Molecular Docking and Dynamics Simulation Study ...

Send Orders for Reprints to [email protected] Combinatorial Chemistry & High Throughput Screening, 2014, 17, 891-903

891

Molecular Docking and Dynamics Simulation Study on the Influence of Zn2+ on the Binding Modes of Aggrecanase with its Inhibitors Panneer S.R. Suganya, Sukesh kalva and Lilly M. Saleena* Department of Bioinformatics, SRM University, SRM Nagar, Kattankulathur, Kancheepuram District, Chennai 603203, India Abstract: Zinc plays a vital role in structural organization, regulation of function and stabilization of the folded protein, which ultimately activates or inactivates the binding sites of the protein. Its transition makes a major change in the protein and its binding affinity. The ligand binding aggrecanases can be influenced by Zn2+ ions; therefore the study focuses on checking the binding mode in the presence and absence of zinc using Docking and Molecular dynamics simulation. The crystal structure with zinc was considered as wild type (ADAMTS-4-1Zn2+, ADAMTS-5-1Zn2+) and the crystal structure without zinc was considered as the mutant type (ADAMTS-4-0Zn2+, ADAMTS-5-0Zn2+). Mutations were made manually by deleting the zinc atom. ADAMTS-4-1Zn2+ had the best Glide score of -12.66 kcal·mol−1, whereas ADAMTS-4-0Zn2+ had -11.69 kcal·mol−1. ADAMTS-4-1Zn2+ had the best glide energy of -72.29 kcal·mol−1, whereas ADAMTS-4-0Zn2+ had -68.44 kcal·mol−1. ADAMTS-4-1Zn2+ had the best glide e-model of -116.34, whereas ADAMTS-4-0Zn2+ had -104.264. The RMSD value for ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+ was 1.9. These results suggested that the absence of zinc decreases the binding affinity of ADAMTS-4 with its inhibitor. ADAMTS-5-1Zn2+ had the best Glide score of -8.32 kcal·mol−1, whereas ADAMTS-5-0Zn2+ had -6.62 kcal·mol−1. ADAMTS-5-1Zn2+ had the best glide energy of -70.28 kcal·mol−1, whereas ADAMTS-5-0Zn2+ had -66.02 kcal·mol−1. ADAMTS-5-1Zn2+ had the best glide e-model of -108.484, whereas ADAMTS-5-0Zn2+ had -93.81. The RMSD value for ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+ was 0.48Å. These results confirmed that the absence of zinc decreased the binding affinity of ADAMTS-5 with its inhibitor whereas the presence extended the docking energy range and strengthened the binding affinity. Per-residue interaction study, MM-GBSA and Molecular Dynamics showed that all the four complexes underwent extensive structural changes whereas the complex with zinc was stable throughout the simulation period.

Keywords: Aggrecanase, docking, MM-GBSA, molecular dynamics, zinc. 1. INTRODUCTION Zinc is one of the important metals present in many metalloproteins. Zinc plays a vital role in many metalloproteinases as they organize the structure, regulate the function and stabilize the folded protein and make the protein biologically active. Its transition can make a major change to the protein and its binding affinity. Zinc -ligand interactions also differ from time to time in a biological function. The effect of the ligands on the metal ion or the effect of the metal ion on the ligands is important for the functionality of the protein. Zinc may activate or inactivate a protein by switching the protein by acting as acceptor or donor. Zinc and Calcium ions play important role in modulating ADAMTS-13 activity [1]. In vitro studies proved that the removal and addition of zinc are possible, therefore in this study in silico methods were used to study the effect of zinc in ADAMTS. In spite of the metal ion and its position of ligation to the protein, there are other features which can change the stability of the protein. The metal is bound by a shell of hydrophilic atomic groups and this hydrophilic shell is embedded within a larger shell of hydrophobic atomic groups. Identifying ligand, which has binding affinity to the zinc ion is challenging, because of *Address correspondence to this author at the Department of Bioinformatics, SRM University, Kattankulathur, Tamilnadu- 603203, India; Tel: +919840506562; E-mail: [email protected] 1875-5402/14 $58.00+.00

polarization, charge transfer, multiple coordination geometries, and lack of accurate force fields. These limitations have slowed down the computer aided drug designing of novel potent and selective ligands for zinc metalloproteins. The bonded model and the non-bonded model are two basic ways to model the zinc ion using a purely classical potential function. In the bonded models, explicit bond and angle terms are introduced into the potential energy function to account for interactions between the metal and the protein and ligand atoms. In the alternative non-bonded model, only the van der Waals and electrostatic terms are included for the zinc ion [2]. Therefore in this study, the influence of zinc over the binding mode of aggrecanase is studied to decide whether a ligand should bind to zinc or not to be functional. Metalloproteinases are zinc dependent enzymes present in all species. These enzymes have catalytic zinc which is mostly bonded to three histidine’s. The presence of zinc named the protein as Zincins. This is further classified as gluzincins, aspzincins, and metzincins. The conserved Met residue at the active site named the family as Metzincin. Matrix metalloproteinases (MMPs), a disintegrin and metalloproteinases (ADAMs), and ADAM proteases with thrombospondin motifs (ADAMTSs) comes under Metzincin. 30 ADAM genes and 18 ADAMTS genes are found in humans. These enzymes play major role in degradation, maintenance and generation of new tissue [3].

© 2014 Bentham Science Publishers

892

Combinatorial Chemistry & High Throughput Screening, 2014, Vol. 17, No. 10

ADAMTS proteins lack transmembrane domain, therefore they are not membrane-anchored proteins. “They have a signal peptide, a prodomain, a metalloproteinase domain, a disintegrin domain, a thrombospondin type I motif, a spacer domain, and a second thrombospondin module of a variable number of repeats at the C-terminal region. ADAMTSs possess a conserved thrombospondin type 1 like repeat that is believed to function as a binding domain for sulfated glycosaminoglycans, which is present on proteoglycans” [4]. “Osteoarthritis (OA) is a disease of the joints that affect most of the elderly population, the main features of which are cartilage damage, subchondral bone remodeling, joint space narrowing, synovial inflammation and osteophyte formation at the joint margins” [5]. Damage to aggrecan causes degradation of collagen which leads to OA. Aggrecan is a proteoglycan, which provides elasticity to the collagen network [6]. This gets damaged when there is increased proteolytic cleavage which is caused by aggrecanases. “ADAMTS-4 and ADAMTS-5 are the most efficient aggrecanases and have generally been considered to play a role in the pathological mechanisms of OA” [7]. “The ADAMTS-4 and ADAMTS-5 proteinases have been implicated in aggrecan degradation in OA, on the basis of mRNA and protein expression. Suppressing the expression of ADAMTS-4 and ADAMTS-5 in human cartilage significantly decreased the aggrecan release [8]. Many experiments are conducted to check the significance of ADAMTS-4 and ADAMTS-5. When ADAMTS-4 and ADAMTS-5 were deleted ex vivo, it showed protection against the degradation of aggrecan and decreased the severity of OA in animal models. Therefore, from literature study show that ADAMTS-4 and ADAMTS-5 play a major role in controlling OA [9]. In this work, wild and mutant type of Aggrecanase I and Aggrecanase II were constructed by including and excluding the catalytic Zn2+ in the catalytic domain. Finally the flexibility of the four models and the influence of zinc on the binding affinity of the protein with the ligand is studied by Molecular Dynamic Simulation. When a inhibitor is designed targeting zinc metalloproteases, this study will help to identify a zinc binding-group on nonzinc binding-group can be selected with respect to facilitating inhibitor orientation, and hence binding, in the enzyme’s active site [10]. Aggrecanase-1 (ADAMTS-4) and Aggrecanse-2 (ADAMTS-5) are the members of ADAMTS family which are expressed in OA cartilage [11]. “Both enzymes were purified from interleukin (IL)-1-stimulated bovine nasal cartilage and are identified by following their activity with an assay using the neo-epitope antibody BC-3, which detects the new N-terminus, ARGS, formed by specific cleavage at the Glu373–Ala374 bond in the IGD of the aggrecan core protein. ADAMTS-4 and ADAMTS-5 cleave at least four other sites in the chondroitin sulfate-rich CS-2 region of bovine aggrecan: GELE1480 ~ GRGD, KEEE1667 ~ GLGS, TAQE1771 ~AGEG, and VSQE1871 ~ LGQR” [12]. “The structures of the two closely related proteins ADAMTS-4 and ADAMTS-5 (with 48% sequence identity outside the pre- and prodomains) present a common fold comprising an N-terminal metalloprotease domain and a C-terminal disintegrin-like domain joined through an extended linker and stabilized by four disulfide bonds. The metalloprotease

Suganya et al.

catalytic domain (residues 214–428 in ADAMTS-4 and residues 264–476 in ADAMTS-5) contains the active site, the catalytic zinc, and two calcium-binding sites and has an α/β structure characteristic of the metzincin family” [13]. 2. MATERIALS AND METHODS 2.1. Protein Structure The Crystal structure of ADAMTS-4 (PDB ID- 2RJP) was obtained from Protein Data Bank [14]. ADAMTS-4 has single calcium and zinc moiety, which is bonded with three histidine residues His361, HIS365 and HIS 371. Likewise, the Crystal structure of ADAMTS-5 (PDB ID- 2RJQ) has two molecules of Zinc, Calcium, and Chlorine. From the 2 zinc moiety, one zinc bonds with HIS410, HIS414, HIS420, and this is the core zinc, while the other zinc bonds with HIS374 and 2 chlorine moieties. The distance between the 2 zinc is 11.90 nm. In our study the crystal structure with zinc is considered as wild type and the crystal structure without zinc is considered as the mutant type. Mutations were made manually by deleting the zinc moiety using Protein Preparation wizard in Schrodinger [15]. Mutation study was carried out in ZINC metal atom to identify the interactions and amino acid residues that are essential for binding, which led to an enhancement or decline in the binding affinity. To clarify the role zinc played in electrostatic and hydrophobic interactions, mutations were introduced into the system after analyzing the Wild type interaction between aggrecanase and its ligands in the presence of zinc. A single point deletion mutation of zinc metal was performed. The mutation led to the loss of Zinc and proved the role in the interaction [16]. 2.2. Protein Preparation Protein preparation was done using the protein preparation wizard in maestro. For the wild type (ADAMTS4-1Zn2+ and ADAMTS-5-1Zn2+) the zinc atom was not deleted, whereas all the other co-crystallized atoms like calcium and chlorine were deleted. “1” in front of zinc represents presence. For the mutated type (ADAMTS-40Zn2+ and ADAMTS-5-0Zn2+) zinc was also deleted. “0” in front of zinc represents absence. Crystal ligand was not deleted because it would be used for grid generation. Finally, all the four models developed were optimized and then minimized using OPLS2005 force field. 2.3. Ligand Preparation Ligand preparation was performed using LigPrep, which generates variations on the ligands, eliminates unwanted and optimizes the ligands. Crystal ligands from PDB were used for this study. 886 (N-({4'-[(4-isobutyrylphenoxy) methyl] biphenyl- 4-yl} sulfonyl)-D-valine) is the crystal ligand for ADAMTS-4 and BAT (4-(N-Hydroxyamino)-2r-Isobutyl-2s(2-Thienylthiomethyl) Succinyl-L-Phenylalanine-N-Methylamide) is the crystal ligand binding in the active site of ADAMTS-5. NAG (N-Acetyl-D-Glucosamine) is also bound to ADAMTS-5 but not in the active site therefore that was deleted in the protein preparation wizard. These ligands were taken for energy minimization using OPLS-2005 force field.

Influence of Zn2+ on the Binding Modes of Aggrecanase


2.4. Receptor Grid Generation Receptor grid generation was performed using glide to decide the location and size of the active site. The shape and properties of the receptor are represented on a grid to ensure that possible actives are not missed. The centroid of crystal ligand (886 and BAT) is used to generate grids using default value of protein atom scaling (1.0) Å within a cubic box. The force field used for grid generation is the OPLS_2005 [17, 18].

893

energies of unliganded receptor and ligands. ΔGsol is the difference in the GBSA solvation energy of the receptorligand complex and the sum of energies of unliganded receptor and ligands. ΔGSA is the difference in the surface area energies for the receptor ligand complex and the sum of energies of unliganded receptor and ligands [21]. 2.8. Molecular Dynamics (MD) Simulation

Extra precision (XP) glide docking was performed for ADAMTS-4 (ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+) and ADAMTS-5 (ADAMTS-5-1Zn2+ and ADAMTS-50Zn2+) separately using GLIDE program. Glide starts with extensive conformational search for each input ligand followed by a comprehensive search for possible locations and orientations of each core conformation over the active site of the protein. The generated ligand poses pass through a series of hierarchical filters that evaluate the spatial fit of each pose to the defined binding site and examine the ligandreceptor interactions using GlideScore. In the next step, the set of candidate molecules are subjected to energy minimization and the energy minimized poses are re-scored with the standard Glide Score function.

The Highest scoring pose from docking was subjected to energy minimization by Protein preparation wizard. Molecular Dynamics Simulation was performed for the minimized protein ligand complex by Desmond. This uses OPLS-2005 force field. SPC (simple point charge) method was used for the solvent model [22-24]. SPC water molecules were then added to the protein ligand complex with orthorhombic dimensions 10Å×10Å ×10Å approximately, to cover whole surfaces of the complexes. The systems were minimized with a maximum of 2000 steps and pre-equilibrated using the default relaxation protocol of Desmond. The prepared system was used for molecular simulation for 10 ns. The structural modification and dynamic activities of the protein in the presence and absence of Zinc were analyzed by calculating the Energy, root mean square deviation (RMSD), root mean square fluctuation (RMSF) and radius of gyration (ROG).

2.6. Per-Residue Interaction

3. RESULTS AND DISCUSSION

Per- residue interaction is done to study the interactions between a ligand and receptor residues. This per-residue interaction scores include Coulomb, Van Der Waals, and Hydrogen bonding scores. The sum of these scores (interaction energy, Eint), and the distances are calculated between the ligand and the particular amino acid residues. These values can be used to identify essential amino acid residues favoring the protein ligand interaction. This can be done by enabling the ‘Write per-residue interaction scores’ option in Ligand Docking [19, 20].

3.1. Docking Analysis

2.7. MM-GBSA

The two models were docked with co-crystal ligand 886. The Glide XP score glide energy and glide emodel scores were compared for both the models. Hydrogen bond is formed with Gly 331 and Leu 330 in both the models. But for ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+, the bond is formed with the oxygen atom of sulphur dioxide and the oxygen atom of acetic acid respectively in the ligand 886. ADAMTS-4-1Zn2+ also bonds with Gln362. Hydrogen bond distance is showed in Table 2. ADAMTS-4-1Zn2+ had the best Glide score of -12.66, whereas ADAMTS-4-0Zn2+ had -11.69 kcal·mol−1. ADAMTS-4-1Zn2+ had the best glide energy of -72.29 kcal·mol−1, whereas ADAMTS-4-0Zn2+ had -68.44 kcal·mol−1. ADAMTS-4-1Zn2+ had the best glide

2.5. Molecular Docking

Prime MM-GBSA (Molecular Mechanics/Generalized Born Surface Area) predicts the free energy of binding for the receptor inhibitor complex. The MM-GBSA approach employs molecular mechanics, the generalised Born model and solvent accessibility. The binding free energy of each ligand was calculated using following equation:

DG bin = DE mm +DG sol +DG SA where ΔEmm is the difference in the minimised energies between the receptor ligand complex, and the sum of Table 1.

The objective of this study was aimed at exploring the binding affinity of the crystal ligand with aggrecanases in the presence and absence of zinc. Docking results of ADAMTS4 and ADAMTS-5 in the presence and absence of Zinc are shown in Table 1. The binding of co-crystal ligand in the presence and absence of zinc is discussed below: 3.1.1 Comparing ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+

Glide score, emodel and glide energy for ADAMTS-4 and ADAMTS-5 ligands with and without zinc.

Protein Complex

Glide Score

Glide Energy

Glide Emodel

Electro

Lipo eVdW

ADAMTS-4withzinc

-12.65

-72.28

-116.33

-4.62

-7.793

ADAMTS-4withoutzinc

-11.68

-68.44

-104.26

-0.73

-8.7

ADAMTS-5withzinc

-8.32

-70.88

-108.48

-2

-3.36

ADAMTS-5withoutzinc

-6.62

-66.02

-93.81

-1.68

-2.89

RMSD(Å)

1.913

0.479

894


emodel of -116.34, whereas ADAMTS-4-0Zn2+ had -104.264 (Fig. 1a, b). The RMSD value for ADAMTS-41Zn2+ and ADAMTS-4-0Zn2+ is 1.9Å. These results suggest that the absence of zinc decreases the binding affinity of ADAMTS-4 with its inhibitor. The glide score, emodel and glide energy for both forms are tabulated in Table 1. Both the models were aligned together to check their binding coordination and Fig. (2) shows that there are differences in the binding mode in the presence and absence of zinc. Perresidue interaction scores had high variation in the presence and absence of Zinc. Interaction energy (Eint) is plotted for the catalytic domain residues to note the changes. Negative value means attraction whereas positive value means repulsion. Per-residue interaction scores had high variation in the presence and absence of Zinc [25]. The Fig. (3) shows that there is a change in the interaction energy in the presence and absence of zinc and this shows that there are high fluctuations proving that Zinc plays a major role in maintaining the interactions between the protein and the ligand [26]. Catalytic site residues of ADAMTS4 are checked for interaction in which residue Val390, Val394 and Met395 from S1 loop produces significant interaction energy of above −3.0 kcal·mol−1 to which the complex formation could be connected. Asp399 goes on the positive side indicating the repulsion energy it has produced. Pro393 has a Table 2.

major variation in the absence of zinc. In the presence of zinc, Pro393 has an interaction energy of -1.0 kcal·mol−1 whereas in the absence of zinc the interaction energy is +1.0 kcal·mol−1. This shows that the residue shows repulsion towards the ligand. Prime MM-GBSA was carried out to calculate the drug receptor binding energy calculation. The binding energies of the different compound receptor complexes were in accordance with the Glide score. ADAMTS4-1Zn2+ had the best Binding score of -112.90 kcal·mol−1 which was contributed mostly by Van Der Waals interaction with a coulomb energy of -19.74 kcal·mol−1 and a Generalized Born electrostatic solvation energy (Bind Solv GB) of 23.10 kcal·mol−1. For ADAMTS-4-0Zn2+ the Binding score was -92.82 kcal·mol−1, which was also contributed mostly by Van Der Waals interaction. However, the coulomb energy was 34.33 kcal·mol−1 which shows that the protein had repulsion from the ligand in the absence of zinc, and the Bind Solv GB was -12.72 kcal·mol−1 (Table 3). 3.1.2. Comparing ADAMTS-5-1Zn2+ and ADAMTS-50Zn2+ The two models were docked with co-crystal ligand BAT. The Glide XP score, glide energy and glide emodel scores were compared for both the models. Hydrogen bond

Interactions and distances between ADAMTS-4 and ADAMTS-5 with its inhibitors. Complex

Donor---H---Acceptor

Distance (Å)

Gly331---N-H---O(886)

2.29

ADAMTS-4withzinc

Leu330---N-H---O(886)

1.99

Gln362---N-H---O(886)

2.22

ADAMTS-4withoutzinc

ADAMTS-5withzinc

ADAMTS-5withoutzinc

Table 3.

Suganya et al.

Gly331---N-H---O(886)

2.32

Leu330---N-H---O(886)

2.06

Leu379---N-H---O(BAT)

2.07


1.78

(BAT)---N-H---O Asp377

2.90

(BAT)---N-H---O Ser441

2.03

(BAT)---O-H---O Glu411

1.88

(BAT)---N-H---O Glu411

2.21

(BAT)---N-H---O Gly380

2.12


1.82


2.02

(BAT)---N-H---O Asp377

1.86

(BAT)---N-H---O Ser441

1.93

(BAT)---O-H---O Glu411

2.06

(BAT)---N-H---O Glu411

2.08

Docking score and energy of binding of receptor inhibitor complex calculated using Prime MM-GBSA method.

Title

Docking Score (kcal/mol)

MMGBSA dG Bind

MMGBSA dG Bind Coulomb

MMGBSA dG Bind Covalent

MMGBSA dG Bind Solv GB

MMGBSA dG Bind vdW

ADAMTS-4 withzinc

-12.66

-112.90

-19.74

6.98

23.10

-64.97

ADAMTS-4withoutzinc

-11.69

-92.82

34.33

14.05

-12.72

-69.58

ADAMTS-5withzinc

-8.33

-102.68

-60.73

0.68

54.74

-55.10

ADAMTS-5withoutzinc

-6.63

-88.26

-42.73

3.90

39.37

-47.68



(a)

895

(b)

Fig. (1). Interaction diagram of ADAMTS-4-1Zn2+ (a) and ADAMTS-4-0Zn2+ (b) with ligand 886.

is formed with LEU379, LEU443, and SER441 in both the models, but for ADAMTS-5-1Zn2+ the bond is also formed with GLY380 (Fig. 4a, b). ADAMTS-5-1Zn2+ had the best Glide score of -8.32 kcal·mol−1, whereas ADAMTS-5-0Zn2+ had -6.62 kcal·mol−1. ADAMTS-5-1Zn2+ had the best glide energy of -70.28 kcal·mol−1, whereas ADAMTS-5-0Zn2+ had -66.02 kcal·mol−1. ADAMTS-5-1Zn2+ had the best glide emodel of -108.484, whereas ADAMTS-5-0Zn2+ had -93.81. The RMSD value for ADAMTS-5-1Zn2+ and ADAMTS-50Zn2+ is 0.48Å. Both the models were aligned together to check their binding co-ordination and Fig. (5) shows that there are differences in the binding mode in the presence and absence of zinc. These outcomes imply that the absence of zinc decreases the binding affinity of ADAMTS-5 with its inhibitor. In the catalytic site of ADAMTS-5 SER441, Ile442 and Leu443 contribute more to the interaction between the protein and ligand and the fluctuations in interaction energy show the influence of zinc (Fig. 6). MM-GBSA binding energies of the different compound receptor complexes were in accordance to the Glide score and all the scores of

ADAMTS-5-0Zn2+ were less when compared to the scores of ADAMTS-5-1Zn2+ (Table 3). 3.2. Molecular Dynamics Results and Discussion Aggrecanase binding with the inhibitor is a vibrant process. Stable binding is necessary to study the interaction between protein and ligand. Molecular Dynamics will validate the binding affinity and structure stability. The four structures were studied to check the influence of Zinc over aggrecanase. RMSD, RMSF, ROG and H-bond distance is calculated to check the structural stability. 3.2.1. Root Mean Square Deviation (RMSD) RMSD is a measure of overall stability of any protein system. RMSD of the Cα atoms were evaluated over the 10 ns time period. RMSD for ADAMTS-4-1Zn2+ was stable throughout the simulation and was within 1 Å initially and the deviation increased along with time to 2.4 Å till 4000ps. And the deviation is constantly maintained at 1.5 Å

Fig. (2). Docking pose of ADAMTS-4-1Zn2+ (blue) and ADAMTS-4-0Zn2+ (green) aligned together.

896


Suganya et al.

Fig. (3). Per-residue binding energy decomposition of the two inhibitor-protein complexes(ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+) showing total residue contribution to the total binding energy.

Fig. (4). Interaction diagram of ADAMTS-5-1Zn2+ (a) and ADAMTS-5-0Zn2+ (b) with ligand BAT.

throughout the simulation. Initially, in ADAMTS-4-0Zn2+, the deviation was 1.8 Å till 4000ps and the deviation kept oscillating throughout the simulation to 3 Å (Fig. 7). It is inferred from the RMSD plot of ADAMTS-5-1Zn2+ that the

deviation of the Cα atoms was fluctuating throughout the simulation between 1.5 to 2.5Å, whereas ADAMTS-5-0Zn2+ deviation followed the deviation plot of ADAMTS-5-1Zn2+ with a difference of 1 Å (Fig. 8).



897

Fig. (5). Docking pose of ADAMTS-5-1Zn2+ (blue) and ADAMTS-5-0Zn2+ (green) aligned together (colored illustration is available online).

Fig. (6). Per-residue binding energy decomposition of the two inhibitor-protein complexes(ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+) showing total residue contribution to the total binding energy.

Fig. (7). RMSD of the backbone atoms of ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+ over a time period of 10 ns.

898


Suganya et al.

Fig. (8). RMSD of the backbone atoms of ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+ over a time period of 10 ns.

3.2.2. Root Mean Square Fluctuation (RMSF)

3.2.3. Measure of the Radius of Gyration (ROG)

RMSF is analyzed to calculate the fluctuation of the residues over time and it is also used as a measure of protein structural integrity. RMSF of most of the residues were within the limit of 2.5 Å. Fluctuations for a few residues exceeded 3 Å. The lower atomic fluctuations of the backbone atoms indicated small conformational changes. RMSF for ADAMTS-4-0Zn2+ was unstable after 4 ns and crossed 5 Å in many places. The RMSF values of N-terminal regions residue segment 214-225 and C-terminal region residue segment 470-480 revealed that ADAMTS-4 experienced larger side chain rearrangements in these regions (Fig. 9). ADAMTS-5-1Zn2+ was stable and was within 1.50 Å to 2.50 Å, whereas ADAMTS-5-0Zn2+ was steep and crossed 3 Å in a few places. The RMSF values of N-terminal regions residue 264 and C-terminal region residue segment 530-540 revealed that ADAMTS-5 experienced larger side chain rearrangements in these regions (Fig. 10).

The ROG shows the compactness of the protein structure. Therefore, variation in the ROG values illustrates the variation in the compactness in the protein structure throughout the simulation. ADAMTS-4-1Zn2+ had an ROG value of around 19.2 Å before simulation. The ROG fluctuated slightly during the initial 4 ns of simulations followed by higher fluctuations up to 8 ns. After 7 ns, the protein ROG settled around 19.2 Å. The ROG graph indicates that there has been an increase in the inter-atomic distances of protein during the simulations, and a consistent ROG of 19.2Å during the last 8 ns confirmed the development of stable protein conformations. ADAMTS-40Zn2+ had an ROG value of around 19 Å before the simulation. The ROG fluctuated slightly during the initial 1 ns of simulation followed by higher fluctuations up to 8 ns. Higher ROG of simulated structures than that of the initial structure indicated that protein had expanded during the MD. The ROG graph indicates that there has been an increase in the inter-atomic distances of protein during the simulations; in a few places, the ROG has crossed 19.3 which shows the

Fig. (9). RMSF of all residues of ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+ during the MD simulations.



899

Fig. (10). RMSF of all residues of ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+ during the MD simulations.

structure is unstable in the absence of zinc (Fig. 11). ADAMTS-5-1Zn2+ had an ROG value of around 19.4 Å before the simulation. The ROG fluctuated slightly during the initial 2 ns of simulations followed by higher fluctuations up to 10ns. After 5 ns, the protein ROG settled around 19.5 Å. The ROG graph indicates that there has been an increase in the inter-atomic distances of protein during the simulations and a consistent ROG of 19.5Å during the last 5 ns confirmed the development of stable protein conformations. ADAMTS-5-0Zn2+ had an ROG value of around 19.1 Å before the simulation. The ROG fluctuated slightly during the initial 1ns of simulations followed by higher fluctuations up to 10 ns. Higher ROG of simulated structures than that of the initial structure indicated that protein had expanded during the MD. ROG graph indicates that there has been an increase in the inter-atomic distances of protein during the simulations; in a few places, the ROG has crossed 19.7 which show the structure is unstable in the absence of zinc (Fig. 12).

3.2.4. H-Bonds Distance The total number of H-bonds formed throughout the simulation was ranging from 2 to 3 for ADAMTS-4-1Zn2+. In the case of ADAMTS-4-0Zn2+, initially, the number of Hbonds was 0-5 till 5000ps and after which the number of Hbonds fluctuated between 0 and 1 throughout the simulation indicating that the ligand has a steady interaction in the presence of Zinc in ADAMTS-4 (Fig. 13). In ADAMTS-51Zn2+, the number of H-bonds till 500ps varied between 6 and 8 after which the number reduced to 4 till 2000ps. Later, the number of H- bonds was at 3 throughout the simulation. For ADAMTS-5-0Zn2+, for the first 5000ps the number of H- bonds was between 0 and 3. The next 5000ps showed no drastic change in the number of H- bonds that was between 0 and 1 signifying that lesser the H- bonds lower the stability in the absence of Zinc in ADAMTS-5 (Fig. 14).

Fig. (11). ROG of ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+ over a time period of 10 ns.

900


Suganya et al.

Fig. (12). ROG of ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+ over a time period of 10 ns.

Fig. (13). H-bonds of ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+ with ligand 886.

fluctuation in the next 5000ps indicating that the ligand establishes a stable interaction with Leu 330 only in the presence of zinc (Fig. 15).

3.2.5. Comparison of Distance Between ADAMTS-4 with Ligand 886 and Residues 330, 331 in the Presence and Absence of Zinc The influence of Zinc on the binding mode was studied in detail by analyzing the residues which bond to ligand 886 around zinc. For ADAMTS-4, Leu330 and Gly331 is around zinc, which bond with ligand. Therefore, these two residues are taken for comparison in the presence and absence of zinc. (i)

Distance Plot between Leu330 and Ligand 886 in the presence and absence of Zinc: In the presence of Zinc, the distance between Leu 330 and ligand was between 2 to 3 Å throughout the simulation. The distance between Leu 330 and ligand in the absence of Zinc during first 5000ps was 2 to 4 Å. The distance inconsistently increased and showed

(ii)

Distance Plot between Gly331 and Ligand 886 in the presence and absence of Zinc: The distance between Gly 331 and ligand with ADAMTS-4 in the presence of Zinc was between 2 to 4 Å during first 6000s after which the distance increased to 5 Å and remained the same throughout the simulation. In the absence of Zinc, the distance between Gly 331and the ligand was between 2 to 6 Å during the first 5000ps. In the later 5000ps, the distance drastically increased and was between 5 and 9 Å (Fig. 16). The distance fluctuation of ADAMTS4 in the absence of Zinc suggests that the ligand interaction can be widely disturbed when the Zinc atom is absent.



901

Fig. (14). H-bonds of ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+ with ligand BAT.

Fig. (15). Distance plot between Leu330 and ligand for ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+.

3.2.6. Comparison of Distance Between ADAMTS-5 with Ligand BAT and Residues 411 in the Presence and Absence of Zinc For ADAMTS-5, Glu411 is around zinc, which bonds with the ligand BAT. Therefore, this residue is taken for comparison in the presence and absence of zinc. (i)

Distance plot between Glu411 and ligand BAT in the presence and absence of zinc: In the presence of zinc, the distance between the ligand and Glu 411 was between 2 to 3 Å till 8000ps. Later, the residual fluctuation increased to 4 Å till the

end of simulation. In the absence of Zinc, first 500ps showed a distance of 2 Å after which the distance oscillated between 2 to 4 Å till 2000ps. Soon after, the residual fluctuation increased to 4 to 6.5 Å till the end of simulation (Fig. 17). Thus, it can be inferred that in the presence of Zinc, the ligand-residue interaction was stable when compared to that in the absence of Zinc, where the distance increased randomly. Therefore, after MD simulation it is confirmed that the presence of Zinc in aggrecanase stabilizes the structure and increases the binding affinity of the protein with its ligand.

902


Suganya et al.

Fig. (16). Distance Plot between GLY331 and ligand for ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+.

Fig. (17). Distance plot between Glu411 and ligand for ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+.

CONCLUSION

CONFLICT OF INTEREST

The binding process of aggrecanase (ADAMTS-4 and ADAMTS-5) was studied in the presence and absence of Zinc. During docking, Aggrecanase had best binding affinity and more H- bond formation in the presence of zinc when compared with the complex without zinc. Following perresidue interaction study, MM-GBSA and Molecular Dynamics showed that all the four complexes underwent big structural changes whereas the complex with zinc stabilized at the end of simulation.

The authors confirm that this article content has no conflict of interest. ACKNOWLEDGEMENTS I thank SRM University for their constant support and funding. REFERENCES [1]

Anderson, P.J.; Kokame, K.; Sadler, J.E. Zinc and calcium ions cooperatively modulate ADAMTS13 activity. J. Biol. Chem., 2006, 281(2), 850-857.

Influence of Zn2+ on the Binding Modes of Aggrecanase [2]

[3] [4] [5] [6] [7] [8] [9] [10]

[11]

[12] [13]

[14] [15]


Jain, T.; Jayaram, B. Computational protocol for predicting the binding affinities of zinc containing metalloprotein-ligand complexes. Proteins, 2007, 67(4), 1167-1178. Gomis-Rüth, F.X. Structural aspects of the metzincin clan of metalloendopeptidases. Mol. Biotechnol., 2003, 24(2), 157-202. Porter, S.; Clark, I.M.; Kevorkian, L.; Edwards, D.R. The ADAMTS metalloproteinases. Biochem. J., 2005, 386(Pt 1), 15-27. Nigel, A.; Nevitt, M.C. "Osteoarthritis: epidemiology. Best Pract. Res. Clin. Rheumatol., 2006, 20(1), 3-25. Vertel, B.M.; Aggrecan, RA. Proteoglycans: Structure, biology and molecular interactions. Thesis, The Stony Book University: New York, 2000. pp. 343-377. Huang, K.; Wu, L.D. Aggrecanase and aggrecan degradation in osteoarthritis: a review. J. Int. Med. Res., 2008, 36, 1149-1160. Tang, B.L. ADAMTS: a novel family of extracellular matrix proteases. Int. J. Biochem. Cell Biol., 2001, 33(1), 33-44. Fosang, A.J.; Rogerson, F.M.; East, C.J.; Stanton, H. ADAMTS-5: the story so far. Eur. Cell Mater., 2008, 15, 11-26. Maingot, L.1.; Leroux, F.; Landry, V.; Dumont, J.;, Nagase, H.; Villoutreix, B.; Sperandio, O.; Deprez-Poulain, R.; Deprez, B. New non-hydroxamic ADAMTS-5 inhibitors based on the 1,2,4-triazole3-thiol scaffold. Bioorg. Med. Chem. Lett., 2010, 20(21), 62136216. Yamanishi, Y.; Boyle, D.L.; Clark, M.; Maki, R.A.; Tortorella, M.D.; Arner, E.C.; Firestein, G.S. Expression and regulation of aggrecanase in arthritis: the role of TGF-beta. J. Immunol., 2002, 168(3), 1405-1412. Nagase, H.; Kashiwagi, M. Aggrecanases and cartilage matrix degradation. Arthritis Res. Ther., 2003, 5(2), 94-103. Mosyak, L.; Georgiadis, K.; Shane, T.; Svenson, K.; Hebert, T.; McDonagh, T.; Mackie, S.; Olland, S.; Lin, L.; Zhong, X.; Kriz, R.; Reifenberg, E.L.; Collins-Racie, L.A.; Corcoran, C.; Freeman, B.; Zollner, R.; Marvell, T.; Vera, M.; Sum, P.E.; Lavallie, E.R.; Stahl, M.; Somers, W. Crystal structures of the two major aggrecan degrading enzymes, ADAMTS4 and ADAMTS5. Protein Sci., 2008, 17(1), 16-21. http://www.rcsb.org/pdb/home/home.do. Sastry, G.M.; Adzhigirey, M.; Day, T.; Annabhimoju, R.; Sherman, W. Protein and ligand preparation: parameters, protocols, and influence on virtual screening enrichments. J. Comput. Aided Mol. Des., 2013, 27(3), 221-234.

Received: February 21, 2014

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25] [26]

903

Sun, T.Y.; Wang, Q.; Zhang, J.; Wu, T.; Zhang, F. TrastuzumabPeptide interactions: mechanism and application in structure-based ligand design. Int. J. Mol. Sci., 2013, 14(8), 16836-16850. Friesner, R.A.; Banks, J.L.; Murphy, R.B.; Halgren, T.A.; Klicic, J.J.; Mainz, D.T.; Repasky, M.P.; Knoll, E.H.; Shelley, M.; Perry, J.K.; Shaw, D.E.; Francis, P.; Shenkin, P.S. Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J. Med. Chem., 2004, 47(7), 1739-1749. Halgren, T.A.; Murphy, R.B.; Friesner, R.A.; Beard, H.S.; Frye, L.L.; Pollard, W.T.; Banks, J.L. Glide: a new approach for rapid, accurate docking and scoring. 2. Enrichment factors in database screening. J. Med. Chem., 2004, 47(7), 1750-1759. Moreira, I.S.; Fernandes, P.A.; Ramos, M.J. Bahadur, R.P.; Chakrabarti, P.; Rodier, F.; Janin, J. A dissection of specific and non-specific protein-protein interfaces. J. Mol. Biol., 2004, 336, 943-955. Bahadur, R.P.; Chakrabarti, P.; Rodier, F.; Janin, J. A dissection of specific and non-specific protein-protein interfaces. J. Mol. Biol., 2004, 336, 943-955. Kumar, R.; Malla, P.; Verma, A.; Kumar, M. Design of potent human steroid 5a-reductase inhibitors: 3D-QSAR CoMFA, CoMSIA and docking studies. Med. Chem. Res., 2013, 22, 45684580. Shivakumar, D.; Williams, J.; Wu, Y.; Damm, W.; Shelley, J.; Sherman, W. Prediction of absolute solvation free energies using molecular dynamics free energy perturbation and the OPLS force field. J. Chem. Theory Comput., 2010, 6, 1509-1519. Guo, Z.; Mohanty, U.; Noehre, J.; Sawyer, T.K.; Sherman, W.; Krilov, G. Probing the alpha-helical structural stability of stapled p53 peptides: molecular dynamics simulations and analysis. Chem. Biol. Drug Des., 2010, 75(4), 348-359. Bowers, K. J.; Chow, E.; Xu, H.; Dror, R. O.; Eastwood, M. P.; Gregersen, B. A.; Klepeis, J. L.; Kolossvary, I.; Moraes, M. A.; Sacerdoti, F. D.; Salmon, J. K.; Shan, Y. Shaw, D. E. Scalable Algorithms for Molecular Dynamics Simulations on Commodity Clusters. Proceedings of the ACM/IEEE Conference on Supercomputing (SC06), Tampa, Florida, 2006, pp.11-17. Sun, T.Y.; Wang, Q.; Zhang, J.; Wu, T.; Zhang, F. TrastuzumabPeptide interactions: mechanism and application in structure-based ligand design. Int. J. Mol. Sci., 2013, 14(8), 16836-16850. Haliloglu T, Erman B. Analysis of correlations between energy and residue fluctuations in native proteins and determination of specific sites for binding. Phys. Rev. Lett., 2009, 102(8), 088103.

Revised: October 20, 2014

Accepted: November 12, 2014