Many models which have formed the basis of novel drug design may be incorrect. This is the sobering and at the same time very important discovery made by two Leiden researchers from the Leiden/Amsterdam Center for Drug Research (LACDR). Prof. Ad IJzerman, head of the Medicinal Chemistry division, and senior researcher Dr Rob Lane worked together with colleagues Dr Ray Stevens and Dr Veli-Pekka Jaakola from the Scripps Institute in La Jolla in the United States on clarifying the crystal structure of a particular protein - the adenosine A2A receptor. This protein is the main target in the human body for caffeine and has also been linked to Parkinson's disease. The findings of the research, which was part-funded by Topinstituut Pharma, were reported last week in Science Express; the editorial board of Science had decided to bring forward publication of the articl, highlighting its importance.

Photo: Prof. Ad IJzerman
 
Epidemiological research has shown that coffee-drinkers suffer less frequently from Parkinson's disease; this was associated with a daily intake of caffeine -  in the region of 100 mg - that you might expect in a cup of espresso or cappuccino.  In addition animal tests have shown that caffeine protects against attempts to induce Parkinson's disease. This suggests that coffee, besides its invigorating effects as a stimulant, may have other beneficial actions.  There are at least two receptors which are blocked by caffeine.  One of these receptors helps us to fall asleep.  Caffeine inhibits the effect of this receptor, resulting in a 'sharper' brain, but this also causes the racing heart which some people suffer after drinking several cups of coffee. In addition to its association in Parkinson's disease, the other receptor, which IJzerman and Lane studied, plays a critical role in inflammation.

Determining the crystal structure of these types of receptors is the best way to find out how medicines work at an atomic level. 'This is something researchers all over the globe have been trying to do for decades,' IJzerman informs us.  'It's quite understandable that so much effort should be put into this, because this class of proteins forms the target for roughly half of all medicines available in pharmacies.' It was a bit like looking for a needle in a haystack.  IJzerman: 'Particularly because these proteins are found in the walls of cells, in a fat-like environment.  They are themselves fats and it's a simple fact that fats do not readily crystallise.'  

Photo: Dr Rob Lane

The researchers in La Jolla have found a clever solution for this problem.  By connecting the receptor protein to another protein which crystallises readily, they have managed to form minuscule crystals of the fusion product.  In combination with the use of very advanced equipment, this is enough to decipher the code of the architecture of the protein using very advanced equipment.  They had managed to do this previously for another receptor, and now it was the turn of an adosine receptor. IJzerman: ‘These receptors have for many years been the focus of research in the Medicinal Chemistry division of the LACDR.  So it was understandable that the American researchers should contact us.  We joined forces and were able to determine the characteristics of the receptor biochemically and pharmacologically in Leiden, while the crystallisation triams were performed on the other side of the ocean.'  

Image: The structure of the adenosine A2A receptor. Brown - receptor; red - ZM241385; green - fats; yellow - disulphide bridges.

The first crystals of a good enough quality were produced and analysed at the end of June 2008. ‘Then we realised the great surprise which will undoubtedly have enormous consequences for researchers in the pharmaceutical industry: the binding site for medicines is very different in this receptor from in the other two receptors whose crystal structure we already knew,' says Lane.  'In the adenosine A2A receptor, a small molecule with the prosaic name ZM241385 is also crystallised; this is a compound which we already knew to have good affinity with the receptor.  You could describe this molecule as a kind of super-caffeine, substances which we have already had some success with in Leiden.  The molecule fits very differently within the receptor than could have been expected based on the other crystal structures.  Researchers both in academia and in the pharmaceutical industryhave use models derived from these previous structures to desig new medicines. This new crystal structure indicates that many of these models could be incorrect. This is the sobering yet significant discovery which we have made.'