders, the h-CLAT was recommended by EURL
ECVAM (3) to be used as part of an IATA to support the discrimination between sensitisers and non-
sensitisers for the purpose of hazard classification and labelling. Examples of the use of h-CLAT data in
combination with other information are reported in the literature (4) (5) (6) (7) (8) (9) (10) (11).
3. The h-CLAT method proved to be transferable to laboratories experienced in cell culture

techniques and flow cytometry analysis. The level of reproducibility in predictions that can be expected
from the test method is in the order of 80% within and between laboratories (3) (12). Results generated in
the validation study (13) and other published studies (14) overall indicate that, compared with LLNA
results, the accuracy in distinguishing skin sensitisers (i.e. UN GHS Cat.1) from non-sensitisers is 85%
(N=142) with a sensitivity of 93% (94/101) and a specificity of 66% (27/41) (based on a re-analysis by
EURL ECVAM (12) considering all existing data and not considering negative results for chemicals with a
Log Kow greater than 3.5 as described in paragraph 4). False negative predictions with the h-CLAT are
more likely to concern chemicals showing a low to moderate skin sensitisation potency (i.e. UN GHS
subcategory 1B) than chemicals showing a high skin sensitisation potency (i.e. UN GHS subcategory 1A)
(4) (13) (15). Taken together, this information indicates the usefulness of the h-CLAT method to contribute
to the identification of skin sensitisation hazards. However, the accuracy values given here for the h-CLAT
as a stand-alone test method are only indicative, since the test method should be considered in combination
with other sources of information in the context of an IATA and in accordance with the provisions of
paragraphs 7 and 8 in the General introduction. Furthermore, when evaluating non-animal methods for skin
sensitisation, it should be kept in mind that the LLNA test as well as other animal tests may not fully reflect
the situation in humans.
4. On the basis of the data currently available, the h-CLAT method was shown to be applicable to test
chemicals covering a variety of organic functional groups, reaction mechanisms, skin sensitisation potency
(as determined in in vivo studies) and physicochemical properties (3) (14) (15). The h-CLAT method is
applicable to test chemicals soluble or that form a stable dispersion (i.e. a colloid or suspension in which
the test chemical does not settle or separate from the solvent/vehicle into different phases) in an
appropriate solvent/vehicle (see paragraph 14). Test chemicals with a Log Kow greater than 3.5 tend to
produce false negative results (14). Therefore negative results with test chemicals with a Log Kow greater
than 3.5 should not be considered. However, positive results obtained with test chemicals with a Log Kow
greater than 3.5 could still be used to support the identification of the test chemical as a skin sensitiser.
Furthermore, because of the limited metabolic capability of the cell line used (16) and because of the
© OECD 2018
6│ 442E OECD/OCDE
experimental conditions, pro-haptens (i.e. substances requiring enzymatic activation for example via P450
enzymes) and pre-haptens (i.e. substances activated by oxidation) in particular with a slow oxidation rate
may also provide negative results in the h-CLAT (15). Fluorescent test chemicals can be assessed with the
h-CLAT (17), nevertheless, strong fluorescent test chemicals emitting at the same wavelength as
fluorescein isothiocyanate (FITC) or as propidium iodide (PI), will interfere with the flow cytometric
detection and thus cannot be correctly evaluated using FITC-conjugated antibodies or PI. In such a case,
other fluorochrome-tagged antibodies or other cytotoxicity markers, respectively, can be used as long as it
can be shown they provide similar results as the FITC-tagged antibodies (see paragraph 24) or PI (see
paragraph 18) e.g. by testing the proficiency substances in Appendix II. In the light of the above, negative
results should be interpreted in the context of the stated limitations and together with other information
sources within the framework of IATA. In cases where there is evidence demonstrating the non-
applicability of the h-CLAT method to other specific categories of test chemicals, it should not be used for
those specific categories.
5. As described above, the h-CLAT method supports the discrimination between skin sensitisers from
non-sensitisers. However, it may also potentially contribute to the assessment of sensitising potency (4) (5)
(9) when used in integrated approaches such as IATA. Nevertheless, further work, preferably based on
human data, is required to determine how h-CLAT results may possibly inform potency assessment.
6. Definitions are provided in Appendix I.
PRINCIPLE OF THE TEST
7. The h-CLAT method is an in vitro assay that quantifies changes of cell surface marker expression
(i.e. CD86 and CD54) on a human monocytic leukemia cell line, THP-1 cells, following 24 hours exposure
to the test chemical. These surface molecules are typical markers of monocytic THP-1 activation and may
mimic DC activation, which plays a critical role in T-cell priming. The changes of surface marker
expression are measured by flow cytometry following cell staining with fluorochrome-tagged antibodies.
Cytotoxicity measurement is also conducted concurrently to assess whether upregulation of surface marker
expression occurs at sub-cytotoxic concentrations. The relative fluorescence intensity of surface markers
compared to solvent/vehicle control are calculated and used in the prediction model (see paragraph 26), to
support the discrimination between sensitisers and non-sensitisers
DEMONSTRATION OF PROFICIENCY
8. Prior to routine use of the test method described in this Annex to Test Guideline 442E, laboratories
should demonstrate technical proficiency, using the 10 Proficiency Substances listed in Appendix II.
Moreover, test method users should maintain an historical database of data generated with the reactivity
checks (see paragraph 11) and with the positive and solvent/vehicle controls (see paragraphs 20-22), and
use these data to confirm the reproducibility of the test method in their laboratory is maintained over time.
© OECD 2018
OECD/OCDE 442E │7
PROCEDURE
9. This test method is based on the h-CLAT DataBase service on ALternative Methods to animal
experimentation (DB-ALM) protocol no. 158 (18) which represents the protocol used for the EURL
ECVAM-coordinated validation study. It is recommended that this protocol is used when implementing
and using the h-CLAT method in the laboratory. The following is a description of the main components
and procedures for the h-CLAT method, which comprises two steps: dose finding assay and CD86/CD54
expression measurement.
Preparation of cells
10. The human monocytic leukaemia cell line, THP-1, should be used for performing the h-CLAT
method. It is recommended that cells (TIB-202™) are obtained from a well-qualified cell bank, such as the
American Type Culture Collection.
11. THP-1 cells are cultured, at 37°C under 5% CO2 and humidified atmosphere, in RPMI-1640 medium
supplemented with 10% foetal bovine serum (FBS), 0.05 mM 2-mercaptoethanol, 100 units/mL penicillin
and 100 µg/mL streptomycin. The use of penicillin and streptomycin in the culture medium can be
avoided. However, in such a case users should verify that the absence of antibiotics in the culture medium
has no impact on the results, for example by testing the proficiency substances listed in Appendix II. In any
case, in order to minimise the risk of contamination, good cell culture practices should be followed
independently of the presence or not of antibiotics in the cell culture medium. THP-1 cells are routinely
seeded every 2-3 days at the density of 0.1 to 0.2 × 106 cells/mL. They should be maintained at densities
from 0.1 to 1.0 × 106 cells/mL. Prior to using them for testing, the cells should be qualified by conducting a
reactivity check. The reactivity check of the cells should be performed using the positive controls, 2,4-
dinitrochlorobenzene (DNCB) (CAS n. 97-00-7, ≥ 99% purity) and nickel sulfate (NiSO4) (CAS n. 10101-
97-0, ≥ 99% purity) and the negative control, lactic acid (LA) (CAS n. 50-21-5, ≥ 85% purity), two weeks
after thawing. Both DNCB and NiSO4 should produce a positive response of both CD86 and CD54 cell
surface markers, and LA should produce a negative response of both CD86 and CD54 cell surface markers.
Only the cells which passed the reactivity check are to be used for the assay. Cells can be propagated up to
two months after thawing. Passage number should not exceed 30. The reactivity check should be
performed according to the procedures described in paragraphs 20-24.
12. For testing, THP-1 cells are seeded at a density of either 0.1 × 106 cells/mL or 0.2 × 106 cells/mL, and
pre-cultured in culture flasks for 72 hours or for 48 hours, respectively. It is important that the cell density
in the culture flask just after the pre-culture period be as consistent as possible in each experiment (by
using one of the two pre-culture conditions described above), because the cell density in the culture flask
just after pre-culture could affect the CD86/CD54 expression induced by allergens (19). On the day of
testing, cells harvested from culture flask are resuspended with fresh culture medium at 2 × 106 cells/mL.
Then, cells are distributed into a 24 well flat-bottom plate with 500 µL (1 × 106 cells/well) or a 96-well
flat-bottom plate with 80 µL (1.6 × 105 cells/well).
Dose finding assay
13. A dose finding assay is performed to determine the CV75, being the test chemical concentration that
results in 75% cell viability (CV) compared to the solvent/vehicle control. The CV75 value is used to
determine the concentration of test chemicals for the CD86/CD54 expression measurement (see paragraphs
20-24).
Preparation of test chemicals and control substances
© OECD 2018
8│ 442E OECD/OCDE
14. The test chemicals and control substances are prepared on the day of testing. For the h-CLAT method,
test chemicals are dissolved or stably dispersed (see also paragraph 4) in saline or medium as first solvent/vehicle
options or dimethyl sulfoxide (DMSO,  99% purity) as a second solvent/vehicle option if the test chemical is
not soluble or does not form a stable dispersion in the previous two solvents/vehicles, to final
concentrations of 100 mg/mL (in saline or medium) or 500 mg/mL (in DMSO). Other solvents/vehicles
than those described above may be used if sufficient scientific rationale is provided. Stability of the test
chemical in the final solvent/vehicle should be taken into account.
15. Starting from the 100 mg/mL (in saline or medium) or 500 mg/mL (in DMSO) stock solutions of the
test chemicals, the following dilution steps should be taken:
− For saline or medium as solvent/vehicle: Eight stock solutions (eight concentrations) are prepared,
by two-fold serial dilutions using the corresponding solvent/vehicle. These stock solutions are then
further diluted 50-fold into culture medium (working solutions). If the top final concentration in the
plate of 1000 µg/mL is non-toxic, the maximum concentration should be re-determined by
performing a new cytotoxicity test. The final concentration in the plate should not exceed 5000
µg/mL for test chemicals dissolved or stably dispersed in saline or medium.
− For DMSO as solvent/vehicle: Eight stock solutions (eight concentrations) are prepared, by two-fold
serial dilutions using the corresponding solvent/vehicle. These stock solutions are then further
diluted 250-fold into culture medium (working solutions).The final concentration in plate should not
exceed 1000 µg/mL even if this concentration is non-toxic.
The working solutions are finally used for exposure by adding an equal volume of working solution to the
volume of THP-1 cell suspension in the plate (see also paragraph 17) to achieve a further two-fold dilution
(usually, the final range of concentrations in the plate is 7.81–1000 µg/mL).
16. The solvent/vehicle control used in the h-CLAT method is culture medium (for test chemicals
solubilised or stably dispersed (see paragraph 4) either with medium or saline) or DMSO (for test
chemicals solubilised or stably dispersed in DMSO) tested at a single final concentration in the plate of
0.2%. It undergoes the same dilution as described for the working solutions in paragraph 15.
Application of test chemicals and control substances
17. The culture medium or working solutions described in paragraphs 15 and 16 are mixed 1:1 (v/v) with
the cell suspensions prepared in the 24-well or 96-well flat-bottom plate (see paragraph 12). The treated
plates are then incubated for 24±0.5 hours at 37°C under 5% CO2. Care should be taken to avoid
evaporation of volatile test chemicals and cross-contamination between wells by test chemicals, e.g. by
sealing the plate prior to the incubation with the test chemicals (20).
Propidium iodide (PI) staining
18. After 24±0.5 hours of exposure, cells are transferred into sample tubes and collected by
centrifugation. The supernatants are discarded and the remaining cells are resuspended with 200 µL (in
case of 96-well) or 600 µL (in case of 24-well) of a phosphate buffered saline containing 0.1% bovine
serum albumin (staining buffer). 200 µL of cell suspension is transferred into 96-well round-bottom plate
(in case of 96-well) or micro tube (in case of 24-well) and washed twice with 200 µL (in case of 96-well)
or 600 µL (in case of 24-well) of staining buffer. Finally, cells are resuspended in staining buffer (e.g. 400
µL) and PI solution (e.g. 20 µL) is added (for example, final concentration of PI is 0.625 µg/mL). Other
cytotoxicity markers, such as 7-Aminoactinomycin D (7-AAD), Trypan blue or others may be used if the
alternative stains can be shown to provide similar results as PI, for example by testing the proficiency
substances in Appendix II.
© OECD 2018
OECD/OCDE 442E │9
Cytotoxicity measurement by flow cytometry and estimation of CV75 value
19. The PI uptake is analysed using flow cytometry with the acquisition channel FL-3. A total of 10,000
living cells (PI negative) are acquired. The cell viability can be calculated using the following equation by
the cytometer analysis program. When the cell viability is low, up to 30,000 cells including dead cells
should be acquired. Alternatively, data can be acquired for one minute after the initiation of the analysis.
Number of living cells × 100

Cell Viability =
Total Number of acquired cells
The CV75 value (see paragraph 13), i.e. a concentration showing 75% of THP-1 cell survival (25%
cytotoxicity), is calculated by log-linear interpolation using the following equation:
(75 – c) × Log (b) – (75 – a) × Log (d)

Log CV75 =
a–c
Where:
a is the minimum value of cell viability over 75%

c is the maximum value of cell viability below 75%
b and d are the concentrations showing the value of cell viability a and c respectively
100
a
Cell viability (%)
50
0
b d
1 10 100 1000
Test dose (g/mL)
Other approaches to derive the CV75 can be used as long as it is demonstrated that this has no impact on
the results (e.g. by testing the proficiency substances).
© OECD 2018
10 │ 442E OECD/OCDE
CD86/CD54 expression measurement
Preparation of the test chemicals and control substances
20. The appropriate solvent/vehicle (saline, medium or DMSO; see paragraph 14) is used to dissolve or
stably disperse the test chemicals. The test chemicals are first diluted to the concentration corresponding to
100-fold (for saline or medium) or 500-fold (for DMSO) of the 1.2 × CV75 determined in the dose finding
assay (see paragraph 19). If the CV75 cannot be determined (i.e. if sufficient cytotoxicity is not observed
in the dose finding assay), the highest soluble or stably dispersed concentration of test chemical prepared
with each solvent/vehicle should be used as starting concentration. Please note that the final concentration
in the plate should not exceed 5000 µg/mL (in case of saline or medium) or 1000 µg/mL (in case of
DMSO). Then, 1.2-fold serial dilutions are made using the corresponding solvent/vehicle to obtain the
stock solutions (eight concentrations ranging from 100×1.2 × CV75 to 100×0.335 × CV75 (for saline or
medium) or from 500×1.2 × CV75 to 500×0.335 × CV75 (for DMSO)) to be tested in the h-CLAT method
(see DB-ALM protocol NO. 158 for an example of dosing scheme). The stock solutions are then further
diluted 50-fold (for saline or medium) or 250-fold (for DMSO) into the culture medium (working
solutions). These working solutions are finally used for exposure with a further final two-fold dilution
factor in the plate. If the results do not meet the acceptance criteria described in the paragraphs 29 and 30
regarding cell viability, the dose finding assay may be repeated to determine a more precise CV75. Please
note that only 24-well plates can be used for CD86/CD54 expression measurement.
21. The solvent/vehicle control is prepared as described in paragraph 16. The positive control used in the
h-CLAT method is DNCB (see paragraph 11), for which stock solutions are prepared in DMSO and diluted
as described for the stock solutions in paragraph 20. DNCB should be used as the positive control for
CD86/CD54 expression measurement at a final single concentration in the plate (typically 4.0 µg/mL). To
obtain a 4.0 µg/mL concentration of DNCB in the plate, a 2 mg/mL stock solution of DNCB in DMSO is
prepared and further diluted 250-fold with culture medium to a 8 µg/mL working solution. Alternatively,
the CV75 of DNCB, which is determined in each test facility, could be also used as the positive control
concentration. Other suitable positive controls may be used if historical data are available to derive
comparable run acceptance criteria. For positive controls, the final single concentration in the plate should
not exceed 5000 µg/mL (in case of saline or medium) or 1000 µg/mL (in case of DMSO). The run
acceptance criteria are the same as those described for the test chemical (see paragraph 29), except for the
last acceptance criterion since the positive control is tested at a single concentration.
22. For each test chemical and control substance, one experiment is needed to obtain a prediction. Each
experiment consists of at least two independent runs for CD86/CD54 expression measurement (see
paragraphs 26-28). Each independent run is performed on a different day or on the same day provided that
for each run: a) independent fresh stock solutions and working solutions of the test chemical and antibody
solutions are prepared and b) independently harvested cells are used (i.e. cells are collected from different
culture flasks); however, cells may come from the same passage. Test chemicals and control substances
prepared as working solutions (500 µL) are mixed with 500 µL of suspended cells (1x106 cells) at 1:1
ratio, and cells are incubated for 24±0.5 hours as described in paragraphs 20 and 21. In each run, a single
replicate for each concentration of the test chemical and control substance is sufficient because a prediction
is obtained from at least two independent runs.
© OECD 2018
OECD/OCDE 442E │ 11
Cell staining and analysis
23. After 24±0.5 hours of exposure, cells are transferred from 24 well plate into sample tubes, collected
by centrifugation and then washed twice with 1mL of staining buffer (if necessary, additional washing
steps may be done). After washing, cells are blocked with 600 µL of blocking solution (staining buffer
containing 0.01% (w/v) globulin (Cohn fraction II, III, Human: SIGMA, #G2388-10G)) and incubated at
4°C for 15 min. After blocking, cells are split in three aliquots of 180 µL into a 96-well round-bottom plate
or micro tube.
24. After centrifugation, cells are stained with 50 µL of FITC-labelled anti-CD86, anti-CD54 or mouse
IgG1 (isotype) antibodies at 4°C for 30 min. The antibodies described in the h-CLAT DB-ALM protocol
no. 158 (18) should be used by diluting 3:25 (v/v, for CD86 (BD-PharMingen, #555657; Clone: Fun-1)) or
3:50 (v/v, for CD54 (DAKO, #F7143; Clone: 6.5B5) and IgG1 (DAKO, #X0927)) with staining buffer.
These antibody dilution factors were defined by the test method developers as those providing the best
signal-to-noise ratio. Based on the experience of the test method developers, the fluorescence intensity of
the antibodies is usually consistent between different lots. However, users may consider titrating the
antibodies in their own laboratory's conditions to define the best concentrations for use. Other
fluorochrome-tagged anti-CD86 and/or anti-CD54 antibodies may be used if they can be shown to provide
similar results as FITC-conjugated antibodies, for example by testing the proficiency substances in
Appendix II. It should be noted that changing the clone or supplier of the antibodies as described in the h-
CLAT DB-ALM protocol no. 158 (18) may affect the results. After washing twice or more with 150 µL of
staining buffer, cells are resuspended in staining buffer (e.g. 400 µL), and the PI solution (e.g. 20 µL to
obtain a final concentration of 0.625 µg/mL) or another cytotoxicity marker's solution (see paragraph 18) is
added. The expression levels of CD86 and CD54, and cell viability are analysed using flow cytometry.
DATA AND REPORTING
Data evaluation
25. The expression of CD86 and CD54 is analysed with flow cytometry with the acquisition channel FL-
1. Based on the geometric mean fluorescence intensity (MFI), the relative fluorescence intensity (RFI) of
CD86 and CD54 for positive control (ctrl) cells and chemical-treated cells are calculated according to the
following equation:
MFI of chemical-treated cells − MFI of chemical-treated isotype control

RFI = cells x100
MFI of solvent/vehicle-treated ctrl cells − MFI of solvent/vehicle-treated isotype ctrl

cells
The cell viability from the isotype control (ctrl) cells (which are stained with mouse IgG1 (isotype)
antibodies) is also calculated according to the equation described in paragraph 19.
Prediction model
26. For CD86/CD54 expression measurement, each test chemical is tested in at least two independent runs
to derive a single prediction (POSITIVE or NEGATIVE). An h-CLAT prediction is considered POSITIVE
if at least one of the following conditions is met in 2 of 2 or in at least 2 of 3 independent runs, otherwise
the h-CLAT prediction is considered NEGATIVE (Figure 1):
© OECD 2018
− The RFI of CD86 is equal to or greater than 150% in at least one tested concentration (with cell
viability ≥ 50%);
− The RFI of CD54 is equal to or greater than 200% in at least one tested concentration (with cell
viability ≥ 50%).
27. Based on the above, if the first two runs are both positive for CD86 and/or are both positive for CD54,
the h-CLAT prediction is considered POSITIVE and a third run does not need to be conducted. Similarly,
if the first two runs are negative for both markers, the h-CLAT prediction is considered NEGATIVE (with
due consideration of the provisions of paragraph 30) without the need for a third run. If however, the first
two runs are not concordant for at least one of the markers (CD54 or CD86), a third run is needed and the
final prediction will be based on the majority result of the three individual runs (i.e. 2 out of 3). In this
respect, it should be noted that if two independent runs are conducted and one is only positive for CD86
(hereinafter referred to as P1) and the other is only positive for CD54 (hereinafter referred to as P2), a third
run is required. If this third run is negative for both markers (hereinafter referred to as N), the h-CLAT
prediction is considered NEGATIVE. On the other hand, if the third run is positive for either marker (P 1 or
P2) or for both markers (hereinafter referred to as P12), the h-CLAT prediction is considered POSITIVE.
© OECD 2018
Figure 1: Prediction model used in the h-CLAT test method. An h-CLAT prediction should be considered in the
framework of an IATA and in accordance with the provision of paragraphs 7 and 8 in the General introduction. P1: run
with only CD86 positive; P2; run with only CD54 positive; P12: run with both CD86 and CD54 positive; N: run with
neither CD86 nor CD54 positive. *The boxes show the relevant combinations of results from the first two runs,
independently of the order in which they may be obtained. #The boxes show the relevant combinations of results from
the three runs on the basis of the results obtained in the first two runs shown in the box above, but do not reflect the
order in which they may be obtained.
28. For the test chemicals predicted as POSITIVE with the h-CLAT, optionally, two Effective
Concentrations (EC) values, the EC150 for CD86 and EC200 for CD54, i.e. the concentration at which the
test chemicals induced a RFI of 150 or 200, may be determined. These EC values potentially could
contribute to the assessment of sensitising potency (9) when used in integrated approaches such as IATA
(4) (5) (6) (7) (8). They can be calculated by the following equations:
EC150 (for CD86) = Bconcentration + [(150 - BRFI) / (ARFI - BRFI) × (Aconcentration - Bconcentration)]
where
Aconcentrationis the lowest concentration in µg/mL with RFI > 150 (CD86) or 200 (CD54)
© OECD 2018
Bconcentration is the highest concentration in µg/mL with RFI < 150 (CD86) or 200 (CD54)
ARFI is the RFI at the lowest concentration with RFI > 150 (CD86) or 200 (CD54)
BRFI is the RFI at the highest concentration with RFI < 150 (CD86) or 200 (CD54)
For the purpose of more precisely deriving the EC150 and EC200 values, three independent runs for
CD86/CD54 expression measurement may be required. The final EC150 and EC200 values are then
determined as the median value of the ECs calculated from the three independent runs. When only two of
three independent runs meet the criteria for positivity (see paragraphs 26-27), the higher EC150 or EC200
of the two calculated values is adopted.
Acceptance criteria
29. The following acceptance criteria should be met when using the h-CLAT method (22) (27).
- The cell viabilities of medium and solvent/vehicle controls should be higher than 90%.
- In the solvent/vehicle control, RFI values of both CD86 and CD54 should not exceed the positive
criteria (CD86 RFI  150% and CD54 RFI  200%). RFI values of the solvent/vehicle control are
calculated by using the formula described in paragraph 25 ("MFI of chemical" should be replaced
with "MFI of solvent/vehicle", and "MFI of solvent/vehicle" should be replaced with "MFI of
(medium) control").
- For both medium and solvent/vehicle controls, the MFI ratio of both CD86 and CD54 to isotype
control should be > 105%.
- In the positive control (DNCB), RFI values of both CD86 and CD54 should meet the positive criteria
(CD86 RFI  150 and CD54 RFI  200) and cell viability should be more than 50%.
- For the test chemical, the cell viability should be more than 50% in at least four tested concentrations
in each run.
30. Negative results are acceptable only for test chemicals exhibiting a cell viability of less than 90% at
the highest concentration tested (i.e. 1.2 × CV75 according to the serial dilution scheme described in
paragraph 20). If the cell viability at 1.2 × CV75 is equal or above 90% the negative result should be
discarded. In such a case it is recommended to try to refine the dose selection by repeating the CV75
determination. It should be noted that when 5000 µg/mL in saline (or medium or other solvents/vehicles),
1000 µg/mL in DMSO or the highest soluble concentration is used as the maximal test concentration of a
test chemical, a negative result is acceptable even if the cell viability is above 90%.
Test report
31. The test report should include the following information.
Test chemical
- Mono-constituent substance
 Chemical identification, such as IUPAC or CAS name(s), CAS number(s), SMILES or InChI
code, structural formula, and/or other identifiers;
 Physical appearance, Log Kow, water solubility, DMSO solubility, molecular weight, and
© OECD 2018
additional relevant physicochemical properties, to the extent available;

 Purity, chemical identity of impurities as appropriate and practically feasible, etc.;
 Treatment prior to testing, if applicable (e.g. warming, grinding);
 Concentration(s) tested;
 Storage conditions and stability to the extent available;
 Justification for choice of solvent/vehicle for each test chemical.
- Multi-constituent substance, UVCB and mixture
 Characterisation as far as possible by e.g. chemical identity (see above), purity, quantitative
occurrence and relevant physicochemical properties (see above) of the constituents, to the
extent available;
 Physical appearance, water solubility, DMSO solubility and additional relevant
physicochemical properties, to the extent available;
 Molecular weight or apparent molecular weight in case of mixtures/polymers of known
compositions or other information relevant for the conduct of the study;
Controls
- Positive control
 Physical appearance, Log Kow, water solubility, DMSO solubility, molecular weight, and
additional relevant physicochemical properties, to the extent available and where applicable;
 Reference to historical positive control results demonstrating suitable run acceptance criteria,
if applicable.
- Negative and solvent/vehicle control
© OECD 2018
 Physical appearance, molecular weight, and additional relevant physicochemical properties in
the case other control solvent/vehicle than those mentioned in the Test Guideline are used and
to the extent available;
Test method conditions
- Name and address of the sponsor, test facility and study director;
- Description of test method used;
- Cell line used, its storage conditions and source (e.g. the facility from which they were obtained);
- Flow cytometry used (e.g. model), including instrument settings, globulin, antibodies and cytotoxicity
marker used;
- The procedure used to demonstrate proficiency of the laboratory in performing the test method by
testing of proficiency substances, and the procedure used to demonstrate reproducible performance of
the test method over time, e.g. historical control data and/or historical reactivity checks’ data.
Test acceptance criteria
- Cell viability, MFI and RFI values obtained with the solvent/vehicle control in comparison to the
acceptance ranges;
- Cell viability and RFI values obtained with the positive control in comparison to the acceptance
ranges;
- Cell viability of all tested concentrations of the tested chemical.
Test procedure
- Number of runs used;

- Test chemical concentrations, application and exposure time used (if different than the one
recommended)
- Duration of exposure (if different than the one recommended);
- Description of evaluation and decision criteria used;
- Description of any modifications of the test procedure.
© OECD 2018
Results
- Tabulation of the data, including CV75 (if applicable), individual geometric MFI, RFI, cell viability
values, EC150/EC200 values (if applicable) obtained for the test chemical and for the positive control
in each run, and an indication of the rating of the test chemical according to the prediction model;
- Description of any other relevant observations, if applicable.
Discussion of the results
- Discussion of the results obtained with the h-CLAT method;

- Consideration of the test method results within the context of an IATA, if other relevant information
is available.
Conclusions
© OECD 2018
LITERATURE
1. Ashikaga T, Yoshida Y, Hirota M, Yoneyama K, Itagaki H, Sakaguchi H, Miyazawa M, Ito Y, Suzuki

H, Toyoda H. (2006), Development of an in vitro skin sensitization test using human cell lines: The
human Cell Line Activation Test (h-CLAT) I. Optimization of the h-CLAT protocol. Toxicol. In Vitro
20, 767–773.
2. Miyazawa M, Ito Y, Yoshida Y, Sakaguchi H, Suzuki H. (2007), Phenotypic alterations and cytokine
production in THP-1 cells in response to allergens. Toxicol. In Vitro 21, 428-437.
3. EC EURL-ECVAM (2013), Recommendation on the human Cell Line Activation Test (h-CLAT) for
skin sensitisation testing. Accessible at: https://eurl-ecvam.jrc.ec.europa.eu/eurl-ecvam-
recommendations
4. Takenouchi O, Fukui S, Okamoto K, Kurotani S, Imai N, Fujishiro M, Kyotani D, Kato Y, Kasahara T,

Fujita M, Toyoda A, Sekiya D, Watanabe S, Seto H, Hirota M, Ashikaga T, Miyazawa M. (2015), Test
battery with the human cell line activation test, direct peptide reactivity assay and DEREK based on a
139 chemical data set for predicting skin sensitizing potential and potency of chemicals. J Appl Toxicol.
35, 1318-1332.
5. Hirota M, Fukui S, Okamoto K, Kurotani S, Imai N, Fujishiro M, Kyotani D, Kato Y, Kasahara T, Fujita
M, Toyoda A, Sekiya D, Watanabe S, Seto H, Takenouchi O, Ashikaga T, Miyazawa M. (2015),
Evaluation of combinations of in vitro sensitization test descriptors for the artificial neural network-
based risk assessment model of skin sensitization. J Appl Toxicol. 35, 1333-1347.
6. Bauch C, Kolle SN, Ramirez T, Fabian E, Mehling A, Teubner W, van Ravenzwaay B, Landsiedel R.
(2012), Putting the parts together: combining in vitro methods to test for skin sensitizing potencials.
Regul Toxicol Parmacol. 63, 489-504.
7. Van der Veen JW, Rorije E, Emter R, Natch A, van Loveren H, Ezendam J. (2014), Evaluating the
performance of integrated approaches for hazard identification of skin sensitizing chemicals. Regul
Toxicol Pharmacol. 69, 371-379.
8. Urbisch D, Mehling A, Guth K, Ramirez T, Honarvar N, Kolle S, Landsiedel R, Jaworska J, Kern PS,
Gerberick F, Natsch A, Emter R, Ashikaga T, Miyazawa M, Sakaguchi H. (2015), Assessing skin
sensitization hazard in mice and men using non-animal test methods. Regul Toxicol Parmacol. 71, 337-
351.
9. Jaworska JS, Natsch A, Ryan C, Strickland J, Ashikaga T, Miyazawa M. (2015), Bayesian integrated
testing strategy (ITS) for skin sensitization potency assessment: a decision support system for
quantitative weight of evidence and adaptive testing strategy. Arch Toxicol. 89, 2355-2383.
10. Strickland J, Zang Q, Kleinstreuer N, Paris M, Lehmann DM, Choksi N, Matheson J, Jacobs A, Lowit
A, Allen D, Casey W. (2016), Integrated decision strategies for skin sensitization hazard. J Appl
Toxicol. DOI 10.1002/jat.3281.
11. Nukada Y, Ashikaga T, Miyazawa M, Hirota M, Sakaguchi H, Sasa H, Nishiyama N. (2012),

Prediction of skin sensitization potency of chemicals by human Cell Line Activation Test (h-CLAT)
and an attempt at classifying skin sensitization potency. Toxicol. In Vitro 26, 1150-60.
© OECD 2018
12. EC EURL ECVAM (2015), Re-analysis of the within and between laboratory reproducibility of the
human Cell Line Activation Test (h-CLAT). Accessible at: https://eurl-ecvam.jrc.ec.europa.eu/eurl-
ecvam-recommendations/eurl-ecvam-recommendation-on-the-human-cell-line-activation-test-h-clat-
for-skin-sensitisation-testing
13. EC EURL ECVAM (2012), human Cell Line Activation Test (h-CLAT) Validation Study Report
Accessible at: https://eurl-ecvam.jrc.ec.europa.eu/eurl-ecvam-recommendations
14. Takenouchi O, Miyazawa M, Saito K, Ashikaga T, Sakaguchi H. (2013), Predictive performance of the
human Cell Line Activation Test (h-CLAT) for lipophilic with high octanol-water partition coefficients.
J. Toxicol. Sci. 38, 599-609.

NishiyamaN, Itagaki H. (2010), A comparative evaluation of in vitro skin sensitisation tests: the
human cell-line activation test (h-CLAT) versus the local lymph node assay (LLNA). Altern. Lab.
Anim. 38, 275-284.
16. Fabian E., Vogel D., Blatz V., Ramirez T., Kolle S., Eltze T., van Ravenzwaay B., Oesch F.,
Landsiedel R. (2013), Xenobiotic metabolizin enzyme activities in cells used for testing skin
sensitization in vitro. Arch Toxicol 87, 1683-1969.
17. Okamoto K, Kato Y, Kosaka N, Mizuno M, Inaba H, Sono S, Ashikaga T, Nakamura T, Okamoto Y,
Sakaguchi H, Kishi M, Kuwahara H, Ohno Y. (2010), The Japanese ring study of a human Cell Line
Activation Test (h-CLAT) for predicting skin sensitization potential (6th report): A study for
evaluating oxidative hair dye sensitization potential using h-CLAT. AATEX 15, 81-88.
18. DB-ALM (INVITTOX) (2014), Protocol 158: human Cell Line Activation Test (h-CLAT), 23pp.
Accessible at: http://ecvam-dbalm.jrc.ec.europa.eu/
19. Mizuno M, Yoshida M, Kodama T, Kosaka N, Okamato K, Sono S, Yamada T, Hasegawa S, Ashikaga
T, Kuwahara H, Sakaguchi H, Sato J, Ota N, Okamoto Y, Ohno Y. (2008), Effects of pre-culture
conditions on the human Cell Line Activation Test (h-CLAT) results; Results of the 4th Japanese inter-
laboratory study. AATEX 13, 70-82.
20. Sono S, Mizuno M, Kosaka N, Okamoto K, Kato Y, Inaba H, , Nakamura T, Kishi M, Kuwahara H,
Sakaguchi H, Okamoto Y, Ashikaga T, Ohno Y. (2010), The Japanese ring study of a human Cell Line
Activation Test (h-CLAT) for predicting skin sensitization potential (7th report): Evaluation of volatile,
poorly soluble fragrance materials. AATEX 15, 89-96.
21. OECD (2005), Guidance Document No.34 on “the Validation and International Acceptance of New or
Updated Test Methods for Hazard Assessment”. OECD Series on Testing and Assessment.
Organization for Economic Cooperation and Development, Paris, France, 2005, 96 pp.
22. OECD (2012), The Adverse Outcome Pathway for Skin Sensitisation Initiated by Covalent Binding to
Proteins. Part 1: Scientific Evidence. Series on Testing and Assessment No. 168. Available at:
http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=ENV/JM/MONO(2012)10/P
ART1&docLanguage=En
23. United Nations UN (2013), Globally Harmonized System of Classification and Labelling of Chemicals
(GHS). Fifth revised edition. New York & Geneva: United Nations Publications. ISBN: 978-92-1-
© OECD 2018
117006-1. Available at: [http://www.unece.org/trans/danger/publi/ghs/ghs_rev05/05files_e.html].
24. ECETOC (2003). Contact sensitization: Classification according to potency. European Centre for
Ecotoxicology & Toxicology of Chemicals (Technical Report No. 87).
25. Ashikaga T, Sakaguchi H, Okamoto K, Mizuno M, Sato J, Yamada T, Yoshida M, Ota N, Hasegawa S,
Kodama T, Okamoto Y, Kuwahara H, Kosaka N, Sono S, Ohno Y. (2008), Assessment of the human
Cell Line Activation Test (h-CLAT) for Skin Sensitization; Results of the First Japanese Inter-
laboratory Study. AATEX 13, 27-35.
© OECD 2018
APPENDIX I
DEFINITIONS
Accuracy: The closeness of agreement between test method results and accepted reference values. It is a
measure of test method performance and one aspect of relevance. The term is often used interchangeably
with concordance to mean the proportion of correct outcomes of a test method (21).
AOP (Adverse Outcome Pathway): sequence of events from the chemical structure of a target chemical
or group of similar chemicals through the molecular initiating event to an in vivo outcome of interest (22).
CV75: The estimated concentration showing 75% cell viability.
EC150: the concentrations showing the RFI values of 150 in CD86 expression
EC200: the concentrations showing the RFI values of 200 in CD54 expression
Flow cytometry: a cytometric technique in which cells suspended in a fluid flow one at a time through a
focus of exciting light, which is scattered in patterns characteristic to the cells and their components; cells
are frequently labeled with fluorescent markers so that light is first absorbed and then emitted at altered
frequencies.
Hazard: Inherent property of an agent or situation having the potential to cause adverse effects when an
organism, system or (sub) population is exposed to that agent.
IATA (Integrated Approach to Testing and Assessment): A structured approach used for hazard
identification (potential), hazard characterisation (potency) and/or safety assessment (potential/potency and
exposure) of a chemical or group of chemicals, which strategically integrates and weights all relevant data
to inform regulatory decision regarding potential hazard and/or risk and/or the need for further targeted and
therefore minimal testing.
Medium control: An untreated replicate containing all components of a test system. This sample is
processed with test chemical-treated samples and other control samples to determine whether the
solvent/vehicle interacts with the test system.
Mixture: A mixture or a solution composed of two or more substances in which they do not react.
Mono-constituent substance: A substance, defined by its quantitative composition, in which one main
constituent is present to at least 80% (w/w).
Multi-constituent substance: A substance, defined by its quantitative composition, in which more than
one main constituent is present in a concentration ≥ 10% (w/w) and < 80% (w/w). A multi-constituent
substance is the result of a manufacturing process. The difference between mixture and multi-constituent
© OECD 2018
substance is that a mixture is obtained by blending of two or more substances without chemical reaction. A
multi-constituent substance is the result of a chemical reaction.
Positive control: A replicate containing all components of a test system and treated with a substance
known to induce a positive response. To ensure that variability in the positive control response across time
can be assessed, the magnitude of the positive response should not be excessive.
Pre-haptens: chemicals which become sensitisers through abiotic transformation
Pro-haptens: chemicals requiring enzymatic activation to exert skin sensitisation potential
Relative fluorescence intensity (RFI): Relative values of geometric mean fluorescence intensity (MFI) in
chemical-treated cells compared to MFI in solvent/vehicle-treated cells.
Relevance: Description of relationship of the test to the effect of interest and whether it is meaningful and
useful for a particular purpose. It is the extent to which the test correctly measures or predicts the
biological effect of interest. Relevance incorporates consideration of the accuracy (concordance) of a test
method (21).
Reliability: Measures of the extent that a test method can be performed reproducibly within and between
laboratories over time, when performed using the same protocol. It is assessed by calculating intra- and
inter-laboratory reproducibility and intra-laboratory repeatability (21).
Run: A run consists of one or more test chemicals tested concurrently with a solvent/vehicle control and
with a positive control.
Sensitivity: The proportion of all positive/active chemicals that are correctly classified by the test. It is a
measure of accuracy for a test method that produces categorical results, and is an important consideration
in assessing the relevance of a test method (21).
Staining buffer: A phosphate buffered saline containing 0.1% bovine serum albumin.
Solvent/vehicle control: An untreated sample containing all components of a test system except of the test
chemical, but including the solvent/vehicle that is used. It is used to establish the baseline response for the
samples treated with the test chemical dissolved or stably dispersed in the same solvent/vehicle. When
tested with a concurrent medium control, this sample also demonstrates whether the solvent/vehicle
interacts with the test system.
Specificity: The proportion of all negative/inactive chemicals that are correctly classified by the test. It is a
measure of accuracy for a test method that produces categorical results and is an important consideration in
assessing the relevance of a test method (21).
Substance: Chemical elements and their compounds in the natural state or obtained by any production
process, including any additive necessary to preserve the stability of the product and any impurities
© OECD 2018
deriving from the process used, but excluding any solvent which may be separated without affecting the
stability of the substance or changing it composition.
Test chemical: The term "test chemical" is used to refer to what is being tested.
United Nations Globally Harmonized System of Classification and Labelling of Chemicals (UN
GHS): A system proposing the classification of chemicals (substances and mixtures) according to
standardised types and levels of physical, health and environmental hazards, and addressing corresponding
communication elements, such as pictograms, signal words, hazard statements, precautionary statements
and safety data sheets, so that to convey information on their adverse effects with a view to protect people
(including employers, workers, transporters, consumers and emergency responders) and the environment
(23).
UVCB: substances of unknown or variable composition, complex reaction products or biological

materials.
Valid test method: A test method considered to have sufficient relevance and reliability for a specific
purpose and which is based on scientifically sound principles. A test method is never valid in an absolute
sense, but only in relation to a defined purpose (21).
© OECD 2018
APPENDIX II
PROFICIENCY SUBSTANCES
Prior to routine use of the test method described in this Annex to Test Guideline 442E, laboratories should
demonstrate technical proficiency by correctly obtaining the expected h-CLAT prediction for the 10
substances recommended in Table 1 and by obtaining CV75, EC150 and EC200 values that fall within the
respective reference range for at least 8 out of the 10 proficiency substances. Proficiency substances were
selected to represent the range of responses for skin sensitisation hazards. Other selection criteria were that
the substances are commercially available, and that high-quality in vivo reference data as well as high
quality in vitro data generated with the h-CLAT method are available. Also, published reference data are
available for the h-CLAT method (3) (14).
Table 1: Recommended substances for demonstrating technical proficiency with the h-CLAT method
CV75 h-CLAT results h-CLAT results

Physical In vivo Reference for CD86 for CD54
Proficiency substances CASRN
state prediction1 Range in (EC150 Reference (EC200 Reference
g/mL2 Range in μg/mL)2 Range in μg/mL)2
Sensitiser Positive Positive
2,4-Dinitrochlorobenzene 97-00-7 Solid 2-12
(extreme) (0.5-10) (0.5-15)
Sensitiser Positive Negative

4-Phenylenediamine 106-50-3 Solid 5-95
(strong) (<40) (>1.5)3

Nickel sulfate 10101-97-0 Solid 30-500
(moderate) (<100) (10-100)
Sensitiser Negative Positive

2-Mercaptbenzothiazole 149-30-4 Solid 30-400
(moderate) (>10)3 (10-140)
Sensitiser Negative Positive

R(+)-Limonene 5989-27-5 Liquid >20
(weak) (>5)3 (<250)
Imidazolidinyl urea 39236-46-9 Solid 25-100
(weak) (20-90) (20-75)
Negative Negative
Isopropanol 67-63-0 Liquid Non-sensitiser >5000
(>5000) (>5000)
Negative Negative
Glycerol 56-81-5 Liquid Non-sensitiser >5000
(>5000) (>5000)
Negative Negative
Lactic acid 50-21-5 Liquid Non-sensitiser 1500-5000
(>5000) (>5000)
Negative Negative
4-Aminobenzoic acid 150-13-0 Solid Non-sensitiser >1000
(>1000) (>1000)
Abbreviations: CAS RN = Chemical Abstracts Service Registry Number
1
The in vivo hazard and (potency) prediction is based on LLNA data (3) (14). The in vivo potency is
derived using the criteria proposed by ECETOC (24).
2
Based on historical observed values (13) (25).
3
Historically, a majority of negative results have been obtained for this marker and therefore a negative
result is mostly expected. The range provided was defined on the basis of the few historical positive
results observed. In case a positive result is obtained, the EC value should be within the reported
reference range.
© OECD 2018
ANNEX II: IN VITRO SKIN SENSITISATION:

U937 CELL LINE ACTIVATION TEST (U-SENS™)
1. The U-SENS™ method quantifies the change in the expression of a cell surface marker
associated with the process of activation of monocytes and dendritic cells (DC) (i.e. CD86), in the human
histiocytic lymphoma cell line U937, following exposure to sensitisers (1). The measured expression levels
of CD86 cell surface marker in the cell line U937 is then used for supporting the discrimination between
skin sensitisers and non-sensitisers.
2. The U-SENS™ method has been evaluated in a validation study (2) coordinated by L’Oreal and
subsequently independent peer reviewed by the European Union Reference Laboratory for Alternatives to
Animal Testing (EURL ECVAM) Scientific Advisory Committee (ESAC) (3). Considering all available
evidence and input from regulators and stakeholders, the U-SENS™ was recommended by EURL
ECVAM (4) to be used as part of an IATA to support the discrimination between sensitisers and non-
sensitisers for the purpose of hazard classification and labelling. In its guidance document on the reporting
of structured approaches to data integration and individual information sources used within IATA for skin
sensitisation, the OECD currently discusses a number of case studies describing different testing strategies
and prediction models. One of the different defined approaches is based on the U-SENS assay (5).
Examples of the use of U-SENS™ data in combination with other information, including historical data
and existing valid human data (6), are also reported elsewhere in the literature (4) (5) (7).
3. The U-SENS™ method proved to be transferable to laboratories experienced in cell culture

techniques and flow cytometry analysis. The level of reproducibility in predictions that can be expected
from the test method is in the order of 90% and 84% within and between laboratories, respectively (8).
Results generated in the validation study (8) and other published studies (1) overall indicate that, compared
with LLNA results, the accuracy in distinguishing skin sensitisers (i.e. UN GHS Cat.1) from non-
sensitisers is 86% (N=166) with a sensitivity of 91% (118/129) and a specificity of 65% (24/37).
Compared with human results, the accuracy in distinguishing skin sensitisers (i.e. UN GHS Cat.1) from
non-sensitisers is 77% (N=101) with a sensitivity of 100% (58/58) and a specificity of 47% (20/43). False
negative predictions compared to LLNA with the U-SENS™ are more likely to concern chemicals
showing a low to moderate skin sensitisation potency (i.e. UN GHS subcategory 1B) than chemicals
showing a high skin sensitisation potency (i.e. UN GHS subcategory 1A) (1) (8) (9). Taken together, this
information indicates the usefulness of the U-SENS™ method to contribute to the identification of skin
sensitisation hazards. However, the accuracy values given here for the U-SENS™ as a stand-alone test
method are only indicative, since the test method should be considered in combination with other sources of
information in the context of an IATA and in accordance with the provisions of paragraphs 7 and 8 in the
General introduction. Furthermore, when evaluating non-animal methods for skin sensitisation, it should be
kept in mind that the LLNA test as well as other animal tests may not fully reflect the situation in humans.
4. On the basis of the data currently available, the U-SENS™ method was shown to be applicable to
test chemicals (including cosmetics ingredients e.g. preservatives, surfactants, actives, dyes) covering a
variety of organic functional groups, of physicochemical properties, skin sensitisation potency (as
determined in in vivo studies) and the spectrum of reaction mechanisms known to be associated with skin
sensitisation (i.e. Michael acceptor, Schiff base formation, acyl transfer agent, substitution nucleophilic bi-
molecular [SN2], or nucleophilic aromatic substitution [SNAr]) (1) (8) (9) (10). The U-SENS™ method is
© OECD 2018
applicable to test chemicals that are soluble or that form a stable dispersion (i.e. a colloid or suspension in
which the test chemical does not settle or separate from the solvent/vehicle into different phases) in an
appropriate solvent/vehicle (see paragraph 13). Chemicals in the dataset reported to be pre-haptens (i.e.
substances activated by oxidation) or pro-haptens (i.e. substances requiring enzymatic activation for
example via P450 enzymes) were correctly predicted by the U-SENS™ (1) (10). Membrane disrupting
substances can lead to false positive results due to a non-specific increase of CD86 expression, as 3 out of
7 false positives relative to the in vivo reference classification were surfactants (1). As such positive results
with surfactants should be considered with caution whereas negative results with surfactants could still be
used to support the identification of the test chemical as a non-sensitiser. Fluorescent test chemicals can be
assessed with the U-SENS™ (1), nevertheless, strong fluorescent test chemicals emitting at the same
wavelength as fluorescein isothiocyanate (FITC) or as propidium iodide (PI), will interfere with the flow
cytometric detection and thus cannot be correctly evaluated using FITC-conjugated antibodies (potential
false negative) or PI (viability not measurable). In such a case, other fluorochrome-tagged antibodies or
other cytotoxicity markers, respectively, can be used as long as it can be shown they provide similar results
as the FITC-tagged antibodies or PI (see paragraph 18) e.g. by testing the proficiency substances in
Appendix II. In the light of the above, positive results with surfactants and negative results with strong
fluorescent test chemicals should be interpreted in the context of the stated limitations and together with
other information sources within the framework of IATA. In cases where there is evidence demonstrating
the non-applicability of the U-SENS™ method to other specific categories of test chemicals, it should not
be used for those specific categories.
5. As described above, the U-SENS™ method supports the discrimination between skin sensitisers
from non-sensitisers. However, it may also potentially contribute to the assessment of sensitising potency
when used in integrated approaches such as IATA. Nevertheless, further work, preferably based on human
data, is required to determine how U-SENS™ results may possibly inform potency assessment.
7. The U-SENS™ method is an in vitro assay that quantifies changes of CD86 cell surface marker
expression on a human histiocytic lymphoma cell line, U937 cells, following 45±3 hours exposure to the
test chemical. The CD86 surface marker is one typical marker of U937 activation. CD86 is known to be a
co-stimulatory molecule that may mimic monocytic activation, which plays a critical role in T-cell
priming. The changes of CD86 cell surface marker expression are measured by flow cytometry following
cell staining typically with fluorescein isothiocyanate (FITC)-labelled antibodies. Cytotoxicity
measurement is also conducted (e.g. by using PI) concurrently to assess whether upregulation of CD86 cell
surface marker expression occurs at sub-cytotoxic concentrations. The stimulation index (S.I.) of CD86
cell surface marker compared to solvent/vehicle control is calculated and used in the prediction model (see
paragraph 19), to support the discrimination between sensitisers and non-sensitisers.
8. Prior to routine use of the test method described in this Annex to Test Guideline 442E,
laboratories should demonstrate technical proficiency, using the 10 Proficiency Substances listed in
Appendix II in compliance with the Good in vitro Method Practices (11). Moreover, test method users
© OECD 2018
should maintain a historical database of data generated with the reactivity checks (see paragraph 11) and
with the positive and solvent/vehicle controls (see paragraphs 15-16), and use these data to confirm the
reproducibility of the test method in their laboratory is maintained over time.
PROCEDURE
9. This test method is based on the U-SENS™ DataBase service on ALternative Methods to animal
experimentation (DB-ALM) protocol no. 183 (12). The Standard Operating Procedures (SOP) should be
employed when implementing and using the U-SENS™ method in the laboratory. An automated system to
run the U-SENS™ can be used if it can be shown to provide similar results, for example by testing the
proficiency substances in Appendix II. The following is a description of the main components and
procedures for the U-SENS™ method.
10. The human histiocytic lymphoma cell line, U937 (13) should be used for performing the U-
SENS™ method. Cells (clone CRL1593.2) should be obtained from a well-qualified cell bank such as the
American Type Culture Collection.
11. U937 cells are cultured, at 37°C under 5% CO2 and humidified atmosphere, in RPMI-1640
medium supplemented with 10% foetal calf serum (FCS), 2 mM L-glutamine, 100 units/mL penicillin and
100 µg/mL streptomycin (complete medium). U937 cells are routinely passaged every 2-3 days at the
density of 1.5 or 3 × 105 cells/mL, respectively. The cell density should not exceed 2 × 106 cells/mL and
the cell viability measured by trypan blue exclusion should be ≥ 90% (not to be applied at the first passage
after thawing). Prior to using them for testing, every batch of cells, FCS or antibodies should be qualified
by conducting a reactivity check. The reactivity check of the cells should be performed using the positive
control, picrylsulfonic acid (2,4,6-Trinitro-benzene-sulfonic acid: TNBS) (CASRN 2508-19-2, ≥ 99%
purity) and the negative control lactic acid (LA) (CASRN 50-21-5, ≥ 85% purity), at least one week after
thawing. For the reactivity check, six final concentrations should be tested for each of the 2 controls
(TNBS: 1, 12.5, 25, 50, 75, 100µg/mL and LA: 1, 10, 20, 50, 100, 200µg/mL). TNBS solubilised in
complete medium should produce a positive and concentration-related response of CD86 (e.g. when a
positive concentration, CD86 S.I. ≥ 150, is followed by a concentration with an increasing CD86 S.I), and
LA solubilised in complete medium should produce negative response of CD86 (see paragraph 21). Only
the batch of cells which passed the reactivity check 2 times should be used for the assay. Cells can be
propagated up to seven weeks after thawing. Passage number should not exceed 21. The reactivity check
should be performed according to the procedures described in paragraphs 18-22.
12. For testing, U937 cells are seeded at a density of either 3 x 105 cells/mL or 6 × 105 cells/mL, and
pre-cultured in culture flasks for 2 days or 1 day, respectively. Other pre-cultured conditions than those
described above may be used if sufficient scientific rationale is provided and if it can be shown to provide
similar results, for example by testing the proficiency substances in Appendix II. In the day of testing, cells
harvested from culture flask are resuspended with fresh culture medium at 5 × 105 cells/mL. Then, cells are
distributed into a 96-well flat-bottom plate with 100 µL (final cell density of 0.5 × 105 cells/well).
Preparation of test chemicals and control substances
© OECD 2018
13. Assessment of solubility is conducted prior to testing. For this purpose, test chemicals are
dissolved or stably dispersed at a concentration of 50 mg/mL in complete medium as first solvent option or
dimethyl sulfoxide (DMSO,  99% purity) as a second solvent/vehicle option if the test chemical is not
soluble in the complete medium solvent/vehicle. For the testing, the test chemical is dissolved to a final
concentration of 0.4 mg/mL in complete medium if the chemical is soluble in this solvent/vehicle. If the
chemical is soluble only in DMSO, the chemical is dissolved at a concentration of 50 mg/mL. Other
solvents/vehicles than those described above may be used if sufficient scientific rationale is provided.
Stability of the test chemical in the final solvent/vehicle should be taken into account.
14. The test chemicals and control substances are prepared on the day of testing. Because a dose
finding assay is not conducted, for the first run, 6 final concentrations should be tested (1, 10, 20, 50, 100
and 200 µg/mL) into the corresponding solvent/vehicle either in complete medium or in 0.4% DMSO in
medium. For the subsequent runs, starting from the 0.4 mg/mL in complete medium or 50 mg/mL in
DMSO, solutions of the test chemicals, at least 4 working solutions (i.e. at least 4 concentrations), are
prepared using the corresponding solvent/vehicle. The working solutions are finally used for treatment by
adding an equal volume of U937 cell suspension (see paragraph 11 above) to the volume of working
solution in the plate to achieve a further 2-fold dilution (12). The concentrations (at least 4 concentrations)
for any further run are chosen based on the individual results of all previous runs (8). The usable final
concentrations are 1, 2, 3, 4, 5, 7.5, 10, 12.5, 15, 20, 25, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, 120, 140,
160, 180 and 200 µg/mL. The maximum final concentration is 200 µg/mL. In the case of a CD86 positive
value at 1 µg/mL is observed, then 0.1 µg/mL is evaluated in order to find the concentration of the test
chemical that does not induce CD86 above the positive threshold. For each run, the EC150 (concentration
at which a chemical reaches the CD86 positive threshold of 150%, see paragraph 19) is calculated if a
CD86 positive concentration-response is observed. Where the test chemical induces a positive CD86
response not concentration related, the calculation of the EC150 might not be relevant as described in the
U-SENS™ DB-ALM protocol no. 183 (12). For each run, CV70 (concentration at which a chemical
reaches the cytotoxicity threshold of 70%, see paragraph 19) is calculated whenever possible (12). To
investigate the concentration response effect of CD86 increase, any concentrations from the usable
concentrations should be chosen evenly spread between the EC150 (or the highest CD86 negative non
cytotoxic concentration) and the CV70 (or the highest concentration allowed i.e. 200 µg/mL). A minimum
of 4 concentrations should be tested per run with at least 2 concentrations being common with the previous
run(s), for comparison purposes.
15. The solvent/vehicle control used in the U-SENS™ method is complete medium (for test
chemicals solubilised or stably dispersed) (see paragraph 4) or 0.4% DMSO in complete medium (for test
chemicals solubilised or stably dispersed in DMSO).
16. The positive control used in the U-SENS™ method is TNBS (see paragraph 11), prepared in
complete medium. TNBS should be used as the positive control for CD86 expression measurement at a
final single concentration in plate (50 µg/mL) yielding > 70% of cell viability. To obtain a 50 µg/mL
concentration of TNBS in plate, a 1 M (i.e. 293 mg/mL) stock solution of TNBS in complete medium is
prepared and further diluted 2930-fold with complete medium to a 100 µg/mL working solution. Lactic
acid (LA, CAS 50-21-5) should be used as the negative control at 200 μg/mL solubilised in complete
medium (from a 0.4 mg/mL stock solution). In each plate of each run, three replicates of complete medium
untreated control, solvent/vehicle control, negative and positive controls are prepared (12). Other suitable
positive controls may be used if historical data are available to derive comparable run acceptance criteria.
The run acceptance criteria are the same as described for the test chemical (see paragraph 12).
© OECD 2018
17. The solvent/vehicle control or working solutions described in paragraphs 14-16 are mixed 1:1
(v/v) with the cell suspensions prepared in the 96-well flat-bottom plate (see paragraph 12). The treated
plates are then incubated for 45±3 hours at 37°C under 5% CO2. Prior to incubation, plates are sealed with
semi permeable membrane, to avoid evaporation of volatile test chemicals and cross-contamination
between cells treated with test chemicals (12).
Cell staining
18. After 45±3 hours of exposure, cells are transferred into V-shaped microtiter plate and collected
by centrifugation. Solubility interference is defined as crystals or drops observed under the microscope at
45 ± 3 hours post treatment (before the cell staining). The supernatants are discarded and the remaining
cells are washed once with 100 µL of an ice-cold phosphate buffered saline (PBS) containing 5 % foetal
calf serum (staining buffer). After centrifugation, cells are re-suspended with 100 µL of staining buffer and
stained with 5 µL (e.g. 0.25 µg) of FITC-labelled anti-CD86 or mouse IgG1 (isotype) antibodies at 4°C for
30 min protected from light. The antibodies described in the U-SENS™ DB-ALM protocol no. 183 (12)
should be used (for CD86: BD-PharMingen #555657 Clone: Fun-1, or Caltag/Invitrogen # MHCD8601
Clone: BU63; and for IgG1: BD-PharMingen #555748, or Caltag/Invitrogen # GM4992). Based on the
experience of the test method developers, the fluorescence intensity of the antibodies is usually consistent
between different lots. Other clones or supplier of the antibodies which passed the reactivity check may be
used for the assay (see paragraph 11). However, users may consider titrating the antibodies in their own
laboratory's conditions to define the best concentration for use. Other detection system e.g. fluorochrome-
tagged anti-CD86 antibodies may be used if they can be shown to provide similar results as FITC-
conjugated antibodies, for example by testing the proficiency substances in Appendix II. After washing
with 100 µL of staining buffer two times and once with 100 µL of an ice-cold PBS, cells are resuspended
in ice-cold PBS (e.g. 125 µL for samples being analysed manually tube by tube, or 50 µL using an auto-
sampler plate) and PI solution is added (final concentration of 3 µg/mL). Other cytotoxicity markers, such
as 7-Aminoactinomycin D (7-AAD) or Trypan blue may be used if the alternative stains can be shown to
provide similar results as PI, for example by testing the proficiency substances in Appendix II.
Flow cytometry analysis
19. Expression level of CD86 and cell viability are analysed using flow cytometry. Cells are
displayed within a size (FSC) and granularity (SSC) dot plot set to log scale in order to clearly identify the
population in a first gate R1 and eliminate the debris. A targeting total of 10,000 cells in gate R1 are
acquired for each well. Cells from the same R1 gate are displayed within a FL3 or FL4 / SSC dot plot.
Viable cells are delineated by placing a second gate R2 selecting the population of propidium iodide-
negative cells (FL3 or FL4 channel). The cell viability can be calculated using the following equation by
the cytometer analysis program. When the cell viability is low, up to 20,000 cells including dead cells
could be acquired. Alternatively, data can be acquired for one minute after the initiation of the analysis.
Number of living cells × 100

Cell Viability =
Total number of acquired cells
Percentage of FL1-positive cells is then measured among these viable cells gated on R2 (within R1). Cell
© OECD 2018
surface expression of CD86 is analysed in a FL1 / SSC dot plot gated on viable cells (R2).
For the complete medium / IgG1 wells, the analysis marker is set close to the main population so that the
complete medium controls have IgG1 within the target zone of 0.6 to 0.9%.
Colour interference is defined as a shift of the FITC-labelled IgG1 dot-plot (IgG1 FL1 Geo Mean S.I. ≥
150%).
The stimulation index (S.I.) of CD86 for controls cells (untreated or in 0.4% DMSO) and chemical-treated
cells are calculated according to the following equation:
% of CD86+ treated cells - % of IgG1+ treated cells
S.I. = x 100
% of CD86+ control cells - % of IgG1+ control cells
% of IgG1+ untreated control cells: referred to as percentage of FL1-positive IgG1 cells defined with the
analysis marker (accepted range of ≥ 0.6% and < 1.5%, see paragraph 22) among the viable untreated cells.
% of IgG1+/CD86+ control/treated cells: referred to as percentage of FL1-positive IgG1/CD86 cells
measured without moving the analysis marker among the viable control/treated cells.
DATA AND REPORTING
Data evaluation
20. The following parameters are calculated in the U-SENS™ test method: CV70 value, i.e. a
concentration showing 70% of U937 cell survival (30% cytotoxicity) and the EC150 value, i.e. the
concentration at which the test chemicals induced a CD86 stimulation index (S.I.) of 150%.
CV70 is calculated by log-linear interpolation using the following equation:
CV70 = C1 + [(V1 - 70) / (V1 – V2) * (C2 – C1)]
Where:
V1 is the minimum value of cell viability over 70%
V2 is the maximum value of cell viability below 70%
C1 and C2 are the concentrations showing the value of cell viability V1 and V2 respectively.
% Viability (Mean)
V1
70
V2
Dose (µg/ml)
C1 CV70 C2
Other approaches to derive the CV70 can be used as long as it is demonstrated that this has no impact on
the results (e.g. by testing the proficiency substances).
EC150 is calculated by log-linear interpolation using the following equation:
© OECD 2018
EC150 = C1 + [(150 – S.I.1) / (S.I.2 – S.I.1) * (C2 – C1)]
Where:
C1 is the highest concentration in µg/mL with a CD86 S.I. < 150% (S.I. 1)
C2 is the lowest concentration in µg/mL with a CD86 S.I. ≥ 150% (S.I. 2).
CD86-IgG1 S.I.
S.I.2
150
S.I.1
Dose (µg/ml)
C1 EC150 C2
The EC150 and CV70 values are calculated

- for each run : the individual EC150 and CV70 values are used as tools to investigate the
concentration response effect of CD86 increase (see paragraph 14),
- based on the average viabilities, the overall CV70 is determined (12) ,
- based on the average S.I. of CD86 values, the overall EC150 is determined for the test chemical
predicted as POSITIVE with the U-SENS™ (see paragraph 21) (12).
Prediction model
21. For CD86 expression measurement, each test chemical is tested in at least four concentrations
and in at least two independent runs (performed on a different day) to derive a single prediction
(NEGATIVE or POSITIVE).
- The individual conclusion of an U-SENS™ run is considered Negative (hereinafter referred to as N) if

the S.I. of CD86 is less than 150% at all non-cytotoxic concentrations (cell viability ≥ 70%) and if no
interference is observed (cytotoxicity, solubility: see paragraph 18 or colour: see paragraph 19 regardless
of the non-cytotoxic concentrations at which the interference is detected). In all other cases: S.I. of CD86
higher or equal to 150% and/or interferences observed, the individual conclusion of an U-SENS™ run is
considered Positive (hereinafter referred to as P).
- An U-SENS™ prediction is considered NEGATIVE if at least two independent runs are negative (N)
(Figure 1). If the first two runs are both negative (N), the U-SENS™ prediction is considered NEGATIVE
and a third run does not need to be conducted.
- An U-SENS™ prediction is considered POSITIVE if at least two independent runs are positive (P)
(Figure 1). If the first two runs are both positive (P), the U-SENS™ prediction is considered POSITIVE
and a third run does not need to be conducted.
- Because a dose finding assay is not conducted, there is an exception if, in the first run, the S.I. of CD86 is
higher or equal to 150% at the highest non-cytotoxic concentration only. The run is then considered to be
NOT CONCLUSIVE (NC), and additional concentrations (between the highest non cytotoxicity
concentration and the lowest cytotoxicity concentration - see paragraph 20) should be tested in additional
runs. In case a run is identified as NC, at least 2 additional runs should be conducted, and a fourth run in
case runs 2 and 3 are not concordant (N and/or P independently) (Figure 1). Follow up runs will be
considered positive even if only one non cytotoxic concentration gives a CD86 equal or above 150%, since
© OECD 2018
the concentration setting has been adjusted for the specific test chemical. The final prediction will be based
on the majority result of the three or four individual runs (i.e. 2 out of 3 or 2 out of 4) (Figure 1).
Figure 1: Prediction model used in the U-SENS™ test method. An U-SENS™ prediction should be considered in
the framework of an IATA and in accordance with the provision of paragraph 4 and of the General introduction
paragraphs 7, 8 and 9.
N: Run with no CD86 positive or interference observed;

P: Run with CD86 positive and/or interference(s) observed;
NC: Not Conclusive. First run with No Conclusion when CD86 is positive at the highest non-cytotoxic
concentration only;
#
: A Not Conclusive (NC) individual conclusion attributed only to the first run conducts automatically to
the need of a third run to reach a majority of Positive (P) or Negative (N) conclusions in at least 2 of 3
independent runs.
$: The boxes show the relevant combinations of results from the three runs on the basis of the results
obtained in the first two runs shown in the box above.
°: The boxes show the relevant combinations of results from the four runs on the basis of the results
obtained in the first three runs shown in the box above.
Acceptance criteria
22. The following acceptance criteria should be met when using the U-SENS™ method (12).
© OECD 2018
- At the end of the 45±3 hours exposure period, the mean viability of the triplicate untreated U937 cells
had to be > 90% and no drift in CD86 expression is observed. The CD86 basal expression of
untreated U937 cells had to be comprised within the range of ≥ 2% and ≤ 25%.
- When DMSO is used as a solvent, the validity of the DMSO vehicle control is assessed by calculating
a DMSO S.I. compared to untreated cells, and the mean viability of the triplicate cells had to be >
90%. The DMSO vehicle control is valid if the mean value of its triplicate CD86 S.I. was smaller than
250% of the mean of the triplicate CD86 S.I. of untreated U937 cells.
- The runs are considered valid if at least two out of three IgG1 values of untreated U937 cells fell
within the range of ≥ 0.6% and < 1.5%.
- The concurrent tested negative control (lactic acid) is considered valid if at least two out of the three
replicates were negative (CD86 S.I. < 150%) and non-cytotoxic (cell viability ≥ 70%).
- The positive control (TNBS) was considered as valid if at least two out of the three replicates were
positive (CD86 S.I. ≥ 150%) and non-cytotoxic (cell viability ≥ 70%).
Test report
23. The test report should include the following information.
Test Chemical
- Mono-constituent substance
 Physical appearance, complete medium solubility, DMSO solubility, molecular weight, and
additional relevant physicochemical properties, to the extent available;
- Multi-constituent substance, UVCB and mixture:
extent available;
 Physical appearance, complete medium solubility, DMSO solubility and additional relevant
© OECD 2018
compositions or other information relevant for the conduct of the study;
Controls
- Positive control
 Physical appearance, DMSO solubility, molecular weight, and additional relevant
physicochemical properties, to the extent available and where applicable;
 Reference to historical positive control results demonstrating suitable run acceptance criteria,
if applicable.
- Negative and solvent/vehicle control
the case other control solvent/vehicle than those mentioned in the Test Guideline are used and
to the extent available;
Test method Conditions
- Cell line used, its storage conditions and source (e.g. the facility from which they were obtained);
© OECD 2018
- Flow cytometry used (e.g. model), including instrument settings, antibodies and cytotoxicity marker
used;
- The procedure used to demonstrate proficiency of the laboratory in performing the test method by
testing of proficiency substances, and the procedure used to demonstrate reproducible performance of
the test method over time, e.g. historical control data and/or historical reactivity checks’ data.
Test Acceptance Criteria
- Cell viability and CD86 S.I values obtained with the solvent/vehicle control in comparison to the
acceptance ranges;
- Cell viability and S.I. values obtained with the positive control in comparison to the acceptance
ranges;
- Cell viability of all tested concentrations of the tested chemical.
Test procedure
- Number of runs used;

- Test chemical concentrations, application and exposure time used (if different than the one
recommended)
- Duration of exposure;
Results
- Tabulation of the data, including CV70 (if applicable), S.I., cell viability values, EC150 values (if
applicable) obtained for the test chemical and for the positive control in each run, and an indication of
the rating of the test chemical according to the prediction model;
Discussion of the Results
- Discussion of the results obtained with the U-SENS™ method;

- Consideration of the test method results within the context of an IATA, if other relevant information
is available.
Conclusions
© OECD 2018
LITERATURE
1. Piroird, C., Ovigne, J.M., Rousset, F., Martinozzi-Teissier, S., Gomes, C., Cotovio, J., Alépée, N.
(2015). The Myeloid U937 Skin Sensitization Test (U-SENS) addresses the activation of dendritic cell
event in the adverse outcome pathway for skin sensitization. Toxicol. In Vitro 29, 901-916.
2. EURL ECVAM (2017). The U-SENS™ test method Validation Study Report. Accessible at:
http://ihcp.jrc.ec.europa.eu/our_labs/eurl-ecvam/eurl-ecvam-recommendations
3. EC EURL ECVAM (2016). ESAC Opinion No. 2016-03 on the L'Oréal-coordinated study on the
transferability and reliability of the U-SENS™ test method for skin sensitisation testing. EUR 28178
EN; doi 10.2787/815737. Available at:
[http://publications.jrc.ec.europa.eu/repository/handle/JRC103705].
4. EC EURL ECVAM (2017). EURL ECVAM Recommendation on the use of non-animal approaches
for skin sensitisation testing. EUR 28553 EN; doi 10.2760/588955. Available at:
[https://ec.europa.eu/jrc/en/publication/eur-scientific-and-technical-research-reports/eurl-ecvam-
recommendation-use-non-animal-approaches-skin-sensitisation-testing].
5. Steiling, W. (2016). Safety Evaluation of Cosmetic Ingredients Regarding their Skin Sensitization
Potential. doi:10.3390/cosmetics3020014.Cosmetics 3, 14.
6. OECD (2016). Guidance Document On The Reporting Of Defined Approaches And Individual
Information Sources To Be Used Within Integrated Approaches To Testing And Assessment (IATA)
For Skin Sensitisation, Series on Testing & Assessment No. 256, ENV/JM/MONO(2016)29.
[http://www.oecd.org/env/ehs/testing/series-testing-assessment-publications-number.htm].
7. Urbisch, D., Mehling, A., Guth, K., Ramirez, T., Honarvar, N., Kolle, S., Landsiedel, R., Jaworska, J.,
Kern, P.S., Gerberick, F., Natsch, A., Emter, R., Ashikaga, T., Miyazawa, M., Sakaguchi, H. (2015).
Assessing skin sensitization hazard in mice and men using non-animal test methods. Regul. Toxicol.
Pharmacol. 71, 337-351.
8. Alépée, N., Piroird, C., Aujoulat, M., Dreyfuss, S., Hoffmann, S., Hohenstein, A., Meloni, M.,
Nardelli, L., Gerbeix, C., Cotovio, J. (2015). Prospective multicentre study of the U-SENS test
method for skin sensitization testing. Toxicol In Vitro 30, 373-382.
9. Reisinger, K., Hoffmann, S., Alépée, N., Ashikaga, T., Barroso, J., Elcombe, C., Gellatly, N.,
Galbiati, V., Gibbs, S., Groux, H., Hibatallah, J., Keller, D., Kern, P., Klaric, M., Kolle, S., Kuehnl, J.,
Lambrechts, N., Lindstedt, M., Millet, M., Martinozzi-Teissier, S., Natsch, A., Petersohn, D., Pike, I.,
Sakaguchi, H., Schepky, A., Tailhardat, M., Templier, M., van Vliet, E., Maxwell, G. (2014).
Systematic evaluation of non-animal test methods for skin sensitisation safety assessment. Toxicol. In
Vitro 29, 259-270.
10. Fabian, E., Vogel, D., Blatz, V., Ramirez, T., Kolle, S., Eltze, T., van Ravenzwaay, B., Oesch, F.,
Landsiedel, R. (2013). Xenobiotic metabolizin enzyme activities in cells used for testing skin
sensitization in vitro. Arch. Toxicol. 87, 1683-1696.
11. OECD. (2017). Draft Guidance document: Good In Vitro Method Practices (GIVIMP) for the
Development and Implementation of In Vitro Methods for Regulatory Use in Human Safety
Assessment. Organisation for Economic Cooperation and Development, Paris. Available at:
[http://www.oecd.org/env/ehs/testing/OECD%20Draft%20GIVIMP_v05%20-%20clean.pdf].
© OECD 2018
12. DB-ALM (2016). Protocol no. 183: Myeloid U937 Skin Sensitization Test (U-SENS™), 33pp.
Accessible at: [http://ecvam-dbalm.jrc.ec.europa.eu/].
13. Sundström, C., Nilsson, K. (1976). Establishment and characterization of a human histiocytic
lymphoma cell line (U-937). Int. J. Cancer 17, 565-577.
14. OECD (2005). Series on Testing and Assessment No. 34: Validation and International Acceptance of
New or Updated Test Methods for Hazard Assessment. Organisation for Economic Cooperation and
Development, Paris. Available at: [http://www.oecd.org/env/ehs/testing/series-testing-assessment-
publications-number.htm].
15. United Nations UN (2015). Globally Harmonized System of Classification and Labelling of
Chemicals (GHS). ST/SG/AC.10/30/Rev.6, Sixth Revised Edition,New York & Geneva: United
Nations Publications. Available at:
[http://www.unece.org/fileadmin/DAM/trans/danger/publi/ghs/ghs_rev06/English/ST-SG-AC10-30-
Rev6e.pdf].
16. OECD (2012). Series on Testing and Assessment No. 168: The Adverse Outcome Pathway for Skin
Sensitisation Initiated by Covalent Binding to Proteins. Part 1: Scientific Evidence. Organisation for
Economic Cooperation and Development, Paris. Available at:
17. ECETOC (2003). Technical Report No. 87: Contact sensitization: Classification according to potency.
European Centre for Ecotoxicology & Toxicology of Chemicals, Brussels. Available at:
[https://ftp.cdc.gov/pub/Documents/OEL/06.%20Dotson/References/ECETOC_2003-TR87.pdf].
© OECD 2018
APPENDIX I
DEFINITIONS
AOP (Adverse Outcome Pathway): sequence of events from the chemical structure of a target chemical
CD86 Concentration response: There is concentration-dependency (or concentration response) when a

positive concentration (CD86 S.I. ≥ 150) is followed by a concentration with an increasing CD86 S.I.
CV70: The estimated concentration showing 70% cell viability.
Drift: A drift is defined by i) the corrected %CD86+ value of the untreated control replicate 3 is less than
50% of the mean of the corrected %CD86+ value of untreated control replicates 1and 2; and ii) the
corrected %CD86+ value of the negative control replicate 3 is less than 50% of mean of the corrected
%CD86+ value of negative control replicates 1and 2.
EC150: the estimated concentrations showing the 150% S.I. of CD86 expression.
Flow cytometry: a cytometric technique in which cells suspended in a fluid flow one at a time through a
focus of exciting light, which is scattered in patterns characteristic to the cells and their components; cells
are frequently labeled with fluorescent markers so that light is first absorbed and then emitted at altered
frequencies.
identification (potential), hazard characterisation (potency) and/or safety assessment (potential/potency and
exposure) of a chemical or group of chemicals, which strategically integrates and weights all relevant data
to inform regulatory decision regarding potential hazard and/or risk and/or the need for further targeted and
therefore minimal testing.
© OECD 2018
one main constituent is present in a concentration ≥ 10% (w/w) and < 80% (w/w). A multi-constituent
substance is the result of a manufacturing process. The difference between mixture and multi-constituent
substance is that a mixture is obtained by blending of two or more substances without chemical reaction. A
multi-constituent substance is the result of a chemical reaction.
known to induce a positive response. To ensure that variability in the positive control response across time
can be assessed, the magnitude of the positive response should not be excessive.
Pre-haptens: chemicals which become sensitisers through abiotic transformation, e.g. through oxidation.
Pro-haptens: chemicals requiring enzymatic activation to exert skin sensitisation potential.
Relevance: Description of relationship of the test to the effect of interest and whether it is meaningful and
useful for a particular purpose. It is the extent to which the test correctly measures or predicts the
method (14).
S.I.: Stimulation Index. Relative values of geometric mean fluorescence intensity in chemical-treated cells
compared to solvent-treated cells.
Solvent/vehicle control: An untreated sample containing all components of a test system except of the test
chemical, but including the solvent/vehicle that is used. It is used to establish the baseline response for the
samples treated with the test chemical dissolved or stably dispersed in the same solvent/vehicle. When
tested with a concurrent medium control, this sample also demonstrates whether the solvent/vehicle
Staining buffer: A phosphate buffered saline containing 5% foetal calf serum.
© OECD 2018
process, including any additive necessary to preserve the stability of the product and any impurities
United Nations Globally Harmonized System of Classification and Labelling of Chemicals (UN
standardized types and levels of physical, health and environmental hazards, and addressing corresponding
communication elements, such as pictograms, signal words, hazard statements, precautionary statements
and safety data sheets, so that to convey information on their adverse effects with a view to protect people
(including employers, workers, transporters, consumers and emergency responders) and the environment
(16).

materials.
sense, but only in relation to a defined purpose (14).
© OECD 2018
APPENDIX II
Prior to routine use of the test method described in this Annex to Test Guideline 442E, laboratories should
demonstrate technical proficiency by correctly obtaining the expected U-SENS™ prediction for the 10
substances recommended in Table 1 and by obtaining CV70 and EC150 values that fall within the
respective reference range for at least 8 out of the 10 proficiency substances. Proficiency substances were
selected to represent the range of responses for skin sensitisation hazards. Other selection criteria were that
the substances are commercially available, and that high-quality in vivo reference data as well as high
quality in vitro data generated with the U-SENS™ method are available. Also, published reference data are
available for the U-SENS™ method (1) (8).
Table 1: Recommended substances for demonstrating technical proficiency with the U-SENS™ method
U-SENS™ U-SENS™ U-SENS™
Proficiency Physical In vivo Solvent/ CV70 EC150

CASRN
substances state prediction1 Vehicle Reference Reference
Range in Range in
µg/mL2 μg/mL2
Sensitiser Complete Positive
4-Phenylenediamine 106-50-3 Solid <30
(strong) medium3 (≤10)
Sensitizer Complete Positive
Picryl sulfonic acid 2508-19-2 Liquid >50
(strong) medium (≤50)
Sensitiser Positive
Diethyl maleate 141-05-9 Liquid DMSO 10-100
(moderate) (≤20)
Sensitiser Complete Positive
Resorcinol 108-46-3 Solid >100
(moderate) medium (≤50)
Sensitiser Positive
Cinnamic alcohol 104-54-1 Solid DMSO >100
(weak) (10-100)
Sensitiser Positive
4-Allylanisole 140-67-0 Liquid DMSO >100
(weak) (<200)
Negative
Saccharin 81-07-2 Solid Non-sensitiser DMSO >200
(>200)
Complete Negative
Glycerol 56-81-5 Liquid Non-sensitiser >200
medium (>200)
Complete Negative
Lactic acid 50-21-5 Liquid Non-sensitiser >200
medium (>200)
Negative
Salicylic acid 69-72-7 Solid Non-sensitiser DMSO >200
(>200)
Abbreviations: CAS RN = Chemical Abstracts Service Registry Number
1
The in vivo hazard and (potency) prediction is based on LLNA data (1) (8). The in vivo potency is derived
using the criteria proposed by ECETOC (17).
2
3
Complete medium: RPMI-1640 medium supplemented with 10% foetal calf serum, 2 mM L-glutamine, 100
units/mL penicillin and 100 µg/mL streptomycin (8).
© OECD 2018
ANNEX III: IN VITRO SKIN SENSITISATION: IL-8 LUC ASSAY
1. In contrast to assays analysing the expression of cell surface markers, the IL8-Luc assay
quantifies changes in IL-8 expression, a cytokine associated with the activation of dendritic cells (DC). In
the THP-1-derived IL-8 reporter cell line (THP-G8, established from the human acute monocytic leukemia
cell line THP-1), IL-8 expression is measured following exposure to sensitisers (1). The expression of
luciferase is then used to aid discrimination between skin sensitisers and non-sensitisers.
2. The IL-8 Luc method has been evaluated in a validation study (2) conducted by the Japanese
Centre for the Validation of Alternatives Methods (JaCVAM), the Ministry of Economy, Trade and
Industry (METI), and the Japanese Society for Alternatives to Animal Experiments (JSAAE) and
subsequently subjected to independent peer review (3) under the auspices of JaCVAM and the Ministry of
Health, Labour and Welfare (MHLW) with the support of the International Cooperation on Alternative
Test Methods (ICATM). Considering all available evidence and input from regulators and stakeholders,
the IL-8 Luc assay is considered useful as part of IATA to discriminate sensitisers from non-sensitisers for
the purpose of hazard classification and labelling. Examples of the use of IL-8 Luc assay data in
combination with other information are reported in the literature (4) (5) (6).
3. The IL-8 Luc assay proved to be transferable to laboratories experienced in cell culture and
luciferase measurement. Within and between laboratory reproducibilities were 87.7% and 87.5%,
respectively (2). Data generated in the validation study (2) and other published work (1) (6) show that
versus the LLNA, the IL-8 Luc assay judged 118 out of 143 chemicals as positive or negative and judged
25 chemicals as inconclusive and the accuracy of the IL-8 Luc assay in distinguishing skin sensitisers (UN
GHS Cat. 1) from non-sensitisers (UN GHS No Cat.) is 86% (101/118) with a sensitivity of 96% (92/96)
and specificity of 41% (9/22). Excluding substances outside the applicability domain described below
(paragraph 5), the IL-8 Luc assay judged 113 out of 136 chemicals as positive or negative and judged 23
chemicals as inconclusive and the accuracy of the IL-8 Luc assay is 89% (101/113) with sensitivity of 96%
(92/96) and specificity of 53% (9/17). Using human data cited in Urbisch et al. (7), the IL-8 Luc assay
judged 76 out of 90 chemicals as positive or negative and judged 14 chemicals as inconclusive and the
accuracy is 80% (61/76), sensitivity is 93% (54/58) and specificity is 39% (7/18). Excluding substances
outside the applicability domain, the IL-8 Luc assay judged 71 out of 84 chemicals as positive or negative
and judged 13 chemicals as inconclusive and the accuracy is 86% (61/71）with sensitivity of 93% (54/58)
and specificity of 54% (7/13). False negative predictions with the IL-8 Luc assay are more likely to occur
with chemicals showing low/moderate skin sensitisation potency (UN GHS subcategory 1B) than those
with high potency (UN GHS subcategory 1A) (6). Together, the information supports a role for the IL-8
Luc assay in the identification of skin sensitisation hazards. The accuracy given for the IL-8 Luc assay as a
standalone test method is only for guidance, as the method should be considered in combination with other
sources of information in the context of an IATA and in accordance with the provisions of paragraphs 7
and 8 in the General introduction. Furthermore, when evaluating non-animal methods for skin sensitisation,
it should be remembered that the LLNA and other animal tests may not fully reflect the situation in
humans.
© OECD 2018
4. On the basis of the data currently available, the IL-8 Luc assay was shown to be applicable to test
chemicals covering a variety of organic functional groups, reaction mechanisms, skin sensitisation potency
(as determined in in vivo studies) and physicochemical properties (2) (6).
5. Although the IL-8 Luc assay uses X-VIVOTM 15 as a solvent, it correctly evaluated chemicals
with a Log Ko/w >3.5 and those with a water solubility of around 100 µg/ mL as calculated by EPI SuiteTM
and its performance to detect sensitisers with poor water solubility is better than that of the IL-8 Luc assay
using dimethyl sulfoxide (DMSO) as a solvent (2). However, negative results for test chemicals that are
not dissolved at 20 mg/ml may produce false negative results due to their inability to dissolve in X-
VIVOTM 15. Therefore, negative results for these chemicals should not be considered. A high false
negative rate for anhydrides was seen in the validation study. Furthermore, because of the limited
metabolic capability of the cell line (8) and the experimental conditions, pro-haptens (substances requiring
metabolic activation) and pre-haptens (substances activated by air oxidation) might give negative results in
the assay. However, although negative results for suspected pre/prohaptens should be interpreted with
caution, the IL-8 Luc assay correctly judged 11 out of 11 pre-haptens, 6/6 pro-haptens, and 6/8 pre/pro-
haptens in the IL-8 Luc assay data set (2). Based on the recent comprehensive review on three non-animal
methods (the DPRA, the KeratinoSens™ and the h-CLAT) to detect pre and prohaptens (9), and based on
the fact that THP-G8 cells used in the IL-8 Luc assay is a cell line derived from THP-1 that is used in the
h-CLAT, the IL-8 Luc assay may also contribute to increase the sensitivity of non-animal methods to
detect pre and pro-haptens in the combination of other methods. Surfactants tested so far gave (false)
positive results irrespective of their type (e.g. cationic, anionic or on-ionic). Finally, chemicals that
interfere with luciferase can confound its activity/measurement, causing apparent inhibition or increased
luminescence (10). For example, phytoestrogen concentrations higher than 1µM were reported to interfere
with luminescence signals in other luciferase-based reporter gene assays due to over-activation of the
luciferase reporter gene. Consequently, luciferase expression obtained at high concentrations of
phytoestrogens or compounds suspected of producing phytoestrogen-like activation of the luciferase
reporter gene needs to be examined carefully (11). Based on the above, surfactants, anhydrides and
chemicals interfering with luciferase are outside the applicability domain of this assay. In cases where
there is evidence demonstrating the non-applicability of the IL-8 Luc assay to other specific categories of
test chemicals, the method should not be used for those specific categories.
6. As described above, the IL-8 Luc assay supports discrimination of skin sensitisers from non-
sensitisers. Further work, preferably based on human data, is required to determine whether IL-8 Luc
results can contribute to potency assessment when considered in combination with other information
sources.
8. The IL-8 Luc assay makes use of a human monocytic leukemia cell line THP-1 that was obtained
from the American Type Culture Collection (Manassas, VA, USA). Using this cell line, the Dept. of
Dermatology, Tohoku University School of Medicine, established a THP-1-derived IL-8 reporter cell line,
THP-G8, that harbours the Stable Luciferase Orange (SLO) and Stable Luciferase Red (SLR) luciferase
genes under the control of the IL-8 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH)
promoters, respectively (1). This allows quantitative measurement of luciferase gene induction by
detecting luminescence from well-established light producing luciferase substrates as an indicator of the
activity of the IL-8 and GAPDH in cells following exposure to sensitising chemicals.
© OECD 2018
9. The dual-colour assay system comprises an orange-emitting luciferase (SLO; max = 580 nm)
(12) for the gene expression of the IL-8 promoter as well as a red-emitting luciferase (SLR; max = 630
nm) (13) for the gene expression of the internal control promoter, GAPDH. The two luciferases emit
different colours upon reacting with firefly D-luciferin and their luminescence is measured simultaneously
in a one-step reaction by dividing the emission from the assay mixture using an optical filter (14)
(Appendix II).
10. THP-G8 cells are treated for 16 hours with the test chemical, after which SLO luciferase activity
(SLO-LA) reflecting IL-8 promoter activity and SLR luciferase activity (SLR-LA) reflecting GAPDH
promoter activity are measured. To make the abbreviations easy to understand, SLO-LA and SLR-LA are
designated as IL8LA and GAPLA, respectively. Table 1 gives a description of the terms associated with
luciferase activity in the IL-8 Luc assay. The measured values are used to calculate the normalised IL8LA
(nIL8LA), which is the ratio of IL8LA to GAPLA; the induction of nIL8LA (Ind-IL8LA), which is the
ratio of the arithmetic means of quadruple-measured values of the nIL8LA of THP-G8 cells treated with a
test chemical and the values of the nIL8LA of untreated THP-G8 cells; and the inhibition of GAPLA (Inh-
GAPLA), which is the ratio of the arithmetic means of quadruple-measured values of the GAPLA of THP-
G8 cells treated with a test chemical and the values of the GAPLA of untreated THP-G8 cells, and used as
an indicator for cytotoxicity.
Table 1. Description of terms associated with the luciferase activity in the IL-8 Luc assay
Abbreviations Definition
GAPLA SLR luciferase activity reflecting GAPDH promoter activity
IL8LA SLO luciferase activity reflecting IL-8 promoter activity
nIL8LA IL8LA / GAPLA
Ind-IL8LA nIL8LA of THP-G8 cells treated with chemicals / nIL8LA of untreated cells
Inh-GAPLA GAPLA of THP-G8 treated with chemicals / GAPLA of untreated cells
CV05 The lowest concentration of the chemical at which Inh-GAPLA becomes <
0.05.
11. Performance standards (PS) (15) are available to facilitate the validation of modified in vitro IL-8
luciferase test methods similar to the IL-8 Luc assay and allow for timely amendment of this Test
Guideline for their inclusion. Mutual Acceptance of Data (MAD) will only be guaranteed for test methods
validated according to the PS, if these test methods have been reviewed and included in this Test Guideline
by the OECD (16).
12. Prior to routine use of the test method described in this Annex to Test Guideline 442E,
laboratories should demonstrate technical proficiency, using the 9 Proficiency Substances listed in
Appendix III in compliance with the Good in vitro Method Practices (17). Moreover, test method users
should maintain a historical database of data generated with the reactivity checks (see paragraph 15) and
with the positive and solvent/vehicle controls (see paragraphs 21-24), and use these data to confirm the
reproducibility of the test method in their laboratory is maintained over time.
© OECD 2018
PROCEDURE
13. The Standard Operating Procedure (SOP) for the IL-8 Luc assay is available and should be
employed when performing the test (18). Laboratories willing to perform the test can obtain the
recombinant THP-G8 cell line from GPC Lab. Co. Ltd., Tottori, Japan, upon signing a Material Transfer
Agreement (MTA) in line with the conditions of the OECD template. The following paragraphs provide a
description of the main components and procedures of the assay.
14. The THP-G8 cell line from GPC Lab. Co. Ltd., Tottori, Japan, should be used for performing the
IL-8 Luc assay (see paragraphs 8 and 13). On receipt, cells are propagated (2-4 passages) and stored frozen
as a homogeneous stock. Cells from this stock can be propagated up to a maximum of 12 passages or a
maximum of 6 weeks. The medium used for propagation is the RPMI-1640 culture medium containing
10% foetal bovine serum (FBS), antibiotic/antimycotic solution (100U/mL of penicillin G, 100µg/mL of
streptomycin and 0.25µg/mL of amphotericin B in 0.85% saline) (e.g. GIBCO Cat#15240-062),
0.15μg/mL Puromycin (e.g. CAS:58-58-2) and 300μg/mL G418 (e.g. CAS:108321-42-2).
15. Prior to use for testing, the cells should be qualified by conducting a reactivity check. This check
should be performed 1-2 weeks or 2-4 passages after thawing, using the positive control, 4-nitrobenzyl
bromide (4-NBB) (CAS:100-11-8, ≥ 99% purity) and the negative control, lactic acid (LA) (CAS:50-21-5,
≥85% purity). 4-NBB should produce a positive response to Ind-IL8LA (≥1.4), while LA should produce a
negative response to Ind-IL8LA (<1.4). Only cells that pass the reactivity check are used for the assay. The
check should be performed according to the procedures described in paragraphs 22-24.
16. For testing, THP-G8 cells are seeded at a density of 2 to 5 × 105 cells/mL, and pre-cultured in
culture flasks for 48 to 96 hours. On the day of the test, cells harvested from the culture flask are washed
with RPMI-1640 containing 10% FBS without any antibiotics, and then, resuspended with RPMI-1640
containing 10% FBS without any antibiotics at 1 × 106 cells/mL. Then, cells are distributed into a 96-well
flat-bottom black plate (e.g. Costar Cat#3603) with 50µL (5 × 104 cells/well).
Preparation of the test chemical and control substances

17. The test chemical and control substances are prepared on the day of testing. For the IL-8
Luc assay, test chemicals are dissolved in X-VIVOTM 15, a commercially available serum-free medium
(Lonza, 04-418Q), to the final concentration of 20 mg/mL. X-VIVOTM 15 is added to 20 mg of test
chemical (regardless of the chemical’s solubility) in a microcentrifuge tube and brought to a volume of
1mL and then vortexed vigorously and shaken on a rotor at a maximum speed of 8 rpm for 30 min at an
ambient temperature of about 20°C. Furthermore, if solid chemicals are still insoluble, the tube is sonicated
until the chemical is dissolved completely or stably dispersed. For test chemicals soluble in X-VIVOTM
15, the solution is diluted by a factor of 5 with X-VIVOTM 15 and used as an X-VIVOTM 15 stock solution
of the test chemical (4 mg/mL). For test chemicals not soluble in X-VIVOTM 15, the mixture is rotated
again for at least 30 min, then centrifuged at 15,000 rpm (≈20,000g) for 5 min; the resulting supernatant is
used as an X-VIVOTM 15 stock solution of the test chemical. A scientific rationale should be provided for
the use of other solvents, such as DMSO, water, or the culture medium. The detailed procedure for
dissolving chemicals is shown in Appendix V. The X-VIVOTM 15 solutions described in paragraphs 18-23
are mixed 1:1 (v/v) with the cell suspensions prepared in a 96-well flat-bottom black plate (see paragraph
16).
© OECD 2018
18. The first test run is aimed to determine the cytotoxic concentration and to examine the skin
sensitising potential of chemicals. Using X-VIVOTM 15, serial dilutions of the X-VIVOTM 15 stock
solutions of the test chemicals are made at a dilution factor of two (see Appendix V) using a 96-well assay
block (e.g. Costar Cat#EW-01729-03). Next, 50 μl/well of diluted solution is added to 50 μl of the cell
suspension in a 96-well flat-bottom black plate. Thus for test chemicals that are soluble in X-VIVO TM 15,
the final concentrations of the test chemicals range from 0.002 to 2 mg/mL (Appendix V). For test
chemicals that are not soluble in X-VIVO TM 15 at 20 mg/mL, only dilution factors that range from 2 to 210,
are determined, although the actual final concentrations of the test chemicals remain uncertain and are
dependent on the saturated concentration of the test chemicals in the X-VIVO TM 15 stock solution.
19. In subsequent test runs (i.e. the second, third, and fourth replicates), the X-VIVOTM 15 stock
solution is made at the concentration 4 times higher than the concentration of cell viability 05 (CV05; the
lowest concentration at which the Inh-GAPLA becomes <0.05) in the first experiment. If Inh-GAPLA does
not decrease below 0.05 at the highest concentration in the first run, the X-VIVOTM 15 stock solution is
made at the first run highest concentration. The concentration of CV05 is calculated by dividing the
concentration of the stock solution in the first run by dilution factor for CV05 (X) (dilution factor CV05
(X); the dilution factor required to dilute stock solution to CV05) (see Appendix V). For test substances not
soluble in X-VIVO at 20 mg/ml, CV05 is determined by the concentration of the stock solution x 1/X. For
run 2 to 4, a second stock solution is prepared as 4 x CV50 (Appendix V).
20. Serial dilutions of the X-VIVOTM 15 second stock solutions are made at a dilution factor of 1.5
using a 96-well assay block. Next, 50 μl/well of diluted solution is added to 50 μl of the cell suspension in
the wells of a 96-well flat-bottom black plate. Each concentration of each test chemical should be tested in
4 wells. The samples are then mixed on a plate shaker and incubated for 16 hours at 37°C and 5% CO 2,
after which the luciferase activity is measured as described below.
21. The solvent control is the mixture of 50 µL/well of X-VIVOTM 15 and 50 µL/well of cell
suspension in RPMI-1640 containing 10% FBS.
22. The recommended positive control is 4-NBB. 20 mg of 4-NBB is prepared in a 1.5-mL

microfuge tube, to which X-VIVOTM 15 is added up to 1 mL. The tube is vortexed vigorously and shaken
on a rotor at a maximum speed of 8 rpm for at least 30 min. After centrifugation at 20,000g for 5 min, the
supernatant is diluted by a factor of 4 with X-VIVOTM 15, and 500 μl of the diluted supernatant is
transferred to a well in a 96-well assay block. The diluted supernatant is further diluted with X-VIVOTM 15
at factors of 2 and 4, and 50 μl of the solution is added to 50 μl of THP-G8 cell suspension in the wells of a
96-well flat-bottom black plate (Appendix VI). Each concentration of the positive control should be tested
in 4 wells. The plate is agitated on a plateshaker, and incubated in a CO2 incubator for 16 hours
(37°C, 5% CO2), after which the luciferase activity is measured as described in paragraph 29.
23. The recommended negative control is LA. 20 mg of LA prepared in a 1.5-mL microfuge tube, to
which X-VIVOTM 15 is added up to 1 mL (20 mg/ mL). Twenty mg/mL of LA solution is diluted by a
factor of 5 with X-VIVOTM 15 (4 mg/mL); 500 μl of this 4 mg/mL LA solution is transferred to a well of a
96-well assay block. This solution is diluted by a factor of 2 with X-VIVOTM 15 and then diluted again by
a factor of 2 to produce 2 mg/mL and 1 mg/mL solutions. 50 μl of these 3 solutions and vehicle control
(X-VIVOTM 15) are added to 50 µl of THP-G8 cell suspension in the wells of a 96-well flat-bottom black
plate. Each concentration of the negative control is tested in 4 wells. The plate is agitated on a plateshaker
and incubated in a CO2 incubator for 16 hours (37°C, 5% CO2), after which the luciferase activity is
measured as described in paragraph 29.
© OECD 2018
24. Other suitable positive or negative controls may be used if historical data are available to derive
comparable run acceptance criteria.
25. Care should be taken to avoid evaporation of volatile test chemicals and cross-contamination
between wells by test chemicals, e.g. by sealing the plate prior to the incubation with the test chemicals.
26. The test chemicals and solvent control require 2 to 4 runs to derive a positive or negative
prediction (see Table 2). Each run is performed on a different day with fresh X-VIVOTM 15 stock solution
of test chemicals and independently harvested cells. Cells may come from the same passage.
Luciferase activity measurements

27. Luminescence is measured using a 96-well microplate luminometer equipped with optical filters,
e.g. Phelios (ATTO, Tokyo, Japan), Tristan 941 (Berthold, Bad Wildbad, Germany) and the ARVO series
(PerkinElmer, Waltham, MA, USA). The luminometer must be calibrated for each test to ensure
reproducibility (19). Recombinant orange and red emitting luciferases are available for this calibration.
28. 100µL of pre-warmed Tripluc® Luciferase assay reagent (Tripluc) is transferred to each well of
the plate containing the cell suspension treated with or without chemical. The plate is shaken for 10 min at
an ambient temperature of about 20°C. The plate is placed in the luminometer to measure the luciferase
activity. Bioluminescence is measured for 3 sec each in the absence (F0) and presence (F1) of the optical
filter. Justification should be provided for the use of alternative settings, e.g. depending on the model of
luminometer used.
29. Parameters for each concentration are calculated from the measured values, e.g. IL8LA, GAPLA,
nIL8LA, Ind-IL8LA, Inh-GAPLA, the mean ±SD of IL8LA, the mean ±SD of GAPLA, the mean ±SD of
nIL8LA, the mean ±SD of Ind-IL8LA, the mean ±SD of Inh-GAPLA, and the 95% confidence interval of
Ind-IL8LA. Definitions of the parameters used in this paragraph are provided in Appendices I and IV,
respectively.
30. Prior to measurement, colour discrimination in multi-colour reporter assays is generally achieved
using detectors (luminometer and plate reader) equipped with optical filters, such as sharp-cut (long-pass
or short-pass) filters or band-pass filters. The transmission coefficients of the filters for each
bioluminescence signal colour should be calibrated prior to testing, per Appendix II.
© OECD 2018
DATA AND REPORTING
Data evaluation
31. Criteria for a positive/negative decision require that in each run:
- an IL-8 Luc assay prediction is judged positive if a test chemical has a Ind-IL8LA  1.4 and the lower
limit of the 95% confidence interval of Ind-IL8LA  1.0
- an IL-8 Luc assay prediction is judged negative if a test chemical has a Ind-IL8LA < 1.4 and/or the lower
limit of the 95% confidence interval of Ind-IL8LA < 1.0
Prediction model
32. Test chemicals that provide two positive results from among the 1st, 2nd, 3rd or 4th runs are
identified as positives whereas those that give three negative results from among the 1 st, 2nd, 3rd or 4th runs
are identified as supposed negative (Table 2). Among supposed negative chemicals, chemicals that are
dissolved at 20 mg/ml of X-VOVOTM 15 are judged as negative, while chemicals that are not dissolved at
20 mg/ml of X-VOVOTM 15 should not be considered (Figure 1).
Table 2. Criteria for identifying positive and supposed negative

1st run 2nd run 3rd run 4th run Final prediction
Positive Positive - - Positive
Negative Positive - Positive
Negative Positive Positive
Negative Supposed
negative
Negative Positive Positive - Positive
Negative Positive Positive
Negative Supposed
negative
Negative Positive Positive Positive
Negative Supposed
negative
Negative - Supposed
negative
© OECD 2018
Figure 1. Prediction model for final judgment
Positive
Chemicals Soluble at 20 mg/ml Negative

Supposed
negative
Insoluble at 20 mg/ml Should not be
considered
Acceptance criteria
33. The following acceptance criteria should be met when using the IL-8 Luc assay:
- Ind-IL8LA should be more than 5.0 at least in one concentration of the positive control, 4-NBB, in each
run.
- Ind-IL8LA should be less than 1.4 at any concentration of the negative control, lactic acid, in each run.
- Data from plates for which the GAPLA of control wells with cells and Tripluc but without chemicals is
less than 5 times of that of well containing test medium only (50 µL/well of RPMI-1640 containing 10%
FBS and 50 µL/well of X-VIVOTM 15) should be rejected.
- Data from plates for which the Inh-GAPLA of all concentrations of the test or control chemicals is less
than 0.05 should be rejected. In this case, the first test should be repeated so the highest final concentration
of the repeated test is the lowest final concentration of the previous test.
TEST REPORT
34. The test report should include the following information:
Test chemicals
- Mono-constituent substance:
 Physical appearance, water solubility, molecular weight, and additional relevant
 Purity, chemical identity of impurities as appropriate and practically feasible, etc;
© OECD 2018
 Solubility in X-VIVOTM 15. For chemicals that are insoluble in X-VIVOTM 15, whether
precipitation or flotation are observed after centrifugation;
 Justification for choice of solvent/vehicle for each test chemical if X-VIVOTM 15 has not been
used.
- Multi-constituent substance, UVCB and mixture:
extent available;
 Physical appearance, water solubility, and additional relevant physicochemical properties, to
the extent available;
compositions or other information relevant relevant for the conduct of the study;
 Solubility in X-VIVOTM 15. For chemicals that are insoluble in X-VIVOTM 15, whether
precipitation or flotation are observed after centrifugation;
 Storage conditions and stability to the extent available.
 Justification for choice of solvent/vehicle for each test chemical, if X-VIVOTM 15 has not
been used.
Controls
- Positive control:
 Physical appearance, water solubility, molecular weight, and additional relevant
physicochemical properties, to the extent available and where applicable;
 Reference to historical positive control results demonstrating suitable acceptance criteria, if
applicable.
© OECD 2018
- Negative control:
 Chemical identification, such as IUPAC or CAS name(s), CAS number(s), and/or other
identifiers;
the case other negative controls than those mentioned in the Test Guideline are used and to the
extent available;
 Justification for choice of solvent for each test chemical.
Test method conditions
- Cell line used, its storage conditions, and source (e.g. the facility from which it was obtained);
- Lot number and origin of FBC, supplier name, lot number of 96-well flat-bottom black plate, and lot
number of Tripluc reagent;
- Passage number and cell density used for testing;
- Cell counting method used for seeding prior to testing and measures taken to ensure homogeneous cell
number distribution;
- Luminometer used (e.g. model), including instrument settings, luciferase substrate used, and
demonstration of appropriate luminescence measurements based on the control test described in
Appendix II;
- The procedure used to demonstrate proficiency of the laboratory in performing the test method (e.g. by
testing of proficiency substances) or to demonstrate reproducible performance of the test method over
time.
Test procedure
- Number of replicates and runs performed;

- Test chemical concentrations, application procedure and exposure time (if different from those
recommended);
- Description of study acceptance criteria used;
Results
© OECD 2018
- Measurements of IL8LA and GAPLA;
- Calculations for nIL8LA, Ind-IL8LA, and Inh-GAPLA;
- The 95% confidence interval of Ind-IL8LA;
- A graph depicting dose-response curves for induction of luciferase activity and viability;
Discussion of the results
- Discussion of the results obtained with the IL-8 Luc assay;

- Consideration of the assay results in the context of an IATA, if other relevant information is available.
Conclusion
© OECD 2018
LITERATURE
1. Takahashi T, Kimura Y, Saito R, Nakajima Y, Ohmiya Y, Yamasaki K, and Aiba S. 2011. An in

vitro test to screen skin sensitizers using a stable THP-1-derived IL-8 reporter cell line, THP-G8. Toxicol
Sci 124:359-69.
2. OECD (2017), Validation report for the international validation study on the IL-8 Luc assay as a
test evaluating the skin sensitizing potential of chemicals conducted by the IL-8 Luc Assay. Series on
Testing and Assessment No. 267, ENV/JM/MONO(2017)19. Organisation for Economic Cooperation and
Development, Paris. Available at: [http://www.oecd.org/env/ehs/testing/series-testing-assessment-
publications-number.htm].
3. OECD (2017), Report of the Peer Review Panel for the IL-8 Luciferase (IL-8 Luc) Assay for in
vitro skin sensitisation. Series on Testing and Assessment No. 258, ENV/JM/MONO(2017)20.
2. OECD (2016), Guidance Document On The Reporting Of Defined Approaches And Individual
Information Sources To Be Used Within Integrated Approaches To Testing And Assessment (IATA) For
Skin Sensitisation, Series on Testing & Assessment No. 256, ENV/JM/MONO(2016)29. Organisation for
Economic Cooperation and Development, Paris. Available at: [http://www.oecd.org/env/ehs/testing/series-
testing-assessment-publications-number.htm].
3. van der Veen JW, Rorije E, Emter R, Natsch A, van Loveren H, and Ezendam J. 2014.
Evaluating the performance of integrated approaches for hazard identification of skin sensitizing
chemicals. Regul Toxicol Pharmacol 69:371-9.
4. Kimura Y, Fujimura C, Ito Y, Takahashi T, Nakajima Y, Ohmiya Y, and Aiba S. 2015.

Optimization of the IL-8 Luc assay as an in vitro test for skin sensitization. Toxicol In Vitro 29:1816-30.
5. Urbisch D, Mehling A, Guth K, Ramirez T, Honarvar N, Kolle S, Landsiedel R, Jaworska J, Kern

PS, Gerberick F, et al. 2015. Assessing skin sensitization hazard in mice and men using non-animal test
methods. Regul Toxicol Pharmacol 71:337-51.

Nishiyama N, and Itagaki H. 2010. A comparative evaluation of in vitro skin sensitisation tests: the human
cell-line activation test (h-CLAT) versus the local lymph node assay (LLNA). Alternatives to laboratory
animals : ATLA 38:275-84.
7. Patlewicz G, Casati S, Basketter DA, Asturiol D, Roberts DW, Lepoittevin J-P, Worth A and
Aschberger K (2016) Can currently available non-animal methods detect pre and pro haptens relevant for
skin sensitisation? Regul Toxicol Pharmacol, online.
8. Thorne N, Inglese J, and Auld DS. 2010. Illuminating insights into firefly luciferase and other
bioluminescent reporters used in chemical biology. Chem Biol 17:646-57.
9. OECD (2016), Test No. 455: Performance-Based Test Guideline for Stably Transfected
Transactivation In Vitro Assays to Detect Estrogen Receptor Agonists and Antagonists, OECD Publishing,
Paris. http://dx.doi.org/10.1787/9789264265295-en.
© OECD 2018
10. Viviani V, Uchida A, Suenaga N, Ryufuku M, and Ohmiya Y. 2001. Thr226 is a key residue for
bioluminescence spectra determination in beetle luciferases. Biochem Biophys Res Commun 280:1286-91.
11. Viviani VR, Bechara EJ, and Ohmiya Y. 1999. Cloning, sequence analysis, and expression of
active Phrixothrix railroad-worms luciferases: relationship between bioluminescence spectra and primary
structures. Biochemistry 38:8271-9.
12. Nakajima Y, Kimura T, Sugata K, Enomoto T, Asakawa A, Kubota H, Ikeda M, and Ohmiya Y.
2005. Multicolor luciferase assay system: one-step monitoring of multiple gene expressions with a single
substrate. Biotechniques 38:891-4.
13. OECD (2017), To be published - Performance Standards for the assessment of proposed similar
or modified in vitro skin sensitisation IL-8 luc test methods. OECD Environment, Health and Safety
Publications, Series on Testing and Assessment. OECD, Paris, France.
14. OECD (2005), Guidance Document the Validation and International Acceptance of New or
Updated Test Methods for Hazard Assessment. OECD Environment, Health and Safety publications,
OECD Series on Testing and Assessment No.34. OECD, Paris, France.
15. OECD (2017), Draft Guidance document: Good In Vitro Method Practices (GIVIMP) for the
Development and Implementation of In Vitro Methods for Regulatory Use in Human Safety Assessment.
[http://www.oecd.org/env/ehs/testing/OECD%20Draft%20GIVIMP_v05%20-%20clean.pdf].
16. JaCVAM (2016), IL-8 Luc assay protocol, Available at.

http://www.jacvam.jp/en_effort/effort02.html.
17. Niwa K, Ichino Y, Kumata S, Nakajima Y, Hiraishi Y, Kato D, Viviani VR, and Ohmiya Y.
2010. Quantum yields and kinetics of the firefly bioluminescence reaction of beetle luciferases. Photochem
Photobiol 86:1046-9.
18. OECD (2012), The Adverse Outcome Pathway for Skin Sensitisation Initiated by Covalent
Binding to Proteins, Part 1: Scientific Evidence. OECD Environment, Health and Safety Publications,
Series on Testing and Assessment No. 168. OECD, Paris, France.
19. United Nations (2015), Globally Harmonized System of Classification and Labelling of
Chemicals (GHS). Sixth revised edition. New York & Geneva: United Nations Publications. ISBN: 978-
92-1-117006-1. Available at: http://www.unece.org/trans/danger/publi/ghs/ghs_rev05/05files_e.html.
© OECD 2018
APPENDIX I
DEFINITIONS
AOP (Adverse Outcome Pathway): Sequence of events from the chemical structure of a target chemical
CV05: Cell viability 05. Minimum concentration at which chemicals show less than 0.05 of Inh-GAPLA.
FInSLO-LA: Abbreviation used in the validation report and in previous publications regarding the IL-8
Luc assay to refer to Ind-IL8LA. See Ind-IL8LA for definition.
GAPLA: Luciferase Activity of Stable Luciferase Red (SLR) (max = 630 nm), regulated by GAPDH
promoter and demonstrates cell viability and viable cell number.
identification (potential), hazard characterisation (potency) and/or safety assessment (potential/potency
and exposure) of a chemical or group of chemicals, which strategically integrates and weights all relevant
data to inform regulatory decision regarding potential hazard and/or risk and/or the need for further
targeted and therefore minimal testing.
II-SLR-LA: Abbreviation used in the validation report and in previous publications regarding the IL-8
Luc assay to refer to Inh-GAPLA. See Inh-GAPLA for definition
IL-8 (Interleukin-8): A cytokine derived from endothelial cells, fibroblasts, keratinocytes, macrophages,
and monocytes that causes chemotaxis of neutrophils and T-cell lymphocytes.
IL8LA: Luciferase Activity of Stable Luciferase Orange (SLO) (max = 580 nm), regulated by IL-8
promoter.
Ind-IL8LA: Fold induction of IL8LA. It is obtained by dividing the nIL8LA of THP-G8 cells treated with
chemicals by that of non-stimulated THP-G8 cells and represents the induction of IL-8 promoter activity
by chemicals.
Inh-GAPLA: Inhibition of GAPLA. It is obtained by dividing GAPLA of THP-G8 treated with chemicals
with GAPLA of non-treated THP-G8 and represents cytotoxicity of chemicals.
Minimum induction threshold (MIT): the lowest concentration at which a chemical satisfies the positive
criteria
© OECD 2018
one of the main constituents is present in a concentration ≥ 10% (w/w) and < 80% (w/w). A multi-
constituent substance is the result of a manufacturing process. The difference between mixture and multi-
constituent substance is that a mixture is obtained by blending of two or more substances without chemical
reaction. A multi-constituent substance is the result of a chemical reaction.
nIL8LA: The SLO luciferase activity reflecting IL-8 promoter activity (IL8LA) normalised by the SLR
luciferase activity reflecting GAPDH promoter activity (GALPA). It represents IL-8 promoter activity after
considering cell viability or cell number.
nSLO-LA: Abbreviation used in the validation report and in previous publications regarding the IL-8 Luc
assay to refer to nIL8LA. See nIL8LA for definition
known to induce a positive response. To ensure that variability in the positive control response across
time can be assessed, the magnitude of the positive response should not be excessive.
Pre-haptens: Chemicals which become sensitisers through abiotic transformation.
Pro-haptens: Chemicals requiring enzymatic activation to exert skin sensitisation potential.
Relevance: Description of relationship of the test to the effect of interest and whether it is meaningful
and useful for a particular purpose. It is the extent to which the test correctly measures or predicts the
method (16).
SLO-LA: Abbreviation used in the validation report and in previous publications regarding the IL-8 Luc
assay to refer to IL8LA. See IL8LA for definition.
SLR-LA: Abbreviation used in the validation report and in previous publications regarding the IL-8 Luc
assay to refer to GAPLA. See GAPLA for definition.
Solvent/vehicle control: An untreated sample containing all components of a test system except of the
test chemical, but including the solvent/vehicle that is used. It is used to establish the baseline response
© OECD 2018
for the samples treated with the test chemical dissolved or stably dispersed in the same solvent/vehicle.
When tested with a concurrent medium control, this sample also demonstrates whether the solvent/vehicle
process, inducing any additive necessary to preserve the stability of the product and any impurities
Surfactant: Also called surface-active agent, this is a substance, such as a detergent, that can reduce the
surface tension of a liquid and thus allow it to foam or penetrate solids; it is also known as a wetting agent.
(TG437)
THP-G8: An IL-8 reporter cell line used in IL-8 Luc assay. The human macrophage-like cell line THP-1
was transfected the SLO and SLR luciferase genes under the control of the IL-8 and GAPDH promoters,
respectively.
United Nations Globally Harmonized System of Classification and Labeling of Chemicals (UN
standardised types and levels of physical, health and environmental hazards, and addressing
corresponding communication elements, such as pictograms, signal words, hazard statements,
precautionary statements and safety data sheets, so that to convey information on their adverse effects
with a view to protect people (including employers, workers, transporters, consumers and emergency
responders) and the environment (21).

materials.
sense, but only in relation to a defined purpose.
© OECD 2018
APPENDIX II
PRINCIPLE OF MEASUREMENT OF LUCIFERASE ACTIVITY AND DETERMINATION

OF THE TRANSMISSION COEFFICIENTS OF OPTICAL FILTER FOR SLO AND SLR
MultiReporter Assay System -Tripluc- can be used with a microplate-type luminometer with a multi-
colour detection system, which can equip an optical filter (e.g. Phelios AB-2350 (ATTO), ARVO
(PerkinElmer), Tristar LB941 (Berthold)). The optical filter used in measurement is 600–620 nm long
or short pass filter, or 600–700 nm band pass filter.
(1) Measurement of two-colour luciferases with an optical filter.

This is an example using Phelios AB-2350 (ATTO). This luminometer is equipped with a 600 nm
long pass filter (R60 HOYA Co.), 600 nm LP, Filter 1) for splitting SLO (max = 580 nm) and SLR
(max = 630 nm) luminescence.
To determine transmission coefficients of the 600 nm LP, first, using purified SLO and SLR
luciferase enzymes, measure i) the intensity of SLO and SLR bioluminescence intensity without filter
(F0), ii) the SLO and SLR bioluminescence intensity that passed through 600 nm LP (Filter 1), and
iii) calculate the transmission coefficients of 600 nm LP for SLO and SLR listed below.
Transmission coefficients Abbreviation Definition

Filter 1
kOR60 The filter’s transmission
SLO Transmission
coefficient for the SLO
coefficients
Filter 1 The filter’s transmission
SLR Transmission kR R60
coefficient for the SLR
coefficients
When the intensity of SLO and SLR in test sample are defined as O and R, respectively, i) the
intensity of light without filter (all optical) F0 and ii) the intensity of light that transmits through 600
nm LP (Filter 1) F1 are described as below.
F0=O+R
F1=OR60 x O + RR60 x R
These formulas can be rephrased as follows:
F0 1 1 O
( )=( )( )
F1 κ𝑂𝑅60 κ𝑅𝑅60 R
Then using calculated transmittance factors (OR60 and RR60) and measured F0 and F1, you can
calculate O and R-value as follows:
−1
O 1 1 F0
( )=( ) ( )
R κ𝑂𝑅60 κ𝑅𝑅60 F1
Materials and methods for determining transmittance factor
© OECD 2018
(1) Reagents
∙ Single purified luciferase enzymes:
Lyophilised purified SLO enzyme
Lyophilised purified SLR enzyme
(which for the validation work were obtained from GPC Lab. Co. Ltd., Tottori, Japan with THP-G8
cell line)
∙ Assay reagent:
Tripluc® Luciferase assay reagent ( for example from TOYOBO Cat#MRA-301)
∙ Medium: for luciferase assay (30 ml, stored at 2 – 8°C)
Final conc. in Required

Reagent Conc.
medium amount
RPMI-1640 - - 27 ml
FBS - 10 % 3 ml
(2) Preparation of enzyme solution

Dissolve lyophilised purified luciferase enzyme in tube by adding 200 μl of 10 ~ 100 mM Tris/HCl or
Hepes/HCl (pH 7.5 ~ 8.0) supplemented with 10% (w/v) glycerol, divide the enzyme solution into 10
μl aliquots in 1.5 ml disposable tubes and store them in a freezer at -80°C. The frozen enzyme
solution can be used for up to 6 months. When used, add 1 ml of medium for luciferase assay (RPMI-
1640 with 10% FBS) to each tube containing the enzyme solutions (diluted enzyme solution) and
keep them on ice to prevent deactivation.
(3) Bioluminescence measurement

Thaw Tripluc® Luciferase assay reagent (Tripluc) and keep it at room temperature either in a water
bath or at ambient air temperature. Power on the luminometer 30 min before starting the measurement
to allow the photomultiplier to stabilise. Transfer 100 μl of the diluted enzyme solution to a black 96
well plate (flat bottom) (the SLO reference sample to #B1, #B2, #B3, the SLR reference sample to
#D1, #D2, #D3). Then, transfer 100 μl of pre-warmed Tripluc to each well of the plate containing the
diluted enzyme solution using a pipetman. Shake the plate for 10 min at room temperature (about
25°C) using a plate shaker. Remove bubbles from the solutions in wells if they appear. Place the plate
in the luminometer to measure the luciferase activity. Bioluminescence is measured for 3 sec each in
the absence (F0) and presence (F1) of the optical filter.
Transmission coefficient of the optical filter was calculated as follows:
Transmission coefficient (SLO (OR60))= (#B1 of F1+ #B2 of F1+ #B3 of F1) / (#B1 of F0+ #B2 of
F0+ #B3 of F0)
Transmission coefficient (SLR (RR60))= (#D1 of F1+ #D2 of F1+ #D3 of F1) / (#D1 of F0+ #D2 of
F0+ #D3 of F0)
Calculated transmittance factors are used for all the measurements executed using the same
luminometer.
Quality control of equipment
The procedures described in the IL-8 Luc protocol should be used (18).
.
© OECD 2018
APPENDIX III
Prior to routine use of the test method described in this Annex to Test Guideline 442E, laboratories
should demonstrate technical proficiency by obtaining the expected IL-8 Luc assay prediction for the 9
substances recommended in Table 1 and by obtaining values that fall within the respective reference
range for at least 8 out of the 9 proficiency substances (selected to represent the range of responses for
skin sensitisation hazards). Other selection criteria were that the substances are commercially available,
and that high-quality in vivo reference data as well as high quality in vitro data generated with the IL-8
Luc assay are available. Also, published reference data are available for the IL-8 Luc assay (6) (1).
Table 1: Recommended substances for demonstrating technical proficiency with the IL-8 Luc assay
Solubility Reference range
in (μg/mL) 3
X- In vivo IL-8 Luc
Proficiency substances CAS no. State
VIVO15 prediction1 prediction2 IL-8 Luc
CV054
at 20 MIT5
mg/mL
Sensitiser
2,4-Dinitrochlorobenzene 97-00-7 Solid Insoluble Positive 2.3-3.9 0.5-2.3
(Extreme)
Sensitiser
Formaldehyde 50-00-0 Liquid Soluble Positive 9-30 4-9
(Strong)
Sensitiser
2-Mercaptobenzothiazole 149-30-4 Solid Insoluble Positive 250-290 60-250
(Moderate)
Sensitiser
Ethylenediamine 107-15-3 Liquid Soluble Positive 500-700 0.1-0.4
(Moderate)
Ethyleneglycol Sensitiser
97-90-5 Liquid Insoluble Positive >2000 0.04-0.1
dimethacrylate (Weak)
Sensitiser
4-Allylanisole (Estragol) 140-67-0 Liquid Insoluble Positive >2000 0.01-0.07
(Weak)
Non-
Streptomycin sulphate 3810-74-0 Solid Soluble Negative >2000 >2000
sensitiser
Non-
Glycerol 56-81-5 Liquid Soluble Negative >2000 >2000
sensitiser
Non-
Isopropanol 67-63-0 Liquid Soluble Negative >2000 >2000
sensitiser
Abbreviations: CAS no. = Chemical Abstracts Service Registry Number

1
The in vivo potency is derived using the criteria proposed by ECETOC (19).
2
3
CV05 and IL-8 Luc MIT were calculated using water solubility given by EPI SuiteTM.
4
CV05: the minimum concentration at which chemicals show less than 0.05 of Inh-GAPLA.
5
MIT: the lowest concentrations at which a chemical satisfies the positive criteria.
© OECD 2018
APPENDIX IV
INDEXES AND JUDGMENT CRITERIA
nIL8LA (nSLO-LA)
The j-th repetition (j = 1-4) of the i-th concentration (i = 0-11) is measured for IL8LA (SLO-LA) and
GAPLA (SLR-LA) respectively. The normalised IL8LA, referred to as nIL8LA (nSLO-LA), and is
defined as:
nIL8LA ij = IL8LA ij / GAPLA ij.
This is the basic unit of measurement in this assay.
Ind-IL8LA (FInSLO-LA)
The fold increase of the averaged nIL8LA (nSLO-LA) for the repetition on the i-th concentration
compared with it at the 0 concentration, Ind-IL8LA, is the primary measure of this assay. This ratio is
written by the following formula:

Ind-IL8LAi = 1 / 4  nIL8LA /1/ 4 nIL8LA
j ij j 0j .
The lead laboratory has proposed that a value of 1.4 corresponds to a positive result for the tested
chemical. This value is based on the investigation of the historical data of the lead laboratory. Data
management team then used this value through all the phases of validation study. The primary outcome,
Ind-IL8LA, is the ratio of 2 arithmetic means as shown in equation.
95% confidence interval (95% CI)

The 95% confidence interval (95% CI) based on the ratio can be estimated to show the precision of
this primary outcome measure. The lower limit of the 95% CI  1 indicates that the nIL8LA with the i-th
concentration is significantly greater than that with solvent control. There are several ways to construct
the 95% CI. We used the method known as Fieller’s theorem in this study. This 95% confidence interval
theorem is obtained from the following formula:
  B  B 2  4AC  B  B 2  4AC 
 , ,
 2A 2A 
2
sd02 sdyi
where A  x 0  t 0.975    , B  2  x  y , C  y i  t 0.975   
2 2 2 2
, and n 0 =4,
n0 n yi
x 0 = 1 / n0   jnIL8LA 0 j , sd02  1 / n0  1 jnIL8LA 0 j  x 0 2 ,
 
n yi =4, y i  1 / n yi  jnIL8LA ij  , sd2y  1 / n y  1  jnIL8LA ij  y i 2 .
i j
t 0.975  is 97.5 percentile of the central t distribution with the  of the degree of freedom, where
© OECD 2018
 sd2 sd2 
2
 sd2  2  sd2yi 
2

 =  0  yi
 n0



/  0  / n 0  1  
 ny 
 
 / ny  1 
.
 0 
 n yi   n  i 
i


Inh-GAPLA (II-SLR-LA)
The Inh-GAPLA is a ratio of the averaged GAPLA (SLR-LA) for the repetition of the i-th
concentration compared with that with solvent control, and this is written by

Inh-GAPLA i = 1 / 4   GAPLA /1/ 4  GAPLA
j ij j 0j .
Since the GAPLA is the denominator of the nIL8LA, an extremely small value causes large variation in
the nIL8LA. Therefore, Ind-IL8LA values with an extremely small value of Inh-GAPLA (less than 0.05)
might be considered poor precision.
© OECD 2018
APPENDIX V
THE SCHEME OF THE METHODS TO DISSOLVE CHEMICALS FOR THE IL-8 LUC
ASSAY.
(a) For chemicals dissolved in X-VIVOTM 15 at 20 mg/mL
If the chemical is soluble in X-VIVOTM 15

(1st run ) a dilution factor of 5
20 mg/mL Stock solution
a dilution factor of 2 (4 mg/mL)
1 mg/mL 4 mg/mL
0 2 mg/mL
Add to cell suspension in a 96 well plate (50 mL : 50 mL)
0.5 mg/mL 2 mg/mL

0 1 mg/mL
Determine the highest concentration of the following experiments
CV05; the lowest concentration at which Inh-

Inh-GAPLA
GAPLA becomes <0.05 (the concentration of

stock solution x 1/dilution factor (X) )
0.05
Final concentration in 2nd, 3rd and 4th experiment

1 mg/mL 4 mg/mL
0 0.5 mg/mL 2 mg/mL
2nd, 3rd or 4th run

dilute
x1 4xCV05 (4 x the concentration
X-VIVOTM 15 control
a dilution factor of 1.5 of stock solution x 1/X )
Addition to cells in a 96 well plate (50 mL : 50 mL)

2xCV05
© OECD 2018
(b) For chemicals insoluble in X-VIVOTM 15 at 20 mg/mL
If the chemical is insoluble in X-VIVOTM 15

(1st run) Rotate and
20 mg/mL centrifuge Supernatant
x1 (Stock solution)
A dilu on factor of 2
x1/1024 x1/256 x1/64 x1/16 x1/4 x1

0 x1/512 x1/128 x1/32 x1/8 x1/2
Add to cell suspension in a 96 well plate (50 mL : 50 mL)
x1/2048 x1/512 x1/128 x1/32 x1/8 x1/2

0 x1/1024 x1/256 x1/64 x1/16 x1/4 Final dilu on
Inh-GAPLA
CV05; the lowest concentration at which Inh-

GAPLA becomes <0.05 (the concentration of
stock solution x 1/dilution factor (X) )
0.05
Final concentration in 2nd, 3rd and 4th experiment

x1/2048 x1/8 x1/2
0 x1/1024 x1/16 x1/4
1/X
2nd, 3rd or 4th run

dilute
X-VIVOTM 15 control x1 4xCV05 (4 x the concentration

a dilution factor of 1.5 of stock solution x 1/X )
Addition to cells in a 96 well plate (50 mL : 50 mL)

2xCV05
© OECD 2018
APPENDIX VI
THE SCHEME OF THE METHOD TO DISSOLVE 4-NBB FOR THE POSITIVE CONTROL
OF THE IL-8 LUC ASSAY.
The positive control : 4-NBB (insoluble in X-VIVOTM 15)
a dilution factor of 4
Rotate and
centrifuge Supernatant x1/4
20 mg/ml
x1
dilution at factor of 2
x1/16 x1/8 x1/4

X-VIVOTM 15
control
Addition to cells in
a 96 well plate (50 mL : 50 mL)
X-VIVOTM 15 x1/32 x1/16 x1/8

control
© OECD 2018

Oecd/Ocde 442E: Key Event-Based Test Guideline

Uploaded by

Copyright:

Available Formats

Oecd/Ocde 442E: Key Event-Based Test Guideline

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Oecd/Ocde 442E: Key Event-Based Test Guideline

Uploaded by

Copyright:

Available Formats

OECD/OCDE 442E

KEY EVENT-BASED TEST GUIDELINE

IN VITRO SKIN SENSITISATION ASSAYS ADDRESSING THE KEY

Activation of dendritic cells Key Event based Test Guideline

1. United Nations UN (2015), Globally Harmonized System of Classification and Labelling of

ANNEX I: IN VITRO SKIN SENSITISATION: HUMAN CELL LINE

INITIAL CONSIDERATIONS AND LIMITATIONS

3. The h-CLAT method proved to be transferable to laboratories experienced in cell culture

6. Definitions are provided in Appendix I.

PRINCIPLE OF THE TEST

Dose finding assay

Preparation of test chemicals and control substances

Application of test chemicals and control substances

Propidium iodide (PI) staining

Cytotoxicity measurement by flow cytometry and estimation of CV75 value

Number of living cells × 100

(75 – c) × Log (b) – (75 – a) × Log (d)

a is the minimum value of cell viability over 75%

Preparation of the test chemicals and control substances

Application of test chemicals and control substances

Cell staining and analysis

DATA AND REPORTING

MFI of chemical-treated cells − MFI of chemical-treated isotype control

MFI of solvent/vehicle-treated ctrl cells − MFI of solvent/vehicle-treated isotype ctrl

31. The test report should include the following information.

additional relevant physicochemical properties, to the extent available;

- Multi-constituent substance, UVCB and mixture

- Negative and solvent/vehicle control

Test method conditions

Test acceptance criteria

- Number of runs used;

Discussion of the results

- Discussion of the results obtained with the h-CLAT method;

1. Ashikaga T, Yoshida Y, Hirota M, Yoneyama K, Itagaki H, Sakaguchi H, Miyazawa M, Ito Y, Suzuki

4. Takenouchi O, Fukui S, Okamoto K, Kurotani S, Imai N, Fujishiro M, Kyotani D, Kato Y, Kasahara T,

11. Nukada Y, Ashikaga T, Miyazawa M, Hirota M, Sakaguchi H, Sasa H, Nishiyama N. (2012),

15. Ashikaga T, Sakaguchi H, Sono S, Kosaka N, Ishikawa M, Nukada Y, Miyazawa M, Ito Y,

CV75: The estimated concentration showing 75% cell viability.

Pre-haptens: chemicals which become sensitisers through abiotic transformation

Pro-haptens: chemicals requiring enzymatic activation to exert skin sensitisation potential

UVCB: substances of unknown or variable composition, complex reaction products or biological

CV75 h-CLAT results h-CLAT results

Sensitiser Positive Negative

Sensitiser Positive Positive

Sensitiser Negative Positive

Sensitiser Negative Positive

ANNEX II: IN VITRO SKIN SENSITISATION:

INITIAL CONSIDERATIONS AND LIMITATIONS

3. The U-SENS™ method proved to be transferable to laboratories experienced in cell culture

6. Definitions are provided in Appendix I.

PRINCIPLE OF THE TEST

Preparation of test chemicals and control substances

Application of test chemicals and control substances

Flow cytometry analysis

Number of living cells × 100

DATA AND REPORTING

CV70 is calculated by log-linear interpolation using the following equation:

CV70 = C1 + [(V1 - 70) / (V1 – V2) * (C2 – C1)]

EC150 is calculated by log-linear interpolation using the following equation: