Estimation of cell cycle parameters from double labeling experiments

Double labeling of cell populations with radioactive thymidine yields two types of differently labeled nuclei. Their numbers and the number of unlabeled nuclei can be used to estimate doubling times, T, and S-phase lengths, S. As of yet, such estimations have been performed either for stationary populations in which proliferation and losses are in balance, or for exponentially growing populations in which all cells have the same cycle duration. We calculate S and T for the more general type of cell population with arbitrarily distributed frequencies of cycle durations. The calculations do not require more mathematical or computational effort. We obtain three main results: (i) The estimation of T and S does not require explicit knowledge of the frequency distribution of cycle durations; (ii) in particular, equivalent estimates for T and S are obtained for both types of growing cell populations without losses, one with arbitrarily distributed cycle durations and one with the same cycle duration for all cells; and (iii) for small labeling indices, the estimate for S from the general model approaches the S-phase length of a stationary population and the estimate for T from the general model approaches the generation time of a stationary population, multiplied by the constant factor 1n(2). These relationships are valuable tools for reinterpreting results derived under the assumption of stationarity, which are considerably easier to obtain.