21st Century COE Program Human Nutritional Science on Stress Control
Research Aim
Dr. Takeda's Labo
Dr. Terao's Labo
Dr. Miyamoto's Labo
Dr. Nakaya's Labo
Dr. Rokutan's Labo
Dr. Chuman's Labo
Dr. Kaji's Labo
Dr. Ohmori's Labo
Dr. Sei's Labo
Dr. Chuman's Labo
Prof. Hiroshi Chuman
Department of Theoretical Chemistry for Drug Discovery
Dr. Chuman's LaboCOE Appointed Associate Professor Noriaki Okazaki
COE Researcher Takashi Kinoshita
Modern society is suffering from various kinds of stresses which cause serious problems of stress-related illness and depression. In recent years, the concept of nutrition and diet therapy has been widely accepted as a preventive measure against these illnesses and some of the food compounds like flavonoids are attracting much attention for their potential efficacy. We have been studying functional mechanisms of functional food compounds using the various methodologies developed in the modern drug discovery research such as QSAR (quantitative structure-activity relationships) analysis, chemoinformatics and biopathway simulation. Our major research projects include (1) construction of the integrated database of flavonoids for nutrition and food sciences, (2) classification and screening of antidepressants, (3) estimation of Log P (n-octanol/water partition coefficient) and its application to nutrition and food sciences, (4) QSAR studies on the inhibitory activity of flavonoids against myeloperoxidase, and (5) biopathway analysis for the network-based drug/food design.
Novel Functional Foods
1. Integrated database of flavonoids for nutrition and food sciences
Recently, various databases of chemicals have been compiled to help biological and/or chemical research works, but no comprehensive database of food constituents is available yet. To further advance the research on this area we have compiled an integrated database which constitutes chemical structures, biological activities, physicochemical properties of flavonoids and the related compounds, as shown in Fig. 1. This database is thought to be useful for systematic analyses based on QSAR and chemoinformatics to design new functional foods, as described in the following sections.
2. Classification and screening of antidepressants
Antidepressants make up one of the largest therapeutic areas of the current drug market. The understanding of their QSARs will be helpful for finding and designing new functional foods. We have developed virtual screening methods for a huge number of target chemicals related to depression. By applying a novel informatics method called “Support Vector Machine”, we succeeded in developing new models that can predict the target enzyme of each antidepressant only based on the chemical structure at the ~70% confidence level. We have also carried out structure pattern-activity relationship studies of flavonoids for lipoxygenase inhibition activity by the means of Support Vector Machine.
Figure 2. Log <i>P</i>3. Estimation of Log P and its application to nutrition and food sciences
Molecular hydrophobicity plays an essential role in the processes such as transporting a chemical to its target organ(s) and binding it with its target receptor(s) in a body. Log P, a measure of hydrophobicity, has been taken as the first choice among various QSAR descriptors (Fig.2). We have found the nice linear relationships between the Log P values and experimental values derived from the HPLC measurement. By this method the Log P values of various structurally diversified compounds which include flavonoids can be now estimated with high accuracy.
Figure 1. Construct of flavonoid DB
Figure 3. Docking simulation4. QSAR studies on the inhibitory activity of flavonoids against myeloperoxidase
We have performed QSAR analyses of inhibitory activities of flavonoids for myeloperoxidase (MPO). MPO is one of the major enzymes of the antimicrobial system of mammalian neutrophils. The excess activation of an immune system during inflammation possibly causes host cell/tissue damages via harmful oxidants. Inhibiting MPO activities from excess activation of neutrophils during acute and/or chronic inflammation may prevent tissue damages. Quercetin, particularly, strongly inhibits dityrosine formation from the MPO-catalyzed oxidation of tyrosine. In this research, the radical scavenging and inhibitory activities of twenty-three flavonoids including quercetin are studied by QSAR.
The correlation equations show that the electronic factors like ESOMO govern the radical scavenging activities. However, Log P as well as the electronic factors is significant for the inhibitory activity. These results suggest that flavonoids interact with MPO in the inhibition of dityrosine formation. We predict that the B-ring in flavonoids can bind to the heme in the active site based on the docking simulation of quercetin with MPO (Fig.3).
Figure 4. three-dimensional ligand-protein complex5. Understanding the function of food compounds at atomic and electronic levels
Predicting enzymatic reactions is a crucial problem of biochemistry and food sciences, many methods to estimate biding properties for protein–ligand complexes have been reported. QSAR is the most widely used among them. However, a structure of target receptor has been usually treated as a “black box” in QSAR analyses. Recently, many three-dimensional structures of biologically important proteins have been solved by means of X-ray and NMR techniques. In this context, we have developed a novel QSAR method based on a complex structure between a drug and its target protein, combined with molecular dynamics and molecular orbital calculations. Being successful in finding some logical links between results of classical QSAR and molecular level simulation for papain hydrosis, we understand how the classical QSAR descriptors can be interpreted in a detailed three-dimensional ligand-protein complex. (Fig.4)
The novel QSAR will enable us to show what is occurring in the binding processes at atomic and electronic levels and suggest us how to design drugs to control their activities and functions. It is also applicable for the elucidation of functions of various food compounds.
6. Biopathway analysis for the network-based drug/food design
Foods and drugs interact with their target molecules such as receptors and enzymes, which triggers a series of biochemical reactions to produce their specific physiological effects on a human body. For the successful in silico drug/food design in terms of reliability and efficiency, it is essential to analyze the dynamics of relevant biochemical reaction networks (biopathway analysis), which fills in the large gap lying between chemical knowledge at the molecular level and biological knowledge at the cellular level.

Figure 5. Analysis of apoptosis network by <i>React</i> systemWe are currently working on the dynamical analysis of biopathway networks by using the home-made reaction compiler system React with the aim of developing a novel network-based drug design system. The network-based drug/food design will be utilized not only to identify the target molecules by finding the appropriate action points in the network but also to predict the effects and side-effects in a more quantitative manner. The computational biopathway simulation is thus expected to reduce the risk, costs and time required for drug discovery by eliminating the false drug leads at the early stage, which contributes to the avoidance of unnecessary animal and preclinical tests.
For the quantitative analysis of biopathway networks, one needs to construct a set of reaction rate equations for each constituent species (ordinary differential equation system). We have developed a reaction compiler system React which enables automatic generation of reaction rate equations from the list of reactions written in a highly readable format. Due to its high modularity, React can be used as a component tool for the integrated simulation platform which facilitates various nonlinear system analysis as well as animated visualization of the reaction dynamics (Fig. 5b). Additionally, React provides the interface to SBML (systems biology markup language) which allows easier linkage to other external software.
We have carried out systematic analysis of the Fas signaling-induced apoptosis network (Fig. 5a) by using React, especially focusing on the phenotypic difference of type I/II cells. The type I/II cells are characterized by fast/slow caspase-3 activation and low/high sensitivity to Bcl-2 overexpression, respectively. The cell-type difference is mechanistically accounted for by the different contributions of extrinsic pathway (D channel) and intrinsic pathway (M channel), where the pathway dominance is found to be controlled primarily by caspases-8 and -9 (Fig. 5c). We propose a simplified network model which well reproduces the dynamic behavior of the full network model.

We would like to thank former COE members, Dr. X. Liu, Dr. Z. Lepp and the graduate student for their significant contributions to our project.
  1. Itokawa D, Nishioka T, Fukushima J, Yasuda T, Yamauchi A, Chuman H, “Quantitative Structure - Activity Relationship Study of Binding Affinity of Azole Compounds with CYP2B and CYP3A”, QSAR & Combinational Science, 26(7), 828-836, 2007.
  2. Kinoshita T, Lepp Z, Kawai Y, Terao J, Chuman H, “An Integrated Database of Flavonoids”, Bio Factors, 26, 179-188, 2006.
  3. Yoshida T, Lepp Z, Kadota Y, Satoh Y, Itoh K, Chuman H, “Comparative Analysis of Binding Energy of Chymostatin with Human Cathepsin A and its Homologous Proteins by Molecular Orbital Calculation”, J. Chem. Info. Model., 46(5), 2093-2103, 2006.
  4. Lepp Z, Kinoshita T, Chuman H, “Screening for New Antidepressant Leads of Multiple Activities by Support Vector Machines”, J. Chem. Inf. Model., 46, 158-167, 2006.
  5. Chuman H et al, "Drug Discovery Using Grid Technology (Chapter 12)”, In Modern Methods for Theoretical Physical Chemistry for Biopolymers, Edited by Straikov EB, Lewis .JP, Tanaka S, Elsevier, 227-247, 2006.
  6. Liu X, Tanaka H, Yamauchi A, Testa B, Chuman H, “Determination of lipophilicity by reversed-phase high-performance liquid chromatography: Influence of 1-octanol in the mobile phase”, J Chromatogra A, 1091, 51-59, 2005.
  7. Liu X, Tanaka H, Yamauchi A, Testa B, Chuman H, “Lipophilicity Measurement by Reversed-Phase High-Performance Liquid Chromatography (RP-HPLC): A Comparison of Two Stationary Phases Based on Retention Mechanisms”, Helv. Chim. Acta., 87, 2866-2876, 2004.
  8. Okazaki N, Asano R, Kinoshita T, Chuman H, “Simple computational models of typeI / type II cells in Fas-induced apoptosis network”, J. Theor. Biol., in press.
  9. Ishii Y, Shimada T, Okazaki N, Hasegawa T, “Wetting-dewetting oscillations of liquid films during solution-mediated vacuum deposition of rubrene”, Langmuir 23, 6864-6868, 2007.
  10. Okazaki S, Okazaki N, Hirose Y, Furubayashi Y, Hitosugi T, Shimada T, Hasegawa T, “Quantitative Analysis of Thin-film Conductivity by Scanning Microwave Microscope”, Appl. Surf. Sci., in press.

All Rights Reserved,Copyright(c)University of Tokushima.