Shengyuan Wang, PhD
School of Systems Biology
George Mason University
Manassas, VA 20110

Investigate the mechanism of how Ibuprofen molecules block the aggregation process of Aβ peptide

Introduction

Alzheimer's Disease (AD) pathogenesis is widely believed to be driven by the production and deposition of the amyloid-β peptide (Aβ), which is a ∼ 4 kDa peptide. It has been found that chronic orally administered Ibuprofen, the most commonly used nonsteroidal anti-inflammatory drug, was able to produce significant diminution in the ultimate number and total area of β-amyloid deposits in transgenic mouse model with AD. However, the mechanism is still not fully understood. To investigate the mechanism of how Ibuprofen molecules block the aggregation process of Aβ, designed molecular simulations were performed in this study. The effect of Ibuprofen molecules on the aggregation of Aβ16-22 fragment was studied in the simulation. The secondary structures of Aβ16-22 fragment were measurable outcomes in the simulation to answer the research question.


Materials and Methods

Aβ16-22 fragment structure was obtained from PDB, solvated either in the water system or water+Ibuprofen system. Simulation was completed in four steps, minimization, heating, equilibration and quenching. The consistency of each simulation was ensured by examining the corresponding output file. The secondary structure of each residue was obtained from quenching output files, which were based on the coordinates of each atom and identified by STRIDE in VMD.


Results

• At the initial phase of production, Aβ16-22 fragment had a high fraction of α-helix in both Experimental and Control systems. After 1500 picosecond, α-helices in the Control system (without Ibuprofen) had higher chance to be dissolved compared to those in the Experimental system (with Ibuprofen). The mean of α-helix fraction in Experimental system was 0.64 with a standard deviation of 0.35, while only 0.38 in Control system with a standard deviation of 0.35. The mean time step in which half of the initial helix fraction was dissolved was 15,550 in Experimental system and 11,891 in Control system. The greater mean time step in the Experimental system indicated that that α-helix in Aβ16-22 fragment was more stable when Ibuprofen existed. In both system, 310-helix appeared when α-helix dissolved.

Figure1. Fragment of all helix, alpha helix and 3-10 helix within system in different trajectories


• Secondary structure propensity for each residue in Aβ16-22 fragment was calculated. α-helix appeared at residue 1-6 but not at residue 7. The fraction of secondary structures of α-helix ranged between approximate 0.4 (Trajectory 3) and 0.9 (Trajectory 1) in Experimental system, while it ranged between approximate 0.1 (Trajectory 1) and 0.65 (Trajectory 2). Overall, the fraction of secondary structures of α-helix was higher in Experimental than Control system.
  The number of secondary structures of 310-helix was also obtained. It reached peak of approximate 0.15 at residue 3 and 4 in Trajectory 3 in Experimental system, while the peak was also approximate 0.15 but at residue 2, 3 and 4 in Trajectory 3 in control system. Therefore, there was no difference in fraction of 310-helix secondary structure between Experimental and Control systems.

Figure2. Propensities of different secondary structure for each residue


• In both systems, α-helix may reform or break once it was dissolved. From the aspect of hydrogen bond formation and break down, my assumption is that water molecules will affect the hydrogen bonds in α-helix, either donor or acceptor in water molecules repulse or attract the electrons so that α-helix break. When water molecules dispart from the Aβ16-22 fragment, there is possibility that α-helix reforms, unless the peptide is dissolved. Ibuprofen molecule has a carboxyl group, it may attract those water molecules that combine with Aβ16-22 fragment, which help Aβ16-22 fragment reforms the α-helix. From the aspect of potential energy, α-helix is assigned from atomic coordinates based on the combined use of hydrogen bond energy and statistically derived backbone torsional angle information in STRIDE. It means that the difference between α and α-helix or α and non-α-helix may vary. Once the residue is defined as non-α-helix, there's may just a slight difference compare to α-helix, so it would easy for these residues redefined as α-helix. Besides, α-helix is a relatively stable structure formation. Once it is formed, energy barrier must be crossed so that the α-helix can be dissolved. If the STRIDE algorithm defines that one α-helix just dissolved, but it does not cross the energy barrier, the α-helix will be easily reformed. On the contrary, if it crosses the energy barrier, the α-helix will not be reformed.


• Another interesting observation is 310-helix will appear once α-helix dissolved in both Experimental and Control system. Dynamics simulations of helices have shown a tendency to some formation of 310-helix in predominantly a-helical molecules, and the 310-helix conformation has been implicated as an intermediate in unfolding of α-helix to form extended conformation. Out simulations analysis show the correlation between α-helix and 310-helix is high negative correlated (around -0.98) and the correlation was statistically significant (P-value = 2.2e-16). Besides, 310-helix at residue 3 and 4 which are Valine and Phenylalanine, showing a small peak appearance in both systems.