Multiple ß-sheet molecular dynamics of amyloid formation from two ABl-SH3 domain peptides.

Molecular dynamics simulations in explicit water were carried out for two stacks, each composed of six 10-strand antiparallel ß-sheets for two peptides corresponding to the diverging turn of two homologous Abl-SH3 domains. The first system, referred to as 10×6×MK contained the DLSFMKGE sequence from the Drosophila, while the second one, referred to as 10×6×KK, contained the human DLSFKKGE sequence. It was found that the 10×6×MK ß-sheet stack is stable, but the 10×6×KK ß-sheet stack is not. The stability of the 10×6×MK ß-sheet stack results from the hydrophobic interactions of the methionine and phenylalanine residues and the leucine residues of the neighboring sheets. The Met, Phe, and Leu hydrophobic units make a hydrophobic core for the stack of ß-sheets. During the MD run, the Met, Phe, and Leu residues of the neighboring ß-sheets acted as a conformational switch moving the ß-sheets so that the Phe residue interacted with the Met residue from the neighboring ß-sheet. Replacement of Met by Lys destroys the hydrophobic core, which is the stability factor of the ß-sheet stack. For the 10×6×KK system, individual ß-sheets were preserved during simulations, but the interactions between the ß-sheets were lost. The calculations of a six ß-sheet stack confirm the conclusion drawn from our earlier studies of single ß-sheet systems that the ß-sheets must form stacks to be stabilized. These results suggest that the two conserved basic residues at the diverging turn of SH3 domains could act as gatekeepers to avoid aggregation.

Molecular dynamics study of amyloid formation of two Abl-SH3 domain peptides.

Molecular dynamics (MD) simulations were carried out for two-strand and ten-strand beta-sheets constructed from two peptides corresponding to the diverging turn of two homologous Abl-SH3 domains, DLSFMKGE (MK; from Drosophila) and DLSFKKGE (KK; from man), in explicit water at the temperatures of 30, 170/190 and 300 K. It was found that the 2 x MK beta-sheet is more stable than the 2 x KK beta-sheet, and that the 10 x MK beta-sheet is more stable than the 10 x KK beta-sheet; this suggests that the MK systems are fibril-creating and the KK systems are not. These results might explain why most SH3 domains possess two conserved basic residues at the diverging turn, which may act as gatekeepers in order to avoid aggregation.

Towards gelsolin amyloid formation.

Amyloid diseases result from protein misfolding and aggregation into fibrils. Some features of gelsolin amyloidogenic fragments comprised of residues 173-243 (G173-243) and residues 173-202 (G173-202) were investigated by the method of molecular dynamics (MD). The alpha-helical structure of G173-243 present in the whole protein unwinds during the course of MD simulation of the fragment G173-243, suggesting that the G173-243 structure is not stable and could unfold before becoming involved in gelsolin amyloid fibril formation. Twelve fragments of G173-202 were used to build a possible beta-fibril. During the course of the simulation, G173-202 fragments formed hydrogen bonds and tended to turn by an angle of 10 degrees -20 degrees towards each other.

The methylester of gamma-butyrobetaine, but not gamma-butyrobetaine itself, induces muscarinic receptor-dependent vasodilatation.

Gamma-butyrobetaine (GBB) is known mostly as a bio-precursor of carnitine, a key molecule in the regulation of myocardial energy metabolism. The metabolites of carnitine and GBB were investigated for acetylcholine-like activity decades ago. The present study shows that the methylester of GBB (GBB-ME) exerts its biological activity by binding to muscarinic acetylcholine receptors. GBB-ME dose-dependently decreased the blood pressure in anaesthetised rats and also produced endothelium-dependent vasodilation in the isolated guinea-pig heart. The biological effects of GBB-ME were inhibited partially by the NOS inhibitor N(omega)-nitro-L-arginine methylester (L-NAME) and abolished by the acetylcholine receptor antagonist atropine, thus supporting the hypothesis that GBB-ME acts as muscarinic agonist. Moreover, we have shown here for the first time that GBB-ME binds directly to transfected human muscarinic (m) acetylcholine receptors, the potency order being m2>m5> or =m4> or =m1>m3. GBB itself showed neither biological activity nor significant affinity for the m1-5 receptors. We conclude that GBB-ME, but not the parent GBB, possesses acetylcholine-like activity in vivo and in vitro.

Molecular dynamics study of a gelsolin-derived peptide binding to a lipid bilayer containing phosphatidylinositol 4,5-bisphosphate.

Gelsolin is an actin-severing protein whose action is initiated by Ca(2+) and inhibited by binding to phosphorylated inositol lipid or phosphoinositides. The regions of gelsolin responsible for phosphoinositide binding are comprised of residues 150-169 (G150-169) and 135-142 (G135-142). The corresponding peptides possess similar binding potency as native gelsolin. Their common feature is the presence of arginine and lysine residues that can bind to negatively charged phosphate groups of phosphoinositides. In this work the binding of the G150-169 peptide to a phosphatidylinositol 4,5-bisphosphate (PIP2) cluster in a lipid membrane model was investigated by molecular dynamics calculations (MD) with the AMBER 4.1 force field, taking into account explicit solvent molecules. Initially the structure of G150-169 was simulated by using the electrostatically driven Monte Carlo (EDMC) and MD methods, and the resulting structure agreed within 3.7 A backbone-atom root mean square deviation with the corresponding experimentally derived structure (PDB code: 1SOL). Using this model for the peptide, a subsequent MD simulation of G150-169 in a periodic box containing a model of dimyristoyl-phosphatidylcholine (DMPC) lipids with a cluster of four PIP2 molecules was carried out. During the simulation G150-169 interacted strongly with PIP2 molecules, initially by formation of salt bridges between its N-terminal basic groups and the phosphate groups of PIP2, followed by formation of hydrophobic bonds between the hydrophobic side chains of the peptide and the fatty acid tail of the lipid. As a result of the formation of hydrophobic bonds, the PIP2 molecules were pulled out from the lipid bilayer. This mode of binding differs from those of other PIP2-binding protein motifs such as PH domains that interact solely with the hydrophilic head group of PIP2. These results suggest that dissociation of gelsolin from actin by PIP2 lipids may involve entering of the PIP2 molecules to the gelsolin-actin interface, thereby weakening the interactions between these proteins.