Advances in Immunopeptidomics

Immunopeptidomics is an emerging sub-discipline of proteomics used to identify and quantify immunopeptides presented by major histocompatibility complex (MHC) molecules on the surface of cells. Applications include the investigation of how cells process and present peptide antigens, which is integral to the development of T cell-based treatments. In this article we explore how mass spectrometry, artificial intelligence, and commercial immunopeptidomics services are advancing the field as well as the development of life-saving immunotherapeutics and vaccines.

Immunopeptidomics aims to uncover the vast repertoire of peptides that can be recognized by T cells, allowing researchers to gain insights into the immune system's recognition of peptides. “Immunopeptidomics research can identify those peptides that trigger an immune response and, thus, might be important for the development of immunotherapies against cancer, autoimmune disease, and infectious diseases,” notes Torsten Mueller, Business Development Manager for Proteomics at Bruker Daltonics.

For example, understanding a cancer patient’s immunopeptidome—or the specific antigens presented on their cancer cells—could provide information about potential targets for a personalized medicine approach. Immunopeptidomics is also helpful for developing vaccines, as it enables the identification of peptide targets that stimulate specific immune responses against pathogens.

Why mass spectrometry?

There are challenges associated with immunopeptidomics, however. For instance, many immunopeptides are only present at extremely low abundances, while their short sequences make them much harder to identify compared with full proteins.

Mass spectrometry-based methods are currently the most common approach to immunopeptidomics. According to Mueller, the sensitivity of mass spectrometry is invaluable in capturing and analyzing the diversity of the immunopeptidome. “Mass spectrometry provides a direct readout of what the immune system can see, distinguishing it from sequencing-based technologies that outline potential cell surface presentation.” Finally, mass spectrometry allows for the study of post-translational modifications.

4D proteomics

Mueller reports that there is a need for ultra-high-sensitivity mass spectrometry, such as Bruker’s timsTOF range of mass spectrometers, to analyze low-abundance immunopeptides. “High-sensitivity methods are crucial for clinical isolates like needle biopsies, where a tiny amount of sample tissue is used for many different tests.”

Trapped ion mobility mass spectrometry coupled with time-of-flight mass spectrometry (called timsTOF) adds a fourth dimension to traditional mass spectrometry and allows for what is referred to as 4D proteomics. “With 4D proteomics, we can determine the retention time, mass-to-charge ratio, fragment ion spectra, and, most importantly, the specific collisional cross-section of peptides.” Mueller adds that the additional dimension of the collisional cross-section helps to distinguish peptides with very similar mass-to-charge ratios.

Mueller further highlights Bruker’s new timsTOF Ultra, the company’s most sensitive instrument to date. “By design, all of our timsTOF systems have been developed to be highly sensitive, robust, and selective. However, years of advances have culminated in the timsTOF Ultra.” The timsTOF Ultra and the CaptiveSpray Ultra ion source have been specifically designed for optimal sensitivity while maintaining all timsTOF benefits.

Immunopeptidomics has also benefitted from the development of de novo peptide sequencing. This method is performed without prior knowledge of the amino acid sequence. “With de novo peptide sequencing, you can obtain the peptide sequences without a database. Instead, de novo sequencing uses computational approaches to deduce the peptide sequence directly from the spectra,” Mueller explains.

Because the data generated through analysis are complex and the original source of the peptides is often unknown, Bruker has introduced software such as the Bruker ProteoScape with the BPS Novor de novo analysis module (developed in cooperation with Rapid Novor) to increase confidence in the results. The BPS Novor module has been trained on Bruker timsTOF data and can process over 1,000 spectra per second.

Making immunopeptidomics more accessible

Biogenity and Cayman Chemical are two companies that offer immunopeptidomics workflow services, starting with the enrichment of the MHC-peptide complex, followed by the identification of the peptide by mass spectrometry. Both companies perform immunoaffinity purification techniques to enrich MHC-peptide complexes from various samples.

According to Julie Rumble, Ph.D., Director of Immunology & Cell Biology at Cayman Chemical, the company offers a wide range of services to support immunopeptidomics studies, from primary cell cultures, treatments, and transfections to LC-MS/MS analysis and transcriptome sequencing.

In contrast to Bruker’s timsTOF systems, Biogenity and Cayman Chemical use LC-MS/MS analysis to identify peptide sequences. “LC-MS/MS analysis provides an efficient and cost-effective deep sequence analysis, and we have established protocols for isolating and sequencing both MHC class I- and MHC class II-associated peptides from both human and mouse samples,” says Rumble.

“What sets Cayman Chemical apart is our flexibility in providing customized solutions tailored to each client's needs,” Rumble explains. “For us, it is not a one-size-fits-all approach.” She adds that the company’s clients have leveraged results from Cayman’s immunopeptidomics studies to identify tumor-specific antigens for personalized anti-cancer therapies, pathogen-derived peptides for vaccine development, and autoantigens for potential autoimmune treatments.

Cayman Chemical recently completed a study in collaboration with a laboratory at the University of Michigan to perform immunopeptidomics on human brain tumor cells—more specifically, glioma cells—grown as tumors in immunodeficient mice. “We were excited about how our data could be used to identify common or unique neoantigens in glioma cells, furthering the search for potential therapeutics in this devastating disease.”

AI approaches to recognize cancer cells

Rune Hertz Larsen, Scientist and Customer Consultant for Biogenity, reports that an essential component of Biogenity’s services involves statistical analyses to identify peptides specific to cancer cells. “We can also perform a more advanced bioinformatics analysis to identify which peptide sequence motifs are more common in cancer than normal cells.”

Biogenity also applies machine learning and deep learning models to predict the binding efficiency of peptides to the MHC molecules. This aids in the selection of peptides that are efficiently presented to the immune system. “These machine learning models are particularly important for vaccine development,” he emphasizes.