Serving a Smorgasbord of Phospho-Specific Antibodies

Protein post-translational modifications (PTMs) generally, and phosphorylation in specific, are crucial for understanding cellular functioning. Researchers have used different ways to look at phosphorylated proteins, from phosphorus-32 radioisotope labeling to mass spectrometry. But the lion’s share goes to affinity reagents, especially anti-phospho antibodies. These come in various flavors—monoclonal and polyclonal, modification-specific, site-specific and motif-specific—and are used in a range of downstream applications, from ELISA to Western blotting to immunohistochemistry.

Production of phospho-specific antibodies follows essentially the same recipes useful for making other antibodies … with perhaps an extra dash of salt here, a little less pepper there and maybe an extra herb to spice things up. Here are some particularly pertinent ways to season a phospho antibody sauce—including taste tests to assure that it’s ready to serve.

Site-specific phospho antibodies

There are different flavors of antibodies raised against protein modifications such as phosphorylations. Take the Akt (protein kinase B or PKB) protein as an example. Akt is phosphorylated on a number of specific sites, resulting in its activation that may be related to specific regulatory events. Many companies offer site-specific antibodies that detect such specific events, recognizing the phospho amino acid within the context of the surrounding residues. This can uncover a valuable clue for understanding cellular events.

A typical immunogen for generating such site-specific phospho antibodies would be a peptide hapten corresponding to the epitope of choice and containing a centrally located phospho amino acid residue, conjoined to a carrier protein, such as keyhole limpet hemocyanin (KLH). Limiting the hapten to at most six residues on either side of the phospho amino acid helps focus the immune response rather than allowing more dominant epitopes to overwhelm the response [1].

According to Matt Baker, director of R&D and business development for Thermo Fisher Scientific’s antibody and immunoassay business, “in general, you can make a specific antibody to just about any phosphorylation site in the proteome, with the exception of some highly conserved or completely conserved motifs. Antibodies to these motifs tend to cross-react with conserved motifs on other proteins.”

As with any immunization, it’s important to assure that the chosen hapten is specific to the site and the protein of interest to avoid any cross-reactivity of the antigen. This can sometimes be difficult, because “when you’re making a site-specific antibody, you’re handcuffed,” says Anthony Couvillon, scientific marketing project manager at Cell Signaling Technology (CST). For example, the surrounding residues may turn out to be nonimmunogenic or proline rich (which will cause a lot of structural kinks in the peptide). Potential options include shifting the phospho amino acid to one end of the peptide, trying to minimize the number of problematic amino acids or using derivatives of the natural amino acid to mitigate its impact on the peptide. But these, in turn, introduce additional layers of validation to assure specificity.

Numerous algorithms are available to help design immunogens, but “their success rate is not particularly better than chance,” says Michael Browning, CEO of PhosphoSolutions, which relies on a spreadsheet of 1,000 or so antibodies the company has tried to make. PhosphoSolutions ranks these antibodies on a scale of zero to three, to learn from both successes and failures. “We use that database as a sort of guide to decide where and what pieces of the peptide adjacent to the phospho site that we’ll include in the antigen, and which we won’t include,” he explains.

Numerous algorithms are available to help design immunogens, but “their success rate is not particularly better than chance

Another option is to link multiple copies of the peptide into a branched structure with defined antigenic, chemical and physical properties, like the multiple antigen-presenting system (MAPS) [2], says Birte Aggeler, director of antibody development for Bio-Techne, whose portfolio includes antibodies from the R&D Systems and Novus Biologicals brands. The branched peptides make it possible for four to eight copies to be of the antigen epitope and eliminate the need for a carrier protein. In addition, this approach affords the opportunity to add other motifs driving “dendritic cell-recognition sequences, to make the sure they really get pulled into the system to mount an immune response.”

A panoply of pan antibodies

Other phospho-specific antibodies can detect the phosphorylated residue free of context: “They don’t care what the amino acids are surrounding the site—they’re actually just detecting the modified amino acid. Those are useful for generically detecting the tyrosine phosphorylation status or the threonine status of proteins in a lysate,” for example, notes Couvillon. “They’re also handy for proteomics studies, because you can use those antibodies to enrich the amino acids you’re interested in studying and look for modulation based on stimulation, drug treatment or disease state.”

There are also antibodies that will recognize a phosphorylated amino acid motif. “Akt, for example, prefers to phosphorylate serine residues, but only when there are arginine residues N-terminal, and CST has pioneered making antibodies against that type of sequence,” says Couvillon. The company’s website notes that the “Phospho-Akt Substrate (RXRXXS*/T*) (23C8D2) Rabbit mAb recognizes endogenous proteins containing phospho-Ser/Thr preceded by Arg at positions -5 and -3 in a manner largely independent of the surrounding amino acid sequence” [3]. Substrate motif antibodies enable researchers to detect sites based on the kinase—such as Akt, PKC, ERK, etc.—they are studying, Couvillon explains. Polyclonal motif antibodies can be made by immunizing with a library of motif-conforming haptens. Multiple monoclonal antibodies—each specific to a specific epitope—can also be combined into a single reagent.

These phospho-specific antibodies are sometimes referred to as “pan antibodies.” But the term is not consistently applied, so it can be ambiguous for people like Aggeler, who says, “The use of the term ‘pan antibodies’ can vary from company to company.” The term is also sometimes used for antibodies that will recognize a site whether or not it is phosphorylated. A third usage of “pan” is for antibodies that will recognize different isoforms of a protein—so a pan-Akt antibody would detect Akt1, Akt2 and Akt3, for example. This is one more example of the ongoing antibody-nomenclature standardization discussion among researchers and vendors that needs to be addressed to assist researchers in identifying the correct reagents to use in their experiments.

But is it valid?

There are many ways to create antibodies, ranging from traditional immunizations to recombinant, phage display and in silico methodologies. There are some advantages and disadvantages to monoclonal antibodies, pools of monoclonal antibodies (what Thermo Fisher calls oligoclonals, and CST terms monoclonal antibody mixes) and polyclonal antibodies. And antibodies need to be screened, selected and validated for particular downstream applications. In these respects, phospho-recognizing antibodies are not significantly different from those with other specificities and must undergo the same scrutiny and verification by researchers and vendors

Yet validating a phospho antibody—and modification-specific antibodies, in general—may be easier in the sense that “you have a binary system of some type,” says Couvillon. “You can use activators or inhibitors. You can block with the phospho-specific peptide vs. the nonphospho peptide. You can use phosphatase treatments. You can make site mutants (such as switching a serine to alanine to show that the antibody no longer works). There are all sorts of tricks you can use. So to a certain degree making a go/no-go call on a phospho project is a little bit easier than on a total antibody project, mostly due to it being a much easier model system.”

Are serine, threonine and tyrosine all there are?

Nine of the 20 naturally coded amino acids can be phosphorylated. Yet nearly all phospho antibodies recognize phospho-serine or -threonine (or both), or -tyrosine. One of the reasons for this is that these phospho-ester (P-O) modifications are stable in blood (allowing the immunogen to be recognized before it is degraded), whereas modifications like phospho-histidine use the far more labile phosphoramidate (P-N) linkage [4] . Recently, Princeton’s Tom Muir (professor in the Chemistry department) designed stable phospho-histidine analogs that have been used separately by his team and Salk Institute’s professor of Biology, Tony Hunter to produce antibodies recognizing the phospho-histidine post-translational modification.

Commercial phospho-histidine antibodies are available from MilliporeSigma. “The problem with phospho-his is not really the ability to make an antibody, it’s the ability to stop the modification from being removed upon cell preparation,” says Couvillon. “CST is planning on releasing [our phospho-his antibody] as a kit with all the reagents necessary to get a successful Western blot.”

It won’t stop there, either. Researchers and vendors are working on other phospho antibodies, such as anti-phospho glutamate. If other modifications are found to be relevant to cell signaling, cancer, brain research or any other topic of interest, be sure that reagents are on their way.

References

[1] Arur, S, Schedl, T, “Generation and purification of highly specific antibodies for detecting post-translationally modified proteins in vivo,” Nat Protoc, 9:375-95, 2014. [PMID: 24457330]

[2] http://www.cellsignal.com/products/primary-antibodies/phospho-akt-substrate-rxrxxs-t-23c8d2-rabbit-mab/10001

[3] Zhang, F, Lu, Y-J, Malley, R, “Multiple antigen-presenting systems (MAPS) to induce comprehensive B-and T-cell immunity,” Proc Natl Acad Sci, 110:13564-13569, 2013. [PMID: 23898212] 

[4] Fuhs, SR, et al., “Monclonal 1- and 3-phosphohistidine antibodies: new tools to study histidine phosphorylation,” Cell, 162:198-210, 2015. [PMID: 2614597]