Effect Of Glycation On Substrate Binding Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

A calcium dependent esterase enzyme, human paraoxonase 1 (huPON1) causes the hydrolysis of wide range of substrates including organophosphates, ester, detoxification of nerve agents, lactones and paraoxon. Despite of all these facts the natural substrate of PON family is still unknown. In present study we try to identify the effect of production of advanced glycation end product at the active site of protein. For this purpose the molecular dynamic simulation and docking tools were used. Structure activity studies were performed using lactones and paraoxon as substrates. We found the structural alteration in the glycated model of HuPON1 and weak interactions of substrate with that of active site cleft. The residues crucial for the lactonase activity including His115, Asn224, Asn168, Asp269 shows some conformational changes which causes steric hindrance in substrate binding and hence hydrolysis. In case of paraoxon, we found that the involvement of His115 and Asn168 in the hydrolysis of HuPON1 and Phe222 shows hydrophobic interaction with in the active site moiety. The disturbance in the orientation of these crucial residues was also observed in molecular dynamic simulation studies. In conclusion, this study gives an idea about the fluctuation and deviation in the structural kinetics of protein after the production of advanced glycation end product. The active residues and the residues of active site lid upon glycation causes hindrance in the hydrolysis of substrate.

Key words: human paraoxonase 1,


Human Paraoxonase 1 (HuPON1) is a calcium dependent esterase located on chromosome 7. It belongs to a multigene family containing three members, PON1, PON2 and PON3. [1]. Although all three members share 65% homology at sequence level, they differ in their expression pattern. PON1 and PON3 are primarily expressed in liver and are circulated in blood in association with high density lipoproteins (HDL) while PON2 is ubiquitously expressed in various tissues but is absent in blood [2, 3]. Natural substrate and hence the physiological role of these enzymes is still unknown but PON1 has paraoxonase, arylesterase and lactonase activity and is capable of neutralizing toxic compounds such as organophosphates and xenobiotics [4, 7]. PONs offer protective role against LDL oxidation thus decreasing the level of cholesterol [8] indicating their antiatherogenic property. Animal model studies have indicated that PONs expression is capable of inhibiting or reversing atherosclerosis development via reducing oxidative stress thus providing ample evidence suggesting role of PONs as an established anti-oxidant. Decreased PON1 activity has been observed in different disorders such as cancers, alzheimer, cardiovascular diseases, diabetes and cataract which are not surprising as oxidative imbalance has been implicated in etiology of these disorders. Oxidative stress is also suggested to be a consequence of hyperglycemia due to possibility of glucose auto-oxidation and generation of advanced glycation end products (AGEs). Indeed, loss of PON1 activity was found to be greater in diabetic subjects with respect to controls.

PON1 contains 355 amino acid residues and has been studied in detail regarding it's catalytic site, binding interactions with HDL and antioxidant capability at structural level. Crystal structure of PON1 available at PDB is of recombinant PON that attains a six bladed β propeller structure having two calcium ions for the stability and catalytic activity of protein [9]. Ca1, being catalytic calcium, is found on surface while Ca2 is structural calcium located in the central region and both are about 3Å apart [10]. Ca1 interacts with side chain oxygen of five residues including Asn224, Asn270, Asn168, Asp269 and Glu53 along with the water molecule and oxygen of phosphate ion [11]. The catalytic calcium facilitates the formation of enzyme-substrate complex and accelerates the breakdown of intermediates into enzyme and products [12]. The residues which are found to be crucial for PON1 activity include the hydrophobic pocket residues Leu240, Val346, Phe292, Leu291, Phe222, Leu69, His115, His285, Thr332, Leu267, Phe347 in which His115-His134 dyad seems to be essential [9, 13]. Recent investigation leads to finding that paraoxonase activity is mediated by helix2 and loop residues such as Tyr71, Pro72, Arg192, Leu191, Met196, Asp188 and Pro189 [14].

Current study was aimed to evaluate the effect of glycation on the binding capability of PON with its respective substrate, Paraoxon and lactones in active site of HuPON1 using docking software. In order to examine glycation induced conformational variations, molecular dynamics simulation studies of PON1 with and without glycation on picoseconds time scale has been carried out. In this article this is one of the first attempts to use molecular dynamic simulation to have insights on structural/functional variation of HuPON1 upon glycation.


Computational resources

The bioinformatics analysis of protein molecules was performed under licensed window Xp and the model was constructed using MODELLER 8v1. The docking study was done by utilizing FlexX windows version and visual interference was carried out using CHIMERA and DS Pro Viewer. Molecular dynamic (MD) simulation were carried out on AMD opteron Model 2220 dual care (1.8 GHz) equipped with Linux environmentruning under Redhat 5.1. Amber was utilized for MD simulation while VMD and chimera were in use in visualization of trajectories

Modeling of human paraoxonase 3-D structure

The glycated and non-glycated models of HuPON1 were constructed as described earlier [ 15]. In brief, sequence of human PON1 (accession no: P27169) was retrieved from SWISSPROT database [16] and was subjected to BLAST (Basic Local Search Tool) algorithm [17] against PDB. PON1 showed 83% similarity with recombinant Paraoxonase (PDB ID: 1VO4) which was selected as a template. Twenty models of PON were constructed using MODELLER version 8v1 [18] and best model was selected on the basis of stereochemical properties after evaluation by the PROCHECK [19]. To create a glycated mutant, pentosidine, an AGE, was inserted at Lys70 and Lys75 using molecular graphic program DS Viewer Pro [20]. The reason for selection of Lys70 and Lys75 was twofold: firstly, Lys is most susceptible residues for glycation and secondly, both residues are located at active site lid, the site of substrate entry. After modification, glycated models were reconstructed as mentioned earlier and the best model was selected for docking and simulation studies.

Docking studies

Docking of huPON1 glycated and non-glycated models with lactones and paraoxon was carried out using FlexX software version 3.1.1 [21]. The ligands were obtained from Pubcomp library of NCBI database. Docking interactions were visualized by using ligplot [22]. During docking through FlexX, the receptor conformation was kept fixed while the conformation of ligand remained flexible. FlexX optimizes the torsion angles of ligand without altering its bond length and angles. The receptor models were without water molecule while the calcium which is crucial for activity [10] was used as reference ligand. Essential residues for PON1 activity were included in the binding cavity of ligand [9] during analysis. Default parameters and SIS algorithm of FlexX were utilized for this work.

Statistical Analysis

Statistical analysis was performed using SPSS 10.0 in order to compare the correlation rate of glycated model with that of non-glycated model on hydrolysis of lactone depending upon the side chain residues. Dependent variables were used for this study. Negative correlation was found by Pearson correlation coefficient.

Molecular Dynamics (MD) simulations

MD simulations were carried out for modeled systems in explicit solvent using AMBER 10 package and the force field 03 [23]. To maintain the neutrality of the system, 11 and 16 Na+ ions were added in glycated and nonglycated PON1, respectively. Both models were solvated by 30,462 and 66,545 water molecules, respectively in a rectangular box around the solute unit. The sizes of cells were 100.59 Å x 118.34 Å x 117.32 Å consisted of 10,154 atoms and 98.62 Å x 94.43 Å x 107.53 Å consisted of 22,182 atoms for glycated and nonglycated PON1, respectively. The rectangular solvate box was created by using Xleap interface of Leap. The solvated protein systems were subjected to a thorough energy minimization before MD simulations. In order to relax the system and to avoid any steric conflicts, minimization of water molecules was done while holding the solute fixed (1000 steps using steep descent algorithm followed by 1000 steps of conjugate gradient minimization of the whole system). Bond lengths involving hydrogen were constrained with SHAKE [24] algorithm with harmonic restraints of 25 kcal/mol A. The time step for all MD simulations was set to 2 fs and nonbonded cutoff of 8 Å was used while periodic boundary conditions were applied to simulate a continous system. The system was then subjected to a gradual temperature increase from 0 to 300 K over 100 ps, and then equilibrated for over 100 ps at 300 K, followed by the production run of 5 ns. Temperature (298 K) and pressure (1 atm) were kept constant by utilizing Berendsen coupling algorithm [24] with a time constant for heat-bath coupling of 0.2 ps. Dielectric constant was set to 1.0. Long-range electrostatic calculations were carried out by particle mesh Ewald method. The resulting trajectories were analyzed using PTRAJ module of the AMBER package.

Result and Discussion

Contemporary developments in computational approaches such as docking studies and molecular dynamics simulation methods have paved the way to open new horizons in the investigation of interactions within the protein molecule and with respect to its substrate. These tools are beneficial in identification of structural variations due to the alteration in single amino acid residue and can find the orientation that makes a stable interaction having minimized energy [25, 26, 27, 28, 29, and 30]. Current study focused on the structural alterations induced in glycated and non-glycated models of paraoxonase after docking with lactones and paraoxons. Furthermore, molecular dynamic simulation studies were performed in order to depict the interaction of molecules within the picoseconds time scale.

Glycated and non-glycated models of HuPON1

Previous studies indicated that upon glycation the activity of proteins decreases due to the production of advance glycation end products (AGEs) [31]. These AGEs cause conformational changes in the structure of a protein molecule which in turn effect the substrate-ligand interactions. Earlier studies show that upon glycation at Lys70 and Lys75 some conformational changes occur at the side chain of Asn168, Asn224 and Asp269. These residues are found to be involved in maintaining the size and orientation of calcium binding moiety [11].

Docking Analysis

The glycated and non-glycated models of HuPON1 were used as receptor and 14 lactones and 3 paraoxon were selected as ligand molecules. Each ligand was docked deep into the active site of modified and normal protein molecules respectively. On the basis of docking energies, best interactions were selected. There is a remarkable increase in docking energy in glycated protein interactions as compared to non-glycated protein as shown in Table 1. The crucial residues for the lactonase activity of HuPON1 include Leu69, His115, His134, Asp69, Phe222, Asp269, His285, Phe292, Thr332, Val346, Trp281 [ 9, 13]. Mutational studies indicate that replacement of His115 causes decreased lactonase activity but retains the paraoxon activity of PON1 [32].

Docking of glycated and non-glycated PON1 with lactones

Diagramatic analysis of docking results shows the interaction of His115 with lactones. In non-glycated model the residues, which play an important role in lactone hydrolysis includes Tyr71, Asn168, Asn224, His115, Leu69, Val346 and Phe222. His115, Asn168 and Asn224 show hydrogen bonding with lactones having bond length 2.90Å, 3.15Å and 2.78Å respectively. The oxygen (O1) atom of lactone molecule interacts with NE2 nitrogen of His115, while oxygen O2 atom of lactone shows hydrogen bonding with that of ND2 of Asn168 and Asn224. Recent studies showed the important role of Tyr71 in binding of substrate with the active site of human PON1 (14). This Tyr71 backbone along with the Leu69 is the opening residues of active site lid which play significant role in substrate selectivity [9]. These residues attain hydrophobic interactions with that of ligand molecule as shown in Fig 1. While in case of glycated model, the hydrophobic interactions of ligand were observed with Glu53, Tyr71, Phe222 and Val346. The entrance of ligand molecule within the active site cleft is same as in normal model. But due to the production of AGE at Lys70 and Lys75 the active site lid attains some conformational changes which causes steric hindrance in such a way that ligand is unable to be fixed properly in the active site cleft of protein (results shown only for Dihydrocoumarin).

Docking of glycated and non-glycated PON1 with paraoxon

The mechanism of hydrolysis of paraoxon by protein paraoxonase remained a question. In this work the crucial residues for the paraoxonase activity of PON1 are perdicted, the ligands paraoxon, mioticol and nitrophenylhalon were used and docked with glycated and non-glycated protein. They all differ in their side chains. The resulting outcome indicates the hydrogen bonding of His115 with that of all three molecules. The NE2 of His115 interacts with O7 of paraoxon with a distance of 2.93Å. On the other hand ND2 of Asn168 also has hydrogen bonding with that of O7 of paraoxon with a lesser distance of 2.75Å. Tyr71 and Phe222 show hydrophobic interactions with paraoxon as shown in figure 2. Earlier studies indicate that mutation of Phe222 results in decreased PON activity [14]. After the glycation some conformational changes occurred in the crucial residues of HuPON1 [15]. This results in weak interaction of paraoxon with that of glycated protein as shown in Fig 2. The residues His134 provides the proton shuttle mechanism and supports His115 for its activity. Due to the modification of protein His134 along with active side lid residues Tyr71, Pro72, Gly73, Lys75 shows hydrophobic interaction with ligand molecule.

We also try to identify the correlation between the two dependent variables including docking energy and length of the aliphatic side chain of lactones and found a strong inverse correlation among them. The correlation graph for glycated and non glycated models and their interaction with lactone were constructed in order to find out the differences between the two. The R value obtained for non glycated and glycated models are -.902 and -.771 respectively.

Comparasion of glycated model with non glycated model using molecular dynamic simulation

After having certain deviations in the interaction of substrate with that of glycated and nonglycated model of huPON1 we try to compare the dynamics of these structures using MD simulation approach. For this purpose, the total energy of the whole system and root mean square deviation (RMSD) from the starting structure are observed in order to determine sustainability and convergence of MD simulations. These changes in the structural conformation were monitored in terms of RMSD and root mean square fluctuations (RMSF). Figure 4 shows plots of RMSDs calculated for the backbone atoms of the two simulated systems over the entire simulation of 5 ns. The average RMSD reached below approximately 2.12 Å in case of glycated-PON. The non-glycated-PON, on the other hand, exhibits a higher value for the RMSD (around 4.79 Å) with an increasing trend (see Fig 4). This results mostly from a higher flexibility in a non-glycated system, which upon the formation of advanced glycation end products, displays comparatively stable conformation throughout the MD simulation. In a recent study, same trend of RMSDs has been found for the MD simulation of apo and ligand bound complex form of huPON [14].

Dynamic fluctuations (RMSF) of proteins around their average conformations act as an important indicator of many biological processes i.e. enzymatic activity. The RMSF for our system of interest was examined for the C-alpha atoms of each of the residue representing the average displacement of these atoms. Figure 5a and 5b displays RMSF plots of glycated-PON and non-glycated PON with an average value of 4.24 Å and 4.38 Å respectively. Figure 5a shows the fluctuations resulting from truncated pdb of non-glycated model while the whole protein taken in to account for MD simulation of glycated-PON. Inspite of these higher values the atomic fluctuations are slightly lower in case of glycated PON suggests that average RMSF decreases upon the formation of AGEs. This decrease in value predicts that there is some hindrance in the enzymatic activity of protein upon glycation. This finding is complete coherence with the outcome of our docking results. Due to the unavailability of the x-ray structure of glycated protein RMSF value cannot be compared.

The nonglycated PON1 demonstrates significant fluctuation within the regions surrounding the active site as compared to the glycated PON1 where the overall fluctuation shows the stability of the system towards the end of simulation. The active site appears to be structurally deviated in case of nonglycated PON1 model as compared to the glycated one. Certain flexibilities were observed in the conformation of the catalytic His115 which was also observed in docking studies. After glycation the fluctuation and deviation was observed at Tyr71 and Lys70 which play important part in the recognition of substrate [14] and provide its backbone for the stability of active site [9]. The interaction of calcium with five protein residues including Asn224, Asn270, Asn168, Asp 269 and Glu53 seems to be highly disturbed in the glycated model. This results in decrease stability which is provided by calcium to a protein molecule.


This study provides information related to the active site architecture of HuPON1 along with its substrate binding. The three dimensional structure of nonglycated and glycated huPON1 was constructed and substrate interactions was observed by docking and MD simulation tools. Our study suggests the importance of His115-His134 dyad in both the lactonase and paraoxonase activity of HuPON1. We found the hydrogen bonding interaction of His115 with that of lactones as well as paraoxon while Tyr71 and Phe222 show some hydrophobic interactions. Upon glycation at Lys70 and Lys75, the structural conformation of crucial residues changes such that the interactions of substrate to active site are highly disturbed. The residue in the active site lid and Tyr71 which is helpful in substrate recognition shows weak interactions with that of substrate.