A cost-effective method for the automatic quantitative analysis of fibroblastic colony-forming units

A great deal of the work characterizing stromal cell precursors in the bone marrow has been performed using the fibroblastic colony-forming unit (CFU- f) assay. However, the assay is limited in its usefulness by the necessity for manual colony counting which means that assay quantitation is highly su bjective, time consuming, and much information regarding the colony size is lost. To rectify this, we have developed a computer-automated method for t he analysis of CFU-f. Bone marrow cells were cultured at low density and tr eated with either prostaglandin E-2 (PGE(2)), basic fibroblast growth facto r (bFGF), or dexamethasone, and colony formation was assessed by staining w ith methylene blue. After staining, the dishes were photographed over a Lig ht box using a digital camera and the image was then analyzed using Bioimag e "Intelligent Quantifier" image analysis software which automatically loca tes and quantifies each individual colony. The data can then be imported to a spreadsheet program and processed. We have shown that this system can ac curately identify, assign coordinates, and quantitate each individual colon y. Colony numbers obtained with this method and manually counting showed a linear relationship with a correlation coefficient of 0.99. In addition, us ing the colony intensity and surface area data, the colony size can be calc ulated. With this methodology, we have shown that dexamethasone, PGE(2),and bFGF can all modulate total cell numbers in bone marrow stromal cells (BMS C) cultures but modulating both colony number and colony size.