Metabolomics and systems pharmacology: why and how to model the human metabolic network for drug discovery

Abstract

Metabolism represents the 'sharp end' of systems biology, because changes in metabolite concentrations are necessarily amplified relative to changes in the transcriptome, proteome and enzyme activities, which can be modulated by drugs. To understand such behaviour, we therefore need (and increasingly have) reliable consensus (community) models of the human metabolic network that include the important transporters. Small molecule 'drug' transporters are in fact metabolite transporters, because drugs bear structural similarities to metabolites known from the network reconstructions and from measurements of the metabolome. Recon2 represents the present state-of-the-art human metabolic network reconstruction; it can predict inter alia: (i) the effects of inborn errors of metabolism; (ii) which metabolites are exometabolites, and (iii) how metabolism varies between tissues and cellular compartments. However, even these qualitative network models are not yet complete. As our understanding improves so do we recognise more clearly the need for a systems (poly)pharmacology.

Figure 1

The change in drug discovery strategy from ‘classical’ function-first approaches (in which the assay of drug function was at the tissue or organism level), with mechanistic studies potentially coming later, to more-recent target-based approaches where initial assays usually involve assessing the interactions of drugs with specified (and often cloned, recombinant) proteins in vitro. In the latter cases, effects in vivo are assessed later, with concomitantly high levels of attrition.

Figure 2

A Mind Map summarising this paper.

Figure 3

Two views of the role of solute carriers and other transporters in cellular drug uptake. (a) A more traditional view in which all so-called ‘passive’ drug uptake occurs through any unperturbed bilayer portion of membrane that might be present. (b) A view in which the overwhelming fraction of drug is taken up via solute transporters or other carriers that are normally used for the uptake of intermediary metabolites. Noting that the protein:lipid ratio of biomembranes is typically 3:1 to 1:1 and that proteins vary in mass and density (a typical density is 1.37 g/ml [441]) as does their extension, for example, see , normal to the ca. 4.5 nm lipid bilayer region, the figure attempts to portray a section of a membrane with realistic or typical sizes and amounts of proteins and lipids. Typical protein areas when viewed normal to the membrane are 30% , membranes are rather more ‘mosaic’ than ‘fluid’ and there is some evidence that there might be no genuinely ‘free’ bulk lipids (typical phospholipid masses are ∼750 Da) in biomembranes that are uninfluenced by proteins . Also shown is a typical drug: atorvastatin (Lipitor^®) – with a molecular mass of 558.64 Da – for size comparison purposes. If proteins are modelled as cylinders, a cylinder with a diameter of 3.6 nm and a length of 6 nm has a molecular mass of ca. 50 kDa. Note of course that in a ‘static’ picture we cannot show the dynamics of either phospholipid chains (e.g. [448]) or lipid (e.g. [449–451]) or protein diffusion (e.g. [452,453]).

Figure 4

The steps in a workflow that uses constraints based on (i) metabolic network stoichiometry and chemical reaction properties (both encoded in the model) plus, and (ii) absolute (RNA-Seq) transcript expression profiles to enable the accurate modelling of pathway and exometabolite fluxes. The full strategy and results are described in .