ADAPTED FROM WIKIPEDIA/SIMIN LIM
Thirty years after it came into use, the protein microarray—a high-throughput tool that tracks the presence, activity, and interactions of proteins—is still going strong, even as its DNA counterpart is being replaced by next-generation sequencing. Protein arrays are workhorses in biomarker discovery, immune-response monitoring, and other areas of biology.
Traditional protein microarrays—slides, beads in a liquid, or plates that are specked with hundreds or thousands of proteins, probed and analyzed, often with fluorescent labels—allow an unbiased look at protein function. Unlike other proteomics techniques, arrays immobilize even low-abundance proteins. “The nice thing about protein arrays is that every protein gets its chance. Every protein is there at reasonable levels,” says Josh LaBaer, director of the Virginia G. Piper Center for Personalized Diagnostics at Arizona State University’s Biodesign Institute.
Protein array technology has continued to improve in throughput and sensitivity, allowing researchers to better capture and probe proteins and their interactions. “Don’t be afraid of the technology,” says Manuel Fuentes, scientist at the Cancer Research Center in Salamanca, Spain. “It’s very high density and you can get a lot of data, which you need to analyze carefully.” However overwhelming the setup of these large-scale assays can seem to a beginner, a single array can answer all of your questions, he adds.
The Scientist talked to experts about what they’ve done for protein microarrays lately—and what protein microarrays have done for them. Here’s what they said.
Approach: One widely used type of protein microarray known as NAPPA (nucleic acid programmable protein array) relies on DNA templates arranged as an array. When these strands are exposed to an in vitro transcription/translation (IVT) system—essentially, cells that have been lysed—the proteins coded by the templates self-assemble. In NAPPA, developed by LaBaer and his team, each DNA template contains an expression cassette for both the gene of interest and the gene expressing glutathione-S-transferase (GST), a naturally occurring protein that facilitates protein folding. The template is attached to a glass slide alongside an anti-GST antibody that captures the newly made protein. As a more recent alternative to the antibody capture, the group uses Promega’s HaloTag, a fusion protein that can covalently attach proteins to the array, allowing researchers to treat it with stronger denaturants.
What’s new: Among other improvements to the tool, in 2013 LaBaer and his team started using lysate from HeLa cell lines on these arrays. Compared with the conventionally used lysate from rabbit reticulocytes, the human lysate boosts protein expression on the array 10-fold and behaves more consistently from run to run (Prot Clin App, 7:372-77, 2013). For studying human proteins, “you feel a lot more confident when you’re making proteins in their native milieu,” LaBaer says. The 1-Step Human Coupled IVT kit is sold commercially by Thermo Scientific for $158 (8 reactions of 25 μL each), although anyone can make it themselves, LaBaer says.
Applications: Researchers have used NAPPA in a variety of contexts, such as for identifying autoantigens from patients with rheumatoid arthritis, ankylosing spondylitis, and cancer. In 2011, LaBaer and his collaborators used it to identify a protein signature that could be used in early detection of breast cancer.
Getting started: It’s possible to set up NAPPA using the published methods. (Check out a step-by-step video tutorial at Current Protocols in Protein Science, 2011, Chapter 27: Unit 27.2.) However, because array-printing machines are less common nowadays, many researchers choose to outsource the printing steps. LaBaer’s NAPPA core facility prints custom arrays and ships them at cost. “[But] we routinely tell people that if they want to do it, we can have them come to our lab for a few days and show them, hands on, tricks of the trade,” he says. Because the arrays are completely customizable, you’ll want to spend some time up front thinking through whether and how a chip can help answer your research questions.
Cost: Varies widely, depending on how many proteins you produce on a single array and how many arrays you want to make. Commercial arrays can cost $2,000 apiece, but are closer to $150 when purchased through LaBaer’s core facility. With NAPPA arrays, it’s not inconceivable to do a study with 300 patients, each patient with his or her own array, LaBaer says.
Caught in a sandwich
Approach: First described in 2008 by scientists at the Babraham Institute in Cambridge, U.K., DAPA (DNA array to protein array) is another method in which the array self-assembles after DNA templates are exposed to transcription and translation machinery (Nature Methods, 5:175-77, 2008). In contrast to NAPPA, where DNA and captured proteins sit together on the same surface, proteins in DAPA are caught on a second slide coated with a capture reagent. The second slide is aligned face-to-face with the DNA template slide. Traditionally, the meat of this slide sandwich was composed of a membrane soaked with the cell extract.
Theoretically, DAPA has a cost-savings advantage over other cell-free protein arrays because it offers the possibility of reusing the DNA template slides for creating multiple protein prints. However, such a step might be more trouble than it’s worth. “It’s something that we don’t do that much at the moment, because we find it’s easier to make more DNA arrays,” says Oda Stoevesandt, a postdoc at the Babraham Institute. Generating protein arrays of consistent quality from reused DNA arrays will require further optimization of the immobilization and washing protocols, she adds.
What’s new: Recently, the researchers found that the membrane middle of the DAPA sandwich wasn’t necessary, says Stoevesandt. It acts as a spacer between the slides and a spongy reservoir for the cell-free mixture. But a fluid-filled gap that held the slides an even distance apart generated much more even pictures, she says (New Biotechnology, 29:586-88, 2012).
Applications: Stoevesandt and her colleagues are using DAPA to characterize the specificity of antibodies. “Basically we’ve got an antibody, it’s supposedly for protein X. What happens if we put it on an array with protein X alongside others?” she says. The group can also use the arrays to determine targets of autoimmune antibodies, though they haven’t tried this yet.
Getting started: Researchers at Babraham have published several protocols—most recently in Methods in Molecular Biology (1118:245-55, 2014). To use it, you’ll need the equipment (a DNA array printing machine) to spot the DNA arrays, as well as molecular biology skills. Assembling the DAPA sandwich requires careful layering of materials, much like putting together a Western blot. Although it’s not especially tricky, says Stoevesandt,“a certain dexterity comes with handling it more than once.” In addition, she says, there’s no reason why human lysate from HeLa cells couldn’t be used with DAPA for protein transcription and translation.
Cost: Varies widely, because of the possibility for customization. A DNA arrayer can cost around $100,000 or more.
Beads of a color
ILLUSTRATION BY BURCU AYOGLU, COURTESY OF JOCHEN SCHWENKApproach: In contrast to planar protein microarrays, liquid arrays depend on the use of beads attached to protein-specific antibodies and are distinguished based on their color, size, or shape. The sample is added to the bead solution and, after a series of incubation and washing steps, the mixture can be analyzed using a flow cytometer or similar machine. Compared to planar arrays, bead-based arrays are more flexible and make it easier to analyze a larger number of patient samples in a single run using standard microtiter plates, says Jochen Schwenk of the KTH-Royal Institute of Technology in Stockholm, Sweden.
What’s new: Schwenk and his colleagues describe a workflow combining both planar and bead-based arrays to profile autoantibodies (Molecular & Cellular Proteomics, 12:2657-72, 2013). “Our methodology is an entry point in selecting interesting protein profiles, and then doing subsequent verification to provide evidence that the indications are true and lead you to the protein you want to study,” says Schwenk. In proteomics, platform changes are a good way to filter out spurious findings that could be linked to a particular technology, he adds.
Applications: Bead-based arrays are a go-to method to profile bodily fluids of healthy and diseased individuals as part of the Swedish Human Protein Atlas project, which aims to systematically map the human proteins in organs, tissues, and cells.
Getting started: Nowadays, bead-based arrays can be performed using a variety of commercially available consumables. For liquid arrays you’ll need access to one piece of equipment—a readout flow cytometer–like instrument (available from Luminex or Bio-Rad), which analyzes as many as 500 of Luminex’s analytes at once. That’s compared to two systems—an arrayer plus an imager—for planar arrays. It’s possible, but time-consuming, to make your own color-coded reagents. Fridtjof Lund-Johansen, who leads the protein array laboratory at Oslo University Hospital in Norway, has generated his own microsphere sets with 1,700 fluorescent bar codes.
Cost: $800 to $3,400 for Luminex bead panels that target as many as 30 different proteins at once. Depending on the instrument, Luminex’s xMAP analyzers range in cost from $25,000–$150,000. Prices for Bio-Rad’s Bio-Plex multiplex products, available only by contacting the company, depend on the system you choose. The most economical option is the small footprint Bio-Plex MAGPIX model, which sells for about the same price as an ELISA plate reader, but high-throughput automation-ready models can cost more than $100,000.
COURTESY OF SOMALOGICApproach: First described by Boulder, Colorado-based SomaLogic in 2010, SOMAscan gives researchers the ability to capture 1,129 distinct proteins on a single array. Streptavidin beads linked to SOMAmers—short strands of DNA aptamers that contain chemical modifications formulated to tightly bind specific proteins—are mixed into a biological sample, allowing SOMAmers to bind to their cognate proteins. SOMAmer-protein complexes are then bound to a second set of streptavidin beads on an array, after which they are isolated and analyzed with microarrays and other DNA technologies.
What’s new: SomaLogic is regularly increasing the size—that is, the number of proteins—of the assay, says chief medical officer Stephen Williams. The company is making thousands of new reagents—modified aptamers specific for a broad range of proteins—per year, and will continue to increase the multiplexing as the market demands.
Applications: SOMAscan is mostly used for biomarker discovery in humans. By comparing protein concentrations between healthy and diseased individuals, SomaLogic has obtained protein signatures for 50 diseases and conditions. Because the company has just started selling the service to academic scientists, Williams expects more findings on SOMAscan to be published in the next year. “If [scientists] already know what they want to measure, then it’s probably the wrong tool,” says Williams. “If they want to find out something unexpected—as well as the expected things—then that’s what it’s useful for.”
Getting started: Although now offered strictly as a service, as part of a new partnership with Agilent, the SOMAscan assay will soon become available in a handful of academic core facilities. Part of the deployment will include technical support and training from SomaLogic. “The actual mechanics of running the assay are not terribly complicated,” Williams says, though decent high-throughput expertise is needed. “The equipment is quite simple. A good lab scientist could do it.” In addition, the company offers a free data-visualization software program to its customers.
Cost: $1,800 for one SOMAscan array, though the company offers “substantial discounts for reasonable volumes, so most people don’t pay that,” Williams says.