Commercial Sample Preparation Materials for Isolation of Intact Proteins from Biological Samples are Absent in the Marketplace

We have been working to develop the use of liquid chromatography–triple quadrupole–mass spectrometry (LC–MS/MS) for quantification of intact proteins from biological fluids (1). The idea is to essentially mimic a small molecule workflow, and to quantify proteins in a top-down fashion using multiple reaction monitoring. It works.

The motivation for developing such methodology includes the growing desire to track protein biomarker targets for disease diagnosis, prognosis, and treatment effectiveness. The biopharmaceutical industry is also interested in developing new protein-based drugs and needs to follow these through clinical trials to determine absorption, distribution, metabolism, and excretion.

Initial studies showed that this top-down approach is not as sensitive as traditionally developed bottom-up-based protein quantitative assays (1,2). Modern triple-quadrupole instruments are optimized for small molecule analysis. Yet, a top-down approach offers the potential for more absolute quantitation and comparisons between laboratories. Bottom-up methods are primarily based on relative quantitation.

The main problem is that the technology to perform top-down quantitation of multiple intact protein targets from biological samples is not advanced far enough to be commercially available. Such preparations are performed routinely for large panels of small molecules using combinations of on- and offline solid phase extraction, protein precipitation, and isotope dilution. The offerings are lacking for targeting protein molecules.

For example, I have yet to find a selection of large-pore-size solid phase extraction products. These would allow some strategizing (reversed-phase, hydrophilic interaction, or ion exchange) for discrimination of various intact proteins (different isoelectric point [pI] values and molecular weights [MWs]) from phospholipids. Just like for small molecules, the presence of phospholipids can kill electrospray ionization efficiency of target proteins.

Worthy to note is that we are not considering immunoaffinity based methods, because we want to be able to prepare and analyze multiple diverse proteins in a single run. Mixed immunoaffinity extraction systems could be considered, but these would still be less generally applicable (one antibody is needed for each specific class of proteins) than having some traditional isolation phases available.

There are some other interesting variations possible. There is certainly potential for some monolithic technologies; polymeric monoliths have been shown to be especially useful for separation of proteins, and these also boast wide pH range compatibility. Silica gel may not be optimal for preparing proteins with more extreme pI values, as it can be much harder to manipulate their overall charge in solution.

Coacervate formation is an intriguing possibility. There are some interesting mixtures of fluorinated and surfactant-based compounds that cause phase separation in aqueous-based systems. Phospholipids in many biological matrices may be complementary associates for some perfluorinated compounds; when added, these fluorinated compounds can associate with phospholipids and also enrich hydrophobic proteins in the formed coacervate phase.

Molecular weight cut-off filters are mainstays in the biochemical laboratory. Various centrifugal spin filters, which contain polymeric membranes permeable to some molecular weight cutoffs, can be used to isolate fractions of molecules that are greater than or less than 3000, 5000, 10000, or 30000 Da. However, these cutoffs are not very accurate. For sure, the protein conformation will affect its ability to be filtered in the desired molecular weight range. Further, protein-protein and protein-small molecule interactions can hurt expected recovery values for such treatments.

Because modern liquid chromatography–triple quadrupole mass spectrometry (LC–MS) instrumentation is not optimized for large-molecule analysis, there is need to concentrate target proteins during sample preparation to reach desired detection limits. If we are to target multiple biomarkers to follow disease progression in a routine fashion, then we need to have broadly applicable and reproducible commercial sample preparation solutions. These all probably need to be in microplate format. I think we are far from there. For all innovators in industry and academia, that means there is more work to do. If someone who reads this has a potential solution, please contact me if you are interested in having us try it.

References

1. E.H. Wang, P.C. Combe, and K.A. Schug, J. Am. Soc. Mass Spectrom. 27, 886–896 (2016).
2. E.H. Wang, D.K. Appulage, E.A. McAllister, and K.A. Schug, J. Am. Soc. Mass Spectrom.28, 1977–1986 (2017).

Kevin A. Schug is a Full Professor and the Shimadzu Distinguished Professor of Analytical Chemistry in the Department of Chemistry & Biochemistry at The University of Texas at Arlington. Direct correspondence to: kschug@uta.edu