Cytometry of asynchronous proliferating cell populations produces data with an extractable time-based feature embedded in the frequency of clustered, correlated events. Here, we present a specific case of general methodology for calculating dynamic expression profiles of epitopes that oscillate during the cell cycle and conversion of these values to the same scale.

The timing of DNA synthesis, mitosis and cell division is regulated by a complex network of biochemical reactions that control the activities of a family of cyclin-dependent kinases. The temporal dynamics of this reaction network is typically modeled by nonlinear differential equations describing the rates of the component reactions. This approach provides exquisite details about molecular regulatory processes but is hampered by the need to estimate realistic values for the many kinetic constants that determine the reaction rates. It is difficult to estimate these kinetic constants from available experimental data. To avoid this problem, modelers often resort to 'qualitative' modeling strategies, such as Boolean switching networks, but these models describe only the coarsest features of cell cycle regulation. In this paper we describe a hybrid approach that combines the best features of continuous differential equations and discrete Boolean networks. Cyclin abundances are tracked by piecewise linear differential equations for cyclin synthesis and degradation. Cyclin synthesis is regulated by transcription factors whose activities are represented by discrete variables (0 or 1) and likewise for the activities of the ubiquitin-ligating enzyme complexes that govern cyclin degradation. The discrete variables change according to a predetermined sequence, with the times between transitions determined in part by cyclin accumulation and degradation and as well by exponentially distributed random variables. The model is evaluated in terms of flow cytometry measurements of cyclin proteins in asynchronous populations of human cell lines. The few kinetic constants in the model are easily estimated from the experimental data. Using this hybrid approach, modelers can quickly create quantitatively accurate, computational models of protein regulatory networks in cells.

B cyclins regulate G2-M transition. Because human somatic cells continue to cycle after reduction of cyclin B1 (cycB1) or cyclin B2 (cycB2) by RNA interference (RNAi), and because cycB2 knockout mice are viable, the existence of two genes should be an optimization. To explore this idea, we generated HeLa BD™ Tet-Off cell lines with inducible cyclin B1- or B2-EGFP that were RNAi resistant. Cultures were treated with RNAi and/or doxycycline (Dox) and bromodeoxyuridine. We measured G2 and M transit times and 4C cell accumulation. In the absence of ectopic B cyclin expression, knockdown (kd) of either cyclin increased G2 transit. M transit was increased by cycB1 kd but decreased by cycB2 depletion. This novel difference was further supported by time-lapse microscopy. This suggests that cycB2 tunes mitotic timing, and we speculate that this is through regulation of a Golgi checkpoint. In the presence of endogenous cyclins, expression of active B cyclin-EGFPs did not affect G2 or M phase times. As previously shown, B cyclin co-depletion induced G2 arrest. Expression of either B cyclin-EGFP completely rescued knockdown of the respective endogenous cyclin in single kd experiments, and either cyclin-EGFP completely rescued endogenous cyclin co-depletion. Most of the rescue occurred at relatively low levels of exogenous cyclin expression. Therefore, cycB1 and cycB2 are interchangeable for ability to promote G2 and M transition in this experimental setting. Cyclin B1 is thought to be required for the mammalian somatic cell cycle, while cyclin B2 is thought to be dispensable. However, residual levels of cyclin B1 or cyclin B2 in double knockdown experiments are not sufficient to promote successful mitosis, yet residual levels are sufficient to promote mitosis in the presence of the dispensible cyclin B2. We discuss a simple model that would explain most data if cyclin B1 is necessary.