Researchers in the Netherlands have used an enzyme found in a Japanese mushroom to develop a method they say can make it significantly easier to identify peptides whose genome is unknown, and help decipher post-translational modifications.
The method, described in a proof-of-principle study in the current issue of Nature Methods, was developed by proteomics researchers at Utrecht University and employs an enzyme called metalloendopeptidase found in the Japanese mushroom Grifola frondosa. The mushroom, also called Maitake or dancing mushroom, cleaves proteins on the amino side of lysine, Lys-N.
Using electron transfer dissociation, the researchers found that peptides containing a single lysine residue that were digested by Lys-N produced spectra dominated by c-type fragment ions, “provid[e] simple ladders for sequence determination,” they write in the article. They add that the method can be “a valuable strategy for de novo sequencing and the analysis of post-translation modifications.”
For proteomics, such a strategy, if replicated, could become a valuable tool because the genomes of most species remain to be sequenced and de novo protein sequencing can be a painstaking and imperfect task.
Lys-N was originally isolated by Japanese researchers whose work on a model protein predicted that the protease would cleave at the Lys-N terminus of peptides. Picking up on that, the Dutch team decided to test Lys-N and its applicability for proteomics work, Albert Heck, the corresponding author on the article, and a professor at Utrecht University in the department of biomolecular mass spectrometry, told ProteoMonitor this week.
Normally when a peptide is fragmented, an amino acid sequence of the peptide is generated and the protein is partially sequenced. When done in a mass spectrometer, the sequencing is performed from both the N- and C-termini of the peptide, however. This makes the spectra complicated because a researcher does not know beforehand which are C-terminus fragments and which are N-terminus fragments.
“When we started to explore this new fragmentation technique [ETD] … we had the hypothesis that if we can really put the charge of the peptide on one side of the peptide, that that might drive the fragmentation that we would observe, and that brought us to the idea that if we generate peptides where the charge is really localized, in this case the N-terminus, we still would form the same fragments, but the charge would always remain on the N-terminus, not on the C-terminus,” Heck said. “And therefore we hypothesized that we would only see one sort of fragment ion, and that makes it much simpler to read the sequence of the peptide.”
“We actually found out that in other species, there are similar proteases, so we want to explore if we can find Lys-N proteases in nature.”
Because Lys-N cleaves proteins on the N-terminus side of a lysine, it generates peptides that have a lysine on the N-terminus, which also has an amine. The combination of the amine and lysine “that has a free amide on the side chain makes these peptides very favorable,” Heck said. “Their charge will be very localized on that side of the peptide.”
In contrast, using Lys-C — an endoprotease that cleaves proteins on the carboxy side of the lysine — creates a free amide, which wants to be positively charged, on the N-terminus, and a lysine, which also wants to be positively charged, on the C-terminus.
This, said Heck, is “exactly what you don’t want because you have a charge on both sides of the peptide, and therefore you get fragments from both sides of the peptide.”
With their method, only c-ions, and not z-ions, can be formed, reducing the number of ions formed by half. More importantly, the “space” in which a researcher would need to search to identify the fragment ions is reduced by a “factor which is way higher, and therefore your assignments become very easy,” Heck said.
“Someone who just knows what the masses are, [or the] amino acids, can read off the sequence from there,” he said. “You can just read it directly off the peptide peaks that are formed.”
In work testing their strategy for in-gel digestion, Heck and his colleagues also found that their Lys-N approach to be “at least as efficient” as trypsin for in-gel proteolysis of bovine serum albumin, they said in the article.
The team’s approach can also be used to analyze post-translation modifications, Heck said, because the strategy makes it easier to read the protein sequence.
“With this method, the reading off becomes so much easier that I think it would make a difference,” Heck said. “This method does not make it an easy [analysis] but at least it helps.”
The caveat to their strategy is that because it cuts at a lysine, it may not work on modified lysines, an issue that the researchers are continuing to study. The strategy also works for about only half of the peptides that may be generated in an experiment — those that contain a lysine but no arginine.
But, Heck said, in proteomics “you always know that you don’t analyze everything. You can turn it around and say ‘I’ll do a targeted approach on these peptides that contain only a single lysine.’” He and his co-researchers are also exploring ways to work around this drawback, he said.
In continuing research, they are testing their method to sequence proteins in ostrich and avocado, both of which have not had their genome sequenced, in order to show that it can work just as well as other, more established proteomics methods for sequenced genomes.
Because the team is preparing a paper on the next phase of its work, Heck declined to provide many details or data, but said that, to date, it had identified more than 1,000 proteins in ostrich. Many of the proteins look like chicken proteins, “but then they have certain amino acid replacements that are different.”
In the avocado, he said fewer than 1,000 proteins have been identified, and overall the fruit does not have “that many proteins.”
Other researchers have also approached them to sequence the zebrafish as well as organisms whose genomes have not been sequenced, including the goldfish. They are also looking at sources for Lys-N other than Grifola frondosa, Heck said.
“We actually found out that in other species, there are similar proteases, so we want to explore if we can find Lys-N proteases in nature,” he said.