Molecular dynamics simulation of thermophile and mesophile ortholog pairs
Thermophilic (high temperature loving) and mesophilic (medium temperature loving) organisms have adapted their live style to survive and thrive in environments of different temperature. This requires that proteins in thermophilic organisms have similar dynamic properties as proteins from mesophilic organisms at lower temperature. Here we will test this hypothesis by performing molecular dynamics simulations of several orthologous proteins from extreme thermophilic, thermophilic and mesophilic organisms.
Thermophilic bacteria, first discovered in 1966, typically thrive at temperatures between 45 and 70 degrees. More recently discovered hyperthermophilic organisms reproduce well at temperatures between 80 and 105 degrees. A common feature of these organisms is that proteins, particularly enzymes, extracted from them are unusually stable at high temperatures and capable of performing the same function at temperatures many tens of degrees above which their mesophilic counterparts denature. It is therefore hypothesised that one of the main mechanisms by which these microbes have adapted to their extreme environment is through the evolution of proteins stable and functional at high temperatures.
Thermophilic microbes have had an important impact on the biotechnology industry. The Taq DNA polymerase from the bacterium Thermus aquaticus, is the gold standard for PCR reactions as it does not undergo significant degradation with the melting phases of the reaction. Other thermophile derived enzymes, in particular polymer-degrading enzymes such as amylases, pullulanases, xylanases, proteases, and cellulases are expected to play an important role in the food, chemical, pharmaceutical, paper, pulp, and waste-treatment industries [Bruins ME at al 2001]. Extensive research is being conducted on the potential uses of thermostable enzymes as detergents, drugs, for toxic waste removal and in drilling for oil [Haki GD et al 2003].
The deliberate design of proteins functional at high temperatures is therefore receiving considerable attention among the protein engineering community [See for example Renugopalakrishnan V et al 2005] for its potential to lead to the design of more efficient industrial catalysts, experimental reagents and even biopharmaceuticals. Some notable design successes have been achieved, for instance Hagihara and colleagues [Hagihara Y et al 2007] were able to significantly increase the stability of an immunoglobulin fold (increased melting temperature of 10 degrees) in an antibody against human chorionic gonadotropin by the addition of an engineered disulfide bond. Fibroblast growth factor 1 (FGF-1) and related factors have potential medical applications in the treatment of wound and fracture healing, cardiovascular disease and neurodegenerative diseases which are limited by the proteins very low thermal stabilities, current approaches to engineering FGFs with enhanced stability therefore hold great promise [Zakrzewska M et al 2008]. Current achievements tend to have been obtained on an ad-hoc trial and error basis with guidance only from a large collection of empirical observations - a sound theoretical understanding of the mechanism of high temperature stability is lacking although important progress is being made on many fronts [Mozo-Villiarías A et al 2006].
The naturally isolated thermophilic protein, despite their extraordinary operating temperature are often very similar in both sequence and structure to their mesophilic counterparts, for example in their molecular dynamics simulation studies Huang X and colleagues used homologous cold shock proteins from a thermophile and a mesophile that differed in less than 11 of their 66 amino acids [Huang X et al 2006]. Such observations have raised the important and largely unresolved question of just what evolutionary strategies have been employed by these proteins to enhance their stability in high temperature environments.
Numerous theories have been suggested, the packing of the proteins core due to hydrophobic interactions and the role of charged amino acids are generally seen as fundamental [Ladenstein R, et al 1998]. Giver [Giver L et al 1998] and Van den Burg [Van den Burg B et al 1998] claimed that the increased stability of thermophilic proteins was due to a decrease in conformational flexibility, numerous studies support these claims [Veronese FM et al 1984, Vieille C et al 1996] but there is also a growing body of evidence that many thermophilic proteins have a greater flexibility than their mesophilic homologues [Grottesi et al 2002, Fitter J et al 2000].
Ion-Ion interactions, or salt bridges, are also known in many cases to play an important role [Perutz MF 1978] although at least one ion-ion interaction previously thought to be important in the stability of adenylate kinases has since been shown, by molecular dynamics simulation, to play an insignificant role [Bae et al 2005].
Structural features, such as the presence of beta sheets and alpha helices and the stabilization (capping) of alpha helix ends are known to be important contributors to thermal stability [Grottesi et al 2002]. Debate continues regarding the degree to which global interactions play a role in high temperature stabilization when compared with local interactions.
Experimentally a great body of evidence has been accumulated concerning the nature of the structural variations present in thermophilic proteins. It seems clear that many different strategies are used by nature in creating proteins stable at high temperatures and that different protein families may use radically different mechanisms, and often the mechanisms are distributed over the protein and difficult to determine [Ladenstein R, et al 1998].
Another natural avenue of approach for the investigation of protein thermal stability is to undertake comparative studies between mesophilic proteins and engineered mutants, typically single residue substitutions, that show enhanced stability. Such investigations include those by Standfuss on the crystal structure of an unusually stable rhodopsin mutant and those by Adinolfi on frataxins [Standfuss J et al 2007 and Adinolfi S et al 2004]
Querol and colleagues [Querol et al 1996] investigated the structural variations present in over 196 single residue mutants showing increased thermal stability that had been reported elsewhere in the literature. From this they were able to generate a large collection of rules which they intended as a guide to protein engineering efforts. It was concluded that amino-acid substitutions which enhance thermal stability are likely to maintain or enhance secondary structures, contribute to neutralise the dipole moment of helices and strands and to lead to increased hydrogen bonding and hydrophobic driven core packing.
A proper theoretical understanding of the thermal stability of these remarkable proteins would likely lead to an enhanced understanding of protein folding mechanisms, for example investigations into the thermal stability of the cytochrome c protein from Hydrogenobacter thermophilus by Sambongi and colleagues [Sambongi Y et al 2002] has yielded important insight into the maturation and folding pathways of the cytochrome protein family.
Because thermophilic proteins often perform an essentially identical function to their mesophilic counterparts at their elevated operating temperature it is reasonable to conclude that they must exhibit, at high temperatures, dynamics similar to those displayed by their mesophilic counterparts at much lower temperatures.
The use of molecular simulations, in particular molecular dynamics simulations, is therefore expected to play a very important role in future investigations of thermophilic proteins and numerous groups have already conducted important work in this regard [See Mozo-Villiarías A et al 2006 for a nice overview]. Grottesi and colleagues ran several simulations comparing the dynamics of homologous mesophilic and thermophilic rubredoxin proteins and from these were able to observe that the principal motions of the proteins differed significantly. They also concluded that the hyperthermophilic rubredoxin was more flexible than its mesophilic counterpart although both exhibited similar flexibility at their respective optimal temperatures [Grottesi et al 2002].
An enormous amount of additional molecular dynamics work has been done to investigate the dynamics and folding of thermophilic rubredoxins. For example, Lazaridis T and colleagues [Lazaridis T et al 1997] investigate the unfolding pathways of mesophilic and thermophilic rubredoxins and find them to follow similar sequences while Bradley and colleagues [Bradley EA et al 1993] in solvent-free simulations conclude that the dynamic behaviour of thermophilic and mesophilic rubredoxins is markedly different and draw similar conclusions on the sampling of conformational space by hyperthermophilic rubredoxins as does Grottesi.
A number of experimental investigations of the molecular dynamics of important proteins from thermophiles and other extremophiles including halophiles (extreme salt concentration lovers) which complement the results of numerous simulations have also been undertaken using techniques such as neutron spectroscopy [Tehei M et al 2005].
Molecular Dynamics Introduction