Solvent channel

MIR), requires the introduction of new x-ray scatterers into the unit cell of the crystal. These additions should be heavy atoms (so that they make a significant contribution to the diffraction pattern) there should not be too many of them (so that their positions can be located) and they should not change the structure of the molecule or of the crystal cell—in other words, the crystals should be isomorphous. In practice, isomorphous replacement is usually done by diffusing different heavy-metal complexes into the channels of preformed protein crystals. With luck the protein molecules expose side chains in these solvent channels, such as SH groups, that are able to bind heavy metals. It is also possible to replace endogenous light metals in metal-loproteins with heavier ones, e.g., zinc by mercury or calcium by samarium. 

FIGURE 6.5 Packing and solvent channels present in Brio cartridges employed in /rPLC systems. 

The original objective in binding metal complexes to insoluble supports was undoubtedly to get over the solubility problems of homogeneous catalysts which made their separation after reaction such a problem. However, there have been other spin-off advantages. In particular the selectivity of the supported catalysts is often greater than their homogeneous counterparts because, in addition to the electronic and steric selectivity present in the free complex, further selectivity arises because the solvent channel leading up to the active site has both a size restriction and polar properties. This results in the diffusion rate of two... 

Step Time (min)a Flow Rate (mL) Solvent Channel A Solvent Channel ... 

Isomorphous replacement is now employed in the determination of the structures of biological macromolecules. These molecules crystallize with 50% or more of the crystal volume filled with solvent molecules. Murray Vernon King, working with David Harker, conceived the idea of soaking protein crystals in solutions of compounds containing a heavy atom. These heavy-atom compounds are diffused into the crystals through the solvent channels and settle on preferred sites on protein molecules. The diffraction patterns of the unperturbed crystal (described as "native ) and the heavy-atom derivative are then compared in such a way that an electron-density map for the protein results. The method of isomorphous replacement, and the manner by which it is used to derive relative phases, are described in detail in Chapter 8. 

Fig. 5. The arrangement of the electron density in a tetragonal crystal of human serum albumin. Prominent features of the molecular packing arrangement are large (90 x 90 A) solvent channels (shown in white) that pass through the crystal parallel to the crystallographic c axis. The unit cell and symmetry operations parallel to the c axis are illustrated. Reproduced with permission from Carter et al. (1989) American Association for the Advancement of Science (AAAS).

Figure 11.13 shows the percolation type configuration, in which solvent flows through a shallow bed of solids. The shallow bed is used because the bed is subjected to less compression, and solvent channeling is less likely. This extractor is more suited for plant materials that float in the solvent. Also, the raw material must have sufficient structure to allow good distribution of the solvent through the bed of solids. 

The first step is to introduce heavy atoms into the protein crystal. This is usually done by soaking the crystals in a solution containing 0.1—10 mmol 1 1 of the heavy atom compound (Hg, Pt, Au, U compounds are often used) but sometimes the macromolecule is also co-crystallized with the heavy atom compound. As discussed in Section 9.03.4, protein crystals contain large solvent channels, which allow the diffusion of small molecules within the crystal. An important caveat is that the binding of the heavy atom compound must not distort the crystal appreciably neither the overall unit cell dimensions nor the conformation of the macromolecule. If it does, the underlying assumption that we can subtract away the protein component is false. In other words, the native (no heavy atom) and derivative (with heavy atom) must be isomorphous, and the techniques are called in general isomorphous replacement. 

The selectivity of the supported catalyst is often greater than their homogeneous analogs since further selectivity is possible because of solvent channels leading up to active sites. This observation has been used often in constructing Pt-supported catalysts. The diffusion rate of a reactant to the active site of the catalyst is dependent on a) the relative molecular size and, b) the polarity of the species. A number of Pt catalysts on supports have been used and the supports are shown in Table 6.2. 

Figure 2.18 Identified inhibitors of TGFpRl identified through inhibitor hybridisation. The red region shows the selectivity pocket fragment, the hinge region binding element is black and the solvent channel unit is coloured blue.

A macromolecular crystal usually consists of between 30% and 80% solvent of crystallisation. Hence, the enzyme active site, for example, can be accessible to these solvent channels and able therefore to catalyse conversion of a reactant to product. Because of this observation alone one is able to say that the crystal structure is directly relevant in helping to determine the macromolecule s functional state. Of course, the results themselves, defining the structure, do make chemical sense. 

Figure 2.1 The basic steps of macromolecular crystal structure solving are illustrated with respect to the enzyme, PNP MW 30000x3 D. (a) A crystal of human PNP space group R32. (b) Monochromatic oscillation diffraction photograph recorded at the Daresbury SRS resolution limit of outermost diffraction spots =3 A. (c) Electron density map, calculated at 6 A resolution, viewed down the hexagonal c axis. The diameter of the central solvent channel is =130A. Six trimers are visible. (d) A portion of the 3 A electron density map with fitted molecular model. (e) The PNP trimer molecular model, (f) The PNP trimer with bound inhibitor the protein here is represented in ribbon format for a-helix and ft sheet (see chapter 3 for details of macromolecular structure). Based on Ealick et al (1990). These figures kindly supplied by Dr S. Ealick and reproduced with permission.

A protein crystal usually consists of between 30% and 80% solvent of crystallisation and as such is an unusual state of crystalline matter. In figure 2.1(c) we saw a view down the three-fold axis of the electron density distribution of a crystal of the enzyme PNP which contains nearly 80% solvent. In this view it is clear that relatively few intermole-cular contacts contribute to making up the crystal lattice. The bulk of the protein molecular surface projects into the solvent channels of the crystal. 

---