Background: Flow cytometry of immunofluorescence and DNA content provides measures of cell-cycle-related gene expression (protein and/or epitope levels) for asynchronously growing cells. From these data, time-related expression through S phase can be directly measured. However, for G1, G2, and M phases, this information is unavailable. We present an objective method to model G1 and G2 kinetic expression from an estimate of a minimum biological unit of positive immunofluorescence derived from the distribution of specific immunofluorescence of mitotic cells.
Methods: DU 145 cells were stained for DNA, cyclin B1, and a mitotic marker (p105) and analyzed by flow cytometry. The cyclin B1 immunofluorescence (B1) distribution of p105-positive cells was used to model the B1 distribution of G2 and G1 cells. The G1/S and S/G2 interface measurements were used to calculate expression in S phase and test the validity of the approach.
Results: B1 at S/G2 closely matched the earliest modeled estimate of B1 in G2. B1 increased linearly through G1 and S but exponentially through G2; mitotic levels were equivalent to the highest G2 levels. G1 modeling of B1 was less certain than that of G2 due to low levels of expression but demonstrated general feasibility.
Conclusions: By this method, the upper and lower bounds of cyclin B1 expression could be estimated and kinetic expression through G1, G2, and M modeled. Together with direct measurements in S phase, expression of B1 throughout the entire cell cycle of DU 145 cells could be modeled. The method should be generally applicable given model-specific assumptions.