Improving the reproducibility of the MCF-7 cell proliferation assay for the detection of xenoestrogens

The MCF-7 cell proliferation assay is potentially a simple and highly reproducible tool for the identification of estrogenic compounds. However, its widespread use has been complicated by the lack of a standardised protocol, resulting in considerable inter-laboratory variability. We have explored the sources of variability both in relation to cell lines and test regimens and report on optimised procedures for the identification of estrogenic agents. Two supposedly identical MCF-7 parent cen lines (designated UCL and SOP), and the BUS subline were cultured according to an existing protocol, and responses to 17-estradiol (E2) assessed. Despite yielding almost identical EC,, values, the proliferative response varied widely between cell lines from 0.98-fold over controls (UCL) to 8.9-fold (BUS) indicating major differences between them. The underlying causes may be genetic, and to assess this we used comparative genomic hybridisation (CGH), a technique which allows the detection of DNA sequence copy number changes on a genome-wide scale. Although numerous similarities existed between the different cell lines, the least oestrogen-responsive line (MCF-7/UCL) exhibited the greatest number of cytogenetic changes, many of which were not seen in MCF-7/SOP cells. We suggest that care must be taken, therefore, when choosing a cell line for MCF-7 cell-based experiments. Selecting the MCF-7/SOP line for further work, we carried out a thorough and systematic optimisation of the MCF-7 cell proliferation assay, finding that a 72-h period in oestrogen-free medium before treatment strongly influenced the cells response to E2. With 1 nM E2, proliferation increased from 1.5-fold to 6.5-fold relative to vehicle-treated controls, a response similar to that seen with MCF-7/BUS cells in the E-SCREEN protocol devised by Soto et al. With parent MCF-7 cells, other laboratories have reported only 4.5-fold increases as maximal. Here we present evidence that the choice of cell line and culture conditions are crucial in determining test outcomes, and once chosen and adhered to the assay yields reproducible results. (C) 2000 Elsevier Science B.V. All rights reserved.