Authors

A Willms¹; W Schaefer¹; LM Philipp²; S Sebens²; B Werdelmann¹; A Guledani¹; M Stoehr¹; R Geisen¹; M Pirsch¹; 
¹ SYNENTEC GmbH, Germany;  ² Institute for Experimental Cancer Research Kiel, Germany

Discussion

Stem cells are a powerful tool in different research fields due to their high proliferative and self-renewal capabilities. To determine stem cell properties in vitro, colony formation assays (CFAs) are commonly used. Here, single cells are seeded, and the resulting colonies are distinguished into three
colony types termed holo-, mero- and paraclones. Holoclones occur as large colonies with a smooth, circular perimeter. They have an extensive proliferative potential and self-renewal capacity and are regarded to derive from stem cells. In contrast, paraclones are supposed to originate from late-stage transit-amplifying cells generating small colonies with an irregular boundary and are limited to further proliferate. Meroclones have an irregular outline, representing an intermediate type, which forms growing and terminal colonies.

In classical CFAs, the clones are fixed, stained and manually evaluated under a microscope. Since this method is time-consuming, and colony discrimination requires training and experience, we aimed to automate this assay using our imager CELLAVISTA^®, YT-SOFTWARE^® and our automation system. For this purpose, we performed single cell cloning experiments of cancer cells and automatically monitored colony growth over-time. Once colonies had formed, they were automatically detected and classified into the three colony types using the image analysis application Single Cell Cloning (Holo, Mero, Para) of YT-SOFTWARE^®. To validate the software-based classification, we analyzed isolated holo- and paraclones for the expression of stem cell markers CD44 and Nestin by immunofluorescence stainings using NYONE^® Scientific, CELLAVISTA^® and the Virtual Cytoplasm (1F) application. Stainings revealed a higher expression of CD44 and Nestin in holoclones compared to paraclones.

Our application simplifies CFAs enormously making it a great tool for research and drug discovery.