Embracing (w)holism for drug discovery

Systems Biology can be of great help in reducing the drawbacks of the current drug discovery process. Dr Abhay Jere, Associate Vice President, Life Science Business Unit, Persistent Systems, writes

Drug discovery is an extremely complex, expensive undertaking and currently facing multiple challenges as large number of promising candidates prove toxic or ineffective during clinical trials. Fundamentally, these failures are attributed to the poor understanding of disease processes and the biological systems which these candidates target. For complex diseases, especially like cancer, auto-immune or neurodegenerative disorders, which were initially believed to be the resultant of relatively subtle dysfunctions of multimodal cellular pathways, the drug discovery approach against them based on high affinity/high specificity compound ('one disease — one-target-one drug') paradigm (which dominated the pharmaceutical industry for decades) is proving unsuccessful. Moreover, many successful drugs are currently in the market having moderate or low selectivity, affinity and target multiple processes while the exact mechanism of action is still unclear. Hence, now there is a growing realisation that to increase the productivity of drug discovery, a more thorough understanding of the underlying molecular mechanisms associated with the disease is essential. Also, simultaneously we need to consider the complete biological context of the drug targets.

Role of systems biology

Dr. Abhay Jere

With recent advances in various 'omics' technologies (proteomics, transcriptomics, genomics and metabolomics), mathematical modeling and scientific computing, a combinatorial approach commonly referred as Systems Biology has evolved. This field has further helped in recognising the complexity and flexibility of cellular dynamics thus offering a new platform for drug discovery. systems biology is 'predictive, quantitative and dynamical' biology claiming to use a new paradigm of '(w)holism' in contrast to the traditional 'reductionistic' perspective (of studying target protein in isolation). Its systematically networks multiple and diverse biological processes in the context of complex physiological milieu and offers insight into the combined behaviour of varied molecular species. This is only achieved through the integration of experimental, mathematical and computational sciences in an iterative manner. Thus system biology is convergence of theory of systems and control with molecular and cell biology to predict a reliable computational model for the cell and an 'integrative systems physiology' model for the organism.

Systems approach has now helped realise that numbers of molecular species in cellular network because of post-translational modifications (like methylation, phosphorylation, ADP-ribosylation as well as proteolytic processing) are far greater than those anticipated by earlier genomic analysis. Another major realisation is that the key biological reactions are characterised by low affinities and even low selectivities, thus offering enormous flexibility and redundancy in cellular circuits. Hence a high specificity/high affinity compound or 'one-disease, one-target, one molecule' paradigm WILL not always be unsuccessful.

The new approach

With robotics and high-throughput screening being more accessible, a more rational approach (for drug discovery) which focuses on agents that modulate multiple targets simultaneously is gaining popularity. Now researchers are targeting cellular functions as a system rather than single protein molecule which has significantly increased the number of drugable interactomes. This approach will help introduce novel classes of multi-target drugs with less adverse events and toxicity. For instance, recently J Sun and his colleagues constructed the first Schizophrenia Molecular Network (SMN) based on protein interaction network, pathways and literature survey. They identified 24 pathways over-represented in SZGenes and studied their interactions and crosstalk. They also observed that these pathways were related to neurodevelopment, immune system, and retinoic X receptor (RXR). Thus system biology approach helped reveal that schizophrenia is a dynamic process caused by dys-regulation of the multiple pathways and this network/pathway approach could be used to identify novel candidate genes, some of which are currently being verified.

Another example is of the ubiquitous cytokine transforming growth factor-beta1 (TGF-beta1) which is one of the most potent metastatic inducers. Functional interactomic mapping using high-throughput proteomic and genomic data provided valuable insights into the regulation of tumor suppressive and metastatic attributes of TGF-beta1. These insights are of immense value in the development of effective cancer therapeutics. Moreover, TGF-beta1 interactomic nodes are also useful in discovering novel cancer biomarkers.

Further, (w)holistic approach has also helped characterise multiple complexities associated with insulin receptor substrate 1 serine/threonine phoshphorylation sites and the interactome analysis of the human TNFa/NF?B network members and the ErbB/EGF receptors. Also, its application in modern vaccine design has been highlighted by a recent publication from R Rappuoli and his group. They claim that this new approach has helped identify about 10-100 times more candidates in last one or two years.

With growing number of researchers embracing systems based multiple target drug development approach, soon we will have a large pipeline of novel class drugs against complex diseases like cancer, schizophrenia, Alzheimer, atherosclerosis and various autoimmune diseases etc.

The road ahead

Although the systems biology approach is promising, it also faces some challenges. Robust methodologies and tools still need to be developed for extracting screenable biomarkers related information from biological systems. Another challenge is the high computational cost. High performance computing, parallel processing, grid computing can help, but these technologies require specific and robust computational tools which are currently lacking. Finally, for the systems biology approach to be successful, indeed a more widespread collaboration between mathematicians, computer scientists, physicians and experimental biologist is required.

At Persistent Systems, we have a good pool of biologist/bioinformaticians and with our core strength in computation sciences and mathematics; we are well poised to undertake projects in the area of systems biology. Also, we have plans to collaborate with premier research institutions and pharmaceutical companies to support their drug discovery initiative in the near future.

(Disclaimer: The views expressed in this article are strictly the personal views of the author. The trademarks or trade names mentioned in this paper are property of their respective owners and are included for reference only and do not imply a connection or relationship between the author and these companies.)

(The author can be contacted at Abhay_Jere@persistent.co.in)