(This article is sponsored by The Boston Group)

The elucidation of the sequence of the human genome last year was a defining event in biology and medicine. We now posses a comprehensive list of the DNA sequences in the ~30,000 genes that define human beings. The information embedded in these genes encodes proteins, the molecules responsible for the carrying out the structural and functional activities of the cell. In addition to the vast array of information encoded in the genome, there are a number of mechanisms that increase the diversity of the protein complement of the genome, known as the proteome. The product of the DNA sequences in the genome and these protein modification strategies is a proteome of more than 100,000 distinct molecules. There now exists an unprecedented opportunity to understand the mechanistic basis of major human diseases and develop novel therapeutics to improve human health. However, given the tremendous increase in the number of potential molecular targets for therapeutic intervention, significant advances need to be made in technologies that uncover the roles that these proteins play in disease and then lead to the rapid discovery and development of drugs to treat these diseases at a scale and scope that complement the genome.

Disease Targets
Linus Pauling first advanced the notion that diseases have a molecular basis 50 years ago. He correlated physical-chemical differences found in hemoglobin chains derived from sickle-cell patients when compared with those from healthy individuals, with the observation that sickle cell disease is an inherited disorder, to conclude that the disease is caused by a change produced in a protein molecule by an allelic variant in a single gene. Advances in molecular biology, genetics and information technology over the past 25 years have led to the identification of many gene mutations that underlie inherited diseases. Included in this list are the CFTR gene and cystic fibrosis, the IT15 gene in Huntington's disease, the LDL Receptor in familial hypercholesterolemia, and the Bcr-Abl fusion gene in myeloid leukemia. The absolute correlation between the presence of these genetic variants and disease pathology has provided further support for the molecular basis of disease and resulted in a major shift in drug discovery efforts in the pharmaceutical industry from activity-based screens to molecular target-based approaches.

Target Validation
The result of the application of the genomics strategies listed above over the past few years is the discovery of thousands of genes and proteins with disease-association hypotheses. Companies and academic institutions have filed tens of thousands of patents on these disease-associated genes and proteins for use as targets to discover new drugs. Clearly this presents a tremendous opportunity to realize the true value of the genome in the development of new therapeutic agents to treat human diseases. However, given that the time required to bring a new chemical entity from the point of discovery to product launch is 10-15 years at an estimated cost of $800 million, the existence of an interesting scientific hypothesis based on mRNA expression patterns or the conservation of interesting structural motifs is not sufficient to support this level of commitment. What is needed are more physiologically relevant biological criteria linking the target or pathway to disease pathology. This process, known as target validation, involves a number of technologies employed in mammalian cell culture and animal models.

One strategy to validate targets is to over-express the target protein in cell culture or in transgenic animals and then study the phenotype. In the case of beta-amyloid and Alzheimer's disease, over-expression of human Amyloid Precursor Protein and human Presenilin-1 in mice resulted in animals with high concentrations of beta-amyloid in their brains and the histological landmarks and behavioral traits resembling those in humans with Alzheimer's disease.

Another approach to validate targets is to lower the concentration of, or remove the protein of interest in mammalian cells, or in animal models, and observe the phenotype. There are a number of methods for reducing protein levels that include antisense RNA, RNA interference and ribozymes. However, while these methods are effective in cell lines grown in culture, they are rarely applicable to mammalian models of disease due to issues of poor stability and low tissue penetration. Antibodies generated against a protein target to neutralize its activity in vivo present another opportunity to validate protein targets, although they are generally applicable only to extracellular targets and in acute disease models.

A much more definitive method of linking targets with disease phenotype in vivo is to remove the gene encoding the target protein in an animal 'knock-out' experiment. Dramatic results have been observed in a number of cases where disease association hypotheses have been confirmed. In the case of the melanocortin-4 (MC4-R) receptor, where circumstantial evidence had linked it to metabolic pathways, it was the obese phenotype observed in MC4-R targeted mice that validated it as an obesity target.

The need for target validation to support drug discovery from the genome is clear, however there are a number of shortfalls associated with the current technologies. First, the methods described above have technical limitations that prevent them from being applied to all targets emerging from the genome. For example, over-expression of proteins in cells or whole organisms is often toxic due to titration by interaction or 'squelching' of other necessary cellular proteins. Likewise, the targeted disruption of genes necessary for embryonic development often results in developmental lethality preventing functional evaluation in adult animals. And, both over-expression and knockout approaches suffer from the potential effects of compensatory changes in the regulation and expression of other proteins that could complicate the resulting phenotype. Finally, the greatest drawback associated with all of these methods is that they do not scale to the numbers of targets or to a time frame that is practical to address the thousands of potential targets that the genome has provided. They all require an empirical process of investigating the biological pathways associated with a given target and understanding its potential role in a disease process that is both expensive and time consuming.

Conclusions
The sequencing of the human genome has created a great deal of excitement due to the unprecedented number of new genes and proteins that have been uncovered. While this is very important from a scientific perspective, the real value derived from the genome is yet to be realized in the form of novel therapeutics to treat serious human diseases. To achieve this goal, significant technological progress needs to be made to industrialize the processes of target validation and drug discovery.

|