Active site geometry

SA DePriest, D Mayer, CB Naylor, GR Marshall. 3D-QSAR of angiotensm-convertmg enzyme and thermolysm inhibitors A comparison of CoMEA models based on deduced and experimentally determined active site geometries. I Am Chem Soc 115 5372-5384, 1993. 

The reversibility of halohydrin dehalogenase-catalyzed reactions has been used for the regioselective epoxide-opening with nonnatural nucleophiles (an example is given in Scheme 10.34) [133]. The stereoselectivity of the enzyme results in the resolution of the racemic substrate. At the same time, the regioselectivity imposed by the active site geometry yields the anti-Markovnikov product. [128]... 

In the end, what is unique about computational methods is their ability to describe transition states and intermediates. This is why the calculation of reaction mechanisms has achieved such a prominent position in quantum biochemistry. We will therefore spend a considerable amount of time to describe when improved active-site geometries can be expected to give important beneficial effects on reaction energies. In addition, we will try to describe how the non-bonded interactions between active site and surrounding protein affect relative energies. 

As briefly mentioned above, the reduced form of MMO reacts with oxygen to initiate substrate oxygenation. To further analyze the protein effects on this reaction, the dioxygen-binding step was treated with two-layer ONIOM (B3LYP Amber) [25], The overall setup was similar to the one used for evaluating active-site geometries. 

One reason for the relatively large RMS deviations, compared to the active sites of MMO and RNR, is that the active-site residues are not coordinated to the selenium (see Figure 2-8). The lack of a structural anchor leads to a relatively unstable active-site geometry. An alternative formulation is that the presence of a metal center with strong ligand interactions is one reason the active-site model works comparatively well for many metal enzymes. 

For the present reaction, the presence of surrounding protein only marginally affects the barrier (it increases by 0.7 kcal/mol). A possible reason for the small protein effects could be that in the present model, the active site is not deeply buried inside the enzyme instead it is located on the interface of two monomers. Still, addition of the protein environment had effects on the active-site geometry. The reason this does not affect the total barrier height is that when comparing transition state and reactant, the protein effect appears to be relatively constant. 

In all the studied systems addition of the surrounding protein in an ONIOM model clearly improves the calculated active-site geometries. This is clearly illustrated in Figure 2-13, which shows the root-mean-square deviation between calculated and experimental structures for four of the studied enzymes. 

There seems to be two major reasons for this improvement in calculated geometries improved hydrogen-bond networks and a better description of metal coordination. Apart from these two effects, we did not find any major changes in active-site geometry that could be attributed to the surrounding protein. 

In many cases the most interesting results of a computational study are the relative energies of transition states and intermediates because they determine the reaction mechanism. In this section we will try to outline when improved active-site geometries can be expected to have important effects on relative energies. 

Figure 7. (a) Stereo view of comparison of the main chain of the X-ray structure of the HIV-1 protease complex with compound 2 (red) with the main chains of the minimized complex (yellow) and a 20ps average dynamical structure of the same complex of HIV-1 protease (green), (b) Stereoview of the active-site geometry of the crystal structure (in half-bond color) of the HIV-1 protease complexed with the compound 2 (with the indole and phenyl groups shown in red) as revealed by X-ray crystallography. 

The Franck-Condon principle states that there must be no movement of nuclei during an electronic transition therefore, the geometry of the species before and after electron transfer must be unchanged. Consequently, the active site geometry of a redox metalloenzyme must approach that of the appropriate transition state for the electronic transfer. Every known copper enzyme has multiple possible copper oxidation states at its active site, and these are necessary for the enzyme s function. 

DePriest, S.A., Mayer, D., Naylor, C.B., Marshall, G.R. 3D-QSAR of angiotensin-converting enzyme and thermo-lysin inhibitors a comparison of CoMFA models based on deduced and experimentally determned active site geometries./. Am. Chem. Soc. 1993, 3 35, 5372-5384. 

Lewis, D. V. F. (1986) Physical methods in the study of the active site geometry of cytochrome P450. Drug Metab. Rev. 17, 1-66. 

Lewis, D. F. and Moereels, H. (1992) The sequence homologies of cytochromes P-450 and active-site geometries../. Comput. Aid. Molec. Design 6, 235-252. 

Distortment of active site geometry by solvent penetration... 

Roberts et al. reported a 27 kDa monomeric carbonic anhydrase, TWCAl, from the marine diatom Thalassiosira weissflogii (221). X-ray absorption spectroscopy indicated that the catalytic zinc is coordinated by three histidines and a water molecule, similar to the active sites of the a- and y-CAs (222). Also, the active site geometry is similar to that of a-CAs. Based on these results the catalytic mechanism is expected to be similar to that of the -class carbonic anhydrases. Tripp et al. (223) proposed that this TWCAl is the prototype of a fourth class carbonic anhydrase designated as 8-class CAs. In the... 

Metalloenzymes pose a particular problem to both experimentalists and modelers. Crystal structures of metalloenzymes typically reveal only one state of the active site and the state obtained frequently depends on the crystallization conditions. In some cases, states probably not relevant to any aspect of the mechanism have been obtained, and in many cases it may not be possible to obtain states of interest, simply because they are too reactive. This is where molecular modeling can make a unique contribution and a recent study of urease provides a good example of what can be achieved119 1. A molecular mechanics study of urease as crystallized revealed that a water molecule was probably missing from the refined crystal structure. A conformational search of the active site geometry with the natural substrate, urea, bound led to the determination of a consensus binding model[I91]. Clearly, the urea complex cannot be crystallized because of the rate at which the urea is broken down to ammonia and, therefore, modeling approaches such as this represent a real contribution to the study of metalloenzymes. 

The enzyme can also catalyze the transfer of the terminal phosphoryl of ATP to water i.e., it acts as an ATPase but at a rate 5 x 106 times slower than the above reaction. The basic and nucleophilic properties of water versus the C-6 hydroxyl of glucose are sufficiently similar to suggest no marked differences in rate. Therefore, the explanation of the rate difference is that glucose induces a conformational change that establishes the correct active-site geometry in the enzyme, whereas a water molecule is too small to do so. 

Although S-cinnamoyl-thiosubtilisin and Se-cinnamoyl-selenosubtilisin are cleaved by water at comparable rates, their hydrolysis is slower than that of native 0-cinnamoyl-subtililsin by more than two orders of magnitude. The decrease in hydrolytic rate constant may reflect subtle structural changes within the relevant active site geometries the van der Waals radii of sulfur and selenium are similar (1.85 and 2.0 A, respectively) but that of oxygen is much smaller... 

---