Jaiswal et al (2017). ATM/Wip1 activities at chromatin control Plk1 re-activation to determine G2 checkpoint duration
April 2018, model of the month by Rahuman Sheriff
Original model: BIOMD0000000641
Cell cycle is a complex process that is controlled by many regulators to warrant accurate cell division. Cells are directed through a specific sequence of cell cycle phases guarded by dedicated check points to produce two daughter cells [1, 2]. If a DNA damage such as a double-strand break (DSB) occurs, the cell cycle is halted and DNA damage response (DDR) is launched within the cell. DSBs trigger ATM-/ATR-dependent signalling that arrests cells at G2 phase by inhibiting the activities of mitosis-promoting kinases such as cyclin-dependent kinase 1 (Cdk1), Polo-like Kinase 1 (Plk1), and Aurora A. ATM kinase phosphorylates over 700 different proteins to initiate DNA damage response. Surprisingly, the cell cycle inhibition is released before all DNA lesions are repaired. Plk1 plays a vital role in reversing cell cycle inhibition after damage and promotion to mitosis .
Using a FRET activity biosensor for ATM/ATR Kinase and Plk1, Jaiswal et al  Jaiswal et al showed that the DNA damage induced ATM activity spreads across the chromatin, inhibiting the activation of Plk1. During recovery, while ATR is still active, chromatin bound phosphatase Wip1 counters ATM activity. This leads to the reactivation of Plk1, which results in cell cycle progression. To further understand the checkpoint regulation and its duration after DNA damage, Jaiswal et al mathematically modelled the interplay between ATM, ATR, Plk1 and Wip1 dependent pathways.
A simple mathematical ODE model, assuming Michaelis-Menten kinetics, was assembled. It contains four main functional entities corresponding to ATM-, Wip1-, ATR- and Plk1-dependent pathways (figure 1). Other entities included DNA damage as well as DNA repair by homologous recombination (H.R) and non-homologous end joining. The effect of Wip1 countering ATR-activation was considered as a constant factor in the model that corresponds to the ratio between Wip1 and ATR in chromatin. In the model, an increase in Plk1-dependent pathways' activity represents progression of cell cycle and high level of this entity is considered as a readout for mitosis. ATR-dependent pathways suppress Plk1 through the phosphorylation of Bora and Wee1. Also, Plk1 pathways suppress ATR via phosphorylation of Claspin and others. Due to lack of evidence supporting a major effect of Plk1 on ATM activity, this interaction was ignored in the model.
Figure 1. Schematic representation of a simplified version of the mathematical model with arrows representing ODEs. Figures adapted from .
Model simulation with the addition of DNA damage in the beginning revealed that ATR prolonged the time taken to reach high Plk1 pathways activity in the absence of ATM (figure 2 a,b). During DNA damage, the balance between ATR and the Plk1 activities potentially determined the duration of the cell cycle arrest. DNA damage activates ATM resulting in suppression of Plk1 activity. Following this, with the help of the positive feedback loop in the Plk1 pathways that essentially promotes self-activation, over the course of time Plk1 activity surges high enough to inhibit ATR pathways and promote entry of cells into mitosis (figure 2b).
ATR pathway, that gets activated to its peak by high DNA damage, was rapidly suppressed by Wip1 when the damage declined. This pulse in the ATR activity increased ATM activation which, in turn, inhibited Plk1 pathways, slowing down cell cycle progression (figure 2c ). Low levels of Plk1 or cell cycle activities inefficiently suppressed ATR and this introduced a delay until Plk1 self-activation kicked in gradually (figure 2 b,c). Thus, a pulse of ATM activation upon DNA breaks, re-tunes cell cycle and introducing a delay before mitotic entry, thus allowing DNA damage repair. This model provides a simplistic system that explains how ATM- and ATR- signalling establishes minimal duration of checkpoint.
Figure 2. Model simulation. Simulation of the model in the absence of DNA damage (A) and ATM (B) and presence of all components (C). Steady state simulation of the optimised model indicating the Plk1 activity for a range of sustained DNA damage levels (D). Simulation performed in Copasi, figures generated in Matlab.
Jaiswal et al , also fitted the time-course data for ATM/ATR and Plk1 activity reporter to the model to estimate model parameters in U2OS cells, a human bone osteosarcoma epithelial cell line used to study DDR. Comparison of model simulation with experimental data showed that ATM did not fully supress Plk1 activity but rather delayed the time take to reach full activity at low levels of DNA damage. In such conditions, the initial levels of Plk1 activity influenced the duration of arrest. In contrast, at high levels of DNA damage, initial levels of Plk1 activity will have minimal effect on the duration of the delay as the cell cycle activities are reset.
Jaiswal et al  then simulated sustained DNA damage to study its impact on the system. This revealed that sustained DNA damage can lead to a steady state with sustained ATM and ATR activity and low Plk1 activity (figure 2d). This simulation also divulged the presence of a threshold for DNA damage above which the cells will not enter mitosis, as the Plk1 activity will be maintained at low levels. Contrarily below this threshold, Plk1 activity can increase through self-activation and eventually suppress ATR activity to progress to mitosis. This prediction was validated experimentally by treating cells with radiomimetic neocarzinostatin, a DNA cleaving agent.
Jaiswal et al  used mathematical models in conjunction with experimental studies to investigate the regulation of the duration of cell cycle arrest caused by double stranded DNA damage. The mathematical model implied that the G2 check point is released before the DNA damage is completely repaired because the check point is regulated by the balance between ATM and Wip1 levels on the chromatin that controls its duration. The model revealed that the G2 check point guarded by ATM- and ATR- signalling imposes a delay in Plk1 activity or cell cycle progression to allow DDR. Simulation of sustained DNA damage suggested a thresholding mechanism for ATM activity that determines whether a cell will recover from the check point. Although this simplistic model missed several key players in the DDR and G2 check points, it still provided valuable insight as it was well coupled with experimental studies and the predictions are validated.
- Vermeulen K, Van Bockstaele DR, Berneman ZN (2003). The cell cycle: a review of regulation, deregulation and therapeutic targets in cancer.Cell Prolif., 36(3):131-49.
- Schafer KA (1998). The cell cycle: a review. Vet Pathol., 35(6):461-78.
- Jaiswal H, Benada J, Mullers E, Akopyan K, Burdova K, Koolmeister T, Helleday T, Medema RH, Macurek L, Lindqvist A. (2017). ATM/Wip1 activities at chromatin control Plk1 re-activation to determine G2 checkpoint duration..EMBO J. 14;36(14):2161-2176. doi: 10.15252/embj.201696082.