Computational prediction and SLN formulation of Narcissin for reverse transcriptase inhibition and controlled drug delivery applications

The aim of this paper was to construct a stable drug delivery mechanism of Narcissin, which is phytoconstituent of Aerva lanata that has Reverse Transcriptase potential of anti-breast cancer. The Rand Forest Classifier was the most successful machine learning algorithm with an accuracy of 86.43 and independent test set validation of 80.85. High binding affinity to Narcissin (13.3 kcal/mol) with five hydrogen bonds and positive hydrophobic interactions were observed in molecular docking. Simulations of Narcissin-Reverse Transcriptase complex using molecular dynamics revealed that it did not exhibit significant changes in RMSD, which meant that the complex was stable. MMGBSA analysis has displayed a good binding free energy of 51.12 kcal/mol with the van der Waals forces playing a major role (62.2 kcal/mol). The Narcissin loaded solid lipid nanoparticles (SLN) had the highest encapsulation efficiency (90.12%), mean particle size of 80 nm, and zeta potential of 20 mV. In vitro release experiments revealed a zero-order, diffusion-controlled release, which was controlled and the cumulative release at pH 7.4 was 93.24%. The MTT assay exhibited dose- and time-dependent cytotoxicity particularly at 100 µg/mL indicating that Narcissin has a potential to be used as a bioactive agent in the treatment of breast cancer in SLN-based formulations.