"Deep Visual Proteomics" technology provides cell-specific, protein-based information for the analysis of cancer diseases
How does cancer develop? How does the cellular composition of a tumor change its malignant properties? These questions are essential and difficult to answer. Nonetheless, they are crucial to understanding cancer and finding a lasting cure. A German-Danish team led by Matthias Mann has now developed a pioneering technology, "Deep Visual Proteomics". This technology provides researchers and clinicians with protein-based information and helps to understand cancers at the resolution of individual cell types. The technology demonstrates its potential in a first-time application to cancer cells.
Proteins are the key pieces of the puzzle for a variety of diseases. They are also referred to as the "molecular workhorses of the cell". Their correct function determines the functionality of a cell and thus also that of an individual. Matthias Mann explains: "If something in our cells isn't working properly and we get sick, you can be sure that proteins are involved in a wide variety of ways. For this reason, mapping the protein landscape can help us find out: Why was a tumor able to develop in a particular patient? What are the weak points of this tumor and which treatment method is advantageous?” explains Matthias Mann from the Max Planck Institute for Biochemistry near Munich and the Center for Protein Research of the Novo Nordisk Foundation at the University of Copenhagen in Denmark.
Inspired by these questions, an interdisciplinary research team led by Matthias Mann has developed an innovative new method. In the study, visual features of a tumor will be determined using a deep profiling technique to analyze proteins in abnormal cell clusters that are adjacent to surrounding healthy cells. This approach can give researchers unprecedented insight into cancer and help oncologists create targeted strategies for diagnosis and therapy.
Deep Visual Proteomics combines four technologies
"Deep Visual Proteomics" integrates for the first time the advantages of four different technologies in a single methodology. First, modern microscopy creates high-resolution tissue maps. Second, machine learning and artificial intelligence algorithms are used to classify cells based on their shape, size or protein localization before collecting individual cells using highly accurate laser microdissection. Third, after sorting normal or diverse diseased cell populations, thousands of proteins within these cell populations are simultaneously determined using ultra-sensitive mass spectrometry. Fourth, sophisticated bioinformatics analyzes generate protein maps that enable spatial resolution of proteins in highly complex diseases such as cancer. Such protein maps are valuable tools for clinicians to better understand the mechanisms of health and disease.
“Our new concept 'Deep Visual Proteomics' could become a paradigm shift for molecular pathology in the clinic. With this method we take a tissue sample with tumor cells and can identify thousands of proteins within a very short time and with little effort. These proteome analyzes unveil mechanisms that drive tumor development. Thus, new therapeutic targets can be derived directly from a single tissue section of a patient biopsy. It shows a cosmos of molecules within these cancer cells," says Andreas Mund, associate professor at the Center for Protein Research and part of the team led by Matthias Mann, who drove this development at the Center for Protein Research and at the Max Planck Institute for Biochemistry.
Relevance to clinical pathology
In the study, the researchers were able to apply "deep visual proteomics" to cells from patients with salivary gland and skin cancer. Lise Mette Rahbek Gjerdrum, Consultant Advisor and Associate Professor of Clinical Research at the Department of Pathology at Seeland University Hospital in Roskilde and the Department of Clinical Medicine at the University of Copenhagen describes: "This unique method combines analysis of tissue architecture with analysis of the proteome, which is essential for the selected cells are specific. We were recently able to diagnose a clinically highly complex case using 'Deep Visual Proteomics' analysis.”
Fabian Coscia, one of the two first authors of the study published in Nature Biotechnology and head of the "Spatial Proteomics" research group at the Max Delbrück Center for Molecular Medicine in the Helmholtz Association in Berlin since June 2021, says: "The technology can also be used for characterization other types of tumors can be applied in a similar way." His goal is to use the archived data from the biobanks to reveal new points of attack for individual cancer therapies and thus to develop forms of therapy tailored to the patient – even for tumors that have previously been resistant to therapy.
It's not just cancer that can be better understood using deep visual proteomics. The methodology can also be applied to other diseases. "For example, you can analyze the proteins of a nerve cell to find out what exactly happens in a cell during the course of neurodegenerative diseases such as Alzheimer's or Parkinson's," continues Coscia. "By combining microscopy, artificial intelligence and highly sensitive, mass spectrometry-based proteomics , we have developed a very powerful method to understand the molecular circuitry of healthy and diseased cells, which could help physicians to identify targets for future drugs and diagnoses,'' explains Matthias Mann.
