German Title: Zeitaufgelöste Analyse von Zellkoloniewachstum in vitro nach Bestrahlung

PDF, English Download (11MB) | Terms of use

Abstract

Here I developed a novel method to investigate the growth of cell colonies in vitro. The method is inspired by and augments the standard in vitro clonogenic assay (IVCA). While the field of application is radiobiological research, the approach can be applied to any domain where colony growth of adherent cells is of interest. The method utilizes high numbers of time-resolved microscopy image series and hence requires largely automated image data acquisition, image processing, quantitative data extraction and single-colony growth characterization. I designed a multi-step analysis framework to implement these steps. This contrasts with traditional approaches relying on visual examination of cell culture containers and manual classification of cell colonies. This new approach allows yet unattained insights into growth behaviors and growth rates of large numbers of individual cell colonies. In applying the new method to five different cell lines (H3122, H460, RENCA, SAT, UTSCC-5) in different experimental settings, the following main results were found: a) For some of the cell lines, the initial seeding density influences the growth dynamics of the resulting colonies in densities commonly used in standard experiments. b) Pre-experimental cell culture conditions influence the growth dynamics in two tested cell lines (SAT, UTSCC-5) without irradiation. c) Exponential growth rates of two tested cell lines (H3122, RENCA) are normally distributed independent of irradiation dose, but the average growth rate decreases linearly across commonly used doses. d) Some colonies growing from photon-irradiated cells exhibit a distinct delayed abortive growth behavior, as observed for the two analysed cell lines (H3122, RENCA). The frequency of this behavior increases with increasing dose. e) Survival rates, as traditionally determined via the standard IVCA, clearly depend on experimental readout choices, namely the time of readout and the size threshold used to score survival of colonies. My analysis indicates that this dependence emerges from observations c) and d). f) The observed influence of readout choices propagates into relative biological effectiveness quantification for carbon irradiation for three examined cell lines (H460, RENCA, UTSCC-5). Hence, I demonstrate that the presented method can be used to inform experimental design decisions in standard IVCA experiments, to perform robustness analyses on these assays, and to find distinct types of growth behavior. Still, the application in its current form is limited to adherently growing cell lines forming contiguous colonies. In addition, due to the multi-step procedure,some underlying assumptions and methodological decisions need to be made which potentially influence the resulting findings. I discuss these aspects in a dedicated chapter. In future work, potential extensions and combinations with quantitative single-cell analysis methods such as FACS, fluorescent live-cell imaging or single cell omics methods can make this method a cornerstone application to build on in order to understand not only how, but also why colonies grow the way they do. In conclusion, the presented method elucidates colony growth in unprecedented detail. The presented results showcase the potential relevance of these details. However, to establish this method as a standard tool for applied research, a unified analysis framework is necessary to standardize the methodological aspects, from image acquisition to colony growth type classification.