Over the last decade, bottom-up, peptide-based techniques have dominated the field of proteomics. In recent years, however, technological advances and growing appreciation of the importance of protein isoforms and post-translational modifications have led to increased interest in top-down, intact protein work.
Bottom-up approaches are still by far the most common, but new tools have enabled workflows that are moving intact protein research out of the realm of specialists and into more widespread use.
Instrumentation advances have been key to increasing accessibility of top-down techniques, said Ben Garcia, assistant professor of molecular biology at Princeton University.
"When you ionize an [intact] protein it can have anywhere from 10 to 50 charges on it, so it takes a really high resolution to determine the charge state, and if there are any variants or modifications of the sequence," he told ProteoMonitor.
For this reason, he said, researchers have typically used Fourier transform ion cyclotron resonance mass spectrometry for top-down work, "and FT-ICR instruments are very specialized instruments."
Thermo Electron's 2003 release of its ion trap-FT hybrid mass spectrometer, the LTQ-FT, marked the beginning of top-down's migration to less complicated machines, Garcia said, noting that "it was a breakthrough because it put the two together so that basically if you could run an ion trap you could run the ion trap hybrid."
Since then, he said, releases like Thermo Fisher Scientific's Orbitrap machines have continued to increase top-down proteomics' accessibility. Top-down researchers are also interested in exploring the use of TOF mass specs such as those produced by Agilent, Bruker, and AB Sciex, Garcia added, saying that "time-of-flight analyzers are interesting because they probably have the highest mass range of most instrumentation, so you could theoretically look at very large proteins" with them.
Another potential interesting technology for top-down proteomics is ion mobility, he said. This feature, currently available on Waters' Synapt instruments, allows researchers to further separate ions based on their mobility in a carrier buffer gas, and could be useful in analyzing protein isomers, Garcia suggested.
In terms of the FT-ICR instruments traditionally used for top-down work, Bruker remains the primary player, said Northwestern University professor of chemistry Neil Kelleher. The company has "taken a fairly aggressive approach to that fairly small but high-profit-margin market," he told ProteoMonitor.
Kelleher is the developer of ProSight PC, an informatics package for top-down proteomics-data analysis that is sold by Thermo Fisher. According to Garcia, such software is probably now more key to the field's continued development than instrumentation advances.
"I think you're going to see a lot more people attempting top-down proteomics because it's just become a lot more readily accessible, but now I think the problem is going to be data analysis," he said. "Now that it's becoming possible to do proteome-wide top-down proteomics, it's going to be incredibly important to get the software to analyze that data robustly."
Software "has been one of the limiting issues," Kelleher agreed. "ProSight itself is pretty complicated. It's got five different search modes in it. And there's a little bit of a barrier there because of the complexities in top-down that are different from bottom-up in terms of database searching."
In particular, Kelleher noted, top-down informatics requires error-tolerant searching to account for mass discrepancies caused by proteolytic events or polymorphisms. The technique, he said, allows researchers to "identify with great confidence a protein even though it has a mass discrepancy from what the best database candidate is."
"The larger population that's doing proteomics doesn't do error-tolerant searching," he said. "They search for peptides that match their data. But top-down proteomics [searching] really needs to be error tolerant."
Sample prep has also posed a challenge for top-down research, said Andreas Huhmer, proteomics marketing director at Thermo Fisher Scientific. As opposed to peptides, which can be separated using liquid chromatography, intact proteins are usually separated via some form of electrophoresis. This step typically requires the use of detergents that must be removed prior to mass-spec analysis – a step, Huhmer told ProteoMonitor, that has proved a bottleneck.
At last year's American Society for Mass Spectrometry conference, Thermo Fisher introduced its Pierce detergent removal spin columns, which filled a key gap in the company's top-down workflow, Huhmer said.
"As soon as that technology became available, a lot of people suddenly realized now [they] could do intact proteins because the bottleneck always was how … [to] remove the detergent [from electrophoresis] but still keep the proteins in solution," he said.
Currently, Huhmer noted, Thermo Fisher offers all the components needed for intact protein workflows with the exception of the front-end electrophoresis.
Top-down "is absolutely a growth opportunity for proteomics and for the vendors that sell into that space," he said. "I think we're putting a lot more effort into really promoting [the company's top-down] workflow as we see it."
The fundamental idea motivating top-down proteomics is that an understanding of protein variants and isoforms is key to basic biological and clinical research and that the best — and in some cases only — way to gather such information is through the study of intact proteins.
Bottom-up methods can provide information on what proteins are present at what levels in complex samples like plasma, but because they use peptides a surrogates for full-length proteins they aren't as good for discerning what specific isoforms are present, and in what proportions.
"With bottom up you're trying to go from a peptide to the full-length protein to the different forms of the full-length protein that actually do the work in your body," said Randall Nelson, a researcher at Arizona State University's Biodesign Institute. "For instance – take a protein that has four post-translation modifications on it. To try to quantify all four of those simultaneously can't be done in a surrogate mode."
Such isoform information could be the key to protein biomarker-based diagnostics, he suggested. Through proteomics firm Intrinsic Bioprobes, of which he is president and founder, Nelson is working on a variety of tests for diagnostically useful isoforms of proteins, including parathyroid hormone (PM 07/16/2010) and certain low-density lipoproteins.
At the beginning of 2011, the company began commercial production of a cisplatin-C assay designed to detect kidney damage. It is currently offered at the University of Medicine and Dentistry of New Jersey's CLIA-certified laboratory.
Another primary focus of top-down research is histone proteins, which are key to DNA transcription. These proteins are highly modified and it's thought that a better understanding of these patterns of modification might lead to insights regarding gene regulation.
"If we do a bottom-up analysis, we can identify all the modifications on a given peptide … Some of these modifications are associated with gene activation, some of them are associated with gene silencing, [and] what we really want to know if which of these modifications are seen together," Garcia said. "Maybe there's a modification code on proteins, and if you're going to try to break the code, you have to know the whole code, not just parts of it."
"There's been an enormous explosion of literature around histones where this type of [top-down] work is absolutely necessary," Huhmer said. "So I think this is basically where you see the first step of top-down going more towards the mainstream."
"Top down has been the domain of experts, but if you do a PubMed search and look at the numbers you can see a steady climb in the number of citations," Kelleher said.
Indeed, a PubMed search of co-occurrences of "top down" and "proteomics" reveals no citations in 2001, two in 2002, 19 in 2005 and 50 in 2010.
"We're ... not at the crest of the wave, but it's building," he said.
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.