The review by Paul D. Dobson, Yogendra Patel and Douglas B. Kell deals with an interesting concept intended to improve the value of drug screening libraries and, as a consequence, the quality of hits and leads derived from it.
As the search for pharmaceutically active drugs with ever more desirable physicochemical and drug metabolism properties, yet with negligible side effects, continues inexorably, it is becoming more and more important to consider the approach as a multiobjective optimisation problem covering an enormous search space of ‘possible’ drugs [1,2]. Screening campaigns generally begin by looking for hits which, following other studies, may be promoted to lead status [3] because, according to Oprea et al. [4], ‘lead structures exhibit, on the average, less molecular complexity (less MW, less number of rings and rotatable bonds), are less hydrophobic (lower cLogP and LogD), and less druglike’ than actual drugs. The process of optimising a lead into a drug with favourable ADMET properties [5,6] almost inevitably results in more complex structures [7] and system approaches [8,9,10,11,12] that consider not only a molecular target, but also biochemical networks that may be of value.
In attempts to narrow the search space of chemically-diverse candidate compounds, cheminformatic methods can be used to constrain the compounds screened in such a way that they tend to display ‘lead-likeness’ or ‘drug-likeness’ (and even ‘CNSlikeness’).
The authors outline how present drug screening libraries are constrained by biophysical properties that predict desirable pharmacokinetics and structural descriptors of ‘drug-likeness’ or ‘lead-likeness’. This has been an approach that has been popular over the last decade or so and popularised by a number of individuals, not least Chris Lipinski in his pioneering studies. This, briefly summarised, states that poor absorption or permeation of a compound is more probable when there are more than five hydrogen-bond donors, the molecular mass is above 500 Da, the lipophilicity is high (clogP > 5) and when the sum of nitrogen and oxygen atoms is greater than 10. Other rules or filters consider generic and calculable properties, such as the number of rotatable bonds and the polar surface area or the ligand efficiency.
More recently, however, it has been suggested that, in order to enter cells, most drugs require solute carriers that normally transport the naturally-occurring intermediary metabolites and many drugs are likely to interact in a similar manner. As the information from the human metabolome continues to increase, this will allow a more comprehensive assessment of the concept of ‘metabolite-likeness’.The similarity of known drugs and library compounds to naturally-occurring metabolites (endogenites) are compared using relevant cheminformatics molecular descriptor spaces in which known drugs are more akin to such endogenites than are most library compounds.
Descriptors such as those developed by Lipinski and colleagues are essentially biophysical, rather than structural in nature, and despite their widespread use it is not altogether clear how they should be understood mechanistically, given the enormous structural diversity of both drugs and libraries.
If drugs are mainly transported by carriers, this might help to explain why general descriptors will frequently not be effective in individual cases; it also promotes the view that we need to understand the specificities for existing and candidate drugs at known drug transporters much better than we now do at a mechanistic level.
This article was the sixth most downloaded review in Drug Discovery Today in the first quarter of 2009.
by Paul D. Dobson, Yogendra Patel and Douglas B. Kell
Drug Discovery Today (2009) 14(1/2), 31-40
1 C. Lipinski and A. Hopkins. Navigating chemical space for biology and medicine. Nature (2004) 432, 855–861.
2 G.V. Paolini et al. Global mapping of pharmacological space. Nat. Biotechnol. (2006) 24, 805–815.
3 K.H. Bleicher et al. Hit and lead generation: beyond high-throughput screening. Nat. Rev. Drug Discov. (2003)2, 369–378.
4 T.I. Oprea et al. Is there a difference between leads and drugs? A historical perspective. J. Chem. Inf. Comput. Sci. (2001) 41, 1308–1315.
5 J. Hodgson. ADMET – turning chemicals into drugs. Nat. Biotechnol. (2001) 19, 722–726.
6 M.P. Gleeson. Generation of a set of simple, interpretable ADMET rules of thumb. J. Med. Chem. (2008) 5, 817–834.
7 C.A. Lipinski. Drug-like properties and the causes of poor solubility and poor permeability. J. Pharmacol. Toxicol. Methods (2000) 44, 235–249.
8 D. Noble et al. Biological simulations in drug discovery. Drug Discov. Today 4 (1999), pp. 10–16.
9 E.C. Butcher. Can cell systems biology rescue drug discovery? Nat. Rev. Drug Discov. (2005) 4,461–467.
10 C.R. Cho et al. The application of systems biology to drug discovery. Curr. Opin. Chem. Biol. (2006) 10, 294–302.
11 D.B. Kell. Systems biology, metabolic modelling and metabolomics in drug discovery and development. Drug Discov. Today (2006), 11, 1085–1092.
12 D.B. Kell. The virtual human: towards a global systems biology of multiscale, distributed biochemical network models. IUBMB Life (2007) 59, 689–695.