(283d) Optimization Based Pathway Analysis to Elucidate the Effects of Burn Injury On the Hepatic Metabolism

The hypermetabolic state in liver after burn injury is characterized by a significant upregulation of glucose, fatty acid and amino acid turnover with increasing demand for hepatic energy ultimately resulting in organ dysfunction. To better characterize the hepatic metabolic state and elucidate the mechanism underlying the hypermetabolic response to burn injury, metabolic engineering tools can be used in combination with the available experimental data. This can provide clues about the new experimental points to be obtained and the potential targets of intervention in order to restore liver functionality after burn injury.

Metabolic flux analysis is used to determine the flux distribution vector of a metabolic network. Assuming that the internal metabolites are at pseudo steady state, the mass balance of the network is constructed by using stoichiometric matrix of metabolic network, where rows correspond to the metabolites and columns correspond to the reactions. One of the important problems in metabolic flux analysis is that the number of measured fluxes is not enough to obtain a determined system. To overcome this limitation, a range of possible values for each flux is assigned using linear programming where each flux is maximized or minimized while allowing all other flux values to vary [1]. This optimization problem not only defines the boundaries of the solution space, but also provides a number of flux distribution vectors within the solution space. Singular Value Decomposition analysis can then be used to analyze the solution space and find the best approximate flux-distribution that captures as much original variation as possible.

Pathway analysis based on elementary modes and extreme pathways are successfully used to reveal inherent properties of the metabolic networks. An elementary mode consists of the minimum number of reactions that it needs to exist as a functional unit, whereas an extreme pathway is the systemically independent subset of elementary modes [2]. Since every flux distribution can be written as a linear combination of the elementary modes or extreme pathways, a weight can be assigned for each corresponding pathway which can be interpreted as an indication of the importance of that pathway in the network. However, the decomposition of a steady state flux vector into pathways is not necessarily unique. Therefore, optimization based methods have been used for the unique decomposition of a steady state flux vector into pathways and identifying the range of possible values for the importance of each pathway [3,4,5].

In this work, considering the hepatic metabolic network involving main liver specific pathways and the experimentally obtained external fluxes [6], metabolic flux analysis is used to determine the values of all unknown fluxes. Then, pathway analysis is applied to elucidate the main pathways connecting network uptakes and outputs. Considering the experimentally obtained external fluxes, all exchange reactions in the network are assumed to be irreversible. Thus the sets of meaningful elementary modes and extreme pathways coincide [7]. Finally, different objective functions are used to determine the corresponding weight values that represent the importance of the pathway. These optimization problems result from minimization of the elementary mode activity [4,8], maximization of the activity of pathways responsible for urea production and/or oxygen consumption, and maximization of the number of active pathways [3].

A number of important results are obtained using pathway and metabolic flux analysis. The pathways including formation of glucose via gluconeogenesis from lactate, glutamine or asparagine, and the pathway responsible for the production of urea from arginine are found to be very important although different methods result in different weight values. Detailed comparison of active pathways obtained from different optimization-based methods also reveals critical information regarding metabolic objectives. It is found that the results of maximizing the number of active pathways and the maximizing the activity of urea production and O2 consumption are very similar which implies that the system tries to maximize redundancy after burn injury. This means that the liver uses as many pathways as possible, thus increasing the demand for oxygen and the production of urea. Identifying all possible metabolic pathways with their weight values and analyzing the external fluxes in conjunction with the inputs of pathways provide important knowledge on metabolic mechanism that can help to elucidate the effects of a perturbation on the liver metabolism.

References:

1. Llaneras F, Picó J (2007) An interval approach for dealing with flux distributions and elementary modes activity patterns. Journal of Theoretical Biology 246: 290-308.

2. Papin JA, Stelling J, Price ND, Klamt S, Schuster S, et al. (2004) Comparison of network-based pathway analysis methods. Trends in Biotechnology 22: 400-405.

3. Nookaew I, Meechai A, Thammarongtham C, Laoteng K, Ruanglek V, et al. (2007) Identification of flux regulation coefficients from elementary flux modes: A systems biology tool for analysis of metabolic networks. Biotechnology and Bioengineering 97: 1535-1549.

4. Schwartz JM, Kanehisa M (2005) A quadratic programming approach for decomposing steady-state metabolic flux distributions onto elementary modes. Bioinformatics 21: 204-205.

5. Wiback SJ, Mahadevan R, Palsson BØ (2003) Reconstructing metabolic flux vectors from extreme pathways: defining the [alpha]-spectrum. Journal of Theoretical Biology 224: 313-324.

6. Lee K, Berthiaume F, Stephanopoulos GN, Yarmush ML (2003) Profiling of dynamic changes in hypermetabolic livers. Biotechnology and Bioengineering 83: 400-415.

7. Klamt S, Stelling J (2003) Two approaches for metabolic pathway analysis? Trends in Biotechnology 21: 64-69.

8. Schwartz JM, Kanehisa M (2006) Quantitative elementary mode analysis of metabolic pathways: the example of yeast glycolysis. Bmc Bioinformatics 7.