Structure - based drug design is usually referred to as the designing of the drug based on the knowledge of three dimensional structure of biological target. For the rational drug design, structural information on both protein and ligand is necessary. In this approach, all of the available structural information is considered to improve the ligand's affinity by optimizing its interaction with the protein target. Recent studies suggest that the influence of protein flexibility and mobility are fundamental properties to be included in the drug design approaches. As Matrix Metalloproteinases (MMP's) possess mobile active sites, designing of their inhibitors is difficult. This paper mainly focuses on the influence of mobility of active sites of MMP on its drug design. Docking of MMP2 and MMP9 with its inhibitor Marimastat showed difference in interaction energy in case of flexibility of active site. From these observations, it is clear that the mobility is an important facot which should be taken into consideration in case of structure-based drug design for MMP.
Keywords: Structure - based drug design, mobility of active site, docking, MMP
INTRODUCTION
Structure based drug design or direct drug design depends on the three dimensional structure of biological target. In this technique, candidate drugs that can bind with high selectivity and affinity to the target can be designed using structure of target. This method can be divided into two categories namely ligand-based drug design and receptor -based drug design. Former is about finding ligands for a given receptor in which a number of ligands are screened to find those fitting the binding pocket of receptor. Latter is by building ligands within the constraints of binding pocket of receptor and thus novel structure can be obtained. This paper deals with structure based drug design for MMP's.
Matrix Metalloproteinases (MMP's) are zinc dependent endopeptidases involved in extracellular matrix and basement membrane degradation. This family of protein consists of about 26 endopeptidases with conserved domain structures. MMP's have role in different types of pathology mainly tissue destruction in rheumatoid arthritis, osteoarthritis, gastric ulcer, cancer invasion and metastasis, fibrosis in liver cirrhosis and multiple sclerosis, weakening of matrix in aortic aneurysm and restenotic lesion. MMP's are classified based on the substrate specificity and cellular localization.They are involved in the cleavage of cell surface receptors,release of apoptotic ligands. They also play major role in apoptosis, differentiation, cell proliferation, migration and host defence. Synthetic MMP inhibitors can be developed by structure-based drug design. Binding pockets or active site of MMP possess an inherent mobility. Thus for the design of inhibitors for various MMP this mobility is taken into consideration.
Mobility of active site is the origin for receptor plasticity and enables the binding partners or ligands to conformationally adapt to each other. Receptor mobility is not only limited to steric complementarity, but also additional energetic and entropic contributions may arise due to the binding affinity. Mobility of active site of receptor leads to a disfavorable reorganization energy that may be large, even for preorganized binding sites. Configurational entropy contributions result from the changes in the receptor flexibility upon complex formation. If flexibility is transferred to other protein parts, it leads to the redistribution of protein configurational entropy. Configurational entropy can be computed by MD simulation. Carrying out interactional studies between MMP and its inhibitor will help to study about the influence of protein mobility on drug design.
THEORY
Molecular Dynamics (MD) simulation is one of the most accurate computational techniques used in the field of macromolecular computation. MD simulation has been utilized for the determination of receptor plasticity which contributes to structure-based drug design. Informations related to structure-based drug design that can be analyzed with this include flexibility and mobility, generating multiple conformations of a protein, examining the influence of mutants and investigating allosteric mechanisms. Atomistic simulation techniques such as molecular dynamics (MD) have become a powerful tool in the field of nanotechnology as they provide a physical insight in understanding various phenomena on atomic scale and enable one to predict some properties of nanomaterials. The simulation is performed over a period of time using well defined mathematical approximations of the known physics of atomic interactions. MD allows a user to probe the interactions of atoms, at the atomic scale, in molecules, on surfaces and in bulk solids. The complete MD calculation process is multidisciplinary, where the relevant MD theory is well established from a mathematics, physics, and chemistry point of view. Furthermore, it employs computational algorithms developed from computer science and information theory, which provide speedy and efficient numerical solutions. The MD method is commonly used today in fields, such as materials engineering and biomolecular research.
The molecular dynamics simulation method is based on Newton's second law or the equation of motion, F=ma, where F is the force exerted on the particle, m is its mass and a is its acceleration. Knowing the force on each atom, it is possible to determine the acceleration of each atom in the system. Integration of the equations of motion then yields a trajectory that describes the positions, velocities and accelerations of the particles as they vary with time. From this trajectory, the average values of properties can be determined. The method is deterministic; once the positions and velocities of each atom are known, the state of the system can be predicted at any time in the future or the past. Molecular dynamics simulations can be time consuming and computationally expensive.
Newton's equation of motion is given by
Where Fi is the force exerted on particle i, mi is the mass of particle i and ai is the acceleration of particle i.
The force can also be expressed as the gradient of the potential energy
Combining these two equations yields
where V is the potential energy of the system.
The simplest choice of V is to write it as a sum of pair wise interactions:
The clause j>I, in the second summation has the purpose of considering each pair wise interaction only once.
Newton's equation of motion can then relate the derivative of the potential energy to the changes in position as a function of time.
Taking the simple case where the acceleration is constant
obtained an expression for the velocity after integration
and since
we can once again integrate to obtain
Combining this equation with the expression for the velocity, the following relation is obtained which gives the value of x at time t as a function of the acceleration, a, the initial position, x0 , and the initial velocity, v0.
The acceleration is given as the derivative of the potential energy with respect to the position, r,
Therefore, to calculate a trajectory, one only needs the initial positions of the atoms, an initial distribution of velocities and the acceleration, which is determined by the gradient of the potential energy function. The equations of motion are deterministic, e.g. the positions and the velocities at time zero determine the positions and velocities at all other times t. The initial positions can be obtained from experimental structures, such as the x-ray crystal structure of the molecule.
Docking is a Lock and Key process which predicts the preferred orientation of receptor to a ligand when bound to each other to form a stable complex. It is used to predict the binding orientation of drug candidates to the protein targets. Hence docking plays an important role in the rational drug design. There are different types of docking tools in Accelrys Discovery Studio mainly CDOCKER, LigandFit and Flexible Docking.
CDOCKER is a molecular docking method that employs CHARMm forcefield. Here the receptor is held rigid and ligand is flexible. For each pose CDOCKER energy and interaction energy have been calculated. LigandFit docking based on initial shape match to the binding site. Dock score and ligand internal energy is obtained and thus ligand interaction energy is calculated. Flexible docking allows flexibility of receptor during docking of flexible ligands. The side-chains of amino acids in the active site are moved during docking allowing the receptor to adapt to ligands in induced-fit model. CDOCKER energy and CDOCKER interaction energy is obtained for each pose.
MATERIALS AND METHODS
The MMP proteins - MMP2 and MMP9 - present in human were collected from Protein Data Bank. These proteins have been subjected to sequence and structural analysis. The primary structure analysis done using ProtParam and the parameters computed from the tool include the molecular weight, amino acid composition, theoretical pI, estimated half-life, instability index, aliphatic index and grand average of hydropathicity. The secondary structure analysis done with SOPMA from which the percentage of alpha helix, beta turn and random coil were obtained. The minimization of the protein structures has been done using Accelrys Discovery Studio at the molecular mechanics level using CHARMm forcefield. Potential energy, Van der Waals energy, and RMS gradient values were computed. Electrostatic energy of the proteins was computed using MGL Tool in Python Molecule Viewer.
The surface scanning of the proteins was carried out using Computed Atlas of Surface Topography of Proteins (CASTp) tool. To simulate the dynamics behaviour, the proteins were subjected to Dynamics (Heating or Cooling) protocol in Accelrys Discovery Studio. The protein structure obtained after Dynamics (Heating or Cooling) was again scanned with CASTp to determine the variation in the surface structure.
The chemical structure of the drug Marimastat for MMP was obtained from DrugBank and it was optimized with Accelrys Material Studio. As Marimastat was given as the inhibitor of only two isoforms of MMP - MMP2 and MMP9 - interaction studies were done with these isoforms. Both rigid and flexible docking was carried out for MMP2 and MMP9 with Marimastat using CDOCKER, LigandFit and Flexible Docking tools of Accelrys Discovery Studio.
RESULTS AND DISCUSSIONS
10 proteins of MMP, one for MMP2 (1RTG) and nine for MMP9 (1GKC, 1GKD, 1ITV, 1L6J, 2OVX, 2OVZ, 2OW0, 2OW1, 2OW2) were retrieved. Results of ProtParam show that the proteins are hydrophilic in nature. It has been found that the proteins are stable in nature from the instability index values.1GKC and 1GKD was found to have highest estimated half-life. The secondary structure of protein mainly consisted of random coils, alpha helix and beta turn.
Potential energy of the modeled protein structure was highly negative and exothermic for most of the proteins which implies high thermodynamic stability. The Van der Waals energy was negative for all the proteins, indicating high possibility of secondary bonding. The RMS gradient is approximately 0, which indicates the stability of protein structure. There was only a slight difference in electrostatic energy between the proteins, computed using MGL.
The binding pockets of protein analyzed using CASTp. Number of pockets was more for 1ITV and 1GKC which indicates that they are less stable and status of mutation is high. Area and volume of the binding pockets of 1L6J, 2OW1, 2OW0 and 2OW2 are more compared to other proteins indicating that their pockets are less dense. Most of the binding pockets are hydrophilic in 1L6J, 2OW2, 2OW0, 2OVX, 1GKD, 2OW1, 2OVZ and 1ITV when compared to other proteins. CASTp analysis of the protein structure obtained after increasing the temperature to 312K and 314K by Dynamics (Heating or Cooling) showed that the number of binding pockets increased for 1RTG, 1GKD, 2OVZ, 2OW2 and 1L6J.
Interaction studies done for all the 10 proteins with optimized structure of Marimastat using CDOCKER, LigandFit and Flexible Docking. Result showed that there is a difference in interaction energy and number of poses for the proteins in different types of docking. Interaction energies for the proteins for which there is a significant difference in docking is shown in the table-1
Table 1 Interaction studies of proteins with Marimastat
CDOCKER
LigandFit
Flexible Docking
PROTEIN
SITE
CDOCKER INTERACTION ENERGY
# OF POSES
LIGAND INTERACTION ENERGY
# OF POSES
CDOCKER INTERACTION ENERGY
# OF POSES
1RTG
1
-21.144
10
-38.362
10
-31.39
30
2
-7.063
1
1L6J
2
-41.343
10
-26.581
10
-35.063
38
3
-41.058
10
-34.872
10
-35.841
4
5
-26.741
10
-31.636
10
-27.144
15
2OVZ
1
-35.596
10
-20.887
1
-31.38
130
2OW1
1
-45.856
10
-30.421
10
-32.748
90
3
-26.111
10
-18.279
2
2OW2
1
-42.314
10
-21.58
10
-41.202
70
From the results it is clear that Flexible Docking is having higher number of poses compared to CDOCKER and LigandFit. Interaction energies are most negative for CDOCKER in case of 1L6J, 2OVZ, 2OW1 and 2OW2 and thus are most favourable to binding. As flexibility of the active site of receptor increases, it influences the interaction between the ligand and the receptor.
CONCLUSION
