THEMATICS: A simple computational predictor of enzyme function from structure

Abstract

We show that theoretical microscopic titration curves (THEMATICS) can be used to identify active-site residues in proteins of known structure. Results are featured for three enzymes: triosephosphate isomerase (TIM), aldose reductase (AR), and phosphomannose isomerase (PMI). We note that TIM and AR have similar structures but catalyze different kinds of reactions, whereas TIM and PMI have different structures but catalyze similar reactions. Analysis of the theoretical microscopic titration curves for all of the ionizable residues of these proteins shows that a small fraction (3-7%) of the curves possess a flat region where the residue is partially protonated over a wide pH range. The preponderance of residues with such perturbed curves occur in the active site. Additional results are given in summary form to show the success of the method for proteins with a variety of different chemistries and structures.T here is a need to develop methods to predict protein function from structure; this need becomes particularly acute as novel protein folds are discovered for which there are no proteins of similar structure with known function. We show that theoretical microscopic titration curves (THEMATICS) can be used to identify active-site residues in enzymes of known structure. This information will prove to be important in the current challenging quest to predict protein function from sequence and structure.Knowledge of protein sequences and structures has burgeoned very recently as a result of genome sequencing (1) and structural genomics efforts. To translate such information into tangible benefits to humankind, the next step is to develop methods that enable one to predict and to establish function from structure. Techniques used to predict function from sequence or function from structure are in their infancy and rely either on analogies to related proteins of known function (2-8) or on computational searches for binding sites by docking (9) of selected sets of small molecules onto the structure. For instance, recent work has searched the Protein Data Bank (PDB; ref. 10; http:͞͞ www.rcsb.org͞pdb͞) for previously unrecognized cation-binding sites (11).However, there is as yet no reliable method to identify active sites of enzymes or other interaction sites of proteins in the absence of biochemical data, even when the structure is known. There is a wealth of biochemical data that demonstrates that active-site residues involved in specific kinds of chemistry possess predictable chemical properties that enable one to identify them as active-site residues. The most important of these properties is a perturbed pK a that can be determined experimentally by pH titrations of the activity of the enzyme. Herein we demonstrate that theoretical titration functions can identify active-site residues that are involved in Brønsted acid-base chemistry for a variety of proteins, thereby identifying the active site from the location of the residues.In an experimental titration curve, one typically plots pH as a function of the volume of...

Computed Energetics of Nucleotides in Spatial Ribozyme Structures: An Accurate Identification of Functional Regions from Structure

“…Electrostatic effects would provide an important contribution to enzyme catalysis [15]. These general considerations are confirmed by a number of theoretical and biochemical studies of individual protein enzymes [16,17,18,19,20,21,22] and can be summarized as follows: Relative physicochemical properties of the functional residues of the enzymes are very distinct from the other residues in a molecule. Now, this is the central idea of the present method: If some relative physicochemical property can be calculated from a structure of a ribozyme, then the bases involved in catalytic site can be identified.…”

“…Functional residues destabilize proteins/enzymes [16,17,18,19] and the energetic contribution of individual residues can be reliably calculated ( [14,20,21] and unpublished data). It is also known that catalytic residues in enzymes also have unusual regions on theoretical microscopic titration curves [22]. Thus, once again, catalytic residues are distinct from all of the remaining residues in an enzyme [22].…”

“…Thus, once again, catalytic residues are distinct from all of the remaining residues in an enzyme [22]. For proteins with known spatial structure, the functional residues can be predicted using these theoretical methods [21,22].…”

Sequence and structure continuity of evolutionary importance improves protein functional site discovery and annotation

“…Approaches based on preexisting structure may be grouped broadly into those using energetics, [12][13][14][15] and others using structural and geometric analysis. [16][17][18][19] Here, we focus on comparative, or evolutionary approaches, [20][21][22][23][24][25][26] and specifically on the evolutionary trace (ET).…”

Application of New Multiresolution Methods for the Comparison of Biomolecular Electrostatic Properties in the Absence of Global Structural Similarity

“…Some effort has also been made to pursue more "informatics"-based approaches to the interpretation of electrostatic properties. Much of this work includes identification of functionally-relevant residues in biomolecules by looking at electrostatic destabilization of conserved residues [18], highly shifted pK a values [44], clusters of charged residues [59], protein-membrane interactions [40], and other structural characteristics [55]. Other research has focused on comparisons of electrostatic potentials including global analyses of the biomolecular structure [38,9,40,51,37,30,36,47,43,8,53,34,46,35,52] both in threedimensional space over the entire biomolecular structure and at localized regions such as active sites [52,6,22].…”