Network Pharmacology Research Approaches for Chinese Herbal Medicines
DALE E. JOHNSONab
network pharmacology represents the integration of several multidisciplinary concepts including biochemical, bioinformatics, and systems bio- logy.1-8 Zhang et al.2 have shown the advantages of this concept in traditional Chinese medicine (TCM) where the concept of a single target is replaced by a holistic view of multi-target effects which can be used to uncover combination effects of several active components of a formulation. From a drug discovery standpoint, this also includes the understanding of the regulation of signaling pathways with multiple channels, a potential efficacy increase in drug and drug-herb combinations, and the potential reduction in side effects.2,7,8 Two main network pharmacology approaches are currently being used:
Issues in Toxicology No. 31
Computational Systems Pharmacology and Toxicology Edited by Dale E. Johnson and Rudy J. Richardson © The Royal Society of Chemistry 2017 Published by the Royal Society of Chemistry, www.rsc.org
- • by establishing a network model through the utilization of online databases and tools, which would include both disease and chemical interaction nodes. This type of network model defines predictions of mechanism of action and potential synergy between different chemical entities and provides the opportunity to revise components within a multi-chemical formulation;2’8’9
- • by combining the disease-target(s) model with high-throughput screening, targeted assays, and directed modeling to provide validation of the predictions or forecasts and also to probe different modified structures of phytochemical analogues.2’7’10’11 Key compound components of these methods include the prediction of oral bioavailability and drug-likeness’ along with projected target screening. Important rules relating to physicochemical properties and chemical structure include lipo- philicity (octanol-water partition)’ molecular weight’ molecular flexibility estimates calculated by the number of hydrogen bond donors and acceptors’ number of rotatable bonds’ and polar surface area.12 As an example’ Kumar et al.13 have described a prediction model for oral bioavailability by a support vector machine-based kernel learning approach using physicochemical properties. Possible gastrointestinal absorption problems are alerted if any two of the following conditions are satisfied: molecular weight >500; number of hydrogen-bond acceptors >10; number of hydrogen-bond donors >5; calculated logP >5.0 (if ClogP is used) or >4.15 (if MlogP is used).14
One of the difficulties in predictive drug or phytochemical research is the estimated potency of the chemical at the biological target or at multiple targets proposed. Desirable potency at the target would be a high value for pIC50’ which reduces the risk of non-specific’ off-target pharmacological effects and could allow for lower total doses. Yu et al.15 designed a set of in silico tools incorporating chemical’ genomic’ and pharmacological information. In the framework’ molecular descriptors were combined with structural and physicochemical property descriptors. Importantly for drug research’ the typical process which follows in sequence after in silico prediction includes initial target identification and validation’ assay development’ high-throughput screening’ hit identification’ lead optimization’ and selection of candidate clinical molecule(s). With TCM therapeutics’ the starting point typically follows a physiology-based concept where herbs and/or concoctions have been used for certain diseases or conditions’ and the disease phenotype and associated biological pathways are then explored with interactions with active phytochemicals. Zhang et al.2 discuss the importance of database development in these processes’ including the incorporation of several TCM databases available online (discussed below). Another aspect of TCM systems pharmacology developing currently is the definition of susceptible individuals: those most likely to respond positively or potentially negatively to a TCM concoction and/or TCM plus Western therapeutic(s). Some of these projections into personalized therapies will require an increased knowledge and incorporation of pharmacogenomics and metabolic enzyme variances due to specific polymorphisms, particularly as it relates to difference seen in various ethnic groups.