Providing pain relief with computers

Oct 23 2008

Pain relief is big business - very big business - which is why Dr Adrian Mulholland in the School of Chemistry has just won an EPSRC Leadership Fellowship grant of nearly £1million, part of which will be used to further his research into designing drugs for pain relief. However, he never goes near a lab, doing it all from his computer.

Providing relief from pain is what every doctor wants for their patients and people have sought ways of doing this ever since they first encountered the stinging nettle - we have all tried rubbing a dock leaf on the sting in an attempt to alleviate the pain. Today pain relief comes from drugs, but the problem with many drugs is their side effects. Drugs may promise great things, but when tested on humans they are sometimes found to cause adverse reactions that outweigh the benefits. One such compound that has been hotly debated over the past ten years is THC, the active ingredient in marijuana. Unfortunately, although THC is quite effective as a pain suppressant, it also creates a whole range of side effects.

More recently, an enzyme called FAAH, found in the brain, has been identified as a target for new pain relief drugs. When you feel pain, the body releases certain chemicals which provide a degree of natural pain relief, but the effectiveness and duration of their activity is determined by how fast they are broken down. In particular, when the body senses pain the brain releases 'anandamide' (the name comes from the Sanskrit for bliss), which nullifies the pain by blocking the pain signal. However, the effect is weak and short-lived, as FAAH quickly breaks down the anandamide. It does this because, of course, it is important for us to know when we have hurt ourselves. Conversely, if we could harness that natural mechanism by blocking FAAH and preventing it from breaking down anandamide, we might be able to develop a mild but more targeted version of pain relief. This is where Mulholland's work starts. It is known what the FAAH enzyme looks like and that there is a hollow tube within it, in which the drug molecule sits. The drug works by forming a new bond to the enzyme, but for a long time the question was which way up should the drug molecule be in order to form the most effective bond? Working with biochemists in Italy and California, Mullholland resolved this question by testing various scenarios on his computer. The results of this computer model explain experimental data and should help in designing newer and more effective versions of the drug, which is now ready to enter clinical trials.

Increasingly, drugs are being designed on computers because it is just too difficult to manually handle the vast numbers of possible drugs and the many ways they could bind to proteins in our bodies. Software packages aim to predict how drugs fit into proteins, but current methods are very approximate and often unreliable. Mulholland will work on developing better methods to predict how changing the structure of a potential drug molecule may make it bind more tightly to its target, making it more effective. He will also work on methods to model how different drugs react in the body, and whether this makes them more efficient or even whether or not they are toxic. The calculations require vast amounts of computer power, recently facilitated by the installation of High Performance Computers in the University. But even with these state-of-the-art machines, the fastest calculations take a few hours and the slowest may take a week or even longer. The new funding will last for five years, helping Mulholland's research team develop new methods and new drugs designed to help us live longer and feel better as we do so.