- Guarded Neutrality: Diplomacy and Internment in the Netherlands during the First World War?
- Related Articles.
- Why Does Cargo Spend Weeks in Sub-Saharan African Ports?: Lessons from Six Countries (Directions in Development)?
- Renormalization: From Lorentz to Landau (And Beyond)!
- Publication details.
- (IUCr) CADEE: Computer-Aided Directed Evolution of Enzymes.
FAD and choline were kept all of the time in the quantum mechanics QM region, and main active-site residues—His, His, Trp, and Glu—were incorporated one by one to the QM region while placing all other residues into molecular mechanics MM region. Our second approach to model the hydride-transfer process did not utilize simplified versions of residues. A total of 93 atoms were placed in the DFT region with a net negative charge and 83 residues totaling atoms were placed in the MM region with an overall negative charge.
In this sense, we tried to address several key features involving the details of the reaction mechanism, noncovalent interactions, and the energetics of the hydride-transfer processes. The main objective was to formulate a model active site including main amino acids that has considerable chemical interactions with the substrate, choline, and then to study the hydride-transfer process from choline to FAD.
One of the carboxylate O atom in glycine betaine is in close proximity to the two histidine residues, His and His, Ser, and Asn, suggesting that choline might display H-bonding interactions with these residues during the hydride-transfer process. Glycine betaine was converted to choline by transforming the carboxylate group into alkoxy functionality considering that the hydride-transfer step occurs after the deprotonation of the alcohol group in the choline.
Histidines were kept as neutral, and glutamate was kept as negatively charged. Optimization of the model geometry including choline, FAD, and all amino acid residues, including His, His, Glu, and Trp, failed to converge in a reasonable time using g basis set. It is more likely that the initial geometry extracted from crystal structure is far from an ideal starting point for the optimization process. We did not utilize any semiempirical or molecular mechanical methods to obtain a better initial geometry since the accuracy of DFT methods is expected to provide more reliable initial structures.
To circumvent initial convergence problems, we adopted an atomistic approach to reach the optimized geometries in a reasonable time for the model active site together with choline and FAD. In the first step, the hydride-transfer process from choline to FAD was studied in the absence of all amino acid residues. Numbering of C4a is based on the systematic numbering of isoalloxazine ring.
Computational Approaches for Studying Enzyme Mechanism Part A, Volume - 1st Edition
Thus, we were able to reach five optimized geometries for the model enzyme—substrate complexes in a systematic way. H atoms connected to N3 atoms at His and His residues interacts with O atom in choline. The H-bonding distance for His, 1. This type of interaction is caused by the electrostatic interaction between the positively charged N atom and the electron cloud of aromatic Trp The distances between O atoms in Glu and two closest H atoms in the choline are around 2. This indicates the importance of Glu in the positioning and binding of choline in the active site. Structure of optimized model active site in the presence of choline, four amino acid residues, and FAD model QM-5 obtained with using MX functional with g basis set using tube models excluding H atoms except the ones shown in green.
It has to be noted that our model active site could only be optimized in the presence of substrate, choline. It should be also noted that the model QM systems for the enzymatic reactions entail simplified, flexible, and isolated active-site residues in the calculations.
Enzyme mechanisms: fast reaction and computational approaches.
However, in real cases, the flexibility of residues is limited, and the contribution of the other residues is grossly ignored. It is the recognition of the noncovalent interactions between each residue and choline by the QM methods that yields an optimized model active site, which is similar to the crystal structure. Without choline, optimization might not converge or might converge to a totally unrelated structure.
Another consideration might involve a possible noncovalent interaction between FAD and choline. This observation suggests that there might be a dipole interaction between negatively charged O atom in choline and the electropositive C4a atom in FAD. To address the limitations of model active sites obtained through QM calculations, we adopted a hybrid approach, ONIOM, incorporating not only more residues around active site but also rigidity and three dimensionality.
In addition, this may eliminate the need to incorporate a protein solvation model into QM calculations. In the same fashion, this approach might, in principle, provide a better snapshot of the active site. To generate model systems, FAD and choline were kept always in QM region, while four amino acid residues—His, His, Trp, and Glu—were incorporated into QM region one by one in four steps in a successive manner, and in each step, geometry optimization process was repeated.
In total, 83 residues were always kept in the MM region. Thus, we were able to reach optimized geometries for the model enzyme—substrate complexes, RC, in a systematic way leading to five model systems. The geometries of each model enzyme—substrate complexes were included in the supporting document. As expected, the orientation of residues is different from QM-5, which is imposed by the MM restriction; the orientations and locations of residues bear more similarity with the crystal structure.
In the QM-5 model, the isoalloxazine ring is nearly planar. However, in the QM-MM-5 model, an obvious bent is apparent between benzene and pyrimidine moieties. In a previous study, this case was observed for a flavo-enzyme, cholesterol oxidase, in the presence of its native substrate, cholesterol.
Our second computational approach, QM-MM model, corroborates this finding, and this particular observation is actually a computational verification of an experimental result.
In future studies, we are planning to reserve more focus on this subject. Structure of optimized model active site in the presence of choline, four amino acid residues, and FAD while excluding all residues to MM region model QM-MM-5 obtained with MX functional with g basis set using tube models excluding H atoms except the ones shown in green. Structure of optimized model enzyme—substrate complex including choline, four amino acid residues, and FAD while excluding all residues to MM region model QM-MM-5 obtained with MX functional with g basis set using tube models. MM region residues are shown with line-bond models.
It is also noteworthy to point out that QM-MM-1 model system, which places all residues into MM region except FAD and choline, did not furnish appreciably H-bonding distances between choline and His— 4. For Trp and Glu residues, both QM and MM methods predicted similar interactions with choline based on their distances.
The PES scan process was repeated four more times to include all of the main amino acid residues, His, His, Trp, and Glu, in the active site to locate the TS structure corresponding to the hydride-transfer step. Structure of optimized TS structure in the presence of choline, four amino acid residues, and FAD QM-5 obtained with MX functional with g basis set using tube models excluding H atoms except the ones shown in green. This illustrates that H-bonding interactions in the RC is stronger than in the TS, which is a quite expected phenomenon for a negatively charged O atom in RC compared to a neutral aldehyde-like TS.
Both O atoms in Glu are around 2. As clearly visible, the isoalloxazine ring is also distorted, and the increase in the H-bonding distances from RC to TS between His residues and O atom in choline is more prominent. Structure of optimized TS in the presence of choline, four amino acid residues, and FAD while excluding all residues to MM region model QM-MM-5 obtained with MX functional with g basis set using tube models excluding H atoms except the ones shown in green.
For QM calculations, adding amino acid residues of His and His to the model systems increased the activation barriers for forward process, whereas Glu caused a considerable decrease. The deprotonated negatively charged O atom in choline is expected to stabilize this positive charge density. Inclusion of His, which forms QM-2 Table 1 , increased the activation barrier by 5. Since His forms H-bonding interaction with the negatively charged O atom in choline, the negative charge density on the O atom is expected to decrease. The effect, despite being moderate, could be stated for the QM-MM-2 model having 2.
In the same fashion, addition of Glu to QM-5 model, which forms QM-5 Table 1 , exert a similar effect on the activation barrier.
- Contemporary Debates in Philosophy of Religion.
- Stanley Cavell (Contemporary Philosophy in Focus).
- Working World: Careers in International Education, Exchange, and Development.
- Financial Statistics and Mathematical Finance: Methods, Models and Applications?
- Application of Molecular Simulations Toward Understanding Cellulase Mechanisms | SpringerLink!
- Emad Tajkhorshid!
- Table of Contents.
- Shop and Discover over 51, Books and Journals - Elsevier.
- U.S. Commercial Remote Sensing Satellite Industry: An Analysis of Risks?
- Paints & Coating Industry January 2012!
- Biological Microarrays: Methods and Protocols.
- Pin Me, Pay Me!: Have Boots, Will Travel!
- Matrix III, Volume 1: The Psycho-Social, Chemical, Biological and Electromagnetic Manipulation of Human Consciousness.
This observation aligns with the study reporting lack of choline oxidase activity in the mutant enzyme in which Glu was replaced with alanine. Furthermore, the corresponding RC structure with these two residues did not reveal any prominent chemical interaction with either FAD or choline. There was not any notable trend in the activation barrier of the reverse hydride-transfer process for QM models. However, inclusion of first residue causes a significant decrease in the activation energy, and further addition of residues causes erratic increases or decreases for both QM models, B3LYP and MX.
This might be due to the absence of any constraints to preserve the geometry of model active site. Even though mobility of residues in the RC and TS models are limited, PC structures through QM-1—5 show considerable mobility and variability in terms of positions of residues and substrate. Inclusion of residues to QM region steadily increased the activation barrier. This observation suggests that the interaction between betaine aldehyde and active-site residues including FAD is tighter than the interaction with choline.
However, Trp and Glu are closer to betaine aldehyde than choline. This might explain part of the stability conferred to PC.
Moreover, a considerable interaction exists between reduced FAD and betaine aldehyde to such an extent that a distance of 2. One more consideration has to be kept in mind that the transition from RC to TS does not entail any major shift in the relative position of choline with respect to FAD. As we switch more residues from MM region to QM region, a significant decrease in the energies of PC models is visible through QM-1 to QMmodels presumably as a result of rearrangement of active site causing favorable interactions between reduced FAD, betaine aldehyde, and surrounding residues.
Structure of optimized PC in the presence of choline, four amino acid residues, and FAD while excluding all residues to MM region model QM-MM-5 obtained with MX functional with g basis set using tube models excluding all H atoms except the ones shown in green.
The optimized geometries were very close to each other obtained with these two basis sets, and the difference in the forward and backward activation barriers was in the range of 0. This observation indicates that calculations with g basis set can provide proper quantitative and qualitative mechanistic information for the hydride-transfer process. Furthermore, during the hydride-transfer step, it was also suggested that His stays protonated based on a study in which His was replaced with alanine. However, addition of neutral imidazole did not contribute any recovery to the enzymatic activity.
In the previous model systems, which were formulated for the hydride-transfer step, His was kept neutral and all of the calculated activation energies were based on the assumption that His is neutral during the hydride-transfer process. To study the effect of protonation state of His on the energetics of hydride-transfer process, His N1 position was protonated. Then, the resulting systems were subjected to geometry optimizations, followed by PES scans to locate TS structures representing hydride-transfer process in the presence of protonated His Optimized structures of RC and TS for the hydride-transfer process for the model active site QM-2 in the presence of protonated His obtained with the MX functional with g basis set using tube models excluding all H atoms except the ones shown in green.
Instead of N3 position of His, the proton is attached to O atom in choline. This observation is not surprising considering the p K a values of choline, According to this model, it could be concluded that the deprotonation of choline and the hydride transfer are coupled and the deprotonation happens on the way while the hydride ion moves from choline. However, a collection of experimental studies indicated that the deprotonation of choline and the transfer of hydride steps are decoupled and deprotonation should occur in a separate step before the hydride transfer.
The activation barrier for the hydride-transfer step from protonated choline to FAD was found to be always 10—15 kcal more than the deprotonated models QM-2 models vs QM-2 in Table 1. These results clearly point out that energetically the hydride transfer is more favorable when choline is deprotonated and His is neutral. It is quite likely that His is indeed responsible for the deprotonation of choline along with a couple of residues who might accept proton from His in a cascade of proton-transfer steps.
The activation energy turned out to be a very high value, A closer analysis of TS-1H 2 O structure reveals the apparent reason for this high value. The TS structure looks like a highly strained four-membered ring, which is expected to have a high ring strain. Indeed, the calculated activation energy turned out to be The energetics of these two models can be compared to QM-1 models QM-1 for MX in Table 1 for the initial hydride-transfer step since they do not contain any residue.
The hydride-transfer process requires We also tested these two model systems with QM-MM calculations and we observed similar high activation energies for a kinetically fast step. Optimized structures of RC, TS-1H 2 O, TS-2H 2 O for two water molecules , and PC for the hydration of betaine aldehyde in the presence of one water molecule obtained with the MX functional with g basis set using tube models excluding all H atoms except the ones shown in green.
It is likely that His or His residues might act as base catalyst deprotonating water molecule and forming hydroxide ion. It should not be also ruled out that a hydroxide anion might diffuse to active site. We tried to locate a TS structure through PES scans for a model system including a hydroxide ion, a protonated His residue, FAD, and betaine aldehyde to simulate a model system as if His residue was the basic residue that deprotonated the water molecule.
However, the protonated His residue could not hold its proton, and it was transferred to hydroxide ion forming water molecule.