Purpose: A novel method for accurate measurement of cell colony growth in complex in-vitro culture conditions is presented, suitable for radiation cell survival and growth assays involving fluorescing cells.
Methods: An Algorithm using CellProfiler was created to measure mCherry-labelled 4T1 murine mammary carcinoma cell colonies grown in 3D organotypic culture on rat brain slices. The accuracy of the algorithm was investigated by application to realistic simulated image sets. 10,000 mCherry-labelled 4T1 were plated on 400 µm rat brain slices on top of transwell-inserts within 6 well plates. The cells were imaged once a day for 5 days filtered for the mCherry signal. Representative simulated images corresponding to days 1-5 were created using MATLAB with similar colony intensity, distribution, size distribution, variable background, colony overlap, and background noise. A CellProfiler pipeline was developed to measure the integrated intensity of the mCherry signal. Parameters in the pipeline were varied to determine an optimal pipeline which gave values closest to ground truth values for the simulations. The optimized CellProfiler pipeline was then applied to the original 4T1 tumor cell images to determine colony growth over five days.
Results: Key parameters to be optimized in the pipeline include colony diameter, minimum intensity, background subtraction method and size. The optimized CellProfiler pipeline performed extremely well on 10 realistic simulated data sets, with an accuracy of 99.23% ± 0.75%. Further analysis on images with increasing colonies, background, and noise showed the pipeline was accurate for variable features. Analysis of the 4T1 real cell image data showed an increase in integrated intensity following day 2 approximately consistent with the 13.6±1.5 hour doubling time for 4T1 cells.
Conclusion: The new algorithm has strong potential to bring much needed improvement in rigor and systematic consistency to the counting of fluorescing cell colonies in radiation survival and cell growth assays.