ROBUSTESS OF THE CALUX BIOASSAY: STATISTICAL AALYSIS OF THE BETWEE-WELL VARIABILITY FOR THE H1L6.1C3 MOUSE HEPATOMA CELL LIE Croes K , Van Langenhove K, Elskens M, Goeyens L, Baeyens W Vrije Universiteit Brussel (VUB), Department of Analytical and Environmental Chemistry, Pleinlaan 2, 1050 Brussels, Belgium. Abstract The variation of the CALUX response between different wells on a 96-well plate was investigated for the H1L6.1c3 mouse hepatoma cell line. Analysis of 6 plates, spiked with 1.5 pg/µL of a 2, 3, 7, 8 -TCDD standard solution, on 3 different days indicated significant differences in RLU values between the wells. The mean RLU values were significantly lower (p = 1.43E-12) in the outer circle of the plate (i.e. rows A and H, and columns 1 and 12) and there were significant differences found between both columns and rows in a two-way ANOVA (p < 0.001). Even after rejection of the outer wells of the plate, significant differences were still found between the rows, columns (both p-values < 0.001) and circles (p = 0.002). Therefore, it is important to pay attention to the spatial distribution of the samples, standards and blanks on the plate and to take into account the between-well variability on a 96-well plate. Introduction The CALUX (Chemical Activated LUciferase gene eXpression) assay is a reporter gene mammalian cell bioassay that is nowadays often used as a rapid and inexpensive screening and semi-quantitative method for the analysis of PCDD/Fs in several matrices, such as milk 1 , blood plasma 2 , sediments 3 and marine biological matrices 4 . The recombinant cells used in the CALUX bioassay contain a stably transfected AhR-responsive firefly luciferase reporter gene, which responds by the induction of luciferase. The measured luminescence is converted into a bioassay toxic equivalency value (CALUX-TEQ) by the comparison of the response for a given sample to a dose-response curve obtained with 2, 3, 7, 8-TCDD standards 5, 6 . Generally, a four parameters Hill-plot is used to fit a sigmoid curve through the standard solutions and 3 quality control solutions are added in duplicate to the cell plate to calculate a coefficient of variation (CV). Since many samples contain low concentrations of dioxin-like compounds and are often available in small amounts (e.g. in biomonitoring studies), the extracts are sometimes dosed on the cell plate in a single well. Therefore, it is important to have an estimate of the variability. This paper presents between-well variability of the cell response for the H1L6.1c3 mouse hepatoma cell line, both within and between 96-well plates. Materials and Methods Calux assay The cell line used in the bioassay was a modified cell line H1L6.1c3, which was obtained from M. Denison, University of California-Davis, USA. Cell treatment and measurement were based on the protocols described by Windal et al 4 . and the XDS method 4435 7 . The cells were grown in a 75 cm² culture flask containing 20 µL RPMI 1640 supplemented with 8% FCS and 1% penicillin/streptomycin (Gibco, UK). After trypsinizing, the cells were counted and diluted to a concentration of ± 70E4 cells per mL. For a reliable statistical interpretation, six 96-well plates were analyzed Vol. 71, 2009 / Organohalogen Compounds page 000551