Dissecting the Proteome To Understand Disease
Advances in proteomic methods and data analysis tools are assisting in the identification and quantification of more proteins in smaller samples.
Complete the form below to unlock access to ALL audio articles.
The large-scale study of proteins, known as proteomics, provides unique insights into the regulation of biological processes and mechanisms of disease. Unlike the large-scale study of genes (genomics), protein expression changes over time, between cells and according to environmental conditions, offering a much more dynamic and complex picture.
“Proteomics gives you a valuable snapshot of what is actually going on inside a biological system,” says Claire Eyers, professor of Biological Mass Spectrometry in the Department of Biochemistry, Cell and Systems Biology and director of The Centre for Proteome Research (CPR) at the University of Liverpool, UK. “It is relevant to all areas of biology, there is no syndrome or disease which will not benefit from proteomic analyses.”
However, studying the proteome is arguably more challenging than analyzing the genome. Unlike genomics and transcriptomics methods, proteins can’t be amplified before their analysis. As Fabian Coscia, group Leader at the Max Delbrück Center for Molecular Medicine (MDC), Berlin, Germany, explains, there is no PCR method equivalent in proteomics. “We need to develop near lossless sample preparation methods that allow us to deliver trace sample amounts to the analytical device, but also highly sensitive analytical tools and instruments that can robustly analyze them.”
Advances in proteomic methods and data analysis tools are helping researchers identify and quantify more proteins in smaller samples, faster and more robustly. In this article, we explore some of the approaches that scientists are using to find early markers of disease, new drug targets and strategies to overcome resistance to existing treatments. We also take a look at progress in understanding the range of protein structures that can arise from a single gene (proteoforms) and protein sequencing, which will help take proteomics to the next level.
Insights from MS-based approaches in proteomics
Of all the established methods to study proteomics, mass spectrometry (MS)-based approaches are still the most widely used to detect and quantify protein levels. MS involves vaporizing samples under the influence of high voltage to create charged ions, separating them according to their mass-to-charge ratio and then detecting and measuring the abundance of each.
“The sensitivity of MS has rapidly evolved in recent years so now, for the first time, can analyze a couple of thousands of proteins from very little sample amount, including single cells,” Coscia says. Furthermore, advances in robotics and artificial intelligence have enabled the automation of workflows, making sample preparation and analysis much more streamlined. “With a robotics platform, the time and cost of processing a serum sample has dramatically decreased and reproducibility has improved,” says Eyers.
Eyers and her team are developing MS-based methods to quantify specific protein modifications in disease contexts. Post-translational modifications (PTMs), such as phosphorylation and sulfation, change how proteins behave and are associated with various biochemical pathways involved in cancer and infection. “The energy required to displace these small chemical groups in the mass spectrometer is different, so you can use different energetics to discriminate between the two,” she explains.
Protein phosphorylation is a reversible and dynamic PTM that can quickly affect protein–protein interactions and cell signaling events. In collaboration with colleagues at Newcastle University, Eyers has been exploring how aberrant phosphorylation of the transcription factor NF-κB in B cell lymphomas contributes to the development of resistance to inhibitors of the DNA damage checkpoint kinase CHK1.1 This type of study could help identify combination inhibitors to minimize cancer drug resistance.
A recent pan-cancer study involving samples from over 1000 patients identified shared PTM profiles across multiple cancer types linked to cancer-related processes such as DNA repair and immune evasion.2 These patterns may have gone undetected in smaller cohorts or by genomic studies. Further understanding how PTMs affect the function of proteins will reveal new mechanisms underlying disease that can potentially lead to better diagnostics and treatments.
Progress in spatial proteomics
Coscia’s team is establishing methods for performing high-resolution spatial proteomics to shed light on cancer cell properties. Spatial proteomics is a branch of proteomic research that allows researchers to examine how proteins are spatially organized in cells and tissues. This is particularly useful to study cancer cell heterogeneity and the role that the microenvironment plays in tumor development and progression.
“We know that some cells thwart the aggressive behavior of neighboring cells, while others help them to spread through the body, but this dynamic interplay is inadequately understood to date,” Coscia says. His team is using Deep Visual Proteomics to map the proteins of cancer cells and of neighboring cells.
Deep Visual Proteomics involves four steps: (i) imaging a tissue sample slice with a high-resolution microscope, (ii) identifying and classifying cells by phenotype using AI, (iii) isolating individual cells from the tissue with an automated laser beam and (iv) performing ultra-sensitive MS to determine the protein composition and projecting the results onto the original image.
With this tool, researchers are able to examine how the proteome is influenced by the type and state of neighboring cells and learn how diverse cell interactions are linked to disease outcomes and therapeutic responses. “Such data are a true treasure trove for the identification of novel therapeutic targets and disease-specific biomarkers,” Coscia says.
Although the technique is currently used to retrospectively analyze cancer patient samples, they hope to apply it prospectively to aid treatment decisions in the future. Importantly, Deep Visual Proteomics could be useful to understand the role of cell interactions in other disease contexts such as infection and neurodegeneration.
The promise of top-down proteomics and single-molecule sequencing
Depending on how proteins are processed, MS-based methods are described as “bottom-up” or “top-down”. In bottom-up approaches all proteins in the sample are enzymatically or chemically digested into peptides that serve as input to the mass spectrometer. The resulting peptide sequences are compared to existing databases to infer identity of the original proteins in the sample. By contrast, top-down proteomics aims to identify and profile intact proteoforms, so proteins in a sample are first separated and then analyzed by MS as intact protein ions.
Although the bottom-up approach remains the method of choice for protein identification and characterization, only a fraction of the total peptide population of a given protein is identified and, hence, information on only a portion of the protein sequence is obtained.3
Top-down proteomics can provide access to the complete protein sequence and has obvious advantages when it comes to detecting protein isoforms, degradation products and site-specific PTMs. However, intact proteins (vs. the smaller peptides used in bottom-up) are more difficult to efficiently fragment inside the instrument and identify using current search algorithms.
Advances in separation technologies, MS instrumentation and data analysis tools are rapidly improving the sensitivity and throughput of top-down proteomics.4 In 2023, using a highly sensitive single-cell top–down proteomics method, Jake Melby et al. were able to detect multiple isoforms of a large motor protein that drives muscle contraction and establish a direct relationship between proteoforms and muscle fibre types.5 This study highlights the potential of top-down proteomics for understanding how proteoforms modify cell function.
Initiatives such as the Human Proteoform Project, which aims to generate a definitive reference set of the proteoforms produced from the genome, are expected to revolutionize our understanding of human health and disease.6 At the time of writing, the Human Proteoform Atlas, a resource linking experimentally identified proteoforms to human cells, tissues and disease, contains over 60,000 unique proteoforms.
Eyers is keeping a close eye on ways to identify and characterize proteoforms. “MS may not be the technology that ends up coming to the forefront,” she says. Though still in early stages, non-MS-based single-molecule methods, such as single-molecule sequencing, enable researchers to detect all the different ways individual proteins are modified.
Efforts to adapt nanopore-based sequencing to proteins are starting to yield interesting results. Nova et al. detected PTMs at the single-molecule level on immunopeptide sequences with cancer-associated phosphate variants. This was achieved by chemically linking the peptides to a DNA oligonucleotide that is translocated in a stepwise manner through a nanopore using a DNA motor enzyme, as in nanopore DNA sequencing.7
Meanwhile, Sauciuc et al. engineered an electroosmotic flow that can translocate natural polypeptides across nanopores, remarkably increasing the feasibility of protein sequencing.8
Despite great excitement around these developments, there is a risk of spending time and money on using a new technology that is not relevant to the problem under investigation, Eyers points out. When it comes to moving proteomics towards the clinic, methods need to be robust and high-throughput. It’s not just about being able to acquire precise data on clinically important biomarkers but making sure the technology “fits” into drug discovery and development pipelines, or that the results be translated into new assays for use in a clinical chemistry laboratory.
Professor Jennifer Van Eyk and her team at Cedars-Sinai Medical Centre in California are developing and optimizing proteomic technologies for clinical applications. They have established a standardized MS-based workflow that can accommodate different types of biofluid sample, while achieving the precision and reproducibility required to develop translatable clinical biomarkers.9 They are also contributing to the development of best practice guidelines for performing, benchmarking and reporting single-cell proteomics experiments.10
“My hope is that by combining high-throughput cell-based screening and single-cell proteomics, we can move towards both personalized diagnostics and individualized therapies. That is, at least, where my lab is heading,” Van Eyk says.
About the interviewees:
Dr. Claire Eyers is professor of Biological Mass Spectrometry in the Department of Biochemistry, Cell and Systems Biology at the University of Liverpool, UK, with an interest in proteomics method development and application primarily in the area of cell signaling and disease. She is also director of The Centre for Proteome Research (CPR) and Associate Pro Vice Chancellor (Research and Impact) for the Faculty of Health and Life Sciences at the University of Liverpool.
Dr. Fabian Coscia is a group Leader at the Max Delbrück Center for Molecular Medicine (MDC) in Berlin, Germany. In 2023 he was awarded an European Research Council (ERC) grant to establish methods for performing high-resolution spatial proteomics and apply them to complex tumor tissue to understand how cancer cells become drug resistant.
Professor Jennifer Van Eyk is an international leader in clinical proteomics interested in the molecular basis behind a variety of cardiovascular disorders. Her lab, affiliated with the Cedars-Sinai Smidt Heart Institute and the Advanced Clinical Biosystems Research Institute at Smidt Heart Institute, in California, USA, is developing large-scale quantitative mass spectrometry methods to decipher the role of protein expression profiles on disease progression.
References:
1. Hunter JE, Campbell AE, Butterworth JA, et al. Mutation of the RelA(p65) Thr505 phosphosite disrupts the DNA replication stress response leading to CHK1 inhibitor resistance. Biochem J. 2022. doi: 10.1042/BCJ20220089
2. Geffen Y, Anand S, Akiyama Y, et al. Pan-cancer analysis of post-translational modifications reveals shared patterns of protein regulation. Cell. 2023;186(18):3945-3967.e26. doi: 10.1016/j.cell.2023.07.013
3. Po A, Eyers CE. Top-down proteomics and the challenges of true proteoform characterization. J Proteome Res. 2023;22(12):3663-3675. doi: 10.1021/acs.jproteome.3c00416
4. Catherman AD, Skinner OS, Kelleher NL. Top Down proteomics: Facts and perspectives. BRRC. 2014;445(4):683-693. doi:10.1016/j.bbrc.2014.02.041
5. Melby JA, Brown KA, Gregorich ZR, et al. High sensitivity top–down proteomics captures single muscle cell heterogeneity in large proteoforms. PNAS. 2023;120(19):e2222081120. doi: 10.1073/pnas.2222081120
6. Smith LM, Agar JN, Chamot-Rooke J, et al. The human proteoform project: Defining the human proteome. Sci Adv. 2021. doi: 10.1126/sciadv.abk0734
7. Nova IC, Ritmejeris J, Brinkerhoff H, Koenig TJR, Gundlach JH, Dekker C. Detection of phosphorylation post-translational modifications along single peptides with nanopores. Nat Biotechnol. 2023. doi: 10.1038/s41587-023-01839-z
8. Sauciuc A, Morozzo della Rocca B, Tadema MJ. et al. Translocation of linearized full-length proteins through an engineered nanopore under opposing electrophoretic force. Nat Biotechnol. 2023. doi: 10.1038/s41587-023-01954-x
9. Mc Ardle A, Binek A, Moradian A, et al. Standardized workflow for precise mid- and high-throughput proteomics of blood biofluids. Clin Chem. 2022;68(3):450-460. doi: 10.1093/clinchem/hvab202
10. Gatto L, Aebersold R, Cox J, et al. Initial recommendations for performing, benchmarking and reporting single-cell proteomics experiments. Nat Methods. 2023;20(3):375-386. doi: 10.1038/s41592-023-01785-3