Apolipoprotein E4 (Apo E4) is the major genetic risk factor in the causation of Alzheimer’s disease (AD). In this study we utilize virtual screening of the world’s largest traditional Chinese medicine (TCM) database and investigate potential compounds for the inhibition of ApoE4. We present the top three TCM candidates: Solapalmitine, Isodesacetyluvaricin, and Budmunchiamine L5 for further investigation. Dynamics analysis and molecular dynamics (MD) simulation were used to simulate protein-ligand complexes for observing the interactions and protein variations. Budmunchiamine L5 did not have the highest score from virtual screening; however, the dynamics pose is similar to the initial docking pose after MD simulation. Trajectory analysis reveals that Budmunchiamine L5 was stable over all simulation times. The migration distance of Budmunchiamine L5 illustrates that docked ligands are not variable from the initial docked site. Interestingly, Arg158 was observed to form H-bonds with Budmunchiamine L5 in the docking pose and MD snapshot, which indicates that the TCM compounds could stably bind to ApoE4. Our results show that Budmunchiamine L5 has good absorption, blood brain barrier (BBB) penetration, and less toxicity according to absorption, distribution, metabolism, excretion, and toxicity (ADMET) prediction and could, therefore, be safely used for developing novel ApoE4 inhibitors.
Alzheimer’s disease (AD) is the most common harmful neurological disorder affecting patients over the age of 65 [
ApoE, which exists in three different isoforms in the brain and periphery, ApoE2 (epsilon 2), ApoE3 (epsilon 3), and ApoE4 (epsilon 4), have an essential role in the regulation of cholesterol metabolism [
We utilized computer-aided drug design (CADD) in this research in order to design potential lead drugs for AD therapy. CADD is an efficient approach for the rapid identification of potential lead compounds in target therapy [
We used 61,000 TCM compounds for database screening, which were obtained from TCM Database@Taiwan [
The LibDock program [
The molecular dynamic simulation was performed with GROMACS 4.5.5 package [
All bonds were constrained with the linear constraint solver (LINCS) algorithm for fixing all bond lengths. The TIP3P model was employed for water simulation. Topology files and parameters of small compounds in protein-ligand complexes were generated for GROMACS simulation by the SwissParam web server; the concentration of NaCl model was set as 0.145 M in the solvent system. The Steepest Descent algorithm performed the 5,000 cycle steps used for energy minimization. This was followed by equilibration performed under position restraints for 1 ns under constant temperature dynamics (NVT type) conditions. Following this step, all production dynamics simulations were performed for 5000 ps under constant pressure and temperature dynamics (NPT type). The temperature in all of the simulation systems was set as 310 K. MD conformations were saved every 20 ps for trajectory analysis.
We used the LibDock score to select potent TCM compounds which have high affinity with ApoE4. The results of the docking score are listed in Table
ADMET prediction of top ten TCM compounds from docking results.
Name | LibDock score | aAbsorption | bBBB Level | cCYP2D6 | dHepatotoxicity |
---|---|---|---|---|---|
Solapalmitine |
|
|
|
|
|
Isodesacetyluvaricin |
|
|
|
|
|
Budmunchiamine L5 |
|
|
|
|
|
Hemiariensin | 115.334 | 0 | 2 | 0 | 0 |
Niranthin | 112.802 | 0 | 1 | 0 | 0 |
Platyphyllonol | 112.212 | 0 | 2 | 0 | 0 |
Triptofordin B1 | 111.954 | 0 | 2 | 0 | 0 |
Aglaiduline | 111.306 | 0 | 2 | 0 | 0 |
Aurantiamide | 110.823 | 0 | 2 | 0 | 0 |
Lobelanidine | 108.696 | 0 | 2 | 0 | 0 |
aAbsorption: good absorption = 0; moderate absorption = 1; low absorption = 2; bBBB level (blood brain barrier): very high penetration = 0; high penetration = 1; medium penetration = 2; low penetration = 3; undefined penetration = 4.
cCYP2D6: noninhibitor = 0, inhibitor = 1.
dHepatotoxicity: noninhibitor = 0, inhibitor = 1.
Chemical scaffold of (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5.
Docking poses of top three candidates: (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5. The small molecular and amino acids are colored in green and yellow, respectively.
H-bond and hydrophobic analysis of docking poses by Ligplot plus tool for each docked ligand in ApoE4: (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5.
In comparison with the docking study, PONDR-FIT was used to predict the disorder region among all residues on ApoE4 (Figure
Disorder prediction of sequence of ApoE4 from the results of PONDR-FIT. The value of disorder disposition above 0.5 in disorder disposition.
Complexes of ApoE4 with docked ligands were performed by MD simulation at 5000 ps, and ApoE4 with no ligand (Apo protein) were regarded as the control for comparison. Each plot of the root mean square deviation (RMSD), mean square displacement (MSD), and radius of gyration (
Plots of (a) protein RMSD, (b) ligand MSD, and (c) radius of gyration from ApoE4 with docked ligand or no ligand (apo) with a simulation time of 5000 ps.
Total energy of ApoE4 with docked ligand: (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5 from all simulation times; the no-ligand binding protein (d) was used as the control.
Root mean squared fluctuation (RMSF) was carried out to analyze the fluctuation of residues on ApoE4 protein (Figure
RMSF values of ApoE4 with docked ligand or no ligand (Apo) (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5 with simulation times of 5000 ps; the no-ligand binding protein (d) was used as the control.
Matrix of smallest distance between each pair of amino acids in the complex with (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5; the no-ligand binding protein (d) is used as the control.
A cluster algorithm was employed to select the most stable conformation over all simulation times. All MD snapshots with docked ligands were grouped into two or four individual clusters (Figure
Time of middle structure in each cluster from all MD simulation times.
Cluster | Time of middle flame (ps) | ||
---|---|---|---|
Solapalmitine | Isodesacetyluvaricin | Budmunchiamine L5 | |
1 | 1540 | 60 | 640 |
2 | 4040 | 280 | 40 |
3 | — | 1920 | 1480 |
4 | — | 4240 | 4340 |
Clustering analyses of protein conformations: (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5 with simulation times of 5000 ps.
The middle structure from each final clustering group with docked ligand: (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5. The small molecular and amino acids are colored green and yellow, respectively.
Ligand pathway prediction form protein conformations: (a) Solapalmitine, (b) Isodesacetyluvaricin, and (c) Budmunchiamine L5 and (d) Apo form of APOE4 with simulation times of 5000 ps.
Solapalmitine is the top candidate by LibDock score but displays significantly increasing MSD values due to unstable binding with ApoE4 over 5000 ps simulation time. Isodesacetyluvaricin has H-bond with Arg158; unfortunately, the H-bond is missing during MD simulation. The LibDock score of Budmunchiamine L5 is not the highest score from the TCM database screening; however, the dynamics simulation shows that the docked ligand complex of ApoE4 is stable. In snapshot analysis, Budmunchiamine L5 still forms an H-bond with Glu77; the binding pose is the same as the initial docking pose, suggesting that Budmunchiamine L5 does not change over all simulation times and stably binds to ApoE4. In terms of ADMET analysis, Budmunchiamine L5 has good absorption, BBB penetration, and less toxicity in the human liver and may therefore be regarded as a safe lead drug for designing a novel ApoE4 inhibitor for AD therapy.
The authors declare that there is no conflict of interests.
Hung-Jin Huang and Hsin-Yi Chen contributed equally to this work.
The research was supported by grants from the National Science Council of Taiwan (NSC102-2325-B039-001, NSC102-2221-E-468-027-), Asia University (ASIA100-CMU-2, ASIA101-CMU-2, 102-ASIA-07), and China Medical University Hospital (DMR-102-051, DMR-103-058, DMR-103-001, DMR-103-096). This study is also supported in part by Taiwan Department of Health Clinical Trial and Research Center of Excellence (DOH102-TD-B-111-004), Taiwan Department of Health Cancer Research Center of Excellence (MOHW103-TD-B-111-03), and CMU under the Aim for Top University Plan of the Ministry of Education, Taiwan. Our gratitude goes to Tim Williams, Asia University.