A team led by Swiss Federal Institute of Technology Zurich researcher Ruedi Aebersold has published a quantitative characterization of the proteome of the human tissue culture cell line U2OS.
Detailed in a paper in the current edition of Molecular Systems Biology, the study identified approximately 10,000 proteins, quantifying roughly 7,300 of these across a concentration range of seven orders of magnitude up to 20 million copies per cell.
The effort is one of the most comprehensive quantitative mappings of a human cell line’s proteome to date, and, said Aebersold, offers a mass spec workflow for studying proteome-wide changes in a clinically relevant human cell line in response to various perturbations.
The technique uses an initial set of LC-MS/MS runs to experimentally determine the best flying proteotypic peptides for each protein. This allows for a mass inclusion list-driven strategy that focuses MS-sequencing time on only these best flying PTPs, maximizing protein coverage. Quantification of the detected proteins is achieved by correlating the average of the signal intensities of the three best responding peptides per protein with a calibration curve built using a set of isotopically labeled reference peptides.
In July, Aebersold’s team published a similar study, also in MSB, on a mass spec workflow for high-throughput quantitation of full proteomes (PM 7/22/2011). Developed using the human pathogen Leptospira interrogans, the method presented in that project was limited to relatively simple proteomes, like those of single-cell organisms. By combining the method with extensive upfront fractionation, the ETZ scientists were able to adapt it to more complex human cells.
“The quantification schema is essentially the same idea,” Aebersold told ProteoMonitor. ”But it had to be done in fractions because the proteome of the human cell is substantially more complex than the [bacterial] one done” in the initial study.
The researchers used isoelectric focusing via off-gel electrophoresis with an Agilent 3100 OFFGEL fractionator to accomplish the sample fractionation. This method, Aebersold noted, offers higher resolution than separation methods like ion exchange chromatography, which was a key advantage for integrating the quantitative data across the multiple peptide fractions.
Because a given peptide typically appeared in more than one fraction, to quantitate their targets the researchers had to “recognize which fractions a peptide is in and then basically sum up the signals across the fractions in which the peptide elutes,” he said. The high resolution of the off-gel electrophoresis method meant that most peptides eluted in no more than two fractions, he added, making summing across fractions “fairly straightforward.”
“That would have been different if we have used ion exchange chromatography where the resolution is lower,” Aebersold said.
The ETH team followed the fractionation with mass spec analysis on an Thermo Scientific Orbitrap Velos instrument, analyzing each fraction three to four times in shotgun mode and then two to five times in directed mode using the mass inclusion data generated in the first series of runs. They validated their measurements by counting nuclear pore complexes in U2OS cells via high-resolution confocal fluorescence microscopy, finding that the numbers of these complexes agreed well with the copy numbers of the relevant proteins measured by mass spec.
The protein measurements generated by the technique “are not super precise,” Aebersold noted, adding that it can detect differences of roughly one-and-a-half-fold and greater. Even at this level of precision, however, the technique is “quite useful” for making quantitative measurements of the proteomes of perturbed cells, he said.
To demonstrate this application, the researchers repeated their initial experiment with U2OS cells arrested in mitosis, quantifying approximately 6,800 proteins. Proteins showing copy number variations compared to non-synchronized cells “were significantly enriched for biological processes carrying out mitotic functions,” they noted in the MSB paper.
The study also enabled an analysis of the relationships between protein abundance and function. While the data largely reinforced the established notion that a relatively small number of high-abundance proteins are responsible for core cellular processes while a larger number of low-abundance proteins handle regulatory functions, some interesting variations from this trend emerged, Aebersold said.
In particular, he said, the researchers found that “protein kinases span a very broad range of cellular abundance,” countering the current understanding that such signaling proteins are consistently present at low levels.
The research, Aebersold said, was more to establish a workflow for quantitative proteome-scale comparisons across multiple cellular states than to achieve an exhaustive mapping of the U2OS cell line. However, he noted, the study does represent one of the largest collections of proteins quantified in a human cell.
According to the MSB paper, previous studies in U2OS cells topped out at 5,399 proteins, while studies investigating HeLa cells peaked at around 3,000 proteins. A study led by Max Planck Institute researcher Matthias Mann also published in the current MSB identified 10,255 proteins in a HeLa cell line.
Based on the Mann study and his own team’s data, Aebersold said he believes that in the U2OS study the ETH researchers essentially saturated the proteome accessible via the LC-MS/MS workflow that they used.
“That’s not to say that there are not other types of proteins in these cells,” he said. “But we would claim that with this workflow — this particular type of cell lysis, this particular type of digestion, this particular type of LC-MS/MS — we would be unlikely to discover many additional proteins even if we kept sequencing.”
“If instead of trypsin you used a different protease, you might uncover a different protein, or if you used harsh lysis conditions to extract membrane proteins, you could certainly uncover additional proteins,” Aebersold said.
"It is possible and likely that there are proteins that are simply not solubilized and so don’t make it into the sample, and how many of these there are we simply do not know,” he said. However, he added, "I think we can be fairly confident that whatever was in the sample that was extracted and digested into peptides was actually identified."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.