Biology subjects

Makhatadze, G. I.

Publications and source records attributed to Makhatadze, G. I..

4 recordsLinked to original sources

bioRxiv2026

Do AI Models for Protein Structure Prediction Get Electrostatics Right?

A variant of the U1A protein containing four substitutions to ionizable residues was generated serendipitously due to a miscommunication. Biophysical measurements show that this variant has at least twice as much helical structure as the wild-type U1A and is trimeric in solution, in contrast to the monomeric wild type. In sharp contrast, structures predicted by deep-learning AI tools (AlphaFold2 and RoseTTAFold2) and transformer-based tools (OmegaFold and ESMFold) are all highly similar to the wild-type U1A (backbone RMSD < 1 [A]). Even more surprising, two of the substituted ionizable residues are predicted to be fully buried in the non-polar core of the protein, an outcome that contradicts well-established physico-chemical principles, as ionizable residues are normally located on the protein surface. To explore this effect further, we generated sequences containing up to all twelve residues that make up the non-polar core of U1A. Across thousands of sequences, and depending on the AI model used, the majority of predicted structures contained fully buried ionizable residues while still maintaining the overall U1A fold. We then examined two additional proteins of comparable size, acylphosphatase and the de novo-designed TOP7 fold, and observed the same phenomenon: AI models frequently predicted structures with buried ionizable residues that nevertheless retained the parent fold. When these AI-predicted structures were subjected to short (50 ns) molecular dynamics simulations using physics-based force fields such as CHARMM or AMBER, the structures rapidly relaxed into ensembles that exposed ionizable residues. We conclude that while AI-based structure prediction tools perform extremely well on naturally occurring sequences, they do not reliably encode the physico-chemical principles governing the placement of ionizable residues. A straightforward remedy is to include a brief molecular dynamics simulation as a final validation step for AI-generated structures.

biophysics↗

bioRxiv2025

The Effects of Pressure and Temperature on the Thermodynamics of α-Helices.

In light of the recent realization that a large fraction of microbial biomass lives under high hydrostatic pressure, there is a renewed interest in understanding molecular details by which proteins in these organisms modulate their functional native state. The effects of pressure on protein stability are defined by the volume changes between native and denatured states. The conformational ensemble of the denatured state can depend on several extrinsic variables, such as pH and ionic strength of solvent, temperature, and pressure. The effect of the latter on the elements of the secondary structures, and -helical structures has been inconclusive. This has been largely due to the inherent difficulties of high-pressure experiments. Here, we adapted the method of choice, circular dichroism spectroscopy, on a well-established series of model peptides to study helical structure formation, while focusing on the pressure and temperature dependencies of the helix-coil transition. We find that at low temperatures, pressure stabilizes the helical structure, suggesting that the volume of the helix-coil transition is positive. However, at higher temperatures (>40{degrees}C), the volume changes become negative, and pressure destabilizes the helical structure.

biophysics↗

bioRxiv2024

Quantitative Detection of Amyloid Fibrils using Fluorescence Resonance Energy Transfer (FRET) Between Engineered Yellow and Cyan Proteins

Over 20 human diseases are caused by or associated with amyloid formation. Developing diagnostic tools to understand the process of amyloid fibril formation is essential for creating therapeutic agents to combat these widespread and growing health problems. Here, we capitalize on our recent striking discovery that green fluorescent protein (GFP), one of the most used proteins in molecular and cell biology, has a high intrinsic binding affinity to various structural intermediates along the fibrillation pathway, independent of amyloid sequence. Using engineered GFP with the fluorescence properties of Aquamarine and mCitrine, we developed a FRET-based sensor to quantitatively monitor amyloid fibrils. The proof-of-principle characterization was performed on a test system consisting of PAPf39 fibrils.

biophysics↗

bioRxiv2024

Modulation of Electrostatic Interactions as a Mechanism of Cryptic Adaptation of Colwellia to High Hydrostatic Pressure

The role of various interactions in determining the pressure adaptation of the proteome in piezophilic organisms remains to be established. It is clear that the adaptation is not limited to one or two proteins, but has a more general evolution of the characteristics of the entire proteome, the so-called cryptic evolution. Using the synergy between bioinformatics, computer simulations, and some experimental evidence, we probed the physico-chemical mechanisms of cryptic evolution of the proteome of psychrophilic strains of model organism, Colwellia, to adapt to life at various pressures, from the surface of the Arctic ice to the depth of the Mariana Trench. From the bioinformatics analysis of proteomes of several strains of Colwellia, we have identified the modulation of interactions between charged residues as a possible driver of evolutionary adaptation to high hydrostatic pressure. The computational modeling suggests that these interactions have different roles in modulating the function-stability relationship for different protein families. For several classes of proteins, the modulation of interactions between charges evolved to lead to an increase in stability with pressure, while for others, just the opposite is observed. The latter trend appears to benefit enzyme activity by countering structural rigidification due to the high pressure.

biophysics↗