Identification of individual proteins with nanopores and supercomputers

Aleksei Aksimentiev, professor of physics at the University of Illinois, and PhD student Jingqian Liu, co-authors of the most recent study on protein identification. (Image credit: TACC)

The main author Henry Brinkerhoff, who promoted this work as a postdoc in the laboratory of the physicist Cees Dekker, compares the protein to a necklace with pearls of different sizes. “Imagine turning on the faucet while slowly moving the necklace down the drain, in this case the nanopore,” he said. “When a large pearl clogs the drain, the water just flows in a trickle; if you have smaller pearls in the chain directly at the drain, more water can flow through. “

With their technology, the researchers can measure the level of the ion current very precisely – but not exactly because the step-by-step passage through the pore is irregular. However, by loading the liquid medium with helicases, researchers can get many separate, overlapping reads of the same molecule, or they can “rewind” the protein and reread its amino acid sequence. This reduced the errors from 13% to practically zero.

Their approach enabled the researchers to distinguish variants of peptides that differed only by a single amino acid – which they proved by developing synthetic peptides with only one altered amino acid and showing that the system could distinguish between them.

However, in order to read out the individual amino acids, they first had to know which signal each one generates on its journey through the pore. Some of these signals could be counter-intuitive, the researchers found.

For example, when the bulky tryptophan amino acid moved through the constriction, the ion current initially decreased and then increased, absurdly, relative to the small and medium-sized variants.

To understand the origin of these patterns, the team relied on supercomputer simulations by computational biologist Aleksei Aksimentiev (UIUC) performed on several of the fastest supercomputers available to academic researchers worldwide: Frontera at the Texas Advanced Computing Center; Blue Waters, at the National Center for Supercomputing Applications; and Expanse at the San Diego Supercomputer Center.

Aksimentiev’s team used a method called molecular dynamics simulation to simulate the behavior of the nanopore, proteins and the surrounding medium with atomic resolution. Such simulations cannot fully capture the true timescale of nanopore activity, which extends to seconds. However, by generating 40 to 50 initial states at different positions and then running 70 simulations in parallel, the team was able to derive statistics for various confirmations of peptides. From this they calculated the current and compared it with experiments. This computer work was directed by Jingqian Liu, a PhD student in biophysics in Aksimentiev’s laboratory.

The simulations comprised 30,000 atoms that interacted over 200 to 500 nanoseconds and were able to agree with experimental results. More importantly, they showed why certain amino acids generate conflicting signals as they pass through the nanopore. In the tryptophan variant, the signal could be traced back to binding of the peptide side chain to the nanopore surface above the constriction.

“For each specific conformation, we could see what happened to the side chain, whether it interacted with the surface or remained in the pore,” said Aksimentiev, professor of physics at UIUC. “Then we were able to determine directly that the connection to the sidechain increases the current.”

It took weeks to generate the simulations on Frontera, currently the tenth fastest supercomputer in the world and the most powerful of all universities. But with the kind of computer clusters available at most universities, it would have taken them years. The research to identify individual proteins – for which a global race for success is taking place – was published online by Science on November 4, 2021 as a “First Release”. The research was sponsored by the Dutch Research Council, the US National Institutes of Health and. supports US National Science Foundation, among others.

“There are tremendous opportunities to develop diagnostics by reading individual proteins with this nanopore approach,” said Aksimentiev. “Computation will play a huge role in the development of these technologies. It is amazing that we can reproduce experiments with computer models and recognize which interactions are taking place on the nanoscale. “

In addition, computer models offer another modality to design, allowing researchers to test nanopores of different sizes or with strategically placed residues that can generate enhanced signals.

More work is required to read more than 20 amino acids and identify heterogeneously charged amino acids, but Aksimentiev dares to develop a working model in three to five years.

“We believe that with our new approach we can recognize post-translational changes,” says Dekker, “and thus shed light on the proteins that we carry around with us.”

Source: Aaron Dubrow, Texas Advanced Computing Center


Comments are closed.