to BioTechniques free email alert service to receive content updates.
Tearing the Top Off ‘Top-Down’ Proteomics
Jeffrey M. Perkel, Ph.D.
BioTechniques, Vol. 53, No. 2, August 2012, pp. 75–78
Full Text (PDF)

Imagine a group of cars are obliterated in a parking lot explosion. Based on license plates and hood ornaments, it's easy to determine the type and number of cars that were involved. But to go beyond this, to determine which starter corresponds to which vehicle or which car had the fuzzy dice hanging from the rear-view mirror, that's a tougher nut to crack.

In a sense, this is what happens in the proteomics workflow known as “bottom up,” in which protein mixtures are digested into peptides, fractionated, and then sequenced and characterized in a tandem mass spectrometer. From these data researchers can determine which proteins were present in the sample, and whether or not they were phosphorylated.

But suppose one protein actually contained phosphorylation sites at either end of its sequence. This protein can exist in any of four states: completely unphosphorylated, doubly phosphorylated, or phosphorylated at either site, explains Detlev Suckau, head of MALDI Applications and Proteomics Research & Development at Bruker Daltonik in Bremen, Germany.

“A bottom-up approach does not distinguish between the four forms,” says Suckau. It can identify them, but not determine how frequently or under what conditions they appear. Nor can it determine whether those two modifications coexist on the same protein at the same time, or are mutually exclusive.

To answer those questions, researchers need the so-called “top-down” approach. Top-down proteomics explores proteins without digesting them—that is, from the “top down.” Because top-down analyses consider intact proteins rather than peptide pools, they can distinguish our four hypothetical isoforms based on their different molecular weights—not to mention closely related family members and splicing isoforms that bottom-up strategies often cannot differentiate.

Researchers have used top-down approaches on individual proteins or protein families, and are pushing toward the proteomics scale, where it can interrogate hundreds or even thousands of proteins in a single sample. The approach even has a dedicated research alliance: the Consortium for Top Down Proteomics was formed in March 2012 with the goal of “promot[ing] innovative research, collaboration and education accelerating the comprehensive analysis of intact proteins.”

But making that transition won't be easy. The technique challenges mass spectrometrists on every level—from protein separation, to the mass spectrometry itself, to bioinformatics.

A Question of Resolution

The top-down workflow clearly has the upper hand when it comes to proteome analyses: If post-translational modifications are molecular toggles that tweak protein activity, then it makes sense that researchers would want to study the different isoforms that wax and wane as cells develop, respond to stimuli, and die.

Yet because it requires separating and sequencing large protein isoforms differing by just a methyl group or two, the workflow demands the highest of high-end instrumentation, the kind of mass specs that can weigh molecular masses out to two or three decimal places.

Ljiljana Paša-Tolić, an investigator in the Consortium for Top Down Proteomics and lead for mass spectrometry at the Environmental Molecular Sciences Laboratory, uses a Fourier-transform ion cyclotron resonance (FTICR) mass spectrometer to drive her top-down analyses.

Driven by massive, supercooled, super-conducting magnets, FTICRs are like the Lamborghinis of mass spectrometry, offering the very highest resolution and mass accuracy. “In FTICR all of the performance parameters scale with the magnetic field linearly or quadratically,” explains Paša-Tolić. “So, the higher the field, the better performance you have.”

Currently, her lab uses a 15-Tesla instrument from Bruker Daltonics, with which she can routinely study LC-separated proteins of 70-80kDa. But using a 21-Tesla instrument, currently under construction with Department of Energy funding, Paša-Tolić estimates they may be able to push that to 150 kDa. “That would then enable us to assess almost the whole human proteome,” she says.

Consortium member and top-down evangelist Neil Kelleher of Northwestern University, who has long advocated FTICRs for top-down proteomics, recently switched to a Thermo Fisher Scientific Orbitrap Elite, a high-resolution, high-mass accuracy instrument that's in many ways a cross between a traditional ion trap and an FTICR, but with slightly lower resolution.

  1    2    3    4