by Caitlin Smith
Proteins are complex beasts. The variety in their characteristics confounds our abilities to demystify them. Thus in proteomics—the study of the proteins comprising the proteome—there is no single best way to prepare a sample for analysis. It depends on the protein you are interested in—and importantly, the vast multitude of other proteins that happen to be there too.
“As proteomics has evolved, researchers have realized that sample preparation prior to downstream protein analysis is critical for producing meaningful, reproducible results,” says Kate Smith, product manager for expression proteomics at Bio-Rad. Results so meaningful, in fact, that they can be used for quantification. “Undoubtedly the most important recent development in proteomics has been its transition from being an ‘identification method’ to being a relatively accurate quantitation technique,” says Ole Vorm, chief scientific officer at Proxeon. “With quantitative information and the ability to study proteins dynamically, proteomics has become a crucial cell biology technique that should benefit virtually every cell biology study for the next hundred years.” Here are some ways in which sample preparation is being refined for proteomics researchers.
Fractionating to simplify
How do you obtain useful quantitative data from, say, a complex cell homogenate? “Fractionation” is Smith’s answer, which she defines as the separation of protein mixtures into discrete fractions that may be analyzed independently. “Fractionation has become the most widely used sample preparation technique,” she says. “Researchers have realized that due to sensitivity limitations of their analysis methods, and the wide concentration range and vast number of proteins in a sample, they must fractionate to gain meaningful information. As such, the trend in recent years in proteomics has become a more targeted approach as opposed to a broad exploration of the entire proteome.”
One of many tools that Bio-Rad offers to reduce sample complexity is the MicroRotofor cell, a preparative isoelectric focusing (IEF) device that fractionates complex protein samples in free solution. Using small sample volumes, the resulting fractions can be pooled and refractionated for further purification. Fractions also can be further analyzed by two-dimensional gel electrophoresis or mass spectrometry techniques. “As a result of the fractionation,” says Smith, “researchers can detect more proteins and are able to more easily study differential expression.” Bio-Rad also offers MicroRotofor lysis kits for easier sample preparation prior to loading the MicroRotofor cell, with protocols for different sample types such as mammalian, plant, bacterial, and yeast.
Novagen
offers different ways to fractionate samples before further analysis. Their ProteoEnrich ATP-Binders Kit enriches for protein kinases and other ATP-binding proteins using an affinity resin containing immobilized ATP; subsequently, eluted kinases retain their activity. For enriching a sample for low molecular weight proteins, the ProteoEnrich CAT-X SEC Kit uses a chromatography resin that is penetrable by polypeptides whose globular size is less than 20 kDa. These proteins bind to the surface of the inner pores according to net charge, and can be fractionated with a salt gradient.
But don’t forget that in order to fractionate your sample, you need to protect its contents first. Homogenizing cells to begin sample preparation results in liberating multiple types of proteases that chew up many types of proteins, possibly including your own. An easy way to prevent this is the addition of protease inhibitors. Roche Applied Science’s Complete Protease Inhibitor Cocktail Tablets contain an optimized mix of various protease inhibitors. You simply add one tablet, which is soluble in all aqueous solutions, directly to your sample for protein protection. To prevent dephosphorylation of your proteins of interest, Roche also offers their new and similarly convenient PhosSTOP Phosphatase Inhibitor Cocktail Tablets.
Improving on immunodepletion
A related complicating factor of working with protein mixtures is that there is a large range of concentrations: some proteins, such as keratin, are abundant, while others are scant, and only show small but important changes during experiments. The abundant proteins can obscure small changes in the lower-abundance proteins. One way to mitigate this problem is to deplete the complex mixture of known interfering proteins with specific antibodies. While still widely used, technology is improving on this tried-and-true enrichment method.
“The ProteoMiner kits are an excellent tool for reducing the concentration range of samples,” says Smith, “and allow researchers to dig deeper in the proteome and uncover low-abundance proteins that would otherwise be masked by high-abundance proteins.” Bio-Rad’s new ProteoMiner kits use a combinatorial library of hexapeptides that serve as unique binding partners for proteins. Bound to a bead matrix, the population of hexapeptides is diverse enough that most or all proteins in a sample should find a binding partner. “Since there are a limited number of binding sites per protein, high-abundance proteins quickly saturate their binding partners, while low-abundance proteins continue to bind,” explains Smith. “Following elution, the low-abundance proteins are enriched while the high-abundance proteins are reduced. This allows for the detection and identification of low-abundance proteins that cannot be discovered using other methods.” ProteoMiner differs from immunodepletion in that it isn’t specific for any particular species, is not antibody-based, and has a much higher capacity for lower-abundance proteins.
The iMac of proteomics
Proxeon has simultaneously advanced ease-of-use and separation power with the recent introduction of their Easy-nLC, a nano-flow HPLC that applies a gradient to the sample in the nL/min range without splitting the solvent stream. This is important because, according to Vorm, “split-free systems are inherently more robust and easier to use in the low nanoliter flow ranges (50 – 300 nL/min) than systems with split solvent flows.” He says that while Eksigent also offers a split-free system, Proxeon’s Easy-nLC is unique because it “is exceedingly easy to use and has been optimized for coupling to a mass spectrometer. It is not designed to be a multi-purpose chromatography system.” The complete self-contained system is intended to arrive at the user’s lab completely optimized and ready to go. “A bit like an iMac,” says Vorm. “The idea behind this design decision is that nanoLC historically has been very challenging, and state-of-the-art nanoLC-MS has been mastered by only very few labs. With our Easy-nLC, non-expert users can also obtain state-of-the-art data with ease.”
Proxeon recently showed that the Easy-nLC could be used for two-dimensional separation prior to mass spectrometry. Previous groups had abandoned efforts to succeed with this method because the synchronization of additional lab equipment made the technique unfeasible. “As a result, 2D-LC-MS has been a niche-technique despite the impressive data that some groups demonstrate from time to time,” says Vorm. The Easy-nLC can perform automated two-dimensional separation without extra equipment, simply and reliably, even for non-expert users. “We hope this will become a proteomics revolution,” says Vorm. “At least a small one.”
Challenges ahead
Despite the technical successes in different fractionation methods, proteomics still has some significant hurdles looming in sample preparation. Most protein scientists face the challenge of “difficult” proteins, says Smith, such as membrane proteins, very acidic or basic proteins, and post-translationally modified proteins. In addition, Vorm cites the ongoing challenge of samples with a wide concentration range. He thinks that several orders of magnitude may be gained within the next few years via improvements in mass spectrometry and the use of nanoLC separation prior to it. “With the natural difference in protein abundance, the overall workflow should ideally have a dynamic range of 1010,” says Vorm. “We are currently far from that performance, with more than a factor of 100,000 still missing.” No doubt more improvements in sample preparation technology will see us there.
