Differential proteomics studies are among the more common approaches to identifying potential protein biomarkers. However, the wide dynamic range of protein expression in common samples like serum can impede the detection and quantitation of low-abundance proteins.
Many proteins that offer promise for early detection of cancer and other diseases are low-abundance, which has complicated biomarker discovery efforts, necessitating sample prep strategies like depletion by immune-affinity capture or precipitation by acetonitrile to reduce levels of common analytes prior to analysis.
Recently, several research groups have investigated the use of libraries of hexapeptide ligands to compress dynamic range in serum and saliva. Offered commercially by Bio-Rad under the name ProteoMiner, these reagents have demonstrated potential as tools for improved analysis of low-abundance proteins in complex media.
The ProteoMiner beads are derived via combinatorial chemistry, resulting in libraries typically comprising between 5 and 10 million hexapeptide ligands. According to Roumen Bogoev, senior product manager at Bio-Rad, when samples are run through the beads, these ligands capture effectively all of the low-abundance proteins.
The very high-abundance proteins, on the other hand, saturate their hexapeptide baits, meaning, that a large proportion of these proteins remain unbound and are subsequently depleted in the flow-through, Bogoev said. This results in depletion of high-abundance analytes and enrichment of low-abundance ones, allowing for an increase in protein IDs.
Importantly for differential proteomics work, lower-abundance proteins within samples of interest retain their quantitative accuracy after treatment with the ProteoMiner beads provided they're in concentrations of 1 micromolar or lower, said Sricharan Bandhakavi, a senior scientist in Bio-Rad's life sciences group. Above that, he told ProteoMonitor, the proteins will saturate the hexapeptide ligands, diluting the actual amount of quantitation.
In a paper published in this month's edition of Clinica Chimica Acta, a team led by researchers at the French National Institute of Health and Medical Research in collaboration with Bio-Rad scientists investigated use of the ProteoMiner beads for plasma biomarker discovery.
They reported that with hexapeptide treatment they were able to detect 3,191 protein spots via 2D electrophoresis, compared to 538 in untreated plasma. They also tracked the relative levels of inflammation marker C-reactive protein to confirm that hexapeptide treatment retained differences in protein abundance, finding that the treated samples maintained their original relative levels of CRP.
CVs for the runs using the beads were in the 15 percent to 30 percent range, the study noted, putting them in the same range as for the total protein assays. The method's variability "is a question that has now been addressed in quite a few publications," Bandhakavi said, "and they've found that at least for serum it's highly reproducible. If you take 10 to 100 microliters of beads, you have enough hexapeptide diversity that the variability is minimized."
The technique is potentially most interesting for samples other than serum, however. While conventional immunodepletion methods are well-developed for blood and serum, they aren't as useful for other samples like saliva, Bandhakavi said.
"Immunodepletion columns typically target 10 or 20 [of the] most abundant proteins in plasma or serum," he said. "If you're interested in a sample other than plasma or serum immunodepletion columns won't be of as much use because the most abundant proteins in saliva or urine or cell lysates are probably going to be different."
In two papers published recently in the Journal of Proteome Research, Bandhakavi, along with scientists at the University of Minnesota – where he was a research assistant professor prior to joining Bio-Rad – looked at using hexapeptide libraries in proteomic analyses of whole human saliva. In particular, they investigated its usefulness for studying post-translational modifications phosphorylation and glycosylation.
In the first study, which was published in December, the researchers identified 259 glycosites in human saliva using hexapeptide beads compared to 127 glycosites in untreated samples. They also developed a workflow for discriminating a priori high abundance proteins from lower-abundance proteins, providing a practical method for determining which analytes in a given sample will retain their quantitative accuracy after hexapeptide treatment.
In the second study, they identified 129 phosphorylation sites, only 11 of which had been previously identified in saliva.
According to Bandhakavi, the combined results of the two papers suggest that, in addition to enriching for low-abundance proteins, hexapeptide libraries can be used to enrich for PTMs as well. While researchers have yet to test the technique across a wide range of PTMs, the fact that it worked with both glycosylation and phosphorylation indicates it could be broadly applicable.
"We chose glyco- and phospho-enrichment for two reasons," Bandhakavi said. "One is that they're both biologically interesting. Two is that glycosylation modifies the physico-chemical properties of proteins very dramatically whereas phosphorylation has a modest effect, relatively speaking. So if ProteoMiner is able to enrich both of them, it should probably do everything else in between as well."
One potential limitation of the system is the relatively large amount of sample needed. The recommended protocol calls for 50 mg of proteins per 100 microliter of beads.
"Basically, you need milligram quantities of proteins," Bogoev said. "It may be possible to go further down, but we don't have data on that."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.