BOX 3.1 Rational Drug Design
The traditional discovery of new drugs is an empirical process that starts with a compound of marginal biological activity. This "lead" compound either is discovered serendipitously by the random screening of a large number of compounds (often obtained from libraries of previously synthesized molecules) or is obtained by preparing analogues of a natural ligand (i.e., a small molecule such as a hormone that binds to a biomacromolecule such as an enzyme). Chemical intuition and experience as well as ease of synthesis serve to suggest other closely related molecules (analogues) for synthesis. This process is iterative and continues until a compound is discovered that not only possesses the requisite activity toward the target but also possesses minimal activity toward other biomacromolecules (i.e., it is selective and nontoxic). The compound must also have a desirable duration of action in a suitable dosage form, its synthesis must not be too costly so that its use will be cost-effective, it must be patentable, etc. This process can take many years, can cost millions of dollars, and often does not result in a marketed product. Any method that would make this process more efficient is clearly useful. Thus, chemists in the pharmaceutical industry have sought a more rational procedure for the discovery and design of new drugs.
If the three-dimensional structure of the target biomacromolecule has been determined (e.g., by using X-ray diffraction or nuclear magnetic resonance spectroscopic techniques), a technique that has been termed structure-based drug design can be used for the design of new molecules with the potential to become useful therapeutic agents. If the three-dimensional structure of the target is unavailable, then a hypothetical model is formulated with the goal of describing the molecular features required if a particular compound is to elicit a desired biological response. This model, of course, can be validated only after a number of compounds have been synthesized and tested for their biological activity so that a statistical relationship between biological activity and physical molecular properties (i.e., a quantitative structure-activity relationship, or QSAR) can be established. Nonetheless, such a model is highly useful for focusing the synthetic effort on compounds that have the greatest chance of exhibiting increased biological activity. Rational drug design is heavily dependent on computational chemistry techniques, and advances in rational drug design are tightly coupled to advances in new algorithms for computer-assisted molecular modeling.
To design a new ligand for a biomacromolecule of interest using the three-dimensional structure of the target biomacromolecule as a guide, the structure of the target must have been found with sufficient resolution to be of utility. One must then attempt to predict the bound geometry and intermolecular interactions responsible for the high binding affinity of novel potential ligands (or molecular fragments) associated with the biomacromolecular target. Computer algorithms have been developed over the past few years that aid in the identification of potential docking modes. These algorithms have also been used to identify, from three-dimensional databases, molecules that can potentially dock (and hence bind) to a biomacromolecular target. The prediction of biological activity of a potential ligand prior to synthesis represents another essential activity for the structure-based design of new drugs. This endeavor represents an enormous challenge for structure-based drug design, but some progress has been made using statistical mechanics-based free energy perturbation techniques that involve computer simulations employing molecular dynamics or Monte Carlo methods or by using QSAR methods that rely on the three-dimensional properties of the bound ligand.
In spite of the obstacles associated with employing an analytical approach to the design of new drugs, rational drug design has, nonetheless, been of enormous utility to the pharmaceutical industry. The QSAR method has played a rule in the development of a number of drugs currently undergoing clinical trials and there are marketed products for which QSAR has been instrumental. A number of potential drugs that have been discovered using structure-based drug design techniques are currently under preclinical or clinical investigation for the treatment of diseases that include cancer, AIDS, rheumatoid arthritis, psoriasis, and glaucoma.