Intercellular propagation of individually programmed growth bursts in FRTL-5 cells. Implications for interpreting...

Five methods are commonly used to quantify FRTL-5 cells' and other thyrocytes' growth in vitro and the impact of growth inhibiting or stimulating maneuvers: Total cell count, mitotic index, DNA measurement, total [3H]thymidine incorporation, and the fraction of [3H]thymidine labeled cells. All of them assess cell growth as though all cells were homogeneous with an identical response to growth factors. We demonstrate here that this assumption is not valid. Rather, some intrinsically growth-prone cells appear to pass a growth signal to neighboring cells so that variably sized colonies of synchronized cells within each cluster growing from monodispersed cells are formed. This is true for FRTL-5 cells growing in vitro in monolayers and in three-dimensional, collagen embedded spheroids. The pattern is the same when cell suspensions or collagen-embedded spheroids are implanted onto nude mice. Patches with alternating high and low growth become particularly prominent in the large tumor-like organoids grown from monodispersed cells in nude mice. The pattern much reminds of similar observations in growing intact thyroids. Since there is no significant correlation between the fraction of [3H]thymidine labeled cells and the size of two- or three-dimensional clusters in any experiment, growth of signal-spreading cells is assumed to occur in leaps and bounds. Growth velocity in each subclone of a cell population depends on the mean interval between bursts of replications and on the number of cells synchronized by cell-to-cell diffusion of the growth signal emanating from one dividing cell. Thus, growth-promoting and growth-inhibiting factors may not only act on the mean interval between successive growth bursts, but they may also change cell-to-cell spreading of growth signals.