(-)-Kusunokinin as a Potential Aldose Reductase Inhibitor: Equivalency Observed via AKR1B1 Dynamics Simulation

Abstract

(-)-Kusunokinin performed its anticancer potency through CFS1R and AKT pathways. Its ambiguous binding target has, however, hindered the next development phase. Our study thus applied molecular docking and molecular dynamics simulation to predict the protein target from the pathways. Among various candidates, aldo-keto reductase family 1 member B1 (AKR1B1) was finally identified as a (-)-kusunokinin receptor. The predicted binding affinity of (-)-kusunokinin was better than the selected aldose reductase inhibitors (ARIs) and substrates. The compound also had no significant effect on AKR1B1 conformation. An intriguing AKR1B1 efficacy, with respect to the known inhibitors (epalrestat, zenarestat, and minalrestat) and substrates (UVI2008 and prostaglandin H₂), as well as a similar interactive insight of the enzyme pocket, pinpointed an ARI equivalence of (-)-kusunokinin. An aromatic ring and a γ-butyrolactone ring shared a role with structural counterparts in known inhibitors. The modeling explained that the aromatic constituent contributed to π-π attraction with Trp111. In addition, the γ-butyrolactone ring bound the catalytic His110 using hydrogen bonds, which could lead to enzymatic inhibition as a consequence of substrate competitiveness. Our computer-based findings suggested that the potential of (-)-kusunokinin could be furthered by in vitro and/or in vivo experiments to consolidate (-)-kusunokinin as a new AKR1B1 antagonist in the future.

Figure 1

Binding energy of (−)-kusunokinin with protein candidates in the CSF1R and AKR1B1 pathways. A number denotes the binding energy in kcal/mol.

Figure 2

(−)-Kusunokinin binding site on AKR1B1. The site is composed of anion binding pocket (orange), small hydrophobic pocket (green), and large hydrophobic pocket (yellow).

Figure 3

(−)-Kusunokinin and ARI interaction with AKR1B1. Interaction of ARIs include epalrestat (orange), zenarestat (blue), minalrestat (red), and (−)-kusunokinin (lime green). Tyr48 and His110 bound ligands using hydrogen bonds, while Trp20 and Trp111 were mainly responsible for π–π interactions. The π–π interactions are shown as black dashed lines, and the hydrogen bonds are shown as red dashed lines.

Figure 4

MSD and RMSF of MD trajectory from the AKR1B1 simulation. The interesting simulations were the AKR1B1 and AKR1B1 structures with six docked compounds. RMSD was plotted from 150 ns simulations (a). RMSF was computed from a 90 to 150 MD trajectory (b).

Figure 5

Distance pattern among ligand-free AKR1B1 and drug-AKR1B1 MD simulations. The distance pattern of all simulations was similar to the ligand-free AKR1B1, giving a hint that a bound ligand showed no significant effect on AKR1B1 conformation.

Figure 6

Binding interaction of (−)-kusunokinin and other inhibitors/substrates with AKR1B1. Six compounds were chosen for MD simulation: (−)-kusunokinin (a), zenarestat (b), minalrestat (c), epalrestat (d), UVI2008 (e), and prostaglandin H₂ (PGH₂) (f). The black, red, and green dotted lines represent hydrogen bonds, π–π interaction, and π–alkyl interaction, respectively.

Figure 7

His110 catalytic role in PGH₂ conversion. PGH₂ conversion to PGF_2α using proton transfer from His110 (a). Position alignment between (−)-kusunokinin (red) and dioxabicycloheptane ring of PGH₂ (blue) at the ARK1B1 catalytic site (His110) (b).

Figure 8

Proposed binding model of ARI in AKR1B1 anion binding pocket. (−)-Kusunokinin (a), carboxylate ARIs (b), hydantoin ARIs (c); red dashed lines: hydrogen bond; orange arrow: π–π interaction.

Figure 9

(−)-Kusunokinin, arctigenin, and arctiin structures.