NEW YORK (GenomeWeb) – Scientists from the University of Cambridge have taken another step towards cell-specific proteomics by mass spectrometry analysis.
Using an unnatural amino acid tagging technology, a team led by first author Thomas Elliott and senior author Jason Chin demonstrated a method of protein enrichment for mass spec-based proteomics.
Advancing upon previous work on a method of protein tagging called stochastic orthogonal recoding of translation (SORT), the researchers were able to enrich and capture labeled proteins, aiding in their identification and quantification by mass spec.
In a proof-of-concept study published last month in Cell Chemical Biology, the team enriched and identified SORT-tagged proteins from Escherichia coli and ovary germ cells in Drosophila melanogaster.
While Chin and colleagues were not available to comment by deadline, David Tirrell, a proteomics researcher at the California Institute of Technology who is also working on cell-specific proteomics but who was not involved in the Cambridge study, told GenomeWeb that while it's early in the development of tagging technologies for cell-specific proteomics, it's "a potentially very important analytical technology, though it's not yet clear whether one will turn out to be more effective than the others."
Single-cell proteomics using mass spec isn't likely to be a reality, because of the lack of sample material; however, because different cells are creating lots of different kinds of proteins and protein profiles, getting more granular information from specific cell types will be an important advancement. "You're no longer just averaging over many cells doing different things at different times," Tirrell said.
Amino acid tagging methods could be the technology that enables cell-specific proteomics. They've been around for about a decade and can allow fluorescent tagging or selective dye labeling. So far, scientists like Chin, Tirrell, Daniela Dieterich of Germany's Otto-von-Guericke University, and others have gotten precise analyses — like proteins made at a certain time point — in "lower" model organisms including bacteria, Caenorhabditis elegans, and fruit flies. "I believe it will turn out to be very general," Tirrell said.
Previously, Chin's team had used SORT to tag proteins for tissue- and developmental stage-specific analyses. "In SORT an unnatural amino acid is recognized by an orthogonal pyrrolysyl-tRNA synthetase, and used to aminoacylate the cognate tRNA bearing a sense-decoding anticodon," the authors explained. "This leads to the substoichiometric incorporation of the unnatural amino acid in response to the targeted sense codons."
The new paper builds on that method, enriching low-abundance proteins for a more complete picture of proteins in a specific cell type within a tissue. The scientists dubbed this method SORT-E, for "enrichment."
"SORT-E does not preferentially identify proteins by molecular weight, but shows a slight bias toward the identification of low-abundance proteins, which should aid their identification, the authors wrote, adding that using the technology at different codons "leads to the enrichment of many proteins with different efficiencies, suggesting that proteome coverage may be increased by performing SORT experiments with several anticodon variants."
After incorporating the unnatural amino acid, the tagged proteins were extracted with a tetrazine diazobenzene biotin compound and captured on streptavidin beads. After washing and elution, analysis was performed with mass spectrometry.
Recent advancements in mass spectrometry instrumentation have been crucial to the development of this kind of analysis, Tirrell said. While Tirrell's lab uses a proteomics core featuring Orbitrap instruments from Thermo Fisher Scientific, Chin's team used LC-MS analysis using the Agilent 6130 Quadrupole instrument.
In the study, Chin's team showed enrichment and identification of labeled proteins, especially low-abundance proteins, differential enrichment of proteins based on tagging at different codons, as well as application in a multi-celluar system, the Drosophila ovary.
"It will be possible to label distinct biomolecules in a single sample with azides or alkynes and cyclopropenes, and selectively and independently enrich and identify each population of labeled biomolecules," the authors wrote. "Given the diversity of biomolecules that can now be labeled with azides and strained alkenes, this opens up many exciting new experimental possibilities."
So far, no team has gotten to the point where it can differentiate, say, one type of neuron from another. Tirrell said his lab has tried to do exactly that in C. elegans, but they haven't been able to turn it into a reliable proteomic analysis.
But as labeling and instrumentation technologies continue to improve, he holds out hope.
"These cell-selective methods will get even better," Tirrell said. "The sensitivity will get better, we'll be able to detect more proteins, [and] we'll need fewer cells."
And there will be no shortage of biological problems that can both drive advancement of cell-specific proteomics and in turn benefit from those enhancements, he predicted.
"It'll be very important in micobiomics. There are thousands of strains of bacteria and we'd like to be able to monitor what's going on in one of them," he said. "Same thing in environmental samples. It's really any kind of a system where you find multiple cell types behaving in multiple ways and don't want to average over that complexity."