Modifying samples and conditions as little as possible makes up a crucial objective in experimental analysis to identify biologically relevant changes. Nonetheless, many analytical approaches, such as fluorescence-based methods, require labels. Still, label-free systems exist for mass spectrometry (MS), Raman spectroscopy, and other technologies. Making measurements without any labeling allows scientists to keep a sample in more physiological conditions, which can benefit basic and applied research.
Taking a label-free approach can be especially useful in drug discovery.1 Here, Matt Cooper, a professor at the Institute for Molecular Bioscience at the University of Queensland, points out that label-free methods provide “rapid assay development, confirmation of receptor binding, and off-rate—dissociation—kinetics that govern receptor-ligand occupancy time.”
Image: In discovering tomorrow’s drugs, modifying samples and conditions as little as possible might improve the process, and that requires label-free analysis for some applications. (FDA photo by Michael J. Ermarth.)
Despite the great benefits of label-free technology, every form of it comes with some limitations. Cooper notes that “most label-free techniques require a high-quality, high-purity receptor, and surface immobilization.” He adds, “The latter, in particular, can impact receptor function/binding ability that can produce artifacts on occasion.”
Performance on proteins
Label-free approaches can be employed for the quantitation of proteins by MS. In quantitative proteomics, scientists identify and quantify peptides or proteins in a sample with primarily one of three techniques: stable isotope labeling with amino acids in cell culture (SILAC), tandem mass tags (TMT), and label-free MS.2 There are two approaches to label-free quantification, and they are differentiated by how the identification (MS/MS) data is acquired. This is done by either data-dependent acquisition (DDA) or data-independent acquisition (DIA).3
With data-dependent, label-free MS, it includes “some element of stochastic selection of precursors—meaning that which peptides are selected for fragmentation is slightly stochastic,” says Aaron Robitaille, a senior manager of product marketing for mass spectrometry at Thermo Fisher Scientific. “So, there can be small differences between experiments on which peptides are identified.” Consequently, statistical and computational approaches are used to reduce the variability across experiments and minimize the number of missing values used for quantitation. Alternatively, data-independent acquisition can decrease stochastic selection by fragmenting all precursors in a sample. However, this approach requires a time-intensive generation of a spectral library prior to quantitation of the unknown sample, and post-acquisition deconvolution of the data. This makes a DIA approach more suited for repeat measurement of the same sample type, such as proteomic profiling of plasma.
“The biggest advantage of label-free technology for MS is that you don’t have to treat cells by metabolically or chemically labeling the proteins,” says Robitaille. “With label-free MS, a sample can be freshly collected, preserved, or even previously frozen from a historical specimen collection.”4 Label-free MS can be performed with minimal sample processing to achieve relative quantification.
For quantification using label-free MS, Robitaille says that the biggest challenge is throughput. SILAC and TMT can simultaneously run up to 3 or 16 samples, respectively.5 “With label-free MS, you can’t take advantage of multiplexing,” Robitaille adds. So, if a researcher needs a 300-cohort study, it requires at least 300 liquid chromatography (LC)-MS runs, or 900 to perform a statistical analysis on triplicate measurements. “With label-free methods, time to results can be extended dramatically,” Robitaille says. “It can become quite challenging for large-scale cohorts, especially for proteomics and complex matrices that require hours for each experiment.”
When sensitivity is crucial in label-free MS, Thermo Scientific’s Orbitrap Exploris 480 mass spectrometer with a FAIMS Pro interface makes a good choice. It provides high-resolution, accurate-mass (HRAM) performance combined with differential ion mobility and “you can do both label-free and labeled quantification with higher selectivity and sensitivity,” Robitaille says.
The Orbitrap Exploris 480 MS platform combined with FAIMS Pro interface can even be applied to single-cell applications. “The incredible sensitivity of the mass spectrometer enables both label-free and labeled MS to be used in single-cell proteomics,” Robitaille says. “This is very exciting and opens the possibilities for many new discoveries in diverse areas from developmental biology to tumor heterogeneity. However, due to the biological variation at the single-cell level, many replicate samples must be analyzed to achieve statistical significance, further increasing the throughput requirements.”
This technology is currently for research-use-only. Getting this technology into clinical use is “the dream and the goal, but there are some challenges on how you do this efficiently, getting to a clinically relevant time scale as well as making the instrumentation easier to use,” Robitaille says. “There have been some great improvements—particularly with our latest instrument control software—that have made the mass spectrometers easier to use with pre-build method templates and simplified calibration routines” Robitaille says. “So, you spend less time setting up a method and more time quantifying biologically relevant samples.”
Cost/benefit trade offs
Like most analytical approaches, there’s room for both labeled and label-free MS technologies. In fact, a lab might work most efficiently and make the most discoveries by using a little bit of both approaches depending on the scale and application.
Robitaille explains that label-free methods can have higher coefficients of variation. “If you get 15% variance from label-free MS, a labeled approach’s variance can be smaller—say, 5%,” he says. “If you’re looking for very small changes measured reproducibly, you need that extra precision.” For example, isobaric labeling might be preferred for post-translational modifications, and phosphorylation is a common application of TMT, because of the improved quantitation precision. “At the protein level, label-free quantification can average out variance by rolling up the abundance from each peptide,” Robitaille says. “When you have 7 to 9 peptides per protein, one can achieve a reasonable coefficient of variation—15%.” He adds, “At the post-translational modification level, there are single amino-acid residues that are modified on one peptide, so the accuracy and precision of the quantitation must be high for each individual measurement, which isobaric labeling methods can achieve, ensuring one can detect small changes in expression.”
“Basically, labeled MS is a good choice where high-precision and throughput are needed,” Robitaille adds, “such as large cohort studies and modifications of proteins. Whereas if the sample type and species varies often, a label-free approach would be appropriate.”
Research with Raman
Some methods, such as coherent Raman scattering, are inherently label-free. Chi Zhang, a research scientist at the Beckman Institute, University of Illinois Urbana, describes coherent Raman scattering as: “a label-free technology using lasers to probe molecular vibrations.” Zhang adds that “coherent Raman spectroscopy and microscopy use two laser beams to excite vibrational transitions of molecules to generate strong Raman scattering signal, and the Raman signal contains chemical information of the specimen.”
In this way, Zhang and his colleagues can get chemical information from a sample without any chemical labeling. “We only use lasers,” he says. “In addition, the imaging speed is as fast as fluorescence microscopy.”
Still, scientists face some challenges in working with Raman-based methods. “Raman scattering, even with coherent enhancement, is still not as sensitive as fluorescence labeling,” Zhang says. “Currently, super-resolution technologies for coherent Raman microscopy are still under development and are facing a lot of challenges.”
Nonetheless, scientists use label-free Raman spectroscopy in many applications. As Zhang and his colleagues wrote: “Advancements in coherent Raman scattering (CRS) microscopy have enabled label-free visualization and analysis of functional, endogenous biomolecules in living systems.”6
There might always be a need for labeled and label-free methods of analysis. Some scientific challenges might lean toward one solution as easier than another. Always, though, scientists will seek ways to study the world without sample-preparation artifacts, and that makes label-free methods appealing where possible.
References
1. Halai, R., Cooper, M. Using label-free screening technology to improve efficiency in drug discovery. Expert Opin. Drug Discov. 2012. 7(2):123–131. [PMID: 22468914]
2. O’Connell, J, Paulo, J, et al. Proteome-wide evaluation of two common protein quantification methods. J. Proteome Res. 2018. 17(5):1934–1942. [PMID: 29635916]
3. Muntel, J, Kirkpatrick, J, Bruderer, R et al. Comparison of protein quantification in a complex background by DIA and TMT workflows with fixed instrument time. J. Proteome Res. 2019. 18(3):1340–1351. [PMID: 30726097]
4. Tsutaya, T, Mackie, M, Koenig, et al. 2019. Palaeoproteomic identification of breast milk protein residues from the archaeological skeletal remains of a neonatal dog. Sci. Rep. 9(1):12841. [PMID: 31492911]
5. Qendro, V., Lundgren, D.H., Rezaul. K., et al. 2014. Large-scale proteomic characterization of melanoma expressed proteins reveals nestin and vimentin as biomarkers that can potentially distinguish melanoma subtypes. J Proteome Res. 13(11):5031–5040. [PMID: 25322343]
6. Zhang, C., Zhang, D, Cheng, J.X. Coherent Raman scattering microscopy in biology and medicine. Annu. Rev. Biomed. Eng. 2015. 17:415–445. [PMID: 26514285]