4-6 Feburary 2008 Tokyo International Forum, Tokyo, Japan
Human beings are now facing crisis. Crisis is in multiple fronts. Global climate change is the most significant crisis that weight loss pills has enormous impacts to destiny of man kind as well as numbers of species currently exists bankers life on this planet. At the same time, we are facing crisis on medical issues as well. Many industrial countries are facing rapidly aging society where medical systems may not be able to cope with. Developing countries are still struggling even to ensure descent medical services and access to drugs. While these crisis (among with other outstanding issues) are not solvable only within efforts in one scientific discipline, these are problems that link building biological sciences can provide significant contributions as these issues boil down to how can we understand and control biological systems at the difference scale: individual organism (medical crisis) and global ecology (global climate and resources crisis). Systems biology has potential to offer significant contribution to possible ramification of the urgent crisis we are facing today. It requires in-depth understanding of biological systems and their dynamics as well as controllability. A novel drug design paradigm, for example, focuses more on network-based and robustness-based approach that may result in more efficient and cost-effective drugs that may significantly mitigate problems for both industrial and developing countries, but in different ways.
At the same time, we must face the reality that the solution requires beyond biological science and involves technology, economics, and politics. While politics are often uncontrollable and unpredictable, it is important that science provides global view of how problem can be / should be solved. Interestingly, both biological systems and economic systems are highly interconnected, non-linear, and dynamic complex systems. Deeper understanding of linkage between bio-sphere and econo-sphere, as well as other inter-linked systems, is critically important for our future. This is profound challenge that is beyond the scope of a single workshop discussion. However, it is important to get things started to find cure for individual organisms and a global ecological system.
Stefan Hohmann (University of Gothenburg, Sweden)
Signal transduction pathways are the cellular information transmission routes that control metabolism, homeostasis, proliferation and differentiation. Hence, signaling pathways are most important targets for treatment of diseases. At present, we still do not have a complete picture of the architecture of international portfolio inc all signaling pathways, not even in yeast, and we lack an understanding of the systems level control of cell signaling. Understanding those properties and exploit signaling pathways in an educated manner as drug targets requires combining quantitative experimentation with mathematical modeling in a systems biology approach. It also requires the use the suitable experimental model organisms.
Our group studies two different signaling systems: Mitogen-Activated Protein Kinase (MAPK) pathways as well as the AMP-activated Kinase (AMPK) pathway. MAPK pathways control growth, proliferation, differentiation and stress responses in yeast, as in all eukaryotes and hence play roles in cancer, Garden Spot Village immunity and more. The yeast MAPK network with its four MAPKs is relatively simple but still its complexity is a challenge. By quantifying different steps in signaling and cellular responses combined with mathematical modeling we try to understand pathway control and pathway cross-talk. We have also employed a genetic overexpression approach (gTOW) to probe robustness against perturbation of protein copy number in part of the system. AMPK controls energy homeostasis in all eukaryotic cells and hence has relevance to diabetes and the metabolic syndrome. Using genetic and molecular analyses we try to understand energy sensing and signal transmission. Also in this case we use quantitative time course experimentation combined with modeling to understand dynamic properties of the pathway and the mechanisms controlling activation and deactivation. Using suitable model organisms inKey Environmental Inc systems biology as in genetics and genomics is an important step to drive discovery as well as driving lessons leeds and exploitation in drug development.
Mapping the global sensitivity of cellular network dynamics: sensitivity heat maps, a new global summation law and applications to network estimation and experimental design
David Rand (University of Warwick, UK)
I develop a more global approach to sensitivity analysis that studies the variation of the whole solution rather that focusing on just one output variable. This more global approach allows one to address which observable variables are affected by which parameter without having to choose the variable or parameter in advance. The results of this analysis can be summarised in (a) the sensitivity heat map and (b) the parameter sensitivity spectrum. The crucial observation that makes the theory applicable in practice by ensuring that for a given tolerance the above Stafford DUI lawyer objects are compact and manageable is that such network systems have a local geometric rigidity – a fact that is very important for other reasons also. This results lead to new approaches to parameter estimation and experimental optimisation. The latter aspect provides a framework to predict which experimental protocols and perturbations best reveal which aspects of the system and to provide a cost-benefit analysis of the different possibilities for data collection.
I will also briefly outline how this theory contributes to the optimization of structure and parameter estimation from molecular and imaging time-series data such as that performed in the following work: (i) to study extrinsic and intrinsic stochasticity in mammalian gene expression with the White group in Liverpool by inserting multiple reporters for the same gene into a cell, reconstructing the transcription rates of each of these and then comparing them; (ii) to analyse the roles of regulatory modules in regulating the promoter ppi claims activity of certain mammalian (with the Koentges group in Warwick); and (iii) to analyse network structure and trancription logic of key genes in the plant circadian clock with the Millar group (Edinburgh). I also hope to explain the role of the theory to optimal experimental design for nonlinear dynamic systems.
Reiko Tanaka (RIKEN)
One of the salient features of biological control systems, compared to man-made ones, is their ability to change their structures and/or functions to match the situation. This plasticity enables biological organisms to adapt to any environmental change and to take appropriate actions. Plasticity appears, depending on the time scale, as evolution, growth, or learning. Whereas man-made systems use specific regulatory mechanisms corresponding to the environmental change, biological systems, with their limited resources, have to exhibit a broad range of flexible actions with one regulatory mechanism in response to a wide variety of environmental changes. My research focuses on this characteristic and aims to reveal the essential elements of biological control at the cellular level, i.e. how biological systems adapt to wide variation in their environment and what regulatory mechanisms they use given their limited genetic and metabolic system resources. We coined the term compound control 1 for the characteristic mode of biological control that attains the flexibility. The key idea of compound control is the observation that a tremendous variety of complex cellular regulations result from spatial and temporal combinations of simple homogeneous computational media, such as interactions among different molecules (proteins, DNA), corresponding to various compound environmental changes. The homogeneous computational media at the cerebral level is considered to be neuron firings. This realization of complex control by a combination of homogeneous computational media is advantageous in terms of plasticity, adaptability, and evolvability in the sense that regulations newly required in the course of evolution are easily acquired through assembly of homogeneous computational elements. In this talk, I will present our recent attempts towards a theoretical basis for compound control: specifically, on (1) mathematical classification of regulatory mechanisms to handle the simultaneous occurrence of a huge number of environmental changes 2 and (2) a computational theory for intracellular biochemical processes by homogeneous computational media 3. Both studies are based on a mathematical framework for compound control, called gene regulatory units 1, and deal with the essential aspects of compound control.
R.J. Tanaka, H. Okano and H. Kimura (2006). Mathematical description of gene regulatory units, Biophysical J., vol. 91, 1235-1247
R.J. Tanaka and H. Kimura (2007). Mathematical classification of regulatory logics for compound environmental changes, J. Theor. Biol., in press
H. Kimura, H. Okano and R.J. Tanaka (2007). Stochastic approach to molecular interactions and computational theory of metabolic and genetic regulations, J. Theor. Biol., 248, 590-607
Hisao Moriya (PRESTO/JST, Cancer Institute)
Robustness is a property of a system that attempts to maintain its functions against internal and external perturbations. It is one of the fundamental and ubiquitously observed properties among biological systems. Understanding the cellular robustness is important, not only to gain insights in biology, Patrick Beharelle but also to identify potential therapeutic targets. Robustness is estimated by measuring how much parameters can be perturbed without disrupting essential functions; comprehensive, as well as quantitative perturbations of intracellular parameters, such as gene expression, are essential for solid robustness analysis.
We have developed a genetic screening method named “genetic tug-of-war” (gTOW) that allows systematic measurement of upper limit gene copy number. gTOW was applied for the robustness analysis of cell division cycle system in the model eukaryotes, S. cerevisiae (budding yeast) and S. pombe (fission yeast), and revealed an evolutionally-conserved fragile core in the system. The gTOW method is particularly suitable for systems biology research and demonstrates the value of comprehensive and quantitative perturbation experiment to uncover system-level properties of the cellular system.
Moriya H., Yoshida-Shimizu, Y. & Kitano H. (2006) In vivo robustness analysis of cell division cycle genes in Saccharomyces cerevisiae.
Nick Monk (University of Nottingham: BBSRC Centre)
The developing root of the plant Arabidopsis thaliana provides an excellent model system in which to study the integration of cellular signalling and mechanics. The establishment and maintenance of the basic architecture of the root depends on the coordination in space and time of cell proliferation, cell fate assignment, cell shape changes, and cell differentiation. The Nottingham Centre for Plant Integrative Biology (CPIB;www.cpib.info) has been established to develop an Indoor Fountains integrative approach to understanding this problem. By bringing together research teams with expertise in genetics, cell biology, 4D imaging, tissue mechanics and mathematical modelling, CPIB is working towards the ultimate objective of developing a dynamic multi-scale virtual devloping Arabidopsis root.
I will describe recent work towards this goal, focusing on the assignment of cell fates during the development of the root epidermis. I will show how the genetic and biochemical data can be integrated in a modelling formalism that allows an exploration of both the sufficiency of known network interactions and the extent to which additional assumptions about the model can account for wild type and mutant data. Our new model shows that an existing hypothesis concerning the auto-regulation of WEREWOLF does not account fully for the expression patterns of components of the network, a prediction that we have confirmed experimentally in transgenic plants. Rather, our modelling suggests that patterning depends critically on the directed movement of the CAPRICE and GLABRA3 transcriptional regulators between epidermal cells. These movements underlie a novel mechanism for pattern formation in planar groups of cells, centred on mutual support of two cell fates rather than diffusion-driven local activation and lateral inhibition.
Mariko Hatakeyama (RIKEN Genome Sciences Center, Japan)
A mechanism how cell utilizes identical signaling cassettes and yet induces distinct cellular phenotypes has been largely unknown. Quantitative analysis of signaling pathway revealed that transient and sustained ERK activities resulted in cellular proliferation and differentiation, respectively. And behind the distinct signaling profiles, pathway structures, such as positive- and negative-feedbacks, often has a profound effect. However, knowledge on how such a quantitative change in signaling pathways induces a qualitative change of the cells is still lacking. ErbB receptors are gate keepers of intracellular signaling network that ultimately define cell fate, and its deregulation is highly analysis together with signaling pathway analysis revealed that levels of ligand-induced immediate early transcription (IET) are controlled by levels of ErbB receptor activities. At the same time, duration of intracellular signaling activities is also controlled by the same receptor. Thus, levels of cellular activities are initially quantitatively controlled by the activities of ErbB receptors. However, a qualitative change in gene transcription appears after IET where IET products are stabilized by remaining prolonged signaling activities. Accordingly, cell uses a feedforward loop embedded in signaling and transcription network and transforms graded response into an all-or-none digital output to lead qualitative change. Such a strategy is seemed to be observed in various cells to stabilize its cellular states. Examples on breast and lung cancer cell lines that express ErbB receptors are discussed in this talk.
Ligand-Dependent Responses of the ErbB Signaling Network: Experimental and Modeling Analysis. Birtwistle, MR, Hatakeyama, M, Yumoto, N, Ogunnaike, BA, Hoek, JB & Kholodenko, BN. Molecular Systems Biology 3:144 doi:10.1038/msb4100188, 2007.
Quantitative transcriptional control of ErbB receptor signaling undergoes graded to biphasic response for cell differentiation. Nagashima, T, Shimodaira, H, Ide, K, Nakakuki, T, Tani, Y, Takahashi, K, Yumoto, N & Hatakeyama, M. J. Biol. Chem. 282, 4045-4056, 2007.
Bela Novak (Univ. Oxford, UK)
The eukaryotic cell cycle is a simple, cyclic developmental process with alternating DNA replication (S phase) and chromosome segregation (M phase). This alternation of S and M phases during mitotic cell cycle is enforced by irreversible transitions at G1/S, G2/M boundaries and exit from mitosis. It is common to explain these irreversible cell cycle transitions by irreversible degradation of certain regulatory proteins (see e.g. Lodish et al., 2004). However this simple and appealing view of irreversible cell cycle transitions is based on ambiguous notion of ‘irreversibility’. I will argue that irreversible transitions in the cell cycle (or in any other molecular control system) cannot be attributed to a single molecule or reaction, but derive rather from feedback signals (positive feedback or mutual antagonism) in molecular regulatory networks (Novák et al., 2007). This systems-level view of irreversibility is supported by theoretical considerations and by many experimental observations. I will provide an overview based on the underlying molecular networks and their mathematical models how the cell cycle transitions are made irreversible. I will show that the irreversible transitions are the consequences of the underlying ‘toggle-switches’ in the molecular mechanisms which have the properties of bistability and hysteresis. In general, the eukaryotic cell cycle engine can be described by two irreversible switches controlling G1/S and G2/M transitions and a mitotic ‘clock’.
Lodish H, Berk A, Matsudaira P, Kaiser CA, Krieger M, Scott MP, Zipusky L, Darnel J (2004) Molecular Cell Biology, 5th edn: Freeman, New York.
Novák, B., Tyson, J.J., Győrffy, B. & Csikász-Nagy, A. (2007): Irreversible cell cycle transitions due to systems-level feedback. Nature Cell Biology 9: 724-728.
The Centre for Integrated Systems Biology of Ageing and Nutrition (CISBAN) at Newcastle University is one of six Systems Biology Centres established in the UK over the last three years. It is funded by the BBSRC and EPSRC. We investigate the network of mechanisms contributing to cellular ageing in vitro, and how cell defects contribute to ageing in vivo. We use human cell cultures, ageing mice and the yeast Saccharomyces cerevisiae, as model organisms, and target systems fundamental to the ageing process, including the roles of mitochondrial dysfunction, telomere erosion, oxidative stress and protein homeostasis. Our approach is highly interdisciplinary, with bioinformaticists, statisticians and computational modellers working closely with wet lab biologists in an iterative cycle of data generation, analysis and modelling.
One of the approaches we use is interactome analysis. The interactome of a cell can be defined as the complete set of interactions between all of the cell’s genes and gene products. Interactomes are constructed using data integration over a range of omice data sets, each of which captures different aspects of the cell biology. Large-scale data integration is an active field of research, involving both computational (eScience, grid technology) and statistical (Bayesian probability) problems to overcome issues of data set size and noise. Interactomes can then be used for tasks such as the assignment of putative function to unknown genes, the identification of functional modules, the identification of previously unrecognized interactions between genes of interest, and informing low-level computational modelling.
This talk will present a broad overview of the organization and research interests of CISBAN, followed by an introduction to interactome analysis, focussed upon some of the applied research being carried out in CISBAN.
Stephen Muggleton (Imperial College, London UK: BBSRC Centre)
Huaiyu Mi (SRI International, Menlo Park, CA, USA)
With the availability of whole genomes from various model organisms and increasing amount of experimental data of genes, proteins, and biological pathways, it becomes obvious that a systems approach to biological research is essential. Such an approach not only provides us an opportunity to study how proteins interact and function in the context of pathway within one particular living system, but also allows us to examine how pathway networks have evolved.
The PANTHER Classification System (1,2) brings a module for representing pathways and networks together with a module for representing molecular evolution. The PANTHER Pathway module is an expert curated pathway ontology generated using the SBML standard (3) with a pathway network editing software called CellDesigner (4). As a result, detailed molecular events of biochemical reactions are captured on the diagrams, and stored in file in SBML format, which keeps consistency between the data and the diagrams. PANTHER library module is a large collection of protein families that have been subdivided into functionally and evolutionarily related subfamilies (2). All pathway components are directly linked to protein sequences from the PANTHER library module through manual curation, connecting pathways to molecular phylogenetic and genomic data. (1)
The infrastructure of PANTHER Pathway and its unique relationship with the phylogenetic information in the PANTHER Library enables us to explore the relationship between molecular sequence evolution, and concomitant evolution of pathways. The phylogenetic trees for protein families in the PANTHER Library are built such that orthologous and paralogous groups of proteins are easily depicted. As most of regulatory pathway research has been done in non-human organisms, the phylogenetic trees allow us to reliably infer the human proteins likely to be involved in a particular pathway, using its evolutionary relationships with proteins in other organisms. We will also present some examples of how duplicated genes have recruited relatively conserved reactions or pathway “modules” to create new pathways. The understanding of pathway evolution maybe helpful to our future research in drug discovery.
Mi, H., Guo, N., Kejariwal, A. and Thomas, P.D. (2007) PANTHER version 6: protein sequence and function evolution data with expanded representation of biological pathways. Nucleic Acids Res, 35, D247-252.
Thomas, P.D., Campbell, M.J., Kejariwal, A., Mi, H., Karlak, B., Daverman, R., Diemer, K., Muruganujan, A. and Narechania, A. (2003) PANTHER: a library of protein families and subfamilies indexed by function. Genome Res, 13, 2129-2141.
Hucka, M., Finney, A., Sauro, H.M., Bolouri, H., Doyle, J.C., Kitano, H., Arkin, A.P., Bornstein, B.J., Bray, D., Cornish-Bowden, A. et al. (2003) The systems biology markup language (SBML): a medium for representation and exchange of biochemical network models. Bioinformatics, 19, 524-531.
Kitano, H., Funahashi, A., Matsuoka, Y. and Oda, K. (2005) Using process diagrams for the graphical representation of biological networks. Nat Biotechnol, 23, 961-966.
Michael Hucka (California Institute of Technology, USA)
A computational model represents a modeler's understanding of the structure and function of part of a biological system. As the number of quantitative models continues to grow, and they become ever more sophisticated, they collectively represent a significant accumulation of knowledge about the structural and functional organization of biological systems. Enabling effective sharing of such quantitative models is the driving vision behind SBML and several adjunct efforts that I will describe.
The Systems Biology Markup Language (SBML) is a machine-readable exchange format for computational models in biology. By supporting SBML as an input and output format, different software tools can operate on the same representation of a model, removing chances for errors in translation and assuring a common starting point for analyses and simulations. SBML is widely supported today and continues to be evolved to support more modeling needs.
Precise, unambiguous representations of models is critical for allowing researchers to upon each others' prior work, for example by locating BioModels Database (a public database of human-curated and -annotated computational models that have been published in peer-reviewed literature) has been driving the development of supporting ontologies, standards and resources. One is MIRIAM (Minimum Information Requested in the Annotation of Models), a set of guidelines for model curation, and another is SBO (Systems Biology Ontology), an ontology of mathematical concepts used in computational models. Additional recent efforts include KiSAO (Kinetic Simulation Algorithm Ontology), TEDDY (TErminology for the Description of DYnamics) and MIASE (Minimum Information About a Simulation Experiment).
Infrastructure such as SBML and BioModels Database, standardized vocabularies such as SBO, and guidelines such as MIRIAM, greatly enhance our efforts are, at their core, a means of improving our ability to communicate our discoveries and understanding.
Adreiano M. Henney (AstraZeneca plc.)
The opportunities for Systems Biology in helping to bring innovative and effective new medicines to the market largely arise from the challenges that the pharmaceutical industry now faces. Over the last ten years the cost of developing new drugs has escalated hugely, matched by a significant decline in the number of new medicines reaching the market. Further, compared with the 1960s, the time taken to develop a new drug has doubled to approximately 12 years. The available data from across the industry shows an increased productivity in the discovery phase, but this has not been matched by success in development, with a number of well-publicised failures recently of drugs in the later stages of the development pipeline. This suggests that whilst the output of projects from discovery into development has increased, the quality of the output has not.
Pharmaceutical R&D generally has been an empirically data-driven, qualitatively oriented, activity. Whilst targets being studied in Discovery may be placed within a network, it is not necessarily clear which of the many targets is the one worth pursuing therapeutically, because of the complexity of the biological system itself, compounded by variability between individuals. Typically, each drug and target combination tends to be considered in isolation, using target-driven, high throughput, approaches that are removed from their physiological context. The key weakness of reductionism is that it cannot be used to understand or predict how the wider system will behave in a quantitative way. In other words the dynamic “context” (pathway, cell, patient or population of patients) is missing. Systems Biology is seen increasingly as an approach that can help tackle this problem, especially in articles in the technical and industry press. It has also been the subject of recent UK and European initiatives. This talk will offer a perspective on this topic, touching on opportunities and challenges that it presents, including some thoughts on industry/academia collaboration.
Hans Westerhoff (UManchester Centre for Integrative Systems Biology, Manchester Doctoral Training Centre for Systems Biology, and Netherlands Institute for Systems Biology, UK)
Now that we know what it is, it is high time to make Systems Biology work. This in itself requires a Life Sciences revolution. Both in biotechnology and in medicine, the approach that has been so successful for identifying and managing mono-factorial diseases, is failing vis-à-vis the major diseases plaguing humankind today, such as type-2 diabetes, obesity, arthritis and cancer. Making living systems do what we want them to, meets with their considerable and adverse robustness. The subjects in disease, cosmetics, and biotechnology are self-managed by highly adaptive, yet robust and often elusive networks. The diseases are Systems Biology diseases, i.e. diseases of the network rather than a molecule. To cure or manage such a network, one must address the network. And this is a different scientific paradigm.
Of course, the exclusively molecular approach to these diseases is highly successful, …… , in generating fascinating science, hence highly appreciated by us, scientists. Hence the old paradigm tends to sustain itself, but this does not mean it succeeds.
New, network-directed strategies need to be developed, relevant for biotechnology, parasitology, mammalian cell biology and physiology. They should address drug effectiveness/safety, remote or dual targets, network mapping and inadvertent adaptation. Objectives include low-cost invalidation of ‘obvious’ drug leads, comprehension of multi-factorial, molecule initiated pathologies, and flavour.
Any phenomenon of interest for Health, Disease and Wealth may depend on all macromolecules in the human host or the biotechnological production organism. Consequently, no research group, institute, company or European country is large enough to address any Life science problem completely. Consequently, the new research paradigm requires also the organization of Life Science research to be comprehensive: The information needs to be comprehensive, though linked with multiple originators. Pre-competitive consortia of industrial and academic centres should create comprehensive sets of tools, knowledge and inspiration. Teams need to be comprehensive, engaging all the required disciplines. And the individuals need to be comprehensive rather than to shy away from experimental biology or mathematics. The Manchester Doctoral Training Centre for Systems Biology introduces new ways of training both Ph D students and Industry staff in multi-, inter- and trans- disciplinary Systems Biology.
Joseph Lehár & Alexis Borisy (CombinatoRx Incorporated, Cambridge MA USA.)
Systematic testing of chemical combinations in cell-based disease models can yield novel information on how proteins interact in a biological system, and thus can make important contributions to biological models of those diseases. Such combination screens can also preferentially discover synergies with beneficial therapeutic selectivity, especially when used in high-order mixtures of more than two agents. We will discuss numerical simulations and experimental results which establish the efficacy and selectivity of synergistic combinations in complex biological systems. These studies demonstrate the value obtainable from combination chemical genetics, and reinforce the growing realization that the most useful paradigm for a drug target is no longer a single molecule in a relevant pathway, but instead the set of targets that can cooperate to produce a therapeutic response with reduced side effects.
Douglas Kell, (School of Chemistry and Manchester Interdisciplinary Biocentre)
131 Princess St, Manchester M1 7DN, UK. dbk @ manchester.ac.uk http://mcisb.org/ http://dbkgroup.org/ http://mib.ac.uk
Systems biology contrasts with molecular biology in having its focus less on biological molecules than on the interactions between them. It also involves an iterative cycle of mathematical modelling and experimental testing of those models, and an emphasis on technology development.
At MCISB, we are largely taking a ‘bottom-up’ approach similar to that of the ‘silicon cell’ , starting with purified recombinant proteins whose kinetics are analysed in detail and leading to coupled ODE models that are compared to experimental measurements. Our major set-piece activity is to make and test a full kinetic model of the metabolic network of baker’s yeast. Recent progress in this organism includes the genome-wide analysis of growth control and the production of a full suite of omics data  and the identification of high-flux-control genes . We have also developed novel numerical methods for kinetic model construction , parameter estimation , sensitivity analysis  and mode of action analyses .
Integrative information management is crucial to systems biology, and this requires the use of data standards such as SBML that enable both model exchange and effective semantic reasoning [8; 9]. We have taken the view that only loosely coupled, distributed systems are likely to be efficacious [10-12], and these have many desirable properties allowing the integration of diverse data held in different locations. This makes the problem one of designing suitable systems biology workflows, for which we are using the Taverna environment [13; 14].
Pharmacological intervention for disease amelioration represents a classical problem of systems biology, since nowadays most drug leads are isolated using molecular targets. However, 90% of drug leads that make it as far as being tested in man fail to become marketed products, a phenomenon referred to as ‘attrition’. The costs of attrition to the pharmaceutical industry run into billions of dollars annually. The two chief causes of attrition are toxicity and lack of efficacy. The former is caused mainly by off-target effects, sometimes linked to the concentrative accumulation of drug in unexpected places, while the latter often reflects a drug’s failure to reach its molecular target. In this respect, the blood-brain barrier represents a particularly acute difficulty. At all events, the organismal transport and cellular uptake of drugs are of huge importance in systems pharmacology.
In an attempt to rationalise elements of ‘drug-like’ properties, Lipinski developed his famous ‘Rule of 5’ . This states that poor absorption or permeation of a molecule is more likely when its H-bond donors > 5, its H-bond acceptors > 10, its MW > 500 and its calculated Log P (CLogP) > 5. (Log P is the octanol-water partition coefficient.) Lipinski’s rules, while empirical, are biophysically based, and have been hugely influential. However, the rules assume implicitly that drugs enter cells only by passive diffusion across the cell membrane, and indeed Lipinski explicitly excluded their application to the purportedly few known cases in which drugs ‘hitch-hiked’ into cells via carriers.
A corpus of evidence and arguments [16; 17] suggest that in fact most drugs do not mainly enter cells by passive diffusion but that they do so by using ‘natural’ transporter molecules, and our job is then to work out which ones. The problem thus moves from being one of biophysics to one of systems biology in which mechanistic models are required. This has considerable implications for the design of safe and efficacious drugs, and underscores the need to acquire a (distributed) digital model of human metabolism that may be integrated with more physiological-level models and most importantly will be open access and thereby accessible to all .
-  Teusink, B., Passarge, J., Reijenga, C. A., Esgalhado, E., van der Weijden, C. C., Schepper, M., Walsh, M. C., Bakker, B. M., van Dam, K., Westerhoff, H. V. & Snoep, J. L. (2000). Can yeast glycolysis be understood in terms of in vitro kinetics of the constituent enzymes? Testing biochemistry. Eur J Biochem 267, 5313-29.
-  Castrillo, J. I., Zeef, L. A., Hoyle, D. C., Zhang, N., Hayes, A., Gardner, D. C. J., Cornell, M. J., Petty, J., Hakes, L., Wardleworth, L., Rash, B., Brown, M., Dunn, W. B., Broadhurst, D., O'Donoghue, K., Hester, S. S., Dunkley, T. P. J., Hart, S. R., Swainston, N., Li, P., Gaskell, S. J., Paton, N. W., Lilley, K. S., Kell, D. B. & Oliver, S. G. (2007). Growth control of the eukaryote cell: a systems biology study in yeast. J. Biol 6, 4.
-  Delneri, D., Hoyle, D. C., Gkargkas, K., Cross, E. J., Rash, B., Zeef, L., Leong, H. S., Davey, H. M., Hayes, A., Kell, D. B., Griffith, G. W. & Oliver, S. G. (2008). Identification and characterization of high-flux-control genes of yeast through competition analyses in continuous cultures. Nat Genet 40, 113-7.
-  Smallbone, K., Simeonidis, E., Broomhead, D. S. & Kell, D. B. (2007). Something from nothing: bridging the gap between constraint-based and kinetic modelling. FEBS J 274, 5576-5585.
-  Wilkinson, S. J., Benson, N. & Kell, D. B. (2008). Proximate parameter tuning for biochemical networks with uncertain kinetic parameters. Mol Biosyst 4, 74-97.
-  Lüdtke, N., Panzeri, S., Brown, M., Broomhead, D. S., Knowles, J., Montemurro, M. A. & Kell, D. B. (2008). Information-theoretic Sensitivity Analysis: a general method for credit assignment in complex networks J Roy Soc Interface 5, 223-235.
-  Jayawardhana, B., Kell, D. B. & Rattray, M. (2008). Bayesian inference of perturbations in metabolic pathways via Markov Chain Monte Carlo. Bioinformatics, submitted.
-  Ananiadou, S., Kell, D. B. & Tsujii, J.-i. (2006). Text Mining and its potential applications in Systems Biology. Trends Biotechnol 24, 571-579.
-  Kell, D. B. & Mendes, P. (2008). The markup is the model: reasoning about systems biology models in the Semantic Web era. J Theoret Biol, in press.
-  Kell, D. B. (2006). Metabolomics, modelling and machine learning in systems biology: towards an understanding of the languages of cells. The 2005 Theodor Bücher lecture. FEBS J 273, 873-894.
-  Kell, D. B. (2006). Systems biology, metabolic modelling and metabolomics in drug discovery and development. Drug Disc Today 11, 1085-1092.
-  Kell, D. B. & Paton, N. W. (2007). Loosely coupled bioinformatic workflows for systems biology. In US-EU Workshop on infrastructure needs for systems biology (ed. M. Cassman and S. Brunak), pp. in press.
-  Oinn, T., Li, P., Kell, D. B., Goble, C., Goderis, A., Greenwood, M., Hull, D., Stevens, R., Turi, D. & Zhao, J. (2007). Taverna / myGrid: aligning a workflow system with the life sciences community. In Workflows for e-Science: scientific workflows for Grids (ed. I. J. Taylor, E. Deelman, D. B. Gannon and M. Shields), pp. 300-319. Springer, Guildford.
-  Li, P., Oinn, T., Stoiland, S. & Kell, D. B. (2008). Automated manipulation of systems biology models using libSBML within Taverna workflows. Bioinformatics 24, 287-289.
-  Lipinski, C. A., Lombardo, F., Dominy, B. W. & Feeney, P. J. (1997). Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev 23, 3-25.
-  Sai, Y. & Tsuji, A. (2004). Transporter-mediated drug delivery: recent progress and experimental approaches. Drug Discov Today 9, 712-20.
-  Dobson, P. D. & Kell, D. B. (2008). Carrier-mediated cellular uptake of pharmaceutical drugs: an exception or the rule? Nat Rev Drug Discov, in press.
-  Kell, D. B. (2007). The virtual human: towards a global systems biology of multiscale, distributed biochemical network models. IUBMB Life 59, 689-95.
A quantitative understanding of dynamic cellular processes during detoxification in human hepatocytes (Research network within the German HepatoSys Project)
Matthias Reuss (Istitute of Biochemcial Engineering and Centre Systems Biology, University Stutgart)
Allmandring 31, D-70569 Stuttgart, Germany. E-Mail: reuss @ ibvt.uni-stuttgart.de
The contribution aims at introducing the German government funding initiative on Systems Biology of Hepatocytes (HepatoSys), thereby focussing at the more detailed description of one of the networks within the initiative coordinated by the University Stuttgart. The 12 research groups within this network are addressing the issue of detoxification processes in human hepatocytes. The contribution of the individual partners are outlined in the introduction of the lecture. The second part of the contribution focus at more specific results on dynamic modeling of the detoxification system. The detoxification metabolism shows a high inter-individual variability in the enzyme expression level, especially in the phase I catalyzing cytochrome P450 monooxygenases (CYP). This is caused by individual food and drug treatment, sex, age, diseases, or due to the polymorphism resulting in phenotype plasticity. However, the detoxification functionality has to be maintained against these external and internal perturbations, characterizing its robustness. Based on different mathematical models for structure and dynamics of the detoxification system the important issues of structural and dynamic robustness are discussed. Superimposed to the metabolism responsible for the detoxification we also tackle the complex phenomena related to the genetic regulation of the expression of the various enzymes. Based upon time series of transcript data a Boolean/Probabilistic Boolean framework is presented to reconstruct the regulatory networks governing the activity of a specific CYP in response to a specific drug. The final part of the lecture introduces and discuss a new approach of instationary 13C flux analysis for quantitatively describing the metabolic traffic within the central metabolism under different physiological conditions. This flux analysis provide a convinient basis for quantifying cell physiology in terms of engagement of metabolic pathways in overall cellular processes and also in context with detoxification.
Integrated Data Analysis of Host and Pathogen Genomes – Mining for Antimicrobial Mechanisms and Targets
Chris Rawling (Rothamsted Research)
The search for novel molecular mechanisms driving virulence in pathogens and disease resistance in their hosts is as important to the sustainability of food (and energy) crop production as it is for the maintenance of human and animal health. Understanding these mechanisms will help the identification of new antimicrobial targets and enable the identification of new chemical entities for use in disease control.
One of the major challenges facing systems biologists working in the area of host-pathogen interactions is to integrate all the genomic and functional information that is available from the wide range of internet resources that are useful in this task. The second major challenge is that of identifying high quality gene function information from the scientific literature that can be used as reliable evidence in the assessment of hypothetical new pathways or molecular interactions that might be of interest. These two problems – data integration and knowledge extraction from text are generic requirements for many systems biology projects and solutions to these problems would be of value in pharmaceutical, biotechnology and crop protection research.
At Rothamsted Research we have developed a data integration framework (ONDEX1) that brings together features of semantic data integration, text mining and graph based data analysis and visualization. We have shown that this framework can be used for a broad range bioinformatics problems, such as genome annotation, supporting database curation and the analysis of microarray experiments in the context of metabolic networks and gene regulatory networks. In the next phase of development, ONDEX will be extended in collaboration with the Universities of Manchester and Newcastle2 to create an advanced data integration platform for systems biology.
One area of research that will benefit from this project will be the search for new mechanisms of action in crop fungal pathogens. We will be using the ONDEX system to support curation of the pathogen-host interaction database (PHI-base)3 using text mining and searching for novel molecular interactions involved in fungal pathogenesis using integrated comparative analysis of fungal genomes and fungal transcriptome and metabolome datasets that will be explored in the context of biochemical pathway and other information. PHI-base is a high quality reference database of experimentally verified pathogenicity, virulence and effector genes from bacterial, fungal and Oomycete pathogens of plant, human, animal, insect and other hosts4. It is expected that the approaches developed in this research will have broad application in other areas of host-pathogen interaction research.
Köhler, J., Baumbach, J., Taubert, J., Specht, M., Skusa, A., Rueegg, A., Rawlings, C., Verrier, P. and Philippi, S. (2006) Graph-based analysis and visualization of experimental results with ONDEX. Bioinformatics 22(11):1383-90.
ONDEX Website: http://ondex.sourceforge.net/
ONDEX will be developed with support from the UK Biological and Biotechnology Research Council (BBSRC) under the Systems Approaches to Biological Research initiative.
Winnenburg, R,. Urban, M., Beacham, A., Baldwin, T. K., Holland, S., Lindeberg, M., Hansen, H., Rawlings, C., Hammond-Kosack K. E. and Koehler, J. (2007) PHI-base update: additions to the pathogen-host interactions database Nucleic Acids Research, In Press. doi:10.1093/nar/gkm858
PHI-base Website: http://www.phibase.org/
Baldwin, T. K., Winnenburg, R., Urban, M., Rawlings, C., Köhler, J. and Hammond-Kosack, K.E. (2006) The Pathogen Hosts-interaction database (Phi-Base) Provides Insights into Generic and Novel Themes of Pathogenicity. Mol. Plant. Microbe Interact, 19(12):1451-1462
In spite of an epidemic in obesity and diabetes, five decades of drug discovery efforts have led to a small number of drug classes that do not meet the current medical need. The major predisposing factor for diabetes and obesity is insulin resistance. To discover new drugs to treat these diseases, we need a holistic way to understand how to modulate insulin resistance and complete data sets from which to extract this insight. Here we will discuss the first global analysis of signal transduction in insulin-sensitive and insulin-resistant cells targeted to define a compelling new class of drug targets. Primary biological data will be collected using time-resolved phosphoproteomics and gene expression, chemical genetics and reverse engineering of transcription factor binding events. These data will be used to construct mathematical models, and a control theoretic analysis will then allow the determination of targets to restore insulin sensitivity. This systems biology approach is unique in that it queries properties of the network as a whole rather than of separate parts. As a result, we expect to identify potent, non-obvious points to modulate this process. Although our focus is on insulin resistance pathways, this approach is extendable and can be used to identify novel targets to modulate any complex biological event.
Pierre De Meyts (Novo Nordisk, Denmark)
The two major forms of diabetes mellitus, type 1 and type 2, are both complex diseases of poorly understood pathogenesis. The poor understanding of molecular mechanisms underlying pathogenesis results in a penury of effective drug therapies or prevention strategies. The pathogenesis of type 2 diabetes involves alterations in insulin production and secretion by the pancreatic beta cells, as well as disturbances in the target cells sensitivity to insulin action. Reductionist approaches have failed to unravel the complex interactions between transcriptional networks that regulate beta cell development and differentiated function, and the signal transduction mechanisms in both beta cells and insulin target tissues, the combined disturbance of which results in impaired metabolic homeostasis. Candidate gene approaches have failed to detect major risk-associated genes. A systems biology approach is clearly warranted. All gene mutations associated with the dominant MODY form of type 2 diabetes have turned out to be genes affecting beta cell differentiation and function. The dogma that insulin resistance is central to the pathogenesis of type 2 diabetes is being seriously challenged as well by the results of recent genome-wide scans. These suggest that (like in other complex diseases), unfavorable combinations of polymorphisms in common alleles, each of which alone make small contributions, may result in disease. Interestingly, most of the newly unraveled SNPs affect genes involved in beta cell function. These new diabetogenes will likely reveal novel and so far unsuspected nodes in pathways important for building a model of pathogenesis. Building a systems biology of type 2 diabetes will also require to integrate a.o. epigenetic factors, metabolomic information, tissue to tissue communication including brain to peripheral target cells, and the newly uncovered relationships between our metabolism and our “microbiome”.
Ref.: De Meyts, P. (1993) The diabetogenes concept of NIDDM. Adv. Exp. Med. Biol., 334:89-100.
Sang Yup Lee (KAIST, Korea)
Biological systems are robust to genetic and environmental changes at all levels of organization. Previous studies on single gene or multi gene knock-outs revealed that organisms possess redundant or alternative pathways that can overcome those genes knocked-out. Thus, the number of genes that are lethal to the cell was found to be much less than expected. In this regard, the reaction-centric gene deletion study has a limitation in understanding the metabolic robustness. I will present the results of our study on the use of flux-sum, which is the summation of all incoming or outgoing fluxes around a particular metabolite under pseudo-steady state conditions, as a good conserved property for elucidating such robustness of organism from the metabolite point of view. The essential metabolites were able to maintain a steady flux-sum even against severe perturbation by actively redistributing the relevant fluxes. Disrupting the flux-sum maintenance was found to suppress cell growth. We applied this finding in identifying the drug target genes using Vibrio as an example. Detailed results will be presented. [This work was supported by Korean Systems Biology Research Program of the Ministry of Science and Technology through the KOSEF. Further support by LG Chem Chair Professorship is appreciated.]
Igor Goryanin (University of Edinburgh, UK)
The activities of the Centre for Systems Biology at Edinburgh are outlined. Then the presentation describes the process for the reconstruction of a high quality human metabolic network from the genome information the existing problems in the reconstruction.
The reconstructed metabolic networks provided a unified platform to integrate all the biological and medical information on genes, proteins, metabolites, disease, drugs and drug targets for a system level study of the relationship of metabolism and disease. Furthermore, the complex network organization structure revealed by structural analysis requires us to develop a system-oriented drug design strategy.
System kinetic modelling approach and its applications for drug R&D, Synthetic biology and Bioenergy are discussed.
Johnjoe McFadden (University of Surrey, UK)
An impediment to the rational development of novel drugs against tuberculosis (TB) is a general paucity of knowledge concerning the metabolism of Mycobacterium tuberculosis, particularly during infection. A particular problem is presented by the state of persistence when the organism grows very slowly and is relatively resistant to most drugs. We have applied constraints-based modeling to investigating microbial metabolism of M. tuberculosis in order to identify new drug targets. GSMN-TB, a genome-scale metabolic model of M. tuberculosis, was constructed. The model was calibrated by growing the pathogen in continuous culture and steady state growth parameters were measured. Flux balance analysis (FBA) was used to calculate substrate consumption rates, which were shown to correspond closely to experimentally-determined values. Predictions of gene essentiality were also made by FBA simulation and were compared with global mutagenesis data for in vitro-grown M. tuberculosis. A prediction accuracy of 78% was obtained. Known drug targets were predicted to be essential by the model. The model demonstrated a potential role for the enzyme isocitrate lyase during the slow growth of mycobacteria and this hypothesis was experimentally verified. An interactive web-based version of the model is available. The model thereby provides a means to examine the metabolic flexibility of bacterium, predict the phenotype of mutants, and identify new drug targets.
D Beste, T. Hooper, G. S. Stewart, B. Bonde, C. Avignone-Rossa, M. Bushell, P. R. Wheeler, S. Klamt, A. M. Kierzek, and J. J. McFadden. GSMN-TB: a web-based genome scale network model of Mycobacterium tuberculosis metabolism. Genome Biol 8:R89, 2007.
D. J. Beste, E. Laing, B. Bonde, C. Avignone-Rossa, M. E. Bushell, and J. J. McFadden. Transcriptomic analysis identifies growth rate modulation as a component of the adaptation of mycobacteria to survival inside the macrophage. J.Bacteriol. 189:3969-3976, 2007.
D. J. Beste, J. Peters, T. Hooper, C. Avignone-Rossa, M. E. Bushell, and J. McFadden. Compiling a molecular inventory for Mycobacterium bovis BCG at two growth rates: evidence for growth rate-mediated regulation of ribosome biosynthesis and lipid metabolism. J.Bacteriol. 187 (5):1677-1684, 2005.
Drug discovery usually focuses on candidate molecules that affect individual reactions with presumed essential functions in the cellular reaction network, especially in the development of diseases. Unfortunately, appropriately designed drugs often fail to show the expected biological effect, since the multitude of interactions in the biochemical reaction network buffers the individual changes or causes significant side effects.
With the increasing power of mathematical modelling of biochemical pathways on one side and of experimental means to describe their properties such as concentrations, fluxes or individual parameters appropriately on the other side, it is time to search for drug targets in biochemical networks.
We addressed this problem through a computational approach, which considers the effect of drug application within a generalized biochemical pathway and by studying the effect of changes regarding the type and strength of inhibitors on the reduction of flux. This allowed us to systematically search for the appropriate target and for type and concentration of the optimal inhibitor. To this end, the traditional concept of flux control is used. The effects of minimal perturbation, which can be quantified by flux control coefficients, are compared with effects of large changes. We propose the flux selectivity as a measure for the discrimination of the effect on different pathways. Since the calculation of the flux selectivity is based on flux control coefficients that are calculated in the non-affected state, it is also a means for predicting the inhibitor efficacy. Furthermore we will propose how to increase discriminative inhibition in the case of a parasitic disease by using multi-target drugs.
Gerber, S., et al., Drug-efficacy depends on the inhibitor type and the target position in a metabolic network—A systematic study. J. Theor. Biol. (2007), doi:10.1016/j.jtbi.2007.09.027
Multiple perturbations on cancer cell metabolic pathways as new targets for novel designed therapies
Marta Cascante* (Universitat de Barcelona-IBUB (Institute of Biomedicine University of Barcelona)
E-mail: martacascante @ ub.edu
Several techniques as DNA sequencing, expression arrays, and proteomic and metabolomic experiments have provided us a large amount of new information that cannot be easily interpreted. The integration of all these in vivo information in models is likely to be the most interesting tool to understand and to complete an overview picture of the cellular processes. Metabolic profile is the end point of the signaling events, where changes caused by diseases may be reflected. Using data from the different –omics, incubation with 13C labeled substrates and isotopomer analysis in selected metabolite pools, and appropriate software developed in our laboratory to estimate dynamic flux distribution among the metabolic network we are able to identify the main steps that control a metabolic pathway, which may be used as new therapeutical targets. We are applying this approach to understand metabolic adaptations accompanying different multifactorial diseases as cancer, diabetes and chronic obstructive pulmonary disease (COPD).Applying these strategies we identify the maintenance of pentose phosphate cycle oxidative and nonoxidative unbalance to be critical for cancer cell survival and vulnerable to chemotherapeutic intervention. Additionally, we used Metabolic Control Analysis (MCA) to identify the main enzymes controlling ribose-5-P synthesis and to plan combined target strategies. Finally, we validated the obtained strategies using specific inhibitors. This strategy results of great interest in imminent applications for the study of other multifactorial diseases. In particular, we are applying this strategy to achieve a better understanding of glucose metabolic network to design interventions at a metabolic level in diabetes and COPD. This new principle for rational drug design originates from the integrative, systems biology approach of understanding cell function and opens new ways to develop novel treatments for diseases as diabetes or COPD.
This work was supported by funds of Spanish Government and European Union FEDER SAF2005-01627, AGL2004-07579-C04-03/ALI and ISCIII-RTICC (RD06/0020/0046)); European Comission(FP6) BIOBRIDGEe LSHG-CT-2006-037939;Foundation Marató TV3-042010; and Comission d'Universitats i Recerca de la Generalitat de Catalunya (SGR00204). Coauthors in this work: G. Alcarraz, A.o Ramos-Montoya*; S. Guzman*, J. J. Centelles*, J. Roca#, V. Selivanov*#, P. de Atauri*, P. Vizan*, and S. Marin* *Universitat de Barcelona-IBUB; # IDIBAPS-Hospital Clínic
Zvi Kam (Weizmann Inst., Israel)
The availability of molecular part lists from genome sequences, and tools to relate sequences with functional domains help define cellular mechanisms in terms of molecular pathways. Potential interactions between molecules, and their experimental confirmation generate “interactome” network graphs, but cannot assign weights to the large number of possibilities or depict the small fraction of interactions relevant in a pathway. In addition, interactions between cellular components are modulated by sub-cellular compartmentalization or cell polarity, and their dynamics, typically balanced by opposite process (e.g. phosphorylation), is critical for the progression and build-up of signals affecting cellular decisions. Cascade-like pathway skeletons can be dissected by knockout/down approaches, but quantitative pathway functionality depends on many unknown and hard to measure variables.
An alternative approach considers the cell as a “black box”. Multiple perturbations can be a basis for reverse engineering algorithms, attempting to reconstruct the underlying circuitry responsible for functional outputs(1). Drugs and RNAi libraries offer such perturbations. GFP tagging serve to construct reporter cell lines enabling quantitative reading of effects of such perturbations on specific proteins within the intact cellular systems. Different cell lines, each painting specific molecules, are required to understand the complexity of cellular responses to perturbations. The knowledge accumulated by cell biologists relates cell functions to the localization and redistribution of specific proteins within sub-cellular organelles and structures. These are quantitatively readable by microscopy, and the higher the resolution the richer is the information.
We developed fast, high-resolution screening microscopes (2-6) that can image cells under large number of perturbations, and automated image analysis pipeline for quantitative multi-parametric characterization of the various aspects of cellular responses. While this technology help shrink the complexity of the problem in terms of number of relevant drugs or proteins, reverse engineering of pathways is still a challenge. Some of the experimental and theoretical aspects of this challenge will be discussed.
(1) Kam, Z. (2002) Bull. Math. Biol. 64:131-142.
(2) Liron et al. (2006) J. Microsc. 221: 145-151.
(3) Paran et al. (2006) Meth. Enzymology 414:228,247.
(4) Paran et al. (2007) J. Struc. Biol. 158: 233-243.
(5) Abu-Abied et al. (2006) Plant J. 48:367-79.
(6) Naffar-Abu-Amara et al. (2008) PLoS ONE, in print.
Shinya Kuroda (U-Tokyo/CREST, Japan)
To elucidate how epidermal growth factor (EGF) and nerve growth factor (NGF) specifically encode their distinct physical properties into transient, and transient and sustained extracellular signal-regulated kinase (ERK) activation respectively, we developed a kinetic simulation model of ERK signalling networks by constraining in silico dynamics on the basis of in vivo dynamics in PC12 cells. Measured in vivo dynamics can be consistently reproduced in dose- and temporal-dependent manners in growth factors in silico. This model allowed us to predict in silico and validate in vivo that transient ERK activation depends on rapid increases of EGF and NGF but not on their final concentrations, whereas sustained ERK activation depends on the final concentration of NGF but not on the temporal rate of increase. This system of ERK dynamics is produced by Ras and Rap1 dynamics, which capture the rapid temporal rate of growth factors and the final concentration of NGF, respectively. These results indicate that the Ras and Rap1 systems capture the temporal rate and concentration of growth factors, and encode these distinct physical properties into transient and sustained ERK activation, respectively1. In this symposium, I will discuss the temporal coding of ERK signaling network with our recent analysis.
1 Sasagawa, S., Ozaki, Y., Fujita, K. & Kuroda, S. Nat. Cell Biol. 7, 365-373 (2005).
Makoto Taiji (High-Performance Molecular Simulation Team, Computational and Experimental Systems Biology Group, Genomic Sciences Center, RIKEN) (taiji at gsc.riken.jp)
Recent rapid advancement of structural genomics enhances the possibilities of structure-based computational drug design. Currently, docking programs are usually used, which can screen millions of compounds to fits to pockets of proteins within a few days. However, their results are often unreliable and produce a lot of false positives and false negatives due to their low precision in binding free-energy estimations. To improve the precision of computational drug screening, we are developing the system using molecular dynamics (MD) simulations. Molecular Dynamics (MD) simulation is one of the most powerful methods to analyze thermodynamical or dynamical properties of biological molecules. MD simulations treat the fluctuations of atoms and the effect of solvents explicitly, which often play important role in thermodynamic quantities. Thus, screening with MD simulations proved much higher precision compared to conventional docking program in several reports. However, the high computational cost of MD simulation still limits its use in real drug designs. To solve the problem, we have developed a petaflops-scale special-purpose computer system for MD simulations, MDGRAPE-3. Using the system, we built a pipeline for high-precision drug screening by large-scale MD simulations. Our system can successfully screen thousands of compounds within a few days. Currently we are trying to enhance our method also for lead optimization and drug specificity analysis. Our approach will be effective in future developments of molecular target drugs as well as system target drugs.
Giulio Superti-Furga (Center for Molecular Medicine of the Austrian Academy of Science)
Giulio Superti-Furga, Uwe Rix, Oliver Hantschel, Nora Fernbach, Gerhard Dürnberger, Marc Brehme, Lily Remsing-Rix, Jacques Colinge, Keiryn Bennett and Tilmann Bürckstümmer CeMM, Center for Molecular Medicine of the Austrian Academy of Sciences, Vienna, Austria
Biology relies on the concerted action of a number of molecular interactions of gene products and metabolites operationally organized in so-called pathways and in yet larger molecular networks. However, current appreciation of the “wiring diagram” of these pathways is scanty. Through integrated approaches using proteomics as central “glue” it is possible to obtain physical, functional and knowledge maps of human disease pathways. We have started to map systematically the pathway around Bcr-Abl in leukemia and innate immunity pathways. Moreover, it is feasible to map active chemical compounds on the pathways by identifying the protein interactors of the immobilized compounds. The mode-of-action of drugs can be determined, linked to biological processes by positioning on molecular networks and implemented into novel therapeutic and diagnostic approaches. Protein maps will be annotated with chemical entities to a point where structure-binding relationships will be feasible at a proteome-wide scale and the function of pathways assessable using different chemical “genetic” agents. Such a “systems biology” approach and chemical/protein space databases promise to enable important synergies between different research avenues and inaugurate a truly “postgenomic” pharmacology era.
Technical challenges of chemical proteomics for stimulating medicinal chemists and enhancing signal transduction researches
Yoshiya Oda (Univ. Tokyo, Japan)
Chemical proteomics is an effective approach for drug discovery. Unlike chemical genetics, chemical proteomics directly and comprehensively identifies proteins that bind specifically to candidate compounds by means of affinity chromatographic purification using the immobilized candidate, combined with MS identification of interacting proteins. This chemical proteomics can be used for unbiased large-scale profiling of protein target selectivity; this is impossible with currently available drug screening panels. The new technique can be applied for the rapid and large-scale identification of primary targets of drug candidates and, more generally, protein-ligand pair interactions, allowing us to obtain binder fingerprints on a proteome-wide scale. Such information is potentially very useful for optimization of lead compounds. The data may also serve to define previously unknown protein functions, based on the phenotypes induced by compounds. Chemical proteomics also reveals that some chemical compounds bind many different protein kinases, which means that those compounds are low-specific kinase inhibitors and may be useful as a tool to enrich protein kinases. Thus chemical proteomics together with phosphoproteomics produces comprehensive information about cellular signal transduction related with protein phosphorylation.