Formulation Graphs for Mapping Structure-Composition of Battery Electrolytes to Device Performance.

Advanced computational methods are being actively sought to address the challenges associated with the discovery and development of new combinatorial materials, such as formulations. A widely adopted approach involves domain-informed high-throughput screening of individual components that can be combined together to form a formulation. This manages to accelerate the discovery of new compounds for a target application but still leaves the process of identifying the right "formulation" from the shortlisted chemical space largely a laboratory experiment-driven process. We report a deep learning model, the Formulation Graph Convolution Network (F-GCN), that can map the structure-composition relationship of the formulation constituents to the property of liquid formulation as a whole. Multiple GCNs are assembled in parallel that featurize formulation constituents domain-intuitively on the fly. The resulting molecular descriptors are scaled based on the respective constituent's molar percentage in the formulation, followed by integration into a combined formulation descriptor that represents the complete formulation to an external learning architecture. The use case of the proposed formulation learning model is demonstrated for battery electrolytes by training and testing it on two exemplary data sets representing electrolyte formulations vs battery performance: one data set is sourced from the literature about Li/Cu half-cells, while the other is obtained by lab experiments related to lithium-iodide full-cell chemistry. The model is shown to predict performance metrics such as Coulombic efficiency (CE) and specific capacity of new electrolyte formulations with the lowest reported errors. The best-performing F-GCN model uses molecular descriptors derived from molecular graphs (GCNs) that are informed with HOMO-LUMO and electric moment properties of the molecules using a knowledge transfer technique.

Hetero-oligomeric Amyloid Assembly and Mechanism: Prion Fragment PrP(106-126) Catalyzes the Islet Amyloid Polypeptide ß-Hairpin.

Protein aggregation is typically attributed to the association of homologous amino acid sequences between monomers of the same protein. Coaggregation of heterogeneous peptide species can occur, however, and is implicated in the proliferation of seemingly unrelated protein diseases in the body. The prion protein fragment (PrP106-126) and human islet amyloid polypeptide (hIAPP) serve as an interesting model of nonhomologous protein assembly as they coaggregate, despite a lack of sequence homology. We have applied ion-mobility mass spectrometry, atomic force microscopy, circular dichroism, and high-level molecular modeling to elucidate this important assembly process. We found that the prion fragment not only forms pervasive hetero-oligomeric aggregates with hIAPP but also promotes the transition of hIAPP into its amyloidogenic ß-hairpin conformation. Further, when PrP106-126 was combined with non-amyloidogenic rIAPP, the two formed nearly identical hetero-oligomers to those seen with hIAPP, despite rIAPP containing ß-sheet breaking proline substitutions. Additionally, while rIAPP does not natively form the amyloidogenic ß-hairpin structure, it did so in the presence of PrP106-126 and underwent a conformational transition to ß-sheet in solution. We also find that PrP106-126 forms hetero-oligomers with the IAPP8-20 fragment but not with the "aggregation hot spot" IAPP20-29 fragment. PrP106-126 apparently induces IAPP into a ß-hairpin structure within the PrP:IAPP heterodimer complex and then, through ligand exchange, catalytically creates the amyloidogenic ß-hairpin dimer of IAPP in significantly greater abundance than IAPP does on its own. This is a new mechanistic model that provides a critical foundation for the detailed study of hetero-oligomerization and prion-like proliferation in amyloid systems.

Amyloid ß-Protein Assembly and Alzheimer's Disease: Dodecamers of Aß42, but Not of Aß40, Seed Fibril Formation.

Evidence suggests that oligomers of the 42-residue form of the amyloid ß-protein (Aß), Aß42, play a critical role in the etiology of Alzheimer's disease (AD). Here we use high resolution atomic force microscopy to directly image populations of small oligomers of Aß42 that occur at the earliest stages of aggregation. We observe features that can be attributed to a monomer and to relatively small oligomers, including dimers, hexamers, and dodecamers. We discovered that Aß42 hexamers and dodecamers quickly become the dominant oligomers after peptide solubilization, even at low (1 µM) concentrations and short (5 min) incubation times. Soon after (≥10 min), dodecamers are observed to seed the formation of extended, linear preprotofibrillar ß-sheet structures. The preprotofibrils are a single Aß42 layer in height and can extend several hundred nanometers in length. To our knowledge this is the first report of structures of this type. In each instance the preprotofibril is associated off center with a single layer of a dodecamer. Protofibril formation continues at longer times, but is accompanied by the formation of large, globular aggregates. Aß40, by contrast, does not significantly form the hexamer or dodecamer but instead produces a mixture of smaller oligomers. These species lead to the formation of a branched chain-like network rather than discrete structures.

Inhibiting and Remodeling Toxic Amyloid-Beta Oligomer Formation Using a Computationally Designed Drug Molecule That Targets Alzheimer's Disease.

Alzheimer's disease (AD) is rapidly reaching epidemic status among a burgeoning aging population. Much evidence suggests the toxicity of this amyloid disease is most influenced by the formation of soluble oligomeric forms of amyloid ß-protein, particularly the 42-residue alloform (Aß42). Developing potential therapeutics in a directed, streamlined approach to treating this disease is necessary. Here we utilize the joint pharmacophore space (JPS) model to design a new molecule [AC0107] incorporating structural characteristics of known Aß inhibitors, blood-brain barrier permeability, and limited toxicity. To test the molecule's efficacy experimentally, we employed ion mobility mass spectrometry (IM-MS) to discover [AC0107] inhibits the formation of the toxic Aß42 dodecamer at both high (1:10) and equimolar concentrations of inhibitor. Atomic force microscopy (AFM) experiments reveal that [AC0107] prevents further aggregation of Aß42, destabilizes preformed fibrils, and reverses Aß42 aggregation. This trend continues for long-term interaction times of 2 days until only small aggregates remain with virtually no fibrils or higher order oligomers surviving. Pairing JPS with IM-MS and AFM presents a powerful and effective first step for AD drug development. Graphical Abstract.

In situ AFM visualization of Li-O2 battery discharge products during redox cycling in an atmospherically controlled sample cell.

The in situ observation of electrochemical reactions is challenging due to a constantly changing electrode surface under highly sensitive conditions. This study reports the development of an in situ atomic force microscopy (AFM) technique for electrochemical systems, including the design, fabrication, and successful performance of a sealed AFM cell operating in a controlled atmosphere. Documentation of reversible physical processes on the cathode surface was performed on the example of a highly reactive lithium-oxygen battery system at different water concentrations in the solvent. The AFM data collected during the discharge-recharge cycles correlated well with the simultaneously recorded electrochemical data. We were able to capture the formation of discharge products from correlated electrical and topographical channels and measure the impact of the presence of water. The cell design permitted acquisition of electrochemical impedance spectroscopy, contributing information about electrical double layers under the system's controlled environment. This characterization method can be applied to a wide range of reactive surfaces undergoing transformations under carefully controlled conditions.

Zinc-Induced Conformational Transitions in Human Islet Amyloid Polypeptide and Their Role in the Inhibition of Amyloidosis.

Type-2 diabetes mellitus (T2DM) is a disease hallmarked by improper homeostasis within the islets of Langerhans of the pancreas. The most critical species affected is insulin, which is produced by the ß-cells of the islets, but there are a number of other species copackaged and cosecreted within the insulin granules. This includes zinc, which exists in high (millimolar) concentrations within the ß-cells, and islet amyloid polypeptide (IAPP), which is an amyloid peptide thought to induce ß-cell apoptosis through self-association into toxic amyloid oligomers. Zinc is essential in the packaging of crystalline insulin within the vesicles but it can also bind and interact with IAPP. This implies a complex relationship between all three species and diabetes, particularly in the structure and function of toxic IAPP aggregates. Atypical (low or high) concentrations of zinc generally appear to correlate with increased hIAPP aggregation, whereas physiological zinc concentrations have an inhibitory effect. To better understand how zinc ions alter the monomer and oligomer structure of hIAPP in vitro, we employ a combination of ion mobility mass spectrometry and atomic force microscopy. We observe an increase in the extended ß-hairpin conformation of hIAPP when it is bound to zinc. With sufficiently low concentrations of zinc this could result in an association site for zinc-free hIAPP, promoting amyloid aggregation. At high zinc concentrations, we see the appearance of a secondary zinc association site whose coordination could account for the loss of inhibition at high zinc concentrations. Generally, it appears that zinc preferentially stabilizes the ß-hairpin conformation of hIAPP and the population of zinc-bound hIAPP in solution determines what effect this has on amyloid aggregation.

Distal amyloid ß-protein fragments template amyloid assembly.

Amyloid formation is associated with devastating diseases such as Alzheimer's, Parkinson's and Type-2 diabetes. The large amyloid deposits found in patients suffering from these diseases have remained difficult to probe by structural means. Recent NMR models also predict heterotypic interactions from distinct peptide fragments but limited evidence of heterotypic packed sheets is observed in solution. Here we characterize two segments of the protein amyloid ß (Aß) known to form fibrils in Alzheimer's disease patients. We designed two variants of Aß(19-24) and Aß(27-32), IFAEDV (I6V) and NKGAIF (N6F) to lower the aggregation propensity of individual peptides while maintaining the similar interactions between the two segments in their native forms. We found that the variants do not form significant amyloid fibrils individually but a 1:1 mixture forms abundant fibrils. Using ion mobility-mass spectrometry (IM-MS), hetero-oligomers up to decamers were found in the mixture while the individual peptides formed primarily dimers and some tetramers consistent with a strong heterotypic interaction between the two segments. We showed by X-ray crystallography that I6V formed a Class 7 zipper with a weakly packed pair of ß-sheets and no segregated dry interface, while N6F formed a more stable Class 1 zipper. In a mixture of equimolar N6F:I6V, I6V forms a more stable zipper than in I6V alone while no N6F or hetero-typic zippers are observed. These data are consistent with a mechanism where N6F catalyzes assembly of I6V into a stable zipper and perhaps into stable, pure I6V fibrils that are observed in AFM measurements.

Human Islet Amyloid Polypeptide Assembly: The Key Role of the 8-20 Fragment.

The aggregation of human islet amyloid polypeptide (hIAPP) has been closely associated with the pathogeny of type 2 diabetes mellitus (T2DM) and destruction of pancreatic islet ß-cells. Several amyloidogenic domains within the hIAPP sequence capable of self-association have been identified. Among them is the 8-20 region of hIAPP, which has formed ß-sheet fibrils despite being contained within an α-helical region of full-length hIAPP. To further understand the propensity of this region for self-assembly, two peptide fragments were compared, one consisting of the residues 8-20 (WT8-20) and a mutant fragment with a His18Pro substitution (H18P8-20). The conformational distribution and aggregation propensity of these peptides was determined using a combination of ion mobility mass spectrometry and atomic force microscopy. Our results reveal that the two peptide fragments have vastly differing assembly pathways. WT8-20 produces a wide range of oligomers up to decamer whereas the H18P8-20 mutant produces only low order oligomers. This study confirms the propensity of the 8-20 region to aggregate from its native α-helical structure into amyloid ß-sheet oligomers and highlights the significance of the charged His18 in the aggregation process.

Human Islet Amyloid Polypeptide N-Terminus Fragment Self-Assembly: Effect of Conserved Disulfide Bond on Aggregation Propensity.

Amyloid formation by human islet amyloid polypeptide (hIAPP) has long been implicated in the pathogeny of type 2 diabetes mellitus (T2DM) and failure of islet transplants, but the mechanism of IAPP self-assembly is still unclear. Numerous fragments of hIAPP are capable of self-association into oligomeric aggregates, both amyloid and non-amyloid in structure. The N-terminal region of IAPP contains a conserved disulfide bond between cysteines at position 2 and 7, which is important to hIAPP's in vivo function and may play a role in in vitro aggregation. The importance of the disulfide bond in this region was probed using a combination of ion mobility-based mass spectrometry experiments, molecular dynamics simulations, and high-resolution atomic force microscopy imaging on the wildtype 1-8 hIAPP fragment, a reduced fragment with no disulfide bond, and a fragment with both cysteines at positions 2 and 7 mutated to serine. The results indicate the wildtype fragment aggregates by a different pathway than either comparison peptide and that the intact disulfide bond may be protective against aggregation due to a reduction of inter-peptide hydrogen bonding. Graphical Abstract ᅟ.