In Silico Molecular Modeling and Docking Studies of Aquaporin-3 with Centella asiatica Active Compound

*Corresponding Author: Linda Yulianti, Departemen of Dermatovenereology Faculty of Medicine Tarumanagara University Jakarta Indonesia. E-mail: lindajuliantiwijayadi@gmail.com Citation: Linda Yulianti, et al (2017). In Silico Molecular Modeling and Docking Studies of Aquaporin-3 with Centella asiatica Active Compound.. Int J Pharm Sci & Scient Res. 3:6, 71-73 Copyright: © Linda Yulianti, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Introduction ISSN 2471-6782
Aquaporins (AQPs) are keyfactor in the mechanism of skin hidration as water channel. The most abundant AQP present in the skin and more specifically in plasma membrane of epidermal keratinocytes is Aquaporin 3 (AQP3). Centella asiatica was commonly use in wound healing treatment and in anti aging cosmetic, The biologically active ingredients are triterpenes namely asiatic acid, madecassic acid, asiaticoside, and madecassoside. In this study, we developed a computational model using Autodock4 program to predict possible binding sites and binding energies between Centella asiatica active compounds and AQP3. The protein structure of AQP3 were predicted using Modeller softaware. Ligands structure were downloaded from chEBI database and converted to PDB file using OpenBabel. Prediction of ligand binding sites were done using Q-Site Finder. Docking studies were done in Autodock4 software. Our result suggested that all ligands that we compared were attached on the same binding site at Alanin 234. However, Asiaticossite and Madecasosside had higher mean binding energies. These higher mean binding energies may be caused by some bonds created by glucose-chain of Asiaticoside and Madecasosside Aquaporin, AQP3, Centella asiatica, Triterpenes, Asiatic acid, Madecassic acid, Asiaticoside, Madecassoside, Q-Site Finder, OpenBabel, Autodock4.
Aged skin is manifested by reduced stratum corneum (SC) moisturization, and although transepidermal water loss (TEWL) is know to be normal or improved with age, the epidermal barrier repair capacity after removing the superficial layer of the barrier by tape striping is significantly impaired (1) . The aquaporins (family of protein) that form channel to facilitate the transport of water across membranes has become of significant interest for its role in epidermal water maintenance (2) Aquaporin-3 (AQP3) has been a particular focus because it is an aquaglycerospore, it can cotransport glycerol, an it's absence result in skin dryness, reduce 3C hydration and elasticity and delayed barrier recovery (2,3,4) . Centella asiatica (Apiceae) has been used as traditional herbal medicine in Malaysia and other part of Asia for hundreds of years. It is commonly known as pegaga in Malaysia, pennywort and gotu kola in America, and pegagan in Indonesia (5) . This tropical plant has been used for various medicinal purposes such as wound healing, treatment of asthma, ulcers, leprosy, lupus erythematosus, psoriasis, vein diseases, for memory improvement, as an antidepressant, antibacterial, antifungal, and anti-cancer agent (6,7) . Its also can be used as herbal anti aging cosmetic (6) . The major biologically active ingredients are triterpenes namely asiatic acid, madecassic acid, asiaticoside, and madecassoside (7,8) . Computational (in silico) methods have been developed and widely applied in biomolecular and biotechnological field for pharmacology hypothesis development and testing. (9,10) . Such methods have seen frequent use in the discovery of novel molecules with affinity to a target, absorption, distibution, metabolishm, excretion as for physicochemical characterization of a new drug. (10,11) . For example, a study of molecular docking of compound isolated from fruits of Helicteres isora for antidiabetic activity (10) . In this study, we developed a computational model using Autodock tools program to predict possible binding sites between Centella asiatica active compounds and AQP3. We performed in silico docking study. Molecular docking is a computational procedure that attempt to predict non-covalent binding of macromolecules or more frequently on macromolecule (receptor) and a small molecule (ligand) efficiency (9,11) .

Protein Modeling
First of all, Aqp3's protein sequence downloaded from Genebank database. The sequence aligned by Protein BLAST module in Sali Lab's Modeller Software to search the most possible template to make homology modeling. The output (1ldf.pdb) were used as template for homology modeling by the same software. From this process, ten possible protein models were created. We selected the best model by analyzed and compared DOPE assessment score from each model (12) .

Ligand Preparations
Asiatic Acid, Madecassic acid, Asiaticosside, and Madecassoside chemical structures downloaded from chEBI database at http:// www.ebi.ac.uk/chebi/. The downloaded Mol files converted to pdb format by openBabel software to make them readable by the docking program (Autodock4) (13) .

Molecular Docking
The best Aqp3 protein model were uploaded to Q-site finder site (http://www.modelling.leeds.ac.uk/ qsitefinder/) to find the possible ligant binding sites. The result became the area parameter for the later docking process (14) . We used Autodock Tool 1.5.6 to simplified the docking procedure by visualizing the standard Autodock4's text-rich process. First of all, we downloaded Autodock4, AutoGrid4, and mgltools_1.5.6 (Autodock Tool) and installed them to our processing computer. Then, we used Autodock Tool to visualize the docking process of each ligand (Asiatic Acid, Madecassic acid, Asiaticosside, and Madecassoside) in pdb format to the best Aqp3 protein model with Q-site finder's area parameter by Autodock4. From, these docking processes we obtained each ligand binding site and its binding energy (13) . The binding site of each ligand were visualized by PyMol software and the binding energy compared by table (13,15) .

Result And Discussion
First of all, the protein modeling method were used because the structure of AQP3 were not existed in PDB database. All PDB files we found on various sources on Internet were only predicted versions of AQP3, so, we made our own predicted protein using Modeller which might be the best method available to do homology modeling on transmembrane protein such as AQP3 (12,16) . The software choose 1ldf as the template of AQP3, maybe due 1ldf sequence 68% similarity with AQP3 sequence, which quite ideal for homology modeling (16) .  Madecasosside concentration in Centella asiatica were significantly higher than the concentration of Asiatic acid and Madecasic acid (7) . The result of this study can be used as information for future in vitro studies on using target cell line.

Conclusion
Asiatic acid, Madecassic acid, Asiaticosside, and Madecassoside are most likely have the same binding site at Alanine 234 of AQP3. Asiaticosside tent to have the highest binding energy and asiatic acid tent to be the lowest binding energy. Asiaticosside and Madecasosside maybe have higher binding energy because of their glucose chain can make some additional bonds to the protein. However, to ensure that binding energies were enough to made a stable complex with AQP3, further investigation required.