The purpose of this study was to compare the characteristics and irritation potentials of three commercial reconstructed 3D skin models based on human keratinocytes as alternative skin irritation test, and to perform a proficiency test using one model...
The purpose of this study was to compare the characteristics and irritation potentials of three commercial reconstructed 3D skin models based on human keratinocytes as alternative skin irritation test, and to perform a proficiency test using one model. The 3D skin models used in the validation study were: KeraSkinTM-VM (Biosolution, Korea), EpiDermTM SIT (EPI-200, MatTek Corparation, USA) and LabCyte EPI-MODEL24 (J-Tec, Japan).
The test substances were applied on the surface of insert cells after a pre-incubation period of 24 hrs. The test substances were completely washed, and then models were post-incubated for 48 hrs. Cell viability was measured by MTT assay to find out if test substances had caused skin irritation.
In the preliminary test, three irritants and three non-irritants were selected from the recommended list of OECD Guidelines 439 (2013 version). In order to validate the proficiency test, additional test substances were used for the 2nd test in a LabCyte model. The six test substances used for the 1st test were: Diethyl phathdate, 4-methylthio-benzaldehyde, 1-decanol, heptanal, methyl stearate, and 1-methyl-3-phenyl-1-piparazi, and the additional seven substances used for proficiency test of the LabCyte model were: naphthalene acetic acid, isopropanol, heptyl butyrate, hexyl salicylate, cyclamen aldehyde, 1-bromohexane, and potassium hydroxide (5% aq.).
We used PBS, DPBS and DW as negative controls, and 5% sodium lauryl sulphate (SLS) as a positive control in according to the OECD guidelines 439 (2013).
Quality control indices of models, such as barrier function (TEER), absorbance of negative control, cell viability of positive control, absorbance of extraction buffer used in MTT assay, acceptability range (upper and lower limit), and results range and consistency between inserts, were in agreement with the acceptance criteria.
Most test substances showed consistent results with the GHS standards in OECD Guideline 439, except the 4-methylthio-benzaldehyde (negative substance in GHS) that showed false positives in all 3 models of the preliminary study. In terms of the proficiency test, in the LabCyte model, all of the 10 test substances showed consistent results with OECD Guideline 439.
Taken all together, results for sensitivity, specificity, and accuracy of the three models were 100, 67, and 83%; however, results from the proficient study, the LabCyte model, had 100% sensitivity, specificity, and accuracy-results that were on par with OECD Guideline.
Our present results indicate that the reconstructed 3D skin model is a good alternative model to a skin irritation test, and the Korean KeraSkinTM-VM model would be a good international model provided that several problems were improved. The LabCyte model could also be considered a suitable alternative model to skin irritation test, given its quality and price.