Homology modeling, molecular dolcking and MD simulations of Human Cannabinoid receptor 1 (CNR1) with the bioactive molecules of cyanobacteriaAbstract
Background and Objective: Cannabinoid receptor 1(CNR1) is having a relation with A G-protein coupled receptors that have mostly existed in the organ system of humans. It has a good opportunity as therapeutic targets in certain globally challenged disease (cancer, diabetes, obesity, cardiac dysfunction, CNS disorders, inflammation and pain).
Methods: In this computational study we were carried out in Schrodinger suite packages like prime application, sitemap generation, grid and glide SP docking mode.These are available in version 10.2 Maestro. The target CNR 1 was retrieved from uniprot and cyanobacterial ligands are getting by chemical database. Homology modelling makes a good model for docking study. Results: As a result, homology modelled target showed low sequence similarity thus the molecule was made quality target. Among the 20 cyanobacterial ligands, Symplocamide A had a potent docking score and shown good binding affinities than other ligands.
Conclusion: According to the results we suggested that Symplocamide A is developed as a potential drug molecule for CNR1 targets.
Key words: CNR1, Homology modelling, Cyanpbacterial ligand, Molecular docking, MD simulation, Symplocamide A
Cannabinoid receptor 1 (CNR1) has a relation with A G-protein coupled receptors that are mostly existed in the organs system of humans. It has a good opportunity as therapeutic targets in certain diseases like cancer, diabetes, obesity, cardiac dysfunction, CNS disorders, inflammation and pain 1. CNR1 is a very useful therapeutic target which is involved in a wide range of metabolic regulations. According to Lipina et al, the CNR1 improving and developing drugs are continuously made by pharmaceutical companies 2. CNR1 is also expressed in the central nervous system (CNS) and are particularly rich in certain brain areas such as the basal ganglia, cerebellum, and hippocampus. It is also found in the body parts of periphery, human testis, retina, sperm cells, colonic tissues, peripheral neurons, adipocytes and other organs such as heart, lung, prostate, uterus, and ovary including adrenal gland 3.
Generally, G protein coupled receptors (GPCRs), has a large family of transmembrane proteins and have a key functions of receiving extracellular signals and physiological functions of the entire body 4. G-protein coupled receptors (GPCRs) exist in most internal organ systems and present a wide range of opportunities as therapeutic targets in areas including cancer, cardiac dysfunction, diabetes, CNS disorders, obesity, inflammation, and pain 1.
An important development in neuropharmacological research produce the discovery of specific cannabinoid receptors (CNR1 and CNR2) in central and peripheral mammalian tissues. CNR1 receptor is a therapeutically useful target involved in a wide variety of physiological processes, includes metabolic regulation, craving, pain, and anxiety 5. Anciently, many marine researchers reported that cyanobacterial species contain effective secondary metabolites 6. These metabolites are used to control such a globally and clinically challenged diseases. Moreover, these molecules were also involved in a wide range of pharmacological activities which includes anticancer, antibacterial and antiparasitic activities 7.
Most of the effective healing properties have secondary metabolites which are obtained from marine cyanobacteria which are listed cyclic, short linear and mixed peptide divisions that are looking like a G-protein coupled receptors of mammalian glands in endogenous 8. These molecules are mostly produced the genus of Lyngbya and also that are active on the cannabinoid receptor.
Now a days conventionally available cannabinoid drugs are mostly used to treat chemotherapy agent, that are induced many side effects such as, nausea, vomiting, relieving neuropathic pain, and as an appetite stimulant for AIDS patients 9. Hence, we are seeking natural bioactive molecules with the lack of side effects by using computation. The main aim of this research effective cyanobacterial bioactive molecules were screened for CNR1 target.
2. Computational methods
2.1. Retrieval of sequence and ligandsThe sequence of a human cannabinoid receptor (CNR1 Accession Number: P21554) was retrieved from the universal protein database 10. Totally 472 residues are availed in CNR1. Target residue is starting from MKSIL and end with SAEAL. Ligand molecules of cyanobacterial compounds taken from chemspider database 11.
2.2. Homology modeling
Homology modelling finds the 3D structure of a protein based on its sequence similarity to one or more proteins of known structure. This method support the observation of the structural conformation of a target which is more highly conserved than its amino acid sequence. Homology modelling is finds the following steps: template identification, sequence alignment, BLAST homology search, model building and validation 12. This model with the lowest energy was selected for further refinement, which comprises the loop refinement and energy minimization using OPLS 2005 force field in Schrodinger Prime 13. The minimized model was then validated using Ramachandran plot, generated from PROCHECK at the SAVES version 4 server (Structure Analysis and Verification Server; (http:// services.mbi.ucla.edu/SAVES/)14. This model was optimized prior to docking using the Protein Preparation Wizard in Schrodinger Maestro Suite 2014 15.
2.3. Preparation of Target molecule
The CNR1 is prepared by using protein preparation wizard in Maestro v10.2. It was mainly involved in the process like to add missing hydrogen atoms, arranging bond order assignments, charge states an orientation of various groups 15.
2.4. Validation of binding site
Binding site process is an essential role of target. Because all the ligand molecules are binding with the target binding cavity. Here effective and suitable ligand molecules get favorable on drug preparation. These sites were validated with the help of site score. It means that the active residues are available and get a larger volume of the binding pocket 16.
2.5.Preparation of LigandsTotally 20 cyanobacterial bioactive selected from this molecular docking studies such as Symplocamide A (Chemspider Id: 23314421), Pompanopeptin B (Chemspider Id: 27023335), Lyngbyastatin 5 (Chemspider Id: 23076610), Lyngbyabellin D(Chemspider Id: 9106209), Hoiamide A(Chemspider Id: 28185012), Nostocylopeptide A(Chemspider Id: 215110), Lyngbyabellin H(Chemspider Id: 23314421) and Cryptophycin 327(Chemspider Id: 23314421). These ligands are prepared with LigPrep tool. Ligand preparation includes a series of steps that perform conversions from 2D to 3D and apply corrections to the structure, produce ionization states at biological pH, generate possible tautomers, optimize the geometrics and finally minimize the ring conformations were represented in Fig.1 17.
2.6. Molecular docking
Docking studies were carried out in modeled CNR 1 protein and some cyanobacterial ligands by using commercial software Schrodinger suite version 10.2 Maestro. This study shown efficient ligand molecule against target. For instance Shubani et al, using the software of Schrodinger version Masetro 9.0 13. But we are using the advanced level of tool.
2.7.Prime/MM-GBSA binding-free energy computation
The selected eight cyanobacterial compounds were further analyzed by Prime/MM-GBSA method to calculate the free energy of ligand binding in the receptor-ligand complex. This method was follwed by Kakarala et al, 18. The total free energy of binding is calculated as follows:
?G bind = G complex – (G protein + G ligand)
Where G = MME (molecular mechanics energies) + GSGB (SGB solvation model for polar solvation) + GNP (nonpolar solvation. MM-GBSA values were evolved as the tool of Prime application 15.
2.8.Molecular dynamic simulations
A protein- ligand complex molecular dynamic simulations were implemented Desmond program v 3.8 (https://www.schrodinger.com/desmond)19. A CNR1 is a membrane protein with ligand so I set up a membrane in a predefined model by POPC 300K (Palmitoyl oleoyl phosphatidyl choline) (Vytautas Gapsys et al., 2013)20 and the CNR1 protein with ligand complex are solvated by TIP4P (Tripathi et al., 2013)21 with OPLS-2005 force field parameters by using all calculations. The orthorhombic periodic bounding box of the size 10Å3, after the system was neutralized by adding appropriate counter-ions followed by addition 0.15M of salt resemble of the physiological condition. Before the MD simulation running process the system was relaxed by performing a series short, restrained and non-restrained solute minimization methods followed by 1.2ps simulation steps using NPT ensembles. After the MD simulation were implemented 10 ns using the NPT ensembles. Trajectories were established inside maestro atmosphere and the results were evaluated through RMSD interaction.22.
3.1.Homology modellingAmong the observation confirmed the model of GPCRs super family, human CNR1 was found to be suitable as a template. The early search for templates for homology modeling of human CNR 1 with BLAST search exposed as the template for 27% identity and 47% similarity were made. The sequence alignment of human CNR1 has been shown (Fig. 2). Later, a model validated through Ramachandran plot and also this template represents the distribution of amino acids in phi and psi angles by checking PROCHECK (Table 1 and Fig. 2a). Among the determined values were helpful to this structural determination. The low sequence similarity can also produce a good model provided. When the alignment is done correctly. Hence, the 3D modeled human CNR1 was used as a template for molecular docking (Fig. 3). After finalization a model was taken as a molecular docking study.
3.2. Binding site validation
This method is one of the phenomenon tryouts for molecular docking. Because all the ligand molecules are posed with the active site. Here, the binding cavity is having many active residues that are shown in Table 2.
3.3.Molecular docking studies
In this docking analysis potential ligand molecule only poses a binding pocket on target and showed that their hydrogen bond formation between two molecules (ligand to protein). Totally, 8 cyanaobacterial ligand molecules and CNR1 protein complex were docked with the help of schrodinger tool. At this point, suitable ligand molecule only got a superior docking score than other ligands against the target. Present computational analysis sucessfully find efficient drug for the protein of CNR1. The outcome of this molecular docking studies showed cyanobacterial ligands efficacy, binding affinity (both side chain and back chain Pi-Pi stacking and salt bridge interaction) and so on. Among the 8 ligands, Symplocamide A having a potent docking score than other molecules and also MM-GBSA free energy calculation were shown in Table 3. Similarly, Hua et al, (2016) have predicted the molecular docking on the crystal structure of cannabinoid receptors 1 (CNR1) with the different antagonist of cannabinoid23.
3.4. Symplocamide A
By the way of computational approach the ligand Symplocamide A have an efficient docking score and are having good binding affinities. Hence, this study examined, that of the ligand contacts between protein residues. Fig 2 showed that the target residues were ASP333, ASP403 and LYS402 involved in H-bond contacts. As a result, we measured the distance on Symplocamide A and CNR1molecule. These residues were interacting with ligand oxygen and ammonia groups that are shown four different distance values, respectively 2.030, 2.264, 1.776 (ASP 333) and 2.648 (ASP 403) (fig.5). ASP402 was showed H- bond contact distance value with that ligand oxygen group and ASP403 was showed H-contact distance value with ligand oxygen groups. Symplocamide A 2 dimensional map shown ligand to protein contacts like H- bond side chain, back chain, Pi-Pi stacking and salt bridge . Hence, we displayed the contacts of protein to ligand in (Fig. 5a). The outcome docking analysis to produce two dimensional interaction map, this figure represented two kinds of H-bond lines. The initial one is dotted straight line and another one is a solid straight line. The dotted straight line is called H-bond side chain. The solid straight line is called H-bond Back chain. ASP402, ASP403 and ASP333 residues were formed H-bonds with an efficient ligand molecule of Symplocamide A. ASP 333 is holding 3 H-bond side chains and 1 H-bond back chain with that of a ligand. ASP402 was seen to frame with the ligand oxygen groups and ASP403 holding with the formation of H-bond side chain in ligand ammonia groups.
3.5. Pompanopeptine B
Pompanopeptine B had a second efficient docking score with the target and that was having a good binding affinities. Here, we examined that of the ligand contacts done by the protein residues. Examination of this docked complex Pompanopeptine B and CNR1 showed that the residues ARG150, ASP338 and ASP333 respectively. As a outcome, we measured the distance between ligand to protein H-bonds contacts distances. These residues were interacting with ligand oxygen and ammonia groups that are bind normally and covalently. At this point, shown six different distance values 2.199, and 2.301 (ARG150) 2.795 and 1.576 (ASP338) and 1.981 in ASP333 respectively (Fig 6.). ARG150, ASP338 and ASP333 were formed H-bonds from the ligand Pompanopeptine B. ARG150 is covalently holding two H-bond side chains with ligand oxygen groups. ASP338 was seen to frame one H-bond side chain on small molecule of the ligand ammonia groups and ASP333 framed one H-bond side chain in ligand ammonia groups (Fig 6a).
3.6. Lyngbyastatin 5
Lyngbyastatin 5 had the third potent docking score against the CNR1 protein and it have a good binding affinities. Hence, we examined the ligand contacts by the protein residues. A Lyngbyastatin 5 showed that the H bond contact residues ARG331, ASP333, ARG150 and TYR224 respectively. As a result of this study, we measured the H-bonds contacts distance between ligand & protein. At this point, shown five different distance values, respectively 2.266 (ARG331), 1.669 (ASP333), 1.87 and 2.088 (ARG150) and 1.846 (TYR224) (Fig 7). ARG150 was framed covalently in H-bond side chain from the ligand of Lyngbysatain 5 oxygen groups. ASP333 residue was frame one H-bond back chain on ligand oxygen groups. ARG331 was seen to frame one H-bond side chain and Pi-Pi stacking from this ligand molecule. TYR224 was seen to framed one H-bond chain with a ligand OH group (Fig.7a).
Lyngbyabellin D had the fourth potent docking score against the CNR1 protein and it is having good binding affinities. Hence, we examined that of the ligand contacts by the protein residues. From this docking analysis that the ligand and protein H bond contacs were displayed respectively LYS402, SER146, ARG150, ASP333 and ARG331. As a result, we measured the distance values of H-bonds between ligand to protein. At this point, shown five different distance values 2.728 (LYS402), 1.886 (SER146), 2.082 (ARG150) 1.958 (ASP333) and 2.623 (ARG331) respectively (Fig.8). LYS402 was framed one H-bond side chain from the ligand of Lyngbyabellin D nitrogen groups. SER146 residue was framed one H-bond back chain with ligand OH groups. ARG331 and ARG150 were representing two H-bond side chain from the ligand oxygen groups (Fig.8a).
3.8. Molecular dynamics simulation
The complex of modeled CNR1 and Symplocamide A were implemented in MD simulations for the purpose of structural stability evaluation. MD simulation is t he release the trajectory files. The trajectory file was taken as RMSD calculations. The final trajectory files were taken for calculating the RMSD of the complex structures. Complex of molecules (target and ligand) was running MD simulations by the time of 10 ns, RMSD plot represented in Fig.9). During the period of this time, the potential energy was also stable 1.6 ns to 4.0 ns (Fig. 9a). Whereas the docking studies showed 6 hydrogen bonds, contacts were formed between targets of ligand molecules. Moreover, it makes the stability of the complex structure during the entire simulation time up to 10 ns.
Our study conducts the homology modeling to take as the CNR1 FASTA sequences. And also this model was generated in low sequence similarity with good resolution. Similarly, Kauffman et al., 24 reported that low sequence similarity template can also make a good ideal model for docking study when the alignment is handled correctly. The target showed 5 binding pockets in which site 1 has a suitable cavity for drug target. For instance, Vijayakumar et al, reported the binding cavity importance and to screen the active residues of HBV target. In this study we analyzed that there are mainly five sites from the target in which site-1 is a major active binding site. 25.
Naturally available sources are having a highly therapeutic agent for a lot of ailments. The natural factors include plant, cyanobacteria and such a beneficial microbes. But our study focused only on cyanobacterial bioactive constituents. Many cyanobacterial bioactive constituents potentials were screened against CB1 and CB2 receptors. Especially, Lyngbya sp bioactive molecules are dominantly studied in many aspects together computationally as well as clinically. Some of the cyanobacterial bioactive molecules were phenominantly studied in such a kind of globally challenged diseases. Malyngamide is one of the fatty acid amide compound the molecule is screened for their CB1-CB2 receptors 26.
Semiplenamide and Serinolamide were similar to the endocannabinoid Anandamide and that identical series which belongs to Malyngamide. The cyanobacterial molecule of Semiplenamide A, B and G showed weak binding affinity with CNR1 receptors. But Semiplenamide A was inhibiting the breakdown of Anandamide. The bioactive molecule of Serinolamide and it was showed effective binding affinity with human cannabinoid receptors and found 5-fold to suitable supplemented agonist activity for CNR1 27.
Our study find this Symplocamide A is a lead docking score than other bioactive molecules for this docking analysis. Generally, symplocamide A is an effective cytotoxin molecule. This bioactive constituent is available in marine cyanobacteria, the genus of Symploca sp 28. It is treated against H460 lung cancer and neuro-2A-neuroblastoma cell and to finding effective possible changes in treating cells 29. Furthermore, Kinghorn et al, determined and they reported that a molecule is one of the protease inhibitor, it is used for certain infectious diseases like HIV and HCV 30. Symplocamide A is a similar like other protease inhibitor that has been used before the treatments.
From this computational approach we find out the emerging and effective drug molecule from the natural source of cyanobacteria against CNR1 protein. Our finding showed the Symplocamide A potentiality. It is also placed in effective docking score and good binding affinity than other seven cyanobacterial molecules against this target. Finally, our study suggested that Symplocamide A is a potent therapeutic drug molecule for CNR1. This method of study helps to reduce the complexities, cost and time.
The authors are grateful to the DST-SERB (SB/YS/LS-109/2014) for providing financial assistant in this project. We specially express our thanks to the management of A.V.V.M. Sri Pushpam College (Autonomous), Poondi, for providing them necessary facilities and support to carry out this work.
O. Elbegdorj, R.B.Westkaemper Y. Zhang, A homology modeling study toward the understanding of three-dimensional structure and putative pharmacological profile of the G-protein coupled receptor GPR55, J. Mol. Graph. Model. 39 (2013) 50–60.
C. Lipina, W. Rastedt, A.J.Irving, H.S. Hundal, Endocannabinoids in obesity: brewing up the perfect metabolic storm, WIREs. Membr. Transp. Signal. 2 (2013) 49–63.
P.H. Reggio, Endocannabinoid Binding to the Cannabinoid Receptors: What Is Known and What Remains Unknown. Curr. Med. Chem. 17 (2010) 1468–1486.
M. Khoddami, H. Nadri, A. Moradi, A. Sakhteman, Homology modeling, molecular dynamic simulation, and docking based binding site analysis of human dopamine (D4) receptor. J. Mol. Model. (2015) 21-36.
K. Mackie. Cannabinoid receptors as therapeutic targets. Annu Rev Pharmacol Toxicol 2006; 46: 101–22
R. Dixit , M.R. Suseela, Cyanobacteria: potential candidates for drug discovery, Ant. Van. Leeu. 103 (2013) 947–961.
N.C.Laxa, K.T. Ahmed, C.M. Ignatza, C. Spadaforad, B.J. Kolbera, K.J. Tidgewell, Marine cyanobacteria-derived serotonin receptor 2C active fraction induces psychoactive behavioral effects in mice, Pharma. Biol. DOI: 10.1080/13880209.2016.1181659
F. Hall, F. Helsenki, Poster Presentations, J. Peptide. Sci. 14 (2008) 47–194.
Pertwee RG. Targeting the endocannabinoid system with cannabinoid receptor agonists: pharmacological strategies and therapeutic possibilities, Phil. Trans .R. Soc. B. 367 (2012) 3353–3363.
Tastan O. Protein homology modeling and its use in South Africa. South. Afri. J. Sci. 104 (2008) 2-6.
S. Subhani, A. Jayaraman, K. Jamil, Homology modelling and molecular docking of MDR1 with chemotherapeutic agents in non-small cell lung cancer, Biomed. Pharmaco. 71 (2015) 37–45.
Prime version 3.7, Schrodinger, LLC, New York, NY, 2014.
F. Shaikh, S.W. Siu, Identification of novel natural compound inhibitors for human complement component 5a receptor by homology modeling and virtual screening, Med. Chem. Res. 25 (2016) 1564–1573.
LigPrep, 2011.Version2.5, Schrödinger, LLC, NewYork.
K.K. Kakarala, K. Jamil, V. Devaraji, Structure and putative signaling mechanism of Protease activated receptor (PAR2)–a promising target for breast cancer. J. Mol. Graph. Model. 14 (2014) 132-6.
Desmond, 2011.Version3.0Schrödinger, LLC, NewYork.
V. Gapsys, B.L. de Groot, R. Briones, Computational analysis of local membrane properties, J. Comput. Aided. Mol. Des. 27 (2013) 845–858.
S.K.Tripathi , M.Ravikumar , S.Sanjeev Kumar, Extra precision docking, free energy calculation and molecular dynamics simulation studies of CDK2inhibitors, J. Theo. Biol. 334 (2013) 87–100.
T. Hua, K. Vemuri , M. Pu , L. Qu, G. Won, H. Yiran, S. Zhao, W. Shui, S. Li, Korde, A. Robert, B. Laprairie, E.L. Stahl, J.H. Ho, N. Zvonok, H. Zhou, I. Kufareva, B. Wu, Q. Zhao, M.A. Hanson, L.M. Bohn, A. Makriyannis, R.C. Stevens, Z.F. Liu, Crystal Structure of the Human Cannabinoid Receptor CB1, Cell. 167 (2016) (3): 750–762.e14.
C. Kauffman , H. Rangwala, G. Karypis, Improving Homology Models for Protein-Ligand Binding Sites. file:///C:/Users/admin/Downloads/kauffman08csb.pdfS. Vijayakumar, J.P. Harikrishnan, S. Prabhu, J.E. Morvin Yabesh, P. Manogar, Quantitative Ethnobotanical Survey of Traditional Siddha Medical Practitioners from Thiruvarur District with Hepatoprotective Potentials through In Silico Methods, Achiev. Life. Sci. (2016) 1-16.
R. Montaser, V.J. Paul, H. Luesch, Marine cyanobacterial fatty acid amides acting on cannabinoid receptors, Chem-BioChem 133 (2012) 2676–2681.
M. Gutierrez, A.R. Pereira, H.M. Debonsi, A. Ligresti, V. Di Marzo, GH. Gerwick. Cannabinomimetic lipid from a marine cyanobacterium, J. Natu. Prod. 74 (2011) 2313–2317.
Han B, McPhail KL, Ligresti A, Di Marzo V, Gerwick WH. Semiplenamides A-G fatty acid amides froma Papua New Guinea collection of the marine cyanobacterium Lyngbya semiplena, J. Natu. Prod. 66 (2003) 1364–1368.
G. Roger, Linington, J. Daniel, F. Edwards Cynthia Shuman , L. Kerry Mcphail, M. Teatulohi, W.H.Gerwick, Symplocamide A a Potent and cytotoxin and chrymptophycin Inhibitor from the Marine Cyanobacterium Symploca sp, J. Natu. Prod. 71 (2008) 21-27.
A.D. Kinghorn, Y.W.Chin, S.M. Swanson, Discovery of Product Anticancer Agents from Biodiverse Organisms, Curr. Opin. Drug. Discov. Devel. 12 (2009) 189-196.