INTRODUCTION:

Minichromosome maintenance proteins (MCM2-7) belong to a family of evolutionarily highly sustained protein. These six proteins form the important part of the DNA replication initiation system^1,2. The MCM proteins contribute to restricting the initiation of DNA replication to once per cycle^3,4. A heteromeric complex is formed by binding of these six proteins with each other⁵. One study showed that in addition to the role in DNA replication, the MCM proteins are also crucial for transcription activation⁶.

In humans, the MCM gene encodes the protein called as DNA replication factor MCM5. This protein is structurally homologous to CDC46 protein from Saccharomyces cerevisiae⁷. Many studies have demonstrated that MCM proteins may be better indicators of a wide variety of proliferative or cancer cells in malignant tissues^8-14. In one study it was seen that Aplidin displayed anti-multiple myeloma activity in a mouse xenograft model, which may be due to restraining of a collection of proliferative/ antiapoptic genes (e.g., MCM2, MCM4, MCM5) and up regulation of several possible regulators of apoptosis¹⁵.

Another study demonstrated that Prohibitin, a tumour suppression protein, interacts with MCM proteins and can act as an effective inhibitor of DNA replication¹⁶.

The most common problem in biology is the functional characterization of a protein sequence. There are some proteins which are too big for NMR analysis and also their structure cannot be predicted by X-ray diffraction. This work is normally made easy by precise three dimensional (3-D) structure of the studied protein. When experimental techniques fail then protein modeling is the only way to extract the structural information. Homology modeling estimates the 3-D structure of a given protein sequence (target) based principally on its alignment to one or more proteins of known structures (templates). The estimation process consists of fold assignments, model building and model evaluation^17,18. One study suggested that the combination of structure-based docking and advanced protein structure modeling should be an important approach to the large-scale drug screening and discovery studies¹⁹. The homology modeling has been widely used to predict the protein structure^20,21. In this study, we designed the structure of MCM5 protein, by using homology modeling. Finally the docking of the ligands was done to predict the binding orientation of small drug molecules with their protein target (MCM5) in order to predict the affinity and activity of the small molecules in inhibiting MCM5 so that the MCM heteromeric complex is blocked, which in turn will lead to inhibition of DNA replication.

Materials And Methods:

The hardware used for calculating molecular modeling includes a personal computer with Intel (R) Core (TM) i3 CPU processor, Windows 7 Home Premium 32-bit operating system having RAM of 2.00 GB.

Sequence Alignment

Fast Alignment (FASTA)

The FASTA format is a text based format for representing either nucleotide sequences or peptide sequences, in which nucleotides or amino acids are represented using single letter codes²². The FASTA sequence of MCM5 was acquired from the website of National Centre for Biotechnology Information²³.

Basic Local Alignment Search Tool (BLAST)

The BLAST is an algorithm for comparing primary biological sequence information, such as the amino acid sequence of different proteins or the nucleotides of DNA sequences²⁴. Using the FASTA sequence, the standard protein BLAST was performed on the NCBI. The protein data bank proteins data base was chosen and the BLAST-P was performed.

Three Dimensional Position-Specific Scoring Matrix (3D-PSSM)

The 3D-PSSM is a fast web based method for protein fold recognition using 1D and 3D sequence profiles coupled with secondary structure and salvation potential information²⁵. The FASTA sequence was submitted to 3D-PSSM for fold recognition.

Protein Homology/Analogy Recognition Engine (Phyre)

Phyre is a web based service for protein structure prediction. Phyre is among the most popular methods for protein structure prediction²⁶. The FASTA sequence was submitted to Phyre for amino acid sequence prediction²⁷.

Templates Preparation

The data obtained from combined 3D-PSSM and Phyre was subjected to RCSB protein data bank²⁸. The templates were selected on the basis of their resolution (Å) and R-value. All the above templates were submitted by X-ray crystallography method in PDB²⁹.

Molecular Modeling

Homology modeling of MCM5 was done by using EasyModeller³⁰. Easy Modeller is a graphical user interface to Modeller program. It is a standalone tool in windows platform with Modeller and Python preinstalled³¹. The Swiss-Pdb viewer was installed from the respective site which is an application that provides a user friendly interface allowing analyzing several proteins at the same time³².

Structure Prediction

The six templates were submitted to the EasyModeller and were aligned. The Discreet Optimized Protein Energy (DOPE) score is a statistical tool to assess homology models in protein structure prediction. The model with the minimum score can be chosen as the best possible structure.

Validation of Predicted Model

The validation of all the five models was performed by using Procheck in SAVS, i.e., Structural Analysis and Verification Server³³. The Ramachandran plot validated the result. The residues in the most favoured region are at maximum and those in the generously allowed and disallowed regions are at minimum.

Loop Modeling

The loop region in the given protein very often contribute to binding sites and determine the functional specificity of a given protein framework. The co-ordinate file in PDB format was submitted for loop optimization to ModLoop, i.e., Modeling of Loops in Protein Structures. ModLoop is a web server for automated modeling of loops in protein structures. It has been developed by Andras Fiser^34,35. The resulting co-ordinate file was sent back by e-mail. This structure was validated by using SAVS. The process of loop modeling and subsequent validation was continued till an optimized structured model of protein was obtained.

Ligand Generation

The literature survey revealed the inhibitory action of Aplidin against MCM protein¹⁵. The Aplidin (DB04977) was used to search for the similar drugs in the Drug Bank and ZINC databases^36,37.

Molecular Docking

The molecular docking is an important tool in structural molecular biology and computer-assisted drug designing³⁸. The docking of these drugs was performed by using free online server DockThor Portal. The implemented DockThor program is a flexible-ligand and rigid-receptor grid based method that employs a multiple solution genetic algorithm and the MMFF94S molecular force field scoring function³⁹. Both the macromolecule and ligands were prepared for docking with the help of PyMol and chemBio3D computer software respectively^40,41. The molecular docking of these drugs was done against MCM5 protein model using online docking server DockThor. The default settings of grid dimensions were employed. The default value of spatial discretization of the energy grid (0.25Å) was used and the grid points were fixed at 704969.

Grid Centre: Grid Dimensions (±ΔX; ±ΔY; ±ΔZ):

Xc = 16 X = 11

Yc = 6 Y = 11

Zc = 20 Z = 11

The best compound was selected on the basis of the total energy (intermolecular ligand-receptor + intramolecular ligand energies) or Interaction energy (only intermolecular ligand-receptor energy) and the root mean square deviation.

Results And Discussion

Template Generation

FASTA sequence of MCM5 was retrieved from the website of NCBI. The NCBI reference sequence is CAG30403.1. The BLAST was performed on the NCBI and 18 hits were recorded (Figure 1). The FASTA sequence was subjected to 3D-PSSM and Phyre for prediction of protein structure. The results obtained were combined and ranked in the descending order of % ID. The subjection of this data to RCSB protein data bank led to the selection of six templates (PDB ID: 3F8T, 2ENX, 1VMH, 1OFH, 2C90, 3K1J) with their resolution < or = 3.00 Å and the R- value is < or = 0.5 (Table 1).

Table 1. Generation of templates using 3D-PSSM, Phyre and RCSB protein data bank.

S. No.	Name	ID %	Resolu-tion	R-Value	PDBID
1	3F9VA	37	4.35	0.415	3F9V
2	4FDGB	37	4.10	0.332	4FDG
3	3ZEND	30	7.5	-	3ZEN
4	3F8TA	29	1.9	0.214	3F8T
5	2ENXA	29	2.8	0.196	2ENX
6	1VMHA	29	1.31	0.159	1VMH
7	1OFHA	28	2.5	0.224	1OFH
8	c2c90A	28	2.2	0.209	2C9O
9	C3K1jA	27	2.0	0.206	3K1J
10	2X47A	26	1.7	0.167	2X47
11	3PVCA	26	2.31	0.173	3PVC

Homology Modeling

The five models were generated with the help of EasyModeller and their DOPE score was obtained (Table 2). The model number 1 scored the minimum and was selected.

Validation:

The models were further validated by Procheck in SAVS. The Ramachandran plot validated and supported the earlier decision of selecting the model number 1, as the sum of residues in most favoured region (71.8%) and residues in additional allowed regions (21.7%) comes out to be highest, and the sum of residues in generously allowed region (4.1%) and residues in disallowed regions (2.3%) comes out to be lowest among all the models (Table 2).

Loop modeling

The PDB file format of model number 1 was submitted for loop optimization to ModLoop and the structure was validated by using SAVS. The model was validated as it had maximum percentage of residues in most favoured region (93.7%) and no residue in generously allowed as well as disallowed regions (Figure 2).

The model of MCM5 protein (Figure 3) was successfully submitted to Protein model data base bearing the PMDB ID: PM0079649⁴².

Figure 1. Distribution of 18 BLAST hits on the query sequence (query Id: Gi|47678565|emb|CAG30403.1) in pdb protein database.

Table 2. DOPE score and Ramachandran plot statistics of the five possible models of AMF.

S. No.	Query File Name	Molpdf	DOPE score	GA341 score	Residues in most favoured region (%)	Residues in additional allowed region (%)	Residues in generously allowed regions (%)	Residues in disallowed regions (%)
1	B99990001.pdb	38392.33203	-47071.94922	0.01582	71.8	21.7	4.1	2.3
2	B99990002.pdb	39515.19141	-44707.50781	0.01502	71.2	18.7	5.4	4.7
3	B99990003.pdb	39084.29297	-46161.54688	0.00532	71.7	20.5	5.2	2.6
4	B99990004.pdb	38443.04297	-43528.59375	0.00862	73.4	19.4	4.3	2.9
5	B99990005.pdb	39131.90625	-44203.32813	0.00788	71.8	20.8	4.4	2.9

Table 3. Docking result of ligands against MCM5 as target.

S. No.	Accession No.	Total Energy (T.E) (Kcal/ mol)	Interaction Energy (I.E) (Kcal/mol)	Root Mean Square Deviation (RMSD) (Å)	S. No.	Accession No.	Total Energy (T.E) (Kcal/mol)	Interaction Energy (I.E) (Kcal/ mol)	Root Mean Square Deviation (RMSD) (Å)
1	ZINC15861168	12.855	-0.001	0.000	36	ZINC35340021	41.846	-0.001	0.000
2	ZINC15861121	17.733	-0.001	0.000	37	ZINC37538182	42.792	-0.001	0.000
3	ZINC15861137	18.545	-0.002	0.000	38	ZINC35340007	44.128	-0.002	0.000
4	ZINC15861138	21.200	-0.001	0.000	39	ZINC35340016	44.179	-0.002	0.000
5	ZINC35339905	21.562	-0.001	0.000	40	ZINC35339971	53.633	-0.001	0.000
6	ZINC15861184	21.66	-0.001	0.000	41	DB04191	71.356	-0.002	0.000
7	ZINC15861122	22.966	-0.001	0.000	42	ZINC79221126	75.183	-0.001	0.000
8	ZINC35339900	23.208	-0.001	0.000	43	ZINC79221118	75.386	-0.000	0.000
9	ZINC35339889	23.392	-0.001	0.000	44	ZINC79221195	76.731	-0.002	0.000
10	ZINC35339896	23.933	-0.001	0.000	45	ZINC79221186	76.940	-0.001	0.000
11	ZINC15861185	25.716	-0.001	0.000	46	ZINC67910488	77.062	-0.001	0.000
12	ZINC04665026	25.725	-0.001	0.000	47	ZINC79216746	77.231	-0.001	0.000
13	ZINC35339965	25.828	-0.001	0.000	48	ZINC79216743	77.639	-0.002	0.000
14	ZINC43766843	27.165	-0.001	0.000	49	ZINC79221183	78.273	-0.001	0.000
15	ZINC35339927	28.201	-0.001	0.000	50	ZINC79221177	78.602	-0.001	0.000
16	ZINC35339886	28.282	-0.002	0.000	51	ZINC67910490	82.136	-0.002	0.000
17	ZINC35339892	28.464	-0.002	0.000	52	ZINC67910485	82.671	-0.001	0.000
18	ZINC35339923	29.917	-0.001	0.000	53	ZINC67910538	86.164	-0.001	0.000
19	ZINC35339834	32.181	-0.001	0.000	54	ZINC67910540	86.849	-0.001	0.000
20	ZINC35339829	34.474	-0.001	0.000	55	ZINC67910544	88.506	-0.002	0.000
21	ZINC35339878	34.731	-0.001	0.000	56	ZINC67911060	89.913	-0.003	0.000
22	ZINC35339883	35.397	-0.001	0.000	57	DB05426	90.162	-0.001	0.000
23	ZINC35340003	35.616	-0.001	0.000	58	ZINC67911062	90.189	-0.003	0.000
24	ZINC67297448	35.936	-0.001	0.000	59	ZINC67910545	91.973	-0.001	0.000
25	ZINC35339969	37.072	-0.001	0.000	60	ZINC67911064	93.237	-0.002	0.000
26	DB07219	37.121	-0.001	0.000	61	ZINC67911067	94.252	-0.002	0.000
27	ZINC37538176	37.414	-0.001	0.000	62	DB08889	126.009	-0.000	0.000
28	ZINC35339909	37.814	-0.001	0.000	63	DB03393	131.549	-0.003	0.000
29	ZINC35339999	38.138	-0.001	0.000	64	DB04977	134.034	-0.002	0.000
30	ZINC35339912	38.277	-0.001	0.000	65	DB06663	138.647	-0.001	0.000
31	ZINC67297434	38.509	-0.002	0.000	66	DB08890	150.493	-0.004	0.000
32	ZINC37538178	38.994	-0.001	0.000	67	DB01723	157.154	-0.168	0.000
33	ZINC35339967	39.674	-0.001	0.000	68	DB05128	159.637	-0.003	0.000
34	ZINC37538180	40.857	-0.001	0.000	69	DB01369	187.235	-0.002	0.000
35	ZINC01380951	41.81	-0.001	0.000

Figure 3. Optimized model of MCM 5 protein.

Conclusion:

The model of MCM5 protein was created by using the concepts of homology and loop modeling. The model was validated by the Ramachandran plot. Various ligands were identified using Drug bank and ZINC database. The molecular docking done against MCM5, of these ligands using online docking server Dockthor identified compound ZINC15861168 (acetonyl-cycloheptyl-cyclohexyl-BLAHtrione) with least total energy of 12.855 Kcal/mol. The study suggested that the above said compound bears the minimum binding energy with MCM5 protein and thus has the maximum probability to bind with it. Thus, the usage of this compound can lead to inhibition of the protein MCM5 which will act as a barricade in the formation of heteromeric complex of MCM2-7 proteins thereby assisting in the treatment of multiple myeloma and other cancer cells.

References:

1 Dudderidge TJ et al. Diagnosis of prostate cancer by detection of minichromosome maintenance 5 protein in urine sediments. British Journal of Cancer. 103 (5); 2010: 701-707.

2 Synder M, Huang XY and Zhang JJ. The minichromosome maintenance proteins 2-7 (MCM2-7) are necessary for RNA polymerase II (Pol II)-mediated transcription. Journal of Biological Chemistry. 284 (20); 2009: 13466-13472.

3 Gauthier L et al. The role of the carboxyterminal domain of RNA polymerase II in regulating origins of DNA replication in Saccharomyces cerevisiae. Genetics. 162 (3); 2002: 1117-1129.

4 Hill AL, Phelan SA and Loeken MR. Reduced expression of pax-3 is associated with over expression of cdc46 in the mouse embryo. Development Genes and Evolution. 208 (3); 1998: 128-134.

5 Miyake S and Yamashita S. Identification of sna41 gene, which is the suppressor of nda4 mutation and is involved in DNA replication in Schizosaccharomyces pombe. Genes to Cells: devoted to cellular and molecular mechanisms. 3 (3); 1998: 157-166.

6 Synder M, He W and Zhang JJ. The DNA replication factor MCM5 is essential for Stat1-mediated transcriptional activation. Proceedings of the National Academy of Sciences of the United States of America. 102 (41); 2005: 14539-14544.

7 Wikipedia. MCM5. Wikipedia [cited 2014 Nov 8]. Available from: http://www.en.wikipedia.org/wiki/MCM5

8 Stoeber K et al. Diagnosis of genito-urinary tract cancer by detection of Minichromosome Maintenance 5 protein in urine sediments. Journal of the National Cancer Institute. 94 (14); 2002: 1071-1079.

9 Yu SY et al. Increased expression of MCM5 is significantly associated with aggressive progression and poor prognosis of oral squamous cell carcinoma. Journal of Oral Pathology and Medicine. 43 (5); 2014: 344-349.

10 Liu H et al. Expression of minichromosome maintenance 5 protein in proliferative and malignant skin diseases. International Journal of Dermatology. 46 (11); 2007: 1171-1176.

11 Going JJ et al. Aberrant expression of minichromosome maintenance proteins 2 and 5, and Ki-67 in dysplastic squamous oesophageal epithelium and Barrett’s mucosa. Gut. 50 (3); 2002: 373-377.

12 Giaginis C et al. MCM-2 and MCM-5 expression in gastric adenocarcinoma: clinical significance and comparison with Ki-67 proliferative marker. Digestive Diseases and Sciences. 56 (3); 2011: 777-785.

13 Saydam O et al. Comparative protein profiling reveals minichromosome maintenance (MCM) proteins as novel potential tumor markers for meningiomas. Journal of Proteome Research. 9 (1); 2010: 485-494.

14 Ayaru L et al. Diagnosis of pancreaticobiliary malignancy by detection of minichromosome maintenance protein 5 in bile aspirates. British Journal of Cancer. 98 (9); 2008: 1548-1554.

15 Mitsiades CS et al. Aplidin, a marine organism-derived compound with potent antimyeloma activity in vitro and in vivo. Cancer Research. 68 (13); 2008: 5216-5225.

16 Rizwani W, Alexandrow M and Chellappan S. Prohibitin physically interacts with MCM proteins and inhibits mammalian DNA replication. Cell Cycle. 8 (10); 2009: 1621-1629.

17 Krieger E, Nabuurs SB and Vriend G. Homology Modeling. In Structural Bioinformatics, Edited by Bourne PE and Weissig H. John Wiley & Sons, Inc, Hoboken, NJ, USA. 2005; Vol 44: pp. 507-521.

18 Webb B and Sali A. Comparative protein structure modeling using MODELLER. Current Protocols in Bioinformatics. 47; 2014: 5.6.1-5.6.32.

19 Du H et al. Protein structure prediction provides comparable performance to crystallographic structures in docking-based virtual screening. Methods. 71 (1); 2015: 77-84.

20 Sivakamavalli J et al. Homology modelling, molecular dynamics, and docking studies of pattern-recognition transmembrane protein-lipopolysacccharide and β-1,3 glucan-binding protein from Fenneropenaeus indicus. Journal of Biomolecular structure and Dynamics. 20; 2014: 1-12.

21 Khan N et al. Insilico approach to predict the structure of amyloid precursor protein responsible for Alzheimer disease. Journal of Advanced Bioinformatics Applications and Research. 3 (2); 2012: 333-338.

22 Wikipedia. FASTA. Wikipedia [cited 2014 May 06]. Available from: http://www.en.wikipedia.org/wiki/FASTA

23 National Centre for Biotechnology Information. MCM5 homo sapiens [FASTA]. National Centre for Biotechnology Information [cited 2014 May 06]. Available from: http://www.ncbi.nlm.nih.gov/protein/47678565?report=fasta

24 Wikipedia. BLAST. Wikipedia [cited 2014 May 06]. Available from: http://www.en.wikipedia.org/wiki/BLAST

25 Kelley L. 3D-PSSM (Version 2.6.0). 3D-PSSM protein fold recognition (threading) server [cited 2014 May 06]. Available from: http://www.sbg.bio.ic.ac.uk/~3dpssm/index2.html

26 Wikipedia. Phyre/Phyre2. Wikipedia [cited 2014 May 06]. Available from: http://www.en.wikipedia.org/wiki/Phyre_/_Phyre2

27 Kelley L et al. Phyre² (Version 2). Protein homology analogy recognition server [cited 2014 May 06]. Available from: http://www.sbg.bio.ic.ac.uk/phyre2

28 Berman HM et al. The Protein data bank. Nucleic Acids Research. 28 (1); 2000: 235-242.

29 Research Collaboratory for Structural Bioinformatics-Protein Data Bank. RCSB Protein Data Bank- RCSB PDB (An information portal to biological macromolecular structures) [cited 2014 May 08]. Available from: http://www.rcsb.org

30 Bhusan KK. EasyModeller (Version 2.0) {Computer software} [cited 2014 May 08]. Available from: http://www.sites.google.com/site/bioinformatikz

31 Bhusan KK, Aparoy P and Reddanna P. EasyModeller: A graphical interface to MODELLER. BMC Research Notes. 3; 2010: 226.

32 Guex N et al. DeepView-Swiss-PdbViewer (Version 4.1) {Computer software} [cited 2014 May 06]. Available from: http://www.Spdbv.vital-it.ch/

33 NIH MBI Laboratory for Structural Genomics and Proteomics. Structural Analysis and Verification Server (Version 3). UCLA MBI-SAVES web server [cited 2014 May 10]. Available from: http://www.nihserver.mbi.ucla.edu/SAVES/

34 Fiser A and Sali A. ModLoop: automated modeling of loops in protein structures. Bioinformatics. 19 (18); 2003: 2500-2501.

35 Fiser A. ModLoop (Version r204). ModLoop-ModBase web server [cited 2014 May 11]. Available from: https://modbase.compbio.ucsf.edu/modloop/

36 Shoichet Laboratory. ZINC {Data set} [cited 2014 May 23]. Available from: http://www.zinc.docking.org

37 The Metabolomics Innovation Centre. Drug Bank {Data set} [cited 2014 May 21]. Available from: http://www.drugbank.ca

38 Morris GM and Lim-Wilby M. Molecular docking. Methods in Molecular Biology. 443; 2008: 365-382.

39 GMMSB/LNCC. The DockThor Portal (Version 1.0) {Online docking server} [cited 2014 November]. Available from: http://www.dockthor.Incc.br

40 PerkinElmer. ChemBio3D (Ultra 13.0) {Computer software} [cited 2014 November 12]. Available from: http://www.cambridgesoft.com/Ensemble_for_chemistry/chemBio3D/

41 Schrödinger. PyMol (Version 1.7.2) {Computer software} [cited 2014 November 12]. Available from: http://www.pymol.org

42 CASPUR & University of Rome “La Sapienza. (2014). Protein Model Data Base {Data Base} [cited 2014 August 28]. Available from: http://bioinformatics.cineca.it/PMDB/

Received on 20.02.2015 Accepted on 10.03.2015

Asian J. Pharm. Tech. 2015; Vol. 5: Issue 1, Pg 17-22

DOI: 10.5958/2231-5713.2015.00004.5