Search Results

This chapter focuses on protein crystallography, a specialized branch of crystallography that investigates, by using diffraction techniques on single crystals, the three-dimensional structure of biological macromolecules. Most of them, like proteins, transfer-ribonucleic acid (t-RNA), polysaccharides, polynucleotides, and complexes of all of the above, can give crystals suitable for X-ray analysis. Since good-quality single crystals can be grown more easily from globular proteins than from any other class of macromolecules, protein structures are overrepresented in the biological data banks and the branch of crystallography currently being discussed is usually referred to as ‘protein crystallography’. At any rate, the techniques for structure solution and refinement described in this chapter apply to all cases where crystals with a high portion of unordered solvent in it can be grown.

The two methods to be described here, the Patterson method and the isomorphous replacement method, have made it possible to determine the three-dimensional structures of large biological molecules such as proteins and nucleic acids. In addition, the Patterson function is still useful for small-molecule studies if problems are encountered during the structure analysis. If a crystal structure determination proves to be difficult, the Patterson map should be determined to see if it is consistent with the proposed trial structure. The Patterson method involves a Fourier series in which only the indices (h, k, l) and the |F (hkl)|2 value of each diffracted beam are required (Patterson, 1934, 1935). These quantities can be obtained directly by experimental measurements of the directions and intensities of the Bragg reflections. The Patterson function, P(uvw), is defined in Eqn. (9.1).

The stages in a crystal structure analysis by diffraction methods are summarized in Figure 14.1 for a substance with fewer than about 1000 atoms. The principal steps are: (1) First it is necessary to obtain or grow suitable single crystals; this is sometimes a tedious and difficult process. The ideal crystal for X-ray diffraction studies is 0.2–0.3mm in diameter. Somewhat larger specimens are generally needed for neutron diffraction work. Various solvents, and perhaps several different derivatives of the compound under study, may have to be tried before suitable specimens are obtained. (2) Next it is necessary to check the crystal quality. This is usually done by finding out if the crystal diffracts X rays (or neutrons) and how well it does this. (3) If the crystal is considered suitable for investigation, its unitcell dimensions are determined. This can usually be done in 20 minutes, barring complications. The unit-cell dimensions are obtained by measurements of the locations of the diffracted beams (the reciprocal lattice) on the detecting device, these spacings being reciprocally related to the dimensions of the crystal lattice. The space group is deduced from the symmetry of, and the systematic absences in, the diffraction pattern. (4) The density of the crystal may be measured if the crystals are not sensitive to air, moisture, or temperature and can survive the process. Otherwise an estimated value (about 1.3g cm−3 if no heavy atoms are present) can be used. This will give the formula weight of the contents of the unit cell. From this it can be determined if the crystal contains the compound chosen for study, and how much solvent of crystallization is present. (5) At this point it is necessary to decide whether or not to proceed with a complete structure determination. The main question is, of course, whether the unit-cell contents are those expected. One must try to weigh properly the relevant factors, among which are: (i) Quite obviously, the intrinsic interest of the structure. (ii) Whether the diffraction pattern gives evidence of twinning, disorder, or other difficulties that will make the analysis, even if possible, at best of limited value.

The crystalline state is characterized by a high degree of internal order. There are two types of order that we will discuss here. One is chemical order, which consists of the connectivity (bond lengths and bond angles) and stoichiometry in organic and many inorganic molecules, or just stoichiometry in minerals, metals, and other such materials. Some degree of chemical ordering exists for any molecule consisting of more than one atom, and the molecular structure of chemically simple gas molecules can be determined by gaseous electron diffraction or by high-resolution infrared spectroscopy. The second type of order to be discussed is geometrical order, which is the regular arrangement of entities in space such as in cubes, cylinders, coiled coils, and many other arrangements. For a compound to be crystalline it is necessary for the geometrical order of the individual entities (which must each have the same overall conformation) to extend indefinitely (that is, apparently infinitely) in three dimensions such that a three-dimensional repeat unit can be defined from diffraction data. Single crystals of quartz, diamond, silicon, or potassium dihydrogen phosphate can be grown to be as large as six or more inches across. Imagine how many atoms or ions must be identically arranged to create such macroscopic perfection! Sometimes, however, this geometrical order does not extend very far, and microarrays of molecules or ions, while themselves ordered, are disordered with respect to each other on a macroscopic scale. In such a case the three-dimensional order does not extend far enough to give a sharp diffraction pattern. The crystal quality is then described as “poor” or the crystal is considered to be microcrystalline, as in the naturally occurring clay minerals. On the other hand, in certain solid materials the spatial extent of geometrical order may be less than three-dimensional, and this reduced order gives rise to interesting properties. For example, the geometrical order may exist only in two dimensions; this is the case for mica and graphite, which consist of planar structures with much weaker forces between the layers so that cleavage and slippage are readily observed.

When approximate positions have been determined for most, if not all, of the atoms, it is time to begin the refinement of the structure. In this process the atomic parameters are varied systematically so as to give the best possible agreement of the observed structure factor amplitudes (the experimental data) with those calculated for the proposed trial structure. Common refinement techniques involve Fourier syntheses and processes involving least-squares or maximum likelihood methods. Although they have been shown formally to be nearly equivalent—differing chiefly in the weighting attached to the experimental observations—they differ considerably in manipulative details; we shall discuss them separately here. Many successive refinement cycles are usually needed before a structure converges to the stage at which the shifts from cycle to cycle in the parameters being refined are negligible with respect to their estimated errors. When least-squares refinement is used, the equations are, as pointed out below, nonlinear in the parameters being refined, which means that the shifts calculated for these parameters are only approximate, as long as the structure is significantly different from the “correct” one. With Fourier refinement methods, the adjustments in the parameters are at best only approximate anyway; final parameter adjustments are now almost always made by least squares, at least for structures not involving macromolecules. As indicated earlier (Chapters 8 and 9, especially Figure 9.8 and the accompanying discussion), Fourier methods are commonly used to locate a portion of the structure after some of the atoms have been found—that is, after at least a partial trial structure has been identified. Initially, only one or a few atoms may have been found, or maybe an appreciable fraction of the structure is now known. Once approximate positions for at least some of the atoms in the structure are known, the phase angles can be calculated. Then an approximate electron-density map calculated with observed structure amplitudes and computed phase angles will contain a blend of the true structure (from the structure amplitudes) with the trial structure (from the calculated phases).