Review
In vivo drug target discovery: identifying the best targets from the genome

https://doi.org/10.1016/S0958-1669(01)00271-3Get rights and content

Abstract

A vast number of genes of unknown function threaten to clog drug discovery pipelines. To develop therapeutic products from novel genomic targets, it will be necessary to correlate biology with gene sequence information. Industrialized mouse reverse genetics is being used to determine gene function in the context of mammalian physiology and to identify the best targets for drug development.

Introduction

The biopharmaceutical industry is on a steep trajectory to modernize its drug discovery methodology in the post-genome era. Mountains of information generated by bioinformatics, gene arrays, proteomics, and single nucleotide polymorphisms (SNPs) are confusing correlative data with causal evidence—an error in logic that, if uncorrected, threatens to turn the drug discovery laboratory into a labyrinth. These technologies provide an illusion of speed, but cannot replace the requirement for in-depth analysis of biology. The next generation of successful drug discoverers will be dominated by those who understand the primacy of the gene in biology and how encoded proteins regulate physiology. They must understand which functional genomics guideposts mark the path to productive drug discovery amidst a forest of genomic phenomenology 1., 2., 3•.. Knockout mice have become the interpreters of the human genome, a standard currency of mammalian functional genomics research widely recognized as critical, if not obligatory, in the discovery of new targets for therapeutic intervention [4].

The next generation of blockbuster drugs will be focused on specific molecular targets encoded by genes of known sequence with novel, proven physiological function. Our experience in the creation and in-depth phenotypic analysis of several hundred novel gene knockouts has allowed us to formulate stringent criteria by which we select novel drug targets. First, the validated drug target must be demonstrated in vivo to be a key switch in mammalian physiology with therapeutic applications. Second, the validated drug target must be a member of a class of proteins that is amenable to biopharmaceutical discovery and development (i.e. it must be ‘drugable’). Third, the validated drug target must address an unmet medical need of large market size. Over the past decade, gene knockout technology in mouse embryonic stem (ES) cells has been applied to thousands of genes in hundreds of laboratories across the world, and has proven to be the most robust method for discovering how genes function to regulate mammalian physiology [5].

In the race to discover new targets from genomics, the ability to identify the genes responsible for human genetic disorders is often confused with the identification of high-quality targets for drug development. Over the past decade, the genes responsible for many inherited human diseases have been discovered, yet few of these genes have led to breakthrough drugs. A good example of this is the Brca1 and Brca2 genes involved in hereditary breast cancer. The discovery of these genes has provided critical insights into the mechanisms of carcinogenesis, however, neither gene is a viable target for small-molecule drug discovery. Similarly, the identification of other genes involved in monogenic or multigenic disease will not necessarily lead directly to therapies. Traditionally, targets for blockbuster drugs have been switches in mammalian physiology, which can be pharmaceutically modulated to produce a therapeutic effect. Examples include cyclooxygenase-1 and histamine H2 receptor, the targets for nonsteroidal anti-inflammatory drugs and the inhibitors of gastric acid secretion, respectively. Neither target is associated with human genetic disease, yet both targets are critical switches that can be exploited to treat disease. Mouse knockout technology is perfectly suited to identify similar physiological switches for disease treatment. In this review we discuss how mouse reverse genetics is enhancing drug development in the post-genome era.

Section snippets

Predicting drug activity with gene knockouts

Knockout mice have proven to be invaluable tools for the functional dissection of biological processes relevant to human physiology in, amongst others, the areas of immunology [6], cancer [7], neurobiology 8., 9., cardiovascular biology [10], and obesity [11]. Gene knockouts as genetic antagonists of targets offer several profound advantages in drug target discovery, and the commercial use of gene knockout technology has evolved well beyond its original applications in modeling human genetic

High-throughput mouse functional genomics

Many features make the mouse a superior model system for drug discovery. First, the mouse genome has been sequenced and may be aligned to the human. Second, well-established parallels exist between humans and mice on cellular, biochemical, and physiological levels. Third, robust ES cell technologies exist in the mouse system. On the basis of saturation gene trapping, EST (expressed sequence tag) technology databases, and genomic DNA sequence, we have estimated that there are approximately

In vivo drug target discovery through comprehensive phenotypic screening

The use of phenotypic screens to analyze drug targets is central to the drug discovery process. We have streamlined a series of sophisticated phenotypic tests that facilitate the large-scale, in vivo analysis of gene knockouts while maintaining the high degree of sensitivity that ensures detection of subtle phenotypes. Over a five-year period, we are subjecting 5000 gene knockouts to a comprehensive phenotypic analysis program called Genome5000. Our phenotyping protocol is modeled on human

Results from recent gene knockouts

The generation of mouse genetic models for in vivo target validation on a genome scale holds much promise for modern drug discovery. Examples of drug targets that have been validated using gene knockouts early in the drug development process include cathepsin K, the melanocortin-3 and -4 receptors (MC-3R and MC-4R, respectively), and acetyl-CoA carboxylase 2 (ACC2). Cathepsin K is an osteoclast-specific cysteine protease that cleaves bone matrix proteins such as type I and type II collagen,

The use of gene knockouts to identify side-effects

Mouse gene knockout technology combined with comprehensive phenotypic analysis also provides vital information and tools to enhance preclinical and clinical studies. The phenotype derived from the knockout of a specific gene provides both potential therapeutic value as well as other target-specific effects that may be anticipated for a small-molecule inhibitor of that target. For instance, a target may have a good therapeutic potential in inflammation, but might also be critical for renal

Conclusions

It is clear that in vivo target validation using gene knockouts is fast becoming the common denominator by which new targets from the genome are judged. The fact that many independent commercial and academic laboratories have successfully implemented gene knockout technology is not only indicative of the power of this approach, but also demonstrative of its robust and reproducible nature. Lexicon's Genome5000 project is poised to evaluate thousands of new candidate drug targets in vivo. The

References (24)

  • S. Fazio et al.

    Mouse models of hyperlipidemia and atherosclerosis

    Front Biosci

    (2001)
  • S.W. Robinson et al.

    Genetic models of obesity and energy balance in the mouse

    Annu Rev Genet

    (2000)
  • Cited by (50)

    • Novel Therapeutic Targets for SARS-CoV-2 and COVID-19

      2022, Textbook of SARS-CoV-2 and COVID-19: Epidemiology, Etiopathogenesis, Immunology, Clinical Manifestations, Treatment, Complications, and Preventive Measures
    • Biochemistry and biomedicine of quantum dots: from biodetection to bioimaging, drug discovery, diagnostics, and therapy

      2018, Acta Biomaterialia
      Citation Excerpt :

      It should clarify which functional genomics guideposts show the route to productive drug discovery among the sphere of genomic phenomenology [211,212]. Walke and co-workers have proved that these genes accounted for 10% of the human genome by using gene knockout technology [213]. Therefore, drug target identification is by far one of the chief challenges that the drug development faced with.

    • Drug Targets, Target Identification, Validation, and Screening

      2015, The Practice of Medicinal Chemistry: Fourth Edition
    • Choosing the target

      2013, Drug Discovery and Development: Technology in Transition
    View all citing articles on Scopus
    View full text