Skip to main content
Premium Trial:

Request an Annual Quote

Aebersold and Colleagues Develop Absolute Quantification Method Suitable for All Cell Types

Premium

This story originally ran on July 21 and has been updated to include additional comments.

By Tony Fong

While relative quantification has served an important function in biology, the goal of absolute quantification has largely evaded proteomics.

But using classic mass spectrometry-based proteomics, researchers in Switzerland have developed an absolute quantification method that that they said will become a "cornerstone of quantitative biology and systems biology."

Described in a study published July 15 in the online version of Nature, the method enables the absolute quantification of "a large fraction" of the proteome in genetically unperturbed cells, and for the first time provides a technique for quantifying all cell types, the researchers said.

Applying the method to the human pathogen Leptospira interrogans, the team of researchers generated an absolute protein abundance scale for 83 percent of the mass spec-detectable proteome from cells at different states.

While a number of technologies and methods have been developed for the relative quantification of proteins in cells, absolute quantification, or the concentration of expressed proteins as a function of cellular state, has eluded proteomics researchers. To date, the only organism whose proteome has been absolutely quantified to a considerable fraction, roughly 70 percent, is Saccharomyces cerevisiae.

At the same time, the data that can be determined by absolute quantification is becoming increasingly valuable to the life science community. The advantage of absolute quantification over relative quantification is that it allows researchers to predict how a particular process behaves under specific conditions via computational modeling.

But such techniques require data to feed and test the model. "And absolute quantitative data is highly desirable for this purpose," Ruedi Aebersold, the corresponding author on the Nature study, told ProteoMonitor this week. Aebersold is a professor of systems biology at the Swiss Federal Institute of Technology and co-founder and faculty member of the Institute for Systems Biology.

While relative quantification can also be used to generate such data, absolute quantification essentially counts the molecules in an experiment and results in a protein concentration, making it possible for researchers to determine relationships between proteins, "which is a fairly significant scope of quantification," Aebersold said.

"In systems biology, increasingly the goal is to have a mathematical model of a process, and it turns out their absolute quantities are extremely useful, especially for people who work with differential equations because it gives you the boundary conditions for the equations," he said.

In yeast, absolute quantification is carried out by affinity tagging: Thousands of yeast strains are generated, each containing one affinity-tagged protein, and the concentration of that one protein in each yeast cell is measured.

This method, however, cannot be reproduced in any other organism because it depends on a technique called sited-directed recombination, in which a native protein is taken out of the genome of the cell and replaced with a tagged protein.

Another mass spec-based method for absolute quantification is based on synthesizing a heavy isotope-labeled analog for each peptide, but for proteome-wide quantification, the technique is "simply prohibitively expensive," Aebersold said. As a result, "it was hard to imagine how proteome-wide absolute quantification could happen in other species."

With the technique he and his colleagues developed — combining isotope-labeled reference peptides, label-free quantification, and LC-MS/MS for high-throughput sequencing — they were able to use a "minimal set" of isotope-labeled peptides as "absolute quantification anchor points" and extrapolate from those "the absolute quantity of a large fraction of the [L. interrogans] proteome," Aebersold said.

In the first step of their procedure, Aebersold and his colleagues fractionated tryptic digests of whole-cell protein extracts by isoelectric focusing by off-gel electrophoresis. Then they used HPLC-MALDI and LC-ESI using precursor ion selection to identify the peptides in the fractions.

From this, they identified 2,221 proteins corresponding to 61 percent of the open reading frames predicted for the L. interrogans genome. After more than 90 LC-MS/MS runs, more than 410,000 fragment ion spectra were acquired. More than 145,000 of the spectra were assigned to about 18,000 unique peptides. The identified peptides and proteins were assembled into a PeptideAtlas instance.

In the next phase, Aebersold and his co-authors chose 32 peptides corresponding to 19 proteins, based on the number of matched MS/MS spectra acquired in step one. Using selected-reaction monitoring and heavy-stable-isotope-labeled-reference peptides, they determined the absolute abundance levels for the 19 proteins.

[ pagebreak ]

"By knowing the number of cells used to generate the sample and the amount of the heavy labeled peptides added, the copy number for the selected proteins could be calculated," the authors wrote.

In the final step, they used extracted precursor ion intensities derived from LC-MS maps acquired from trypsinized cell lysates, and then calculated the total protein ion intensity for the respective proteins using the median intensities of the three most intense peptides matching a specific protein.

"The 19 anchor proteins with SRM-determined copy numbers …acted as calibration points for translating the relative abundance measurements based on extracted peptide precursor ion intensities and spectral counting into absolute abundance measurements," the researchers said.

The absolute protein abundance estimates were obtained for 769 proteins using extracted ion intensities, and the absolute abundance for an additional 1,095 proteins was estimated by spectral counting.

According to Aebersold, a key to the method is the assumption that the three most intense peptides matching a specific protein "will have similar response factors, basically a similar propensity for being observed in a mass spectrometer across all proteins.

"Then we simply identify for each protein these three high-flyers. We make a database, we tell the mass spectrometer to only measure the ion currents for these peptides, which can be done with an inclusion list in a modern instrument, and then we use about 30 absolutely quantified peptides, basically as goal posts or benchmarks," he said. "And then, by bootstrapping, we relate the ion currents of the measured high-flying peptides of these absolutely quantified peptides, and therefore we establish an abundance scale of a large fraction of the proteome."

Using their method they generated a proteome map of absolute protein concentrations for 51 percent of the ORFs predicted from the L. interrogans genome, corresponding to 83 percent of the proteome, "observable by deep proteome mapping, with an average error rate of less than threefold," the authors said.

They assessed the accuracy of their findings using cryo-electron tomography, and based on this orthogonal method "we can confirm the estimated accuracy of the determined absolute abundance protein scale," they wrote in the study.

To test the method's performance on differentially perturbed cells, they investigated the quantitative changes in the L. interrogans proteome after it had been treated with the antibiotic ciprofloxacin.

L. interrogans cells were collected and counted at three different time points of treatment: 3 hours, 24 hours, and 48 hours. After extraction and digestion, the proteins were analyzed by mass spectrometry. On average, the authors wrote, about 1,000 proteins were identified and quantified with absolute protein concentrations, and with the set of observed proteins more than 200 proteins changed their abundance more than twofold.

That highlights one limitation of the method, however. While changes of that magnitude "will be accurately detectable," more subtle changes may not be detected by the method. "Let's say if a protein goes up 10 percent, it won't be precise enough," Aebersold said.

Another limitation is that it can't detect every protein, because some proteins cannot be digested by trypsinization and therefore cannot be observed by mass spectrometry, Aebersold added. Similarly, the method won't work for low-abundance proteins that escape detection by mass spectrometry.

In an e-mail to ProteoMonitor, Virginie Brun, senior scientist at the Comissariat à l’Energie Atomique and Institut national de la Santé et de la Recherche Médicale in France who has researched new methods for absolute quantification [see PM 11/15/07] , said that the approach developed by Aebersold and his colleagues represents a "breakthrough" in the field of quantitative proteomics and systems biology.

"The beauty of the study results from synergizing a clever concept with an appropriate experimental model (low complexity and very few PTMs) and a great expertise in quantitative proteomics," she said. "Personally, I look forward to the extension of this approach to more complex proteomes, and to the impact it will have in the clinical setting through the better comprehension of physiopathological mechanisms."

Aebersold and his team are currently testing the method on cells from more complex species such as humans, mice, and rats, though Aebersold said there are no data yet he can share. Such species pose challenges because they present too many peptides in a fraction to be detected by LC-MS, so the researchers are developing new fractionation techniques.

They are testing the method on yeast, as well, and are currently amassing the data.

But for bacterial species, the method provides an "easy" absolute quantification strategy, Aebersold said. "For microbial proteomes, this is a very fast and very valid technique, and also rather cheap."

Once the proteome map has been generated, which would take only a few days for microbes using the technique described in the Nature study, it would take only a few hours mass spec time to do a quantitative measurment of the proteome and a total of about 30 stable isotope labeled peptides, Aebersold sold.

"This is the largest expense but it has to be expended only once per species," he said. "The total one time expense to establish the method is therefore less that $10,000 in material costs and each subsequent measurement is a lot less."

The Scan

For Better Odds

Bloomberg reports that a child has been born following polygenic risk score screening as an embryo.

Booster Decision Expected

The New York Times reports the US Food and Drug Administration is expected to authorize a booster dose of the Pfizer-BioNTech SARS-CoV-2 vaccine this week for individuals over 65 or at high risk.

Snipping HIV Out

The Philadelphia Inquirer reports Temple University researchers are to test a gene-editing approach for treating HIV.

PLOS Papers on Cancer Risk Scores, Typhoid Fever in Colombia, Streptococcus Protection

In PLOS this week: application of cancer polygenic risk scores across ancestries, genetic diversity of typhoid fever-causing Salmonella, and more.