Applying Design of Experiment methodology to PEI toxicity assay on neural progenitor cells

Abstract Design of Experiment (DoE) statistical methodology permits the simultaneous evaluation of the effects of different factors on experimental performance and the analysis of their interactions in order to identify their optimal combinations. Compared to classical approaches based on changing only one factor at a time (OFAT), DoE facilitates the exploration of a broader range of parameters combinations, as well as providing the possibility to select a limited number covering the whole frame. The advantage of DoE is to maximise the amount of information provided and to save both time and money. DoE has been primarily used in industry to maximise process robustness, but recently it has also been applied in biomedical research to different types of multivariable analyses, from determination of the best cell media composition to the optimisation of entire multi-step laboratory protocols such as cell transfection. Our case study is the optimisation of a transfection protocol for neural progenitor cell lines. These cells are very hard to transfect and are refractory to lipidic reagents, so we decided to set-up a protocol based on the non-lipidic Poliethylenimine (PEI) reagent. However, the effect of PEI toxicity on cells has to be correctly evaluated in the experimental design, since it can affect output computation. For this reason, we decided to apply DoE methodology to investigate the effect of PEI, both concentration and type, on cell viability and its interaction with other factors, such as DNA and cell density. The statistics-based DoE approach allowed us to express analytically the neural cell viability dependence on PEI amount/cell and efficiently identify the dose levels of PEI suitable for transfection experiments.