ABSTRACT
We have identified macrocyclic inhibitors of the aspartic protease BACE, implicated in the etiology of Alzheimer's disease. An X-ray structure of screening hit 1 in the BACE active site revealed a hairpin conformation suggesting that constrained macrocyclic derivatives may also bind there. Several of the analogs we prepared were >100x more potent than 1, such as 7 (5 nM K(i)).
Subject(s)
Amyloid Precursor Protein Secretases/antagonists & inhibitors , Aspartic Acid Endopeptidases/antagonists & inhibitors , Macrocyclic Compounds/chemistry , Protease Inhibitors/chemistry , Quinazolines/chemistry , Amyloid Precursor Protein Secretases/metabolism , Aspartic Acid Endopeptidases/metabolism , Binding Sites , Computer Simulation , Humans , Macrocyclic Compounds/chemical synthesis , Macrocyclic Compounds/pharmacology , Protease Inhibitors/chemical synthesis , Protease Inhibitors/pharmacology , Quinazolines/chemical synthesis , Quinazolines/pharmacology , Structure-Activity RelationshipABSTRACT
A new aspartic protease inhibitory chemotype bearing a 2-amino-3,4-dihydroquinazoline ring was identified by high-throughput screening for the inhibition of BACE-1. X-ray crystallography revealed that the exocyclic amino group participated in a hydrogen bonding array with the two catalytic aspartic acids of BACE-1 (Asp(32), Asp(228)). BACE-1 inhibitory potency was increased (0.9 microM to 11 nM K(i)) by substitution into the unoccupied S(1)' pocket.
Subject(s)
Amyloid Precursor Protein Secretases/antagonists & inhibitors , Aspartic Acid Endopeptidases/antagonists & inhibitors , Models, Molecular , Quinazolines/chemical synthesis , Amyloid Precursor Protein Secretases/chemistry , Amyloid beta-Peptides/antagonists & inhibitors , Amyloid beta-Peptides/blood , Amyloid beta-Peptides/metabolism , Amyloid beta-Protein Precursor/genetics , Animals , Aspartic Acid Endopeptidases/chemistry , CHO Cells , Caco-2 Cells , Cell Membrane Permeability , Cricetinae , Cricetulus , Crystallography, X-Ray , Humans , Hydrogen Bonding , Molecular Conformation , Mutation , Oligopeptides/chemistry , Peptide Fragments/antagonists & inhibitors , Peptide Fragments/blood , Peptide Fragments/metabolism , Quinazolines/chemistry , Quinazolines/pharmacology , Rats , Stereoisomerism , Structure-Activity RelationshipABSTRACT
Computing the binding affinity of a protein-ligand complex is one of the most fundamental and difficult tasks in computer-aided drug design. Many approaches for computing binding affinities can be classified as linear interaction energy (LIE) models as they rely on some type of linear fit of computed interaction energies between ligand and protein. We have examined the computed interaction energies of a series of beta-secretase (BACE) inhibitors in terms of van der Waals, coulombic, and continuum-solvation contributions to ligand binding. We have also systematically examined the effect of different protonation states of the protein and ligands. We find that the binding affinities are relatively insensitive to the protonation state of the protein when neutral ligands are considered. Inclusion of charged ligands leads to large deviations in the coulomb, solvation, and even van der Waals terms. The latter is due to increased repulsive van der Waals interactions in the complex due to the strong coulomb attraction found between oppositely charged functional groups in the protein and ligand. In general, we find that the best models are obtained when the protein is judiciously charged (e.g. Asp32-, Arg235+) and the potentially charged ligands are treated as neutral.
Subject(s)
Endopeptidases/metabolism , Protease Inhibitors/chemistry , Protease Inhibitors/metabolism , Amyloid Precursor Protein Secretases , Energy Transfer , Ligands , Linear Models , Models, Molecular , Protein Binding , Static Electricity , ThermodynamicsABSTRACT
Alzheimer's Disease (AD) is a progressive neurodegenerative disorder marked by loss of memory, cognition, and behavioral stability. AD is defined pathologically by extracellular neuritic plaques comprised of fibrillar deposits of beta-amyloid peptide (Abeta) and neurofibrillary tangles comprised of paired helical filaments of hyperphosphorylated tau. Current therapies for AD, such as cholinesterase inhibitors, treat the symptoms but do not modify the progression of the disease. The etiology of AD is unclear. However, data from familial AD mutations (FAD) strongly support the "amyloid cascade hypothesis" of AD, i.e. that neurodegeneration in AD is initiated by the formation of neurotoxic beta-amyloid (Abeta) aggregates; all FAD mutations increase levels of Abeta peptide or density of Abeta deposits. The likely link between Abeta aggregation and AD pathology emphasizes the need for a better understanding of the mechanisms of Abeta production. This review summarizes current therapeutic strategies directed at lowering Abeta levels and decreasing levels of toxic Abeta aggregates through (1) inhibition of the processing of amyloid precursor protein (APP) to Abeta peptide, (2) inhibition, reversal or clearance of Abeta aggregation, (3) cholesterol reduction and (4) Abeta immunization.
