Received: 28th July-2014 Revised: 05th Oct-2014 Accepted ... - IJABPT

2 downloads 0 Views 1MB Size Report
Oct 5, 2014 - Email: kv.ramesh@jainuniversity.ac.in. ABSTRACT: Melanocortin system is composed of four peptide hormones known as α, β, γ and.
Received: 28th July-2014

Revised: 05th Oct-2014

Accepted: 06th Oct-2014 Research article

MOLECULAR DYNAMICS SIMULATION OF Β-MELANOTROPIN STIMULATING HORMONE (Β-MSH) DOCKED ONTO THE RECEPTOR MC4R Mutangana Dieudonné and Ramesh K V* Department of Biotechnology, Centre for Postgraduate Studies, Jain University, 18 /3, 9th Main, Jayanagar 3rd block, Bangalore – 560011, Karnataka, India. [Fax: +918043226507, Ph: +918043226510] Email: [email protected] ABSTRACT: Melanocortin system is composed of four peptide hormones known as α, β, γ and adrenocorticotropic hormone (ACTH), derived from post-translational cleavage of a polypeptide precursor ‘proopiomelanocortin (POMC)’. Among these hormones; β-melanotropin stimulating hormone (β-MSH), an 18 amino acid residue peptide fragment is an important hormone as it is involved in activation of MC4R to induce lean phenotype ‘balance between energy intake and energy expenditure’. In addition to this, MC4R is also involved in the modulation of erectile function, including the spinal cord and pelvic ganglion of rats and the penis of both rats and humans, providing an anatomical basis for melanocortin effects on sexual function. MC4R is one of the five melanocortin receptors (MC1R–MC5R) which have been characterized with tissue-specific expression patterns and different binding affinities for each of the melanocortin hormones to regulate various biological functions. In the present work, 3D models of MC4R and β-MSH have been predicted, followed by docking and molecular dynamics simulation. While the 3D model of MC4R receptor has been predicted through threading approach, structure of βMSH was built based on ab initio technique. The β-MSH model was later successfully docked onto the MC4R protein. Molecular dynamics (MD) simulation for 15 ns was used to compute the electrostatic solvation energy as well as binding energy between MC4R with β-MSH model under implicit solvent conditions. Key words: Binding energy, Melanocortin, Melanocortin 4 receptor, Melanotropin beta, Molecular docking, Molecular dynamics. List of abbreviations: GTP: Guanosine-5'-Triphosphate, ACTH: Adrenocorticotropic Hormone, POMC: Proopiomelanocortin, MCR(s): Melanocortin Receptor (s), THIQ: Tetrahydroisoquinoline, IGD: Isolated Glucocorticoid Deficiency, CNS: Central Nervous System, MSH: Melanocyte-Stimulating Hormone, NMR: Nuclear Magnetic Resonance, MD: Molecular Dynamics, AMBER: Assisted Model Building With Energy Refinement, VMD: Visual Molecular Dynamics, SANDER: Simulated Annealing With NMR-Derived Energy Restraints, EL: Extracellular Loop, CL: Cytoplasmic Loop, RMSD: Root Mean Square Deviation, RMSF: Root Mean Square Fluctuation

INTRODUCTION Five melanocortin heterotrimeric (GTP) binding proteins (MC1R, MC2R, MC3R, MC4R, MC5R) expressed in different tissues have been identified (Adan & Gispen, 1997). Among these, human MC4R (333 amino acids) has been cloned (Ira et al., 1993) and is widely expressed in the central nervous system, including the cortex, thalamus, hypothalamus and brain stem (Abdel-Malek, 2001). Based on the distribution of MC4R, Abdel-Malek (2001) has suggested this receptor to be involved in autonomic and neuroendocrine functions. Pharmacologic characterization of MC4R expressed in heterologous cells indicated that it is very similar to the MC1R in its relative affinities for various melanocortins. Abdel-Malek (2001) has suggested this receptor to be involved in autonomic and neuroendocrine functions. Pharmacologic characterization of MC4R expressed in heterologous cells indicated that it is very similar to the MC1R in its relative affinities for various melanocortins. Even though functional role of all the 5 melanocortin receptors is well defined and has been discussed elsewhere (Giuliani et al., 2012), importance is given only for MC4R in the present study. Experimental investigations on rodent as well as human indicate that activation of MC4R results in a lean phenotype, whereas inactivation leads to obesity (Michael & Hang, 2000).

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 219

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

The order of potency for activation of MC4R, Chhajlani and Wikberg (1992) and Suzuki et al., (1996), by melanocortins is: α-MSH = ACTH > β-MSH > γ-MSH. Direct evidence for MC4R as a regulator of food intake has come from the observations made by Huszar and his group (Huszar et al., 1997); they have demonstrated that ‘mc4r’ gene when knocked out in mice, develop obesity. Hinney et al., (1999) have confirmed that in adolescent humans, mutations in MC4R can lead to the development of obesity. Association of certain mutations in the MC4R gene with obesity in humans suggests that this receptor is significant for regulating food intake in human beings (Gu et al., 1999). Van der Ploeg and his co-workers (2002) elucidated the role of MC4R using a non- peptide agonist “THIQ”. In their studies, THIQ augmented erectile function in wild type, which was absent in mc4r-null mice. Even though this study does not prove that MC4R is the only melanocortin receptor involved in peripheral action of melanocortins on sexual function, it demonstrates that administration of an MC4R agonist is sufficient to elicit melanocortin effects on sexual function. Presently, it is assumed that melanocortin mediated sexual function is due to the action of both central and peripheral nervous systems. Since melanocortin-mediated sexual responses appears to be elicited by peripheral administration of a selective MC4R agonist, this can have potential therapeutic use in pharmaceutical industries (Ira & Tung, 2003). However, use of such agonists may have an undesirable side effect for the treatment of obesity. Role of MC4R in sexual function has received a great deal of attention and has provided framework to explore melanocortin system for treating obesity, other metabolic abnormalities sexual dysfunction. Pogozheva et al.,(2005) have reported interactions between homology modelled human melanocortin 4 receptor and non-peptide (THIQ, MB243) as well as peptide agonist (NDP-MSH) by examining the effects of mutations in the TM domain on agonist binding and ligandinduced receptor-mediated cAMP accumulation. Analogs of α- MSH cyclized through site-specific rhenium (Re) and technetium (Tc), have been structurally characterized and analysed for their ability to bind α-MSH receptors (Giblin, Wang, Hoffman, Jurisson, & Quinn, 1998). Several studies carried out in the past have reported the presence of “HFRW” motif in all the melanocortins (Nakanishi et al., 1979). Based on modelling and site-directed mutagenesis of hMC4R, binding site of this receptor has been determined (Pogozheva et al., 2005). Outcome from this study reveal there are 10 critical amino acids positions distributed among TM2 (transmembrane domain) TM3, TM6 and TM7. However, studies on MC4R and its interaction with ligand have been elucidated without evidences supported by experimental 3D structures. Due to the absence of experimental data, an attempt was done in this research (i) to predict 3D structures of MC4R and β-MSH using computational techniques (2) to analyse receptor-ligand active sites through docking techniques followed by (iii) molecular dynamics simulation to evaluate the effect of solvent on the stability and binding affinity under implicit solvent conditions.

MATERIAL AND METHODS Molecular modeling of MC4R from Human and β-MSH Sequence information of MC4R protein was retrieved from UniProt database (ID: P32245) and was then submitted to TMHMM server (Krogh, Larsson, Von Heijne, & Sonnhammer, 2001) for the prediction of transmembrane helices. Search for potential template structures for this sequence was performed using PSI-BLAST tool with PDB as the database (Altschul et al., 1997). Theoretical structure of MC4R was constructed using threading approaches by accessing I-TASSER server (Zhang, 2008). Based on the reports generated by ERRAT (Colovos & Yeates, 1993) and PROCHECK (Laskowski, MacArthur, & Thornton, 1993), the initial 3D model of the receptor was subjected to loop refinement by accessing ModLoop server (Fiser, Gian Do, & Šali, 2000). Loop refined structure was subjected to energy minimization using the Deep View (Guex & Peitsch, 1997) package. The coordinates of the energy minimized structure were submitted to the DaliLite server (Holm & Rosenström, 2010) which compares it against the structural homologs deposited in the Protein Data Bank. The PSI blast tool (Altschul et al., 1997) could not retrieve any good template structure from PDB for β-MSH (UniProt ID: P01189 (217-234)) having a sequence identity of more than 20 percent which is required for homology modeling or threading (Johnson & Overington, 1993), 3D structure for βMSH was predicted using ab initio strategy followed by molecular dynamics simulations (MD) as implemented in AMBER 10 package (Case et al., 2008). This was performed on SGI Altix UV 10 having SUSSANE LINUX as the operating system. The initial structure of β-MSH (155 atoms ) in the linear polypeptide form was constructed using LEaP program with ‘ff99SB’ force field (Hornak et al., 2006). Initial topology and coordinates files were generated for β-MSH structure. The MD simulations were fully unrestrained and carried out with no periodic boundary conditions in the NPT ensemble using the SANDER module of AMBER package. SHAKE algorithm was applied to constrain bonds involving hydrogen as proposed by Padhi et al., (2012) and Ryckaert et al., (1997). Calculation was started by energy minimization in a total number of 1000 cycles with a switch from steepest descent to conjugate gradient algorithm after 500 cycles. Implicit solvation effect was introduced using generalized Born model as suggested by Still et al.,(1990). This model is based on pairwise de-screening formulation of the calculation of Born radii by Hawkins and coworkers (1996). In heating stage, system was gradually heated from 0 to 300 K spread over 6 stages (an increase of 50 K per stage), with time set to 5 ps for each of the first five stages and to 25 ps for the last stage.

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 220

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Weak-coupling algorithm was used to control temperature (Berendsen, Postma, van Gunsteren, DiNola, & Haak, 1984). Heat bath coupling for the system was set to default during the entire heating stage. Following heating stage, equilibration spread over 10 stages was started. Each stage was run for 5 ns at 300 K, with total simulation time of 50 ns. Output from MD simulation results was subsequently plotted using GRACE software package (Stambulchik, 1997). Conformational stability of predicted structures was analyzed by exploring the formation of polar contacts and salt bridges using VMD package (Humphrey, Dalke, & Schulten, 1996). For validating β-MSH model, it was superimposed onto NMR resolved α-MSH analog (12 residues; PDB id 1BOQ) having conserved motif “HFRW” (Giblin et al., 1998).

Docking studies Docked conformations for ‘MC4R – β-MSH’ were obtained using the protein-protein docking software HEX (v6.3) (Ritchie, 2003). The predicted structures of both, the receptor ‘MC4R’ and the ligand ‘β-MSH’ were initially loaded into HEX software. Docking was started by activating the shape and electrostatic mode. During the docking exercise, dock energy was calculated based on the shape as well as electrostatic interaction using Fast Fourier Transformation docking algorithm, with a default grid spacing of 0.6 Å. Though the program retains a summary of 10,000 highest scoring orientations, the best 100 orientations were retained for viewing.

Molecular dynamics simulation of MC4R_ β-MSH complex, MC4R and β-MSH Preparation of starting structures Molecular dynamics simulations of the docked ‘MC4R_β-MSH’ (2744 atoms) complex as well as the predicted structures of MC4R (2587 atoms) and β-MSH (155 atoms) were performed separately for calculating the free binding energy (∆G). Simulations were carried out on the same supercomputing machine that was used for predicting the 3D structure of β-MSH. All the three 3D models: MC4R_β-MSH complex, MC4R and β-MSH were loaded into AMBER 10 using the LEaP tool with ‘ff99SB’ as the force field. The same protocol used for β-MSH was followed as described above.

Heating and equilibration All the three 3D models were subjected to energy minimization, heating and equilibration using SANDER module of AMBER package. Protocol described earlier was followed with minor modification during equilibration/production stage; cutoff (16 Å) flag was introduced to constrain bonds involving hydrogen atoms for 15 ns. The MD simulation output from all the three models were separately analyzed using the perl script as well as PTRAJ command. Energy and RMSD plots were obtained using GRACE software package (Stambulchik, 1997). Salt bridges formed by MC4R model and MC4R_β-MSH complex, both before and after MD simulation, were analysed by VMD package. Electrostatic part of solvation free energy ∆Gel was calculated using the method described by Smith and Honig (1994).  . This method was summarized in the following equation by Froloff and coworkers (1997).

Where ∆Gel is obtained from the sum of coulomb and reaction (solvation) energies of the complex minus that of the isolated reactants. The electrostatic part of solvation free energy ∆Gel is, according to the theory approximated by an analytical function introduced by Still et al., (1990). Here it may be noted that entropic cost of fixing rotational and translational degree of freedom is not taken into account for our calculations.

where rij is the distance between atoms i and j, qi and qj are partial charges and 1 is the dielectric constant of the solvent. Born radii of interacting atoms Ri and Rj which represent each atom’s degree of burial within the solute was given by the equation developed by (Hawkins et al., 1996). Conformational free energy (Potential energy in our case) of each of the three models in solution was computed using the method that has been described (Smith & Honig, 1994) using finite difference solutions to Poisson-Boltzmann equation.

where ∆Gconf (g) is the conformational energy in gas phase (vacuum) and ∆Gsolv is the gas to water solvation free energy. ∆Gconf (g) was obtained from molecular mechanics force field. These involve sums of terms over all the internal coordinates of the molecule ‘∆Einternal’ (bond, angle, and dihedral energies), ∆Eelect (electrostatic), and ∆EVDW (van der Waals) energies. ∆Gsolv is the conformation change as a result of solvation. In the same analogy, binding free energy (∆Gbinding) of the reaction due to conformational change is estimated by the proposed equation:

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 221

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

where is computed using equation 3. Here it may be noted that entropic cost of fixing rotational and translational degree of freedom was parameterized to be removed at the end of every simulation step of MD simulation while writing to the output file. In other words, values for various variables (bond, angles, dihedrals, electrostatic etc) in the output files are shown after subtracting the entropy energy.

RESULTS AND DISCUSSION Molecular modeling studies Melanocortin 4 receptor The TMHMM server identified seven transmembrane helices (H1-7), four cytoplasmic loops (CL1-4) and four extracellular loops (EL1-4) (Fig 1); based on the MC4R sequence information. These results suggest that the MC4R sequence probably belongs to G- protein coupled receptors superfamily which are commonly characterised by their seven transmembrane-domain architecture (Moukhametzianov et al., 2011). Because the PSI BLAST tool was unable to retrieve any sequence homologs for MC4R sequence having more than 30 % sequence identity from PDB (Table 1), which is important for homology modelling, threading technique was chosen to predict the 3D structure of the protein. This was done by accessing I-TASSER server. The server was successful in generating 5 different models. Among these, the first model is generally considered to be the best one as it recorded the highest C-score. A model with a higher C score indicates a better structural match with the template structure. It also indicates a better confidence in predicting the function of the protein using the template (Wang et al., 2006). Based on Modloop server output, it has been noticed that the quality factor of the loop refined model when re-evaluated by ERRAT increased from 63.889 to 83.025, while the energy of the model decreased from -7642.723 to - 8775.846 kcal mol-1 at the end of energy minimization process. Further validation of the loop refined MC4R model using PROCHECK server revealed that 90.5% residues were in most favoured, followed by 7.8 % in additional allowed and 1.0 % in generously allowed, 0.7% in disallowed regions. Overall G factor for the predicted structure was -0.07. A G value of not less than -0.5 is suggestive of a satisfactory quality of the predicted model (Kleywegt & Jones, 1996). PROCHECK analysis was used by Prabhavathi et al., (2011) to check the quality of the predicted thioredoxin (TRX) protein. .

Structural homologs of MC4R model From the output generated by the DaliLite server, it is apparent that Cα atoms of MC4R model got well superimposed onto its structural homologs deposited in the PDB data bank, which are proteins belonging to the superfamily of Gprotein coupled receptors. The z-score of the top five structural homologs superimposed onto the MC4R model varied from 26.2 to 26.6, with the RMSD varying between 2.3 to 2.5 Å. These results suggest that the structure was of acceptable quality. DaliLite has been used by Prabhavathi et al., (2011), Mukherjee and Zhang (2011) and others for identifying the structural neighbours for the predicted protein structure.

Investigation of MC4R model structure and functions based on superimposition studies Superimposition of the MC4R model onto the X-ray structure of adenosine A2A receptor (PDB ID: 3RFM_A) from human (structural homolog suggested by DaliLite server and structural template proposed by Swiss model server) revealed that both the proteins were structurally similar. The RMSD of Cα atoms between MC4R with 3RFM_A analysed by SPDV (Guex & Peitsch, 1997) package was 1.12 Å. Structural alignment of the MC4R model onto 3RMF_A revealed approx. 18 % residues of the model were completely conserved (Fig 2). Aspartate (D146), the first residue of “DRY” motif of MC4R model being completely conserved was also involved in forming salt bridge with the second residue, arg147 of the same motif of CL2. In separate studies conducted by Moukhametzianov et al., (2011) and Vogel et al., (2008), ‘DRY’ motif of G-protein coupled receptors (3RFM_A and 2YCW_A) were completely conserved. However, arginine, the second residue of ‘DRY’ motif present in H3 formed salt bridge with glutamate located in H6 of G-protein coupled receptors. From the data (Fig 2), it is evident that helices and loop regions of MC4R more or less overlapped with those of 3RFM_A. Most of the conserved residues belong to helix regions: 7 from H1 [val50, 56, 65, leu60, 64, asn62 and ala68], 10 from H2 [phe82, 100, ser85, leu86, 92, arg87, ala89, asp90, val92, 95, and ile102], 7 from H3 [ile125, 137, 143, ser136, 139, leu140,141], 4 from H4 [ile169, 170, trp174, gly181], 1 from H5 [leu207], 6 from H6 [ile249, gly252, phe254, cys257, trp258 and pro260], and 7 from H7 [leu286, 290,asn294,ser295, phe299,ile301,tyr302 and ala303]. The conserved residues were less in loop regions; while val46 of EL1 and phe117 of EL2 were completely conserved, none of the residues of EL3 and EL4 was conserved. A total of 15 residues were conserved in cytoplasmic loops of MC4R; CL1 [asn72, 74, leu75], CL2 [asp146, 165, arg147, tyr148, ile157, leu160 and thr162], CL3 [arg244] and CL4 [arg305, glu308, lys314 and ile317].

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 222

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Structural alignment data shows that ser190, 191 of EL3 have not aligned with 3RFM_A, resulting in the formation of a big loop in MC4R model. Based on this information, structurally conserved regions are present towards both N- as well as C-termini regions of MC4R protein. Outcome from these observations strongly suggest that the predicted MC4R structure is probably a G- protein coupled receptor (GPCRs), characterised by seven transmembrane-domain architecture (Fig 3). Studies carried out by Bokoch et al., (2010), Thomson et al., (2012) and others have described similar architecture for G-protein coupled receptors present in other organisms.

ab initio structure prediction and molecular dynamics simulation of β-MSH AMBER package was able to successfully fold primary sequence of β-MSH into its 3D form having two short beta strands upon completion of 50 ns of MD simulation. Further examination of the folded β-MSH structure through Deep View package revealed involvement of “HFRW” motif in the formation of reverse β-turn. This complies with the observation made by Haskell-Luevano (1996), who have reported similar kind of folding pattern for structurally modified α-MSH peptide. Energy profile of β-MSH remained constant during the equilibration stage of MD simulation, wherein the total energy, kinetic energy and potential energy did not fluctuate during the equilibration stage of MD simulation. The lowest energy structure of β-MSH, which represents the folded state of the protein, was obtained during the third equilibration stage of MD simulation (26th ns) having a potential energy of -660.69 kcal mol-1 (Fig 4 (a)). Upon closer scrutiny using VMD package , the structure displayed two salt bridges formed between O-atoms of glu8,2 with N-atoms of lys17 and arg6 residues (Fig 4 (b)) respectively. The RMSD plot of backbone Cα atoms of β-MSH showed 3 plateau spread across 50 ns time scale (0 to ~ 20000; ~20000 to ~ 32000; ~ 32000 to ~ 50000 ps). The structure appears to have attained stability between 20 and 32 ns (middle plateau) with deviation of Cα atoms less than 0.5 Å (Fig 4). As it can be seen later from the RMSD plot in figure 7, the lowest energy structure of β-MSH when once again subjected to additional 15 ns of simulation for computing free energy, deviation of backbone Cα atoms varied between 0.5 to 1.0 Å; which very well agrees with our earlier observation. These results suggest that the predicted structure was not trapped in a high energy minimum; instead it has evolved into its folded state. Superimposition of β-MSH model onto NMR structure of α-MSH peptide analog showed deviation of Cα atoms (RMSD) between the two structures was 1.39Å. According to Fiser and corkers (Fiser et al., 2000), modelled structure is believed to be reasonably good if RMSD is less than 2.0Å.

Docking studies Among 100 orientations generated by HEX software while docking β- MSH peptide onto MC4R protein, the first docked pose (figure 6 (a)) having the highest docking energy (-493.2 kcal mol-1) was taken for further analysis. Active site residues of the docked complex analysed by PyMOL (DeLano, 2002) revealed that among all the 18 residues of the β-MSH model, 11 residues (asp1,18, glu2, 3, pro4, 16, tyr5, his9, phe10, arg11, lys17) were involved in the interaction with MC4R; among these his9, phe10 and arg11 belong to the group of amino acids known to unify the melanocortins which are rich in his-phe-arg-trp. As reported by Ira and Tung (2003), these amino acids in fact constitute a key pharmacophore necessary for biological activity of melanocortin peptides. This “HFRW” motif was reported by Hruby et al., (1987) to be the binding sites of the melanocortins required for activation of the receptors. PyMOL package has also shown 16 residues of MC4R protein interacting with β-MSH peptide hormone (Fig 6 (a)). Majority of the interacting residues of the receptor protein is hydrophobic in nature and forms a part of transmembrane helices H6 (ile245, 249, thr246, 248, leu250 gly252, val253, 256, cys257) and H7 (ile297, 301, leu304). Remaining interacting residues were located in the cytoplasmic loop regions of MC4R. While three residues [phe216, met241 and lys242] were present in CL3, only one residue (arg305) of CL4 was involved in the interaction. These observations strongly suggest that β-MSH model has got docked into the pocket occupied by the residues present towards the C-terminal (i. e, H6, H7, CL3 and CL4) region of MC4R model. Schiöth et al., (1997) have noticed the absence of H4 and H5 as well as EL2 from the peptide binding pocket of the melanocortin receptors. It may also be noted, none of the residues of H1-5, CL1, CL2 as well as ELs of MC4R protein were involved in interaction with β-MSH peptide hormone. Though, none of the binding site residues of MC4R reported by Pogozheva et al., (2005) was interacting with β-MSH structure; residues belonging to H6 and H7 were involved in the interaction. This could be due to the fact that Pogozheva and his coworkers, have used linear polypeptide agonist “NDP-MSH” for docking against MC4R. Polar contacts Among the 16 residues of MC4R model interacting with β-MSH model, OD1 atom of asp 18 from β-MSH structure established polar contact by forming H bonds with NZ atom of lys242 and NH2 atom of arg304 of MC4R protein (Fig 6 (b)). Although, there are many experimental investigations which prove the interaction of β-MSH peptide with the MC4R protein (Abbott et al., 2000; Biebermann et al., 2006; Schiöth et al., 2002), none of these studies provide details of structural interaction of the proteins. The outcome from the present docking exercise provides a strong evidence of interaction at molecular level between β-MSH and MC4R proteins, thereby authenticating the earlier works.

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 223

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

MD simulation studies for computing free binding energy Various parameters such as temperature, total energy, kinetic energy, potential energy, root mean square deviation (RMSD), root mean square fluctuation (RMSF), free binding energy, structural superimposition, polar contacts and salt bridge formation were analysed from the output generated from the MD simulations of peptide hormone (β-MSH model), receptor (MC4R model) and docked complex structure (MC4R_β-MSH). Based on temperature and energy profiles of βMSH, MC4R and MC4R_β-MSH structures, these two parameters remained steady during the entire period of simulation. Kinetic energy was stable for all the three systems as expected, since temperature, which is directly proportional to kinetic energy also remained stable during the simulation process. Lowest energy structure of β-MSH at the end of additional 15 ns of simulation recorded a potential energy of – 677.01 kcal mol-1, which was lesser compared to the value (-660.69 kcal mol-1) obtained during the initial 50 ns of MD simulation. Further, it is evident that potential energy of the complex is lesser (- 6214.46 kcal mol-1) than that of receptor (-5442.89 kcal mol-1) (Table 2) suggesting the stability of the docked protein – protein complex. In a MD simulation studies conducted by Abhinav et al., (2011) on Hsp90/Cdc37 protein docked with the ligand ‘withaferin’, energy profile of the docked complex was lesser compared to the receptor protein, suggesting the stability of the docked drug-protein complex. Analysis of RMSD plot of MC4R and MC4R_β-MSH generated with reference to their corresponding lowest energy structure showed a similar trend (Fig 7). The Cα RMSDs of both structures showed a steady decrease during the first 9 ns. During later stage of simulation, these two curves showed stable fluctuations. Structural drift measured in terms of deviation of Cα atoms of MC4R and MC4R_β-MSH structures from 9 - 15 ns was about 1 Å with larger RMSD observed in the unbound form than in the complexed form, suggesting that the stability of the receptor structure might have enhanced upon peptide binding. Tao et al., (2010) noticed an increase in the stability of enzyme structure upon substrate binding, as a result of which the complexed form had lower backbone RMSD values than the unbound form. The extent of deviation of Cα atoms in MC4R and MC4R_β-MSH complex was higher during the initial stages of simulation. This is because the initial structures taken for our simulation studies were modelled structures. The deviation however, got diminished during the later stages of simulation. As the simulation progressed, β-MSH in association with MC4R tried to make interactions with the solvent and should have assisted in stabilizing the complex structure. The solvent effect has caused lowering of RMSD trajectory from ~ 9 ns till the end of simulation. Abhinav et al., (2011) have reported lowering of RMSD of the complex protein in relation to the receptor, when subjected to MD simulation. RMSD trajectory of β-MSH has retained its conformational stability when subjected to additional 15 ns of simulation (this was done for computing ∆G). This stability was noticed throughout the simulation period with RMSD values varying between 0.5 Å to 1.5 Å (Fig 7). Data presented in figure 8 shows the plotting of RMSF values for β-MSH, MC4R and MC4R_β-MSH models as a function of residue number. The observed overlapping of several peaks for MC4R and MC4R_β-MSH complex in fact corresponds to the loop regions of these protein structures. Magnitude of fluctuations of the residues in MC4R_βMSH model was however lesser, when compared to MC4R structure alone. This is because binding of receptor either to a peptide or a ligand, can cause decrease in the fluctuations of loop regions of the proteins. Tao et al., (2010) have observed peaks in the RMSF plot generated for serine protein kinase from fungus Tritirachium album when subjected to MD simulation. This has been attributed to the presence of residues located in the surface – exposed loops which are more prone to fluctuations. Binding of the substrate to the kinase has caused reduction in the fluctuations of these residues.

Fig. 1 Prediction of transmembrane helices of MC4Rfrom humans predicted by TMHMM server [Data in (a) shows the length of residues spanning the transmembrane and those that are inside and outside the membrane; (b) Graphical display of helices present as TM (coloured red), inside (coloured blue) and outside (coloured orange)]

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 224

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Similar explanation has been quoted by Tarus et al., (2012) while working on the MD simulation of nucleoprotein of Influenza A virus. Another reason for reduction in the fluctuation of peaks in RMSF plot of MC4R_β-MSH could be due to the formation of higher number of salt bridges, H-bonds, polar contacts made by MC4R with β-MSH compared to MC4R alone. As pointed out by Sandeep et al., (2012), the small range of RMSFs reflects slight structural rearrangement in the docking complex. Based on this outcome it is possible to conclude that loop regions of a protein in its unbound state is prone to more fluctuation rather than in its bound state. Binding electrostatic solvation free energy (∆Gel (bind)) calculated for MC4R_β-MSH was negative (-174.37 Kcal mol-1); thus, strongly favouring complex formation (Table 2). Many researchers have used electrostatic solvation energy as a measurement of stability of different protein conformations (Feig, Onufriev, Lee, & Brooks, 2003; Zhang, Gallicchio, Friesner, & Levy, 2000). However Gilson and Honig (1988) have pointed out that though the total electrostatic energy may provide a useful measure of stability, there is no prior reason to use it alone as a measure of stability. Instead the total free energy (potential energy in our case) of the protein, which is the sum of many contributions should be used (Novotny, Rashin, & Bruccoleri, 1998). In this regard, we have extended our energetic analysis by examining the free binding energy and the contribution of single energy components. Table 1 Template sequence profile (top 20 hits) for MC4R from human generated by PSI- BLAST tool.at the end of 20th iteration with PDB as the database [Threshold E-value = 0.001].

 

Free binding energy (∆G) of MC4R_β-MSH complex at the end of 15 ns simulation (Table 2) showed that, ∆G value was lesser (-136.09 kcal mol-1) in relation to the value obtained before the equilibration stage (-94.23 Kcal mol-1). This clearly demonstrates that complex when subjected to MD simulation has resulted in increasing the binding affinity of the receptor towards the ligand. An increase in the binding affinity of the complex protein could be attributed to the closer interaction of the active site residues of receptor and ligand as a consequence of MD simulation (Ismaila et al., 2012; Skovstrup, Laurent, Taboureau, & Flemming, 2012). Upon closer examination of the contributions of single energy components (table 2), it is evident that Coulomb interaction (referred to as EELEC in table 2) strongly favour complex formation (-775.05 kcal mol-1) opposed by disfavorable contributions (600.05 kcal mol-1) coming from electrostatic component of the energy of interaction with the solvent (referred to as EGB in table 2). The total electrostatic contribution computed however being negative (- 174.34kcal mol-1) favours complex formation. This is in agreement with the observation made by Muegge et al., (1998), Sheinerman and Honig (2002). Even though some of the internal energy components such as electrostatic interactions involving ‘torsion angles and between atoms separated by 3 bonds’ are positive disfavouring the complex formation, van der Waals interaction, one of the main component favouring the complex formation, was negative (-72.67 kcal mol-1). Holger and David (2004) have noticed that the binding affinity Ras-Raf protein arise from van der Waals interactions as well as nonpolar part of solvation free energy. MC4R_β-MSH complex upon MD simulation underwent conformational changes, which resulted in bringing many of the interacting residues of these two individual proteins in close proximity. While the number of interacting residues of MC4R protein increased from 16 before MD simulation to 22 at the end of MD simulation, those from β-MSH protein increased from 11 to 14. Solvation effect introduced during MD simulation caused conformational changes in MC4R_β-MSH structure, which is very clear when individual secondary structural elements (SSEs) of the complex structure were superimposed onto the initial structure and visualized using Chimera (Pettersen et al., 2004) package (Fig 9).

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 225

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

The RMSD data shown in figure 10 throws more insight into the conformational changes that has occurred in MC4R component of the complex protein structure as a result of simulation process. Among all the SSEs, H2 and EL3, followed by H3, H4 and H5, were the least fluctuating as reflected by their lower RMSD values. Since EL3 is a short loop, made up of only 3 residues, is less susceptible to fluctuation. One important reason for helices ‘H2-H5’ showing lesser degree of fluctuation is that they are sandwiched between H1 and H6 on either side; thereby having reduced solvent accessible area. On the contrary, H1 and CL4 being exposed towards the N- and C- terminal respectively, have more solvent accessible area, as a result are more prone to fluctuation. This is confirmed by their larger RMSD values (Fig 10). Between H6 and H7, the former had more residues interacting with β-MSH peptide, which is obvious from our docking studies. This could be one of the reasons for H6 having more fluctuation compared to H7. Due to uniform solvent accessibility, fluctuations noticed with respect to EL1, EL2, EL4, CL1-CL3 of MC4R model were more or less of same magnitude (Fig 10). This explains the effect of solvent in inducing conformational changes in the protein and thus favouring protein-protein interaction. Conformational changes were noticed by Hahnbeom et al., (2011) in ‘chemotaxis aspartate receptor Tar’, a protein involved in signal transduction, when subjected to MD simulation in explicit lipid bilayer.

Fig. 2 Structural alignment between MC4R (human) and 2RFM (human) generated by SPDBV (Guex & Peitsch, 1997) package. The representation of TM helices (H1 to H7) for both the sequences is based on the output generated by the TMHMM server. The dashed boxes represent the conserved motifs. [EL= extracellular loop; CL = cytoplasmic loop]

Fig. 3 Loop refined structure of MC4R from human, predicted by I-TASSER. The image CHIMERA package (Pettersen et al., 2004)

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

was generated by

Page: 226

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Fig. 4 (a) Lowest energy structure of βMSH model predicted by ab initio method (-660.69 kcal mol-1). The image was generated using PyMOL package. (b) Salt bridges formed between the O atom (glu) and N atoms (arg and lys) in the lowest energy structure of βMSH at the end of 50 ns MD simulations. Distances (Å) between the atoms are represented as dotted lines. The image was generated using VMD package

Fig. 5: RMSD of Cα backbone of βMSH model to its lowest energy structure at the end of 50 ns of MD simulation. The plot was generated using GRACE package

Fig. 6 (a) Active site residues of MC4R_βMSH docked complex before MD simulation [residues 1-5, 9-11, 16-18 belong to βMSH model; residues 216, 241, 242, 245, 246, 248 - 250, 252, 253, 256, 257, 297, 301, 304 and 305 belong to MC4R model]. (b) Polar contact (dotted lines) is seen between asp18 (OD1 atom) of β-MSH model and [lys241 (NZ) atom, arg305 (NH2) atom)] of MC4R model. The images were generated using PyMOL package

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 227

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Fig. 7 RMSD of Cα backbones of βMSH model (green), MC4R model (red) and MC4R_βMSH docked complex (black) to their corresponding lowest energy structures at the end of 15 ns of MD simulations. The plots were generated using GRACE package

Fig. 8 Fluctuations of Cα backbone of βMSH model (green), MC4R model (red) and MC4R_ βMSH docked complex (black) plotted as a function of residue number at the end of 15 ns of MD simulations The docked MC4R_ß-MSH complex upon MD simulation underwent a conformational change which has resulted in bringing many of the interacting residues of these two proteins much closer (Table 3). While the number of interacting residues from H6, H7, CL3 and CL4 of MC4R increased from 16 before MD simulation to 22 at the end of MD simulation, those from ß-MSH increased from 11 to 14. A drastic increase in the number of polar contacts is well supporting the fact of conformational change that has happened in MC4R_ß-MSH complex due to simulation. Before the onset of simulation, two N atoms of lys244 (NZ) and arg305 (NH2) residue of MC4R formed only 2 polar contacts with one O atom of asp18 (OD1) from β-MSH. After the completion of MD simulation, there was remarkable increase in number of atoms involved in the formation of polar contact. Figure 11 illustrates polar contacts formed between O and N atoms of various residues of β-MSH and MC4R structures. Oxygen atoms (OD1 and OD2) of asp1, 18 from β-MSH structure interacted with the nitrogen atoms of arg220, 310 (NH1, NH2), lys224 (NZ) and gln307 (NH2) of MC4R protein, resulting in maximum number of polar contacts. Similar interaction was seen between O atoms (O, OE2) of glu2 from β-MSH with N atoms (NH2) of arg220, 305 from MC4R model. In a deviation to this pattern, N atoms of arg11 (NH1) and his9 (NH2) from β-MSH model established polar contacts with O atom of gly252 from MC4R protein. Conformational changes in MC4R_β-MSH complex as a result of MD simulation could be confirmed by noticing an increase in the number of salt bridges (Table 4). Salt bridges were formed by the O atoms of aspartate and glutamate with the N atoms of arginine, histidine as well as lysine. Data presented in table 4 shows substantial increase in the number of salt bridges formed at intra and intermolecular level after the completion of simulation.

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 228

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Table 2 Summary of the contributions of different potential energy components of βMSH model, MC4R model and MC4R- βMSH docked complex considered for calculating binding free energy (∆G). The values reported here (kcal mol-1) are averages calculated at the end of 15 ns of MD simulation MC4R_ βMSH βMSH MC4R docked Contribution Model model complex Bond 64.00 1024.55 1088.30 -0.26 Angle 166.16 2748.65 2911.07 -3.74 Dihed 181.26 3448.57 3670.45 40.62 1-4 NB 59.57 1250.63 1304.35 -5.85 1-4 EEL 788.22 14628.99 15497.40 80.19 VdWaals -106.75 -2644.83 -2824.25 -72.68 EELEC -1395.52 -22441.77 -24612.35 -775.05 EGB -392.42 -3457.68 -3249.42 600.68 -635.47 -5442.90 -6214.46 ∆Gbind = -136.09 EPtot Gel -1787.94 -25899.46 -27861.77 ∆Gel(bind) = -174.37 Note (1) EPtot: Potential energy, Bond: electrostatic interaction between atoms separated by one bond (1-2), angle: electrostatic interaction between atoms separated by 2 bonds (1-3), dihed: electrostatic interactions involving torsion angles, 1-4 NB: van der Waals interactions between atoms separated by 3 bonds, 1-4 EEL: electrostatic interaction between atoms separated by 3 bonds, VdWaals: van der Waals interactions, EELEC: electrostatic interaction excluding energy of 1-4 terms (coulomb), EGB: Electrostatic components of the energy of the interaction with the solvent (2) Binding free energy (∆Gbind) = sum total of EMC4R_ βMSH – sum total of EMC4R – sum total of E βMSH (3) Electrostatic solvation free energy (Gel) = the sum of ∆Geelec and ∆Gsolv of individual model (4) Binding electrostatic solvation free energy (∆Gel (bind)) = Gel complex - Gel receptor- Gel peptide

Fig. 9 Superimposition of secondary structural elements (SSE) of MC4R_ βMSH docked complex (before and after MD simulation). While SSEs of MC4R_ βMSH before MD are represented as grey colour, they are shown as cyan after the completion of simulation. RMSD values (Å) for each of the superimposed structures are shown within parentheses. The images were generated using Chimera package

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 229

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Fig. 10 Root mean square deviation (RMSD) of secondary structural elements (H1 - H7: helices; EL1 - EL4: extracellular loops; CL1 – CL4: cytoplasmic loops) of MC4R model subjected to 15 ns of MD simulation. (The RMSD value computed by Chimera package is based on the superimposition of SSEs of MC4R (after MD simulation) onto its initial structure (before MD simulation)

Table 3: Summary of the residues interacting in docked MC4R_βMSH complex (before and after MD simulation) Before MD simulation MC4R

After MD simulation βMSH

MC4R

SSEs

SSEs

CL3

H6

H7

CL4

phe216 met241

ile245, 249 thr246, 248

ile297, 301 leu304

arg305

242

lys

250

leu

gly

252

val

253, 256

βMSH

257

cys

CL3

H6

H7

CL4

asp1,18

try211

thr248

ile297

arg305

asp1,18

glu2, 3

phe216

ile249,251

leu300

ser306

glu2, 3

pro4, 16

leu217

gly252

ile301

gln307

pro4, 15, 16

tyr5

arg220

val253,255,256

leu304

arg310

tyr5

his9

lys224

his9

phe10

asn240

phe10

arg11

lys242

arg6,11

lys17

trp12 lys17

Total

3

9

3

1

11

7

7

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

4

4

Page: 230

14

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Fig. 11 Three sets (i to iii) of polar contacts (dotted lines) established between the residues of MC4R_βMSH complex after 15 ns of MD simulation. Two O atoms of arg18from βMSH model were involved in 5 polar contacts with N atoms of arg310, gln307and lys242 from MC4R model [(OD2) – arg310 (NH2, NH1) – gln307 (NH2), OD1- gln307 (NH2) – lys242 (NZ)] and two O atoms of asp1 (OD2, OD1) and glu2 (O, OE2) from βMSH were involved in 4 polar contacts with N atoms of lys224 (NZ) arg220 (*NH2) and arg305 (NH2) from MC4R model respectively. O atom of gly252 was chelated with 2 polar contacts with N atoms of arg1 (NH1) and his9 (NH2) from βMSH model Table 4 Summary of salt bridges formed by MC4R_ βMSH complex (before and after MD simulation) computed by VMD package. Average distances between O-N atoms (Å) between the charged residue pairs are shown in parentheses Before MD Intra MC4R

After MD Inter

βMSH

asp327–arg331 (2.9)

glu2–arg6 (3.49)

asp189–arg22 (4.57)

glu8–lys17 (3.73)

Intra

Inter

MC4R_βMSH

MC4R

βMSH

MC4R_βMSH

arg305– asp18 (4.08)

asp327–arg331 (4.54)

glu2–arg6 (3.38)

arg310 – asp18 (3.89)

asp189–arg22 (3.19)

glu8–lys17 (3.83)

lys242 –asp18 (3.62)

glu308 –lys73 (3.89)

arg 305– glu2 (4.62)

asp146–arg147 (3.39)

lys224 – asp1 (3.19)

asp327–lys73 (2.80)

arg220 – asp1 (4.5)

asp189–arg18 (3.99) asp327–lys71 (3.62) glu29 –arg22 (3.43) glu100 – his6 (4.94) glu315–lys311 (3.00) glu29 – lys33 (3.82)

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 231

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

Fig. 12 Inter salt bridges formed by MC4R_ βMSH docked complex after 15 ns of MD simulation. Distances (Å) between O atoms (glu) and N atoms (arg, his and lys) are represented as dotted lines. The images were generated by VMD package from one single frame [Note: While table 4 displays average distances of salt bridges taken from various frames, figure 11 displays the distances calculated based on single conformational pose] 

Intra salt bridges

Two salt bridges that were initially formed within MC4R (asp327 – arg331, asp189 – arg 22) and β-MSH (glu2 – arg6, glu8 – lys17) components of the complex structure were retained even after the successful completion of simulation with minor variation noticed in the distances of the interacting atoms. While the number of intra salt bridges formed with MC4R component of the complex structure drastically increased, no such increase was seen for β-MSH component.

Inter salt bridges

Interestingly, as a consequence of solvent effect, intermolecular salt bridges increased (MC4Rarg310 – β-MSH asp21; 242 – βMSHasp18; MC4Rarg305 – βMSHglu2, MC4Rlys224 – βMSHasp1, MC4Rarg220 – βMSHasp1) (Fig 12), thereby improving MC4Rlys the stability of MC4R_ β-MSH complex structure. It is important to note that the lone inter salt bridge formed between MC4R and β-MSH models (MC4Rarg305– β-MSHasp18) before MD simulation got disrupted as a result of simulation and formed 2 new salt bridges (MC4Rarg310– β-MSHasp18; MC4Rarg305– β-MSHglu2). Bosshard et al., (2004) in a recent review on the role of salt bridges in stabilizing the protein stabilization have concluded that few salt bridges were either formed or dissolved for enhancing the stability as well as to facilitate thermodynamically favourable protein-protein interaction. From these observations, we conclude that a protein structure when subjected to a prolonged period of MD simulation attains conformational stability with increased number of polar contacts and salt bridges. Hahnbeom et al., (2011) in their studies on MD simulation of chemotaxis receptor protein have opined that conformational changes can alter the dynamics and conformation of the protein, which is perhaps a mechanism to deliver the signal from the transmembrane domain to the cytoplasmic domain. By drawing analogy from their observation, it is possible that conformational change as a result of solvent effect noticed in our simulation experiment should be a mechanism of bringing the interacting proteins much closer having higher binding affinity. Output of this research has confirmed that MC4R as a G-protein coupled receptor having affinity towards β-MSH peptide hormone. Docking followed by MD simulation studies has shed more insight in understanding the molecular interactions between MC4R and β-MSH models and conformational stability of the complex structure. This basic information could be useful in the development of melanocortin-based drugs.

Conflict of Interest Conflict of interest declared none ACKNOWLEDGEMENT We are grateful to Dr Chenraj Roychand, Chairman, JGI and Dr Sudha Deshmukh, Dean (Sciences), CPGS, Jain U’sty, Bangalore, India for creating the computational facilities. We also acknowledge Dr Sunil S More and Dr Rathore for giving useful points during the progress of this work. We also thank Mr. Anand of Lintechnokrat, Bangalore, India for the efficient maintenance of our computer cluster system.

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 232

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550 REFERENCES

Abbott, C. R., Rossi, M., Kim, M., AlAhmed, S. H., Taylor, G. M., Ghatei, M. A., Bloom, S. R. (2000). Investigation of the melanocyte stimulating hormones on food intake. Lack Of evidence to support a role for the melanocortin-3-receptor. Brain Res, 869, 203–210. Abdel-Malek, Z. A. (2001). Melanocortin receptors: their functions and regulation by physiological agonists and antagonists. Cell. Mol. Life Sci, 50, 434–441. Abhinav, G., Ashutosh, S., Vibhuti, A., Piyush, P., Virendra, S. B., Durai, S., & Divya, B. (2011). Hsp90/Cdc37 Chaperone/co-chaperone complex, a novel junction anticancer target elucidated by the mode of action of herbal drug Withaferin A. BMC bioinformatics, 12, S30. Adan, R. A. H., & Gispen, H. W. (1997). Brain Melanocortin Receptors: From Cloning to Function. Peptides., 18, 1279–1287. Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller, W., & Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein data base search programs. Nucleic Acids Res., 25, 3389-3402. Berendsen, H. J. C., Postma, J. P. M., van Gunsteren, W. F., DiNola, A., & Haak, J. R. (1984). Molecular dynamics with coupling to an external bath. J. Chem. Phys., 81, 3684–3690. Biebermann, B., Tamara, R. C., van Landeghem, F., von Deimling, A., Escher, F., Brabant, G., Krude, H. (2006). A role for β-melanocyte-stimulating hormone in human body-weight regulation. Cell, 3, 141–146. Bokoch, M. P., Zou, Y., Rasmussen, S. G., Liu, C. W., Nygaard, R., Rosenbaum, D. M., Kobilka, B. K. (2010). Ligand-specific regulation of the extracellular surface of a G-protein-coupled receptor. Nature 463, 108-112. Bosshard, H. R., Marti, D. N., & Jelesarov, I. (2004). Protein stabilization by salt bridges: concepts, experimental approaches and clarification of some misunderstandings. J. Mol. Recognit, 17, 1-16. Case, D. A., Darden, T. A., Cheatham, T. E., Simmerling, C. L., Wang, J., Duke, R. E. (2008). AMBER 10. Chhajlani, V., & Wikberg, J. E. (1992). Molecular cloning and expression of the human melanocyte stimulating hormone receptor cDNA. FEBS Lett, 309, 417-420. Colovos, C., & Yeates, T. O. (1993). Verification of protein structures: patterns of non- bonded atomic interactions. Protein Sc., 2, 1511-1519. DeLano, W. L. (2002). The PyMOL molecular graphics system, Version 1.5.0.4 Schrödinger, LLC. Feig, M., Onufriev, A., Lee, M. S., & Brooks, C. L. (2003). Performance comparison of generalized born and poisson methods in the calculation of electrostatic solvation energies for protein structures. J Comput Chem, 25, 265– 284. Fiser, A., Gian Do, R. K., & Šali, A. (2000). Modeling of loops in protein structures. Protein Sciences, 9, 753–1773. Froloff, N., Windemuth, A., & Honig, B. (1997). On the calculation of binding free energies using continuum methods: Application to MHC class I protein-peptide interactions. Protein Science, 6, 1293-1301. Giblin, M. F., Wang, N., Hoffman, T. J., Jurisson, S. S., & Quinn, T. P. (1998). Design and characterization of alphamelanotropin peptide analogs cyclized through rhenium and technetium metal coordination. Proc.Natl.Acad.Sci.USA, 95, 12814-12818. Gilson, M. K., & Honig, B. (1988). Calculation of the total electrostatic energy of a macromolecular system: Solvation energies, binding energies, and conformational analysis. Proteins Struct. Funct. Genet, 4, 7-18. Giuliani, D., Minutoli, L., Ottani, A., Spaccapelo, L., Bitto, A., Galantucci, M., Guarini, S. (2012). Melanocortins as potential therapeutic agents in severe hypoxic conditions. Frontiers in Neuroendocrinology, 33, 179–193. Gu, W., Tu Z, Kleyn PW, Kissebah A, Duprat L, Lee J, . . . DB., A. (1999). Identification and functional analysis of novel human melanocortin-4 receptor variants. Diabetes, 48, 635-639. Guex, N., & Peitsch, M. C. (1997). SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis, 18, 2714-2723. Hahnbeom, P., Wonpil, I., & Chaok, S. (2011). Transmembrane signaling of chemotaxis receptor tar: Insights from molecular dynamics simulation studies. Biophys. J, 100, 2955–2963 Haskell-Luevano, C., Sawyer, T. K., Trumpp-Kallmeyer, S., Bikker, J. A., Humblet, C., Gantz, I., & Hruby, V. J. (1996). Three-dimensional molecular models of the hMC1R melanocortin receptor: complexes with melanotropin peptide agonists. Drug Des Discov, 14, 197-211. Hawkins, G. D., Cramer, C. J., & Truhlar, D. G. (1996). Parametrized models of aqueous free energies of solvation based on pairwise descreening of solute atomic charges from a dielectric medium. J. Phys. Chem, 100, 19824–19839.  

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 233

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

  Hinney, A., Schmidt, A., Nottebom, K., Heibült, O., Becker, I., Ziegler, A., Hebebrand, J. (1999). Several mutations in the melanocortin-4 receptor gene including a nonsense and a frameshift mutation associated with dominantly inherited obesity in humans. J. Clin. Endocrinol. Metab, 84, 1483-1486. Holger, G., & David, A. C. (2004). Converging Free Energy Estimates_ MM_PBGBSA studies on the protein proten complex ras raf. J. Comput. Chem, 25, 238–250. Holm, L., & Rosenström, P. (2010). Dali server: conservation mapping in 3D. Nucleic Acids Res., 38, W545-549. Hornak, V., Abel, R., Okur, A., Strockbine, B., Roitberg, A., & Simmerling, C. (2006). Comparison of multiple Amber force fields and development of improved protein backbone parameters. Proteins, 65, 712-725. Hruby, V. J., Wilkes, B. C., Hadley, M. E., Al-Obeidi, F., Sawyer, T. K., Staples, D. J., Hintz, M. F. (1987). alphaMelanotropin: the minimal active sequence in the frog skin bioassay. J Med Chem, 30, 2126-2130. Humphrey, W., Dalke, A., & Schulten, K. (1996). VMD - Visual Molecular Dynamics. J. Molec. Graphics, 14, 33-38. Huszar, D., Lynch, C. A., Fairchild-Huntress, V., Dunmore, J. H., Fang, Q., Berkemeier, L. R., Lee, F. (1997). Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell, 88, 131-141. Ira, G., & Tung, M. F. (2003). The melanocortin system. Am. J. Physiol. Endocrinol. Metab., 284, E468-E474. Ira, G., Yoshitaka, K., Takao, T., Yoshimasa, S., Hiroto, M., Gerd, M., . . . Tadataka, Y. (1993). Molecular cloning of a novel melanocortin receptor. J. Biol.Chem, 268, 8246-8250. Ismaila, Y. S., Wong, E. L., Kwee, H. J. T., Mohd, S. S., Haryati, J., & Fahrul, H. (2012). Structure prediction, molecular dynamics simulation and docking studies of d-specific dehalogenase from rhizobium sp. RC1. Intl. J. Mol. Sc, 13, 15724-15754. Johnson, M. S., & Overington, J. P. (1993). A structural basis for sequence comparisons: an evaluation of scoring methodologies. J. Mol. Biol, 233, 716-738. Kleywegt, K. J., & Jones, T. A. (1996). Phi/Psi-chology: Ramachandran revisited. Structure, 4, 1395–1400. Krogh, A., Larsson, B., Von Heijne, G., & Sonnhammer, E. L. L. (2001). Predicting transmembrane protein topology with a hidden Markov model: Application to complete genomes. J. Mol. Biol., 305, 567-580. Laskowski, R. A., MacArthur, M. W., & Thornton, J. M. (1993). PROCHECK: A program to check the stereochemical quality of protein structures. J. Appl. Crystallogr., 26, 283-291. Michael, B. Z., & Hang, S. B. M. (2000). Pro-opiomelanocortin (POMC) deficiency and peripheral melanocortins in obesity. Nutr.Rev., 58, 177–180. Moukhametzianov, R., Tony, W., Patricia, C. E., Maria, J. S. V., Andrew, G. W. L., Christopher, G. T., & Gebhard, F. X. S. (2011). Two distinct conformations of helix 6 observed in antagonist-bound structures of a β 1 adrenergic receptor. Proc. Natl. Acad. Sci. USA, 108, 8228–8232 Muegge, I., Schweins, T., & Warshel, A. (1998). Electrostatic contributions to protein_protein binding affinities_application to Rap_Raf interaction. Proteins, 30, 407-423. Mukherjee, S., & Zhang, Y. (2011). Protein-protein complex structure predictions by multimeric threading and template recombination. Structure, 19, 955-966. doi: 10.1016/j.str.2011.04.006 Nakanishi, S., Inoue, A., Kita, T., Nakamura, M., Chung, A. C. Y., Cohen, S. N., & Numa, S. (1979). Nucleotide sequence of cloned cDNA for bovine corticotropin-beta-lipotropin precursor. Nature, 278, 423–427. Novotny, J., Rashin, A., & Bruccoleri, R. E. (1998). Criteria that distinguish between native proteins and incorrectly folded models. Proteins, 4, 19-30. Padhi, A. K., Kumar, H., Vasaikar, S. V., Jayaram, B., & Gomes, J. (2012). Mechanisms of loss of functions of human angiogenin variants implicated in amyotrophic lateral sclerosis. PLoS ONE, 7, e32479. Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G., Greenblatt, D. M., Meng, E. C., & Ferrin, T. E. (2004). UCSF Chimera: A visualization system for exploratory research and analysis. J. Comput. Chem, 35, 16051612. Pogozheva, I. D., Chai, B.-X., Lomize, I. L., Fong, T. M., Weinberg, D. H., Nargund, R. P., Mosberg, H. I. (2005). Interactions of Human Melanocortin 4 Receptor with Non Peptide and Peptide Agonists. Biochemistry, 44, 11329–11341. Prabhavathi, M., Ashokkumar, K., Geetha, N., & Saradha Devi, K. M. (2011). Homology modeling and structure prediction of thioredoxin (TRX) protein in wheat. Intl. J. Biosci, 1, 20-32. Ritchie, A. W. (2003). Evaluation of protein docking predictions using Hex 3.1 in CAPRI rounds 1 and Proteins. Proteins: Struct., Funct., Bioinf., 52, 98–106. Ryckaert, J. P., Ciccotti, G., & Berendsen, H. J. C. (1997). Numerical integration of the cartesian equations of motion of a system with constraints: Molecular dynamics of n-alkanes. J. Comput. Phys., 23, 327–341. Sandeep, S., Priyadarshini, V., Pradhan, D., Munikumar, M., & Umamaheswari, A. (2012). Docking and molecular dynamics simulations studies of human protein kinase catalytic subunit alpha with antagonist. J. Clin. Sci. Res, 1, 15-23.

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 234

Mutangana Dieudonné and Ramesh

Coden: IJABPT, Copyrights@2014, ISSN : 0976-4550

  Schiöth, H. B., Bouifrouri, A. A., Rudzish, R., Muceniece, R., Watanobe, H., Wikberg, J. E., & Larhammar, D. (2002). Pharmacological comparison of rat and human melanocortin 3 and 4 receptors in vitro. Regul. Pept, 106, 7–12. Schiöth, H. B., Muceniece, R., Mutulis, F., Prusis, P., Lindeberg, G., Sharma, S. D., Wikberg, J. E. S. (1997). Selectivity of w 7 xw 7 x cyclic D-Nal and D-Phe substituted MSH analogues for the melanocortin receptor subtypes. Peptides, 18, 1009–1013. Sheinerman, F. B., & Honig, B. (2002). On the role of electrostatic interactions in the design of protein-protein interfaces. J. Mol. Biol, 318, 161-177. Skovstrup, S., Laurent, D., Taboureau, O., & Flemming, S. J. (2012). A steered molecular dynamics study of binding and translocation processes in the GABA transporter. PLoS ONE, 7, e39360. Smith, & Honig, B. (1994). Evaluation of the conformational free energies of loops in proteins. Proteins Struct. Funct. Genet, 181, 19-132. Stambulchik, E. (Producer). (1997). Xmgrace. Retrieved from http://plasma-gate.weizmann.ac.il/Xmgr Still, W. C., Tempczyk, A., H., R. C., & H., T. (1990). Semianalytical treatment of solvation for molecular mechanics and dynamics. J. Am. Chem. Soc, 112, 6127-6129. Suzuki, I., Cone, R. D., Im, S., Nordlund, J., & Abdel-Malek, Z. A. (1996). Binding of melanotropic hormones to the melanocortin receptor MC1R on human melanocytes stimulates proliferation and melanogenesis Endocrinology, 137, 1627-1633. Tarus, B., Chevalier, C., Richard, C. A., Delmas, B., Di Primo, C., & Slama-Schwok, A. (2012). Molecular dynamics studies of the nucleoprotein of influenza a virus: Role of the protein flexibility in RNA binding. PLoS ONE, 7, e30038. Thompson, A. A., Liu, W., Chun, E., Katritch, V., Wu, H., Vardy, E., . . . Stevens, R. C. (2012). Structure of the nociceptin/orphanin FQ receptor in complex with a peptide mimetic. Nature, 485, 395-399. Van der Ploeg, L. H. T., Martin, W. J., Howard, A. D., Nargund, R. P., Austin, C. P., Guan, X., MacIntyre, D. E. (2002). A role for the melanocortin 4 receptor in sexual function. Current isssues, 99, 11381–11386. Vogel, R., Mahalingam, M., Ludeke, S., Huber, T., Siebert, F., & Sakmar, T. P. (2008). Functional role of the ‘‘ionic lock’’- an interhelical hydrogen-bond network in family A heptahelical receptors. J. Mol. Biol, 380, 648–655. Wang, Z. X., Zhang, W., Wu, C., Lei, H., ; , Cieplak, P., & Duan, Y. (2006). Strike a balance: Optimization of backbone torsion parameters of AMBER polarizable force field for simulations of proteins and peptides. J. Comp. Chem, 27, 781–790. Yan Tao , Zi-He Rao, & Shu-Qun. (2010). Insight derived from molecular dynamics simulation into substrate-induced changes in protein motions of proteinase k. J. Biomol. Struc. Dyn, 28, 143-157. Zhang, Y (2008). I-TASSER server for protein 3D structure prediction. BMC bioinformatics, 9, 40-47. Zhang, Y., Gallicchio, E., Friesner, R. A., & Levy, R. M. (2000). Solvent models for protein–ligand binding: Comparison of implicit solvent poisson and surface generalized Born models with explicit solvent simulations. J Comp Chem, 22, 591–607

International Journal of Applied Biology and Pharmaceutical Technology Available online at www.ijabpt.com

Page: 235