To mangle an oft-posed question, if a protein is expressed in a cell and nobody’s around to see it, does it have an effect?
Of course it does. But that doesn’t mean it’s easy to measure.
The simplest way to study a given protein—whether it’s a question of abundance, location, turnover, interaction partners or post-translational modifications—is by using antibodies. But many proteins don’t have corresponding antibodies available. And there also are some cases in which antibodies simply won’t do—monitoring protein trafficking in live cells, for instance.
Fortunately, tool developers offer a diverse array of protein tagging technologies to handle such situations.
In vivo tags
Tagging technologies can be divided into three fundamental categories. The first is metabolic tagging strategies, or in vivo labeling.
In this approach, cells are fed a chemically tagged nutrient or building block, which they then incorporate into newly synthesized proteins, nucleic acids or metabolites. Researchers can then harvest the cells and isolate these molecules to gain either a global view of cellular behavior—for instance, are the cells producing more or less phosphoproteins, in general—or to examine one protein in particular using an antibody or other capture reagent.
One popular metabolic approach for proteomics researchers is SILAC, which stands for stable isotope labeling with amino acids in cell culture. Used to drive mass spectrometry experiments, SILAC enables researchers to quantify differences in protein abundance between two cultures or cell lines by labeling one with a heavy amino acid (e.g., N15- or C13-lysine). Protein lysates grown under the two conditions are then mixed and analyzed simultaneously, yielding a relative assessment of protein abundance differences under the two conditions.
SILAC reagents are available from both Life Technologies and Thermo Fisher Scientific.
Non-mass spectrometrists can use so-called “bio-orthogonal” chemistries. In a bio-orthogonal system, cells are fed molecular building blocks tagged with some chemical group that is nonreactive toward natural biological reactive groups, such as amines or carboxyl moieties. Addition of an exogenous compound containing that group’s reaction partner instigates a chemical reaction, coupling the tagged biomolecules to a desired functional group, such as a fluorophore.
Thermo Fisher Scientific offers reagents to link azide and phosphine groups in what is called a Staudinger Ligation. “What you do is feed cells with [a building block containing] the azide group, and then after the experiment, you treat with a reagent containing a phosphine group, and the two snap together,” explains Rizwan Farooqui, market segment manager for Thermo Scientific Protein Research Products. Azide and phosphine groups don’t normally exist in biological systems, so these molecules are otherwise inert, Farooqui says.
In one example on the company’s website, HK-2 cells were incubated with N-azidoacetylmannosamine, a sugar analog, to tag newly synthesized glycoproteins. Then they were treated with a phosphine-coupled fluorophore (DyLight 650-phosphine) to visualize those proteins.
Thermo Fisher Scientific offers three different phosphine-fluorophore conjugates, as well as a biotinylated phosphine for protein isolation on streptavidin-conjugated surfaces.
Life Technologies Click-iT® line is analogous, but it is based on the reaction of an azide-modified protein with an alkyne-modified partner in the presence of a copper catalyst. (A copper-free form of the system is also available for live-cell applications, as copper can be toxic to living cells.) “In nature, you have no azide-alkyne chemistry normally,” says Chockalingam “Palani” Palaniappan, vice president of research and development for Flow and Imaging Cytometry Products at Life Technologies. That means the technology should produce relatively low background, he explains. “It’s truly orthogonal chemistry.”
Click-iT building blocks are available for labeling proteins (e.g., an azide methionine analog) as well as post-translational modifications like glycosylation and palmitylation. Labeling reagents are available in seven flavors, including six fluorophores and biotin.
According to Palani, this type of reagent can be used in live cells for “pulse-chase” experiments, in which cells are first fed, say, an azide building block and an alkyne dye to label newly synthesized proteins. Then, after a time, the cells can be fed alkyne-containing building blocks and labeled with a different color azide dye, to distinguish newly synthesized proteins from their older counterparts.
A more common approach to tagging uses recombinant DNA to express a protein tag. In this type of system, researchers clone their gene of interest in-frame with a sequence encoding either an affinity tag or a gene that they can capture or detect, respectively, for instance using antibodies or via fluorescence.
One application of these tags is to enable protein purification in the absence of a protein-specific antibody. For example, companies like Sigma-Aldrich and Agilent Technologies offer vector systems for tagging proteins with peptide antigens such as FLAG®, c-Myc or CBP (calmodulin-binding peptide), for which specific capture reagents exist.
Both Sigma-Aldrich and Agilent also offer vectors encoding “tandem-affinity tags,” dual peptide sequences that can be used in a two-step purification strategy to yield highly purified protein. Sigma-Aldrich’s TAP system, for instance, couples both the FLAG and HA (hemagglutinin) tags to a protein of interest; protein isolation over an anti-FLAG antibody column, followed by an anti-HA column, yields a more purified product than that produced with only one enrichment step, says product manager Caleb Hopkins. This strategy often is used to capture intact protein complexes and evaluate protein-protein interactions.
Genetic fusions with fluorescent proteins enable protein visualization in vivo. Clontech Laboratories now has 22 fluorescent proteins from blue to far-red in its Living Colors product line. These include a handful of “switchable fluorescent proteins.”
According to Michael Haugwitz, Clontech’s associate director of research and development, these protein variants differ in their emission spectra (color) and brightness. And they may function as multimers (e.g., ZsGreen, which is a tetramer) or monomers (e.g., mCherry)—the latter being ideal for fusion protein expression.
Clontech’s newest offerings including photoswitchable protein vectors, which come in three flavors, says Haugwitz. Dendra2 changes color upon photoactivation, from 507 nm to 573 nm. PAmCherry is dark until activated, at which point it fluoresces red; among other applications, this fluorophore has been used in super-resolution microscopy applications such as PALM and STORM. And Timer is, well, a timer; it automatically and slowly matures from green to red at a fixed rate so researchers can, for instance, back-calculate how long a given promoter has been active in a cell based on how much green and red protein is present.
Clontech also offers what Haugwitz calls “protein functional control” tags. The iDimerize tag enables researchers to force their protein of interest to dimerize using a specific ligand. The ProteoTuner tag destabilizes proteins so they degrade very rapidly. Addition of a ligand called Shield1 stabilizes the protein within 15 minutes or so, enabling users to monitor the effect of rapidly inducing a protein. “That’s much faster than transcriptional systems like tetracycline induction,” Haugwitz says.
Several companies have developed customizable systems that enable users to apply any of a number of different chemical tags to a given protein, depending on the application. These include New England Biolabs (SNAP-tag, CLIP-tag, ACP-tag and MCP-tag); Life Technologies (Lumio); Active Motif (LigandLink™); and Promega (HaloTag®).
Each system is unique, but the basic idea in all these cases is that a protein of interest is fused to another protein or peptide, which specifically binds a detection or capture reagent. That event enables the researcher to label or purify the protein, depending on the particular application—kind of like one of those modular ratchet sets in which a common adaptor can be fitted to any of a number of different tools.
Promega’s HaloTag system, for instance, couples the protein of interest to a modified enzyme (the Rhodococcus rhodochrous DhaA protein) that covalently couples to a chloroalkane substrate. If that substrate were bound to a fluorescent dye—the company has an assortment of fluorescent substrates available in different colors, some of which are cell permeable and others of which are not—the result would be a fluorescently labeled protein. If the substrate were attached to a solid surface, however, the result would be an immobilized protein. Several different HaloTag-specific surfaces (called HaloLink), including magnetic and nonmagnetic beads and the HaloLink Protein Array System, are available to capture and display proteins using this approach.
According to Marjeta Urh, research director at Promega, systems like HaloTag offer flexibility that standard fluorescent proteins cannot. “With a fluorescent protein, you are fixed to one color. But with HaloTag we offer many different ligands, which are comprised of different dyes—green, blue, red, permeable ligands and nonpermeable ligands that only label surface proteins. These are things you cannot do if you fuse your protein to a fluorescent protein—it’s always the same color.”
The system can even be used for protein purification, Urh notes, as Promega’s HaloTag vectors include a protease cleavage site between the tag and protein of interest allowing for the release of covalently captured proteins.
Users can prepare their own HaloTag fusions or have them custom-made through Promega. Alternatively, Promega has partnered with the Kazusa DNA Research Institute in Japan to provide more than 8,500 premade human protein-HaloTag fusions; search the gene database at http://www.promega.com/FindMyGene/search.aspx.
Active Motif’s LigandLink system couples a protein of interest to a modified bacterial dihydrofolate reductase, which binds trimethoprim. The company currently offers two trimethoprim-coupled fluorescent ligands, which bind noncovalently to the enzyme. But according to product manager Kyle Hondorp, the company has developed a covalent DHFR (dihydrofolate reductase) variant, as well.
“The covalent bonding increases the time frame [in which] you can do your analysis in vivo,” Hondorp explains. Previously, with the noncovalent, reversible form of the LigandLink reagents, media changes could disrupt labeling equilibria, “whereas with the covalent version, you can put it in there and it’s irreversible.”
New England Biolabs (NEB) also has been upgrading its tagging systems: SNAP-tag, CLIP-tag, ACP-tag and MCP-tag. Each of these systems uses a different tag and ligand; SNAP and CLIP are self-labeling—that is, they will couple to their ligand (O6-benzylguanine, in the case of SNAP) automatically—whereas ACP and MCP require an enzyme (phosphopantetheinyl transferase) to do the job.
According to NEB staff scientist Ivan Correa, systems like SNAP-tag offer several advantages vs. fluorescent proteins. For one thing, fluorescent proteins are always on, but tagged proteins only turn on when the fluorophore is added to the system. Users can also change the color of their proteins simply by selecting a different ligand. And as an added bonus, tagging systems use organic fluorophores, which are typically brighter than fluorescent proteins, he says.
“That’s one of the limitations of fluorescent proteins,” Correa adds. “Their photophysical properties are not as good as most organic dyes, in particular for highly demanding applications such as super-resolution microscopy and single-molecule imaging.”
NEB has about 20 dye-coupled ligands available for its four systems, including a dozen or so for SNAP. Some are cell-permeable, and others are extracellular only. Biotin labels are also available, as are the chemical “building blocks” themselves—kits that enable users to couple their own fluorophores or capture reagents to the SNAP-tag or CLIP-tag ligands.
The company continues to develop new applications, as well. For instance, Correa recently published a paper describing self-quenched SNAP-tag substrates that only fluoresce upon covalent linkage to the SNAP-tag itself. Ordinarily, fluorescent substrates are fluorescent all the time, meaning they must be removed from cells after incubation to lower the fluorescent background. But a self-quenching substrate doesn’t have that problem and can be used for “wash-free labeling” (imaging without removal of unbound substrate), according to the article.
Also described in the article is a faster form of the SNAP protein, called SNAPf, which demonstrates about 10-times faster kinetics than the previous versions. This new variant “allows for reduction of substrate concentration and incubation times,” according to an NEB application note on the protein.
In vitro tagging
The final class of protein tags labels proteins in vitro. In addition to the usual reactive chemistries, such as NHS (N-hydroxysuccinimide) esters, which are used to couple fluorophores to antibodies, for instance, several companies offer protein-tagging kits to enable multiplexed mass spectrometry analysis. These kits label proteins with isobaric tags—that is, a set of molecules that have the same molecular weight, so that they will appear identical to a mass spectrometer, but which fragment in different ways so they can be distinguished during tandem mass spectrometry.
Life Technologies’ iTRAQ reagents, for example, enable researchers to combine up to eight samples, each tagged with a unique isobaric reporter, in one tube for simultaneous (relative) quantification.
Thermo Fisher Scientific’s TMT line of isobaric reagents enables six-plex multiplexing. The company has updated its TMT line with the new iodoTMT reagents. According to technical product manager Monica Noonan, iodoTMT reagents tag only cysteine-containing peptides (whereas standard amine TMT reagents tag all peptides). Thus, researchers can use iodoTMT to specifically study that portion of peptides in which cysteine-specific modifications are of interest, such as S-nitrosylation. Thermo also offers an immobilized anti-TMT resin, which can be used to enrich for these low-abundance, cysteine-containing peptides that might otherwise be undetectable.
Ultimately, says Correa, which tagging approach is right for you depends on your application and an assessment of the pros and cons of each approach. How big is a given tag, for instance? Will it interfere with biological interactions? How many different ligands (or colors) does a given tagging system have, and are they compatible with your application?
“No tag is perfect,” Correa says. “There are always some limitations and advantages, and I think the best way to measure this is to look at the users and papers and the different applications.”
The image at the top of the page is from Promega's HaloTag® Technology.