Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse




Conference Paper

Predicting "Good" drug targets based on Metabolic Control Analysis


Gupta,  Shobhit
Max Planck Society;

There are no locators available
Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available

Gupta, S. (2002). Predicting "Good" drug targets based on Metabolic Control Analysis. In European Conference on Computational Biology 2002 (ECCB 2002) in conjunction with the German Conference on Bioinformatics 2002 (GCB 2002) (pp. P52-P52).

Cite as:
Poster abstracts: "Classical Drug Discovery" is a discovery process wherein screening is done for a large number of compounds and after a series of trials a candidate drug is discovered. During the past decade, with the advent of genomics and crystallographic techniques for compound identification, a more reliable drug development approach has taken shape, which is now known as "Rational Drug Discovery"[1]. The current work proposes another such rational approach to drug discovery. It involves a computational strategy to select the better drug targets, based on the structural properties of the metabolic pathway involved. Such "weak" enzymes would thereby require less inhibitor concentration for the desired inhibition and subsequently the drug then developed would be less susceptible for preclinical failures resulting due to toxic effects of the compound. The methodology is an application of the Metabolic Control Analysis to simplified pathway structures to get some correlation between the position and inhibitory characteristics of the enzymes. Thus, the major objectives are: · Comparing in-vivo and in-vitro conditions · Comparison of different types of inhibitors · Identifying the structures which can be used as "building blocks" for reconstructing any part of the metabolic web · Performing Metabolic Control Analysis simulations to figure out the weak points in the so-called building blocks. The Metabolic Control Analysis tool: Gepasi[2] Gepasi is a tool to simulate various metabolic pathways or any other chemical reaction for that matter. It inputs the metabolic pathway, the reaction kinetics and initial metabolite concentrations. It then simulates the pathway by calculating the various control coefficients[1] and thereby calculates the transient metabolite concentrations. It also computes steady state metabolite concentrations and provides the flexibility to choose between different integration algorithms like newton's integration, backward integration or a combination of the two. The most interesting feature of the tool is that it can scan across some concentration range for specific metabolite(s), thus allowing us to find an optimal concentration for the desired output. This feature in our case would enable us to find minimal inhibitor concentration required for the desired percentage inhibition. Methods A manual analysis of the complete metabolic web was done using the publicly available database of Metabolic Pathways, KEGG[3]. From this analysis a list of structures, building blocks (Fig. 1), was made which can be used for completely reconstructing the web. Subsequently, an analysis of the data present in the enzyme database BRENDA[4] was done to find some representative values of the kinetic parameters required for simulations. The simulations were then done for different building blocks, and different inhibitor types using the software Gepasi. The resulting data was parsed and processed using perl scripts for the ease of automation at later stages. The processed data was then used to plot using Matlab. Simulations All the simulations were done over a range of inhibitor concentration with the idea that the desired percentage inhibition is different for different metabolites producing effective results, which in this case would be the killing of a pathogen or controlling some abnormal enzyme activity. 1.The first set of simulations was done to differentiate between the in-vitro and in-vivo conditions. 2.The next set of simulations compared different types of inhibitors, Viz: competitive, uncompetitive and allosteric. In order to compare the inhibitor types the linear pathway was chosen. 3. In the final set of simulations all the five structures were simulated and competitive inhibitor was chosen as a representative case. Simulation Results 1. It is more difficult to inhibit in-vivo than in-vitro. The plots obtained could be used as an estimate for scaling from in-vitro inhibitory concentration to the concentration required for desired inhibition in in-vivo conditions. 2. It was seen that competitive and allosteric inhibitors behaved in a similar fashion with the exception that the inhibitor concentration required by an allosteric inhibitor is much less than that required by a competitive inhibitor. Uncompetitive inhibitors behave differently and also require much higher inhibitor concentration for desired inhibition. 3. For different structures (ref: Fig. 1) the following "weak" points were identified for competitive and allosteric inhibitors. Additionally the plots can give an estimate of the required inhibitor concentration. a. For Linear pathways, it is best to inhibit at the first or the earliest possible step of the pathway whose end product is to be inhibited. b. For Forward branches, it is best to inhibit at the first step after the branching occurs. c. For Reverse branches it is best to inhibit after the branching occurs. d. For Circular pathways, it is best to inhibit at an earliest opportunity, such that there is no branching from the inhibition point till the product to be inhibited. Future Plans A more comprehensive study of each of the structure would be done, with variations in kinetic parameters. Some more examples of a circular pathway have would be designed and simulated, for a better understanding of it. Finally a more extensive study of the inhibitor types would be done, including more inhibitor types. [1] Fell D. Understanding the control of Metabolism. [2] Mendes P. (1993) GEPASI: A software package for modelling the dynamics, steady states and control of biochemical and other systems. Comput. Applic. Biosci. 9, 563-571. [3] KEGG database of metabolic pathways : [4] BRENDA Enzyme database : [5] Eisenthal R. & Bowden A,C. (1998) Prospects for Antiparasitic Drugs JBC 10, 5500-5505.