There’s been considerable excitement of late about the clinical potential of next-generation DNA sequencing. After all, many diseases, such as cancer, sickle cell anemia and Huntington’s disease, are encoded in mutations or variations in genomic DNA. Yet these diseases physically manifest themselves in the presence and absence of proteins and small molecules. Thus, focused efforts are on cataloging and characterizing these molecules to identify molecular indicators (read: biomarkers) of disease, prognosis, therapeutic response and progression.
At the protein level, such studies are conducted using mass spectrometry, and they can be done in either of two ways. The simplest and most widely used approach is “bottom-up proteomics.” Here, protein extracts—representing, say, cancerous and normal tissues—are broken down into peptides with a proteinase, chromatographically separated and then analyzed in a mass spectrometer. The goal is to identify peptides (and thus, the proteins they represent) whose abundance or post-translational modification (PTM) changes as a result of disease or treatment.
As described in a recent review, there are at least three challenges with implementing the bottom-up approach [1]. First, closely related but distinct proteins (two members of a protein family, for instance) may have peptides in common and thus be indistinguishable in a bottom-up workflow. Secondly, informative peptides may slip through the cracks and go undetected. Lastly, because the proteins are digested prior to analysis, any knowledge of which peptides were physically linked is lost. As a result, researchers are unable to determine if, for instance, two post-translational modifications co-occur in the same molecule.
The alternative approach is “top-down proteomics”. In the top-down method, intact proteins are analyzed directly, then fragmented and analyzed in the mass spectrometer.
“The advantage of top-down is you’re looking at the intact molecule, and in particular, the modified forms of those proteins,” explains Jim Langridge, director of health sciences at Waters Corporation. Many informative proteins, he says, are phosphorylated or glycosylated, and those modifications often affect protein function. “That is a lot easier to study at intact-protein level than at the peptide level.” (One recent study, for instance, identified 74 protein forms of human histone H4, three of which included novel PTM combinations [2].)
Although the top-down approach has some challenges, its popularity and applications are growing. Here we review some key features and tools to consider if adopting this approach for your proteomic studies.
Separation anxiety
Traditionally, proteomics researchers have used bottom-up workflows to home in on candidate molecules and top-down for validation and deeper characterization. It is possible to use top-down in a discovery mode—see, for instance, the work of Neil Kelleher at Northwestern University—but few actually do so.
For one thing, top-down work is largely restricted to small to mid-size proteins. Furthermore, says John Yates III, the Ernest W. Hahn Professor of Chemical Physiology and Molecular and Cellular Neurobiology at The Scripps Research Institute in La Jolla, California, bottom-up is higher throughput. That makes the method more amenable to biomarker discovery studies, which often involve large patient cohorts.
That throughput hit occurs partly because peptides are easily separated on chromatographic columns, but intact proteins are not. “There is no single condition that gives all the proteins and keeps them in solution,” says Andreas Huhmer, director of marketing for life sciences mass spectrometry at Thermo Fisher Scientific. Many researchers thus fractionate samples under a variety of conditions to capture different slices of the proteome.
In one recent study, for instance, Kelleher identified 1,220 proteins—some as large as 80 kDa—and “over 5000 proteoforms”using a multipronged subcellular fractionation and enrichment protocol involving solution isoelectric focusing, GELFrEE (gel-eluted liquid fraction entrapment) separation and liquid chromatography [3].
One chromatographic approach that may simplify top-down separations, Huhmer says, is using so-called “monolithic columns.” Unlike traditional chromatography columns, which separate molecules on beds of packed beads, monolithic columns provide a large “unstructured surface, and that allows you to generate very fast exchange kinetics to get sharp peaks and good separation, which you cannot achieve with traditional reverse-phase approaches.”
Yates described an alternative strategy in 2014 [4]. Called CESI (capillary electrophoresis-electrospray ionization), the method separates proteins via capillary electrophoresis (CE) prior to mass spec injection, rather than traditional chromatography.
“For peptides, CE doesn’t work very well; there’s lots of compromises you have to make,” Yates explains. “But top-down with CE works really well,” providing a 100-fold increase in sensitivity, he says. Applying the method to the Pyrococcus furiosus proteome, a standard proteomics exercise, Yates’ team “identified 134 proteins and 291 proteoforms” using just 270 ng of material [4].
Higher res required
The other challenge of top-down proteomics is the instrumentation required. High mass accuracy, sensitivity and resolution are needed to distinguish the many protein charge and isotopic states that intact-protein analysis generates. Although peptides may be charged to +1, +2 or +3, intact proteins can carry dozens of charges, each of which produces a signal.
Many studies—including Yates’—are accomplished using a Thermo Scientific Orbitrap-based mass analyzer. Some use top-of-the-line Fourier transform ion cyclotron resonance (FT-ICR) instruments; others use more accessible quadrupole/time-of-flight (TOF) hybrids, such as Waters’ Synapt systems.
The Synapt G2-Si has lower mass accuracy and resolution than Orbitrap and FT-ICR instruments (for infusion experiments)—it is rated at 50,000 resolution, compared to 450,000 on some Orbitrap systems. But with their built-in ion mobility separation (IMS), Synapt mass spectrometers can, for instance, differentiate proteoforms and conformational isomers that otherwise may be indistinguishable by conventional mass spectrometry.
“There’s more than one way to create resolving power,” says Gordon Kearney, senior manager of mass spectrometry product management at Waters. “One way is from mass resolution; another way is from a good mass-resolving mass spectrometer, like a TOF, coupled with IMS. We see that as a key enabler for the analysis of complex samples.”
Shattering expectations
Another technical consideration in top-down proteomics occurs after the intact proteins make it into the mass spectrometer. To make any sense of the resulting spectra, researchers must isolate individual proteins and systematically break them down to identify their sequences and post-translational modifications. But that’s easier said than done, says Yates. “As proteins get bigger and bigger, it’s harder to get them to fragment.”
Thermo Fisher’s Orbitrap Fusion mass spectrometer offers three fragmentation approaches, Huhmer says—collision-induced dissociation (CID), higher-energy collisional dissociation (HCD) and electron-transfer dissociation (ETD). In CID and HCD, Huhmer explains, “you just shake the molecule really hard,” causing it to shatter. Oftentimes, PTMs break off as well, making it difficult to figure out where a modification occurred. In ETD, by contrast, a charged molecule transfers an electron to the protein, causing the backbone to break while leaving PTMs largely unaffected. (The Synapt G2-Si also offers CID and ETD fragmentation.)
Another option is surface-induced dissociation (SID), available through Waters in collaboration with Vicki Wysocki at Ohio State University. In SID, Langridge says, molecules “take a glancing blow off a surface,” causing fragmentation processes that reveal such details as the stoichiometry of protein complexes.
One emerging alternative is UV photodissociation (UVPD), developed by Jennifer Brodbelt at the University of Texas at Austin. Among other things, Brodbelt has used the technique to study the “size and linkage type of polyubiquitin chains” via top-down [5].
According to Brodbelt, the method, which uses laser energy to induce fragmentation, “causes extensive fragmentation of intact proteins, thus leading to very high sequence coverage,” without the loss of PTMs. At the moment, she says, the method is restricted to proteins smaller than 40,000 Da, but the lab is working to address those issues.
Yates, for one, has been sufficiently impressed to purchase a laser to implement the method on his own instruments, pairing the method with ETD. “It’s still research-scale, but it looks incredibly promising,” he says.
The same could be said of top-down proteomics itself, with its exciting results and promising potential for novel biomarker characterization.
References
[1] Catherman, AD, et al., “Top down proteomics: Facts and perspectives,” Biochem Biophys Res Commun, 445:683-93, 2014. [PubMed ID: 24556311]
[2] Dang, X, et al., “The first pilot project of the Consortium for Top-Down Proteomics: A status report,” Proteomics, 14:1130-40, 2014. [PubMed ID: 24644084]
[3] Catherman, AD, et al., “Large-scale top-down proteomics of the human proteome: Membrane proteins, mitochondria, and senescence,” Mol Cell Proteomics, 12:3465-73, 2013. [PubMed ID: 24023390]
[4] Han, X, et al., “Sheathless capillary electrophoresis-tandem mass spectrometry for top-down characterization of Pyrococcus furiosus proteins on a proteome scale,” Anal Chem, 86:11006-12, 2014. [PubMed ID: 25346219]
[5] Cannon, JR, et al., “Top-down 193-nm ultraviolet photodissociation mass spectrometry for simultaneous determination of polyubiquitin chain length and topology,” Anal Chem, 87:1812-20, 2015. [PubMed ID: 25559986]
Image: Shutterstock