An enormous amount of information has accumulated about the molecular components that govern the cell division cycle and apoptosis, information that is crucial to the development of new anti-cancer therapies. However, the relevant molecular interaction networks are so complicated that it is often difficult to understand how they function. Comprehensive network diagrams, using a standardized notation, therefore are essential in the same way that circuit diagrams are essential in electronics. Diagramming bioregulatory networks however presents a major challenge, because of the large role played by multimolecular complexes and protein modifications. Several years ago, we devised a notation for molecular interaction maps (MIM) that many in the field think is the best way to represent network functions and interaction details. We have used MIMs in many publications and examples have been prominently displayed in journal issues (such as being featured on the cover). To generate a MIM, one has to adhere to formal MIM notation rules (described in Kohn et al Mol. Biol. Cell 17:1-13, 2006);this requirement imposes a useful discipline of logic that enforces clear understanding of the available information about a network's structure. The MIM notation has important advantages over other proposed notations and has the unique ability to represent combinatorial complexity in molecular interaction networks (Kohn et al Mol. Systems Biol. doi:10.1038/msb4100088, 2006). In order to make MIMs conveniently accessible, we post them on the internet with links to annotations, references, and to other databases (http://discover.nci.nih.gov/mim/). Recently, we showed for the first time how MIMs can represent network events at the DNA and chromatin levels. We prepared a MIM describing the network involved in nucleosome disassembly in front of a DNA replication fork, assembly behind the replication fork, and the copying of epigenetic information onto the replicated chromatin. We used this MIM in a comprehensive review of the molecular interactions taking place during these events, thereby showing the advantages of systematizing information in this way (Kohn et al Mol. Biol. Cell 19:1-7, 2008). We have now extended that work to show how MIMs can be more informative and easier to interpret by means of a hierarchical representation. We applied that hierarchical procedure to map signaling from stalled replication forks to the controllers of cell cycle checkpoints and trans-lesion DNA repair synthesis. That work was recently published (Kohn et al. Cell Cycle 8:14, 2281-2299;15 July 2009). Stalled replication forks signal by way of a complex network of molecular interactions. To clarify those complexities, we designed a novel hierarchical assembly procedure in which the molecular components of the network are mapped at hierarchical levels according to interaction step distance from the DNA region of stalled replication. The hierarchical MIM allowed us to disentangle the networks interlocking pathways and loops and to suggest functionally significant features of network architecture. The MIM shows how parallel pathways and multiple feedback loops can provide failsafe and robust switch-like responses to replication stress. Within the central level of hierarchy ATR and Claspin together appear to function as a nexus that conveys signals from many sources to many destinations. We noted a division of labor between those two molecules that separates enzymatic and structural roles. In addition, the network architecture disclosed by the hierarchical map suggested a model for how molecular crowding and the granular localization of network components in the cell nucleus can facilitate function. This work was recently published (Kohn et al. Cell Cycle 8:14, 2281-2299;15 July 2009). Currently we are developing new and updated MIMs of the complex and intricate signaling systems that determine whether and when cells commit to replication, either during the cell cycle or from quiescence, as well as the signaling connections of those systems to DNA repair processes and the checkpoint responses to DNA damage. A key component of the DNA damage response system is the p53 tumor suppressor, a commonly mutated gene in human cancer. In response to DNA damage or uncontrolled proliferation signals from cancer cells, p53 arrests the progress cells through the cell cycle or causes cells to die by apoptosis. Those responses are governed in large part by the activity of p53 as a transcription factor, controlled through interactions with Mdm2 and MdmX. Although the action of Mdm2 on the control of p53 is well understood, the role of Mdmx on this control network is not. We therefore assembled an updated MIM of the interactions that comprise the p53 control network. From that MIM, we extracted for computer simulation an explicit model network that we thought could be the core of the control system involving both Mdm2 and MdmX. Continuing our work during the previous report period, we carried out computer simulations, in which we surveyed the kinetic parameter space for the possible DNA damage-induced p53 response patterns. The results further support our inference that MdmX may amplify or stabilize DNA damage-induced p53 responses and that the effects of MdmX are mediated by accumulated reservoirs of p53:MdmX and Mdm2:MdmX heterodimers. Regions of oscillatory and switch-like responses were mapped to the kinetic parameter space. Of particular interest was the ability of MdmX to dampen or prevent oscillations. A possible role of MdmX therefore may be to prevent uncontrolled oscillations that could produce erratic cell behavior or unnecessary cell death. The mathematical aspects of this work are assisted by collaboration with Dr. Geoffrey McFadden of NIST. We are conducting experiments to look for such effects in cell culture systems. We are collaborating with an international consortium of researchers to develop systems biology gaphics notations (SBGN) that may a serve as a standard. The SBGN effort was recently summarized by Le Novere et al (38 co-authors, including M. I. Aladjem and K. W. Kohn from our Laboratory) in Nature Biotechnology 27, 735 - 741 (2009). SBGN includes MIM-like entity-relationship diagrams, a formal description of which can be accessed on line at www.sbgn.org/Documents/Specifications, under the sections entitled Entity Relationship language. Our LMP group is independently developing computer facilities specific to MIMs, which, for technical reasons, differs in some respects from SBGNs entity-relationship language. SBGNs language may be more amenable to some computer application and integration with other parts of SBGN. Our MIM notation however seems better suited for depicting network functions for biologists. We are therefore developing a formal computer-readable structure specifically for MIM, while allowing for future compatibility with the SBGN language. In that development, we collaborate with members of the Genomics and Bioinformatics group in our Laboratory, and with Dr. Ruth Nussinov at NCI/Frederick and her computer scientist associates at Bogazici University in Turkey.
Showing the most recent 10 out of 14 publications
