TECHNOLOGY FEATURE

Pieces and parts 290 Blood work, with less sweat 290 Learning to love 2D 291 Box 1: Shaking the branches 290 Box 2: Looking for the hook-up 292 Taking on chromatography’s hard cases As protein chemistry becomes faster and more automated, a few tough problems in purification and analysis

methods still remain. Incremental evolution in chromatography and revolutionary developments in other techniques are finally starting to lower some of the field’s highest hurdles. Alan Dove reports.

Happy molecular biologists are all alike; every “many transmembrane proteins are highly .com/nature e protein chemist is unhappy in his own way. unstable after extraction and undergo con- Purifying nucleic acids, with their compara- formational changes and denaturation, even .natur tively uniform chemistry, simple building in the presence of detergents and stabilizing w blocks and predictable structures, is often agents.” tedious, but usually straightforward. In bio- Companies like Thermo Fisher are work- molecule purification, if you have seen one ing on new strategies, but for many scientists, http://ww

nucleic acid, you have seen them all. Proteins, the quickest solution may be to drive around

oup with their astonishing chemical and structur- the problem. “The chromatography tools r

G al diversity, are a different story. If you have haven’t changed very radically. The radical

seen one protein, you have seen one protein. ©iStockphoto.com/Gannet77 improvements are on the protein identifica- In the past few years, the rise of proteomics tion side,” says Speicher. In just a few years, Chromatography’s merry-go-round still in fashion. lishing from buzzword to bona fide discipline has he explains, mass spectrometers have evolved b simplified protein chemists’ lives consider- from distinguishing a few dozen proteins in Pu ably. Nonetheless, a few challenging classes of separation technologies recently1 (Box 2). a sample to distinguishing hundreds. The molecules still resist easy, large-scale purifica- New liquid chromatography columns from machines cannot quite handle a crude cell

Nature tion. In a cruel twist, these hard cases include PhyNexus are built to fit standard pipettors, lysate yet, but biochemists can now obtain

7 many of the proteins biochemists and drug reducing sample volumes and shortening run clean data from surprisingly dirty samples.

200 developers are most interested in studying: times. High-performance liquid chromatog- Speicher recalls for example a project for

© membrane- and organelle-bound signaling raphy (HPLC), long a workhorse of fast pro- which his group wanted to identify ligands molecules, post-translationally modified pro- tein purification, has been mechanized by GE that interacted with a specific membrane teins (Box 1) and scarce hormones diluted in Healthcare, with its automated AKTAxpress protein. Unfortunately, the researchers were complex biological fluids like serum. system. And Waters has upped the ante⎯or unable to get enough pure protein to analyze, Protein chemists have not given up, though, at least the pressure⎯even further, with just faint bands of potential ligands on a gel. and gradual improvements in chromatogra- ultra-performance liquid chromatography The team abandoned the project and took a phy and separations are chipping away at (UPLC) that promises to shorten processing different tack, “but today if we had that gel, some of the field’s longstanding problems. times and boost resolution. we’d just cut the band out and pop it in the Meanwhile, the rapid evolution of new analyti- “UPLC is something that's certainly on mass spec, and we’d have a protein identifica- cal technologies, especially mass spectrometry, the horizon,” says Gary Siuzdak, director tion, and if it turned out that it was a contam- is simplifying protein purification by allowing of the Center for Mass Spectrometry at the inant, we’d know it immediately, or if it was researchers to analyze dirtier samples. Scripps Research Institute. None of the new a specific ligand we’d get that identification technologies, however, are especially good at very early,” Speicher explains. The revolution will be ionized separating transmembrane proteins, a class Nonetheless, researchers need to be careful “If you talk about how [have] the chroma- that includes many of the most important not to overinterpret mass spectrometry data, tography and tools associated with chroma- signaling molecules in the cell. especially with impure samples. “You cannot tography improved, I think there have been The problem frustrates both biochemists compare it to what we have in nucleic acids. incremental improvements,” says David and product vendors. “Although we offer If you have a negative in a PCR, you can say Speicher, director of the proteomics facility a membrane-protein fractionation kit for ‘this mRNA is not expressed’. But if you can’t at the Wistar Institute. Although soluble- rapidly isolating these proteins from cells, we see something in the proteome, it may just protein purification has become relatively recognize that membrane proteins represent be below the limit of detection,” says Kai te straightforward, Speicher says membrane- a significant challenge for purification and Kaat, global business director for proteins protein purification has lagged behind. analysis,” says Greg Hermanson, director of at Qiagen. He adds that “from a proteomics Indeed, biochemists working on soluble technology at the Pierce division of Thermo technology standpoint, we still have a way proteins have been deluged with handy new Fisher Scientific. Hermanson adds that to go to really make a negative result not

NATURE METHODS | VOL.4 NO.3 | MARCH 2007 | 289 TECHNOLOGY FEATURE

But even with an expression system in hand, BOX 1 SHAKING THE BRANCHES purifying a membrane- or organelle-bound protein can be a nightmare. Biochemists For many proteomics projects, typically start by fractionating cells into their identifying the proteins in a complex cytosolic and membrane compartments, or compartment is only half the battle; using classical techniques like density-gradi- researchers also want to know exactly ent centrifugation. Unfortunately, the centri- how and when those proteins are post- fuge cannot distinguish one membrane from translationally modified. Tracking one another. of the most common and important As te Kaat explains, “If you look at mem- modifications, phosphorylation, has brane-bound proteins, what you usually get been a particularly tough challenge. in homebrew methods…is all membrane- methods Phosphoprotein antibodies or metal bound proteins, so this means you have affinity chromatography can isolate phosphorylated proteins, but neither endoplasmic reticulum, [and] you have the technique yields quantitative information organelles,” as well as plasma membranes. Worse, centrifugation techniques are notori- .com/nature about specific phosphorylation patterns. e Dendrimers shake the branches. (Courtesy of ously touchy, so results often vary from one Recently, though, researchers at the Ruedi Aebersold.) Institute for Systems Biology and the experiment to the next. .natur To address this problem, Qiagen now w University of Zurich solved this problem, using cleverly designed chemical reactions with off-the-shelf reagents2. To prove offers subcellular fractionation kits, includ- that the system works, the team tested it in human T cells. In one experiment, ing one specifically for plasma membranes. “What this kit actually delivers is a very, very http://ww they successfully identified all known tyrosine phosphorylation sites in the

immunoreceptor tyrosine-based activation motifs of the T-cell receptor CD3 chains, high enrichment of the plasma membrane

oup as well as previously unknown phosphorylation sites on 97 other proteins. with, for example, almost complete absence r

G ⎯ of endoplasmic reticulum molecules,” says te

The technique in which investigators use highly branched dendrimer structures to bind chemically modified phosphopeptides, then release and identify the peptides Kaat. The kits use proprietary buffers and col- by mass spectrometry⎯is not entirely new. "This goes back probably six or so umn resins, relying mostly on affinity purifi- lishing years that we started working on it," says Ruedi Aebersold, a professor at both the cation to isolate the desired components. b Institute for Systems Biology and the University of Zurich, and senior author on the Thermo Fisher also offers an extensive Pu paper. The original protocol was too hard for many researchers, though. "It was still series of kits and reagents for subcellular a very complicated procedure, and people who are not chemists had some difficulties fractionation. “We now allow researchers to

Nature doing it," says Aebersold. fractionate cells into cytoplasmic, nuclear,

7 After several simplifications, the new method is considerably more user-friendly. "I membrane, mitochondrial, lysosomal or per-

200 would think that everyone who has good experimental skills on the lab bench would oxisomal fractions,” says Hermanson, adding

© be able to do it," says Aebersold. He has discussed the possibility of commercializing that in many cases, the company's kits yield kits for the procedure, but with all of the reagents available from standard chemical intact organelles that can then be studied in vendors already, there is no need to wait. vitro or subjected to further purification. The New chemistry may soon come to the rescue for researchers working on proteins kits have been big sellers, and Thermo Fisher with other types of post-translational modifications, too. "Phosphorylation is one is working on introducing more. sub-implementation, and there's other stuff that's being worked on, [including] Even researchers who do not need highly carbohydrate-bound proteins," says Aebersold. pure proteins may find organelle fraction- ation useful. “Another important aspect to subcellular fractionation is that it provides something I can’t see but something that through a series of columns and other sepa- important physical context to proteomics isn’t there.” rations. Although this outline has remained analysis. It not only facilitates the identi- essentially unchanged for decades, a few fication and characterization of proteins, Pieces and parts recent improvements can accelerate the but it generates information regarding Although sophisticated mass spectrometry process. For example, many suppliers offer where they naturally exist within a cell,” says systems may allow proteomics researchers expression vectors that can be shuttled easily Hermanson. to avoid some of the challenges of classical between Escherichia coli and more complex purification, even mass spectrometrists agree organisms, drastically simplifying⎯or even Blood work, with less sweat that this approach will not work for everyone. eliminating⎯the tedious process of vector In addition to separating cells into their “If you’re ultimately looking for structural or construction. individual components, biochemists have really detailed information on a particular “Companies are selling libraries of cDNAs also struggled with the problem of frac- protein, ultimately you’re going to have to go in expression-ready systems, so if you’re lucky tionating biological fluids, especially serum. the purification route,” says Siuzdak. enough to be working with a protein or a set For clinical research and new diagnostic The traditional way to purify a difficult of proteins where they’re commercially avail- strategies, measuring changes in blood- protein to homogeneity is to develop a sys- able, you don’t even have to make the expres- borne hormone levels is often the order of tem that overexpresses it, then enrich it sion vector, you just buy it,” says Speicher. the day. Unfortunately, these molecules are

290 | VOL.4 NO.3 | MARCH 2007 | NATURE METHODS TECHNOLOGY FEATURE

usually present in vanishingly small quanti- umns, I would be skeptical that companies present in very low concentrations may sim- ties, immersed in a stew of vastly more abun- are going to,” says Speicher. ply vanish. When looking for relative changes dant but much less interesting proteins. Load For those working on human serum, in the levels of signaling proteins, however, serum onto most liquid chromatography Speicher and Siuzdak say the columns are immunodepletion can work very well. columns, for example, and the result will be quite useful, provided one keeps their limita- a column clogged with albumin. tions in mind. Like most liquid chromatog- Learning to love 2D To answer this challenge, tool makers have raphy techniques, immunoaffinity absorbs No matter what kinds of proteins a labora- been rolling out specialized resins that bind some molecules nonspecifically, so the meth- tory studies, if the goal is a proteomic sur- common proteins like albumin. Overall, the od is not strictly quantitative, and proteins vey, sooner or later two-dimensional (2D) products have gotten good reviews. “One of the things that we’re doing is looking for plasma biomarkers, and one of the incre- methods mental improvements there…was the first commercial products which utilized immu- noaffinity to deplete abundant proteins,” says Speicher, adding that “we’ve found that to be .com/nature e greatly enabling.” The new immunodeple- tion columns have been available since about .natur 2003, and companies like Sigma-Aldrich and w Agilent now offer extensive immunodeple- tion resin product lines. “Because of the success of removing the http://ww

6 top proteins, there’s been interest now in

oup removing the top 15 or so proteins, [because] r

G when you reduce these top-level proteins, it

allows you to delve further into the pro- teome,” says Siuzdak. lishing The new resins are not a panacea for plas- b ma proteomics, though. In particular, man- Pu ufacturers have generally favored products that deplete human serum. “Mouse immu-

Nature nodepletion has really lagged behind,” says

7 Speicher, adding that “there are only a few

200 products, and they don’t work as well as the

© human products do, and there seems to be little commercial interest in developing that.” Researchers using non-mouse animal models may be entirely out of luck. “That’s going to be very much a niche, and given the expense of developing these immunodepletion col-

Mitochondria isolation kit for tissue

One of the available subcellular fractionation kits focusing on mitochondria. (Courtesy of Pierce- Thermo Fisher.)

NATURE METHODS | VOL.4 NO.3 | MARCH 2007 | 291 TECHNOLOGY FEATURE

BOX 2 LOOKING FOR THE HOOK-UP

It’s a common chore in protein chemistry: link a purified protein The company offers linkers for proteins, peptides and to a solid support or to another protein. For many researchers, oligonucleotides, which a small collection of corporate and the available techniques⎯biotin-avidin or thiol-based cross- university customers have been using for everything from vaccine linking kits⎯are adequate, but occasionally annoying. The development to immunological PCR. "We do things routinely reaction conditions can be difficult to control, and it is hard to that people would kind of think 'wow, I didn't think that was quantify how well or poorly the cross-linking went. possible'," Schwartz says. For example, he explains that the At least one company thinks it found a better way. "We've company can attach one linker compound to an oligonucleotide developed a new conjugation chemistry based on a stable Schiff and another to a peptide during synthesis. Mixing the two base, engineered to cross-link any biomolecule to a surface or any biomolecules together produces an oligo-tagged peptide that is methods other biomolecule more easily and with less harsh conditions," says stable at PCR temperatures. David Schwartz, chief scientific officer of Solulink Biosciences. Still, he concedes that the technology has not caught on In Solulink's approach, the reagents are not only gentler than widely yet, possibly because protein chemists have become traditional thiol-based systems, they are also colorimetrically accustomed to the limitations of traditional conjugation .com/nature

e labeled, so researchers can quantify how efficiently they have chemistry, and because Solulink, with only a half-dozen cross-linked their targets. "The reproducibility is exquisite, employees, is not well known. “We’re trying to get to the mass

.natur because you really understand what's going on by all this market, [but] we’re just having trouble getting the message w traceability and quantitation," says Schwartz. out,” says Schwartz. http://ww

gel electrophoresis will come up. Indeed, in Several companies have tried to automate reference samples on a 2D gel, researchers

oup the 12 years since the term “proteomics” first 2D gel handling, but that approach has been scan the gel for the fluorescent labels. The r

G appeared, separating all of a sample’s pro- slow to catch on. “Most people who are doing result is an experimental pattern and a refer-

teins on a 2D gel has been a hallmark of the 2D gels are still doing them the old manual ence pattern, the latter providing alignment field. For just as long, vendors and engineers way,” says Speicher. Chip-based and column- points for comparisons between gels. lishing have been promising that this troublesome based alternatives are also in development, “So if I have a gel-to-gel variation…that’s b technique is on the verge of being replaced but so far, none have managed to supplant going to happen with the reference as well as Pu by something better. the gels. experimental samples,” says Speicher, who The motivation to replace 2D gels is The technology has evolved somewhat in adds that “my experience in talking to col-

Nature strong. The standard technique requires recent years, though. In particular, the dif- leagues is that…an increasing percentage of

7 pouring, running and slicing large poly- ferential in-gel electrophoresis (DIGE) tech- people doing 2D gels are migrating to that

200 acrylamide gels manually, with ample nique is helping many groups overcome the approach.”

© room for errors and variation at each step. legendary reproducibility problems of 2D Tool makers are still working on ways to Results commonly differ widely from gels. In DIGE, the investigator pools aliquots simplify 2D gels or eliminate them entirely, one experiment to the next. Nonetheless, of all the experimental and control samples but they concede that quick solutions are 2D gels have established a durable niche. and labels them with one fluorescent dye, cre- not just around the corner. “2D gel separa- “The 2D gel approach isn’t going to be ating a reference sample. Another aliquot of tions still offer important information on going away anytime soon,” says Siuzdak, each experimental and control sample is then protein modification, such as phosphoryla- echoing a widespread sentiment among labeled individually with another dye. After tion, glycosylation and peptide cleavage that researchers. running the individually labeled and pooled is difficult or impossible to evaluate by [mass spectrometry],” says Hermanson. That theme is typical of the remaining ‘difficult’ problems in protein chemistry. Although researchers still hope the field will have a big break, in the meantime they’re Qproteome optimistic about the incremental evolution mitochondria of present techniques. “The combination of isolation kit these new technologies is…going to far sur- pass what we’ve done in the past, and I think we’re going to see a lot of very interesting things coming out,” says Siuzdak.

1. Chapman, T. Nature 434, 795–798 (2005). Mitochondrial proteins 2. Tao, W.A. et al. Nat. Methods 2, 591–598 (2005). Non-mitochondrial proteins Unknown location Alan Dove is a science writer based in the Mitochondria protein enrichment results using the QIAGEN separation kit. (Courtesy of QIAGEN.) New York area ([email protected])

292 | VOL.4 NO.3 | MARCH 2007 | NATURE METHODS TECHNOLOGY FEATURE

SUPPLIERS GUIDE: COMPANIES OFFERING PRODUCTS FOR SEPARATION AND CHROMATOGRAPHY Company Web address Agarose Bead Technologies http://www.abtbeads.com Agilent http://www.agilent.com Amersham Biosciences/GE Healthcare http://www.amershambiosciences.com Applied Biosystems http://www.appliedbiosystems.com Beckman Coulter http://www.beckman.com BD Biosciences Clontech http://www.bd.com BioChrom Labs http://www.biochrom.com Bio-Rad http://www.bio-rad.com

methods BioScience Beads http://www.bioscience-beads.com Brinkman http://www.brinkmann.com Caliper Life Sciences http://www.caliperls.com Ciphergen http://www.ciphergen.com

.com/nature deltaDOT Ltd http://www.deltadot.com e EMD Biosciences http://www.emdbiosciences.com

.natur Eppendorf http://www.eppendorf.com w Fluidigm http://www.fluidigm.com GenScript http://www.genscript.com Gyros http://www.gyros.com http://ww

Invitrogen http://www.invitrogen.com

oup Mallinckrodt Baker http://http://www.mallbaker.com r G

MDS http://www.mdsintl.com Millipore http://www.millipore.com NCSRT http://www.ncsrt.com/ lishing b New England BioLabs http://www.neb.com Pu

Paragon Bioservices http://www.paragonbioservices.com PhyNexus http://www.phynexus.com Pierce/Thermo Fisher http://www.thermo-fisher.com Nature

7 Polysciences http://www.polysciences.com

200 QBIOgene http://www.qbiogene.com

© Qiagen http://www.qiagen.com Sartorius http://www.sartorius.com SEPIAtec http://www.sepiatec.com Sigma-Aldrich http://www.sigmaaldrich.com Solulink http://www.solulink.com Stratagene http://www.stratagene.com Tecan http://www.tecan.com Trenzyme http://www.trenzyme.com Waters http://www.waters.com Whatman http://www.whatman.com

NATURE METHODS | VOL.4 NO.3 | MARCH 2007 | 293