By Tony Fong
A wide-ranging team of researchers will use a $4.8 million ARRA grant to test the feasibility of measuring all proteins in humans.
The project, which will be funded for two years, will act as a "precursor" study to determine whether the Human Proteome Detection and Quantitation initiative, or hPDQ, can be done.
Participants in the first phase include researchers from the Broad Institute, the Fred Hutchinson Cancer Research Center, the company Plasma Proteome, and others.
First proposed in January in an article published in Molecular and Cellular Proteomics, hPDQ would create a library of anti-peptide antibodies to measure collections of proteins with high sensitivity and specificity [See PM 01/22/09]. The project would serve as a "near-term tactical approach" for providing a more basic map of the human proteome than the Human Proteome Organization's Human Proteome Project.
The current project will determine the feasibility of such a project by testing what it would take to build multiplexed mass spectrometry-based assays, using a multiple-reaction monitoring strategy, to cover the entire human proteome.
Because making and charactering antibodies for such assays are both time-intensive and costly, "it makes sense to first … characterize at the peptide level what assays work really well — what peptides work really well — and then subsequently go in and make antibodies in subsequent phases to those peptides to improve the sensitivity and throughput of the assays," Amanda Paulovich, a co-PI on the project, told ProteoMonitor this week. She is an associate member in the Clinical Research Division at the Fred Hutchinson Cancer Research Center specializing in genetics and oncology.
The project was recently identified by ProteoMonitor as the proteomics-directed research that received the highest stimulus funding in fiscal year 2009 [See PM 10/16/09].
Steven Carr, director of the Proteomics Platform at the Broad Institute, is the other co-PI, and Leigh Anderson, founder and CEO of the Plasma Proteome Institute, is senior advisor on the project.
"So this is the precursor stage where we're focused entirely on the MRM, defining what the best proteotypic peptides are, what the parameters are, and doing exploratory studies and cell lysates to see how well the assays behave," Paulovich added. "And then in a subsequent phase beyond this two-year early phase, we would go on to generate antibodies to improve sensitivity and throughput."
The fact that existing assays can measure only "very few human proteins" currently has been a "major impediment to research," she said. "So the idea here is that if we can achieve the ultimate goal of building a highly specific, sensitive, [and] multiplexible mass spec-based assay for every human protein, that would greatly facilitate not only basic biomedical research but also applied research such as biomarker studies."
Carr emphasized that though the project has a strong clinical angle, the benefits would have a much broader reach for the whole proteomics community.
"It's really about quantitative biology," he said. "While the project is addressing what appears to be a clinical need, it really has extensibility to any biological question you wish to answer in whatever context — tissue, cell line, biofluid — you want to address."
To those ends, Paulovich, Carr, and their colleagues will build assays to proteins detected in frozen breast cancer surgical specimens, then configure and fully characterize them in breast cancer cell lines. The goal is to configure a total of 400 novel assays targeting 400 peptides from 200 proteins.
They will then evaluate the assays in 51 breast cancer cell lines that had been previously characterized from a genomic perspective by Joe Gray's laboratory at the University of California, San Francisco.
Paulovich, Carr, and colleagues chose these particular proteins because they are expressed in human breast cancer and are detectable by mass spectrometry. Just as importantly, they may have some clinical relevance.
They will be evaluating the clinical relevance by mining Matthias Uhlen's Human Protein Atlas to look for proteins that are expressed in breast cancers but not in normal breast epithelium. "Those would be higher priority targets for us," she said.
[ pagebreak ]
Also, they are integrating published DNA and genomic analysis data to find genes whose copy numbers or expression levels are correlated with poor outcome in breast cancer.
"So if we have a protein that we can see in surgical samples by mass spectrometry and [it] appears to be aberrantly expressed, or at a high copy number in poor-outcome breast cancers, or is in Mathias Uhlen's data … those would be the highest priority proteins of all for us for assay development," Paulovich said.
The assays will be characterized with stable isotope standards and, once they are fully optimized and characterized, the researchers will make multiplexed measurements of the proteins across 26 of the 51 cell lines "spanning different molecular subtypes of breast cancer," she said.
Paulovich and her collaborators are awaiting Institutional Review Board approval, and once that occurs "we're good to go." Her and Carr's labs will each develop half of the assays and swap 10 percent of them to assess the reproducibility and portability of the assays, key criteria in their research.
Other important factors being evaluated in determining the feasibility of the hPDQ project are the assay's sensitivity in a commonly used biospecimen, such as cell lysate; their linear range; their precision; and multiplexity, Paulovich said.
Their work will also draw on the efforts of researchers at the Institute for Systems Biology to build an MRM Atlas [See PM 10/23/09]. Paulovich and Carr will be using the MRM atlas to create their assays.
"The Atlas will allow [Paulovich] to quickly establish the assays for the proteins important for the MRM assays without having to recreate the fragmentation profiles initially," Robert Moritz, one of the PIs on the MRM Atlas project, told ProteoMonitor by e-mail.
Paulovich's team "can use the MRM Atlas like a catalogue [to determine] what peptides would be best and the instrumentation conditions needed to carry out the assays all obtained from the Atlas project," he said.
Push to Map Human Proteome
The research being done by Paulovich and her colleagues come also as some in the proteomics community are pushing for efforts to map the human proteome. HUPO first proposed its Human Proteome Project in the spring of 2008 [See PM 05/09/08], and though the 10-year, $1 billion proposal has met resistance from funding agencies, various studies and projects that would carry out aspects of HPP are being conducted.
hPDQ was proposed as a surrogate for HPP to “enable individual biological researchers to measure defined collections of human proteins in biological samples," with high sensitivity and "absolute specificity, at throughput and cost levels that permit study of meaningfully large populations,” or between 500 and 5,000 samples, researchers wrote in the MCP article describing the project.
hPDQ “is aimed at providing immediately useful capabilities of the human biology research community in a way that does not adversely impact funding for individual investigators and does not generate administrative constraints on their ability to set and change courses in the conduct of research,” the researchers added.
Carr said that while he could not say that the current ARRA-funded research would convince funders to start throwing money at the HPP initiative or hPDQ, it should help define times and costs for each stage of such large-scale human protein mapping projects and take them out of the pie-in-the-sky realm.
"Here's what it costs, here's how it's going to get done, here's the likelihood of success. It makes it much more concrete," he said, adding that the Human Genome Project proceeded only when the technology was stabilized, and similarly, it was shown how much such a project would cost and what steps were needed to complete HGM.
On the hPDQ project, Carr said that meetings are taking place, including one that was held this week, to determine the best way to proceed. Those meetings will be opened up in the spring to other groups already developing MRM assays, clinical chemists, such as those working in in vitro diagnostics, and mass spec vendors, to gather their input, he said.
A 'Clinically Accessible' Human Proteome?
Some preliminary work on hPDQ has been done, but Carr declined to comment on any data.
Ultimately, if he and his colleagues are successful in their ARRA-funded project, it could lead to the creation of new methods and technologies for measuring human proteins, which in turn could lead to breakthroughs in clinical proteomics, including the development of new diagnostic tools, screening tests, drugs, and personalized medicine.
In a statement, Henry Rodriguez, director of the National Institute's Clinical Proteomic Technologies for Cancer program, said, "This pilot has the potential of developing the first step toward making the entire human proteome clinically accessible."
Other collaborators on the project include researchers from Massachusetts General Hospital and the University of North Carolina, Chapel Hill. Life Technologies is a commercial partner, as well, and the Applied Biosystems 5500 QTrap is being used on the project.
"In the same way that the Human Genome Project has really enabled quantitative genomics to happen … this really provides the basis for a large-scale quantitative proteome project that would be analogous to that and equally transformative to the biomedical research enterprise," Paulovich said.