NEW YORK (GenomeWeb) – Key to developing successful multiple-reaction monitoring assays is careful selection of peptides on which to perform quantitation.
In addition to selecting proteotypic peptides, researchers also look for peptides that, among other things, fly well in the mass spectrometer and suffer minimal interference from other molecules.
In a paper published this week in Nature Methods, researchers from the UK's Institute of Cancer Research and Cancer Research UK Manchester Institute (CRUK) suggested that researchers should more carefully consider potential changes in particular peptides across samples and conditions.
As the authors note, the assumption in MRM-MS-based quantitation is that levels of individual peptides are accurate and consistent proxies for the levels of the proteins from which they came. However, factors like post-translational modifications and alternative splice forms could lead to situations where the relationship between the level of a peptide and its corresponding protein differ under varying experimental conditions.
Peptides that can be used as accurate proxies for the level of their corresponding proteins are termed quantotypic. And, said Claus Jørgensen, a CRUK researcher and senior author on the paper, while the targeted proteomics field has identified the need for identifying good quantotypic peptides, there has been less actual investigation to date into the issue.
"It's something that has been discussed a bit, but nobody has really looked at the extent of it, what is the actual impact, [how many] peptides that, while proteotypic, don't necessarily constitute good quantotypic peptides," he told ProteoMonitor.
Jørgensen and his colleagues tried to get at this question in their recent work developing MRM assays for the human kinome. Looking to identify human kinases across six cell lines, they used ActivX nucleotide analogs to enrich for kinases, ultimately identifying 219 protein kinases using a Thermo Fisher Scientific Orbitrap Velos instrument.
They followed this by evaluating the proteotypic properties of the tryptic peptides in the identified kinases, using in silico digestion and determining the intensity of the various MRM transitions. Ultimately, they identified 790 promising proteotypic peptides targeting 196 protein kinases, or 38 percent of the human kinome — and 132 of these proteins were covered by three or more proteotypic peptides.
The ICR team then set out to evaluate how quantotypic these peptides were, which they did by determining the ratios between proteotypic peptides from the same protein and then measuring them across multiple conditions – the notion being that, if these peptides are, in fact, quantotypic, the ratio between them should remain the same regardless of changes in the expression level of their corresponding proteins across conditions.
The researchers measured the ratio between all proteotypic peptides on each of 109 protein kinases across six cell lines, finding strong correlations between the ratios for 80 percent of the protein pairs. They then repeated this analysis in a different set of six cell lines, again identifying 80 percent of the peptides as being quantotypic. Of the peptides determined to be quantotypic in the initial six cell lines, 72 percent were again determined to be quantotypic in the second set of six cell lines.
In total, the researchers were able to identify quantotypic peptides for 107 protein kinases – 21 percent of the human kinome.
Jørgensen noted that the question remains, however, whether peptides determined to be quantotypic under a certain set of conditions will remain quantotypic across all conditions.
"I think we still need to learn a lot more about how condition-specific this is," he said.
To begin approaching this question the researchers treated cells with a variety of growth factors and then assessed whether the peptides previously determined to be quantotypic remained so under this stimulation.
That work found that only "a minimal number of peptides were affected by these kinds of stimulations," Jørgensen said. "So that does suggest that the quantotypic peptides that we identified probably work across a greater number of conditions than what we used them in."
He added that the researchers found that "you don't actually need that many different types of samples to evaluate this," noting that they maxed out the number of quantotypic peptides they identified at around 10 different conditions.
"That really suggested that with about 10 different conditions – which if you are doing a large study is not that many – you can validate your quantotypic peptides, and then you can use them across a large number of samples," he said.
Further work is need, though, to more rigorously assess the quantotypic nature of target peptides, Jørgensen said, adding that, in particular, analyses using absolute quantitation are necessary.
"Essentially we based our definition of what constitutes a good set of quantotypic peptides on the fact that their abundance correlates well across a large number of samples," he said. "But in order to really asses this we need to really start doing some more absolute quantitation of a larger number of peptides across a larger number of conditions. We need standards that we can quantify against."
Jørgensen said that he and his colleagues have also begun comparing data generated from peptides validated as quantotypic to that generated by peptides not validated as such, but, he said, they would need a larger dataset before they could draw any strong conclusions from this work.
They plan to continue the work, adding a larger number of peptides and conditions to their analyses, he said. Additionally, the researchers aim to use the developed MRM assays for their cancer proteomics research.
"Where we come from actually is that we are very interested in protein kinase signaling, in particular in the context of tumor development, metastasis, and drug sensitivity," he said. "So that is where we started, trying to develop assays that would allow us to really quantify low levels of protein kinases, and that is how we got into this."