A team led by researchers at the Swiss Federal Institute of Technology Zurich and the Institute for Systems Biology has published a pair of mass spec reference spectral libraries covering nearly the entire yeast proteome.
The libraries, presented in a paper published this week in Nature, provide both ion trap and triple quadrupole-based reference fragment ion spectra for peptides covering 97 percent of the predicted yeast proteome, offering scientists an almost complete set of proteomic assays for performing either discovery-style shotgun experiments or targeted selected-reaction monitoring work in yeast.
The researchers developed the library by generating reference spectra for roughly 28,000 proteotypic peptides representing around 6,400 of the total 6,607 predicted yeast proteins. They collected the ion trap spectra on a Thermo Fisher Scientific LTQ-Orbitrap-XL instrument and the triple-quad spectra on an AB Sciex Q-TRAP 4000.
Using this data, researchers can now measure any of the covered proteins via either shotgun or targeted mass spec without the need for extensive assay development, Ruedi Aebersold, an ETH Zurich researcher and leader of the study, told ProteoMonitor.
The study, Aebersold noted, is part of a larger project to develop such reference spectra libraries covering a wide range of organisms, with the ultimate goal of making mass spec-based protein analysis as straightforward and accessible for the average biologist as Western blots, currently the dominant technology for protein quantitation and a roughly $1 billion annual market.
In addition to the yeast libraries, Aebersold and his colleagues have also via their SRMAtlas project built a comprehensive map of the human proteome – a library that comprises more than 170,000 single reaction monitoring assays, one each for at least five proteotypic peptides for each of the 20,300 human genes currently annotated as protein-encoding.
The researchers announced the completion of an initial draft of that library in 2010 (PM 9/24/2010) and are currently in the process of writing a paper detailing the resource, Aebersold said. They have also completed a similar map for Mycobacterium tuberculosis for which a paper is under review, he said.
Moving forward, the researchers plan to build spectral libraries for other species and to add spectra – for phosphorylated peptides and splice and SNP variants, for instance – to the existing assays, Aebersold noted.
The paper is "a nice step in the direction of being able to take a large-scale, deeper look at the proteome in a sizable number of experiments," Princeton University researcher Leonid Kruglyak, who was not involved in the work, told ProteoMonitor.
The goal of such efforts, he noted, "is to get [proteomic experiments] to look something like microarray experiments, or RNA-seq, where you basically get reasonable measurements of any message that is expressed in your sample without a lot of complicated fractionation steps and this and that."
"It hasn't been the case that you can do that with proteins," he said.
This inability to offer researchers streamlined mass spec assays for protein detection and quantitation has hindered both the uptake of mass spec by biologists and protein research more generally, Aebersold said.
"Right now most measurements of proteins in a biological sample are done by Western blotting," he said. "And, of course, for only a relatively small fraction of protein is there an antibody [suitable for Western blotting]."
This, Aebersold noted, means that the most widely studied proteins are those for which good reagents exist – not necessarily the proteins actually most worthy of study.
Assays like those put forth in the Nature paper have the potential to have "substantial impact for biological and clinical research because [with them] essentially all proteins become measurable, quantifiable," he said. "My vision is that a biologist in a laboratory who is doing quantification of proteins by Western blotting would have access to a mass spectrometer and they could say, 'I need to measure these 60 proteins over 100 strains,' and they could simply download the assays and do the measurements."
A Role for Vendors
Achieving this vision isn't solely a matter of compiling the relevant spectral libraries, Aebersold said. The other half of the equation involves mass spec vendors and their interest in and ability to manufacture instruments that will allow non-expert users to make protein measurements at essentially the touch of a button.
Aebersold said that he had discussed the idea with several vendors and that "they were intrigued," but that there were a variety of hurdles, technical and otherwise, that would need to be overcome.
From a technical and usability standpoint, chromatography would likely present the biggest challenge, Ian Jardine, Thermo Fisher's vice president of global R&D, told ProteoMonitor.
"Biologists usually are not familiar with HPLC, and that is kind of a big step up for them," he said. Additionally, low-flow chromatography systems have typically been needed to obtain the required sensitivity for quantitation of lower abundance proteins, and these systems are notoriously finicky.
There is also the question of cost, Jardine said, noting that "biologists aren't that familiar with shelling out $300,000 [for equipment], so you would want to make [the price] $200,000 or $150,000 or even less if you could."
At that price, he said, "you get into some serious constraints in terms of cost versus performance."
"I think it would be very important to do … a detailed marketing study with biologists to see how much they would spend, how much they could get to spend, and what kind of performance they would really need," Jardine said.
Despite these concerns, he noted that in theory such a push-button device was an attractive idea, adding that Thermo Fisher actually had experience with a similar product, the ProteomeX, an integrated LC-MS/MS system launched by Thermo Finnigan in 2002.
"We actually sold many hundreds of that system," he said. "It did quite well, but it kind of got supplanted when we put out the LTQ-FT and then the LTQ-Orbitrap. The automated, integrated system kind of fell by the wayside at that point."
Like Jardine, Aaron Hudson, senior director of the academic & omics business at AB Sciex, suggested such a system was an attractive idea, noting that "it is the logical extension of where we've been going with targeted proteomics as we expand beyond the mass spec expert to the big biology labs and even small biology labs."
These researchers "don't care about the mass spectra itself. They're more interested in the proteins, are they up- or downregulated or expressed or not," he said. "I think there is a real need to update that technology."
Hudson also noted the challenges inherent in developing such an instrument, particularly with regard to getting the cost low enough to drive broad adoption. He said, though, that AB Sciex's work in applied markets perhaps points a way forward.
"It's almost analogous to what we've been trying to do in applied markets with mass spec for years now, he said, citing the company's Cliquid software, which allows users to "log in, choose an assay, and run it."
"That's been very successfully deployed" for applications like pesticide screening, Hudson said. "We have a lot of those systems out there and it has been very successful."
Dan Kassel, senior director for LC/MS marketing at Agilent, likewise noted that improvements in mass spec sensitivity are a focus, as are continued efforts to reduce costs to help instrument vendors penetrate this type of market.
However, he said, "for both research and clinical, companies are thinking about this. There is no question that there is the desire to go beyond the expert lab and put tools in the hands of the less expert."
One thing that could make a push-button research machine more plausible is the ongoing push of mass spec into the clinic, Jardine said.
"The requirement for push-button automation for clinical diagnostic instruments is extremely high," he said. "So I think what you might find over the next few years is that as those systems are being developed for diagnostics, that type of configuration can be folded back in and used for the biologist."
"The demand in the clinic is so high in terms of ease of use and robustness, that if we and other mass spec vendors can pull that off, I think we will find it is of great benefit in terms of moving into fields like [general biological research]," he said.