Many of the instruments that have been built to study proteins and other molecules were developed for physics, as tools to investigate atoms or energy at work on the atomic scale. These inventions have proven so useful in the study of biology that many of their designers have received Nobel Prizes for their work. One of the most important developments has been the use of X-rays to investigate the structure of biological molecules.
For several decades after the German physicist Wilhelm Röntgen (1845-1923) discovered X-rays, doctors and others used them without being quite sure what they were. Some physicists thought that X-rays were made up of streams of particles; German physicist Max Laue (1879-1960) believed they were waves. Laue thought up an experiment involving crystals that could settle the issue. No one was quite sure what crystals were, either, although a popular hypothesis suggested that they were made up of rows upon rows of atoms, all arranged in the same direction. Individual atoms could not be seen using visible light because its wavelength was much too broad—like trying to play the piano wearing a baseball glove. Hitting single notes (seeing single atoms) would require thinner fingers. Laue reasoned that the wavelength of X-rays ought to be about the same as the distances between atoms in a crystal. This meant that shining them through the crystal ought to produce clean, single "notes"—the waves would be scattered in a pattern that could be interpreted.
The experiments revealed the wavelike behavior of X-rays and showed that crystals are indeed made up of row upon row of small repeated unit cells, each of which contains an identical package turned the same way. Imagine a huge stack of identical shoeboxes, each containing a pair of the same type of shoe, all the same size. The boxes could be stacked in many different ways, but since they are all identical, the shape of the entire stack will have some regular features. This is why crystals have well-organized geometrical shapes. A structural biologist hopes to obtain a three-dimensional image of a single protein, much like trying to determine the shape of a single shoe in one of the boxes.
When X-rays are aimed through a crystal, most of the energy passes straight through. But some of the rays collide with electrons in the atoms in the sample and are diffracted, or radiated off at an angle. These waves are captured on a photographic plate or a detector.
It is hard to describe and imagine this process because an X-ray diffraction experiment does not produce an image that makes sense to the eye, but the following analogy will give a sense of the problem and the solution. Suppose that pairs of identical shoes are taken out of their boxes and stacked on row after row of shelves. Imagine that each shoe is covered with sparkling sequins. If a huge light were shone through the shelves, it would sparkle off the sequins and cast light every-where—off the walls, floor, and ceiling. If the shoes were all lined up identically—which is what happens to atoms in crystals—sequins on all the toes would reflect their light in the same direction, and so would the sequins on the sides and heels. Studying a diffraction pattern is like capturing all of the reflected light and analyzing it to determine the shape and relative positions of individual pairs of shoes.
To obtain a meaningful pattern, a molecule has to be organized in a very regular form, such as the identical unit cells of a crystal or the racks of a shoe store. Another arrangement that can be analyzed is a fiber if it is composed of a string of units that repeat over and over. DNA is such a fiber. In the late 1930s and 1940s William Astbury (1898-1961), a physicist working on biological molecules at the University of Leeds, stretched it and examined it using X-rays. The work was continued and refined in the early 1950s by Maurice Wilkins (1916-2004) and Rosalind Franklin, two physicists working on organic molecules in London. The experiments provided crucial information about the structure of DNA. They showed, for example, that it was arranged in the form of a helix; they revealed the width of the helix and the distance between individual "steps." These measurements were essential when Watson and Crick made their model of the double helix, described in the next chapter.
The first X-ray pictures of molecular structures came from salt and minerals, substances with a small number of atoms arranged in simple patterns. The British researchers Sir Lawrence Bragg (1890-1971) and his father, Sir William Bragg (18621942), shared the 1915 Nobel Prize in physics for this work. Their research laid the groundwork for the structural study of more complex molecules, which was carried out in new crystallography departments that were being set up at British universities, by young scientists who had studied in the Braggs' laboratories. One of them was Astbury. Before working on DNA, he used X-rays to examine protein fibers. His data allowed Linus Pauling (1901-94), an American chemist at the California Institute of Technology, to discover the basic secondary structures of proteins: alpha helices and beta sheets. It was a huge accomplishment that helped secure Pauling the Nobel Prize in chemistry in 1954.
John Bernal (1901-71), a member of William Bragg's laboratory in London, moved north to Cambridge University to start a crystallography unit there. One of his students, Dorothy Crowfoot Hodgkin (1910-94), became interested in the possibility of using X-rays to examine proteins when she was given a sample of the hormone molecule insulin, in crystal form. Bernal and Hodgkin took the first X-ray images of the crystal in 1934, but they had no way of interpreting their results into a model of insulin's structure. Yet, the protein crystals produced clear diffraction patterns, giving the researchers hope that they could someday be used to determine the arrangement of atoms in the molecules.
Another of Bernal's students, Max Perutz (1914-2002), had developed an interest in protein structure. Perutz had come to Cambridge in 1936 from Austria, where he had completed a degree in inorganic chemistry. He left the country when the Nazis invaded Austria and Czechoslovakia; his family became refugees. When war broke out between Britain and Germany, Perutz's nationality made him suspect, and he was briefly imprisoned. Upon his release he was put to work for the British war effort. He had done some work on ice crystals in glaciers and was assigned a crazy project to build artificial icebergs that could be used as aircraft carriers. (The plan did not work.)
Perutz had learned of the work of Pauling and other pioneers in the field during an organic chemistry course in Austria. In postwar Britain, finally free to pursue his own interests, he obtained crystals of the protein hemoglobin, which he hoped to use to obtain a structure. Bernal taught him how to handle the X-ray equipment, and Perutz began what would become a 22-year project to obtain a structural map of hemoglobin. In 1947 he was joined by a British student, John Kendrew (1917-97), who quickly became equally passionate about protein structure.
For many years it seemed as though the molecules would never give up the secrets of their architecture. Proteins are chemically very complex—with 20 amino acids, rather than a single atom like a diamond or the repeated structures of an artificial fiber. The patterns produced in X-ray experiments were so complicated that they did not seem possible to interpret. Additionally, the patterns were very weak. X-rays interact with an atom's electrons, so it is easiest to obtain diffraction patterns from heavy atoms with many electrons, such as metals. Biological molecules are composed almost entirely of light atoms with very few electrons. In addition, diffraction patterns did not capture phase, a type of information that was crucial to matching spots on the pattern to atoms in a molecule. Phase had to do with the fact that two X-ray waves, deflected by a protein crystal, might interfere with each other. Usually they cancel each other out. The X-rays that escape make the diffraction pattern. The spots on the photographic plates recorded the intensity of a wave but not its phase. It was like hearing all the notes of a piece of music played at the same time, without experiencing the sounds separated by time, or structured by a rhythm.
To solve these problems Perutz and Kendrew discovered they could soak protein crystals in heavier atoms, such as mercury, that would attach themselves to the side chains of some of the proteins. These atoms gave off strong, characteristic signals that could be used to locate and chart the positions of other atoms. They could also be used to calculate the phase, by comparing the patterns made by proteins soaked in the atoms with an image from unsoaked crystals.
Producing the first structure of a protein, myoglobin, took five years of work on the part of Kendrew and a team of women "computers" to do calculations by hand, figuring out the mathematics of 250,000 spots on a plate of film. Perutz followed a similar procedure with hemoglobin. As the project progressed, they were able to make use of early computers, which was essential because about 5 billion sums had to be calculated based on the diffraction patterns. Finally, in 1959 the groups had built three-dimensional models of the two proteins at a resolution of two angstroms (an angstrom equals one ten-billionth of a meter), insufficient to detect the lightest atoms, such as hydrogen, but clear enough to reveal the main shapes within the molecule. For their work, the two men were awarded the 1962 Nobel Prize in chemistry. That same year two more of Perutz's former students, Watson and Crick, were presented with the Nobel Prize in physiology or medicine for their work on DNA.
The Nobel ceremonies include a lecture in which the award-ees explain their work. In his presentation Perutz noted that obtaining the structure was one step toward a larger goal: explaining how the protein's building plan allowed it to transport oxygen. He said: "We may hope that the interaction, and the acid shift on which the respiratory functions of hemoglobin depend, will eventually find their explanation in terms of the structural changes of which these new results have just given us a first glimpse." To do so would probably require obtaining an even higher-resolution image of at least one of the two forms, he stated: "Due to the enormous amount of labour involved this may take some time, but not much, perhaps, compared to the 22 years needed for the initial analysis." In fact, just a year after receiving the prize, Perutz was able to report success. A clearer set of data and a new crystal revealed that some of the internal structures of the protein rearranged themselves when oxygen atoms were bound to the molecule. The two different conformations of hemoglobin were like snapshots; Perutz could now assemble them into a "film" showing how the protein's structure changed to carry out its functions.
Protein crystallography has now become the main method used to unravel the structures of proteins. What took Perutz and Kendrew more than two decades can now be done in a single afternoon—once scientists have a usable crystal—thanks to new technology and analytical software. The small X-ray machines used in laboratories in the 1950s and 1960s have been replaced by some of the world's largest machines: synchrotrons, ring-shaped tunnels that are used to smash atomic particles together. In 1969 the British researcher Ken Holmes (1934- ), who had studied with Franklin, carried out a series of experiments on a synchrotron in Hamburg, Germany, showing that the high-energy X-rays produced by these machines could be used to obtain protein structures. He discusses the motivations behind his pioneering experiments and some of the results in the sidebar.
Was this article helpful?