Quantitative analysis of cell cycle phase durations and PC12 differentiation using fluorescent biosensors

Abstract

Cell cycle analysis typically relies on fixed time-point measurements of cells in particular phases of the cell cycle. The cell cycle, however, is a dynamic process whose subtle shifts are lost by fixed time-point methods. Live-cell fluorescent biosensors and time-lapse microscopy allows the collection of temporal information about real time cell cycle progression and arrest. Using two genetically-encoded biosensors, we measured the precision of the G(1), S, G(2) and M cell cycle phase durations in different cell types and identified a bimodal G(1) phase duration in a fibroblast cell line that is not present in the other cell types. Using a cell line model for neuronal differentiation, we demonstrated that NGF-induced neurite extension occurs independently of NGF-induced cell cycle G(1) phase arrest. Thus, we have begun to use cell cycle fluorescent biosensors to examine the proliferation of cell populations at the resolution of individual cells and neuronal differentiation as a dynamic process of parallel cell cycle arrest and neurite outgrowth.

Figure 1

Phase-dependent biosensor localization and combinatorial usage. (A) The G₁ phase biosensor construct consists of the HDHB C-terminus and tdimer2 (a dimeric red fluorescent protein). In G₁ phase (post nuclear envelope formation in the daughter cells) the biosensor is nuclear, as shown in the schematic and transfected HeLa cell (left). As the cell progresses through the cell cycle, the G₁ phase biosensor translocates to the cytoplasm (G₁/S), becoming nuclear excluded in S and G₂ phases. Breakdown of the nuclear envelope in M phase allows for fluorescence to spread throughout the rounded cell, with exclusion only at the condensed chromatin (right). Time is in hours:minutes. (B) The nuclear localized S phase biosensor consists of an NLS (SV40 nuclear location signal), EYFP, a linker (18 hydrophilic amino acids) and PCNA. Epifluorescence time-lapse images of a HeLa cell (top) and a schematic (bottom) show S phase biosensor localization through one cell cycle. The nuclei of the schematic S phase cell have been enlarged to better illustrate puncta formation and their change in morphology, which indicates replicative progression. Time is in hours:minutes. (C) Confocal images of a HeLa cell with a punctate S phase biosensor (green) and cytoplasmic G₁ phase biosensor (red) denotes S phase DNA replication. (D) Single frame analysis of coexpressed G₁ and S phase biosensors allows for the identification of the four phases. When the G₁ phase biosensor is nuclear and the S phase biosensor is nuclear but not punctuate, the cell is in G₁ phase (arrows). A cytoplasmically localized G₁ phase biosensor and a punctate S phase biosensor identify S phase cells (*). G₂ cells have cytoplasmic G₁ phase biosensor fluorescence and non-punctate S phase biosensor fluorescence (circle). Coalignment of the G₁ and S phase fluorescence in the rounded mitotic cell body occurs post nuclear envelope breakdown (square). S phase puncta are not always easy to identify, especially in early and mid S phase (#).

Figure 2

Time-lapse analysis of the combined biosensors measures cell cycle phase durations. (A) NIH3T3 cell example of the G₁ (top, red) and S (bottom, green) phase biosensors in all 4 phases: (G₁) nuclear G₁ and S phase biosensors, (S) cytoplasmic G₁ phase biosensor and punctate S phase biosensor, (G₂) cytoplasmic G₁ phase biosensor and non-punctate nuclear S phase biosensor, and (M) colocalized G₁ and S biosensors filling the entire rounded cell. Durations of the phases (bar graph) were scored for cells expressing both biosensors and had at least one frame in the preceding phase and one frame in the subsequent phase. Error bars are s.e.m. *p < 0.05, **p < 0.01, ***p < 0.001. (B) Histograms of the number of minutes spent by each cell type in G₁ phase. (C) Histograms of the number of minutes spent by each cell type in S phase. Dashed lines are the mean durations of completed G₁ (B) or S (C) phases for the three cell types.

Figure 3

Serum and NGF treated PC12 cells have a marked difference in cell cycle progression. (A) Representative traces of PC12 cells serum starved for 24 hours then treated with serum and imaged after 24 hours for the next 37 hours. (B) Average phase durations of the PC12 cells (shown in A) were tabulated when at least one frame of the preceding and one from the following phase was clearly identified. Error bars are s.d. (C) Histograms of the number of minutes spent in each of the cell cycle phases for the PC12 cells described in (A). (D) Representative traces of PC12 cells serum starved for 24 hours then treated with NGF and imaged after 24 hours for the next 37 hours. The number of frames spent in each phase divided by the total number of frames counted for all cells imaged for 0-37 hours (E) or 24-61 hours (F) post-NGF or serum addition.

Figure 4

Cell cycle effects and neurite extension caused by NGF addition. (A) PC12 cells transfected with both the G₁ phase biosensor (left) and a plasma membrane marker (Kras tail) (right) were imaged to identify cell cycle phases and neurite extensions, respectively. PC12 cells expressing the G₁ phase biosensor and Kras tail were cultured in media containing serum (B) or serum-free media (C) for 24 hours and then cultured in media containing serum + NGF (Control) or NGF. Schematics of the experimental designs are located above the graphs of the corresponding data. Panels on the left give the percent of PC12 cells in G₁ phase, i.e., those with a nuclear-localized G₁ phase biosensor. The middle panels indicate the percent of cells with plasma membrane extensions (neurites) at least 1.5 × their body diameter. Combining the information from the G₁ phase biosensor and the Kras tail allows for the assignment of either G₁ phase or S/G₂ phases to the cells that have formed neurites (panels on the right). Error bars are 95% confidence intervals.