Scientists explore proteomes to identify proteins, quantify their expression and analyze modifications. But proteome coverage and complexity are increasing. This throws up unique challenges for traditional profiling, which relies upon time-consuming and labor-intensive methods that are struggling to keep up with the ever-increasing demands of high-throughput proteomics. Technologies based on high-resolution mass spectrometry (MS), however, can overcome such challenges, making them the method of choice in basic, translational and biopharmaceutical laboratories alike.
Recent advances in this area, notably differential ion mobility technology coupled with liquid chromatography (LC) and tandem MS systems, enable sensitive, accurate, high-throughput proteomic analyses without sacrificing performance. Here, we explore how proteomic methods and coverage can benefit from using high-field asymmetric waveform ion mobility spectrometry (FAIMS) coupled to a high-resolution MS.
We also examine how different modes of data acquisition in shotgun proteomics (also known as bottom-up proteomics), such as data-dependent and data-independent modes, can affect the number of proteins identified, and the confidence with which they are quantified. The promise of these technologies is demonstrated using a case study from the field of paleoproteomics, which corroborates their ability and potential to provide detailed protein identification from precious, rare samples.
A more selective, productive way to profile proteins
Conventional methods for detecting and profiling proteins, such as enzyme-linked immunosorbent assay (ELISA) or western blotting, involve a great deal of manual work, and so are limited in efficiency and throughput. These assays are also not always sensitive enough to detect less abundant proteins. Scientists who rely on these antibody-based methods may struggle to detect and quantify proteins at trace levels in a high throughput manner, limiting the speed at which they can test their hypotheses.
Shotgun proteomics methods using LC-MS bring much-needed high-throughput capabilities to proteome profiling. In the widely used “bottom-up” proteomics approach, complex mixtures of proteins are enzymatically broken down into peptides, separated by chromatography, and analyzed on the mass spectrometer.
To enhance selectivity and improve detection limits, MS can be used in combination with FAIMS. FAIMS is a differential ion mobility technique that spatially separates ionized species in the gaseous phase in the presence of strong and weak electric fields. Placed before the MS entrance, only ions with a specific mobility value and stable flight path exit into the MS to be detected and sequenced. To select which ion groups traverse the interface for MS analysis, a secondary direct current potential, the compensation voltage (CV), is applied to the inner electrode. Changing the CV selects different groups of ions and allows multiple unique precursors to be detected, giving increased coverage over FAIMS-free methods.
A combined FAIMS and MS workflow brings numerous benefits over conventional immunoassays and methods involving only LC-MS. By carrying out online gas-phase fractionation before introducing ions into the mass spectrometer, the method increases proteome coverage without any additional sample preparation steps, making it more productive as a result. FAIMS also accommodates nano and capillary flow applications, helping to conserve precious, size-limited samples. In terms of implementation, some modern technologies offer pre-built method templates with recommended parameters, allowing the workflow to slot seamlessly into a proteomics laboratory without the need for any extra technical expertise.
Data-dependent acquisition (DDA): Improved proteome coverage
Proteomics aims to identify a specified number of proteins with high confidence. Whether or not this is achieved hinges upon a method’s approach to gathering data. Classic shotgun proteomics research acquires data using a method known as data-dependent acquisition (DDA), where the mass spectrometer analyzes separated LC components of a mixture to produce the MS1 scan. In DDA, the highest-intensity precursor ions from MS1 are selected, fragmented into smaller product ions, and further analyzed in the second stage of tandem MS (MS2). The data from MS2 is then subjected to a post-acquisition database search in order to identify peptides.
The stochastic nature of DDA precursor selection means that a percentage of detectable peptides get fragmented and typically 80% of the fragmented ions are shared among all the runs, resulting in poor reproducibility. As only selected precursors are fragmented, peptides are under-sampled, and some proteome information is lost. This is commonly referred to as the missing value problem, and limits quantitation precision across a large number of samples.
In a comprehensive analysis of the complete human proteome from a single-cell type, ~584,000 unique peptide sequences from HeLa cells were analyzed using MS-based shotgun proteomics (1). To improve the detection limits of the DDA approach and maximize the number of proteins identified, the researchers used several sample loads with multiple short LC-MS gradients. This multi-shot strategy increased the dynamic range and proteome coverage compared to single-shot methods, while maintaining a high confidence in identification with a false discovery rate of 1% at the protein level.
The researchers also tried using FAIMS with DDA, running a CV gradient to allow different groups of ions to pass into the mass spectrometer. This enabled them to identify 30% more proteins at a given CV than in the previous FAIMS-free workflow.
Data-independent acquisition (DIA): Comprehensive and unbiased
In data-independent acquisition (DIA)**, all analytes in a predefined window are fragmented and subsequently analyzed in the MS2 scan. The DIA method fragments every single peptide in the sample – not just selected precursors, as in DDA – providing higher reproducibility. During the fragmentation process, all peptides that yield fragment-ion signals above the noise will get consistently recorded on every run, significantly reducing the number of missing values. This unbiased approach to protein profiling makes DIA a promising technique for proteome discovery analysis.
DIA acquires MS2 spectra in a parallel manner, allowing it to identify more peptide precursors than DDA, which instead acquires MS2 spectra sequentially. A study performed by Professor Jesper Olsen’s group at the University of Copenhagen compared the performance of shotgun proteomics using DIA and DDA methods (2). Using the same LC-MS2 gradients, DIA routinely identified more than twice the number of peptides compared to DDA, while also offering a broader dynamic range.
When coupled with FAIMS, DIA brings the best of both worlds to proteomics analysis: higher reproducibility and greater coverage of the proteome. In an experiment by the Olsen Lab to test the sensitivity of the different methods in proteome analysis, a dilution series was performed where gradually increasing concentrations of loaded samples were analyzed. DIA, coupled with FAIMS, consistently yielded higher numbers of quantified proteins compared to DDA or DIA alone, even when the amount of loaded sample was low. For example, at 5 ng of loaded sample, while DDA and DIA identified 26 and 547 proteins respectively, DIA with FAIMS identified 1,074 proteins, making it the most sensitive method in comparison.
Case study: Proteomic analysis of a 43,000-year-old woolly mammoth bone
The field of paleoproteomics, which deals with rare fossil samples to examine species evolution, needs highly sensitive proteomics. Proteins are preserved in fossils far longer than DNA and act as ancient biomarkers, providing invaluable information about evolutionary relationships and molecular phylogenetic inference. Historic bone samples are not amenable to DNA sequencing, and so are subjected to high-resolution tandem MS (such as methods based on LC-FAIMS-Orbitrap MS) to identify ancient proteins.
Using high-sensitivity, high-resolution tandem MS, a research collaboration including the Natural History Museum at Denmark and the Novo Nordisk Foundation Center for Protein Research performed shotgun sequencing on ancient protein remains extracted from a 43,000-year-old woolly mammoth bone preserved in Siberian permafrost (3).
The research team identified 126 unique proteins from the mammoth bone remains, 90% of which have not been previously identified through MS-sequencing from an ancient bone. The protein profile showed a predominance of extracellular matrix (ECM)-associated proteins (49%) and plasma proteins (21%), with the rest comprising membrane and intracellular proteins, collagens, and keratin.
Due to taxonomic placement by sequence variants, the FAIMS technology brought the deep proteome coverage the project needed, illustrating how advances in high-throughput proteomics platforms offer new perspectives in evolutionary biology.
Proteomics for basic and translational research
Proteomics laboratories are facing increasing demands, and these can be met by rolling the selectivity of FAIMS and the high resolution of MS into one robust workflow. Whether the end application is exploratory, as in paleoproteomics, or translational through the identification of clinically relevant biomarkers from patient samples, the high-throughput capabilities brought by these advanced MS-based methods enable rapid screening of hundreds of samples, all while improving quantitative precision and increasing proteome coverage.
1. An Optimized Shotgun Strategy for the Rapid Generation of Comprehensive Human Proteomes https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5493283/
2. Performance Evaluation of the Q Exactive HF-X for Shotgun Proteomics https://pubs.acs.org/doi/10.1021/acs.jproteome.7b00602#
3. Proteomic Analysis of a Pleistocene Mammoth Femur Reveals More than One Hundred Ancient Bone Proteins https://pubs.acs.org/doi/full/10.1021/pr200721u
**The DIA method is currently not yet available in the United States of America