Br. J. Cancer (1983), 48, 863-868 Short Communication Potential biological explanation of stimulation of colony growth in semi-solid agar by cytotoxic agents F.L. Meyskens, Jr., S.P. Thomson, R.A. Hickiel & N.J. Sipes Cancer Center Division and Department of Internal Medicine, University of Arizona, Tucson, Arizona 85724, USA and 'Department of Pharmacology, University of Saskatchewan, Saskatoon. Canada S7N OWO. Growth of single cells into colonies in semisolid medium has been widely used both as a measure of the clonogenic potential of normal and transformed cells (Park et al., 1971; Thomson & Rauth, 1974; Courtenay, 1976; Metcalf, 1977; Buick et al., 1977; Hamburger et al., 1977; Salmon, 1980) and as an in vitro marker for cellular transformation (Macpherson & Montagnier, 1964; MacAllister & Reed, 1968). Growth of cells in semisolid medium also has been used to measure the effect of various cytotoxic and non-cytotoxic agents on clonogenic tumour cells (Salmon et al., 1978; Tveit et al., 1980, 1982; Von Hoff et al., 1981; Meyskens et al., 1981). The relation between the number of cells plated and the formation of clusters and colonies, defined here as growth units, should be clearly defined to assure valid interpretation of perturbations of clonogenic cells. The frequency and extent of proliferation to form growth units has been regarded as a function of the clonogenic capacity of the tumour sample together with the conditions of culture. However, many of the effects of the conditions of culture remain undefined. Therefore, because clonogenic assays are generally closed non re-fed systems, we have examined the relationship between the number of cells plated, the number of growth units formed, the relative frequency of growth units containing different numbers of cells, and the total number of cells formed within the growth units. We found that the cloning efficiency and proliferative characteristics of clonogenic cells in agar is significantly determined by the number of cells plated. Murine melanoma cells (CCL 53.1) were grown in medium (FIO plus 10% horse and 2% heat- inactivated foetal calf-serum) as monolayers in plastic Falcon flasks. Treatment with Tryodes solution removed the cells and produced suspensions of single cells. Different numbers of cells were plated in 1.0ml of medium containing Correspondence: F.L. Meyskens. Received 10 May 1983; accepted 6 September 1983. 0.3% agar (Bacto) over 1.0ml of 0.5% agar in medium in 35mm diameter Petri dishes. The plates were incubated in a humidified, 5% Co2, air atmosphere at 37°C for 18 days. Every group of cells (containing greater than one cell) was counted in randomly selected 1, 2.5, or 6.25 mm2 areas. The mean +s.e. number of growth units per area was multiplied by 908 mm2 per 35mm culture dish to obtain the total number of growth units per culture dish. The number of growth units of different diameters was calculated using the relative frequency of growth units by size, obtained by direct measurement of 200 growth units per plate with a micrometer, and multiplying by the total number of growth units per plate. A nomogram was constructed to determine the number of cells per growth unit. This was done by direct visual observation in growth units containing 80pm in diameter. As the number of cells plated was decreased, the number of larger growth units increased. For example, when 500 cells were plated, only 10% of colonies were The Macmillan Press Ltd., 1983
Read full abstract