ICAT and its variations have become the gold standard for comparing global protein expression in many proteomics labs. But they may soon be rivaled by another strategy, recently published by Steven Gygi’s group at Harvard. Termed “Aqua” for absolute quantification, it promises to allow users to measure a small number of proteins — or their modification states — in large numbers of biological samples. Aqua assays are slated to be commercialized soon by Cell Signaling Technology, a company based in Beverly, Mass.
Last week, Gygi’s group published a description of Aqua, including a number of applications, online in PNAS (Gerber et al., www.pnas.org/cgi/doi/10.1073/pnas.0832254100). Their method is based on an old principle: stable isotope dilution — the use of a labeled molecule as an internal standard. This approach has been widely used in the analysis of small molecules by mass spectrometry for several decades, but has only recently found entry into protein and peptide analysis in biological samples, according to Scott Gerber, a postdoctoral research fellow who developed the technique in Gygi’s lab. “We have taken pieces of existing technology and have put them together [for] protein identification,” he said.
However, others — including Steven Carr at Millennium Pharmaceuticals — have used similar strategies but have not published details of them, according to several researchers in the field. “[The method] is not new, but this is a very nice demonstration of its applications,” said Scott Patterson, CSO of Farmal Biomedicines, who has used a similar method himself.
“Approaches such as the Aqua strategy will become increasingly important as the focus of proteomics shifts from identifying proteins to understanding their functions in detail,” said David Arnott, a mass spectrometrist at Genentech.
Gerber and his colleagues synthesized isotopically labeled peptide standards that correspond to peptides from a digest of specific proteins of interest. After assessing the fragmentation spectra of these standards and selecting a suitable fragment ion for each of them to monitor, they spiked a known amount of the standards into biological samples prior to their trypsin digestion. They then performed an LC-MS/MS experiment using selected-reaction monitoring (SRM) to measure the fragment ion from both the native peptide and the labeled peptide. Knowing the amount of the standard, they were able to calculate the absolute amount of the native peptide and — using certain assumptions — that of the corresponding protein in the biological sample, down to 60 copies per cell.
In the paper, the researchers applied their method in three different ways: to quantify two low-abundance proteins in yeast; to study phosphorylation levels at a specific site of a human protein in a lysate as a function of the cell cycle; and to monitor phosphorylation of a protein at a specific site by different kinases in an in vitro kinase assay. Meanwhile, Gerber said, they have also used the method to quantify protein ubiquitination, methylation, and acetylation.
Aqua differs from ICAT in several ways: While ICAT compares the relative amounts of proteins in a test and reference sample, Aqua determines absolute amounts of proteins in a single sample using internal peptide standards. ICAT allows users to quantify hundreds or thousands of proteins simultaneously, making it suitable for discovery-type experiments, whereas Aqua zooms in on a few specific proteins of interest, or their modification states. Unlike ICAT, Aqua does not require the proteins in the sample to be chemically modified; however it does require specific internal protein standards to be prepared prior to the experiment. Finally, MS/MS spectra in ICAT are used to identify peptides — in Aqua, the identities of the peptides of interest are already known, and the MS/MS spectra only serve for the purpose of quantification.
In terms of applications, Aqua competes with other quantitative assays, including immunoassays that rely on the availability of specific antibodies. While these can be difficult to obtain, “Once you have the isotope-labeled peptide in hand, you can interrogate a large number of samples,” Patterson said.
But the more significant use of the technology might be in monitoring posttranslational modifications. “That’s something that changes rapidly in a cell from one time point to another, and a technique like this can be used to follow phosphorylation and dephosphorylation in signal transduction pathways,” said Arnott.
Harvard University owns a patent application to the Aqua technology that covers both the method — quantification based on stable-isotope labeled peptides using tandem mass spectrometry — and the sale of kits of posttranslationally modified internal standard peptides for monitoring modification changes, according to Gerber.
Cell Signaling Technology holds a reagent license to the patent. The company is planning to develop and commercialize validated Aqua assays for specific proteins, in particular for signal transduction pathways that have branch points, according to John Rush, the company’s director of protein chemistry, who is also an author on the PNAS publication. According to Gerber, Cell Signaling is collaborating with Thermo Finnigan, which has been funding Gerber’s research.
Rush developed an inexpensive method for producing stable isotope peptides for Aqua. While the usual cost for making such a peptide can easily top $2,000, he said, his company plans to offer enough material for up to several hundred assays for a few hundred dollars.
At the moment, Cell Signaling is looking for scientific collaborators to help it develop assays for commercialization, in return for labeled peptides at low or no cost. Rush thinks that the technology might eventually be used by researchers with little expertise in mass spectrometry, using triple quadrupole mass spectrometers, which have traditionally been used for small-molecule rather than protein applications. “They are the instrument of choice because you can tell [them] that you are only interested in a certain parent ion mass going to a certain product ion mass and set it up for SRM assays very easily,” he said.
Gerber will present an update on the Aqua method on June 9 at the ASMS meeting in Montreal, which will include unpublished applications for measuring protein ubiquitination and methylation.
So will techniques like Aqua become as common as ICAT? “I would say it’s a useful addition to the proteomics toolbox,” said Arnott, and its applicability will be proven over time. “It’s hard to say at this point how widespread it will be, but I am excited by it.”
— JK