Researchers at the Beijing Proteome Research Center have developed a fast mass spec workflow that enables profiling of whole mammalian proteomes in as little as half a day.
Detailed in a paper published this week in Molecular & Cellular Proteomics, the method has become the standard mass spec workflow for the Chinese Human Proteome Project, said Jun Qin, a Baylor College of Medicine researcher and leader of the CHPP's mass spectrometry efforts.
Relying on dual reverse phase HPLC followed by MS/MS analysis, the workflow is similar to conventional mass spec approaches but cuts down on the time required by using much shorter gradients, Qin told ProteoMonitor.
In the paper, the researchers used 30-minute gradients for each of the two RP-LC runs, allowing them to identify more than 8,000 proteins in HeLa cells in 12 hours, more than 6,000 proteins in six hours, and roughly 2,800 proteins in 30 minutes of mass spec run time. By way of comparison, recent studies using more conventional, longer gradient times took three days to identify a similar number of proteins, while recent efforts identifying roughly 10,000 proteins in human cells took in the range of two to three weeks (PM 11/18/2011).
In addition to protein identification, the BPRC researchers are applying the workflow to label-free quantitation of proteomes, as well, Qin said. Currently, they are able to perform pair-wise quantitative comparisons of human cell proteomes in around three days, he said, a time that they hope to bring down to as little as a day through further optimization.
Qin and his colleagues performed the experiments on three different mass spec instruments – an AB Sciex TripleTOF 5600, a Thermo Fisher Scientific Q Exactive, and a Thermo Scientific LTQ Orbitrap Velos. Using the 30-minute gradients, they identified 8,154, 8,145, and 6,763 proteins, respectively, on the Q Exactive, TripleTOF 5600, and Orbitrap Velos.
The results, Qin said, indicate that while the capabilities of newer mass specs like the Q Exactive and TripleTOF provide a boost in protein IDs, the method is still applicable to older machines.
Using the newer machines "helps a little bit, but only by around 10 percent – not that significant," he said.
Employing the workflow for quantitation, the researchers analyzed two sets of human umbilical vein endothelial cells – one a control and the other treated with the NAE-inhibitor MLN4924 – quantifying 6,784 proteins and identifying 194 differentially expressed proteins. Quantification took 12 hours of total mass spec time for each cell set, a roughly eight-fold reduction in time compared to conventional workflows, the authors wrote.
The BPRC researchers were led to try the reduced gradients "out of necessity," Qin said, citing the magnitude of the CHPP undertaking.
"We're trying to cover the whole proteome, but it took a month to do those experiments and so that was a problem," he said. "So I felt that the time [requirements] had become critical, because ideally you want to finish the experiment in a day or two."
The CHPP project was officially launched at the Human Proteome Organization's 2010 annual meeting where Chinese researchers announced plans to expand their work on HUPO's Human Liver Proteome Project into an effort to characterize the human proteome in its entirety.
One of the centerpieces of the project is a new national laboratory in Beijing called the Pilot Hub of Encyclopedical proteomIX, or PHOENIX, center, which the government is funding to the tune of ¥1.2 billion ($200 million). Over half of those funds have been slated for the center's new mass spectrometry facility, headed by Qin. When fully operational, the center – which is scheduled to open in 2015 – will employ more than 300 researchers and contain more than 30 high-end mass spectrometers.
Currently, Qin said, there are more than 10 mass spectrometers at centers throughout China running samples around the clock for the CHPP effort. The researchers are working to characterize the proteomes of 10 different organs plus blood, he said, adding that recently these analyses have begun in earnest.
"We've started taking serious data," he said.
Given the vast scope and dispersed nature of the project, arriving at a standard operating procedure for proteome characterization has been a significant challenge, Qin noted.
"I've spent a lot of time doing pilot experiments at centers around China trying to come up with a SOP so that people can use almost the same approach for everything," he said, adding that the workflow presented in the MCP study will now serve as the basis of that SOP.
Beyond speed, the workflow's most significant advantage is that by bringing down analysis time, it reduces the variation introduced into experiments by the inherent instability of the LC and mass spec instrumentation, Qin said.
"That's actually the major benefit," he said. "Because I can keep a machine stable for a day, but I can't keep it stable for a month."
This becomes particularly important in the case of quantitation experiments, Qin said.
"Because we have to do replicates and everything, the best we can do right now for quantification is one [pair-wise comparison] every three days," he said. "But we are still optimizing it. We have to [further] shorten the time to analyze the proteins, because we really need to finish the experiment within a day for pair-to-pair [proteome] comparisons."