Background
 |
|
Figure 1: A ribbon model of the red kidney bean purple acid phosphate enzyme. The inset picture shows the active site, with a Zn and Fe iron. |
Most of the biochemical reactions that occur in the human body do not take place spontaneously - they require the presence of molecules called enzymes to occur at a significant rate. Enzymes act as catalysts in reactions involved with processes such as the digestion of food, signal transduction and muscle contractions. Malfunctioning enzymes can also lead to genetic diseases, and conditions such as mental impairment and cancer.
Enzymes are proteins - long chains of amino acid residues - and are usually much larger than the substrate molecules they act on. Thus the section of the enzyme that actually makes contact with the substrate and causes catalysis - known as the active site - is only a very small part of the protein, typically only 3-4 amino acids in length. In a catalysis reaction, the substrate first binds with the active site, forming an enzyme-substrate complex. In this step the enzyme is thought to change shape slightly in order to better accommodate the substrate. Once bound, the chemical bonds in the substrate molecule are broken, and new bonds are formed, creating the product molecule. The enzyme then releases the product and returns to its original shape, ready to bind to other substrate molecules and repeat the process.
It is thought that the substrate-to-product reaction occurs so much faster than without the enzyme because the enzyme holds the substrate in a configuration that lowers the energy required to break the bonds of the substrate, thus lowering the activation energy of the reaction. The function of an individual enzyme depends on its three dimensional structure - an enzyme will only act on a substrate that can fit into its active site - and so to understand the processes that occur in enzymatic catalysis we must determine this structure.
The Project
Andrew Dick and Associate Professor Mark Riley of the University of Queensland are concurrently performing spectroscopic and computational analysis in an effort to contribute to the understanding of enzymatic catalysis. Experimentally they're performing a variety of spectroscopic techniques to determine the electronic and geometric structures of proteins of interest. To aid in understanding the spectroscopy results, they are using quantum chemistry computing programs to model the active sites of these proteins. In particular they are interested in bimetallic compounds - enzymes with two metal ions at their active site.
 |
|
Figure 2: A model complex with a fragment of the active site of a protein. |
Electronic spectroscopy involves exciting the molecule of interest to an excited electronic state, and using the resultant spectra to calculate the energy of the transition. From these energies it is possible to determine the transitions that occurred, and depending on the type of spectroscopy performed, build up information about the molecule, such as electron configuration and geometric structure. In their work Andrew and Professor Riley use a variety of spectroscopic techniques, including absorption, electron spin resonance and magnetic circular dichroism spectroscopy. As well as structural information, they are also studying the effects of certain condition changes on the active site. For example, how does the structure of the active site change if the pH of the substrate changes, and how does this affect how the active site acts as a catalyst?
However, the presence of metal ions in these molecules makes analysis of the spectroscopy results, along with modelling of the enzyme and the reaction, much more challenging. Typically quantum chemistry calculations are performed for organic compounds - these molecules are composed of lighter molecules, with few electrons. When molecular bonds form between these lighter atoms, the electrons pair up according to spin, and so in performing spectroscopic analysis, you are generally only interested in transitions between singlet states. However, metals such as those commonly found at the active sites of proteins, including, among others, iron, nickel and cobalt, have a large number of electrons, many of which are unpaired. The presence of more than one metal means that electrons from different metals may couple ferromagnetically or anti-ferromagnetically - aligning or anti-aligning - resulting in many more spin states and therefore additional energy levels that can be occupied. Thus there are many more transitions that can occur in spectroscopic analysis. In addition to this, the presence of the extra electrons introduces more forces to include in computational calculations, increasing the complexity of these calculations.
To aid in the calculations for proteins, Andrew and Mark also study model complexes - simpler molecules that can be made in the laboratory. These model complexes - such as that shown in Figure 2 - are simpler to study, as the crystal structure is known. The complexes studied have a fragment of the active site in them, and so studying the transitions occurring in the model complex may give us some idea of the transitions that occur in the protein.
Computational Tools
Modelling of the proteins is performed using the quantum chemical packages Gaussian 03 and Amsterdam Density Functional (ADF). These packages allow Andrew to first perform quantum mechanical calculations on just the active site and the substrate, without taking into the consideration the rest of the enzyme. Then using these results as a foundation, the rest of the enzyme may be incorporated into less computationally expensive molecular dynamics calculations, to give a more realistic description of the system.
The calculations are performed on computing clusters at the Computational Molecular Science (CMS) computing facilities at UQ, and the APAC computing facilities at ANU.
Contacts
Andrew Dick,
Associate Professor Mark
Riley
School of Molecular and Microbial Sciences, UQ
Publications
M. Lanznaster et al., "A new heterobinuclear FeIIICuII complex with a Single Terminal FeIII-O(phenolate) bond. Relevance to purple acid phosphatases and nucleases", Journal of Biological Inorganic Chemistry, 2005, 10, 319-332.
E.G. Moore, P.V. Bernhardt, M.J. Riley and T.A. Smith, "Electronic energy-transfer rate constants for geometric isomers of a bichromophoric macrocyclic complex." Inorganic Chemistry, 2006, 45, 51-58.
Written by T. Curtis, September 2006