Binding promise

Rational drug design is making huge advances in the treatment of disease. Julian Hiscox and Gail Lynagh report

Since the time of Alexander Fleming's fortuitous discovery of penicillin, the search for chemotherapeutic agents against microbes and diseases such as cancer has been one of random chance and mass screening.

Plants and other natural products were the first source of medicinal substances. As the science of chemistry evolved, it became possible to isolate the active components, so that dosage could be regulated more accurately. The discovery of penicillin led to a large-scale screening effort. Thousands of soil microorganisms were grown and tested to see whether they could produce other substances that kill bacteria. Antibiotics such as streptomycin and the tetracyclines resulted from these efforts. Now, new technologies and practices are emerging that will allow the design of therapeutic strategies tailored to individual needs. The most important of these is the concept of rational drug design. Its goal is to use what is known about a disease or an infectious agent to create safer, more effective drugs.

Most drugs work by binding to a specific site called a receptor. A central problem is to find drug molecules that bind strongly to the receptor. In so-called traditional drug design, researchers synthesise potential drugs and experimentally assay them for biological activity. There may be up to 10,000 potential drug molecules in a typical application. Perhaps 1,000 of these will bind to the receptor of interest.

In rational drug design, scientists examine what happens at the molecular level when a drug binds to a receptor, gaining three-dimensional knowledge about a binding site. The aim is to develop drugs that bind to a given receptor with greater selectivity. Accurate chemical modelling helps provide a more detailed understanding of disease and reduces costs of developing new medicines.

One technique that can measure the strength of binding is surface plasmon resonance. A target such as a protein is anchored to a silicon chip and then the molecule of interest is passed over the target. A laser measures how many molecules bind to the target, and how sticky the binding is. One of us (Julian Hiscox) uses this technique to measure the binding of RNA to a virus protein. The idea is to characterise the site of interaction. Then potential drug molecules can be designed to block the interaction, and evaluated in the same system.

This type of basic research, whether it is solving the three-dimensional structure of proteins or designing therapeutic agents, provides great inroads into the treatment of disease. However, universities and government institutes lack the financial resources to go from the laboratory to the marketplace, and this is where pharmaceutical companies enter the picture. While a new drug might bring in sales of £500 million a year, the cost of bringing a drug to market is high and fraught with failure. The development of a drug from basic research to phase III trials (testing the efficacy of the drug on large numbers of patients) can take up to 12 years and cost about £250 million, and there is no guarantee of success. Rational drug and vaccine design should help reduce this cost.

The power of rational design can be illustrated in the treatment of viral and bacterial diseases. Human immunodeficiency virus (HIV) is the causative agent of acquired immune deficiency syndrome (Aids). While prevalent throughout the developed world, it is a scourge of the developing world, with 50 per cent of the population infected in some areas. Apart from the human tragedy, the economic consequences are immense, including the burden of health care and the loss in productivity.

Use of an anti-HIV agent will select for the development and growth of resistant strains unless it blocks all viral proliferation. Because resistance develops readily, and because no single pharmaceutical on the market is powerful enough to suppress HIV on its own, combinations of drugs are used to maximise potency and reduce the likelihood of resistance.

Rational drug design has been used to create chemotherapeutic agents against one of HIV's proteins - the viral protease, whose job is to cleave viral proteins. Without this protein HIV cannot function. The three-dimensional structure of the protein was determined and computer simulations were used to design molecules that could block or disrupt the active site of the protease. Four products are on the market using this approach: nelfinavir, saquinavir, ritonavir and indinavir.

Aventis Pasteur, one of the largest producers of human vaccines, has used what is called in silico antigen discovery to develop vaccines against Chlamydia pneumoniae, a bacterium that causes respiratory disease and possibly arteriosclerosis. Computers were used to analyse the germ's DNA sequence and identify proteins likely to appear on the cell's surface where they can be recognised by antibodies. Such proteins are good candidates for vaccines. More than 46 proteins have been identified and 90 patents have arisen from this approach.

Rational drug design could also help doctors to tailor treatment to an individual's needs, using knowledge of their genetic makeup. This should help to reduce harmful side-effects and slow the spread of drug-resistant pathogens.

Julian Hiscox is a lecturer in virology at the University of Reading and Gail Lynagh is a clinical data coordinator at GlaxoSmithKline.