Research articles

By Mr. S M Sabbir Alam , Mr. Md Shariful Islam
Corresponding Author Mr. S M Sabbir Alam
Department of Microbiology, University of Dhaka, - Bangladesh 1000
Submitting Author Mr. S M Sabbir Alam
Other Authors Mr. Md Shariful Islam
Department of Microbiology, University of Dhaka, - Bangladesh


Trimethoprim, Dihydrofolate Reductase, Protein Modeling, Enzyme Inhibition Assay, Automated Docking.

Alam S, Islam M. Homology Modeling and Docking Studies Showed that Dihydrofolate Reductase from Pseudomonas Putida is a Possible Choice for Diagnosis of Serum Trimethoprim by Enzyme Inhibiton Assay. WebmedCentral BIOINFORMATICS 2011;2(12):WMC002825
doi: 10.9754/journal.wmc.2011.002825

This is an open-access article distributed under the terms of the Creative Commons Attribution License(CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Click here
Submitted on: 30 Dec 2011 02:23:41 PM GMT
Published on: 30 Dec 2011 04:23:25 PM GMT


Trimethoprim is a chemotherapeutic drug mainly used in prophylaxis and treatment of bacterial infections. It belongs to dihydrofolate reducase inhibitors and has bacteriostatic properties. It may cause serious side effects if it is overdosed or used for long time. It may cause renal clearance mechanism impairment, thrombocytopenia, allergic reactions and a number of other clinical complications. An enzyme inhibition assay can be used to determine serum trimethoprim, which may provide advantage in terms of time and cost. This involves inhibition of dihydrofolate reductase by trimethoprim in serum. Dihydrofolate reductase (DHFR) is an enzyme that reduces dihydrofolic acid to tetrahydrofolic acid. In this study DHFR from lactobacillus casei (PDB id: 4DFR:A),  Bacillus anthracis (PDB id:3JW3_A) and Moritella profunda (PDB id: 3IA4_A) are used as templates for building 3D models of DHFR from some other species. Programs used here are MODELLER, SWISS 3D MODEL and GENO3D. Based on overall stereochemical quality (PROCHECK, VARIFY3D, ANOLEA, PROSA) best models were selected, refined and characterized for binding site by CASTp program along with Catalytic Site Atlas (CSA) database. Best models were studied further for structure function relationship with ligand (trimethoprim) and its analogue (dihydrofolate reductase) by using docking approach (AutoDock and AutoDock VINA). The interaction energy between the trimethoprim and modeled enzyme indicated that homology models for DHFR of Pseudomonas putida can account for better regionspecificity of this enzyme towards trimethoprim. Findings from the current study could be utilized to de novo enzyme selection for diagnosis of serum trimethoprim.


Dihydrofolate reductase is the target enzyme for a group of antifolate drugs like methotrexate and trimethoprim [1]. It inactivates dihydrofolate reductase, which functions in conversion of dihydrofolate to tetrahydrofolate. And folate is an essential factor for DNA synthesis and cell division. Due to inhibition of DNA synthesis and replication bacterial growth stops and thus trimethoprim can act as bacteriostatic agent [1,2]. But also trimethoprim may show some side effects. It may cause renal clearance mechanism impairment, thrombocytopenia, allergic reactions and a number of other clinical complications. It may also characterized by nausea, vomiting, swollen face, epigastric pain, headache, & weakness [4]. Diagnosis of serum trimethoprim can aid to determine the efficacy of drug and its effect in system in terms of dose and time. An enzyme inhibition assay can be used to determine serum substrate by testing inhibition of enzyme in serum [5, 6]. For understanding and analyzing protein function it is necessary to understand its 3D structure. Protein 3D structure can be determined by experimental methods such as X-ray crystallography or NMR analysis. It can also predict by computational analysis. By homology modeling a reliable model of protein can be found [7-9]. These models have also been proved as useful for drug design projects and allowed to take actions in compound optimization and chemical adjustment [10]. By docking study the interaction of secondary structure elements in proteins may be demonstrated. It is considered to use matching two separate molecules. It used to show correlations between experimental binding affinities and its mathematical score for various protein-ligand complexes [11].
In present study, by using different programs like MODELLER, GENO3D, and SWISS 3D MODEL was used to generate 3D model of dihydrofolate reductase from different organisms. Dihydrofolate reductase from lactobacillus casei (PDB id: 4DFR:A), Bacillus anthracis (PDB id:3JW3_A) and Moritella profunda (PDB id: 3IA4_A) is used as template for model built up. Validation of these models was done by programs like PROCHECK, VARIFY3D, ANOLEA, PROSA etc. Active site prediction and docking study were performed using CASTp program, Catalytic Site Atlas (CSA) database, AutoDock and AutoDock Vina to analyze functional association of dihydrofolate reductase with trimethoprim.

Materials and methods

2.1 Protein sequence retrieval and 3D modeling
Protein sequence was retrieved from NCBI protein sequence database (accession no: ABZ01067.1, ZP_06637221.1, NP_752010.1, YP_001454852.1, YP_002152055.1, ZP_06192186.1, YP_001439347.1, YP_003537753.1, ZP_06124802.1 and YP_003363702.1). Best template was selected by using NCBI protein blast by using hits against Brookhaven Protein Data Bank (PDB) database [16] to find nearest crystal structure. Dihydrofolate reductase structure from lactobacillus casei (PDB id: 4DFR:A), Bacillus anthracis (PDB id: 3JW3_A) and Moritella profunda (PDB id: 3IA4_A) was selected as template for their maximum sequence identity and E value. ClustalW was used for building pairwise sequence alignment. For 3D modeling MODELLER [14], SWISS 3D MODEL [12, 13] and GENO3D [15] were used.
2.2 Validation of 3D models
By using different software programs (MODELLER, SWISS 3D MODEL, PROCHECK [19], VERIFY3D [18], and PROSA [21]) the validation of structure models were obtained. RamchandranPlot obtained from PROCHECK was used to check stereochemical property. Model constructed from SWISS-3D MODEL and MODELLER was finally chosen for subsequent analysis as they possessed good geometry and energy profile. PROSA was used for final model to check energy criteria and Verify-3D was used to check compatibility of 3D models with its sequences.
2.3 Active site characterization
By aligning with known template with known active site we determined the active site of model structures. Here Catalytic Site Atlas (CSA) database [28-30] and CASTp program [22] was used with combination of PyMOL [3, 24] for visualization and analysis of protein molecular structures. In CSA database catalytic residues and enzyme active sites in 3D structure are documented. In CSA database it consists of two type’s annotated site: original annotated set comprising information directly extracted from primary literature and annotations deduced by PSI-BLAST and sequence alignment with original set [28-30]. After determining catalytic site residues we aligned model sequences with template sequences to find conserved residues and dissimilar catalytic residues. These data was used to set grid parameter for docking approach.
2.4 Retrival of ligand structure
Structure of trimethoprim was obtained from NCBI PubChem [23]. OpenBabelGUI and AutoDock tools were used to convert this chemical format to a suitable format for docking approach. PubChem is a database for small molecules and their biological properties. It provides opportunity of rapid data retrieval, structure selectivity analysis, target selectivity examination etc [23].
2.5 Docking ligand into enzyme 3D model
AutoDock tools and AutoDock Vina [34] was used for docking ligand into enzyme active sites. Previous file formats were reformatted and refined prior to docking approach, utilizing AutoDock tools. AutoDock Vina was used for docking of ligand (trimethoprim) into enzyme active site. AutoDock Vina is a program that facilitates molecular docking and virtual screening approach. It is an automated docking tool which offers greater speed and improved accuracy for binding mode predictions with automated estimation of grid maps and clusters [34].

Results and discussion

3.1 Homology modeling
Homology modeling estimates the 3D structure of a target protein sequence by using its alignment to one or more protein template of known structure [25]. For structure based protein molecule design and function investigation homology modeling is most suitable method [26]. The modeling process involves of target-template selection and alignment, model building and model evaluation. [25] As the number of known protein structures are increasing and protein model software’s are improving, the accuracy of the models are increasing [25]. As dihydrofolate reductase from Lactobacillus casei was previously used in enzyme inhibition assay for methotrexate we analyzed it in terms of trimethoprim[27]. DHFR from some related organisms therefore modeled. Organisms eg. Paenibacillus polymyxa (YP_003870895.1), Cronobacter sakazakii (YP_001439347.1), Erwinia amylovora (YP_003537753.1) Providencia rettgeri (ZP_06124802.1) Citrobacter rodentium (YP_003363702.1). For these organisms it was found that DHFR from L. casei (PDB id: 3DFR:A) is a suitable template. And for DHFR from some common microorganisms like Pseudomonas putida (ABZ01067.1), Serratia odorifera (ZP_06637221.1), Escherichia coli CFT073 (NP_752010.1), Citrobacter koseri (YP_001454852.1) and Proteus mirabilis (YP_002152055.1) it was found that DHFR from Bacillus Anthracis (3JW3_A) and Moritella
Profunda (3IA4_A) are suitable templates. Five models for each sequences was constructed using MODELLER, SWISS 3D MODEL, and GENO3D. Using RamchandranPlot from ProSA, Phi and Psi torsion angles were checked. For each sequence best model was selected for subsequent analysis. These models were further refined for docking purpose using AutoDock tools. Polar hydrogen was added to each structure.
3.2 Model evaluation
The quality of protein model verifies the informatics can be mined from it. So, evaluation of the accuracy of protein modes is essential for their interpretation [25]. For this purpose different programs were used e.g. Swiss 3D model, PROCHECK, VARIFY3D and PROSA. Stereochemical properties of the models were evaluated by ProCheck. A Ramchandran plot was found for every model (Table 01). This plot shows the quality of each model. For each sequence best model then selected. The Ramachandran plot showed that model found from MODELLER (ABZ01067.1, YP_003870895.1, YP_001439347.1 and YP_003537753.1) and from SWISS-3D MODEL (ZP_06124802.1, YP_003363702.1, ZP_06637221.1, NP_752010.1, YP_001454852.1, YP_002152055.1) have most residues in most favorable region and have overall good quality. From Ramchandran plot it was found that for model 6 (ABZ01067.1) 96.20% residues in most favorable region, 2.90% in allowed region, 0.60% in additional and 0.30% are in disallowed region (figure 5) as compared to template 3(3IA4_A) 97%, 2.70%, 0.3 % and 0.0%, respectively. It ensures that most residues are in consistent phi-psi distribution and are reliable for further analysis. Prosa energy plot showed that for each selected model the interaction energy for each residue with rest of the protein in negative and Verify-3D graph showed that for each selected model 3D-1D score is above zero (Table 2), thus side chain environments are acceptable.
3.3. Active site prediction and docking study
To analyze substrate binding and specificity docking study for all homology models was performed. By docking study interactions of substrate into active site can be visualized as protein substrate complex. Active site pockets of templates 4DFR:A, 3JW3_A and 3IA4_A were analyzed. All models ware aligned in order to find the corresponding regions of all structures. By sequence alignment and selecting matched point active site conservation analysis was performed.
It was found that ILE 5 (for template 3 ILE 6), MET 20 (MET 21), ASP27 (GLU 28), LEU28 (LEU 29), PHE31 (PHE 32), LEU54 (LEU55), ILE 94(ILE 96) was highly conserved among all template and models. In homology model 1, 6 and 7 active site residues MET21, GLU28 and LEU29 was different. Changes in conserved residues may change conformational change and binding pattern with substrate. Finally docking study for the protein 3D models was performed to find its relation in terms of ligand binding. Trimethoprim was successive docked onto active site of enzyme models. Table 3 shows output of docking experiments in terms of affinity (kcal/mol). Different model shows significant difference in dock scores. Among them model 5 showed highest dock scores -14.9 illustrated its tight binding with target.


Comparative structural modeling and docking simulations showed significant difference in affinity of dihydrofolate reductase towards trimethoprim. Various model evaluation methods indicated that modeled structures has considerably good geometry and acceptable profiles for all programs. DHFR from Pseudomonas putida showed significant dock scores than others. It suggests its possible application for analysis of serum trimethoprim by enzyme inhibition assay.


We are thankful to Md. Monwarul Islam (Department of Computer science and engineering, University of Dhaka) for his contribution in docking studies.


1. S. J. Benkovic, C. A. Fier1ke, A. M. Naylor. Insights into Enzyme Function from Studies on Mutants of Dihydrofolate Reductase. Science 4 March 1988: Vol. 239 no. 4844 pp. 1105-1110
2. H.Groenendal and F.H.J.Rampen, Methotrexate and trimethoprim-sulphamethoxazole - a potentially hazardous combination, Clinical and Experimental Dermatology 1990; 15: 358-360.
3. Seeliger D, de Groot BL. Ligand docking and binding site analysis with PyMOL and Autodock/Vina. J Comput Aided Mol Des. 2010 May; 24(5):417-22.
4. Ellenhorn, M.J., S. Schonwald, G. Ordog, J. Wasserberger. Ellenhorn's Medical Toxicology: Diagnosis and Treatment of Human Poisoning. 2nd ed. Baltimore, MD: Williams and Wilkins, 1997. p. 236
5. Prachya Kongtawelert and Peter Ghosh. An enzyme-linked immunosorbent-inhibition assay for quantitation of hyaluronan (hyaluronic acid) in biological fluids. Analytical Biochemistry Volume 178, Issue 2, P. 367-372
6. T. Porstmann and S. T. Kiessig. Enzyme immunoassay techniques an overview. Journal of Immunological Methods Volume 150, Issues 1-2, P. 5-21
7. J Mark, S. Johnson, Narayanaswamy Srinivasan, Ramanathan Sowdhamini and Tom L Blundell. Knowledge-Based Protein Modeling. Critical Reviews in Biochemistry and Molecular Biology 1994, Vol. 29, No. 1 Pages 1-68
8. S.K.Burley. An overview of structural genomics. Nature Structural Biology.7(Suppl.)(2000)932–938.)
9. R.G. Bodade, S.D. Beedkar, A.V. Manwar, C.N. Khobragade. Homology modeling and docking study of xanthine oxidase of Arthrobacter sp. XL26. Int. Journal of Biological Macromolecules 47 (2010) 298–303
10. M. C. Peitsch. ProMod and Swiss-Model: Internet-based tools for automated comparative protein modeling. Biochemical Society Transactions (1996) 24, (274–279)
11. Ausiello, G., Cesareni, G., and Helmer-Citterich, M. 1997. ESCHER: A new docking procedure applied to the reconstruction of protein tertiary structure. Proteins 28: 556–567.
12. Kopp J, Schwede T. The SWISS-MODEL Repository: new features and functionalities. Nucleic Acids Res. 2006 Jan 1;34(Database issue):D315-8
13. Arnold K, Bordoli L, Kopp J, Schwede T. The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics. 2006 Jan 15;22(2):195-201. Epub 2005 Nov 13.
14. Jamroz M, Kolinski A. Modeling of loops in proteins: a multi-method approach. BMC Struct Biol. 2010 Feb 11; 10:5.
15. Combet C, Jambon M, Deléage G, Geourjon C. Geno3D: automatic comparative molecular modelling of protein. Bioinformatics. 2002 Jan;18(1):213-4.
16. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE. The Protein Data Bank. Nucleic Acids Res. 2000 Jan 1;28(1):235-42.
17. Lund, O., Nielsen, M., Lundegaard, C., and Worning, P. (2002) CPHmodels 2.0: X3M a Computer Program to Extract 3D Models. In CASP5 Conferencea102.
18. Eisenberg D, Lüthy R, Bowie JU. VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol. 1997;277:396-404.
19. Laskowski R.A.,Macarthur M.W.,Moss D.S..Hornton J.M., Procheck: a program to check the stereochemical quality of protein structures journal of applied crystollography 1993; 26(-):283-291.
20. Melo, F. and Feytmans, E. (1997) "Novel knowledge-based mean force potential at atomic level". Journal of Molecular Biology 267, 207-222.
21. Wiederstein & Sippl (2007) ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Research 35, W407-W410
22. Joe Dundas, Zheng Ouyang, Jeffery Tseng, Andrew Binkowski, Yaron Turpaz, and Jie Liang. 2006. CASTp: computed atas of surface topography of proteins with structural and topographical mapping of functionally annotated residues. Nucl. Acids Res., 34:W116-W118
23. Yanli Wang, Jewen Xiao, Tugba O. Suzek, Jian Zhang, Jiyao Wang, and Stephen H. BryantPubChem: a public information system for analyzing bioactivities of small molecules. Nucleic Acids Res. 2009 July 1; 37(Web Server issue): W623–W633
24. Lill MA, Danielson ML. Computer-aided drug design platform using PyMOL. J Comput Aided Mol Des. 2011 Jan;25(1):13-9. Epub 2010 Oct 30.
25. Mart´-Renom, Ashley C. Stuart, Andr´as Fiser, Roberto S´anchez, Francisco Melo, and Andrej Sali Annu. Comparative protein structuremodeling of genes and genomes. Marc A. Rev. Biophys. Biomol. Struct. 2000. 29:291–325
26. F. Melo, E. Feytmans, Assessing protein structures with a non-local atomic interaction. Journal of Molecular Biology 277 (5) (1998) 1141–1152.
27. T. Atkinson, T. K. Sundaram and D. S. Secher. The Potential of Microbial Enzymes as Diagnostic Reagents. Phil. Trans. R. Soc. Lond. B. 1983 vol-300, P. 399-410
28. The Catalytic Site Atlas: a resource of catalytic sites and residues identified in enzymes using structural data. Craig T. Porter, Gail J. Bartlett, and Janet M. Thornton (2004) Nucl. Acids. Res. 32: D129-D133.

Source(s) of Funding


Competing Interests



This article has been downloaded from WebmedCentral. With our unique author driven post publication peer review, contents posted on this web portal do not undergo any prepublication peer or editorial review. It is completely the responsibility of the authors to ensure not only scientific and ethical standards of the manuscript but also its grammatical accuracy. Authors must ensure that they obtain all the necessary permissions before submitting any information that requires obtaining a consent or approval from a third party. Authors should also ensure not to submit any information which they do not have the copyright of or of which they have transferred the copyrights to a third party.
Contents on WebmedCentral are purely for biomedical researchers and scientists. They are not meant to cater to the needs of an individual patient. The web portal or any content(s) therein is neither designed to support, nor replace, the relationship that exists between a patient/site visitor and his/her physician. Your use of the WebmedCentral site and its contents is entirely at your own risk. We do not take any responsibility for any harm that you may suffer or inflict on a third person by following the contents of this website.

0 comments posted so far

Please use this functionality to flag objectionable, inappropriate, inaccurate, and offensive content to WebmedCentral Team and the authors.


Author Comments
0 comments posted so far


What is article Popularity?

Article popularity is calculated by considering the scores: age of the article
Popularity = (P - 1) / (T + 2)^1.5
P : points is the sum of individual scores, which includes article Views, Downloads, Reviews, Comments and their weightage

Scores   Weightage
Views Points X 1
Download Points X 2
Comment Points X 5
Review Points X 10
Points= sum(Views Points + Download Points + Comment Points + Review Points)
T : time since submission in hours.
P is subtracted by 1 to negate submitter's vote.
Age factor is (time since submission in hours plus two) to the power of 1.5.factor.

How Article Quality Works?

For each article Authors/Readers, Reviewers and WMC Editors can review/rate the articles. These ratings are used to determine Feedback Scores.

In most cases, article receive ratings in the range of 0 to 10. We calculate average of all the ratings and consider it as article quality.

Quality=Average(Authors/Readers Ratings + Reviewers Ratings + WMC Editor Ratings)