- freely available
Molecules 2015, 20(6), 10689-10704; doi:10.3390/molecules200610689
Abstract: Herpes simplex virus infections have been described in the medical literature for centuries, yet the the drugs available nowadays for therapy are largely ineffective and low oral bioavailability plays an important role on the inefficacy of the treatments. Additionally, the details of the inhibition of Herpes Virus type 1 are still not fully understood. Studies have shown that several viruses encode one or more proteases required for the production new infectious virions. This study presents an analysis of the interactions between HSV-1 protease and benzoxazinone derivatives through a combination of structure-activity relationships, comparative modeling and molecular docking studies. The structure activity relationship results showed an important contribution of hydrophobic and polarizable groups and limitations for bulky groups in specific positions. Two Herpes Virus type 1 protease models were constructed and compared to achieve the best model which was obtained by MODELLER. Molecular docking results pointed to an important interaction between the most potent benzoxazinone derivative and Ser129, consistent with previous mechanistic data. Moreover, we also observed hydrophobic interactions that may play an important role in the stabilization of inhibitors in the active site. Finally, we performed druglikeness and drugscore studies of the most potent derivatives and the drugs currently used against Herpes virus.
The Herpesviridae family is one of the major viral families. It includes 100 identified viruses that affect almost all animal species . The infections caused by Herpes simplex virus type 1 (HSV-1) are prevalent worldwide and have been described in the medical literature for centuries, but they remain incurable [2,3]. Clinical manifestations of HSV infections can range from asymptomatic or symptomatic versions (e.g., oral, labial, mucocutaneous, ocular, genital herpes, eczema herpeticum) to central neurological complications such as neonatal herpes, herpes encephalitis and even a fatal dissemination, mainly seen in immunocompromised patients [4,5].
Nowadays, several drugs are used in HSV therapy [6,7,8]. Although the current therapy against these viruses presents a safe profile, the low oral bioavailability plays an important role in compromising the treatment efficacy, therefore, the discovery of new potent and safe molecules with high oral bioavailability is of considerable interest.
In the last 20 years, the literature has described that several viruses that codify one or more proteases involved in protein catalysis or the capsid maturation process, a crucial process for virion production [9,10]. One example is the HSV-1 serine protease, which participates in the viral capsid organization process. This serine protease has emerged as an important therapeutic target for the development of new antiviral agents due to its structural features such as non-homologous folding [11,12,13,14]. Several serine protease inhibitors of herpes viruses have been reported in the recent years. In general, the mechanism of the protease inhibitors involves the classical reaction with the serine residue of the active site (Ser-His-His catalytic triad) thus inactivating the HSV enzyme .
The benzoxazinones were identified in 1960 as secondary metabolites of grasses and currently are known as serine protease inhibitors [16,17,18]. The inhibitory mechanism of benzoxazinones against HSV-1 protease was proposed by Jarvest and coworkers , who suggested the formation of an acyl-enzyme complex with Ser129 from the catalytic triad. Further evidences also pointed to interactions with the residues His61, His148, Arg156 and Arg157 as important for HSV-1 protease inhibition [11,15,18,19].
Advances in computer technology have driven the use of in silico modeling techniques, including molecular docking, in drug discovery and development research . This technique can be very helpful to discover new molecules since it enables the study of intermolecular complexes. In cases where the tridimensional structure of the target is not available, comparative modeling can be used as a tool to predict the tridimensional structure of the protein/enzyme before conducting the docking evaluation . Other parameters such as druglikeness and drugscore calculations may also be used to rationalize how the physicochemical properties may influence in vivo studies .
In order to study the active site profile and the binding mode of HSV-1 protease with a series of potential inhibitors and to propose a lead compound, herein we present the structure-activity relationship (SAR), comparative modeling, molecular docking and in silico pharmacokinetic studies of a series of benzoxazinones .
2. Results and Discussion
2.1. Structure-Activity Relationship
In order to determine the structural and stereoelectronic features that could determine the anti HSV-1 protease profile, first we evaluated the structure-activity relationships (SAR) of the studied benzoxazinone series. The activities were previously obtained by Jarvest and co-workers . The comparison of the two groups (oxy- and aminobenzoxazinones) showed better antiviral activity for the oxybenzoxazinones.
This biological activity can be associated to the electronegativity of the oxygen compared to the nitrogen atom (Table 1) and the capacity to be only a hydrogen bond acceptor. The structure-activity relationship analysis of this series suggested that bulk, lipophilic and aliphatic substituents on the R1 position decreased or even abolished (compound 9) the inhibitory activity. Besides, deactivating groups linked to a phenyl in R1 could also decrease the activity (compound 7), highlighting a high electronic density as a requirement in this region.
Analysis of the R2 position, specially comparing asparagine (Asn) and alanine (Ala) residues, revealed Asn as deleterious for activity for both amino- and oxybenzoxazinone compounds (4, 5, 13 and 14, Table 1). In contrast, substitution of an Ala residue at R2 position increases the activity, which can be related to its hydrophobic feature and volume.
The occurrence of hydrophobic interactions should be considered for these series since this type of interactions is related to atom polarizability and lipophilicity and may be important for the activity. Lipophilic and polarizable groups located on the R3 position seem to be good for the inhibitory activity, as observed in compounds 13 and 15. The same influence is observed for substitutions at R4 position as shown by compound 15 (R4 = Cl). Interestingly, these studies revealed some important structural features of these molecules apparently related to the activity (Table 1). The most active molecules presented molecular weights and the volumes ranging from 269.2 to 445.9 amu and 261.9 to 424.4 Å3, respectively. Molecules showing values above or below the highlighted ranges were less active (Table 1). In contrast, the analysis of the calculated electronic parameters of these compounds (such as EHOMO, ELUMO and dipole moment) and the correlation with activities did not show any direct relationships (data not shown).
|Compound||X||R1||R2||R3||R4||IC50 (µM)||MW (amu)||MV (Å3)||cLog P|
2.2. Comparative Modeling
The crystal structure of HSV-2 protease (PDB ID: 1AT3) was selected as potential template as the result of a BLAST-P search. The three-dimensional structure of HSV-1 protease was constructed employing SWISSMODEL and Modeller (Figure 1), as previously described. Stereochemical validation of the models structures is an important step of comparative modeling, used to evaluate the accuracy of the 3D-structure representation.
The Ramachandran Plot exhibited 90.8% of residues of the model obtained with Modeller localized in most favored regions and none in disallowed regions. Meanwhile the model generated from SWISS-MODEL showed 83.7% of residues localized in most favored regions and 1.5% in disallowed regions. Verify3D results showed 87.9% and 81.1% of the amino acids in the Modeller and SWISS-MODEL models have compatible 1D-3D scores greater than 0.2 (Table 2) suggesting that both models have overall self-consistency in terms of sequence-structure compatibility.
|Ramachandran plot a (%)||90.80||83.7|
|Ramachandran plot b (%)||0||1.5|
|Verify 3D c (%)||87.93||81.3|
a Residues on most favored regions. b Residues in disallowed regions. c Average 3D-1D profile score for each residues in a 21-residue sliding window.
Thus, according to our validation data (Table 2), the best model was obtained with the Modeller program. Both models presented quite separated catalytic residues, allowing the entry and cleavage of a polypeptide substrate at the active site.
2.3. Molecular Docking
2.3.1. Docking Validation
The validation of the docking protocol was carried out through a re-docking using the crystal structure of Cytomegalovirus protease (PDB ID: 1NJU), as it presents 30% of identity with HSV-1 protease with a highly conserved active site. Moreover it presents an inhibitor similar with the benzoxazinones studied herein. The ligand was removed from the enzyme and re-docked keeping the backbone of the peptidomimetic rigid. The comparison of re-docking results with the co-crystallized conformation showed a high success rate, with a root mean square deviation (RMSD) of 1.21 Å. These data supported the hypothesis that the experimental binding mode could be reproduced with accuracy using this protocol.
2.3.2. Docking Analysis
The benzoxazinones backbone was treated as in validated protocol where the peptidic moiety was kept rigid whereas the non-peptidic moiety and side chains were considered flexible (Figure 2). Each benzoxazinone was docked into the HSV-1 protease active site and their interactions were then analyzed.
The most active benzoxazinone (compound 15) is involved in four hydrogen bonds with Thr132 (distance of 3.2 Å), Ser129 (distance of 3.0 Å) and Arg157 (distances of 3.1 and 2.9 Å) residues of HSV-1 protease (Figure 3). The importance of Arg157 and Ser129 to the activity of HSV-1 protease has been previously reported [15,19]. The hydrogen bond between benzoxazinone 15 with Ser129, which is a member of catalytic triad and a nucleophile when the enzyme is activated, seems to be one of the most important interactions to the inhibitory activity as this is the only benzoxazinone that interacts with this residue.
Additionally, the hydrogen bond with Arg157 makes this residue less available to stabilize the nucleophilic attack by Ser129, inactivating the enzyme. Compound 15 also presented hydrophobic interactions with Leu38, Leu130, Ala131, Cys152, Ala153, Ile154 residues, contributing to stabilize this inhibitor in the enzyme (Figure 3).
Comparison of compounds 15 and 6, which are structurally similar, but with different inhibitory profile, revealed that compound 6 conserved only two hydrogen bonds with the oxygen of Ser129 and nitrogen of Arg157 (Figure 4).
The absence of the chlorine atom located at the R3 position could be related to the poor activity of compound 6, corroborating the SAR analysis. Comparison of the docking of compounds 18, 17 and 8 which present similar substituents at the R1 position indicated the influence of the number and type of interactions on activity. Compound 18, the second most active, showed hydrogen bonds with the Thr132 backbone nitrogen and with the side chain nitrogens of the Arg156 and Arg157 residues with distances of 3.4, 2.9 and 3.1 Å, respectively. Compound 17 presented hydrogen bonds with the Thr132 backbone nitrogen (distance of 3.0 Å) and the Arg157 side chain nitrogen (distance of 2.8 Å). Compound 8 is the less active among the three and presented only one hydrogen bond with the Thr132 backbone nitrogen and a cation-π interaction with Arg156 with distances of 3.4 Å and 3.9 Å, respectively. Cation-π interactions present lower energy than hydrogen bonds  and may explain the ca. 6- and 24-fold difference in activity compared to compounds 17 and 18, respectively. It is important to note that 18 is the only compound that presented an interaction with Arg157, which is probably directly related to its good activity (Figure 4).
As can be observed by comparing the docking results of aminobenzoxazinones 2, 4, 5 and oxybenzoxazinones 11, 13, 14, bulky substituents at position R2 as well as the absence of substituents at this position seem to decrease the activity, reinforcing the SAR analysis (Figure 5). In contrast, the presence of chlorine atom at positions R3 and specially R4 increased the activity. The presence of this type of substituent triggers the positioning of the compounds away from the Ser129. Nevertheless, compound 13 has shown an activity profile (IC50 = 7 µM) probably related to a hydrogen bond with the Arg156 side chain nitrogen with a distance of 3.1 Å. Besides hydrogen bonds, results pointed to the orientation of a chlorine atom towards a hydrophobic cavity as important to increase the activity (Figure 5).
Compound 1 presented no hydrogen bonds with HSV-1 protease, which may explain its low activity (Figure 6). However, it was well positioned in the HSV-1 protease catalytic site, probably due to the hydrophobic interaction between the apolar chain and hydrophobic amino acids of the enzyme. The hydrophobic interactions seems to be deeply associated with the positioning of compounds but are not the unique factors to increase activity. We could infer that hydrogen bond interactions are oriented by hydrophobic interactions of the R2 substituents.
The benzoxazinone 15 showed a particular binding mode with HSV-1 protease, characterized by the presence of a hydrogen bond with Ser129. This evidence supports the hypothesis that the mentioned interaction could facilitate the approximation of the nucleophile Ser129 to the carbonyl. The proximity of the atoms consequently could lead to the reaction for covalent bond formation, which would be in accordance with the proposed mechanism of action .
To evaluate this hypothesis, we also performed the covalent docking between benzoxazinone 15 and HSV-1 protease. As expected, the results showed the presence of the covalent bond between the Ser129 side chain and the benzoxazinone (Figure 7). Other interactions were also observed as a cation-π and a hydrogen bond with Arg156 with a distance of 3.1 Å (Figure 7) [15,19].
2.4. In Silico Analysis of Pharmacokinetic Properties
Druglikeness and drugscore studies were carried out to compare the most actives benzoxazinones (4, 13, 15, 18) with some currently available drugs against HSV including acyclovir, penciclovir, famciclovir and trifluridine. It is important to highlight that all benzoxazinones fulfilled Lipinski’s rule of five, and therefore theoretically presented good oral bioavaliability . Although the benzoxazinone 15 was the most active compound, the best druglikeness and drugscore results were achieved for compound 18 that was also better than the commercial drugs (Figure 8). The druglikeness evaluation revealed a positive value only for compound 18 in addition to the higher drugscore for this compound.
We also performed toxicity risks analysis using Osiris Property Explorer. Compounds 15 and 18 showed better mutagenicity, tumorigenicity, irritant and reproductive effect results than the commercial drugs (Figure 9). Taken together these results suggest that 18 could be an interesting lead compound for further studies and structural modifications.
3. Experimental Section
3.1. Structure-Activity Relationship
We constructed the three-dimensional structures of 18 benzoxazinone HSV-1 protease inhibitors previously described by Jarvest and coworkers (Table 1) . All calculations were performed using SPARTAN’10 (Wavefunction Inc. Irvine, CA, USA). The structures were minimized and a conformational analysis was done using Molecular Mechanics Force Field (MMFF) . The equilibrium geometry was obtained in vacuum using a semi-empirical RM1 module. To obtain the electronic properties of the RM1 minimal energy conformation of each benzoxazinone, they were submitted to a Density Functional Theory (DFT) B3LYP calculation with a 6-31G* basis set. The electronic (HOMO and LUMO energy, electrostatic potential and dipole moment) and structural properties (molecular weight, area, number of hydrogen bonds donors and acceptors) were calculated for all compounds.
3.2. Comparative Modeling
Since there is no crystal structure of HSV-1 protease currently available, we built a three-dimensional structure employing comparative modeling. The amino acid sequence was retrieved from the Uniprot database  (access number Q69087). BLAST-P  was used to identify homologous structures by searching the structural database of protein sequences in the Protein Data Bank. Based on the highest percentage of sequence identity (71.7%) the crystal structure of HSV-2 protease was selected as structural template (PDB ID: 1AT3, 2.5 Å resolution) . Models were generated using two different methods: (a) through SWISS-MODEL server on the automatic mode  and (b) performing the alignment with T-Coffee  server followed by model construction using Modeller program [29,30,31,32,33,34,35]. The overall stereochemical quality of the developed models was assessed by using PROCHECK [36,37,38,39] and the environment profile was checked using Verify-3D [40,41]. All validations were performed within SAVES server.
3.3. Molecular Docking
In order to investigate the binding mode of a series of benzoxazinones we performed molecular docking studies. The AutoDock 4.2 program was used  running on a Windows-based PC. First, the 3D structures of all benzoxazinones were built as described in the SAR methodology. The ligands were partially flexibly docked to HSV-1 and the docking files were prepared using AutoDockTools (ADT). The protein was treated as rigid, polar hydrogen atoms were added, nonpolar hydrogen atoms were merged and Gasteiger charges were assigned by default. The grid map dimensions were 36 × 40 × 60 points with 0.375 Å spacing and the coordinates to centralization were defined by validation step. Docking studies were carried out using the empirical free energy function and the Lamarckian Genetic Algorithm (LGA). The default parameters of LGA modified for calculations were elitism (15), crossover rate (0.6), number of energy evaluations (25,000,000) and population size (150). Results were analyzed using PyMol program (The PyMOL Molecular Graphics System, Version 1.3, Schrödinger, LLC, San Francisco, CA, USA).
We also performed a covalent docking between the most active benzoxazinone 15 and the HSV-1 protease, with the software GOLD Suite [43,44,45,46]. The molecular docking started setting the default parameters for Serine Proteases, following the binding cavity centralization in the side chain oxygen of Ser 129. The dimension of 15 radius was selected to the binding cavity. The GOLD fitness function used was Chemscore, allowing the rescore by Goldscore. A total of 50 solutions were obtained and the 5 best solutions were analyzed using PyMol program.
3.4. In Silico Pharmacokinetic Properties
There are many approaches that assess the druglikeness of compounds based on topological descriptors, fingerprints of molecular druglikeness, drugscore, structural keys or other properties as clogP and molecular weights. In this work we used the Osiris program  to calculate, for the most active compounds, the toxicity risks and the fragment-based druglikeness and drugscore, which combines the molecular properties above.
In this work we present the construction and validation of the HSV-1 protease structure and a molecular docking study between this enzyme and eighteen previously tested benzoxazinones. The results confirmed the importance of some residues including Arg156, Arg157 and mainly Ser129 for protease activity and of hydrophobic interactions on ligand anchoring, not yet described. The involvement of Ser129 in an exclusive hydrogen bond with benzoxazinone 15, the most active of the series, reinforced the importance of this residue for activity. Although compound 15 was the most active, druglikeness and drugscore studies revealed compound 18 to have the best overall profile. These findings could help in drug design and indicate compound 18 as a potential hit in the search for new drugs against HSV-1.
This work was supported by Capes and CNPq schoolarships.
J.F.R.M., M.A.B., A.M.T.S., A.C.R.S. and C.R.R. conceived and designed the experiments; J.F.R.M., N.C.B., R.O. performed the experiments; J.F.R.M. analyzed the data; H.C.C., L.M.C., L.A.M. and C.R.R. contributed reagents/materials/analysis tools; J.F.R.M., B.A.A.V., M.A.B., A.M.T.S. and H.C.C. contributed to the writing of the manuscript.
Conflicts of Interest
The authors declare no conflict of interest.
- Marnett, A.B.; Nomura, A.M.; Shimba, N.; Ortiz de Montellano, P.R.; Craik, C.S. Communication between the active sites and dimer interface of a herpesvirus protease revealed by a transition-state inhibitor. Proc. Natl. Acad. Sci. USA 2004, 101, 6870–6875. [Google Scholar] [CrossRef] [PubMed]
- Ding, X.; Sanchez, D.J.; Shahangian, A.; Al-Shyoukh, I.; Cheng, G.; Ho, C.M. Cascade search for hsv-1 combinatorial drugs with high antiviral efficacy and low toxicity. Int. J. Nanomed. 2012, 7, 2281–2292. [Google Scholar]
- Hewlett, G.; Hallenberger, S.; Rubsamen-Waigmann, H. Antivirals against DNA viruses (hepatitis b and the herpes viruses). Curr. Opin. Pharmacol. 2004, 4, 453–464. [Google Scholar] [CrossRef] [PubMed]
- Chida, Y.; Mao, X. Does psychosocial stress predict symptomatic herpes simplex virus recurrence? A meta-analytic investigation on prospective studies. Brain Behav. Immun. 2009, 23, 917–925. [Google Scholar] [CrossRef] [PubMed]
- Kinchington, P.R.; Leger, A.J.; Guedon, J.M.; Hendricks, R.L. Herpes simplex virus and varicella zoster virus, the house guests who never leave. Herpesviridae 2012, 3, 5. [Google Scholar] [CrossRef] [PubMed]
- De Clercq, E. In search of a selective therapy of viral infections. Antiviral Res. 2010, 85, 19–24. [Google Scholar] [CrossRef] [PubMed]
- Eizuru, Y. Development of new antivirals for herpesviruses. Antivir. Chem. Chemother. 2003, 14, 299–308. [Google Scholar] [CrossRef] [PubMed]
- Behrens-Baumann, W. Herpes simplex keratitis. A short overview of the current therapy. Klin. Monbl. Augenheilkd. 2010, 227, 388–392. [Google Scholar] [CrossRef] [PubMed]
- Babe, L.M.; Craik, C.S. Viral proteases: Evolution of diverse structural motifs to optimize function. Cell 1997, 91, 427–430. [Google Scholar] [CrossRef]
- Krausslich, H.G.; Wimmer, E. Viral proteinases. Annu. Rev. Biochem. 1988, 57, 701–754. [Google Scholar] [CrossRef] [PubMed]
- Hoog, S.S.; Smith, W.W.; Qiu, X.; Janson, C.A.; Hellmig, B.; McQueney, M.S.; O’Donnell, K.; O’Shannessy, D.; DiLella, A.G.; Debouck, C.; et al. Active site cavity of herpesvirus proteases revealed by the crystal structure of herpes simplex virus protease/inhibitor complex. Biochemistry 1997, 36, 14023–14029. [Google Scholar] [CrossRef] [PubMed]
- Reiling, K.K.; Pray, T.R.; Craik, C.S.; Stroud, R.M. Functional consequences of the kaposi’s sarcoma-associated herpesvirus protease structure: Regulation of activity and dimerization by conserved structural elements. Biochemistry 2000, 39, 12796–12803. [Google Scholar] [CrossRef] [PubMed]
- Shahiduzzaman, M.; Coombs, K.M. Activity based protein profiling to detect serine hydrolase alterations in virus infected cells. Front. Microbiol. 2012, 3, 1–5. [Google Scholar] [CrossRef] [PubMed]
- Tong, L. Viral proteases. Chem. Rev. 2002, 102, 4609–4626. [Google Scholar] [CrossRef] [PubMed]
- Waxman, L.; Darke, P.L. The herpesvirus proteases as targets for antiviral chemotherapy. Antivir. Chem. Chemother. 2000, 11, 1–22. [Google Scholar] [CrossRef] [PubMed]
- Arroyo, E.; Chinchilla, N.; Molinillo, J.M.; Macias, F.A.; Astola, A.; Ortiz, M.; Valdivia, M.M. Aneugenic effects of benzoxazinones in cultured human cells. Mutat. Res. 2010, 695, 81–86. [Google Scholar] [CrossRef] [PubMed]
- Hedstrom, L.; Moorman, A.R.; Dobbs, J.; Abeles, R.H. Suicide inactivation of chymotrypsin by benzoxazinones. Biochemistry 1984, 23, 1753–1759. [Google Scholar] [CrossRef] [PubMed]
- Jarvest, R.L.; Parratt, M.J.; Debouck, C.M.; Gorniak, J.G.; Jennings, L.J.; Serafinowska, H.T.; Strickler, J.E. Inhibition of hsv-1 protease by benzoxazinones. Bioorg. Med. Chem. Lett. 1996, 6, 2463–2466. [Google Scholar] [CrossRef]
- Hemmateenejad, B.; Elyasi, M. A segmented principal component analysis-regression approach to quantitative structure-activity relationship modeling. Anal. Chim. Acta 2009, 646, 30–38. [Google Scholar] [CrossRef] [PubMed]
- Sant’Anna, C.M.R. Molecular modeling methods in the study and design of bioactive compound: An introduction. Rev. Virtual Quim. 2009, 1, 49–57. [Google Scholar] [CrossRef]
- Zuegg, J.; Cooper, M.A. Drug-likeness and increased hydrophobicity of commercially available compound libraries for drug screening. Curr. Top. Med. Chem. 2012, 12, 1500–1513. [Google Scholar] [CrossRef] [PubMed]
- Bissantz, C.; Kuhn, B.; Stahl, M. A medicinal chemist's guide to molecular interactions. J. Med. Chem. 2010, 53, 5061–5084. [Google Scholar] [CrossRef] [PubMed]
- Lipinski, C.A.; Lombardo, F.; Dominy, B.W.; Feeney, P.J. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev. 1997, 23, 3–25. [Google Scholar] [CrossRef]
- Halgren, T.A. Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94. J. Comput. Chem. 1996, 17, 490–519. [Google Scholar] [CrossRef]
- Consortium, T.U. Reorganizing the protein space at the universal protein resource (uniprot). Nucleic Acids Res. 2012, 40, D71–D75. [Google Scholar] [CrossRef] [PubMed]
- Altschul, S.F.; Gish, W.; Miller, W.; Myers, E.W.; Lipman, D.J. Basic local alignment search tool. J. Mol. Biol. 1990, 215, 403–410. [Google Scholar] [CrossRef]
- Peitsch, M.C. Protein modeling by e-mail. Nat. Biotech. 1995, 13, 658–660. [Google Scholar] [CrossRef]
- Notredame, C.; Higgins, D.G.; Heringa, J. T-coffee: A novel method for fast and accurate multiple sequence alignment. J. Mol. Biol. 2000, 302, 205–217. [Google Scholar] [CrossRef] [PubMed]
- Eswar, N.; John, B.; Mirkovic, N.; Fiser, A.; Ilyin, V.A.; Pieper, U.; Stuart, A.C.; Marti-Renom, M.A.; Madhusudhan, M.S.; Yerkovich, B.; et al. Tools for comparative protein structure modeling and analysis. Nucleic Acids Res. 2003, 31, 3375–3380. [Google Scholar] [CrossRef] [PubMed]
- Marti-Renom, M.A.; Stuart, A.C.; Fiser, A.; Sanchez, R.; Melo, F.; Sali, A. Comparative protein structure modeling of genes and genomes. Annu. Rev. Biophys. Biomol. Struct. 2000, 29, 291–325. [Google Scholar] [CrossRef] [PubMed]
- Marti-Renom, M.A.; Yerkovich, B.; Sali, A. Comparative Protein Structure Prediction. Curr. Protoc. Protein Sci. 2:2.9; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2001. [Google Scholar]
- Melo, F.; Sanchez, R.; Sali, A. Statistical potentials for fold assessment. Protein Sci. 2002, 11, 430–448. [Google Scholar] [CrossRef] [PubMed]
- Pieper, U.; Eswar, N.; Braberg, H.; Madhusudhan, M.S.; Davis, F.P.; Stuart, A.C.; Mirkovic, N.; Rossi, A.; Marti-Renom, M.A.; Fiser, A.; et al. Modbase, a database of annotated comparative protein structure models, and associated resources. Nucleic Acids Res. 2004, 32, 217–222. [Google Scholar] [CrossRef] [PubMed]
- Sali, A.; Blundell, T.L. Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 1993, 234, 779–815. [Google Scholar] [CrossRef] [PubMed]
- Sali, A.; Overington, J.P. Derivation of rules for comparative protein modeling from a database of protein structure alignments. Protein Sci. 1994, 3, 1582–1596. [Google Scholar] [CrossRef] [PubMed]
- Doreleijers, J.F.; Raves, M.L.; Rullmann, T.; Kaptein, R. Completeness of noes in protein structure: A statistical analysis of NMR. J. Biomol. NMR 1999, 14, 123–132. [Google Scholar] [CrossRef] [PubMed]
- Doreleijers, J.F.; Rullmann, J.A.; Kaptein, R. Quality assessment of nmr structures: A statistical survey. J. Mol. Biol. 1998, 281, 149–164. [Google Scholar] [CrossRef] [PubMed]
- Laskowski, R.A.; MacArthur, M.W.; Moss, D.S.; Thornton, J.M. Procheck: A program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 1993, 26, 283–291. [Google Scholar] [CrossRef]
- Laskowski, R.A.; Rullmannn, J.A.; MacArthur, M.W.; Kaptein, R.; Thornton, J.M. Aqua and procheck-nmr: Programs for checking the quality of protein structures solved by NMR. J. Biomol. NMR 1996, 8, 477–486. [Google Scholar] [CrossRef] [PubMed]
- Bowie, J.U.; Luthy, R.; Eisenberg, D. A method to identify protein sequences that fold into a known three-dimensional structure. Science 1991, 253, 164–170. [Google Scholar] [CrossRef] [PubMed]
- Luthy, R.; Bowie, J.U.; Eisenberg, D. Assessment of protein models with three-dimensional profiles. Nature 1992, 356, 83–85. [Google Scholar] [CrossRef] [PubMed]
- Morris, G.M.; Huey, R.; Lindstrom, W.; Sanner, M.F.; Belew, R.K.; Goodsell, D.S.; Olson, A.J. Autodock4 and autodocktools4: Automated docking with selective receptor flexibility. J. Comput. Chem. 2009, 30, 2785–2791. [Google Scholar] [CrossRef] [PubMed]
- Jones, G.; Willett, P.; Glen, R.C. Molecular recognition of receptor-sites using a genetic algorithm with a description of desolvation. J. Mol. Biol. 1995, 245, 43–53. [Google Scholar] [CrossRef]
- Jones, G.; Willett, P.; Glen, R.C.; Leach, A.R.; Taylor, R. Development and validation of a genetic algorithm for flexible docking. J. Mol. Biol. 1997, 267, 727–748. [Google Scholar] [CrossRef] [PubMed]
- Nissink, J.W.M.; Murray, C.; Hartshorn, M.; Verdonk, M.L.; Cole, J.C.; Taylor, R. A new test set for validating predictions of protein-ligand interaction. Proteins Struct. Funct. Genet. 2002, 49, 457–471. [Google Scholar] [CrossRef] [PubMed]
- Verdonk, M.L.; Cole, J.C.; Hartshorn, M.J.; Murray, C.W.; Taylor, R.D. Improved protein-ligand docking using gold. Proteins Struct. Funct. Genet. 2003, 52, 609–623. [Google Scholar] [CrossRef] [PubMed]
- Organic Chemistry Portal. Available online: http://www.organic-chemistry.org/prog/peo (accessed on 20 May 2014).
- Samples Availability: Not available.
© 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).