Inference of cell cycle-dependent proteolysis by laser scanning cytometry

Abstract

Mechanisms that couple protein turnover to cell cycle progression are critical for coordinating the events of cell duplication and division. Despite the importance of cell cycle-regulated proteolysis, however, technologies to measure this phenomenon are limited, and typically involve monitoring cells that are released back into the cell cycle after synchronization. We describe here the use of laser scanning cytometry (LSC), a technical merger between fluorescence microscopy and flow cytometry, to determine cell cycle-dependent changes in protein stability in unperturbed, asynchronous, cultures of mammalian cells. In this method, the ability of the LSC to accurately measure whole cell fluorescence is employed, together with RNA fluorescence in situ hybridization and immunofluorescence, to relate abundance of a particular RNA and protein in a cell to its point at the cell cycle. Parallel monitoring of RNA and protein levels is used, together with protein synthesis inhibitors, to reveal cell cycle-specific changes in protein turnover. We demonstrate the viability of this method by analyzing the proteolysis of two prominent human oncoproteins, Myc and Cyclin E, and argue that this LSC-based approach offers several practical advantages over traditional cell synchronization methods.

Figure 1. Myc levels and phosphorylation are influenced by cell-synchronization

(A) U20S cells were synchronized by treatment with nocodazole to arrest at the G2/M transition. Cells were released from arrest, and samples taken at the indicated time-points for analysis of cell cycle by FC, and protein levels by western blot. The numbers for G0/G1, S, and G2/M percentages were derived from FC. ‘pT58’ refers to an antibody that recognizes the phosphorylated form of residue threonine 58 within Myc. (B) As in (A), except that cells were subjected to double-thymidine (G1/S) block and release. FC profiles are presented in Supplemental Figure 1. Detailed methods are presented in Supplemental Information.

Figure 2. Parallel analysis of Myc RNA and protein levels throughout the cell cycle

(A) Visualization of nuclear Myc and actin transcripts in U2OS cells by RNA FISH. Anti-sense probe cocktails detect bona-fide transcripts; sense probe cocktails reveal background. (B) Quantification of absolute FISH signals. Fluorescence intensities were quantified by LSC in cells binned into G1, S, or G2/M populations, and expressed relative to the signal from G1 phase cells. (C) Quantification of relative FISH signals. Fluorescence intensities were quantified by LSC in cells binned into G1, S, or G2/M populations, normalized to the signal for DNA content in those cells, and expressed relative to the signal from G1 phase cells (n=2, mean +/− S.D.). Arbitrary units of fluorescence are used. (D) Quantification of absolute levels of Myc and cyclin E protein. Fluorescence intensities were quantified by LSC in cells binned into G1, S, or G2/M populations, and expressed relative to the signal from G1 phase cells. (E) Quantification of relative protein levels. Fluorescence intensities were quantified by LSC in cells binned into G1, S, or G2/M populations, normalized to the signal for DNA content in those cells, and expressed relative to the signal from G1 phase cells (n=4, mean +/− S.D.). Detailed methods are presented in Supplemental Information.

Figure 3. The metabolic stability of Myc is constant throughout the cell cycle

HeLa cells were treated with cyclohexamide for the indicated time points, fixed, and Myc (A) and actin (B) levels for each cell cycle subpopulation measured by immunofluorescence and LSC. In each case, the relative signals are normalized to DNA content and presented as a percentage of the zero-time point samples. Detailed methods are presented in Supplemental Information.

Figure 4. Differential regulation of Myc and Cyclin E by Fbw7 during the cell cycle

(A) Effects of siRNA-mediated knockdown of Fbw7 on Myc and Cyclin E. U2OS cells were transfected with the indicated siRNAs [control (=luciferase); Fbw7, and Myc], and Myc and Cyclin E levels determined in each cell cycle subpopulation of cells. Protein expression histograms show that Fbw7 knockdown substantially increases Cyclin E levels in S and G2 phase (rightward shift), whereas it has no significant effect on Myc profiles. (B) Analysis of total steady-state levels of Myc and Cyclin E following Fbw7 knockdown. Western blot analysis of cells analyzed in (A). (C) Quantification of relative Myc protein levels. Fluorescence intensities for Myc were quantified by LSC in cells binned into G1, S, or G2/M populations, normalized to the signal for DNA content in those cells, and expressed relative to the signal from G1 phase. (D) Combined effect of USP28 and Fbw7 knockdown. U2OS cells stably expressing an shRNA against USP28 (or control luciferase shRNA) were transfected with siRNA against Fbw7 (or control luciferase siRNA). Cells were harvested and Myc, cyclin E, and actin levels detected by WB. (E) Effect of USP28 on Myc levels during the cell cycle. Immunofluorescence of endogenous Myc was performed in USOS cells expressing either cells stably expressing an shRNA against USP28, or control luciferase shRNA. Cells were counterstained with Hoechst 33342 and analyzed by LSC. (F) Cell cycle-dependent effects of Fbw7 on Cyclin E. Experiment was performed as in (C), except that Cyclin E was detected by IF. Detailed methods are presented in Supplemental Information.