(699a) Metabolic Network Analysis and Dynamic Modeling of Skeletal Muscle Cells for Understanding the Effects of Plasma Free Fatty Acids

We have applied complex network analysis methods on the metabolic pathways of energy metabolism of skeletal muscle cells and identified key metabolites and reactions. The findings aid us in our ongoing studies on the development and analysis of the dynamic model of the system.

Diabetes is estimated to affect 6% of the world's adult population by the end of the decade [1]. Type 2 diabetes accounts for over 90 percent of 21 million diagnosed cases in the USA [2]. A better understanding of the alterations in insulin utilization in tissues caused by high levels of free fatty acids (FFA) will help identify the key steps to be targeted for treatment of Type 2 diabetes. Elevated FFA and intracellular lipid concentrations are the primary suspects for suppression of glucose transport and causing insulin resistance in liver and muscles. Recent research has revealed important details about glucose and insulin metabolism and the effects of plasma FFA concentrations. However, detailed metabolic networks and dynamic models describing these phenomena remain to be developed.

Energy metabolism of skeletal muscle cell is a complex network where each metabolite can influence the level of a handful of others via biochemical transformations and activation/deactivation of reactions. By utilizing current knowledge on biochemical reactions and activation/deactivation processes of the system, we have developed an "interaction network" which represents every possible direct influence that a certain metabolite can exert on another. The resulting network is a directed graph that consists of more than 64 nodes (metabolites) and 235 arrows (interactions). We have determined the diameter of the system which demonstrates the average number of influences that has to take place for a metabolite to affect the level of another metabolite. The metabolites with key roles within the system were identified according to their connection properties. We have analyzed the degree of emission (number of metabolites whose levels can be influenced by a certain metabolite), and the degree of reception (number of metabolites that influence the level of a certain metabolite) after n number of direct influences. By setting n=1, we have determined several metabolites (e.g. mitochondrial NAD+, mitochondrial FAD) as key metabolites according to the number of direct interactions with other metabolites. Further analysis was conducted by setting n>1 and metabolites such as oxygen were observed to interact with a significant number of metabolites of the system with n being as low as 3.

We have identified the network motifs [3] of the interaction network. Network motifs are patterns of interconnections occurring in complex networks at numbers that are significantly higher than those in randomized networks. Different networks that belong to a certain class (e.g. biological) may share the same unique network motifs. Therefore, motifs can define broad classes of networks, each with specific types of elementary structures. The significance of these structures raises the question of whether they have specific roles in the network. If they do, they might be used to understand the network dynamics in terms of elementary computational building blocks. Dr. Alon and co-workers have observed bi-fan and feed-forward loop motifs as common motifs of biological systems [3]. We have identified more than 3 unique motifs in the interaction network of the energy metabolism of skeletal muscle cells, none being bi-fan or feed-forward loop. We have investigated the number of occurrences of previously defined key metabolites in these motifs and observed that several metabolites participate only in specific motifs.

The knowledge gained from the methods above is utilized in the development and analysis of the dynamic model of the system. The key metabolites and motifs with specific roles are being integrated as the elements of the dynamic model. Principles of cybernetic modeling framework [4] are employed in the case of insufficient experimental data for model development. Simulation studies, bifurcation analysis, time series data analysis and metabolic control analysis tools are employed to identify the key elements and the dynamic properties (e.g. robustness and adaptation) of the system.

REFERENCES

1. Zimmet P, Alberti KG, Shaw J: Global and societal implications of the diabetes epidemic Nature 414: 782-787, 2001

2. Centers for Disease Control and Prevention, National Diabetes Fact Sheet. 2005

3. Milo, R. et. al., Network motifs: Simple building blocks of complex networks, 2004, Science, 303, 1538?1542

4. Varner J, Ramkrishna D., Metabolic engineering from a cybernetic perspective. 1. Theoretical preliminaries, Biotechnol. Prog., 1999, 15, 407-425.