Part One: Approaches to Finding Good Hit Compounds 'The Old and the New'

Written by SelectScience Guest Editor, Dr Peter Simpson, AstraZeneca.

In this first part of this article, Peter Simpson summarizes the limits of high throughput screening for lead identification and the possible solutions to these failures.

Peter Simpson, is Director of Screening Services at AstraZeneca, and recently wrote an article in Nature on the different aspects of 'lead' generation in drug discovery.

From various meta-analyses that have been published over recent years, conventional wisdom seems to have come round to the view that high throughput screening has ‘failed’ in its promise of delivering better, faster medicines by rapid identification of good chemical starting points. New (or indeed old) replacements for high throughput screening have been proposed by various opinion leaders. A substantial movement has emerged with the view that ‘phenotypic’ screening is the solution to at least some of the problems that have made it challenging to translate high throughput screening activity into successful lead molecules.

Firstly, has high throughput screening ‘failed’? Secondly, is phenotypic screening the solution? And finally, what other approaches are emerging to deliver better, faster new molecules?

Traditional high throughput screening is, largely correctly, seen as a reductionist approach to biology. It has involved isolating a single protein, either in solution or by over-expressing it in a non-biologically relevant but convenient cell type; and then running assays based on the robustness over time and high signal to background, rather than necessarily on relevance to disease biology.

Phenotypic screening
Phenotypic screening is inherently more seductive for its stronger scientific rationale. Phenotypic assays involve using a ‘disease relevant’ cell line, such as a human neuron derived from inducible pluripotent stem cells, with native expression of proteins in context with their interacting partners. The key biological outcome that is directly linked to disease pathway is then measured, to identify compounds that prevent or alter this outcome. The final step, for the most part, is working backwards to the molecular target that drives that cellular outcome.

Bio- and chemoinformatics
This is, of course, a ‘back to the future’ approach: it is a new version of the gut-bath pharmacology approaches that were already old hat by the time I studied at university. The major advance since those days has been in the sophisticated bioinformatic and chemoinformatic analyses, and post-screening target engagement, biophysical annotation and triage techniques. These increase the opportunity of identifying the compound’s true molecular target and then progressing through a more traditional lead identification project. Many phenotypic screening key opinion leaders advocate missing that step, and driving to the lead molecule, purely on the phenotypic activity. It is not yet clear how successful that approach will prove to be. A recent whitepaper discusses these challenges and use of next generation phenotypic screening.

Disease focused screening
It can be forgotten that, in some disease areas, high throughput screening has always been phenotypic screening. Tractable infectious disease targets have been difficult to identify at the molecular level, and have not often proved druggable through biochemical approaches. By far the most predominant approach to high throughput screening for many decades, for infectious disease, has been variations on whole organism survival, proliferation and death assays. As these inherently involve a disease relevant endpoint, in a biologically relevant cell type, without a single molecular target, they are the prototype of a phenotypic assay. So high throughput screening is not necessarily a different beast to phenotypic screening. For certain types of diseases, the phenotypic high throughput screening has been tried, and delivered some, though modest, success.

In part two, Dr Peter Simpson covers more on phenotypic high throughput screening and other approaches to replace or supplement the current methods and technologies.