I3 Therapeutic Targetsidentification And Validation

Target discovery, which comprises identification and validation of disease-modifying targets, is an essential early step in the drug discovery pipeline. A number of approaches to target discovery have been described in recent years. These approaches and models incorporate recent technological advances, such as genomics, proteomics, small interfering RNAs (siRNAs), and mouse gene knockout models. In this section of the book, these aspects of modern drug discovery will be described in brief (Figure I.2).

The various techniques applied in target identification and validation can be grouped into two broad target discovery strategies: the "molecular" and "systems" approach. In practice, however, both are used in varying proportions within different therapeutic areas. The "systems" approach should not be confused with the recent emergence of "systems biology," which is an attempt to construct models that explain biological responses using the plethora of information being produced from the molecular sciences.

The molecular approach is focused on the cells implicated in the disease and uses clinical samples and cell models. The molecular approach has been driven by the enormous experimental successes of molecular biology, and in particular genomics. In terms of target classes, the molecular approach

Overview

k.

M

1

Disease model

¬ęTarget identification

Target validation

Drug discovery

V

V

V

1

Disease tissue expression

Path A:

Molecular approaches

Path B:

Systems approaches

Path A:

Molecular approaches

Genomics, proteomics, genetic association

Forward genetics

Genomics, proteomics, genetic association

Forward genetics

Clinical sciences

Forward genetics Reverse genetics

Animal models

Clinical sciences

Forward genetics Reverse genetics

Disease tissue expression mRNA KO, protein overexpression

Animal models:

(conditional) KO, transgenic mice

Drug discovery:

High throughput screening of compound libraries

Structure-based drug design

FIGURE I.2 Target-based drug discovery. Four step overview (top arrows) and detailed schematic outline. Target-based drug discovery may be divided into molecular- (path A) and system-based (path B) approaches. Each approach is composed of three steps: the provision of disease models/tissues (red), target identification (purple), and target validation (blue). The molecular approach (path A) comprises techniques such as genom-ics, proteomics, genetic association, and reverse genetics, whereas systems approach (path B) comprises clinical and other in vivo studies to identify targets. Target validation covers conformational experiments in cell and/or animal models. Subsequently, the drug discovery process is commenced.

is more likely to identify intracellular targets, such as regulatory, structural, and metabolic proteins, and has been most extensively deployed in oncology. In recent years, there has been a significant shift toward the molecular approach in an attempt to identify new targets through an understanding of the cellular mechanisms underlying disease phenotypes of interest.

The systems approach is geared toward target discovery through the study of diseases in whole organisms. In general, this information is derived from the clinical sciences and in vivo animal studies in physiology, pathology, and epidemiology. The systems approach has been traditionally the main target discovery strategy and this remains the case for many diseases, including obesity, arteriosclerosis, heart failure, stroke, behavioral disorders, neurodegenerative diseases, and hypertension, in which the relevant phenotype can be only detected at the organismal level. For these historical reasons, the majority of current drugs was identified through this strategy and includes those that act against both disease phenotypes and intracellular/extracellular targets. Interestingly, because many of these drugs are directed against targets which were identified from physiological studies, rather than being directly implicated in the disease mechanism, they would probably not have been identified by the molecular approach. For example, although changes in P2-adrenoreceptor expression/activity in airway smooth muscle have not been implicated in the mechanism of allergen hyperreactivity that produces airway contraction in asthma; these symptoms are commonly treated with P2-agonist.

The incidence of many chronic diseases is strongly correlated with age, and such diseases are thought to be influenced by both genomic and environmental factors. The overall contribution of genomic factors is still unknown, although it is believed that many diseases are influenced by the presence of susceptibility genes. With the exception of smoking, the role of environmental factors is controversial, although a number of studies have indicated the importance of infection/inflammation and diet in diseases such as arteriosclerosis, central nervous system (CNS) disease, and cancer.

In undertaking target discovery, one would ideally perform clinical studies and obtain cell/ tissue samples using normal and diseased human patients. In reality, this is usually unethical and/or impractical, which means we must rely on cellular and/or animal models. However, such models often suffer from a number of significant problems which make them poor predictors of human disease. In the case of cell models, the central problem lies in simulating the complexity of the in vivo biological interactions, particularly as many of these are unknown. This problem of complexity makes it increasingly difficult to predict the role of a protein as one proceeds from the cellular level to tissue and organism. In addition, the use of immortalized cell lines to overcome the problems of availability prompts the questions regarding their biochemical similarity with primary cells. To overcome the problems of complexity, we often use animal models. However, although these models may reproduce a particular disease phenotype, genomic differences (related to species and strain), and the difficulty of identifying and replicating the long-term environmental influences, imply that the underlying causes could be different.

Was this article helpful?

0 0

Post a comment