Protein nanofibril design via manipulation of hydrogen bonds

Protein self-assembly, particularly fibrillar self-assembly, is driven by non-covalent interactions, playing dual biological roles: aberrant self-assembly forming amyloid fibrils associated with neurodegenerative diseases (like Alzheimer's and Parkinson's), and functional self-assembly creating strong, elastic fibers (functional amyloids) such as silk. The mechanical properties of amyloid fibrils are critically involved in disease progression, while functional amyloids show promise as biomaterials. Understanding how self-assembly pathways shape amyloid material properties is vital for both therapeutics and materials design. Despite variations in sequence, amyloid fibers share a similar molecular organization: β-strands perpendicular to the fibril axis, linked by a dense H-bond network forming extended β-sheets. Sequences rich in A, G, or AG motifs in various proteins are prone to amyloid fibril formation, increasing the risk of Huntington's and mad cow diseases. Interestingly, these motifs are also found in functional amyloids like silk fibroin and Pmel17. Hydrophobic interactions, π-π stacking, and H-bonds stabilize amyloid structures, with H-bonds being a key driving force for aggregation. π-stacking interactions often accelerate fibril formation by promoting directional growth. However, the relationship between these interactions and the resulting fibril mechanical strength remains unclear; Aβ fibrils show an elastic modulus of 2–5 GPa, while FF assemblies exhibit 20–30 GPa. This research aims to understand how molecular interactions, particularly H-bonds and aromatic residue interactions, shape the physical properties of self-assembled amyloid protein constructs.

Extensive research has explored the role of various interactions in amyloid fibrillation. Studies highlight the importance of hydrophobic interactions, pi-pi stacking, and hydrogen bonds in stabilizing amyloid structures. The literature shows that H-bonds are a key driving force for aggregation, while pi-stacking interactions often accelerate the process by promoting directional growth. However, the precise relationship between these interactions and the resulting mechanical properties of amyloid fibrils remains a subject of ongoing investigation. Studies on Aβ and FF assemblies show a significant variation in the elastic modulus, highlighting the complexity of this relationship. This paper builds on existing knowledge by focusing on the controlled manipulation of hydrogen bonds to elucidate their influence on amyloid fibril formation and material properties.

This study employed a peptide design approach to control molecular interactions and H-bond network formation. A representative polypeptide sequence (GAGAGSGA GAGSGAGAGSGAG) with a high propensity for amyloidogenic fibrillar aggregation was chosen. This sequence mimics both functional (silk fibroin) and aberrant (disease-associated) self-assembly. Glycine (G) residues were systematically substituted with aromatic amino acids: Phenylalanine (F), Tyrosine (Y), and Tryptophan (W). These substitutions introduced two types of changes: 1) introduction of aromatic residues in the core, interfering with H-bond network formation, and 2) delocalization of aromatic residues at the fibril interface to promote or limit fibril elongation. The aromatic amino acid substitutes differ in hydrophobicity, polarity, and H-bonding ability. The peptides were dissolved in DMSO (dimethyl sulfoxide), a solvent that, while not entirely mimicking intracellular conditions, has been shown not to abolish biological activity or amyloidogenic aggregation. The high aggregation propensity and low solubility of amyloidogenic peptides often necessitate solubilization in organic solvents like DMSO before analysis. Several techniques were used to analyze the resulting peptide assemblies: * **Atomic Force Microscopy (AFM):** To assess fibril morphology and measure nanomechanical properties (elastic modulus) using the DMT model. * **Transmission Electron Microscopy (TEM):** To further characterize fibril morphology. * **Thioflavin T (ThT) Assay:** To monitor the kinetics of aggregation, determining nucleation and elongation rates. * **Fourier Transform Infrared Spectroscopy (FT-IR):** To characterize the secondary structure of the assemblies, analyzing the amide I and II regions to identify β-sheet, α-helix, random coil, and other structures. * **X-ray Photoelectron Spectroscopy (XPS):** To investigate the H-bonded network by analyzing changes in binding energies of neighboring atoms (carbon and nitrogen) due to charge transfer upon H-bond formation. Electron diffraction analysis was used to study the atomic structures of peptide assemblies, determining inter-strain and inter-sheet distances. The effects of varying the percentage (10% and 30%) and position of aromatic substitutions on fibril morphology, kinetics, secondary structure, hydrogen bonding, and mechanical properties were systematically investigated.

The study revealed that alterations in H-bond networks do not necessarily prevent amyloid formation but significantly impact the final fibril structure and mechanical properties. * **Morphology:** The unsubstituted peptide (P1) formed elongated fibrils. Substitution with F at the termini (P2) resulted in longer fibrils, while substitution with Y (P4) produced shorter fibrils. Substitution with W (P6) led to both spherical and fibrillar structures. Increasing the aromatic residue fraction to ~30% resulted in varied morphologies depending on the amino acid; F led to beaded fibrils and spheres, while Y and W resulted in continuous fibrils. * **Kinetics:** Peptides with low aromatic content nucleated faster than those with high aromatic content, indicating that the aromatic residues increased the thermodynamic stability of the peptides. The presence of aromatic amino acids, especially Y, increased the elongation rate. * **Secondary Structure:** FT-IR analysis confirmed the presence of aggregated β-sheets in all peptide assemblies. The type and amount of aromatic substitution influenced the relative proportions of β-sheet, α-helix, and random coil structures. The presence of F showed a relative increase in the random coil/α-helical fraction. Y residues promoted β-sheet formation while preserving a significant fraction of random coil and α-helix structure. W showed the preservation of disordered random coil/α-helical content. * **Hydrogen Bonding:** XPS analysis indicated that while aromatic substitutions decreased the overall number of H-bonds, they didn't always decrease the C^H/(C^H+C^am) ratio. This suggests that the aromatic side chains can either hinder backbone H-bond formation or serve as alternative H-bond donors (Y, W). The 30% G-to-Y/W substitution resulted in fewer, stronger backbone H-bonds. * **Nanomechanics:** Peptides P3 and P7, which had high aromatic content showed increased elastic modulus values, indicating higher stiffness. The spherical assemblies of the same peptides showed lower modulus values, highlighting the role of the continuous H-bonded network.

The results demonstrate that controlled manipulation of molecular interactions, specifically H-bonds, can effectively tune the properties of amyloid fibrils. The findings highlight the interplay between three key factors: steric hindrance, H-bond formation, and solvent interactions. Steric hindrance from aromatic residues influences nucleation and fibril growth. H-bond formation, particularly via Y and W side chains, creates alternative assembly pathways. Solvent interactions (DMSO) compete with fibril formation, particularly pronounced with F and W. The localization of aromatic residues (at the interface or core) also affects the mechanical properties. The ability to control fibril morphology and mechanical properties by manipulating H-bond networks opens up possibilities in designing biomaterials with tailored properties. The observation of different morphologies and mechanical properties emphasizes the complexity of amyloid self-assembly and the need for a multifaceted approach for understanding and controlling this process. Further investigation into the interplay of these factors with varying peptide sequences is warranted.

This study presents a novel approach to control the properties of amyloid fibrils by manipulating hydrogen bonds through strategic amino acid substitutions. The results highlight the crucial roles of steric hindrance, hydrogen bond formation, and solvent interactions in shaping fibril morphology, aggregation kinetics, and mechanical properties. This work opens new avenues for designing biomaterials with precisely controlled characteristics by tailoring molecular interactions at the peptide level. Future research could focus on exploring a wider range of amino acid substitutions and investigating the impact of these findings on various biological and materials applications.

The study utilized DMSO as a solvent, which might not perfectly replicate intracellular conditions. The analysis primarily focused on a specific peptide sequence; further investigations using other sequences are necessary to generalize the findings. The nanomechanical measurements were performed on dried samples; measurements on hydrated samples would provide a more realistic representation of the mechanical properties under physiological conditions. Finally, the complexity of amyloid self-assembly means these findings represent a snapshot of a highly dynamic process.