PORTFOLIO

The recent developments in experimental and genomic methods allow accumulation of large amounts of data on metabolic and signaling networks in various biological systems ranging from bacteria to human. This data is already analyzed with classical methods to understand the general features of such biologic networks. While these efforts are useful, a full understanding of biologic networks can only be achieved by in depth analysis of their design principles and dynamic behavior. As all biological systems are a result of evolutionary processes, such in depth analysis must employ an evolutionary perspective.
This project aims to provide an understanding of evolutionary forces resulting in observed properties of metabolic and signal pathways. Such understanding will then be employed to explain and predict the behavior and dynamics of these pathways.

This project aims to fill an important gap in the analysis of metabolic and signaling networks by providing an evolutionary perspective and completing the ongoing conventional efforts.

During this two-year project we focused both on metabolic and signal transduction pathways and tried to use evolutionary approaches in order to understand key features of these pathways. Here, we briefly describe our findings for both these systems. A detailed discussion of the methods used and the results obtained can be found in our publications.

Study of Metabolic Pathways
Metabolic pathways control the production of biomass needed for the growth and survival of a cell. In recent years there have been increasing interest in how the structure of metabolic pathways in the cell looks like, whether it contains specific features, and what functional roles such features might have, if any. One discovery that came out of these studies is that certain molecules involved in the metabolic pathways of the cell are involved in such a way that there are many molecules that have few interaction partners, while few interact with extensively large numbers of others. In other words, metabolic pathways contain certain molecules (i.e. metabolites) that act as hubs.
To study the emergence of hub metabolites we implemented computer simulations of a widely accepted scenario for the evolution of metabolic pathways. Our simulations indicate that metabolic pathways with a large number of proteins (i.e. enzymes) that are highly specialized to catalyze certain metabolic reactions may evolve from a few multifunctional enzymes. In the evolutionary simulations we performed, enzymes duplicate and specialize in time, which leads to a loss of certain biochemical reactions and intermediary metabolites that participated in these reactions. This evolutionary process, when coupled with selection for higher growth rate (i.e. production of more biomass) leads to the emergence of hub metabolites in metabolic pathways. Applying different selection schemes and model parameters in our evolutionary simulations, we were also able to show that certain biochemical reactions are essential for the emergence of hubs. A detailed description of this project can be found in the resulting publication.

Study of Signal Transduction Pathways
Signal transduction pathways allow the cell to sense its environment and accordingly produce physiological responses, ranging from changes in the membrane polarity to changes in the gene expression patterns. Compared to metabolic pathways, signalling pathways are much more difficult to study mainly due to difficulties associated with characterizing proteins involved in such pathways. Despite such difficulties, there are few pathways for which we have a substantial understanding of pathway behaviour and properties. These indicate that most signalling pathways are highly complex, show substantial robustness to environmental and internal disturbances (for example changes in pH), and have some modular architecture. While the existing studies indicate such general properties for signal transduction pathways, it is not completely clear how these properties emerge or how they are related to the function of these pathways.
In order to address some of these questions we developed several approaches. Firstly, we simulated and analyzed the evolution of signal transduction pathways. This approach involves optimization of a generic model of signal transduction pathways to mediate a certain biological behaviour. The optimized (i.e. evolved) network can then be analyzed in order to understand the structural and dynamic network features that are necessary for achieving the selected behaviour. The most powerful feature of this approach is that it could highlight key features of a pathway, without any a priori knowledge on its structure or dynamics. We demonstrated the applicability of this approach using the well-studied chemotaxis pathway in bacteria as a benchmark.
Secondly, we addressed the question of structure-function relation in signal transduction pathways. These pathways are composed of different proteins that interact in a certain way to control each other’s activity. Hence, one could speak of a pathway topology, dictating which proteins are interacting with which. In this approach we catalogued the response of pathways for all possible topologies using three or four proteins. Using different assumptions in the modelling of pathways, we could understand the role of different biochemical processes and topology structures for achieving certain responses (i.e. oscillations). A detailed discussion of our findings can be found in the resulting publication.
Finally, we used an abstract model of signal transduction pathways and simulated their evolution under the effect of different mutational mechanisms. This approach allowed us to address the emergence of complexity in such pathways. Our running hypothesis was that complexity could emerge in such pathways, simple due to an imbalance in the effects of different mutations on the response of the pathway. Our results indicate that such a hypothesis could indeed explain the emergence of complexity in signalling pathways and in other systems of interacting units. The resulting publication from this study is currently under review at PNAS.

“The Evolution of Connectivity in Metabolic Networks”, Thomas Pfeiffer, Orkun S. Soyer, and Sebastian Bonhoeffer, PLoS Biology (2005), 3:1269-75.
“Signal Transduction Networks: Topology, Response, and Biochemical Processes”, Orkun S. Soyer, Marcel Salathé, and Sebastian Bonhoeffer, Journal of Theoretical Biology (2006), 238:416-25.
“Simulating The Evolution of Signal Transduction Pathways”, Orkun S. Soyer, Thomas Pfeiffer, and Sebastian Bonhoeffer, Journal of Theoretical Biology (2006), 241(2): 223-32.
“Evolution of Complexity in Signaling Pathways”, Orkun S. Soyer and Sebastian Bonhoeffer, submitted

keine