Validation Study on Five Cytotoxicity Assays by JSAAE IV. Details of the Colony Formation Assay

The inter-laboratory validation study on 5 cytotoxicity assays conducted by JSAAE has been described in the preceding articles in this issue. Presented here are precise data and the protocols on the colony formation (CF) assay with two cell lines, HeLa S3 (SC) that is common to the other 4 assays and BALB/3T3 A31-1-1 that is frequently used in this assay. The CF assay showed high performance rates with both HeLa S3 (SC) and BALB/3T3 A31-1-1 cell lines. Ninty five percent of submitted data files were acceptable before ED50 calculation in this assay. Variations on negative controls in the assay revealed technical differences among laboratories and the characteristics of cells applied. The CF assay with BALB/3T3 A31-1-1 cells resulted in the lowest median values of log(ED50) for Tween 20, Tween 80 and cetylpyridinium chloride monohydrate. For propylene glycol assayed with HeLa S3 (SC) cells, the lowest median value of log(ED50) was observed. Thus the CF assay was able to detect with the highest sensitivity the least toxic chemical (propylene glycol) and the most severely toxic chemical (cetylpyrimidinium chloride monohydrate). CF assay with HeLa S3 (SC) cells gave the largest power for distinction when observing the most and the least toxic chemicals. However, the toxicities of moderately toxic chemicals were not sharply distinguishable from each other. Comparing the mean hinge-spreads, CF assay with BALB/3T3 A31-1-1 cells showed the largest variation of the log(ED50) values among all assay systems examined in this study. These results suggest that the CF assay is a highly sensitive assay but is influenced by many factors such as culture conditions and techniques of cell handling.