Limits...
Dynamics of p53 and NF-κB regulation in response to DNA damage and identification of target proteins suitable for therapeutic intervention.

Poltz R, Naumann M - BMC Syst Biol (2012)

Bottom Line: Simulating therapeutic intervention by agents causing DNA single-strand breaks (SSBs) or DNA double-strand breaks (DSBs) we identified candidate target proteins for sensitization of carcinomas to therapeutic intervention.Further, we enlightened the DDR in different genetic diseases, and by failure mode analysis we defined molecular defects putatively contributing to carcinogenesis.By logic modelling we identified candidate target proteins that could be suitable for radio- and chemotherapy, and contributes to the design of more effective therapies.

View Article: PubMed Central - HTML - PubMed

Affiliation: Institute of Experimental Internal Medicine, Otto von Guericke University, Leipziger Str, 44, Magdeburg, 39120, Germany.

ABSTRACT

Background: The genome is continuously attacked by a variety of agents that cause DNA damage. Recognition of DNA lesions activates the cellular DNA damage response (DDR), which comprises a network of signal transduction pathways to maintain genome integrity. In response to severe DNA damage, cells undergo apoptosis to avoid transformation into tumour cells, or alternatively, the cells enter permanent cell cycle arrest, called senescence. Most tumour cells have defects in pathways leading to DNA repair or apoptosis. In addition, apoptosis could be counteracted by nuclear factor kappa B (NF-κB), the main anti-apoptotic transcription factor in the DDR. Despite the high clinical relevance, the interplay of the DDR pathways is poorly understood. For therapeutic purposes DNA damage signalling processes are induced to induce apoptosis in tumour cells. However, the efficiency of radio- and chemotherapy is strongly hampered by cell survival pathways in tumour cells. In this study logical modelling was performed to facilitate understanding of the complexity of the signal transduction networks in the DDR and to provide cancer treatment options.

Results: Our comprehensive discrete logical model provided new insights into the dynamics of the DDR in human epithelial tumours. We identified new mechanisms by which the cell regulates the dynamics of the activation of the tumour suppressor p53 and NF-κB. Simulating therapeutic intervention by agents causing DNA single-strand breaks (SSBs) or DNA double-strand breaks (DSBs) we identified candidate target proteins for sensitization of carcinomas to therapeutic intervention. Further, we enlightened the DDR in different genetic diseases, and by failure mode analysis we defined molecular defects putatively contributing to carcinogenesis.

Conclusion: By logic modelling we identified candidate target proteins that could be suitable for radio- and chemotherapy, and contributes to the design of more effective therapies.

Show MeSH

Related in: MedlinePlus

Logical interaction hypergraph of the DDR. Genotoxic agents generate DSBs, thereby triggering signal transduction pathways to cell cycle arrest and onset of apoptosis. Arc colours mark earliest (time step 1, black), intermediate (time scale value 2, red) and latest (time scale value 3, blue) interactions, respectively.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3473366&req=5

Figure 1: Logical interaction hypergraph of the DDR. Genotoxic agents generate DSBs, thereby triggering signal transduction pathways to cell cycle arrest and onset of apoptosis. Arc colours mark earliest (time step 1, black), intermediate (time scale value 2, red) and latest (time scale value 3, blue) interactions, respectively.

Mentions: Based on quality-controlled literature data, we built a discrete logical model of the response to SSBs and DSBs in human epithelial cells. The model encompasses 96 regulatory components, connected by 98 interactions. It is represented by a logical interaction hypergraph (Figure 1), and a list of logical functions describing the interactions (Additional file 1: Table S1). The numbers assigned to interactions in Figure 1 correspond to the numbers of the logical functions. The network shows the typical structure of signal transduction networks: the input layer is given by stimuli, which damage the DNA, from where signals are being transmitted to and processed in the intermediate layer, finally reaching the output layer (cell cycle arrest and ‘onset of apoptosis’). We chose ‘onset of apoptosis’ instead of ‘apoptosis’ as an output, as this output corresponds to the beginning of apoptotic processes, but not to completion of apoptosis, ie. cell death. The activity levels of most regulatory components (nodes) are represented by Boolean state variables, i.e. they can only attain the values ‘0’ (inactive or absent) or ‘1’ (active or present). Ternary (three-valued) variables were only assigned to phosphorylated ataxia telangiectasia mutated (ATM-P), phosphorylated inhibitor of kappa B kinase (IKK-complex-P) and inhibitor of kappa B α (IκBα). In that way, we took account for the fact that each of those components differs in its functions, depending on whether its activity is low (‘1’) or high (‘2’). Specifically, a low activity of ATM (ATM-P = 1) is required for inactivation of the ATM phosphatase protein phosphatase 2 A (PP2A) [24] (interaction 22). Once PP2A is inactivated, DSBs can induce high activity of ATM (ATM-P = 2), which is now able to phosphorylate further substrates [25] (interaction 23). Similarly, the IKK complex has a low basal activity (IKK-complex-P = 1), which is sufficient for partial degradation of IκBα (decreasing IκBα’s activity level from ‘2’ to ‘1’), leading to activation of proto-oncogene c-Rel (c-Rel) in absence of induced DNA damage [26,27] (interaction 67). Upon induction of DNA damage, the IKK complex attains high activity (IKK-complex-P = 2), which enables more degradation of IκBα (IκBα = 0), enabling the activation of the NF-κB dimers p50-p65-P and p50-p50 [10] (interactions 68, 69). For some structural analyses, we took account for the limited knowledge of time-dependent signal transmission by assigning each interaction to one of three time scale values. Interactions composing the signal transduction pathways leading to activation/inactivation of components that are directly linked to the components “CELL CYCLE ARREST” or “ONSET OF APOPTOSIS” were assigned to time scale value 1, as long as literature data did not indicate a distinct delay. Examples of components that are directly linked to “CELL CYCLE ARREST” or “ONSET OF APOPTOSIS” are the transcription factors. Time scale value 2 was assigned to interactions that also lead to cell cycle arrest, apoptosis, or anti-apoptosis, but were shown to occur distinctively later than interactions of time scale value 1. For example, p53-induced protein with a death domain (PIDD) binds to NEMO (interaction 44, time scale 1), and later, PIDD binds to RIP1-associated ICH-1/CED-3 homologous protein with a death domain (RAIDD) (interaction 43, time scale 2) [28]. Time scale value 2 was also assigned to interactions linked directly to the regulatory components “CELL CYCLE ARREST” or “ONSET OF APOPTOSIS”. Activation of proteins that initiate switching off parts of the DDR was assigned time scale value 3. This was based on the assumption that these events occur during the latest phase of the DDR. For instance, Wip1 interrupts signal transduction pathways by dephosphorylating ATM and other proteins [29] (e.g. interaction 21). Accordingly, induction of Wip1 expression [30] (interaction 82) has been assigned to time scale value 3. Detailed information on assignments of time scale values are given in Additional file 1: Table S1. For most analyses, we simulated the DDR at time scale value 2, i.e., at a time before feedback inhibition comes into play. Our study focused on inhibitions and molecular defects interfering with cell cycle arrest, apoptosis, or anti-apoptosis. As follows from the considerations above, only time scale value 2 pertains to maximum activity of all components promoting cell cycle arrest, apoptosis, or anti-apoptosis. Hence, for this time scale value, the sensitivity of the simulation results to changes in time scales of interactions should be minimal.


Dynamics of p53 and NF-κB regulation in response to DNA damage and identification of target proteins suitable for therapeutic intervention.

Poltz R, Naumann M - BMC Syst Biol (2012)

Logical interaction hypergraph of the DDR. Genotoxic agents generate DSBs, thereby triggering signal transduction pathways to cell cycle arrest and onset of apoptosis. Arc colours mark earliest (time step 1, black), intermediate (time scale value 2, red) and latest (time scale value 3, blue) interactions, respectively.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3473366&req=5

Figure 1: Logical interaction hypergraph of the DDR. Genotoxic agents generate DSBs, thereby triggering signal transduction pathways to cell cycle arrest and onset of apoptosis. Arc colours mark earliest (time step 1, black), intermediate (time scale value 2, red) and latest (time scale value 3, blue) interactions, respectively.
Mentions: Based on quality-controlled literature data, we built a discrete logical model of the response to SSBs and DSBs in human epithelial cells. The model encompasses 96 regulatory components, connected by 98 interactions. It is represented by a logical interaction hypergraph (Figure 1), and a list of logical functions describing the interactions (Additional file 1: Table S1). The numbers assigned to interactions in Figure 1 correspond to the numbers of the logical functions. The network shows the typical structure of signal transduction networks: the input layer is given by stimuli, which damage the DNA, from where signals are being transmitted to and processed in the intermediate layer, finally reaching the output layer (cell cycle arrest and ‘onset of apoptosis’). We chose ‘onset of apoptosis’ instead of ‘apoptosis’ as an output, as this output corresponds to the beginning of apoptotic processes, but not to completion of apoptosis, ie. cell death. The activity levels of most regulatory components (nodes) are represented by Boolean state variables, i.e. they can only attain the values ‘0’ (inactive or absent) or ‘1’ (active or present). Ternary (three-valued) variables were only assigned to phosphorylated ataxia telangiectasia mutated (ATM-P), phosphorylated inhibitor of kappa B kinase (IKK-complex-P) and inhibitor of kappa B α (IκBα). In that way, we took account for the fact that each of those components differs in its functions, depending on whether its activity is low (‘1’) or high (‘2’). Specifically, a low activity of ATM (ATM-P = 1) is required for inactivation of the ATM phosphatase protein phosphatase 2 A (PP2A) [24] (interaction 22). Once PP2A is inactivated, DSBs can induce high activity of ATM (ATM-P = 2), which is now able to phosphorylate further substrates [25] (interaction 23). Similarly, the IKK complex has a low basal activity (IKK-complex-P = 1), which is sufficient for partial degradation of IκBα (decreasing IκBα’s activity level from ‘2’ to ‘1’), leading to activation of proto-oncogene c-Rel (c-Rel) in absence of induced DNA damage [26,27] (interaction 67). Upon induction of DNA damage, the IKK complex attains high activity (IKK-complex-P = 2), which enables more degradation of IκBα (IκBα = 0), enabling the activation of the NF-κB dimers p50-p65-P and p50-p50 [10] (interactions 68, 69). For some structural analyses, we took account for the limited knowledge of time-dependent signal transmission by assigning each interaction to one of three time scale values. Interactions composing the signal transduction pathways leading to activation/inactivation of components that are directly linked to the components “CELL CYCLE ARREST” or “ONSET OF APOPTOSIS” were assigned to time scale value 1, as long as literature data did not indicate a distinct delay. Examples of components that are directly linked to “CELL CYCLE ARREST” or “ONSET OF APOPTOSIS” are the transcription factors. Time scale value 2 was assigned to interactions that also lead to cell cycle arrest, apoptosis, or anti-apoptosis, but were shown to occur distinctively later than interactions of time scale value 1. For example, p53-induced protein with a death domain (PIDD) binds to NEMO (interaction 44, time scale 1), and later, PIDD binds to RIP1-associated ICH-1/CED-3 homologous protein with a death domain (RAIDD) (interaction 43, time scale 2) [28]. Time scale value 2 was also assigned to interactions linked directly to the regulatory components “CELL CYCLE ARREST” or “ONSET OF APOPTOSIS”. Activation of proteins that initiate switching off parts of the DDR was assigned time scale value 3. This was based on the assumption that these events occur during the latest phase of the DDR. For instance, Wip1 interrupts signal transduction pathways by dephosphorylating ATM and other proteins [29] (e.g. interaction 21). Accordingly, induction of Wip1 expression [30] (interaction 82) has been assigned to time scale value 3. Detailed information on assignments of time scale values are given in Additional file 1: Table S1. For most analyses, we simulated the DDR at time scale value 2, i.e., at a time before feedback inhibition comes into play. Our study focused on inhibitions and molecular defects interfering with cell cycle arrest, apoptosis, or anti-apoptosis. As follows from the considerations above, only time scale value 2 pertains to maximum activity of all components promoting cell cycle arrest, apoptosis, or anti-apoptosis. Hence, for this time scale value, the sensitivity of the simulation results to changes in time scales of interactions should be minimal.

Bottom Line: Simulating therapeutic intervention by agents causing DNA single-strand breaks (SSBs) or DNA double-strand breaks (DSBs) we identified candidate target proteins for sensitization of carcinomas to therapeutic intervention.Further, we enlightened the DDR in different genetic diseases, and by failure mode analysis we defined molecular defects putatively contributing to carcinogenesis.By logic modelling we identified candidate target proteins that could be suitable for radio- and chemotherapy, and contributes to the design of more effective therapies.

View Article: PubMed Central - HTML - PubMed

Affiliation: Institute of Experimental Internal Medicine, Otto von Guericke University, Leipziger Str, 44, Magdeburg, 39120, Germany.

ABSTRACT

Background: The genome is continuously attacked by a variety of agents that cause DNA damage. Recognition of DNA lesions activates the cellular DNA damage response (DDR), which comprises a network of signal transduction pathways to maintain genome integrity. In response to severe DNA damage, cells undergo apoptosis to avoid transformation into tumour cells, or alternatively, the cells enter permanent cell cycle arrest, called senescence. Most tumour cells have defects in pathways leading to DNA repair or apoptosis. In addition, apoptosis could be counteracted by nuclear factor kappa B (NF-κB), the main anti-apoptotic transcription factor in the DDR. Despite the high clinical relevance, the interplay of the DDR pathways is poorly understood. For therapeutic purposes DNA damage signalling processes are induced to induce apoptosis in tumour cells. However, the efficiency of radio- and chemotherapy is strongly hampered by cell survival pathways in tumour cells. In this study logical modelling was performed to facilitate understanding of the complexity of the signal transduction networks in the DDR and to provide cancer treatment options.

Results: Our comprehensive discrete logical model provided new insights into the dynamics of the DDR in human epithelial tumours. We identified new mechanisms by which the cell regulates the dynamics of the activation of the tumour suppressor p53 and NF-κB. Simulating therapeutic intervention by agents causing DNA single-strand breaks (SSBs) or DNA double-strand breaks (DSBs) we identified candidate target proteins for sensitization of carcinomas to therapeutic intervention. Further, we enlightened the DDR in different genetic diseases, and by failure mode analysis we defined molecular defects putatively contributing to carcinogenesis.

Conclusion: By logic modelling we identified candidate target proteins that could be suitable for radio- and chemotherapy, and contributes to the design of more effective therapies.

Show MeSH
Related in: MedlinePlus