A two-channel hypothesis for regulation of cell division and differentiation

We propose the hypothesis that extracellular regulation of cell division and differentiation acts through just two communications channels. These channels consist of a series of redundant components: extracellular messenger hormones; these hormones' receptors; cytoplasmic proteins activated by the hormone-receptor complex; and trans-activating nuclear regulatory proteins. One channel, here labeled "D" ("differentiate"), includes transforming growth factor-beta as one of its hormones; the other, labeled "G" ("growth") includes epidermal growth factor. We postulate that signal reception occurs as a result of competition between different actuating proteins for equilibrium-controlled binding to critical DNA sites. Stem cells commit to mitosis when some high proportion of critical sites is occupied by actuating proteins of the G class, and to terminal differentiation when a high proportion is occupied by "D" actuators. Intermediate occupancy can either lead to division into one differentiated and one stem cell, or to maintenance of cells in the reference state, quiescence. Equilibrium control of binding implies that critical site occupancy will be proportional to the relative concentrations of "D" and "G" actuating proteins in the nuclear fluid. These concentrations depend on the external hormone concentrations, the numbers of receptors on the cell membrane, and the coefficients of the rate-determining steps between internalization of the hormone-receptor complexes and activation of the actuating proteins. All of these quantities can be affected by various factors, including endocrine hormones. This model is consistent with most reported behavior of various growth factors, interferons, etc, toward a variety of cells in culture. It predicts that under otherwise constant conditions, high relative concentrations of a D-hormone (e.g. transforming growth factor-beta) will induce commitment to terminal differentiation, while high relative concentrations of a "G" hormone (e.g. epidermal growth factor) will induce mitosis. We have seen no report of an experiment which adequately tests this prediction. The model implies that cancer causing mutations are those which increase the relative intensity of the "G" signal; this can occur via changes in components of either channel. Such mutant cells should be both more likely to divide and less likely to differentiate than normal stem cells. Conversely, mutations which increase relative sensitivity to the "D" signal during embryonal development can lead to premature differentiation, cessation of growth, and structural abnormalities (terata).