A team led by scientists at the University of Manchester has identified a set of potential protein biomarkers for preeclampsia via a label-free selected-reaction monitoring mass spec workflow.
Presented in a paper published last month in Molecular & Cellular Proteomics, the label-free SRM approach eschews the isotopically labeled peptides typically used in SRM workflows, offering biomarker researchers an inexpensive way to evaluate potential markers identified in discovery experiments before passing them on for more thorough validation, said Richard Unwin, author on the paper and head of the proteomics laboratory at the Manchester Biomedical Research Centre's Centre for Advanced Discovery & Experimental Therapeutics.
In protein quantification via SRM mass spec, researchers typically spike samples with stable isotope labeled peptides as standards. However, Unwin noted, such labeled peptides can be a significant expense, especially in experiments seeking to quantify large numbers of proteins.
"If you want a high-quality, pure isotopically labeled peptide, you can get quotes for around £250 to £300 [or $390 to $470]," he told ProteoMonitor. "So if you have 15 proteins coming out of a discovery experiment and you want to test for five peptides in each protein, that starts to add up."
Exacerbating the situation, Unwin said, is the fact that funding is typically more difficult to obtain for biomarker validation work than for initial discovery experiments.
"We wanted a very inexpensive, relatively quick way of being able to take a list of candidates from a discovery experiment and narrow it down to a better list of proteins that we could then take into a more formal sort of validation," Unwin said.
"The discovery tools [the field] uses are relatively blunt instruments in terms of how accurately they quantify any given peptide," he added. "You don't put large numbers of samples into discovery experiments, and so you do get false positives cropping up."
Beyond helping to weed out likely false positives, a quick and inexpensive initial validation lets researchers pick out the peptides likely to make the best biomarkers from a technical standpoint, Unwin noted, allowing them to identify "the ones that are generated reproducibly by [trypsin] digestion, the ones that chromatograph well, that are well behaving for quantification."
To test their label-free SRM method, the Manchester researchers developed proof-of-principle assays to peptides from candidate proteins — platelet basic protein and several members of the pregnancy specific glycoprotein family — identified via analysis of iTRAQ labeled samples on both an AB Sciex Qstar XL QTOF and a 5800 MALDI TOF-TOF.
To determine that their target peptides were reproducibly produced and detected following digestion, the team performed 12 replicate digests of a single plasma sample and measured its reproducibility.
They likewise used 12 replicate digests to generate dilution curves to determine the linearity of their SRM assays. Taking advantage of the fact that PSG is found only in the plasma of pregnant women, they also measured their assays' linearity by generating a dilution series by spiking early pregnancy plasma into male plasma. This also allowed them to establish that their measurements were sufficiently sensitive to detect relatively small changes in protein concentrations. For instance, they wrote, in a comparison of median levels of the PSG peptides in the 100 percent female plasma versus 80 percent female plasma, all peptides reached high significance.
In all, Unwin and his colleagues used their label-free SRM approach on an Agilent 6460 triple quadrupole instrument to measure 19 PSG peptides and five platelet basic protein peptides in 58 preeclampsia cases and 42 controls. While the label-free method did not allow for absolute quantification, they were able to identify PSG-5 and PSG-9 as differing between preeclampsia cases and controls, while also identifying PBP as a likely false positive.
Moving forward, the researchers plan to test the identified PSGs in larger patient cohorts, Unwin said, noting that they have obtained roughly 250 additional samples along with 500 matched controls.
"So the next stage is hopefully to test these proteins in those cohorts, and then if they are still promising the next step would be to go to a clinical trial where we are looking at thousands of people," he said.
The researchers have not yet patented the markers and have no immediate commercialization plans, Unwin said, adding that although promising, "at the moment the individual PSG markers on their own are not sufficiently predictive."
"We suspect they might [be useful] added to other [preeclampsia] markers," he said. "So if we were to patent or license something, the thing we would look to patent or license is a panel of markers" including the PSGs.
Preeclampsia has drawn a fair amount of interest from proteomics researchers, with two companies — Belgian protein biomarker firm Pronota and Stanford University spinout Carmenta Bioscience — prepping proteomic tests for the condition.
Pronota is currently undertaking a prospective, 2,000-patient clinical study of its preeclampsia panel in support of a US Food and Drug Administration 510(k) submission and an application for a European Union CE mark. The company is also in discussions with several large in vitro diagnostics makers, including Roche, Abbott, and PerkinElmer, about licensing the panel as a complement to their tests for the current standard preeclampsia marker PlGF (PM 7/13/2012).
Carmenta, which this month announced that it has raised $2 million in seed funding, is preparing for a several hundred-patient validation study of panels of seven and 15 preeclampsia markers (PM 2/1/2013).