Josh P. Roberts has an M.A. in the history and philosophy of science, and he also went through the Ph.D. program in molecular, cellular, developmental biology, and genetics at the University of Minnesota, with dissertation research in ocular immunology.
The adage “garbage in, garbage out” certainly holds true for proteomic analysis, whether it’s Western blotting, mass spectrometry (MS), interaction studies or any of the burgeoning fields of subproteomics. To get good results, you need to start with a good preparation. Salts, detergents, protein contaminants, enzymes, sample concentration and a host of other considerations need to be accounted for.
For many downstream applications, it makes sense to start by running a polyacrylamide gel to separate proteins by (apparent) size and remove salts and detergents. From there, it’s relatively straightforward to transfer to a membrane and run a Western blot, or use affinity reagents to visualize your protein of interest directly. Even with a complex sample such as serum with “a bazillion proteins on it, you’re only going to see whatever the antibody reacts with,” says Carol Beach, who runs the University of Kentucky proteomics core facility.
Yet Beach can’t simply take a plug out of that gel and tell you what it is. “Do a protein stain instead of an antibody stain, and that little plug may contain 50 proteins. And you don’t know which one we’re talking about. You need to purify it,” she explains.
Depletion
Many times the problem is that the protein(s) of interest are of low abundance in the sample, so they get swamped out. Researchers often start with a “depletion that will remove abundant proteins and increase the depth of coverage—that’s a pretty common strategy,” says Matt Foster, assistant professor in pulmonary medicine at Duke University.
Such purification is frequently done by liquid chromatography (LC) columns such as Agilent’s MARS [multiple antigen removal systems], which (in the case of the Human 14 cartridge) is designed to remove 14 highly abundant proteins—94% of the total—from human plasma samples. LC and its “high-performance” cousins (HPLC, UPLC, etc.) require expertise and/or specialized equipment that are beyond the scope of this article, but Foster points out that the same technology is available in a spin-column format.
The MARS spin columns don’t come cheap (retail price in the US for the MARS spin cartridge Human 14 is $2,231 according to the Agilent Website), but they’re designed to be regenerated and reused for several hundred runs, according to Agilent technical support.
Other vendors offer similar solutions, such as Sigma-Aldrich’s Seppro IgY14 column, which uses chicken-derived antibodies to capture the 14 most highly abundant proteins in human plasma. Sigma-Aldrich also offers the Seppro Supermix column, which will pull out the next 200-or-so (mid-abundant) proteins. When the two columns are used in series, about 99% of the total protein is removed, leaving behind “the low-abundance proteins, which [are] generally where biomarkers are,” says Aaron Sin, the company’s global product manager for applied markets.
Although the MARS and Seppro columns are good for “serial” applications, a researcher isn’t likely to purchase enough to run many of them in parallel. That’s where kits like the Pierce Abundant Protein Depletion Spin Column from Thermo Fisher Scientific come in, notes Foster. They allow users to simultaneously run 16 samples and just throw the columns out when they’re finished. "That saves time.”
“We’re the first company to have a single-use spin column that’s very cost-effective,” notes Monica O’Hara Noonan, market segment manager for sample preparation at Thermo Fisher Scientific, which launched its Top 2 and Top 12 Columns – respectively targeting the two and twelve most abundant proteins – in January.
Positive selection
Affinity selection also can be used to capture the proteins a researcher is interested in retaining (some even collect albumin by eluting it from albumin-depletion columns). A host of reagents are available to enrich for phosphorylated proteins, glycosylated proteins or secreted proteins.
Foster says he has had very good success pulling down proteins that have been genetically tagged—with GFP or a FLAG-tag, for example—using specific reagent resins like GFP-Trap from ChromoTek GmbH and anti-FLAG M2 affinity gel from Sigma-Aldrich, respectively. He cautions, though, that because other proteins may bind nonspecifically to these resins, “it’s probably best if you have a system where you can elute your epitope specifically” by competing it off with the tag.
The CaptureSelect series of affinity reagents, based on 15 kD, single-chain camelid antibodies (produced in yeast but originally derived from camels) conjugated to a solid support—including small, porous beads for HPLC as well as agarose—has recently become available through Life Technologies. “The true benefit is that you have the ability to affinity-purify a range of molecules where affinity-purification techniques did not previously exist,” says Christine Gebski, project management leader at Life Technologies. A customer would have needed to buy an anti-FSH (follicle stimulating hormone) or anti-GCSF (granulocyte colony stimulating factor), for example, and couple that to an activated bead, she explains, noting that there is not another commercially available anti-FSH resin. Among the other agarose-conjugated offerings are anti-human transferrin, -growth hormone, and -von Willibrand factor.
Tissue-specific, compartment-specific
Scientists often ask to isolate protein from a particular tissue or compartment, and “we’ve developed a series of reagents that are very cell-specific as well as location-specific,” says Noonan. Among Thermo Fisher’s newer compartment-specific tools are the Subcellular Protein Fractionation Kit for Tissues, which allows for extraction of five cellular protein fractions from a tissue sample in less than two hours, introduced last year, and Mem-PER Plus (Membrane Protein Extraction Reagent), introduced in January.
Tissue-specific reagents that debuted early last year include N-PER (Neuronal Protein Extraction Reagent) for extracting proteins from primary cultured neurons and tissues, and Syn-PER (Synaptic Protein Extraction Reagent) for isolating functional synaptosomal complexes, or synapses.
Sigma-Aldrich offers its own lines of compartment- and tissue-specific extraction reagents under the ProteoPrep and CelLytic brands. Many are labeled as specific for mammalian, plant, yeast or prokaryotic cells.
What’s old is new
Not every protein-extraction tool is new, of course. Some old protein-preparation standards also are getting facelifts or turbocharges.
“Protease-inhibitors cocktails have seen a renewed interest” for MS, says Sin, who has seen a surge of interest in phosphatase inhibitors, as well. “We’ve had them for a long time, but now people are looking at customizing the mixes.”
Thermo Fisher Scientific found a way to manufacture MS-grade trypsin much more inexpensively, says Noonan, and in January began selling it “at a fraction of the cost of what some of our competitors sell it for.”
The company also recently introduced an automation-compatible, 96-well dialysis plate with free-floating cassettes, as well as a 250-ml dialysis flask that looks like a tissue-culture flask to get rid of contaminants and get the protein into the buffer needed for the next application.
The point is this: Protein-sample preparation is undoubtedly a mature field. But small and large innovations alike continue to enable easier, faster and perhaps more efficacious protein preparation for a host of downstream analytical applications. So don’t be afraid to update your protocols.