Mass spectrometry (MS) is a powerful tool for characterizing target proteins of interest from complex mixtures. When preparing samples for MS, there are two commonly used approaches: top-down and bottom-up (also called shotgun). An advantage of the top-down method is that proteins remain intact during MS analysis, which is helpful in studying sequence variants or post-translational modifications. However, intact proteins (vs. the smaller peptides used in bottom-up) are more difficult to separate and ionize prior to MS, making the bottom-up method a more widely used choice for protein quantitation and proteome profiling. In the bottom-up approach, proteins are digested enzymatically; the resulting peptides can be separated by liquid chromatography (LC) and then analyzed by MS. Like pieces of a jigsaw puzzle, the MS results are then assembled to gain a better understanding of the intact proteins in the starting sample. The bottom-up approach often involves multiple sample-preparation steps prior to MS. Here we examine some of the latest tools available to assist MS researchers interested in characterizing their proteins using a bottom-up proteomics approach.
Reagents and automation
Lysis of cells and/or tissues is often the first preparation step. Lysis buffers are designed not only to facilitate disruption of the cell but also to prevent protein denaturation. Many vendors offer protein extraction buffers reactive to specific cell or tissue types. In addition, there are commercially available filters for removing post-lysis debris, protease and phosphatase inhibitors, and reducing or alkylating agents.
Before digestion, sample preparation may require removal of detergents, which can interfere with digestion and MS. This typically is accomplished by filter-aided sample preparation (FASP). “Samples can be buffer-exchanged and concentrated through multiple spins without drying and precipitation concerns,” says Ivona Strug, senior biochemical scientist at MilliporeSigma, who notes that researchers recently adapted FASP to a 96-well format using MilliporeSigma’s MultiScreen® Filter Plates for increased throughput [1].
Waters offers another option, acid-labile surfactant RapiGest SF, to remove detergents from a sample. “After solubilization and digestion, you lower the pH of the solution, and RapiGest falls apart. So it doesn’t hamper the LC-MS analysis,” says Hans Vissers, senior manager, science operations, in health sciences research at Waters Corp. The company’s new ProteinWorks products, such as the ProteinWorks eXpress Digest and uElution solid-phase extraction (SPE) kits, also are designed to streamline the sample-preparation workflow.
Digestion enzymes are critical tools in bottom-up proteomic methods. One challenge is “providing proteases that give reproducible results in cases where trypsin does not work well,” says Gary Kobs, senior global product manager at Promega. In such cases, Promega recommends its Trypsin/Lys-C protease blend or alternative proteases. “The use of unique protease blends to provide data that standard tryptic digestions can’t, will be the next step in bottom-up proteomics,” Kobs says.
Liquid chromatography often follows sample preparation. To improve workflow, Waters offers the microfluidics-based ionKey (or iKey) Separation Device, which plugs straight into the MS source. Inside the iKey is a laser-etched microfluidic channel packed with sub-2-micron stationary-phase particles for LC separation. About the size of a smartphone, the iKey is especially helpful for routine work and for “novel microcolumn LC users, who may lack the technical know-how or inclination to troubleshoot the sometimes tricky nanoflow columns that are often used in proteomics research,” says Vissers.
SCIEX automates sample preparation for bottom-up proteomics with its Biomek NXP Span-8 Workstation (from Beckman Coulter), which runs common protocols such as trypsin digestion, protein denaturation, reduction and alkylation. “We combined our protein-preparation kit with an optimized method on the BioMek to provide an out-of-the-box solution to process up to 96 proteomic samples in a five- [to] six-hour time frame with very high reproducibility,” says Christie Hunter, director of ‘omics applications at SCIEX. The company is developing additional modules, such as bead-based immuno-capture.
Agilent Technologies’ AssayMAP Bravo Platform also automates protein and peptide digestion and affinity purification. Automation can reduce human error and other sources of variability and increase throughput and reproducibility. “Minimizing variability from sample preparation and analysis improves throughput and reduces the number of biological samples necessary to get a statistically meaningful result,” says David Edwards, senior director of mass spectrometry marketing at Agilent Technologies.
Depletion and enrichment
Edwards believes sample complexity is a major analytical challenge in bottom-up proteomics. Depletion and enrichment of a sample let you alter a sample’s composition to make it easier to analyze. Both Agilent’s Multiple Affinity Removal System and MilliporeSigma’s Seppro® technology remove the 14 most abundant proteins, reducing sample complexity to help researchers focus on lower-abundance proteins. Seppro technology is now available in a high-throughput plate format for human plasma, “for validating potential biomarkers across thousands of samples,” says Kevin Ray, senior manager of analytical R&D at MilliporeSigma.
In case removing the top 14 proteins isn’t enough, MilliporeSigma’s SuperMix technology targets the next tier of medium-abundance proteins for depletion. According to Ray, this technology currently is enabling a clinical-diagnostics lab to evaluate blood samples for malignant vs. benign lung masses detected by imaging. “SuperMix technology has been quite powerful for allowing people to dig down deep into low-abundance proteins of the plasma proteome,” he says.
Whereas depletion strategies get rid of unwanted proteins, enrichment strategies are useful for targeted protein studies. Tools for the enrichment of phosphoproteins include Clontech’s Phosphoprotein Enrichment Kit, G Biosciences’ FOCUS™ PhosphoRich™ Kit, MilliporeSigma’s Phosphopeptide Enrichment Kit, Revvity’s Phos-trap™ Phosphopeptide Enrichment Kit and Thermo Fisher Scientific’s Phosphoprotein Enrichment Kit. Researchers can enrich for glycosylated proteins using products such as Clontech’s Glycoprotein Enrichment Kit or MilliporeSigma’s ProteoExtract® Glycopeptide Enrichment Kit. In addition, tools such as Thermo Fisher Scientific’s Streptavidin IP-MS and Protein A/G IP-MS Kits allow for enrichment of particular proteins in targeted proteomics. These kits enable researchers to select for and increase the likelihood of detecting the important “needle in a haystack.”
MS features and software
Tool providers are continually evolving the options and features of both their systems and the software to assist researchers with bottom-up proteomics workflow methods. One example is Agilent’s iFunnel technology, which “boosts sensitivity, enhancing the detection of low-level peptides,” says Edwards. Agilent’s Jet Stream Proteomics solutions also enable researchers to “achieve near-nanoflow sensitivity using conventional-flow HPLC, [with] higher speed, greater robustness and better reproducibility for bottom-up proteomics analysis,” says Edwards. Several vendors have added ion mobility to their MS instruments, which gives extra separation power to researchers analyzing complex mixtures. Examples include Agilent’s 6560 Ion Mobility Q-TOF LC/MS system and Waters’ Vion and Synapt systems.
SCIEX’s SWATH® Acquisition, a data-independent acquisition (DIA) technique, is designed for higher-throughput comprehensive quantitation in discovery proteomics on its TripleTOF® 6600 system. The company’s QTRAP® 6500 system’s “SelexION® Technology allows a researcher to use differential mobility separation to remove the tough interferences from a quantitative targeted assay,” says Hunter.
After getting results from the mass spectrometer, researchers need software to help them piece together the results into biologically meaningful data. Waters offers two software products for bottom-up proteomics researchers: Progenesis QI, for label-free quantitation proteomic studies, and ProteinLynx GlobalSERVER, for routine bottom-up proteomics applications, labeled quantitation and examining post-translation modifications.
Agilent’s Spectrum Mill integrates with the open-source software Skyline (from the MacCoss Lab of Biological Mass Spectrometry at the University of Washington), which is designed for targeted, quantitative MS studies. Agilent Mass Profiler Professional (MPP) software statistically analyzes protein or peptide abundances across multiple samples. “The MPP Pathway Architect module takes the results from bottom-up proteomics experiments and maps them onto canonical biological pathways, concurrently analyzing, visualizing and interpreting pathway information,” says Edwards.
SCIEX’s SWATH processing tools are in the BaseSpace Cloud computing environment. “One of the advantages of working in BaseSpace is the ability to process and compare large multi-omics datasets from genomics, transcriptomics and proteomics, all within the same cloud environment, so customers can get a more complete picture of the biology,” says Hunter. For example, users can visualize their data in the context of pathways using third-party applications such as iPathwayGuide from Advaita.
Even as tools continue to emerge, bottom-up proteomics still wrestles with variability. “Lab-to-lab consistency continues to be a huge problem,” says Ray. One main source of variability in proteomics is digestion during sample preparation. Issues such as digestion times, incomplete digestions, digestion-resistant proteins and denaturation prior to digestion can all wreak havoc with consistency. “The discussion around what technological advances are needed to improve proteomics studies is shifting from instrumentation to sample preparation,” says Strug. Thus it’s likely that further improvements to proteomic sample-preparation tools are on the horizon.
Reference
[1] Yu, Y, et al., ”Urine sample preparation in 96-well filter plates for quantitative clinical proteomics,” Analytical Chemistry, 86:5470, 2014. [PMID: 24797144]
Image: Shutterstock