Comissariat à l’Energie Atomique and Institut national de la Santé et de la Recherche Médicale
Who: Virginie Brun
Position: Postdoc at Comissariat à l’Energie Atomique and Institut national de la Santé et de la Recherche Médicale where she is developing strategies for absolute quantitative proteomics
Background: DVM, École Nationale Vétérinaire de Lyon, 1999; PhD in physiology and transcriptomics, Claude Bernard University, Lyon, France
The focus of proteomic studies has moved toward a quantitative approach, but quantifying proteins and biomarkers has remained a challenging problem. In a paper under review by Molecular & Cellular Proteomics, the authors say that while for many protein biomarkers, an immunological approach has been taken, this approach, which depends on the development of high specificity and high affinity antibodies, has found limited success, and “the simultaneous optimization of several protein assays is hardly ever possible.”
Mass spectrometry techniques for quantification, they say, have their own issues. Label-free methods, while relatively inexpensive, have “rather poor accuracy.” Meanwhile, strategies using isotope-labeled peptides have been developed, but they can lead to severe biases, the authors of the study say.
To solve this issue, they propose a method dubbed Protein Standard Absolute Quantification, based on the use of in vitro synthesized full-length isotope-labeled recombinant proteins as quantification standards.
ProteoMonitor spoke with Virginie Brun, the lead author on the study, about the approach. In the work described in MCP, Brun and her colleagues applied their method for detecting and quantifying Staphylococcusaureas superantigenic toxins, a potentially important application considering recent reports of the existence of particularly virulent strains of methicillin-resistant S. aureus.
Though the work was more in the realm of food safety and public health, Brun said that the method can be used for any protein biomarker work.
Below is an edited version of the conversation.
What do you mean by absolute quantification?
We want to determine the exact concentration of the analyte in the sample. Absolute quantification differs from relative quantification because relative quantification is just an analysis between different samples.
Relative quantification does not lead to an exact determination of concentration. It just says this protein is more [abundant] in this sample than the other.
Why is absolute quantification important?
I think it’s very important for the validation of biomarkers because in this step, in the validation or verification of biomarkers, the analysis must be quantitative and very accurate because a threshold between the physiological and pathological samples must be determined.
You need to know the concentration [to see] which samples are normal. For instance, if somebody is ill, the concentration will be increased in comparison to the normal threshold.
Describe Protein Standard Absolute Quantification and the work you did developing it.
To date there are mainly two methods for absolute quantification of proteins in samples. These are AQUA [Absolute Quantitation] and QconCAT [concatemer of standard peptides]. These two methods are based on internal standardization with isotope-labeled peptides. However, these peptides are generally added at late stages of the analytical process.
So they do not allow accurate quantification because they do not take into account the yield of biochemical prefractionation and/or tryptic digestion. We have postulated that the best internal standards for absolute quantification of a protein would be the same protein but with stable isotope. We have synthesized isotope-labeled proteins to serve as internal standards for quantification, and we obtained very good accuracy. This internal standardization takes into account the prefractionation yield and the digestion yield.
We developed it on a biological model, which is the detection of Staphylococcus aureas toxins in water and urine samples. These toxins are involved in toxic shock syndrome and food poisoning. They also represent potential biological weapons. So we were very interested in detecting them and quantifying them in water and urine.
So we have synthesized isotope level toxins to detect and quantify these proteins in water or urine samples.
You say you compared PSAQ with AQUA and QconCAT. What did you find in the comparisons?
By comparing these three methods, we have shown that the PSAQ strategy is exquisite for absolute quantification of proteins in prefractionated samples. We have also shown that this strategy can be used for the quantification of poor protease-sensitive proteins such as toxins. Finally, this method is robust as it offers the largest sequence coverage for quantification.
You say in your study that the use of isotope-labeled marker peptides for MS-based absolute quantification of proteins can lead to severe biases. What did you mean?
If you have a very complex analytical process, for instance with biochemical purification and partial digestion, these peptides won’t allow you to quantify accurately the proteins because they do not allow [you] to take into account the yield of the biochemical purification and proteolysis.
An internal standard has to be added at early stages of analytical processes. It has to be added directly in the clinical sample in urine or serum.
Have you used the technique in your research yet, or is this still a proof-of-principle finding?
We have made the proof of principle and we are now developing this method, so we have worked on serum.
Have you experienced any unseen issues or problems with the method you developed in using it in your research?
There are limits, which are currently the quantification of proteins [that] are post-translationally modified, like phosphorylated of glycosylated. However, we have some ideas to try to quantify these types of proteins. So we are now developing new standards that will allow the quantification of phosphorylated and glycosylated proteins.
Are these standards going to be based on the method described in the Molecular & Cellular Proteomics paper?
You’ve applied your method in urine, water, and serum samples.
Yes, and we are now working on plasma.
Does that make any difference, what biological sample you use? Will it affect the outcome or results?
The method is very accurate whatever the sample is. We have verified this because it was quite easy [in] water but it was also verified in blood and plasma.
And is this a method that can be used easily by other researchers, in terms of the effort involved and costs?
We think it’s not a very difficult method. When people are interested in one protein, usually they have cloned it. They can easily express this protein in an isotope-labeled form. We think this method can be applied in a wide range of labs. What we would like to do is constitute a bank of isotope-labeled proteins.
Is this method more applicable for drug discovery and development?
Any one who wants to quantify a protein in biological sample can use this sample quite easily. The only [requirement] is to have a mass spectrometer.
And this is not vendor- or platform- specific?
It’s not mass spectrometer specific. It can be applied in MALDI, or Q-TOF, or ion trap.