Embeddings from protein Language Models (pLMs) are replacing evolutionary information from multiple sequence alignments (MSAs) as the most successful input for protein prediction. Is this because embeddings capture evolutionary information? We tested various approaches to explicitly incorporate evolutionary information into embeddings on various protein prediction tasks. While older pLMs (SeqVec, ProtBert) significantly improved through MSAs, the more recent pLM ProtT5 did not benefit. For most tasks, pLM-based outperformed MSA-based methods, and the combination of both even decreased performance for some (intrinsic disorder). We highlight the effectiveness of pLM-based methods and find limited benefits from integrating MSAs.

The Ebola delta peptide is an amphipathic, 40-residue peptide encoded by the Ebola virus, referred to as E40. The membrane-permeabilising activity of the E40 delta peptide has been demonstrated in cells and lipid vesicles suggesting the E40 delta peptide likely acts as a viroporin. The lytic activity of the peptide increases in the presence of anionic lipids and a disulphide bond in the C-terminal part of the peptide. Previous in silico work predicts the peptide to show a partially helical structure, but there is no experimental information on the structure of E40. Here, we use circular dichroism spectroscopy to report the secondary structure propensities of the reduced and oxidised forms of the E40 peptide in water, detergent micelles, and lipid vesicles composed of neutral and anionic lipids (POPC and POPG, respectively). Results indicate that the peptide is predominately a random coil in solution, and the disulphide bond has a small but measurable effect on peptide conformation. Secondary structure analysis shows large uncertainties and dependence on the reference data set and, in our system, cannot be used to accurately determine the secondary structure motifs of the peptide in membrane environments. Nevertheless, the spectra can be used to assess the relative changes in secondary structure propensities of the peptide depending on the solvent environment and disulphide bond. In POPC-POPG vesicles, the peptide transitions from a random coil towards a more structured conformation, which is even more pronounced in negatively charged SDS micelles. In vesicles, the effect depends on the peptide-lipid ratio, likely resulting from vesicle surface saturation. Further experiments with zwitterionic POPC vesicles and DPC micelles show that both curvature and negatively charged lipids can induce a change in conformation, with the two effects being cumulative. Electrostatic screening from Na+ ions reduced this effect. The oxidised form of the peptide shows a slightly lower propensity for secondary structure and retains a more random coil conformation even in the presence of PG-PC vesicles.

Beta turns, in which the protein backbone abruptly changes direction over four amino acid residues, are the most common type of protein secondary structure after alpha helices and beta sheets and play key structural and functional roles. Previous work has produced classification systems for turn geometry at multiple levels of precision, but these operate in backbone dihedral-angle (Ramachandran) space, and the absence of a local Euclidean-space coordinate system and structural alignment for turns, or of any systematic Euclidean-space characterization of turn backbone shape, presents challenges for the visualization, comparison and analysis of the wide range of turn conformations and the design of turns and the structures that incorporate them. This work derives a turn-local coordinate system that implicitly aligns turns, together with a set of geometric descriptors that characterize the bulk BB shapes of turns and describe modes of structural variation not explicitly captured by existing systems. These modes are shown to be meaningful by the demonstration of clear relationships between descriptor values and the electrostatic energy of the beta-turn H-bond, the overrepresentations of key side-chain motifs, and the structural contexts of turns. Geometric turn descriptors complement Ramachandran-space classifications, and they can be used to select turn structures for compatibility with particular side-chain interactions or contexts. Potential applications include protein design and other tasks in which an enhanced Euclidean-space characterization of turns may improve understanding or performance. The web-based tools ExploreTurns, MapTurns, and ProfileTurn, available at www.betaturn.com, incorporate turn-local coordinates and turn descriptors and demonstrate their utility.

Secondary structure refers to highly regular local sub-structures formed by the polypeptide backbone through hydrogen bonding. The two main types of secondary structures are α-helices and β-strands (which can form β-sheets). The development of a robust circular dichroism (CD) method for structural analysis of biomolecules requires careful consideration of several key factors. Solvent selection plays a crucial role in maintaining the native or desired conformation of the sample while ensuring transparency in the relevant wavelength regions. Aqueous buffers are often preferred for studying proteins in their native state. Optimizing the sample concentration and path length is essential to achieve an optimal absorbance range and maximize the signal-to-noise ratio. Typical concentrations for far-UV CD measurements range from 0.1 to 1 mg/ml, with shorter path lengths (1 mm) allowing for higher concentrations and longer path lengths (5 mm) suitable for dilute solutions. Instrumental parameters, such as scanning speed, accumulations, and nitrogen flow rate, significantly impact the quality and reliability of the acquired CD spectra. Data processing is a critical step in obtaining accurate and interpretable CD spectra. Baseline correction, smoothing, and conversion to mean residue ellipticity are essential for reliable secondary structure analysis.

To reveal the structural mechanism by which the low-complexity domain of the fused in sarcoma protein (FUS-LC) mediates liquid-liquid phase separation (LLPS), we conducted a vacuum-ultraviolet circular dichroism (VUV-CD) spectroscopic study, a technique to analyze the secondary structures of proteins. The VUV-CD measurements were performed at the BL12 VUV-CD station at the Hiroshima Synchrotron Radiation Center (HiSOR) in Japan. CD spectra were measured between 180 and 260 nm while controlling the temperature of samples from 37°C to 5°C to obtain the LLPS of FUS-LC. The CD spectrum obtained at 37°C exhibited a large negative peak at 195 nm and a small negative shoulder near 220 nm. The peak intensity around 195 nm decreased as the sample temperature decreased. The spectral changes originated from the LLPS formation.

Antifreeze peptide (AFP) including in frozen protein ink is an inevitable trend because AFP can make protein ink suitable for 3D printing after freezing. AFP-based surimi ink (ASI) was firstly investigated, and the AFP significantly enhanced 3D printability of frozen surimi ink. The rheological and textural results of ASI show that the τ0, K, and n values are 321.14 Pa, 2.2259 × 105 Pa·sn, and 0.19, respectively, and the rupture strength of the 3D structure is up to 217.67 g. Circular dichroism, intermolecular force, and differential scanning calorimeter show ASI has more undenatured protein after freezing when compared that surimi ink (SI), which was denatured, and the α-helix changed to a β-sheet due to the destruction of hydrogen bonds and the exposure of hydrophobic groups. The water distribution, water holding capacity, and microstructure indicate that ASI effectively binds free water after freezing, while SI has weak water binding capacity and a large amount of free water is formed. ASI is suitable for 3D printing, and can print up to 40.0 mm hollow isolation column and 50.0 mm high Wuba which is not possible with SI. The application of AFP provides guidance for 3D printing frozen protein ink in food industry.

Oligourea foldamers are known to fold into 2.5-helices, stabilized by three-centered hydrogen bonds, which makes them conformationally more rigid than peptides. Nevertheless, the folding propensity and conformational stability in solution depend on the length of the oligomer, as well as the temperature, solvent, and so forth. In the peptide field, there are many approaches known for constraining the backbone in the folded conformation, including the stapling of side chains by disulfide bridges, lactam formation, ring closing metathesis reaction, and others. In this work, we linked side chains by lactam bridges of short oligoureas (four residues), containing Glu- and Lys-like residues. The designed oligoureas differed in the position of the Glu-like residue. Next, the conformational properties of linear and cyclic compounds were studied in protic solvent (methanol) by nuclear magnetic resonance and circular dichroism. Importantly, it was discovered that larger macrocycles (24-membered) are more tolerated with respect to the helical turn than smaller macrocycles (19-membered) under the studied conditions.

Spike (S) glycoprotein is the largest structural protein of SARS-CoV-2 virus and the main one involved in anchoring of the host receptor ACE2 through the receptor binding domain (RBD). S protein secondary structure is of great interest for shedding light on various aspects, from functionality to pathogenesis, finally to spectral fingerprint for the design of optical biosensors. In this paper, the secondary structure of SARS-CoV-2 S protein and its constituting components, namely RBD, S1 and S2 regions, are investigated at serological pH by measuring their amide I infrared absorption bands through Attenuated Total Reflection Infrared (ATR-IR) spectroscopy. Experimental data in combination with MultiFOLD predictions, Define Secondary Structure of Proteins (DSSP) web server and Gravy value calculations, provide a comprehensive understanding of RBD, S1, S2, and S proteins in terms of their secondary structure content, conformational order, and interaction with the solvent.

The effect of β-sheet ratio and chain length on all-β proteins was investigated by MD simulations. Protein samples composed of different repeating units with various β-sheet ratios or a different number of repeating units were simulated under a broad temperature range. The simulation results show that the smaller radius of gyration was achieved by the protein with the higher proportion of β-sheet secondary structure, which had the lower nonbonded energy with more HBs within the protein. The root mean square deviation (RMSD) and the root mean square fluctuation (RMSF) both increased with temperature, especially in the case of a longer chain. The visible period was also shown according to the repeated secondary structure. Several minimum values of RMSF were located on the skeleton of Cα atoms participating in the β-sheet, indicating that it is a kind of stable secondary structure. We also concluded that proteins with a short chain or a lower ratio of β-sheet could easily transform their oriented and compact structures to other ones, such as random coils, turns, and even α-helices. These results clarified the relationship from the primary level to the 3D structure of proteins and potentially predicted protein folding.

In recent years, there has been a growing focus on the development of medium-sized drugs based on peptides or nucleic acids owing to their potential therapeutic benefits. As some of these medium-sized drugs exert their therapeutic effects by adopting specific secondary structures, evaluating their conformational states is crucial to ensure the efficacy, quality, and safety of the drug products. It is important to assess the structural integrity of biomolecular therapeutics to guarantee their intended pharmacological activity and maintain the required standards for drug development and manufacturing. One widely utilized technique for quality evaluation is secondary structural analysis using circular dichroism (CD) spectroscopy. Given the higher production and quality control costs associated with medium-sized drugs compared with small-molecule drugs, developing analytical techniques that enable CD analysis with reduced sample volumes is highly desirable. Herein, we focused on a microsampling disk-type cell as a potential solution for reducing the required sample volume. We investigated whether CD spectral analysis using a microsampling disk could provide equivalent spectra compared with the standard cell (sample volume: approx. 300 µL). Our findings demonstrated that the microsampling disk (sample volume: 2-10 µL) could be successfully applied to CD spectral analysis of peptide and nucleic acid drugs, paving the way for more efficient and cost-effective quality evaluation processes.