Tumor cells stably transfected with fluorescent proteins enable scientists to visualize many important aspects of cancer in real time at the single cell level. For example, transfected tumor cells have been visualized either through surgically created chronic-transparent windows or directly through the opened skin of living animals . This intravital imaging provides a powerful tool for observing cancer initiation and progression and evaluating the efficacy of candidate cancer drugs in vivo. On the other hand, assays using tumor cells grown in culture provide reliable information about cancer mechanisms, and are amenable to automated high-throughput screening [16–20]. Using a modified fluorescent indicator of cell cycle progression (Fucci2) and cultured immortalized cells, we investigated the mechanism(s) by which anticancer drugs modulate the cell cycle. While population analysis provided statistical data, time-lapse high-resolution imaging analysis allowed us to explore the processes of cell cycle alteration in great detail. Drug-induced cell cycle modulation varied between different cell types or following treatment with different drugs, but, intriguingly, differences were observed between cells treated with different concentrations of the same drug or following drug addition during different phases of the cell cycle (Figure 1).
Three different fates await NMuMG/Fucci2 cells at the G2/M checkpoint after exposure to etoposide. At low concentrations of the drug, cells halt their cell cycle at the checkpoint. At intermediate concentrations, cells override the checkpoint and undergo nuclear mis-segregation. At high concentrations, cells avoid the checkpoint and enter the endoreplication cycle. Endoreplication may render cells resistant to conditions that induce DNA damage, possibly because they have several copies of each gene and do not need to segregate their chromosomes . From a clinical perspective, the tendency of high concentrations of etoposide to induce endoreplication in NMuMG cells may provide insight into the persistence of disease and development of chemotherapy resistance.
When cells are treated with drugs that affect the cell cycle, drug concentration is an important parameter affecting the cellular response. However, other factors including environmental factors may be critical. A previous study showed that hematopoietic cells treated with etoposide halt their cell cycle at G2/M or experience nuclear mis-segregation in the presence or absence of hematopoietic cytokines, respectively . Also, asymmetric cell division may generate different vulnerabilities to drugs. We observed that cell division produced two daughter cells with different fates in the presence of 1 μM etoposide; one underwent nuclear mis-segregation and the other entered the endoreplication cycle (Figure 6C). This observation excludes the possibility that the diversity of drug response results from inherent heterogeneity of the cell lines.
Mammalian megakaryocytes and trophoblast giant cells undergo endoreplication to reach DNA ploidies with a DNA content >1,000-fold higher than that of a normal diploid cell [4, 5, 7–9]. We isolated megakaryocytes and trophoblast giant cells from Fucci transgenic mouse lines, and successfully imaged the endoreplication cycles in these cell types cultured on a coverslip (A. S.-S, and A. M. unpublished results); the nuclei changed color alternately many times between green and red while enlarging. This observation verifies that oscillations in ubiquitination involving the SCFSkp2 and APCCdh1 complexes function during the endoreplication cycle, and Fucci technology allows for the visualization of endoreplication [31–34]. In contrast to the developmentally programmed endoreplication , most of the NMuMG/Fucci2 cells treated with etoposide experienced endoreplication only once, stably remaining tetraploid during our observation period of ≤4 days. Although it is known that anticancer drugs induce endoreplication to generate polyploidy in a variety of cell types, the predominant accumulation of tetraploid cells was also reported to result from inhibition of DNA topoisomerase II in other biological systems . Conventional cell cycle analysis that only quantifies dye-incorporation using FACS may misinterpret this long-lasting tetraploid state as G2 arrest. The information provided by the Fucci probes regarding the cell cycle phase allowed us to identify the single-round of endoreplication in these cells. Although a single round of endoreplication was the principal outcome in this study, a small fraction of the NMuMG cells was observed to go through multiple endoreplication cycles in the presence of 1 - 3 μM etoposide (Figure 1C, Additional files 8 - 11).
Fucci allows for visualizing cell-cycle progression in real time at the single cell level. Combined with information of DNA content obtained by cytometry analysis, Fucci technology can reveal complex aspects of drug-induced cell cycle alteration(s). Importantly, the technology becomes more powerful with long-term time-lapse imaging experiments. This dynamic data will advance our understanding of how individual cells respond to a drug. In our time-lapse imaging experiments where NMuMG/Fucci2 cells were continuously treated with 1 - 3 μM etoposide, for example, nuclear mis-segregation and endoreplication were mutually exclusive; we did not see any transition from one to the other. Although it was reported that chromosomes were condensed during formation of endopolyploid cells [4, 24], the endoreplication observed in the present study lacked all vestiges of mitosis, including chromosome condensation, nuclear envelope breakdown, and the reorganization of microtubules that builds the spindle. The other finding made by time-lapse imaging was that NMuMG/Fucci2 cells entered the endoreplication cycle if they had been in late G1 phase when the Cdk4 inhibitor was added to the medium; otherwise, the cells exposed to the drug showed G1 cell-cycle arrest. These features could not be elucidated by conventional cell cycle analysis using fixed cell samples.
Fluorescence imaging of stably transformed Fucci2-expressing cells in culture will provide reliable pharmaco-dynamic readouts for the proliferation and death of cancer cells. Through full integration of statistical and image data, our multifaceted assay system represents a validated strategy for characterization of drugs that modulate the cell cycle for anticancer drug screening and for cell toxicity studies.