Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
In the cell, proteins are rarely always on or always off. Their activity typically is modulated by the addition or removal of small molecular groups, collectively called post-translational modifications (PTMs). Using these chemical tweaks, which include such modifications as phosphorylation, acetylation, methylation and glycosylation, the cell can dial in protein activity as conditions demand, creating an exquisitely sensitive and responsive system.
One tool for studying PTMs is the protein microarray, the protein equivalent of DNA microarrays, and these are available in a few basic designs. Antibody arrays enable researchers to assess modifications by the ability of capture or detection antibodies to bind the protein of interest. True “protein arrays”—arrays of cellular proteins themselves—can be used to identify the substrates of PTM-generating enzymes. Peptide arrays, which are perhaps easiest to fabricate, offer similar applications.
Relatively few of any of these classes have been commercialized—they are far more difficult to develop and manufacture than their DNA counterparts. But for certain applications, they can be especially powerful. Here, we review some of the available options.
Antibody arrays
Researchers use antibody arrays to quantify the abundance of target proteins, making the assay effectively like a multiplexed ELISA, says Julio Herrera, head of assay development at Cell Signaling Technology (CST). With its focus on PTMs, CST designed its PathScan® Antibody Arrays to quantify changes in protein modification.
Instead of testing for each PTM of interest in a separate ELISA, each glass PathScan slide contains 16 “pads,” or reaction sites, each containing about 20 antibodies targeting modified forms of key members of different signaling pathways, arrayed in duplicate, plus positive and negative controls. Detection is accomplished by applying a cocktail of biotinylated secondary antibodies followed by streptavidin-conjugated fluorophores or chemiluminescent enzymes. In this arrangement, one half of the “sandwich” is a modification-specific antibody, and the other is a total protein antibody.
According to Herrera, these arrays are intended to provide a “focused, carefully validated picture of signaling” rather than a comprehensive view of every member of every pathway. Academics and drug developers can use these tools to zero in on signaling pathways that may be implicated in a particular biological process or targeted by a specific drug—“to get a sense of what’s changing.”
CST has five PTM-focused PathScan arrays available—Akt signaling, receptor tyrosine kinases, EGFR signaling, intracellular signaling and the company’s newest product for stress- and apoptosis-signaling pathways—in both fluorescent and chemiluminescent formats. Customers’ suggestions for new arrays are always welcome, Herrera says, but CST does not generally make custom arrays. Given all the work the required to ensure the antibodies play well together, “a 19-plex array can take six months to validate,” he says.
RayBiotech takes a different approach with its RayBio® Human RTK Phosphorylation Arrays. Instead of testing for site-specific modifications one by one, these arrays, which are spotted with antibodies for 71 different receptor tyrosine kinases, are probed with a pan-anti-phosphotyrosine antibody.
Protein microarrays
Cellular protein microarrays serve different purposes than antibody arrays. Among their most popular applications are assessing protein-protein interactions and identifying autoimmune antibody targets. But these arrays also can be used for PTM analysis, to identify substrates for a given enzyme.
One such tool is the ProtoArray® Human Protein Microarray from Life Technologies (now part of Thermo Fisher Scientific), which has more than 9,000 human proteins arrayed on a single glass slide.
The key to PTM analysis with these tools, says Heath Balcer, product manager for protein biology at Life Technologies, is figuring out a strategy to determine that a modification has taken place. For phosphorylation, for instance, the company suggests incubating the array with the enzyme and radioactive ATP. For ubiquitin, researchers can use Alexa fluor-conjugated ubiquitin, while methyltransferase assays use tritiated S-adenosylmethionine, the enzyme’s methyl donor.
The challenge in building protein arrays is that unlike nucleic acids, proteins do not all behave alike. Cloning, expressing and arraying several thousand proteins in such a way that all fold correctly is no mean feat. Expression yield can vary wildly from protein to protein, as can stability. Michael Snyder, the researcher (then at Yale University) whose lab first developed a yeast predecessor to the ProtoArray, accomplished the trick by expressing all the proteins as glutathione S-transferase (GST) fusions.
An alternative approach, developed by Joshua LaBaer, director of the Virginia G. Piper Center for Personalized Diagnostics at Arizona State University (ASU), is the “nucleic acid programmable protein array,” or NAPPA.
“We developed [NAPPA] to get around one of the challenges of protein microarrays, which is producing proteins separately and arraying them,” LaBaer explains.
Instead of arraying proteins, the NAPPA technique instead spots cDNAs encoding each protein, which is easier to synthesize than protein and far more uniform in behavior, LaBaer says. To create the working protein array, users simply incubate the slide with a HeLa cell extract, which performs both transcription and translation in situ, wash and use.
“We get a much more even distribution of proteins,” LaBaer says. “Approximately 93% of proteins are within two fold of the mean.” Plus, the proteins are usually no more than an hour old, he says. “They are made fresh and remain in solution until testing.”
NAPPA slides contain about 2,300 spots each, a limit caused by the fact that transcripts made on-chip can diffuse. The lab has made higher-density arrays, however, using chemically patterned slides.
Originally, LaBaer's team built their NAPPA arrays from cDNAs encoding GST fusion proteins, which bind to anti-GST antibody-functionalized slides. More recently, they have been using fusions with Promega’s HaloTag®, a modified dehalogenase that forms a covalent linkage to a surface-conjugated chloroalkane group. Using this approach, binding is covalent, meaning washes can be more stringent. Arrayed proteins also can be denatured if desired, for instance to expose internal epitopes.
Marjeta Urh, director of research and proteomics group manager at Promega, says another advantage of HaloTag is that the arrayed proteins are always oriented in the same direction. That’s in contrast to some other arraying strategies, in which proteins bind to the surface randomly.
LaBaer’s lab mostly uses its arrays to study autoimmunity, but researchers in his lab also have used them to identify substrates of AMPylation and citrullination. In the former case, team member Xiaobo Xu used a modified form of ATP called N6pATP (N6-propargyl adenosine 5-triphosphate), which allows the labeling of AMPylated proteins with azide-functionalized fluorescein via Click chemistry.
“The general theme is, if you have a clever chemist and you want to do PTM studies, you modify the substrate so you can come in later to see them by fluorescence,” LaBaer says.
NAPPA microarrays are available through the NAPPA Core Facility at ASU, which has 60,000 to 70,000 cDNAs available. Custom designs are also available.
Peptide macroarrays
A third approach to PTM analysis is the peptide macroarray. EMD Millipore, for instance, uses its AbSurance™ Histone Antibody Specificity Arrays (PDF) in-house to profile its own PTM-specific antibodies.
Comprising 89 19-mer peptides from histones H3, H4, H2A and H2B spotted on a low-fluorescence PVDF membrane, AbSurance macroarrays are intended to give researchers confidence in the quality and specificity of PTM-focused antibodies, says John Rosenfeld, R&D group leader for epigenetics at EMD Millipore.
“This is a method we use for the histone antibodies we develop and our attempt to provide transparency into the quality of our products,” he says. The approach also provides a way to demonstrate antibody specificity if requested by manuscript reviewers, he adds.
But researchers also may use the array to assess activity of histone-modifying enzymes, says Michael Sturges, senior product manager for epigenetics at EMD Millipore; the company currently is developing a product along that line.
“We are intending to commercialize a similar array on a slide, and we will validate that product for protein-protein interaction screening and the ability of chromatin modifiers to bind to modifications and add or remove modifications,” Sturges says.
The current AbSurance macroarray is split over two membranes, one for histone H3 and the other for H2A, H2B and H4, with peptides spotted at two concentrations. The complete set costs $255 for both membranes, or $143 for one.
Fundamentally, of course, a microarray experiment is not the end of a study, but the beginning, a way to generate hypotheses. Follow-up studies are inevitable, typically using more focused technologies such as ELISA or mass spectrometry. “If you are starting out with no knowledge of what you’re working on, an array is a good first step,” Balcer says.
Update (6/11/14): This article was updated to add information about RayBiotech.