Amyloid proteins are inherently disordered, adopting shape after shape as though they were dancers anxious to satisfy a mad choreographer. Seen this way, amyloid proteins might seem pitiable. But amyloid proteins are best kept in a frenzy, suggest researchers based at Cambridge University. These researchers, led by Michele Vendruscolo, PhD, have found that amyloid proteins may be less likely to aggregate and form Alzheimer’s plaques if they are subject to a choreography that keeps them maximally disordered.

“We are used to thinking of proteins as molecules that fold into well-defined structures: finding out how this process happens has been a major research focus over the last 50 years,” said Vendruscolo, co-director of Cambridge’s Centre for Misfolding Diseases. “However, about a third of the proteins in our body do not fold, and instead remain in disordered shapes.”

Vendruscolo’s team, together with scientists from the University of Milan and Google Research, have used machine learning techniques to predict how proteins, particularly those implicated in neurological diseases, completely change their shapes in a matter of microseconds. The researchers focused their attention on the amyloid beta peptide, a protein fragment associated with Alzheimer’s disease, which aggregates to form amyloid plaques in the brains of affected individuals.

They found that amyloid beta hops between widely different states millions of times per second without ever stopping in any particular state. One of the states, an exceptionally disordered state, happens to be a hub state. Other states, relatively structured states, transition to the hub state quickly. Transitions the other way, however, from the hub state to relatively structured states, occur more slowly.

To capture this transitional choreography, the scientists used a molecular dynamics simulation and a new analysis method based on deep learning. Details of the work appeared January 14 in the journal Nature Computational Science, in an article titled, “A kinetic ensemble of the Alzheimer’s Aβ peptide.”

Kinetic ensembles contain information about the molecular structures of proteins, the populations of their metastable states, and the transition rates between these different metastable states.

“We develop a Markov state model and apply it to determine a kinetic ensemble of Aβ42, a disordered peptide associated with Alzheimer’s disease,” the article’s authors wrote. “Through the Google Compute Engine, we generated 315-µs all-atom molecular dynamics trajectories. Using a probabilistic-based definition of conformational states in a neural network approach, we found that Aβ42 is characterized by inter-state transitions on the microsecond timescale, exhibiting only fully unfolded or short-lived, partially folded states.”

Essentially, the kinetic ensemble approach developed by the researchers harnesses the power of Google’s computer network to generate large numbers of short trajectories. The most common motions show up multiple times in these “movies,” making it possible to define the frequencies by which disordered proteins jumps between different shapes.

“By counting these motions, we can predict which states the protein occupies and how quickly it transitions between them,” said first author Thomas Löhr, a graduate student from Cambridge’s Yusuf Hamied department of chemistry.

The scientists also studied a variant of amyloid beta in which one of the amino acids was modified by oxidation. This variant showed a decreased tendency to aggregate. It also changed shape even faster than its unmodified counterpart. With this variant, the population of the completely disordered state was larger (85%) than that of the wild type, and the transition rates from this state to the minor structured states were much slower. These differences may explain the variant’s low aggregation propensity.

“From a chemical perspective, this modification is a minor change,” said Löhr. “But the effects on the states and transitions between them are drastic.”

The shape-shifting behavior of amyloid beta is the main reason amyloid beta has been deemed “undruggable,” Vendruscolo noted. “The constant motion of amyloid beta is one of the reasons it’s been so difficult to target,” he continued. “It’s almost like trying to catch smoke in your hands.”

Nevertheless, a better understanding of amyloid beta’s shape-shifting behavior may lead to new therapeutic strategies. One such strategy was suggested by Vendruscolo: “By making disordered proteins even more disordered, we can prevent them from self-associating in aberrant manners.”

The scientists concluded that their results illustrate how kinetic ensembles provide effective information about the structure, thermodynamics, and kinetics of disordered proteins. They asserted that their approach provides a powerful tool to investigate a class of proteins with fast and disordered motions. To date, such proteins have remained mysterious despite their importance in biology and medicine, mostly because traditional methods tend to address the problem of determining static structures, not structures in motion.