Speed dating for active ingredients

X-ray crystallography is an effective way of searching for new active pharmaceutical ingredients. The idea is to create crystals of the potential active substances together with the molecule they are supposed to attack, such as a particular protein.

These crystals are then illuminated with intense X-rays, for example those produced by PETRA III. As a result, the experts can precisely identify whether a drug candidate binds to the protein as hoped or not. PETRA IV has the potential to make this drug screening much faster and thus more efficient.

At the beginning of the coronavirus pandemic, researchers at DESY managed to examine 7700 crystals in a very short time, each involving a different active ingredient candidate. In the process, a number of substances were identified that were binding to a specific protein of the virus – its main protease. If a drug could be found that blocked this protease, it would knock out the virus.

In subsequent tests, one of these active substances has proved particularly promising, and it is now being studied in more detail. Later, the team also looked at other target proteins, such as cathepsin L, which is produced naturally in the human body, and came across a number of candidates that might be able to inhibit this protein, which is important for the replication of the virus. 

“We hope to increase throughput five- to tenfold, meaning the we will be able to analyse up to ten times as many proteins in the same time, to identify new active substances.”

On top of this, the X-rays from PETRA III that have been used so far have been monochromatic, i.e. with a single wavelength. PETRA IV would allow polychromatic – multicoloured – radiation to be used. This increase in radiation intensity ought to speed up the measurements considerably.

Among other things, the higher throughput could produce a solid foundation of data with which to develop powerful artificial intelligence algorithms. The “AlphaFold” algorithm, for instance, is already capable of calculating the spatial shape of many proteins. Eventually, similar programmes, based on the measurement data from PETRA IV, for example, could predict which active substances will bind particularly effectively to a protein – a potential paradigm shift for drug screening.