This story originally ran on Oct. 22.
Researchers in Germany and Belgium have reported the most precise protein determination of mitochondria to date, identifying a protease that stabilizes mitochondrial proteins, and in the process solving a mystery that has long baffled protein researchers.
Their results are publishedin the Oct. 16 edition of Cell. In the paper, the researchers report the proteome of N-termini in yeast mitochondria, a collection of 615 proteins, and the identification of the enzyme Icp55 as an until-now unknown processing peptidase that serves to stabilize some mitochondrial proteins.
While prior research by other scientists has yielded a collection of 851 mitochondrial proteins in yeast, the N-proteome of 615 proteins described in the Cell study represents "the first systematic resource for the characterization of authentic mitochondrial presequences and their processing, and thus provides an important foundation for the analysis of mitochondrial protein biogenesis and turnover," the authors wrote.
Once thought of strictly as the energy sources of cells, mitochondria is now known to play an important role in the onset and progression of numerous diseases.
"Very simple defects in mitochondrial proteins or mitochondrial function can lead to many, many different diseases," from cancer and diabetes to neurodegenerative disease, Chris Meisinger, a professor of biochemistry at the Institute for Biochemistry and Molecular Biology at the University of Freiburg, and the project leader for the Cell study, told ProteoMonitor this week.
"All these diseases are diseases where mitochondria dysfunction plays a very central role. And, of course, mitochondria can only work properly if all the proteins involved in this function are present and are not degraded. So they have to be there in a stable way. Only then can they fulfill this function."
For many cell organelles, such as mitochondria, proteins are synthesized with signal sequences "that are proteolytically removed by specific enzymes," according to the study. The majority of organellar proteins have not had their N-termini determined, though and what has been observed of mitochondrial proteins has been done so on a small subset of preproteins.
"Since the exact N-termini of the mature proteins, and thus the processing sites, have only been determined for a minor fraction of the mitochondrial proteome, we have only limited information on preprotein processing and its functional relevance," the researchers said.
Meisinger noted that the majority of mitochondrial proteins are encoded in the nucleus of genes, but the mitochondrial genome encodes between only eight and 13 genes for stable expressed mitochondrial proteins. That means that most of the 1,000 mitochondrial proteins have to be imported after they are synthesized in cytosolic ribosomes.
During this process, the signal sequences targeting the proteins to the mitochondria are cleaved, "and we've known for more than 20 years [of] a very special protease which cleaves these signal sequences. Such a processing step occurs for about 70 percent of the mitochondrial proteins, where such a signal sequence is cleaved after the import," Meisinger said.
With the elucidation of the N-proteome, "we have now systematically determined the cleavage sites for all these mitochondrial proteins, and we found that indeed 70 percent of the mitochondrial proteins are cleaved, and 30 percent are not cleaved," he added.
According to co-author Albert Sickmann who oversaw the proteomics analysis part of the study, the identification of the cleavage sites of the processing peptidase is important to cell biologists and the study of mitochondrial-associated diseases.
Sickmann is a professor for applied proteomics at the medical faculty at Ruhr-Universität Bochum and director of the department of bioanalytics at the Institute for Analytical Sciences.
"If you transport these proteins through biological membranes, you cleave off a part of it and afterward you are in the mature part of the protein," he said. "If you have a disease and this is somehow changed and you have an alternative cleavage site, maybe there is a mutation. Or if you cannot import this protein something is going wrong with the protein folding. Of course, there is some implication in disease.
[ pagebreak ]
"Maybe if you have a lack of this processing peptidase, you will have problems with the tertiary and quaternary structures of proteins within mitochondrial metrics," he added.
To perform their analysis, the researchers used a fractionation technology known as combined fractional diagonal chromatography, or COFRADIC, developed by Joël Vandekerckhove and Kris Gevaert, a co-author on the Cell paper. The technology consists of two identical chromatographic separations with a step targeted to a subset of peptides between the two separations.
The researchers started with a mixture of mitochondrial proteins, of which all three amino acids were modified by acetylation. The mixture was digested resulting in two types of peptides, one with N-terminal acetylation — "so this corresponds to the real N-terminal peptides of the original peptides," Meisinger said — and internal peptides that now have a free amino group.
The internal peptides were then modified with a hydrophobic linker. "Then you can separate [the proteins] due to this increased hydrophobicity of the internal peptides from the original N-terminal peptide, which cannot be modified because they have this acetylation group at the N-terminus," Meisinger said. "And then you can sequence these N-terminal peptides, which are then separated from the internal peptides."
After analysis by LC-MS/MS, the N-peptides were compared with the most comprehensive proteomic analysis of yeast mitochondria, the PROMITO database, containing 851 proteins. That led to the N-proteome, the N-termini of 615 proteins different proteins.
According to Sickmann, the method is also easily adoptable for other proteomics-directed work.
"For example, if you do the expression of proteins and you want to do something like X-ray crystallography, what you always need to know is the N-terminus of the protein — if it's an intact N-terminus or if you have an intact C-terminus part of the protein, or if you have any degradation during the purification," he said. "With this method it's quite easy to determine all these aspects of the protein."
During their global analysis, the researchers also came across an unexpected observation: "The most surprising thing in this study was that for many proteins, we not only observed one amino terminus for this protein, but a second one," Meisinger said. "And for a special class of mitochondrial proteins, we observed a second processing site just one amino acid [away]."
What they saw were intermediates that had an additional amino acid at the N-terminus, which suggested that there was a second processing enzyme that cleaved the additional amino acid after the first processing step.
That new processing peptidase was Icp55, or intermediate cleaving peptidase of 55 kDa.
Prior studies had identified the existence of Icp55, but virtually nothing was known about its function, Meisinger said. Icp44 "belongs to a special protease class … called amino peptidase P and this P stands for proline. And it was assumed that all the members of this class of proteases cleaves in front of a proline," he said. "But according to our study, this can't be the case, so this was maybe just a misassignment to this class. … Nobody would have ever expected that we have such a system — a protease system that leads to a stabilization of proteins."
Indeed, rather than searching for systems of stabilization, researchers have been hunting for systems responsible for protein degradation.
"The question was: Why is there an enzyme that cleaves off this additional amino acid? What sense does this make?'" he said. They compared the amino acids that were cleaved in the substrate proteins "and to our surprise, all these amino acids, which are additionally cleaved, or which have to be additionally cleaved, are destabilizing amino acids, according to the Varshavsky N-end rule," which relates the in vivo half-life of a protein to the identity of its N-terminal residue.
According to the rule, if the N-terminus of a protein contains a destabilizing amino acid, the protein degrades more quickly than if it has a stabilizing amino acid at the N-terminus. "And this new enzyme that we identified cleaves off the destabilizing amino acid and therefore stabilizes the protein," Meisinger said.
It is not known, though, what role, if any, Icp55 may have in disease because so much remains unknown about it. While it may be tempting to speculate that some dysfunction of Icp55 in humans may result in disease, "our model here is yeast, so we don't know if we would also have substrate proteins of this novel protease involved in disease-related functions in human mitochondria," Meisinger said.
But now that a function can be assigned to the enzyme, "it's possible that very soon people maybe will identify mutations and families" for it, he added.
He declined to speculate whether other proteases in mitochondria may have a similar function to Icp55. But he said that "there must be a destabilizing protease because we have turnover of mitochondrial proteins. …This is something that nobody doubts."
He and his colleagues are searching for such a protease but have identified no candidates yet.
"We know now which substrates we can use because … from our list we can just choose substrates that have destabilizing amino acids in the N-terminus and use them as a test substrate, and then analyze various mutants of mitochondrial proteases … and then see if there is some mutant of this substrate that does not get degraded," he said. "And by this way, then identify this degradation protease system."