Modelling the checkpoint response to telomere uncapping in budding yeast

Abstract

One of the DNA damage-response mechanisms in budding yeast is temporary cell-cycle arrest while DNA repair takes place. The DNA damage response requires the coordinated interaction between DNA repair and checkpoint pathways. Telomeres of budding yeast are capped by the Cdc13 complex. In the temperature-sensitive cdc13-1 strain, telomeres are unprotected over a specific temperature range leading to activation of the DNA damage response and subsequently cell-cycle arrest. Inactivation of cdc13-1 results in the generation of long regions of single-stranded DNA (ssDNA) and is affected by the activity of various checkpoint proteins and nucleases. This paper describes a mathematical model of how uncapped telomeres in budding yeast initiate the checkpoint pathway leading to cell-cycle arrest. The model was encoded in the Systems Biology Markup Language (SBML) and simulated using the stochastic simulation system Biology of Ageing e-Science Integration and Simulation (BASIS). Each simulation follows the time course of one mother cell keeping track of the number of cell divisions, the level of activity of each of the checkpoint proteins, the activity of nucleases and the amount of ssDNA generated. The model can be used to carry out a variety of in silico experiments in which different genes are knocked out and the results of simulation are compared to experimental data. Possible extensions to the model are also discussed.

Figure 1

A model for the interaction between checkpoint pathways and nucleases at cdc13-1-induced damage (Reproduced with permission from Jia et al. (2004), Copyright Genetics Society of America.). Rad9 inhibits ssDNA production at telomeres of cdc13-1 strains by two routes, a and b. One route, a, depends on Mec1, Rad9 and Rad53 and targets Exo1. The other route, b, is dependent on Rad9, independent of Mec1 and Rad53 and targets ExoX.

Figure 2

Network diagram showing the checkpoint response to uncapped telomeres. (a) Activation of ExoX and Exo1. ExoX requires Rad24 and Rad17 binding for its activation. Exo1 is activated independent of Rad24 and Rad17, although it may also act on telomeres bound by Rad17 and be activated in a Rad24-dependent manner. (b) Binding of single-stranded DNA (ssDNA) by RPA. Each molecule of RPA requires three units of ssDNA to bind. (c) Activation of checkpoint response via Rad53/Dun1 pathway. Activation of this pathway leads to inhibition of nuclease activity. Dashed line indicates an event, where a threshold level of Mec1RPAssDNA activates a kinase (Rad9Kin) which activates Rad9. (d) Activation of checkpoint response via Chk1 pathway does not affect Exo1. (e) Recovery can take place during S phase or G2/M arrest after single-stranded DNA has been removed. The dashed line indicates an event. When the level of ssDNA is equal to 0, the dummy species ‘recovery’ is set to 1 and the recapping reaction can then occur. See §2.1.6 for more details.

Figure 3

Model predictions of the number of capped telomeres in a wild-type cell when k₁=3.6×10⁻⁶ or k₁=0.0005. The output is for one simulation over a period of 12 h.

Figure 4

Model predictions for the kinetics of uncapping in the cdc13-1 strain at the restrictive temperature (with the default parameters). The output is for one simulation over a period of 12 h.

Figure 5

Model predictions for the number of divisions obtained in a wild-type cell. The output is for three simulations over a period of 12 h.

Figure 6

Growth of wild-type and cdc13-1 mutant strains (reproduced with permission from Zubko et al. (2004); Copyright Genetics Society of America). Yeast strains were released from G1 arrest and allowed to form microcolonies for 15 h at 36°C (restrictive temperature) before being photographed at 200× magnification. Cell numbers within microcolonies were estimated from the photographs shown and are given, along with their standard deviations, for each strain: (a) cdc13-1 RAD+(2); (b) cdc13-1 exo1 (8±2); (c) cdc13-1 rad9 (18±5); (d) cdc13-1 rad9 ex01 (54±17); (e) cdc13-1 rad24 (43±22); (f) cdc13-1 rad9 rad24 exo1 (113±34); (g) wild type.

Figure 7

Model predictions for the amount of ssDNA generated in wild-type and cdc13-1 mutant strains. The output is for one simulation for each strain over a period of 250 min.

Figure 8

Model predictions for the number of divisions obtained by wild-type and cdc13-1 mutant strains if a critical threshold of 20 kb ssDNA triggers cell death. The output is for one simulation for each strain over a period of 12 h.

Figure 9

Model predictions for (a) the amount of ssDNA per cell and (b) the number of cell divisions if a critical threshold of 120 kb ssDNA triggers cell death. The output is for one simulation for each strain over a period of 12 h.

Figure A1

Network diagram of the cell-cycle model. G1CdkA and G1CdkI represent the active and inactive G1 Cdk, respectively, and similarly for the other Cdks. The dashed lines connecting the cyclins to the Cdks indicate events. When a cyclin reaches a level of 100, the respective Cdk is activated.