4 min readSystems Biology: An Integrative Approach to Drug Discovery
Experimental investigation is compulsory to penetrate the complexity of biological systems with large-scale coverage of molecular underpinnings. Computational analysis also plays a crucial role in system-level understanding as a vast amount of information is generated via the experimental analysis. Available knowledge aids experiment design and analysis. It is not computationally feasible to analyse data without incorporating previously accumulated knowledge.
Novel knowledge is generated by the analysis of new data in light of available knowledge as shown in Figure 1. Thus the cycle continues, with knowledge generated from efficient computational analysis coupled with available knowledge, newer experiments are designed for further understanding of biological systems.
Historical Roots of the Integrative Thinking
The deduction of transcriptional regulation of the lac operon is probably one of the first few applications of ‘system-level’ thinking of the transcription processes. The elucidation of feedback inhibition in amino acid synthesis in the early 60’s and the glycolytic pathway much before that are also considered novel approaches at system-level understanding. Advances in technology have led to the automation and miniaturisation of biological assays thus contributing to the deluge of information which fuelled the ‘omics’ revolution. However, the actual application of these molecular data types lies in its utilisation in the integrative approach.
Extensive research is required in genomics, proteomics, computational aspects such as bio-simulation and analysis software and high precision measurement techniques such as advances in high content screening to ensure commercialisation of systems biology. Figure 2 below shows the impact of technologies on system-level understanding.
Application in Drug Discovery Process
The integrative thinking concept has a wide range of applications since an understanding of the interactions among components enables useful modification of systems. One such application is the simplification of the drug discovery process as demonstrated in Figure 3.
Metabolic Pathway Elucidation
There have been several approaches to metabolic pathway elucidation. Mathematical modelling techniques, though not a new approach, has enabled researchers to identify interactions between components. Data from feedback inhibition and competitive binding studies have been analysed by mathematicians using Gauss and Stoke’s theorems. Evolution optimisation of metabolic pathways has been studied using game theory for several years now. The accurate description of the feedback mechanisms of the insulin signalling pathway is one of the examples where the accuracy of the mathematical models has been experimentally verified. The presence or absence of an interaction in a pathway can be elucidated by formulating equations for all the components and checking whether the theoretical values (concentrations) obtained hold true for experimental values (concentrations). In case of discrepancies additional interactions are taken into account and more equations are formulated until the values derived theoretically and experimentally are equal.
Disease Pathway Identification
Knowledge of the metabolic pathway enables easier identification of ‘culprit’ regulatory proteins causing disorders. Experimental data is obtained by the detection of anomalous interactions or concentration levels of metabolites in affected individuals via protein detection techniques. Regulatory characteristics of the pathway along with its multitude of components and interactions are studied. The merit of systems biology lies in studying the pathway as a part of the system and not isolating it and treating it as a separate entity.
Target Protein(s) Identification
Advances in technologies for high precision measurements, such as laser desorption ionisation, have increased the accuracy and throughput of protein structure elucidation. Identification of regulatory proteins causing metabolic disorders is essential for effectively utilising pathway information in drug discovery. A simple masking or expression regulation of a single protein usually does not affect the system on a whole as in most cases feedback mechanisms exist that nullify the effects of external factors. Also, a single protein could be involved in the regulation of more than one pathway. Identification of target proteins forms the bottleneck in the drug designing process since all the interactions have to be identified and validated before targeting the protein.
Bioactive molecule identification
Biologically active molecules are usually low molecular weight compounds that cause changes to the pathway in order to alleviate the metabolic disorder. The common mode of action of such molecules is by binding to target proteins, competing with proteins or metabolites for binding with receptors or bind to transcription factors that are involved in the expression of target proteins. High content screening (HCS) aids in observing the system level changes of potential molecules. A major drawback in the present drug discovery process is the exposure of new chemical entities to systems as late as the clinical trials stage. This results in the increase in the number of failures in clinical trials ultimately increasing R&D costs. With HCS, safety of the molecules can be assessed at every stage of the drug discovery process starting from its interaction with proteins in vitro, effect on metabolic pathways and on neighbouring cells in tissues and organs.
The high level of predictive power promised by systems biology has led to a major multi-disciplinary research effort that will enable us to understand biological processes as a system. As informatics tools become more sophisticated and automated the ability to handle huge cell data streams will increase. Data annotation will enable powerful data mining tools to visualise patterns from experimental results. Increasing R&D costs and decreased patent life of new chemical entities has forced the pharmaceutical industry to consider alternate approaches to drug discovery. Such approaches aim at decreasing discovery time and effectively utilise technological advances in related fields. Systems biology provides a common platform to integrate hypothesis-driven and discovery-driven research in life sciences.