Myc and Ras oncogenes engage different energy metabolism programs and evoke distinct patterns of oxidative and DNA replication stress

Abstract

Both Myc and Ras oncogenes impact cellular metabolism, deregulate redox homeostasis and trigger DNA replication stress (RS) that compromises genomic integrity. However, how are such oncogene-induced effects evoked and temporally related, to what extent are these kinetic parameters shared by Myc and Ras, and how are these cellular changes linked with oncogene-induced cellular senescence in different cell context(s) remain poorly understood. Here, we addressed the above-mentioned open questions by multifaceted comparative analyses of human cellular models with inducible expression of c-Myc and H-RasV12 (Ras), two commonly deregulated oncoproteins operating in a functionally connected signaling network. Our study of DNA replication parameters using the DNA fiber approach and time-course assessment of perturbations in glycolytic flux, oxygen consumption and production of reactive oxygen species (ROS) revealed the following results. First, overabundance of nuclear Myc triggered RS promptly, already after one day of Myc induction, causing slow replication fork progression and fork asymmetry, even before any metabolic changes occurred. In contrast, Ras overexpression initially induced a burst of cell proliferation and increased the speed of replication fork progression. However, after several days of induction Ras caused bioenergetic metabolic changes that correlated with slower DNA replication fork progression and the ensuing cell cycle arrest, gradually leading to senescence. Second, the observed oncogene-induced RS and metabolic alterations were cell-type/context dependent, as shown by comparative analyses of normal human BJ fibroblasts versus U2-OS sarcoma cells. Third, the energy metabolic reprogramming triggered by Ras was more robust compared to impact of Myc. Fourth, the detected oncogene-induced oxidative stress was due to ROS (superoxide) of non-mitochondrial origin and mitochondrial OXPHOS was reduced (Crabtree effect). Overall, our study provides novel insights into oncogene-evoked metabolic reprogramming, replication and oxidative stress, with implications for mechanisms of tumorigenesis and potential targeting of oncogene addiction.

Keywords: DNA damage response; DNA fork progression; Energy metabolism; Myc; Ras; Replication stress.

Figure 1

Ras overexpression induces cell hyperproliferation and DNA damage response in BJ cells (BJ Ras). (A) BJ cells expressing the active version of the Ras protein (H‐RasV12) under the control of the Tet‐on system were incubated with doxycycline (Dox) for different time intervals and cell viability was estimated by trypan blue exclusion. Ras overexpression stimulated cell proliferation from day 1 to day 10 in comparison with non‐treated cells. At later time points oncogene‐induced senescence was gradually established (data not shown and (Kosar et al., 2011)). (B) The fraction of cells in the S phase was augmented in Ras overexpressing cells. A 30 min pulse of BrdU was added at specific time points in BJ Ras cells. The highest number of cells in the S phase was observed at day 4 post‐induction and gradually decreased afterwards. (C) Quantification of total γH2AX after Ras overexpression. BJ Ras cells were treated with Dox at different time points and γH2AX was detected by immunofluorescence. The total intensity of γH2AX was measured in individual nuclei using high‐throughput microscopy and the average intensity in each cell was analyzed by ScanR software. The plot shows the average of >4000 cells for each time point. (D) Quantification of the average number of 53BP1 foci after Ras overexpression. BJ Ras cells were treated with Dox at different time points and 53BP1 was detected by immunofluorescence. The total number of 53BP1 foci in G1 cells was quantified in individual nuclei. The average number of foci was analyzed by ScanR software. (E–I) Ras overexpression modifies the speed of replication fork progression. Cells were pulse‐labeled for 20 min with CldU, washed and labeled with IdU for another 20 min. The length of each pulse in individual well spread forks was measured and converted into kb/min. (E–H) Distribution of the fork extension rates (kb/min) in non‐treated cells and cells treated with Dox for 3 (E), 4 (F), 8 (G) and 14 days (H), respectively. Empty bars represent data from control non‐treated cells and black bars from Dox‐treated cells. (I) Quantification of the mean extension rates (kb/min, 1 μm = 2.59 kb) during the first (CldU, 20 min) and the second (IdU, 20 min) pulses. The number of analyzed forks (n) is shown and the probabilities associated to the t‐test (p value) are presented in the last column of the table (t‐test: ‐Dox vs + Dox).

Figure 2

Myc overexpression slows down DNA replication fork progression in BJ cells. (A) BJ cells were infected to express the fusion protein MycER under the regulation of 4‐OHT. After 2 weeks of selection cells were incubated in the presence of 4‐OHT for different time points and cell viability was estimated by trypan blue exclusion. (B) BJ MycER cells were pulse‐labeled for 20 min with CldU, washed and pulse‐labeled with IdU for subsequent 20 min. The red color in DNA fibers is the signal from the first pulse and the green color from the second pulse. The length of each pulse in individual, well spread fibers was measured and converted into kb/min. Examples of DNA fibers from BJ MycER cells non‐treated (6d control) and treated with 4‐OHT (6d 4‐OHT) for 6 days are shown. (C) BJ cells infected with retrovirus containing an empty pBabe vector were treated or not (control) with 4‐OHT for 3 days and analyzed for the fork speed. (D–F) BJ MycER cells were induced for 1 (D), 3 (E) and 6 (F) days, respectively and the total fork speed was analyzed. Plots show the distribution of fork extension rates (kb/min) of the first and the second pulse in non‐treated (empty bars) and 4‐OHT‐treated (black bars) cells. (G) Quantification of the mean extension rates (kb/min) during the first (CldU, 20 min) and the second (IdU, 20 min) pulses. The number of analyzed forks (n) is shown and the probabilities estimated by the t‐test are presented in the last column of the table (t‐test: control vs 4‐OHT). Scale bars are 10 μm (25.9 kb).

Figure 3

Myc overexpression induces DNA damage response and cell death in U2‐OS cells. (A) U2‐OS cells that express MycER were induced with 4‐OHT at different time points and cell viability was assessed by trypan blue exclusion. (B) Live cells non‐treated and 4‐OHT‐treated were stained with Hoechst 33342 and propidium iodide (PI). After 1, 2 and 3 days of 4‐OHT induction the percentage of apoptotic cells was quantified by PI exclusion and nuclear fragmentation. 1d: control n = 2080, 4‐OHT‐induced n = 1655, p associated to the t‐test analysis <0.0098; 2d: control n = 2374, 4‐OHT‐induced n = 2142, p < 0.001; 3d: control n = 1864, 4‐OHT‐induced n = 2080, p < 0.006. (C) U2‐OS MycER cells were treated with 4‐OHT at different time points and γH2AX was detected by immunofluorescence. The total amount of γH2AX was measured in individual cells using high‐throughput microscopy and the average intensity in each cell was analyzed using ScanR software. The plot shows an average of >4000 cells for each time point. (D) In a time course experiment, total proteins from non‐induced and induced U2‐OS MycER cells were extracted, resolved and γH2AX was immunoblotted. Loading control: histone H3.

Figure 4

Myc overexpression slows down DNA replication fork progression and induces fork asymmetry in U2‐OS cells. (A) U2‐OS MycER cells were pulse‐labeled for 20 min with CldU, washed and pulse‐labeled with IdU for another 20 min. The red tracks in DNA fibers are signals from the first pulse and the green tracks from the second pulse. The length of each pulse in individual, well spread fibers was measured and converted into kb/min. Examples of DNA fibers from U2‐OS MycER cells non‐treated (3d control) and treated with 4‐OHT (3d 4‐OHT) for 3 days shown. (B–D) U2‐OS Myc cells were induced for 1 (B), 3 (C) and 6 (D) days, respectively and the total fork speed was analyzed. Plots show the distribution of fork extension rates (kb/min) of the first and the second pulse in non‐treated (empty bars) and 4‐OHT‐treated (black bars) cells. CldU/IdU values are shown in (E–G). When extension rates are similar during both pulses, perfect symmetry is equal 1. From day 1 (E) of Myc overexpression there was a significant increase of highly asymmetric replication forks (ratios above 1.6; control n = 56, 4‐OHT‐induced n = 113; p associated to the t‐test <0.0001). Distribution of the ratio of fork rates during the first and the second pulse after 3 days (F) of Myc activation (ratios above 1.6; control n = 45, 4‐OHT‐induced n = 128; p < 6E‐5) and after 6 days (G) of Myc activation (ratios above 1.6; control n = 115, 4‐OHT‐induced n = 128; p < 2E‐7). (H) Quantification of the mean extension rates (kb/min) during the first (CldU, 20 min) and the second (IdU, 20 min) pulses. The number of analyzed forks (n) is shown and the probabilities assessed by the t‐test are presented in the last column of the table (t‐test: control vs 4‐OHT). Scale bars are 10 μm (25.9 kb).

Figure 5

Myc and Ras overexpression induces changes in cell metabolism. The extracellular acidification rate (ECAR) and the oxygen consumption rate (OCR) were measured simultaneously in the Seahorse XF96e bioanalyzer. (A–D) U2‐OS MycER and BJ MycER cells were left untreated (−) or treated with 4‐OHT (+) and were investigated after 3 days or after 3 and 6 days, respectively. After 3 days of Myc activation in U2‐OS cells there was a significant increase of ECAR (A) and a decrease of OCR (B) levels, with no apparent changes in BJ MycER cells (C, D). (E) Ras overexpression by Dox incubation induced an increase of ECAR after 4 and 14 days in BJ cells. (F) OCR was initially (4d) decreased after Dox induction but was significantly elevated at later time points (14d).

Figure 6

Oxidative RNA/DNA damage following oncogene activation. The 8‐oxoguanine (8‐oxoG/8‐oxo(d)G) levels were analyzed by immunocytochemistry from day 1–6 after Myc activation and from day 1–18 after Ras overexpression. (A) Representative images of U2‐OS MycER and BJ MycER cells after 3 days of 4‐OHT incubation (lower row) are shown. BJ wild‐type cells and BJ Ras cells Dox‐induced for different time points are shown in upper row. (B, C) Levels of total superoxide measured by FACS analysis of dihydroethidium staining of (B) non‐induced BJ Ras or BJ Ras cells after 4 and 6 days of Dox treatment, respectively and (C) BJ pBabe and BJ MycER cells after 3 days of 4‐OHT induction. Measurements of mitochondrial superoxide using MitoSOX Red™ are shown in (D–G). (D) Representative histograms showing the influence of doxycycline‐induced expression of Ras on production of mitochondrial superoxide after 4 and 14 days. (E) Fold increase in superoxide levels in induced compared with non‐induced BJ Ras cells. (F) Representative histograms showing the level of superoxide in U2‐OS MycER and BJ MycER cells after treatment with 4‐OHT for 3 days. (G) Fold increase in superoxide levels in activated MycER cells compared with non‐treated cells. In D‐G, treatment with the mitochondrial complex III inhibitor Antimycin A (5 μM) served as a control for mitochondrial superoxide production.