“Sometimes more is better!”
In its science session, the 6th European XFEL Users’ Meeting focused on applications of free-electron lasers for structural biology. On the agenda: nanocrystallography, sample injection, and new stunning mathematical methods for data analysis.
For decades, X-rays have been an indispensable tool for life sciences. Since Watson and Crick resolved the structure of DNA by means of X-ray crystallography in 1953, tens of thousands of biomolecules have been added to the list of resolved structures. With better and brighter light sources, the rate of progress has grown rapidly.
This development has led to an enormous growth in knowledge, which has not only changed cellular and molecular biology, but also affected chemistry, chemical physics, and drug discovery in profound ways.
But life is so unbelievably complex that researchers are still scratching the surface of the life sciences. The structures of many essential molecules remain a mystery, as conventional methods of crystallography cannot be applied to them.
In X-ray crystallography, you grow crystals of a sample and illuminate them with X-rays. The interaction of the X-rays with the crystals results in diffraction patterns that give insight into the structure of the sample molecules. Why crystals? Because the diffraction pattern of a single molecule would be too weak–at least in the case of conventional synchrotron radiation sources. Using crystals, the diffraction patterns of millions of ordered molecules add to a usable picture.
But crystallization is not feasible for many relevant biological substances–many molecules and molecular complexes, such as many membrane proteins and whole viruses, are thus left out. Therefore, the ultimate goal of future research is to study single biomolecules instead of large crystals. The high intensities of X-ray free-electron lasers and the high repetition rate of the European XFEL make such experimental setups possible. The necessary methods are still in development, but first results at X-ray free-electron lasers show the big progress that has been achieved recently.
A very important step towards providing improved methods for structural biology research , which are ten to a hundred times smaller than the crystals used at conventional synchrotron radiation sources. Since nanocrystals are smaller, they are easier to grow. To make use of these tiny structures, X-ray free-electron lasers provide the high intensities necessary.
Smaller crystals, higher intensity! Done?
Life is not that easy, according to Thomas Barends of the Max Planck Institute for Medical Research in Heidelberg. In his talk, the chemist presented extremely successful experiments with protein nanocrystals at the SLAC Linac Coherent Light Source (LCLS) in the USA. For the very first time using an FELs, the atomic structure of this molecule could be resolved. But the short X-ray pulse length, which is crucial for taking pictures before radiation damage commences, makes data analysis much more complicated than with more conventional approaches. Single snapshots taken with ultrashort pulses of X-rays do not contain an entire data set. As a result, many snapshots have to be composed into a complete pattern. To solve this problem, new computational methods had to be developed and successfully applied for structure determination from FEL data. The remaining problem of the method is that you need lots of crystals, Barends added. And providing vast numbers of nanocrystals is not an easy task. Take proteins from human brain cells, which are available in very small quantities only. The high repetition rate at the European XFEL will help to make the use of probes more efficient. Additionally, developing clever ways of injecting nanocrystals into the X-ray beam is of great importance.
How such sample injection can be achieved was explained by Joachim Schulz. In his talk, the leader of the “Sample Environment” group at the European XFEL presented different methods for providing fresh samples every 220 nanoseconds–the shortest time interval between two X-ray flashes at the European XFEL. Liquid jet injection provides a steady stream of particles where the nanocrystals are contained in liquids like water. Such injection has essential advantages in the case of biomolecules that require a watery environment. To avoid the disadvantage of too strong water scattering, a complementary approach is being developed that will spray single particles into the beam by means of a method called “aerodynamic focusing”. Both methods together will allow single wet and dry particles to be injected into the X-ray beam.
How to use nanocrystals that have grown in living cells was the topic of a talk by Christian Betzel from the University of Hamburg. Nanocrystals are not just curiosities for imaging. Insulin, for instance, is produced as nanocrystals in the pancreas to be slowly released afterwards. Betzel presented investigations of the structure of biomolecules that play an important role in the African sleeping sickness that causes ten thousands of deaths a year. An international team of scientists has, for the first time, crystallised a key enzyme of the pathogen for this disease. The results obtained at the X-ray free-electron laser LCLS will be published in Nature and could lead to a new treatment approach. Resolving the structure of such molecules can help to find and test new drugs. This timely breakthrough is of great importance, as progress in the development of antibiotics has slowed down, and the cost of development compared to the output of useful drugs is increasing. Therefore, new insight gained by using new light sources is highly welcome.
How to deal with the enormous amount of collected data was the subject of the talk by Abbas Ourmazd. The theoretical physicist from the University of Wisconsin focused on the data analysis of experiments that investigate shape-changing molecules, as is most often the case with biomolecules. If you have millions of snapshots, it is particularly difficult to determine which pictures show which shape in which orientation. Ourmazd and collaborators have developed sophisticated methods to solve this problem mathematically–with stunning results that are an important step towards making molecular movies in 3D. With its high repetition rate, the European XFEL will be very well prepared for this job, since all that counts is a very large number of snapshots, Ourmazd said. The more, the better. The rest is done by methods that appear like mathematical magic.
Author: Dirk Rathje
The incredibly high intensity and short pulse lengths of the European XFEL will enable us to look at biological molecules in ways never possible before. This will help us understand them in atomic detail, which is important for fundamental science as well as for applied biomedical research. (Thomas Barends, Max Planck Institute for Medical Research in Heidelberg)
Biological samples are often difficult and expensive to prepare. We'll develop and establish fast and efficient sample delivery methods that are well adapted to the special needs of the European XFEL facility. (Joachim Schulz, European XFEL)
In vivo crystallization within cells, a method producing nanosized crystalline samples of proteins, in combination with the emerging method of femtosecond nanocrystallography at FELs, will offer a novel and innovative approach to obtain structural information of proteins that cannot be investigated by conventional methods of structural biology. New and most valuable structural insights to be used for future drug discovery investigations will be obtained. (Christian Betzel, University of Hamburg)
The high repetition rate of the European XFEL will allow lots of snapshots of a given molecule or molecular dance. These can be combined by mathematical means to get the whole picture–or even a 3D movie. Sometimes more IS better! (Abbas Ourmazd, University of Wisconsin)