1. INTRODUCTION:

The interdisciplinary field of research in system biology aims to study the interaction involved in different toxicological event in the human body¹. The combination of experimental as well as biological knowledge can be a key regulatory method for modelling a gene. The greatest goal of the study in gene modelling is to help evolution of new variety of treatment and drugs opposed to a disease. In the current field of genomic research one of the peak challenges is the genomic variation, which is due to the extensive number of genetic variation in the genome of humans². The gene modelling provides a new protein which can be experimentally tested or serve as a guideline for physical experiments in future.

In each single individual SNPs called single nucleotide polymorphisms constitute the most plentiful genetic variations through the genome of human ranging from between three and four million^3,4. These SNPs are generally neutral, but some among them contribute to the predisposition of diseases by the modification of protein function or as the genetic marker in respect to find nearby mutation causing disease through the association of genetic studies and family related studies. Scientist also believe that variation in SNPs also influence better response to some of the drugs ^5(p1),6. The encoded amino acids can be changed by SNPs called as nonsynonymous single nucleotide polymorphisms (nsSNPs). This type of SNPs forms about half the changes in genome related human diseases, can also cause neutral or deleterious effects in resulting protein structure and function⁷. Thus single nucleotide polymorphism modelling is a key or main factor to remodel the genes involved in various toxicities or diseases so that there will be a better binding affinity with the ligands bound to it^8–10.

The drug induced gastric reaction is the major causative for gastro-duodenal injury in both normal animals as well as human¹¹. This effect can alter the gene performing normal function and induce the mechanism of some toxicity in the gastric system. Gastrotoxicity is the major effect affected primarily due to some toxic alterations in genome or if any exposure happens¹². The altered gene due to this toxicity can be determined by the gene modelling process. The involved genes are identified and its altered different forms are recognized and modeled to get a new gene that can bind to the suitable ligands. Here in this research the genes names ABCB1 and ALDH1A1 has been taken whose different altered forms are identified and SNP model has been done in order to check its binding affinity with the ligands of Madhucalongifolia^13,14. The gene ABCB1 is an multidrug resistance gene and alteration in its form can cause adverse gastric toxicities, ABCB1 or ATB Binding Cassette Subfamily B Member 1 is a gene coding the protein, the diseases associated with its alteration include gastric toxicities, inflammatory Bowel Disease 13 and Colchicine Resistance and the gene ALDH1A1 is Aldehyde Dehydrogenase 1 Family Member A1 an major gene involved in detoxification of active cyclophosphamide metabolism. The major diseases associated with ALDH1A1 include Erythroplakia and Lung Adenoma¹⁵. These two genes are docked with the ligands or bioactive compounds found in bark, stem, fruits and leaves of Madhuca logifolia plant^12,16. The leaf of the plant is used in gastropathy, verminosis, dipsia, bronchitis and hemorrhoids and along with leaves the bark is also used for diabetes mellitis, chronic bronchitis and much more¹⁷. Those bioactive compounds from the different parts of the plant showed a highly efficient binding affinity with the genes involved in gastrotoxicity¹⁸. Here computer visualization technique is mainly involved for the modelling purpose by taking the 3D structure of the proteins and by determining the SNP database we changes the mutated nucleotide to another and checks its affinity. The new protein sequence whose nucleotide is altered is submitted to I-TASSER which brings out an new model of receptor. Here patchdock helps to create a linkage between the receptor and ligand involved and how much affinity they create can be checked by their bonding strength^19,20.

2. MATERIALS AND METHOD:

2.1. Identification of genes involve in gastrotoxicity:

The genes responsible for gastrotoxicity were identified through literature review. It is seen that a lot of genes responsible lack a well modelled tertiary structure. The genes which cause gastrotoxicity are PGDH2, PGD2, PGE2, PGF2α, VGFR, TXB2. Among the genes there are genes which undergo single nucleotide polymorphism and cause gastrotoxicity. The identified SNP genes are ALDHA1 and ABCB1 genes.

2.2. SNP Gene modeling:

ALDH1A1 and ABCB1 genes were selected. The missense mutations were selected and the sequence which was retrieved from Uniprot, were subjected to single nucleotide polymorphism. The FASTA sequences were submitted to ITASSER and the gene models were obtained.

2.3. Quality check of modelled protein:

The genes which have been submitted to ITASSER for modelling is now subjected to quality check to ensure that only the gene model which has high percentage of residues in the Ramachandran plot is selected and subjected to docking with the ligands of Madhuca longifolia. The secondary chain structure of the modelled protein is also analysed.

2.4. Ligands:

The ligands are the active molecules which are present in the seed, fruit, bark and nutshell which is obtained from literature review. The canonical smile of each compound where obtained through PubChem and subjected to Corina molecule to obtain PDB structure. Around 29 ligands were selected for docking with the modelled genes. The ligands are Ethylcinnamate, Sesquiterine alcohol, α-terpeneol, α- and β- amyrin acetates, n-Hexacosanol, Quercetin, Dihydroquercetin, β-sitosterol, Quercitin- 3-glucoside, Icosonoic acid, Octadeca 9,12 dienoic acid, Octadec-9-enoic acid, Tetradecanoic acid, Hexadecanoic acid, Octadecanoic acid, 2 amino propanoic acid, 4 methyl 2(phenyl methoxy carbonyl amine) pentanoic acid, 2-6 diaminohexanoic acid, 2 amino 4 methyl sulfanylbutanoic acid, Pyrrolidine-2 carboxylic acid, 2 amino 3 hydroxypropanoic acid, 2 amino-3 hydroxybutanolic acid, 3,5,7 trihydroxy -2-(3,4,5 trihydroxyphenyl) chromen-4-one, 2(3-4 dihydroxyphenyl)-5,7 dihydroxy-3, Mi-saponin A, Mi-saponin B, 2 amino butanedoicacid, 2 amino-3-[2amino-2carboxy, 2 amino acetic acid and 2 amino 3 methyl pentanoic acid.

2.5. In silico Docking:

In silico docking was performed using patch dock server. In order to perform the docking process, the ligands and receptors were submitted to the online server in PDB format. The results consist of Score, ACE and area.

2.6. Analysis of docked structure:

In order to find the effective interaction between the ligands and receptors the docked structures are visualized through Pymol software. The software helps us to find the the atoms, the residues and their bond length and hence the effective interaction is predicted.

3. RESULTS:

3.1. SNP gene analysis:

The various single nucleotide polymorphism present in the gene ALDH1A1 and ABCB1. The SNP taken into consideration are ASN121SER, PHE466TYR, ALA446VAL and ILE177PHE from ALDH1A1 gene. The SNP taken into consideration from gene ABCB1 includes GLY185VAL and SER893ALA.

3.2. Protein model analysis through Ramachandran plot

The various gene models obtained through ITASSER now is subjected to quality check to ensure the best model is taken into consideration and is used for docking with the ligands of Madhuca longifolia. These receptors are used for further studies. The below table provides the C score TM score and RMSD score and Percentage (Table 1). Fig 1 represent the best PDB structure of modelled protein.

Table 1: Quality Check of the Modelled Protein

Gene	Models	c-score	TM score	RMSD score	%
ALDH1A1 (ASN121SER)	MODEL1	1.53	0.93± 0.06	4.1 ±2.8Å	84.3%
ALDH1A1 (PHE466TYR)	MODEL2	1.46	0.92 ±0.06	4.2 ±2.8 Å	84.3%
ALDH1A1 (ALA446VAL)	MODEL1	0.96	0.84 ±0.08	5.3 ±3.4 Å	82.7%
	MODEL2	0.04	0.84± 0.08	5.3± 3.4 Å	83.1%
	MODEL3	1.32	0.84± 0.08	5.3 ±3.4 Å	81.5%
ALDH1A1 (ILE177PHE)	MODEL1	1.40	0.91± 0.06	4.4 ±2.9 Å	83.4%
BCB1 (GLY185VAL)	MODEL1	0.46	0.77 ±0.10	8.4 ±4.5 Å	79.5%
	MODEL2	0.97	0.77 ±0.10	8.4 ±4.5 Å	75.7%
	MODEL3	-0.07	0.77 ±0.10	8.4 ±4.5 Å	77.9%
	MODEL4	0.09	0.77 ±0.10	8.4 ±4.5 Å	78.7%
	MODEL5	-2.31	0.77 ±0.10	8.4 ±4.5 Å	78.7%
ABCB1 (SER893ALA)	MODEL1	1.09	0.86± 0.07	7.0± 4.1 Å	79.5%
	MODEL2	0.16	0.86± 0.07	7.0± 4.1 Å	83.7%
	MODEL3	-2.00	0.86± 0.07	7.0± 4.1 Å	77.7%
	MODEL4	-1.82	0.86± 0.07	7.0± 4.1 Å	78.6%
	MODEL5	-2.19	0.86± 0.07	7.0± 4.1 Å	82.7%

Fig 1 Best PDB structure of the modelled protein

Fig 2: Ligand binding site

3.4. Docking analysis:

3.4.1. Docked complex with ALDH1A1 (ASN121SER):

The docking analysis is done with ligands and Madhuca longifolia and the six SNP gene models. The modelled protein, ALDH1A1 (ASN121SER) is docked with the 29 ligands. Among the docked complex the best interaction were found in MiSaponin A, MiSaponin B, Sesquiterene alcohol and Quercetin-3-glucoside (Fig 3). The bond with residue, atom and length of MiSsaponin A are HIS293:H:1.4, ARG326:O59:2.8, GLU289:H:1.2 and ARG326:O59:1.9. The bond with residue, atom and length of MiSaponin B are ASN334:O89:1.8, THR14:O30:1.1, GLY376:H:1.5, VAL11:H:1.9, SER4:O12:1.7, ASN377:O74:1.9 and ASN377:H:1.1. The bond with residue, atom and length of Sesquiterene alcohol are GLY226:H71:1.6, SER247:O19:2.6, SER247:O13:1.6, GLU249:H77:1.7 and THR245:H73:2.7. The bond with residue, atom and length of Quercetin-3-glucoside ILE167:O33:3.2, GLU196:H49:2.8, TRP169:O20:1.8, GLN350:O10:3.0 and GLN350:H40:1.8.

Fig 3: Docked complex of ALDH1A1 (ASN121SER)

3.4.2. Docked complex with ALDH1A1 (PHE466TYR):

The modelled protein, ALDH1A1 (PHE466TYR) is docked with the 29 ligands. The major ligands which showed high hydrogen bond with the modelled protein are MiSaponin A and MiSaponin B (Fig 4). The bond with residue, atom and length of MiSaponin A are LEU16:H:2.3, THR14:H:1.7, MET1:H:1.8, SER4:O22:0.8, SER2:O19:2.8, SER2:H:2.1 and MET1:O32:1.4. The bond with residue, atom and length of MiSaponin B are LYS398:O78:2.0, GLU400:O89:2.2, SER247:O33:2.6, LYS353:O20:2.5, ASP347:O93:0.7, ASP347:O93:2.3, GLU349:H:2.1 and SER247:O28:2.3.

Fig 4 Docked complex of ALDH1A1 (PHE466TYR) gene

3.4.3. Docked complex with ALDH1A1 (ALA446VAL):

The modelled protein, ALDH1A1(ALA446VAL) is docked with the 29 ligands. The major ligands which showed high hydrogen bond with the modelled protein are MiSaponin A, MiSaponin B and Quercetin-3-glucoside (Fig 5). The bond with residue, atom and length of MiSaponin A are ASP353:H:0.9, GLU406:O15:2.4, GLU406:O7:2.4, GLU255:H:2.2, THR254:O83:2.5, TYR432:O13:1.0, GLU406:H:1.5, TYR432:O7:2.8 and TYR231:O71:2.3. ASN48. The bond with residue, atom and length of MiSaponin B are ASN48:H:2.4, ASN340:O53:2.6, LEU6:H:0.6, LEU6:O43:2.3, ASN383:H:1.0, ALA1:H:1.6, ALA1:O72:2.0, ALA1:O82:1.5, SER8:O38:1.4 and SER8:O88:2.8. The bond with residue, atom and length of Quercetin-3-glucoside are GLY131:O33:3.4, VAL466:O30:2.6, ASN127:H46:2.5, ASN127:O26:2.6 and ASN127:O29:2.8.

Fig 5 Docked complex of ALDH1A1 (ALA446VAL) gene

3.4.4. Docked complex with ALDH1A1 (ILE177PHE):

The modelled proteinALDH1A1(ILE177PHE) is docked with the 29 ligands. The major ligands which showed more interaction with the modelled protein are MiSaponin B and Quercetin-3-glucoside (Fig 6). The bond with residue, atom and length of MiSaponin B are GLU48:H:0.5, LYS17:O91:2.5, ASP15:H:2.1 and SER3:H:2.2. The bond with residue, atom and length of Quercetin-3-glucoside are GLN350:H40:2.3, GLU400:H39:2.5, SER247:O27:1.8, ILE167:O33:3.3, TRP169:O20:2.1 and GLN197:O30:2.4.

Fig. 6: Docked complex of ALDH1A1 (ILE177PHE) gene

3.4.5. Docked complex with ABCB1 (GLY185VAL):

The modelled protein ABCB1(GLY185VAL) is docked with the 29 ligands. The major ligand which showed more interaction with the modelled protein is Sesquiterene (Fig 7). The bond with residue, atom and length of Sesquiterene are ARG-832:2.3:O15, ASP-6:2.6:H72, ARG-832:1.3:O33 and ARG-832:2.4:O33.

Fig. 7: Docked complex of ABCB1 (GLY185VAL) gene

3.4.6. Docked complex with ABCB1 (SER893ALA):

The modelled protein ABCB1 (SER893ALA) is docked with the 29 ligands. The major ligands which showed more interaction with the modelled protein are MiSaponin A, MiSaponin B and Quercetin (Fig 8). The bond with residue, atom and length of MiSaponin A are LYS12:H:2.6, GLN824:O59:1.8, THR816:O23:2.8, SER646:O19:2.7, GLU643:H:1.9, LYS645:H:2.2, SER644:O85:0.5 and SER644:O85:2.0. The bond with residue, atom and length of MiSaponin B are ASN183:O31:2.4, TRP136:O31:2.7, GLU184:H:1.4, LYS887:O93:1.7 and ARG148:O94:2.3. The bond with residue, atom and length of Quercetin are TYR247:H30:1.9, TYR247:H30:1.9, TYR247:H30:1.9, TYR247:H30:1.9, TYR247:H30:1.9 and TYR247:H30:1.9.

Fig. 8: Docked complex of ABCB1 gene

DISCUSSION:

The drugs that are administered for breast cancer are said to cause gastrotoxicity. This is caused by the mutation of a gene. The main genes that are involved are ABCB1 and ALDH1A1. The researchers have identified many SNPs in these three genes that cause gastrotoxicity¹⁹.So a minute change in the sequence on the specific sites can be done with the help of the gene sequence obtained from Uniprot. With the help of ITASSER the homology structure of the receptor can be obtained.

Madhuca longifolia is a plant that is commonly available in many parts of India. The active components from bark, fruit and nutshell are said to be very effective against hepatotoxicity¹⁸. So those structures of ligands are obtained from pdb and checked their inhibing activity on the receptors obtained from I-TASSER. The structure of the ligand and the receptor is docked to check the interactions between them. The one having the more number of bonds will be the best or the most effective ligand against⁵. Two SNPs were there for gene ABCB1. From our studies the ligand that shown maximum interaction for the receptor ABCB1 (SER893ALA) was MI Saponin A, MI saponin B and Quercitin and model2 (GLY185VAL) was sequesterine alcohol. And for the other receptor ALDH1A1, there were two SNPs. From results the model 1 (ILE177PHE) had shown more interaction with MiSaponin B and Quercitin 3Glucoside. And the second SNP (ALA446VAL) have more interaction with MiSaponin A, MiSaponin B and Quercitin 3 Glucoside. From this experiment we are focusing on the future perspectives.

Further in future, in-vitro studies can be held. The formulation of the specific ligand into a drug can be administered to an animal model and check their activity. Here the gastrotoxicity that is caused in prior can be blocked by formulating the drugs with the specific ligand of Curvularialunata⁵. After the experimental designing of the drugs in vivio studies can also held that test the effect of drug on the organism. So any mutation in a particular gene can be overcome by the specific ligands that are having high affinity to the receptor while docking^21–23. So the ligand of Madhuca longifolia can serve as a better medicine for those who are suffering from gastrotoxicity. It opens a wide range scope in pharmacology and drug development.

4. CONCLUSION:

Breast cancer is one of the most common type of cancer. There are so many therapeutic ways to treat this. The drug administered for this cancer is said to cause mutation in some of the genes and can cause gastrotoxicity. In our study, we just focused on the SNPs of two genes. They are ABCB1 and ALDH1A1. The structures on the SNPs are obtained from I-TASSER. So through this experiment we are proving that the ligands of Curvularia lunata from its bark, nutshell and fruit. The structure of the ligands are obtained from PDB and then the models of SNPs and the ligands were docked in patch dock. The interaction between them have proven that, this can be formulated into a drug in future that can prevent gastrotoxicity. Among the 29 ligands MI Saponina, MI Saponin B and quercitin 3 glucoside have shown their maximum interaction in most of the receptors. In future, in-vitro and in-vivo studies can be held to make an effective medicine for gastrotoxicity

5. ACKNOWLEDGMENT:

We would like to thank the Vellore Institute of Technology for providing us this great opportunity to perform and carry out all the necessary work.

6. CONFLICTS OF INTEREST:

Authors declare that they do not have any conflicts of interests.

7. REFERENCES:

1. Karlebach G, Shamir R. Modelling and analysis of gene regulatory networks. Nature Reviews Molecular Cell Biology. 2008; 9(10): 770-780.

2. Ivanov IV, Qian X, Pal R. Emerging Research in the Analysis and Modeling of Gene Regulatory Networks. IGI Global; 2001. https:// www.igi-global.com/book/emerging-research-analysis-modeling-gene/145652. Accessed October 21, 2019.

3. Wu B, Song H-P, Zhou X, et al. Screening of minor bioactive compounds from herbal medicines by in silico docking and the trace peak exposure methods. Journal of Chromatography. A. 2016;1436: 91-99.

4. Anbarasu K, Mahjabeen A, Mahendran R. Insights on Impact of Missense SNPs of Human USH2A associated with Usher Syndrome II. Research Journal of Pharmacy and Technology. 2017;10(10):3365.

5. Yao S, Sucheston LE, Zhao H, et al. Germline genetic variants in ABCB1, ABCC1 and ALDH1A1, and risk of hematological and gastrointestinal toxicities in a SWOG Phase III trial S0221 for breast cancer. The Pharmacogenomics Journal. 2014;14(3):241-247.

6. Ranjan S, Sharma PK. Association of Brain-Derived Neurotrophic factor (BDNF) gene SNP G196A with Type 2 Diabetes and Obesity: A Meta-Analysis. Research Journal of Pharmacy and Technology. 2017;10(12):4297.

7. Mani A, Ravi L, Krishnan K. Antibacterial and antifungal potential of marine Streptomyces sp. VITAK1 derived novel compound Pyrrolidinyl-Hexadeca-Heptaenone by in Silico docking analysis. Research Journal of Pharmacy and Technology. 2018;11(5):1901.

8. Grupe A, Germer S, Usuka J, et al. In silico mapping of complex disease-related traits in mice. Science (New York, N.Y.). 2001;292(5523):1915-1918.

9. Kumar RS, Kaavya G. Binding Efficiency of Molecules from Medicinal Plants with Fidgetin Like Protein 2-A Novel Target for Diabetic Foot Ulcer. Research Journal of Pharmacy and Technology. 2017;10(11):3757.

10. Sheng T, Lu Y, Yang K, et al. Association Between Single Nucleotide Polymorphisms (SNPs) in the Promoter of Adiponectin Gene, Hypoadiponectinemia, and Diabetes. Research Journal of Science and Technology. 2016;8(1):34.

11. Rainsford KD, Whitehouse MW. Paracetamol [acetaminophen]-induced gastrotoxicity: revealed by induced hyperacidity in combination with acute or chronic inflammation. Inflammopharmacology. 2006;14(3-4):150-154.

12. Rainsford KD. Gastrointestinal and other side-effects from the use of aspirin and related drugs; biochemical studies on the mechanisms of gastrotoxicity. Agents and Actions. Supplements. 1977;(1):59-70.

13. Patil MP, Singh RD, Koli PB, et al. Antibacterial potential of silver nanoparticles synthesized using Madhucalongifolia flower extract as a green resource. Microbial Pathogenesis. 2018; 121:184-189.

14. Singh A, Singh S, Anbarasu A. In silico Evaluation of Non-Synonymous SNPs in IRS-1 gene associated with type II diabetes mellitus. Research Journal of Pharmacy and Technology. 2018;11(5):1957.

15. Annalakshmi R, Mahalakshmi S, Charles A, Sahayam CS. GC–MS and HPTLC analysis of leaf extract of Madhuca longifolia (Koenig) Linn. Drug Invention Today. 2013;5(2):76-80.

16. Govindammal M, Prasath M, Sathya B, Selvapandiyan M. Investigation on the Binding Interaction Between Clomipramine and Doxepin with LeuT by Molecular Docking Analysis. Asian Journal of Research in Chemistry. 2017;10(4):486-490.

17. Sikarwar RLS. Mahua (Madhuca longifolia (Koen.) Macbride) - A paradise tree for the tribals of Madhya Pradesh. In: 2002.

18. Simon JP, Parthasarathy M, Nithyanandham S, Katturaja R, Namachivayam A, Prince SE. Protective effect of the ethanolic and methanolic leaf extracts of Madhuca longifolia against diclofenac-induced toxicity in female Wistar albino rats. Pharmacological Reports. May 2019.

19. Chen Y-C. Beware of docking! Trends in Pharmacological Sciences. 2015;36(2):78-95.

20. Al-Koofee DAF, Mobarak SMH. Primer1: A Network Service for Tetra-Arms PCR Primer Design Based on Well-Known dbSNP. Research Journal of Pharmacy and Technology. 2018;11(8):3633.

21. Mahendran R, Jeyabasker S, Francis A, Manoharan S. Homology Modeling and in silico docking analysis of BDNF in the treatment of Alzheimer’s disease. Research Journal of Pharmacy and Technology. 2017;10(9):2899.

22. Anbarasu K, Jagadeeswari P, Mahendran R. In-silico Screening of Deleterious NF1 SNPs Associated with Neurofibromatosis Type I. Research Journal of Pharmacy and Technology. 2017;10(9):3247.

23. Sudha R, Devi PB, Nithya G. Docking Analysis of various synthesised Benzilic acid derivatives towards Sirutin protein and toxicity studies. Research Journal of Pharmacy and Technology. 2019;12(7):3277.

Received on 20.12.2019 Modified on 17.02.2020

Research J. Pharm. and Tech. 2020; 13(12):5799-5805.

DOI: 10.5958/0974-360X.2020.01011.2

Gene	C Score	PDB hit	Cluster size	Ligand name	Ligand binding site residues
ALDH1A1 (ASN121SER)	0.74	4PZ2A	260	NAD	166,167,168,169,170,175,193,195,196,226,230,231,244,245,246,247,250,253,254,269,270,271,303,400,402,428,466
ALDH1A1 (PHE466TYR)	0.72	4NI4A	245	NAD	166,167,168,169,170,175,193,195,196,226,230,231,244,245,246,247,250,253,254,269,270,271,303,400,402,428,466
ALDH1A1 (ALA446VAL)	0.72	4PZ2A	245	NAD	172,173,174,175,176,181,199,201,202,232,236,237,250,251,252,253,256,259,260,275,276,277,309,406,408,434,472
ALDH1A1 (ILE177PHE)	0.72	2O2QA	239	NAP	166,167,168,169,170,175,193,195,196,225,226,230,231,244,245,246,247,250,253,254,269,270,271,303,400,402,428,466
ABCB1 (GLY185VAL)	0.27	3G60B	10	0JZ	69,72,336,340,343,725,728,732,953,975,978,979,982,985
ABCB1 (SER893ALA)	0.25	3G60B	10	0JZ	69,72,336,340,343,725,728,732,953,975,978,979,982,985