Implication of Caffeic Acid for the Prevention and Treatment of Alzheimer's Disease: Understanding the Binding with Human Transferrin Using In Silico and In Vitro Approaches.

In present times, a switch from chemical molecules towards natural products is taking place, and the latter is being increasingly explored in the management of diseases due to their broad range of therapeutic potential. Consumption of coffee is thought to reduce Alzheimer's disease (AD); however, the mechanism is still unexplored. Primarily, it is thought that components of coffee are the key players in making it a neuroprotectant. Caffeic acid (CA) is found in high quantities in coffee; thus, it is being increasingly explored to decipher its neuroprotection by various mechanisms. Iron is a toxic element in a free form capable of causing oxidative damage and ultimately contributing to the pathogenesis of AD. Thus, maintaining the proper iron levels is vital and human transferrin (Htf), a glycoprotein, is a key player in this aspect. In this work, we explored the binding mechanism of CA with Htf at the atomistic level, employing molecular docking and extensive molecular dynamics simulation (MD) approaches coupled with spectroscopic techniques in a bid to decipher the mode of interaction of CA with Htf. Molecular docking results demonstrated a strong binding affinity between CA and Htf. Furthermore, MD study highlighted the Htf-CA complex's stability and the ligand's minimal impact on Htf's overall structure. In silico approaches were further backed up by experimental approaches. Strong binding of CA with Htf was ascertained by UV-visible and fluorescence spectroscopy observations. Together, the study provides a comprehensive understanding of the Htf-CA interaction, adding to the knowledge of the use of CA in the treatment of AD, thereby adding another feather to its already known neuroprotective role.

Linagliptin and Empagliflozin Inhibit Microtubule Affinity Regulatory Kinase 4: Repurposing Anti-Diabetic Drugs in Neurodegenerative Disorders Using In Silico and In Vitro Approaches.

Type 2 diabetes mellitus (T2DM) and Alzheimer's disease (AD) are significant public health burdens. Many studies have revealed the possibility of common pathophysiology between T2DM and AD. Thus, in recent years, studies deciphering the action mechanism of anti-diabetic drugs with their future use in AD and related pathologies are on high demand. Drug repurposing is a safe and effective approach owing to its low cost and time-saving attributes. Microtubule affinity regulating kinase 4 (MARK4) is a druggable target for various diseases and is found to be linked with AD and diabetes mellitus. MARK4 plays a vital role in energy metabolism and regulation and thus serves as an irrefutable target to treat T2DM. The present study was intended to identify the potent MARK4 inhibitors among FDA-approved anti-diabetic drugs. We performed structure-based virtual screening of FDA-approved drugs to identify the top hits against MARK4. We identified five FDA-approved drugs having an appreciable affinity and specificity toward the binding pocket of MARK4. Among these identified hits, two drugs, linagliptin, and empagliflozin, favorably bind to the MARK4 binding pocket, interacting with its critical residues and thus subjected to detailed analysis. All-atom detailed molecular dynamics (MD) simulations revealed the dynamics of binding of linagliptin and empagliflozin with MARK4. Kinase assay showed significant inhibition of MARK4 kinase activity in the presence of these drugs, implying them as potent MARK4 inhibitors. In conclusion, linagliptin and empagliflozin may be promising MARK4 inhibitors, which can further be exploited as potential lead molecules against MARK4-directed neurodegenerative diseases.

Protein folding is a convergent problem!

Protein folding, tagged as a grand challenge/NP hard problem, has been an open area of research in diverse fields. Exploring folding from the configurational volume perspective in terms of phi (ϕ), psi(ψ) backbone dihedral angles, we asked ourselves a fundamental question as to when do the neighborhood effects on the allowed ϕ, ψ values of each residue take over to assure convergence of proteins to their observed unique tertiary structures. A mapping of the higher order steric correlations beyond Ramachandran plots from ∼43612 protein structures comprising ∼26.5 million amino acid residues reveals that conformational restrictions on allowed ϕ, ψ values of each amino acid residue due to the N and C terminal neighbors - essentially a consideration of the sterically allowed regions of a tripeptide - ensure convergence of the configurational volume.