GRENOBLE in the French Alps has become the home of a new movie industry. The producers are biologists, the stars are minute crystals of proteins, and the stage lighting will come not from floodlights but repeated flashes of X-rays. The end product will never be big at the box office, but its significance could put the biggest Hollywood blockbuster in the shade. That is because the Grenoble movie makers hope to reveal how protein molecules twist and turn as they run through the lightning-fast chemical reactions that underlie biology鈥檚 fundamental processes.
Their work is made possible by X-ray beams at the European Synchrotron Radiation Facility (ESRF), which are the most energetic, intense and brilliant in the world. Researchers liken them to searchlights, against which conventional X-rays glimmer no brighter than candles. Inside the giant doughnut-shaped building that houses the synchrotron, a high energy electron beam runs at close to the speed of light in a storage ring 844 metres in circumference, shielded by thick concrete walls. Stretching out tangentially from this ring are 29 long huts clad in hefty plates of lead. At these 鈥渂eamlines鈥� scientists harvest X-rays produced by the electrons as they pass through carefully controlled magnetic fields (see 鈥淓urope鈥檚 shining new light鈥�, 快猫短视频, 14 March 1992). The ESRF opened for business last September, funded with 拢400 million from 12 European countries.
Physics and chemistry are the sciences most commonly associated with synchrotron research, but X-rays have been used to study the shapes and atomic structures of proteins for more than half a century. When a protein crystal is bombarded with X-rays, the atoms within it scatter the rays to give a diffraction pattern of small dots which can be interpreted as a three-dimensional atomic structure. What makes the ESRF special is the brilliance of its X-ray beams 鈥� the huge numbers of photons of the same wavelength that bombard the sample each second. This is the first of a new generation of synchrotrons 鈥� similar facilities are soon to open in the US and Japan 鈥� with beams up to 10 000 times as brilliant as those from traditional sources. This opens up all sorts of possibilities for biologists. For example, using short, intense bursts of X-rays they will build up a moving picture that allows them to see how an experimental drug interacts with a protein receptor from a cell surface. Chemists could then use this information to fine-tune drugs.
Advertisement
The intense X-ray beams also extend the range of proteins that scientists can study. Many proteins, if they crystallise at all, form tiny crystals no more than 50 micrometres in diameter, which is too small to be detected by conventional X-ray beams. 鈥淲e will be able to solve new biomolecular structures from microcrystals of sizes only 5 to 10 micrometres,鈥� says Bjarne Rasmussen, who is in charge of biological diffraction experiments on the High Brilliance beamline, the most intense of the ESRF鈥檚 single-frequency X-ray sources.
String of amino acids
Proteins consist of amino acids strung together like pearls on a necklace, which then folds into a characteristic shape. The 500 or so proteins whose atomic structures have been uncovered to date are built up from no more than a couple of hundred amino acid units. But some of the most important proteins contain several thousand amino acid units, and so produce diffraction patterns made up of a huge number of densely packed spots. Conventional X-rays are not bright enough to separate these spots. But X-ray beams at the ESRF are, and scientists there plan to look at proteins with tens of thousands of atoms. Others will study the structures of muscle fibres, tendons, hair, spider鈥檚 web, silk, cellulose and even viruses. Ambitions run well beyond working out the structures of individual proteins. 鈥淚n the coming decade the interest will move to large complexes of genes and proteins,鈥� says Carl Branden, director of biological research at the ESRF. Branden and others have their sights fixed on the cell鈥檚 protein factories, the ribosomes.
When the ESRF is fully equipped in 1998, it will have 45 beamlines serving more than 100 experimental stations for 6000 hours each year. So far, nine beamlines have been brought into regular use or are being commissioned, of which three are at least partly dedicated to biological research. Biologists creating moving pictures of proteins will use the White Beam beamline, which uses a spread of wavelengths from visible light to X-rays to create the most powerful beam possible. The Microfocus beamline, which can focus X-rays onto a spot of no more than 2 micrometres square, is ideal for studying microcrystals and biological fibres.
Complicated as proteins are, there is more to them than their static structure. They move, exchange electrons and run through fast chemical changes as they react with cell membranes, genes, other proteins or drugs. Most of these changes take place within fractions of a second, and some are all over in a matter of femtoseconds (10鈭�15 seconds). This puts them well beyond conventional X-ray crystallography, which needs minutes or hours of exposure time to produce a full diffraction pattern. A few live movies of biochemical reactions have been shot at other synchrotrons with a time resolution of seconds to milliseconds. At the ESRF it will be possible to film much faster reactions.
X-ray crystallography normally uses radiation of a single wavelength, as this makes the diffraction patterns easier to interpret. The White Beam beamline is so intense that a flash lasting only 50 picoseconds can record a diffraction pattern for a protein. Repeated flashes 2.8 microseconds apart will be used to follow fast biochemical reactions. But the diffraction patterns produced by white radiation are complicated, so this method will be used mainly on proteins with known structures.
The White Beam beamline has already shown its potential. Last summer, Michael Wulff and his team created a world record by producing a diffraction pattern from a protein in turkey egg-white, using a 50-picosecond flash. 鈥淎 diffraction pattern of a similar quality would have required an exposure time of 15 minutes on a conventional X-ray tube,鈥� says Wulff.
Open and shut case
Armed with the knowledge that such short exposures are feasible, an American group led by Keith Moffat from the University of Chicago has started a study of myoglobin, the protein that stores oxygen in muscle cells. Myoglobin contains a central iron atom, which binds the oxygen molecule inside a group of atoms called the haem pocket. The question is how the oxygen gets in and out of that pocket. Researchers believe the answer lies with a 鈥渄oor鈥� that opens for a matter of nanoseconds to let the oxygen molecule pass through. 鈥淲e hope to prove whether this is true or not,鈥� says Wulff. 鈥淚f we succeed in filming such fast reactions, it would be of great importance for basic biochemistry and for drug research.鈥�
Already, key steps in slower enzyme reactions have been recorded at the White Beam beamline by Janos Hajdu from the Laboratory of Molecular Biophysics at the University of Oxford and Helen Jouve from the Institute for Structural Biology in Grenoble. They studied an enzyme called haem catalase, monitoring how it is oxidised by hydrogen peroxide and then reduced to its original state. Haem catalase is one of a group of enzymes present in all oxygen- using organisms, which decompose hydrogen peroxide to water and oxygen. If these enzymes fail, hydrogen peroxide builds up to dangerous levels, causing problems such as inflammation, ageing in cells and even genetic mutations.
Biochemists are especially interested in how these enzymes help particular reactions along. 鈥淢ost enzymes act as rate accelerators, speeding up biochemical reactions, but this group work as reaction selectors,鈥� says Hajdu. They assist their biochemical reactions with the help of a metal ion that sits in the middle of their folded protein structure. In its high oxidation form the metal ion is capable of reacting with practically anything known to organic chemistry. But access to the centre is restricted. As a result, only certain molecules 鈥� those that fit 鈥� can reach the metal centre. If the detailed structure and function of these enzymes can be discovered, it might be possible to modify them so that they catalyse a different reaction. 鈥淚t may be possible to create artificial enzymes, which can catalyse reactions that are not easily achieved in nature,鈥� says Hajdu.
Proteins hold the key to health and disease. 鈥淚f one understands the normal biosynthesis of proteins, one may be able to find a way of blocking abnormal synthesis in disease,鈥� says Ada Yonath from the Weizmann Institute in Israel and the Max-Planck Research Unit for Ribosomal Structure in Hamburg, Germany. To that end, her group is studying the ribosomes where proteins are made. It is an ambitious project, but the rewards if it succeeds could be huge. For example, many antibiotics work by binding to ribosomes and preventing synthesis of the proteins that bacteria need to survive. Armed with a detailed picture of how this happens, chemists could modify antibiotics to make them more specific or tighter binding.
With a decade of work behind them, Yonath and her team have a fairly detailed picture of how the ribosomes鈥� molecular assembly line works. Studies on other synchrotrons show a structure with two subunits that come together to manufacture a protein. The smaller unit contains a strand of RNA enmeshed with 21 proteins, while the larger contains two RNA molecules and over 36 proteins.
Protein synthesis requires two things 鈥� raw materials and instructions. The building blocks, amino acids, are carried to the ribosomes on molecules called transfer RNA. The instructions for assembling them come in the form of a genetic code delivered from the cell鈥檚 nucleus by a molecule called messenger RNA. Recent studies show that messenger RNA slots into a large gap between the ribosome鈥檚 two units. A prominent tunnel through the larger subunit appears to shield the growing chain of amino acids until the protein is big enough to fold and protect itself. Yonath and her team have just started to use the Grenoble synchrotron for their experiments. 鈥淲e expect to get significantly better data at the ESRF than at other synchrotrons,鈥� she says. 鈥淭here is little doubt that a model for the ribosome will eventually be obtained.鈥�
Meanwhile, one of the first studies at the ESRF has clarified what happens at an early stage of protein synthesis. In a test experiment in 1993, Steve Cusack鈥檚 group from the European Molecular Biology Laboratory Outstation in Grenoble showed how transfer RNA binds to specific amino acids. This is crucial for health, because there is no failsafe mechanism built into the ribosome鈥檚 protein production line. If the wrong amino acid is incorporated into a protein the result is often disease.
Each of the 20 amino acids has its own 鈥渉elper鈥� enzyme known as a synthetase to bind it to the correct transfer RNA molecule. Cusack and his researchers looked at the enzyme that works with the amino acid serine. They found that the synthetase first binds to serine and a molecule of the biochemical fuel called ATP. Then the enzyme binds to transfer RNA and uses the energy from the ATP to link the serine on to it. In March last year, the team published a picture of the binding site derived from X-ray analysis at the ESRF, which showed that the synthetase identifies the correct messenger RNA molecule by its shape.
X-ray crystallography is the only way to solve the atomic structure of viruses, which consist of a core containing genetic material and a coat usually made up of four or five types of protein. David Stuart and a group from Oxford University have already used the ESRF鈥檚 Microfocus beamline to solve the atomic structure of bovine enterovirus, which is endemic in cattle. The virus causes no obvious disease, which makes it an ideal carrier for cattle vaccines. 鈥淲ork is now under way to make a vaccine against foot-and-mouth disease,鈥� says Stuart. 鈥淭his is done by inserting a small part of foot-and-mouth disease virus which is recognised by the cow鈥檚 antibodies into the surface of the bovine enterovirus.鈥�
Cheating mutations
The cattle virus is about 30 micrometres in diameter, which is not particularly large, but Stuart鈥檚 next project is more ambitious. He hopes to carry out experiments on blue tongue virus which, at 80 nanometres across, is larger than any viral structure solved so far. This animal virus is a member of the same family as human rota virus, which causes severe diarrhoea. Once scientists know the structure of such viruses they can start to discover more about how the virus mutates and how the immune system is cheated by these mutations. This line of research could lead to new antiviral drugs and vaccines.
Work on the influenza virus bears this out. The outstanding characteristic of influenza is that each epidemic is caused by a new strain of the virus. Because the different strain is not recognised by the immune system, the infected person becomes ill, and for the same reason previous years鈥� vaccines don鈥檛 work. When the immune system finally reacts, it neutralises the virus by producing antibodies which recognise and bind to haemagglutinin, a large protein in the virus鈥檚 membrane. Mutation of this protein is the primary reason why new strains of influenza virus arise.
At the ESRF, teams led by Marcel Knossow from the Laboratory for Structural Biology in Gif-sur-Yvette near Paris and John Skehel from the National Institute for Medical Research in London are getting important insights into these processes. They have grown several crystals of haemagglutinin combined with different antibodies, and recently looked at the first of these using the High Brilliance beamline. 鈥淭his antibody recognises amino acids in haemagglutinin near the region of the protein that enables the virus to bind to a human cell,鈥� says Thiery Bizebard, one of the French team. 鈥淚t seems that the antibody prevents it from binding,鈥� he adds. Eventually this study should show precisely how influenza viruses are neutralised, which might suggest modifications in vaccine design or in vaccine strategy.
When plans for the ESRF were being drawn up in the 1970s, biological research was merely an appendix. Now it surpasses physics and chemistry. Almost one-third of the 94 experiments done last year were in the field of biology, and less than a year after opening the ESRF has already proved itself invaluable to the study of living things.