High levels of homocysteine are reported as a risk factor for Alzheimer’s disease (AD). Correspondingly, inborn hyperhomocysteinemia is associated with an increased predisposition to the development of dementia in later stages of life. Yet, the mechanistic link between homocysteine accumulation and the pathological neurodegenerative processes is still elusive. Furthermore, despite the clear association between protein aggregation and AD, attempts to develop therapy that specifically targets this process have not been successful. It is envisioned that the failure in the development of efficacious therapeutic intervention may lie in the metabolomic state of affected individuals. We recently demonstrated the ability of metabolites to self-assemble and cross-seed the aggregation of pathological proteins, suggesting a role for metabolite structures in the initiation of neurodegenerative diseases. Here, we provide a report of homocysteine crystal structure and self-assembly into amyloid-like toxic fibrils, their inhibition by polyphenols, and their ability to seed the aggregation of the AD-associated β-amyloid polypeptide. A yeast model of hyperhomocysteinemia indicates a toxic effect, correlated with increased intracellular amyloid staining that could be rescued by polyphenol treatment. Analysis of AD mouse model brain sections indicates the presence of homocysteine assemblies and the interplay between β-amyloid and homocysteine. This work implies a molecular basis for the association between homocysteine accumulation and AD pathology, potentially leading to a paradigm shift in the understanding of AD initial pathological processes.

An atomic view of a main aqueous conformation of cyclosporine A (CycA), an important 11-amino-acid macrocyclic immunosuppressant, is reported. For decades, it has been a grand challenge to determine the conformation of free CycA in an aqueous-like solution given its poor water solubility. Using a combination of X-ray and single-crystal neutron diffraction, we unambiguously resolve a unique conformer (A1) with a novel cis-amide between residues 11 and 1 and two water ligands that stabilize hydrogen bond networks. NMR spectroscopy and titration experiments indicate that the novel conformer is as abundant as the closed conformer in 90/10 (v/v) methanol/water and is the main conformer at 10/90 methanol/water. Five other conformers were also detected in 90/10 methanol/water, one in slow exchange with A1, another one in slow exchange with the closed form and three minor ones, one of which contains two cis amides Abu2−Sar3 and MeBmt1−MeVal11. These conformers help better understand the wide spectrum of membrane permeability observed for CycA analogues and, to some extent, the binding of CycA to protein targets.

Our knowledge of amyloid formation and cytotoxicity originating from self-assembly of α-helical peptides is incomplete. PSMα3 is the only system where high-resolution X-ray crystallography and toxicity data are available. Oligomers of multiple α-helical monomers are less stable than those of β-strands, partially due to the lack of a consistent hydrogen-bonding network. It is challenging to preserve such oligomers in the gas phase where mass-selected structural studies using ion-mobility spectrometry mass spectrometry (IMS-MS) could be performed. As the oligomers fall apart after exiting the drift cell of the mass spectrometer, novel features that have shorter (a loss of charged species) or longer (a loss of neutral species) arrival times than expected are present together with those from the intact species. By obtaining a complete data set of PSMα3 peptides in solution and with n-dodecyl-β-d-maltoside, a micelle-forming detergent, we are able to discern the dissociated from the intact oligomers and detergent-bound complexes and correlate the reported cytotoxicity to the peptide oligomeric structures and their interactions with membrane mimetics. The study sheds new insights into the interpretation of IMS-MS data from biomolecular self-assembly studiesan important and timely topic.

Gas-phase ion-mobility spectrometry provides a unique platform to study the effect of mobile charge(s) or charge location on collisional cross section and ion separation. Here, we evaluate the effects of cation/anion adduction in a series of xylene and pyridyl macrocycles that contain ureas and thioureas. We explore how zinc binding led to unexpected deprotonation of the thiourea macrocyclic host in positive polarity ionization and subsequently how charge isomerism due to cation (zinc metal) and anion (chloride counterion) adduction or proton competition among acceptors can affect the measured collisional cross sections in helium and nitrogen buffer gases. Our approach uses synthetic chemistry to design macrocycle targets and a combination of ion-mobility spectrometry mass spectrometry experiments and quantum mechanics calculations to characterize their structural properties. We demonstrate that charge isomerism significantly improves ion-mobility resolution and allows for determination of the metal binding mechanism in metal–inclusion macrocyclic complexes. Additionally, charge isomers can be populated in molecules where individual protons are shared between acceptors. In these cases, interactions via drift gas collisions magnify the conformational differences. Finally, for the macrocyclic systems we report here, charge isomers are observed in both helium and nitrogen drift gases with similar resolution. The separation factor does not simply increase with increasing drift gas polarizability. Our study sheds light on important properties of charge isomerism and offers strategies to take advantage of this phenomenon in analytical separations.

While soluble forms of amyloid-β (Aβ) and Tau work together to drive healthy neurons into a disease state, how their interaction may control the prion-like propagation and neurotoxicity of Tau is not fully understood. The cross-linking via disulfide bond formation is crucial for Tau oligomers to obtain stable conformers and spread between cells. This work thus focuses on how Aβ42 regulates this critical process. By studying the interactions between Aβ42 and TauPHF43, a construct that mimics the Tau R3 isoform, has a similar length to Aβ42, and contains one cysteine (Cys-322), we discovered that fresh Aβ42 could protect Tau against the formation of disulfide cross-linked dimers. We showed that the monomeric and small Aβ oligomers (the “nonamyloidogenic Aβ”) efficiently disassembled tau dimers and heparin-induced Tau oligomers to recover Tau monomers. Interestingly, Aβ serves the role of an antioxidant to prevent disulfide bond formation, as supported by the experiments of Aβ with cystine. Furthermore, using cyclosporine A (CycA), a macrocyclic β-sheet disruptor, we demonstrated that targeting amyloidogenic Aβ with CycA does not affect the TauPHF43 disassembly driven by Aβ42. Separately, we assessed the initial toxicity of Aβ42 and TauPHF43 in acute brain slices and found that Aβ42 is more toxic than TauPHF43 or the two peptides combined. Our work highlights a potential protective role of Aβ42 monomers in AD that was previously overlooked while focusing on the mechanism behind Aβ42 aggregation leading to tau dysfunction.