Berberine Derivatives as Multi‐Kinase Inhibitors of MAP Kinase and PI3K Pathways: Molecular Modelling, Molecular Dynamics Simulations, and in vitro Evaluation

Berberine is a plant‐based nitrogenous cyclic compound with a structure that is highly similar to that of intercalating agents. Berberine has been shown to inhibit PI3K and MAP kinase pathways in various cancer cell lines. However, the immediate targets of berberine are not well‐understood. As EGF and AKT are upstream of MAP kinase and PI3K pathways, the objective of this study is to perform in silico modelling followed by in vitro analyses to describe the anti‐EGFR and anti‐AKT properties of berberine. Other BBR derivatives are also modelled to predict better inhibitors than berberine based on in silico studies. Material and methods Using molecular docking, thermodynamics parameters such as the binding energy, inhibition constant (Ki), and pKi have been computed for 13 protein crystals and several berberine derivatives. To validate the results, molecular dynamics (MD) simulations were performed, and Radial Distribution Function (RDF) and other MD parameters were computed based on 10 nanosecond simulations. Target‐specific indices such as nBEI (pKi + Log [heavy atoms]) and NSEI (pKi/[N+O]) for each berberine derivative were calculated to compare the effectiveness of berberine as compared with other derivatives. The interactions have been modelled. In vitro studies were performed by using of MCF‐7 and MDA‐MB231 breast cancer cell line. Cytotoxicity assays was done, IC50 was calculated, and apoptosis assay was performed. To evaluate the effects of berberine on EGFR and AKT, ELISA and Western Blotting were used to detect both phosphorylated kinases in different concentrations. We also used lapatinib as a positive control for both in vitro and in silico studies. Results EGFR and AKT were targeted by berberine derivatives especially via LYS745 and LYS179, respectively. The most important interactions were H‐bonding and cation‐π interactions within which aromatic rings of berberine and positive charged amino group of lysine are involved. Average Ki calculated of berberine for EGFR and AKT were 6.3 and 2.5 μM, respectively. BBR‐9, a newly generated ligand, seems to be the more effective than original berberine with less Ki for both targets ( Table 1). According to ELISA quantitative experiments, the activity of four enzymes (EGFR, AKT, P38, and ERK1/2) were affected by berberine with the most effect on EGFR and AKT which the total concentration was highly affected as well as the phosphorylated enzymes. It has been observed a combinatorial treatment of both lapatinib and berberine strongly increased the level of apoptosis compared with lapatinib alone ( Figure 1). These results suggested berberine might be useful to overcome drug resistance which has been reported for inhibitors of MAP kinase pathways. Conclusion Berberine derivatives that can be docked to different targets were selected, and three novel berberine derivatives were generated against EGFR and AKT. It has been suggested that one new ligand may be the best anti‐EGFR and anti‐AKT for further synthesis and experimental evaluation. 1 The computed inhibition constants of selected modelled compounds. BBR‐9 is suggested as a replacement for berberine for further evaluations.Target Selected Compounds Iterations Min. Ki (nM) Max. Ki (nM) Mean Ki (nM) Std. ErrorAKT ATP 20 .00661 5.70000 .4551545 .28095174 BBR‐9 20 33.32 719.38 183.9635 41.09922 Inhibitor control 20 10.24 3800.00 375.6345 189.24955 Berberine 20 1530 7160 2526.50 367.516EGFR ATP 30 .15 53.56 14.8614 2.83217 BBR‐9 30 118.05 645.69 357.6123 31.96655 Lapatinib 30 13.74 14810.00 869.1090 486.85738 Berberine 30 2290 9420 6284.33 508.119