NEW YORK – A team led by researchers at the University of Oxford has demonstrated the ability of protein nanopores to locate protein post-translational modifications like phosphorylation and glycosylation on long polypeptides.
Detailed in a study published last month in Nature Nanotechnology, the work points toward an approach that could prove useful for identifying a variety of modified and variant proteins, ultimately allowing researchers to explore the proteoform content of biological samples, said Hagan Bayley, professor of chemical biology at Oxford and senior author on the study.
The effort also reflects growing interest among researchers in nanopore-based protein analysis using electro-osmotic flow instead of enzymatic approaches to both unfold proteins and translocate them through nanopores.
Bayley is one of the founders of Oxford Nanopore Technologies, which uses nanopores for nucleic acid sequencing. In recent years, he and a number of other researchers have turned their attention to nanopore-based protein sequencing, which, if proven feasible, could allow for single-molecule level protein detection as well as the detection of different proteoforms. Ultimately, it might enable de novo sequencing of peptides and proteins.
Nanopore-based protein analysis is a difficult task. While nucleic acids are composed of just four bases, peptides are composed of 20 different amino acids, which makes it challenging to decode the signals produced by a peptide as it moves through a nanopore and assign the signals to individual amino acids.
A related challenge is translocating target peptides through the nanopore in a controlled enough manner to collect good signal on each individual amino acid as it passes through the pore. Researchers have explored a variety of enzymes that can be used to pull peptides through nanopores, but these enzymes typically move the peptides in an inconsistent fashion.
As an alternative approach, Bayley and study coauthors — including Yujia Qing, associate professor of organic chemistry at Oxford and the primary leader of the effort, Bayley said — used electro-osmosis to unfold and translocate proteins non-enzymatically. The process uses nanopores selective for particular ions in a liquid. When an electric potential is applied to the nanopore, it pulls those ions toward it, with the ions carrying molecules of liquid (water, in the case of the Nature Nanotechnology study). This flow of liquid can be used to unfold proteins and move them through the nanopore.
A number of labs including Bayley's have been exploring the approach as it relates to nanopore-based protein analysis. In January, a team led by Northeastern University researcher Meni Wanunu published a study in Nature Biotechnology using the method to move proteins through nanopores. In a BioRxiv preprint published in February, researchers led by Giovanni Maglia, professor at the University of Groningen and founder of nanopore firm Portal Biotech, detailed the use of electro-osmosis to move polypeptides through a nanopore.
In the Nature Nanotechnology study, the Oxford team used its approach to translocate and analyze proteins more than 1,200 amino acid sequences long, which Bayley noted is much larger than the average protein.
"One of the points of the paper was to show that we could take complete proteins of hundreds of amino acids and get them through the nanopore, whereas many other groups have been working with [smaller] peptides," he said.
The researchers also demonstrated that they could detect post-translational modifications as they translocated full-length proteins through the nanopore, picking up phosphorylation, glycosylation, and glutathionylation.
Bayley suggested the work points toward the use of nanopores to distinguish between differentially modified proteins and other variants allowing researchers to characterize the proteoform landscape of biological samples.
"It's nice work," said Cees Dekker, a professor at Delft University of Technology whose research includes nanopore-based protein analysis. He highlighted the ability of the method to translocate and analyze very long peptide sequences as well as its detection of protein PTMs.
Dekker noted that challenges involved in various enzymatic methods to translocating peptides through nanopores has led researchers like Bayley and others to explore non-enzymatic approaches like electro-osmotic flow to push target peptides through nanopores for analysis.
He said, though, that it remains to be seen how effective these non-enzymatic approaches will be, particularly for de novo sequencing applications. Dekker has explored both enzymatic and non-enzymatic methods for nanopore protein analysis.
“It would be wonderful if you can get the same precision without using an enzyme, but based on current data, I doubt that a bit," he said.
Dekker also noted that the results in the paper don't demonstrate a significant advance in the ability to do de novo protein sequencing, which, he said, remains "the Holy Grail" of nanopore protein analysis.
Bayley said that he believed the ability to sequence proteins de novo was less important than the ability to identify the presence of a modified or variant protein and suggested that protein fingerprinting approaches would be sufficient for many applications.
"With a few exceptions, the polypeptide sequence is encoded in the genomic DNA," he said. "We're assuming that we know the sequence, which is pretty reasonable … and then we're looking for alterations in that sequence."
"There are 20 different amino acids, some of them are quite similar to each other, and we could spend a lot of time trying to do [de novo sequencing], but in my view it's not really worth it," he said. "We want to find those modifications and find out where they are in full-length polypeptides."
Bayley said he believes that initially such approaches might look at populations of specific proteins isolated by, for instance, antibody pull-down, and then analyze the different proteoforms in that population. Longer term, though, he said he believes that nanopore-based analysis would produce signal that, while perhaps not sufficient for de novo sequencing, could be used to identify both specific protein groups and then the presence of different proteoforms within those groups.
Dekker said in the absence of a true de novo sequencing technology, fingerprinting approaches like that being pursued by Bayley's lab have potential.
"Fingerprinting approaches are very useful for certain applications, for diagnostics, for quantitative analysis of samples and things like that," he said.
Bayley said that while his lab receives funding from Oxford Nanopore, the company was not involved in this work. In the paper, he and his coauthors wrote that the approach "will be readily transferrable to nanopore sequencing devices" like Oxford Nanopore's MinIon platform.
First, though, the researchers will need to demonstrate the method in actual biological samples, Bayley said. The recently published work was done using model proteins to which the researchers added specific modifications. The study authors noted that as they move to biological samples, they will likely face a number of challenges including the difficulty of establishing "universal conditions for the translocation of all the protein components of a cell" and the possibility that smaller PTMs may not be directed by the nanopore.
"There's a lot more to be done," Bayley said.