NEW YORK – Researchers from French startup DreamPore along with several academic collaborators have developed a nanopore-based system capable of identifying 15 of the 20 proteinogenic amino acids.
The findings, presented in a study published last week in Nature Biotechnology, are a step in the ongoing effort to develop nanopore-based platforms for protein analysis.
According to DreamPore CEO Luc Lenglet, the company is currently working on a system based on the nanopore technology presented in the study for the detection of a biomarker linked to brain cancer recurrence.
Nanopores are currently used for nucleic acid sequencing, most notably in Oxford Nanopore's various sequencing platforms. Researchers, including some affiliated with Oxford Nanopore, have also been investigating the use of nanopores for protein analysis, either by functionalizing nanopores with affinity molecules like aptamers or antibodies or by translocating proteins or peptides through a nanopore and reading the amino acid sequence as the molecule passes through the pore. Binding of molecules near the pore opening or passage of molecules through the pore cause changes in the electric current across the nanopore, and the molecules can be detected and characterized based on these changes.
In the Nature Biotechnology study, the researchers used an aerolysin nanopore for detection of individual amino acids. They attached the amino acids to polycationic carrier molecules, which ensured that they stayed within the nanopore long enough to collect sufficient information about the change in current. Using a wild-type nanopore, they found that they could distinguish between 13 of the 20 amino acids found in proteins. Using chemical modification of the amino acids they were able to bring that number up to 15 out of 20. While the remaining five registered current changes within the nanopore, these changes were too similar to be clearly distinguished between.
Aleksei Aksimentiev, professor of biological physics at the University of Illinois and one of the senior authors on the paper, said that there were a number of approaches the researchers were exploring to improve on these initial results, including modifying the pore to increase the dwell time of amino acids as well as using different chemical modifications of the target amino acids.
Abdelghani Oukhaled, a researcher at Cergy-Pontoise University and a senior author on the study, noted that reducing the noise of the current change measurements might also provide enough resolution to distinguish between the full set of 20 amino acids.
Oukhaled said he believed that while in the Nature Biotechnology study the researchers did not use the nanopore to distinguish all 20 amino acids, the findings indicate that it is possible.
"I think this gives us confidence," he said. "I would say now that it is possible to sequence proteins."
Oukhaled added that in addition to optimizing the system to identify all 20 amino acids, the next step would be to develop a method to drive the individual amino acids into the pore and keep them there for a sufficient length of time without requiring the polycationic carrier molecules used in the recent study.
To that end, the researchers are exploring an approach that uses water for this purpose, said Juan Pelta, Oukhaled's colleague at Cergy-Pontoise and co-author on the study. Pelta is also a co-founder of DreamPore and a member of its scientific advisory board.
"The idea is to use an enzyme to cleave the peptide [into individual amino acids] and to force the entry of the amino acids into the sensor," he said. "We need a driving force, and our idea is to use just the flow of water."
Pelta said he and his colleagues have been working on nanopore-based protein detection for around 10 years. He said that the aerolysin pore had a number of advantages, including its high stability and the ability, given that it is a protein, to modify it genetically to change its characteristics.
"We can change the pore diameter, we can change the charge of the channel, and that means we can change the interaction between the peptide and the channel," he said.
Beyond working to detect all 20 amino acids with the nanopore, the researchers are also using it to detect modified peptides.
"Being able to read just the amino acids is not sufficient to obtain the sequence of a protein," Pelta said, noting that it will also be important to detect post-translational modifications. He said he and his colleagues had been able to detect various modifications with the pore, including protein phosphorylation.
In the paper, the authors presented a vision of how the technique might be applied to protein sequencing, using chemical or enzymatic methods to sequentially cleave terminal amino acids from copies of a target protein and then direct them into the nanopore. The process, they noted, could be done in parallel across large numbers of "electrically isolated volumes."
Long term, Lenglet said DreamPore hopes to develop the technology for large-scale sequencing of proteins and peptides. In the short term, the company is focusing on developing the platform for targeted detection of a biomarker for brain cancer recurrence. The company has licensed intellectual property covering the nanopore protein sequencing approach from Cergy-Pontoise.
The target is a protein found in circulating tumor cells, and Lenglet said the company believes the nanopore platform could provide the level of sensitivity required to detect this analyte at very low levels.
Lenglet said DreamPore is working on this project in collaboration with Lariboisière Hospital in Paris, where Philippe Manivet, the company's other co-founder and an author on the Nature Biotechnology study, is a researcher.
He said the hospital connection has allowed the company to identify some clinical needs the nanopore technology might help address.
"We don't want to stay in pure science," he said. "We really want to go into the healthcare business."
He added that the company expects to take the next several years proving out the technical validity of the platform and clinical utility of the brain cancer marker with the ultimate goal of packaging it as a CE-IVD marked kit.
DreamPore currently has four employees and is in the process of doubling that headcount to eight, Lenglet said. The company has raised €600,000 ($664,000) to date.
The DreamPore group and its collaborators are one of a number of teams working on nanopore-based protein detection. On the commercial side, Oxford Nanopore is perhaps the most prominent example, though the company has not publicized its efforts in this area beyond noting that it is interested in expanding its technology to protein work.
One the academic side, researchers are pursuing a variety of approaches using nanopores for protein analysis. This month, a team led by scientists at Aarhus University published a study in Nature Communications using DNA to create nanopores that could be used for biosensing applications including protein detection.
Oxford Nanopore founder Hagen Bayley has been working independently of the company on nanopore applications including distinguishing between differentially phosphorylated protein forms.
Last year, Bayley and colleagues demonstrated the use of DNA nanostructure scaffolds that he said could enable construction of larger, more uniform nanopores for applications including DNA and protein detection.
In 2017, researchers at the University of Groningen demonstrated the ability of Fragaceatoxin C (FraC) nanopores to identify peptide and protein biomarkers in simple mixtures and to distinguish between polypeptides differing by as little as a single amino acid.
The same year, a team at the University of California, San Diego published a machine learning-based approach to nanopore protein identification that they said indicated large-scale proteomic profiling via nanopores could be possible.