The Homologous Recombination Mathematical Model and the Role

of miRNAs in ATM/ATR-dependent and BRCA1/BRCA2-dependent

DSBs Repair Pathway

Outline of the Project

Katarzyna Jonak, Monika Kurpas and Krzysztof Puszyński

Systems Engineering Group, Institute of Automatic Control, Silesian University of Technology,

ul. Akademicka 16A, 44-100 Gliwice, Poland

1 INTRODUCTION

Eukaryotic cells are exposed continuously to the

genotoxic stresses caused by various sources, what

may result in formation of DNA double strand

breaks (DSBs) or single strand breaks (SSBs). DSBs

are known to be one of the most cytotoxic lesions,

caused by exposure to ionizing radiation (IR),

clastogenic drugs (Lindahl and Barnes, 2012), but

also formed endogenously during DNA replication

or even as an effect of reactive oxygen species

(ROS) (Lopez-Contreras and Fernandez-Capetillo,

2012). In order to maintain genomic integrity, the

DNA damage response is activated. This biological

signaling pathway is a cascade of the signals from

different types of macromolecules: detectors that

recognize DSBs, proteins mediating signal

transduction, and effectors responsible for activation

of damage response.

DSBs are detected indirectly by ataxia

telangiectasia mutated (ATM) that stabilizes and

activates repair pathways, such as homologous

recombination (HR) or non-homologous end joining

(NHEJ). The proper functioning of repair pathways

is essential to enhance the cellular survival.

For better understanding of the molecular

mechanisms of DNA repair pathways, the useful

approach is presented by systems biology. It

describes complex systems of interactions between

macromolecules with loops of positive and negative

feedbacks in a form of mathematical models. Such

modeling allows to not only understand the complex

interactions between various components of the

regulatory pathways, but also allows to investigate

the impact of DNA-damaging agents on cells that

can lead to such diseases as neurological disorders

or cancerogenesis. Mathematical models can be used

for preliminary analysis of the experimental

hypotheses, as well as for putting the hypotheses on

possible treatment at the level of the cellular

signaling pathways.

2 STAGE OF THE RESEARCH

2.1 DSBs Detector Module

ATM functions as a DSBs detector that sends the

signal about the damage to different mediators and

effectors. We have developed a mathematical model,

where the consequences of ATM activation on two

transcription factors, tumor antigen p53 and NF-κB

(nuclear factor kappa-light-chain-enhancer of

activated B cells), are presented. Both nuclear

factors control several physiological processes from

cell cycle arrest through DNA repair and adaptive

immune response to apoptosis. The model is based

on our previous model of p53-NF-κB interaction

(Puszynski et al., 2009).

Major DNA damage response regulators play an

essential role in ATM-p53-NF-κB pathway. Mdm2

(E3 ubiquitin-protein ligase) facilitates p53

degradation, checkpoint kinase 2 (Chk2) inhibits p53

ubiquitination and degradation, and cellular

transcription factor CREB transcriptionally activates

ATM. Moreover, in this model we linked ATM-p53-

NF-κB pathway components with protein

phosphatase Wip1 that regulates dephosphorylation

events (inactivation of the most of the pathway

components).

The mathematical model is presented as a set of

stochastic and deterministic equations according to

the Haseltine-Rawlings postulate. Stochastic

description and following Gillespie direct method

based simulation were used for slow reactions, like

states of genes change, while deterministic

Jonak K., Kurpas M. and Puszy

nski K..

The Homologous Recombination Mathematical Model and the Role of miRNAs in ATM/ATR-dependent and BRCA1/BRCA2-dependent DSBs Repair

Pathway - Outline of the Project.

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

description based on ordinary differential equations

(ODE) and following Runge-Kutta 4

order

simulation method were used for description of

quick reactions, like activation or degradation of the

proteins involved in repair pathway.

The model is activated upon IR induction and

TNFα (tumor necrosis factor alpha). The signaling

pathway is also stimulated continuously by the small

number of damages that occur spontaneously. The

components of ATM-p53-NF-κB model are

presented mostly in two main states: active and

inactive. Moreover, most of the proteins considered

in the model contain their transcriptional forms

(mRNA). The model has an assumption that each

gene has two copies and among them one can be in

an active state, both, or none.

The output of the model is p53 level that

determinates cell fate. In this model cell death is

recognized as a permanently increase of p53 level

for more than 6 hours, then cell is considered as an

apoptotic and its elements are degraded.

We simulated the cellular response to the

damage combining all of the described elements.

The obtained results shown that ATM pathway is an

effective system for DSBs detection with strong

amplification signal and quick response.

Furthermore, we observed the strong dependence of

the cellular response to the DNA damage on Wip1,

what leads to the conclusion that it plays a role as

a gatekeeper in the ATM-Mdm2-p53 regulatory

loops, essential in the process of DNA damage

repair.

2.2 SSBs Detector Module

Another detector system which was important in our

work was ATR module (ataxia telangiectasia and

Rad3-related protein) responsible for detection of

DNA SSBs caused by, for example, ultraviolet

radiation (UV) or replication fork arrest (Lopez-

Contreras and Fernandez-Capetillo, 2012). The main

subject of the study was to develop a mathematical

model for p53 regulatory pathway with ATR as

a main detector system, perform simulations and

then linked to NF-κB regulatory module, ATM

detection module and deactivation agent Wip1.

Figure 1: Schematic model of HR repair pathway with ATM and ATR as major detector systems of DSBs and SSBs. The

letter “P” next to the protein name indicates phosphorylated form of this protein. Solid lines are transitions between states of

the HR components, dotted lines with arrow-heads are positive regulation (acceleration), and dotted lines with hammer-

heads are negative regulations (inhibition).

BIOSTEC2014-DoctoralConsortium

In ATR-p53 model the first stage of SSBs detection

is an activation of Rad9-Rad1-Hus1 protein complex

(called 9-1-1) and activation of the ATR-ATRIP

(ATR interacting protein) complex. Rad9 subunit

phosphorylated by protein ATR recruits TopBP1

(topoisomerase II binding protein 1), which presence

is necessary for full activation of ATR. Then, ATR

activates checkpoint kinase 1 (Chk1) that activates

p53 and increase degradation rate of Mdm2.

The model was built using the same assumptions

as ATM model, with the usage of stochastic and

deterministic approaches. The activation of the

ATR-p53 model was performed upon UV radiation

at 24 hours after start of the simulations. The effect

of different doses of UVC was studied and the

apoptotic death threshold was chosen. It was

observed in situ that with dose of 18 J/m

more than

half of the cells become apoptotic.

The simulations shown that ATR module acts as

an efficient system to detect even a single DNA

damage. The SSBs detection module is very fast -

the breakage is detected within few seconds after the

occurrence of DNA damage. Moreover, ATR and

ATM mathematical models explain that the base

production and activation level of the p53 protein

and its signaling pathway proteins may be caused by

persistent cellular stress.

2.3 HR as a Repair Pathway

The interesting subject in the area of modeling DSBs

and SSBs detection modules is the role of the

different proteins responsible for specific repair

pathways. Therefore, now we combine the two

detector modules, ATM and ATR, in order to

perform further extensions to homologous

recombination repair pathway.

We have already investigated the main modules

responsible for HR. The main interactions between

them are presented in Figure 1. The detector system

of DSBs is ATM, which together with MRN

complex activates BRCA1 (breast cancer type 1

susceptibility protein) responsible for DNA repair.

This activation leads to resection of DSBs and

activation of ATR module that detects SSBs. Then,

the process of repairing the single strand damages

starts and goes through BRCA2 (breast cancer type

2 susceptibility protein) and Rad51. The model

includes the effects of the damages on checkpoints

proteins, as well as the effect of known specific

microRNAs (miRNAs) responsible for regulation of

transcription of the proteins involved in HR

pathway.

3 OUTLINE OF OBJECTIVES

For a better understanding of the HR pathway and

identification of the abnormalities that may occur in

this process, it is useful to build a mathematical

model describing the dynamics of this module. The

model should be based on experimental data in order

to correspond to reality, hence the biological model

will be built that will allow experimental verification

of the HR mathematical model.

The main goal of the project is to examine the

impact of different agents that cause DSBs on HR in

a single cell as one of the most important repair

pathways in eukaryotic organisms. For this purpose

the models of ATM and ATR pathways will be

combined as detectors active in S/G2 phase of cell

cycle without taking into account interactions with

p53 and NF-κB. The whole HR pathway will be

described, from detector module to ligation of the

DNA strands. Another objective is to identify

substances which can cause DSBs without causing

SSBs. The impact of the DNA damages on miRNAs

will be also investigated with the negative

interaction of these macromolecules on other

components of HR pathway.

Within the project we anticipated three tasks:

 Construction of the deterministic and

stochastic mathematical model of HR repair

pathway based on the information from

existing models described in the literature, as

well as ATM and ATR pathways models

developed by the authors.

 Collection of the data from experiments and

literature on the parameters of the model, such

as activation and inactivation rates,

degradation rates, transcription rates, etc.

 Construction and experimental verification of

the model.

4 RESEARCH PROBLEM

The main research problem focuses on the

interactions between different components of HR

pathway during repair process of DNA double strand

breaks. Because of the fact that HR occurs only in

S/G2 phase, some of the proteins, such as p53, are

not active, what should be taken into account during

the model development. It is necessary to note that

different concentrations of specific detection

components, such as MRN, may lead to various

repair pathways, for example, NHEJ. This

TheHomologousRecombinationMathematicalModelandtheRoleofmiRNAsinATM/ATR-dependentand

BRCA1/BRCA2-dependentDSBsRepairPathway-OutlineoftheProject

phenomena should be also considered in the

theoretical model.

The impact of miRNAs on HR components is

essential during repair. Overexpression of some of

the miRNAs may cause cell cycle arrest and may

forward the cell into apoptosis pathway through p53

protein in G1 phase. The same result may be

obtained by downregulation of BRCA1 or BRCA2

genes and low activation level of these proteins.

Therefore, the exact impact of these components

should be examined, as well as the impact of Wip1 –

the major deactivation agent.

We found necessary to examine which DNA-

damaging factors may lead to DSBs and to

activation of HR pathway, as well as what dose of

these factors may lead to cell cycle arrest and

apoptosis.

The main purpose of the existence of the HR

model with the DNA damages detector modules is to

illustrate the processes occurring during DNA repair.

The project will allow to understand the interactions

between molecules, probably investigate new

components of the HR network, and contribute to

putting the hypotheses on possible treatment of

different diseases at the level of the cellular

pathways.

5 STATE OF THE ART

5.1 DSBs Detection

Activation of a specific mechanism of DNA repair

depends mainly on cell cycle. Thus, initiation of

NHEJ is possible during the whole cycle, however

usually appears in G1/G0 phases, where HR is

limited to late S phase and G2 phase (Langerak and

Russell, 2011). DSBs are detected by ATM and

multiprotein complex MRN. In G1 phase, the

activation of these proteins results in Chk2-mediated

p53-dependent cell cycle arrest. DNA damages

undergo only minor nucleolytic processing being

repaired very fast by NHEJ (Jazayeri et al., 2006). In

case of S/G2 phases and repair by HR, the cell cycle

arrest is p53-independent and the process of repair is

slower than NHEJ (Jazayeri et al., 2006). The choice

between HR and NHEJ repair pathways is partially

determinate by MRN complex activity: in HR when

DSBs require resection the activity of the complex is

much higher (Chowdhury et al., 2013). BRCA1 also

promotes resection and excludes 53BP1 protein,

which is involved in NHEJ process (Chowdhury et

al., 2013).

ATM is activated by DSBs directly and

indirectly by MRN complex. ATM is also

autophosphorylated by the formation of defects in

the chromatin structure (Bakkenist and Kastan,

2003). At the same time exonulease MRN is

activated directly by DSBs and by phosphorylation

of one of its component, Nbs1, by ATM (Bakkenist

and Kastan, 2003). ATM is involved in cell cycle

arrest by phosphorylation of Chk1 and mostly by

phosphorylation of Chk2. The signal of DSBs is

amplified by autophosphorylation of Chk2 induced

by ATM (Ahn et al., 2004).

Figure 2: Scheme of HR repair pathway. DNA helix after

DSBs induction is presented as a light blue ladder, DNA

from sister chromatid is a purple ladder.

5.2 DNA Resection and Repair

MRN together with CtIP protein binds to free ends

of DNA molecule and in presence of BRCA1

degrades one of the two helices from 5' end of DNA

strand. This process results is resection of DSBs.

The resulting free 3' ends are detected and protected

by replication protein A complex (RPA) (Filippo et

al., 2008). These damages of single stranded DNA

(ssDNA) recruits proteins involved in detection of

BIOSTEC2014-DoctoralConsortium

SSBs, such as ATR. ATRIP is capable of attaching

themselves to the RPA-ssDNA, which induces

autophosphorylation of ATR (Nam et al., 2011). In

addition, ssDNA fragment binds a complex of

Rad17-RFC2-5, which allows the attachment of the

9-1-1 complex to the damaged site (Filippo et al.,

2008). ATR recruits TopBP1 and then activates

several proteins involved in the repair pathway, such

as Chk1 and Chk2. Signal strength of the checkpoint

cascade is dependent on the length of RPA-ssDNA

and the possibility of autophosphorylation of ATR

molecules (Filippo et al., 2008).

Activated Chk1 and Chk2 results in cell cycle

arrest by inactivation of cyclin-dependent kinases.

The reduction of activity of these kinases leads to

activation of BRCA2, which recruits Rad51 and

interact with RPA-ssDNA to initiate HR process

(Filippo et al., 2008). Moreover, activated Chk1

phosphorylates Rad51, what leads the cell to the HR.

BRCA2 and RAD51 together with the

accompanying proteins catalyze the invasion of the

free 3' end to the homologous sequence of sister

chromatid. As an effect the D-loop structure is

formed, what allows for the rebuilding of a missing

DNA fragment by polymerase η (Filippo et al.,

2008). The DNA ligase I join the new fragments.

Synthesis is finished when two Holliday’s structures

are formed (Filippo et al., 2008). These structures

are then removed by helicases, like BLM,

topoisomerases IIIα or endonucleases, such as

Mus81/Eme1 (Filippo et al., 2008). The process of

DNA repair by HR is presented in simplified way in

Figure 2.

5.3 MiRNAs and Wip1 in HR Pathway

An essential component in regulation of DSBs

detection and the process of DNA damages repair is

Wip1, which is a deactivation agent for key proteins

involved in HR. The main role of this protein is to

regulate the level of activated proteins mostly after

succeed process of repair, and to unlock the cell

cycle. Wip1 transcription is p53-dependend, what

makes it connected to DNA damages (Lowe et al.,

2012). The most important parts of regulation of the

repair pathways by Wip1 is associated with

inactivation of DNA damages detector modules:

ATM and ATR. Moreover, the cell cycle

checkpoints, Chk1 and Chk2, are dephosphorylated

by Wip1, as well as p53 and Mdm2 (Lowe et al.,

2012).

Small RNAs, called miRNAs, also play

important roles in DSBs repair pathway. Both

miRNA transcription and maturation processes are

altered in response to damages of DNA strands and

repair processes. Biogenesis of these

micromolecules is induced in an ATM-dependent

manner (Chen X. and Chen T., 2011). Activated

ATM phosphorylates KSRP protein (KH-type

splicing regulatory protein) what leads to activation

of some of the miRNAs (Chen X. and Chen T.,

2011). Not all of the miRNAs involved in HR

pathway have been discovered yet, however there

are some which are known, such as miR-100, miR-

101 and miR-421 which suppress ATM, miR-182

which suppress BRCA1 or miR-16 that suppress

Wip1 (Chen X. and Chen T., 2011).

5.4 Existing HR Models

The existing models of HR pathway are based

mainly on the late phase of the repair process: from

RPA coating to the action of DNA ligases.

In the model described in (Taleei et al., 2011)

DSBs detection module is treated as a one

component: MRN. The authors focus on the effects

of resection, without describing the whole DSBs

detection with ATM, and without including

checkpoint proteins. The most important parts there

are MRN, RPA, Rad52 and ligase. Another

theoretical model (Rodríguez et al., 2012) is based

on Boolean network system for the FA/BRCA

pathway involved in HR. The whole model is very

expanded and it takes into account also NHEJ repair

pathway. However, the model do not contain the

effects of miRNAs and is rather focus only on the

whole BRCA pathway.

6 METHODOLOGY

The mathematical model of HR repair pathway will

be built using set of equations that allow simulating

the behavior of one cell treated with different doses

of cytotoxins that cause DSBs. For the mathematical

model stochastic and deterministic approach will be

used as it was performed for the models of ATM and

ATR signaling pathways. Moreover, the Michaelis-

Menten kinetics and the law of mass action will be

used in order to bring the model and its reaction

speed to reality.

In order to investigate the effect of other not-yet-

known modules on described components of the HR

pathway, several biological experiments will be

performed. Northern blotting will be used in order to

investigate miRNAs that regulate different

components of the HR. Western blotting will be

used to investigate level of key proteins (total and

TheHomologousRecombinationMathematicalModelandtheRoleofmiRNAsinATM/ATR-dependentand

BRCA1/BRCA2-dependentDSBsRepairPathway-OutlineoftheProject

phosphorylated forms) involved in the process of

DSBs detection and repair, as well as to collect the

information concerning the kinetics parameters of

the model, such as time of deactivation of the

specific protein. Moreover, the analysis of the

expression profile of genes involved in HR pathway

will be performed with a method of quantitative

real-time PCR in stress conditions.

The analysis of the level of double strand breaks

of DNA after treatment with specific cytotoxins will

be performed with microscope analysis of H2AX

foci assay. The amount of foci reflects the amount of

DNA breaks. The analysis of the total number of

DSBs and SSBs will be performed with comet assay.

The experiments will be performed on

mammalian cancer and normal cell lines with active

or inactive forms of the proteins involved in HR.

The whole methodology is still under development.

7 EXPECTED OUTCOME

The experimental-based mathematical model

(stochastic and deterministic) of HR repair pathway

will be presented. The HR model, together with

ATM and ATR models developed by our group, will

create a comprehensive mathematical model

describing the dynamics of the interactions

occurring in the cell from the inception of DNA

damages to making the decisions about cell fate: to

direct cell to repair by HR, to arrest cell cycle or

direct cell to apoptosis pathway during HR or before

the process of repair.

The expected outcome is an identification of at

least one new component of the HR pathway which

can be even protein or miRNA, and the parameters

of the model for most of the pathway components.

Moreover, it is expected that the developed model

will reflect the experimental data and will be a good

tool for simulation of the cell behavior during HR.

REFERENCES

Ahn, J., Urist, M., Prives, C., 2004. The Chk2 protein

kinase. DNA Repair, 3:1039-1047.

Bakkenist, C. H., Kastan, M. B., 2003. DNA damage

activates ATM through intermolecular

autophosphorylation and dimer dissociation. Nature,

421:499-506.

Chen, X., Chen, T., 2011. DNA repair, InTech. 1

edition.

Chowdhury, D., Choi, Y. E., Brault, M. E., 2013. Charity

begins at home: non-coding RNA functions in DNA

repair. Nature Reviews Molecular Cell Biology,

14:181-189.

Jazayeri, A., Falck, J., Lukas, C., Bartek, J., Smith,

G.C.M., Lukas, J., Jackson, S.P., 2006. ATM- and cell

cycle-dependent regulation of ATR in response to

DNA double-strand breaks. Nature Cell Biology, 8:37-

45.

Langerak, P., Russell, P., 2011. Regulatory networks

integrating cell cycle control with DNA damage

checkpoints and double-strand break repair.

Philosophical Transactions of the Royal Society,

366:3562-3571.

Lindahl, T., Barnes, D. E., 2012. Repair of endogenous

DNA damage. Cold Spring Harb Symp Quant Biol,

65:127-3.

Lopez-Contreras, A. J., Fernandez-Capetillo, O., 2012.

Protein Phosphorylation in Human Health, InTech. 1

edition.

Lowe, J., Cha, H., Lee, M. O., Mazur, S. J., Appella, E.,

Fornace, A. J., (2012). Regulation of the Wip1

phosphatase and its effects on the stress response.

Frontiers in Bioscience, 17: 1480-1498.

Nam, E. A., Zhao, R., Glick, G. G., Bansbach, C. E.,

Friedman, D. B., Cortez, D., 2011. Thr-1989

Phosphorylation Is a Marker of Active Ataxia

Telangiectasia-mutated and Rad3-related (ATR)

Kinase. J Biol Chem, 286(33): 28707–28714.

Puszynski, K., Bertolusso, R., Lipniacki, T., 2009.

Crosstalk between p53 and nuclear factor-kB systems:

pro- and anti-apoptotic functions of NF-kB. IET Syst

Biol., 3(5):356-367.

Rodríguez, A., Sosa, D., Torres, L., Molina, B., Frías, S.,

Mendoza, L., 2012. A Boolean network model of the

FA/BRCA pathway. Bioinformatics, 28(6):858-66.

San Filippo, J., Sung, P., Klein, H., 2008. Mechanism of

Eukaryotic Homologous Recombination. Annu. Rev.

Biochem,. 77:229–57.

Taleei, R., Weinfeld, M., Nikjoo, H., 2011. A kinetic

model of single-strand annealing for the repair of

DNA double-strand breaks. Radiat Prot Dosimetry.,

143(2-4):191-5.

BIOSTEC2014-DoctoralConsortium