Single-molecule proteomics: next-generation analysis of proteins in individual cells

Proteins, the molecules that work together to enable life, are formed from strings of amino acids and encoded by genes. Although we have about 20,000 genes, there are many more than 20,000 "proteoforms"-altered forms of a protein that can function very differently despite sharing the same amino acid sequence. Post-translational modifications (PTMs) are an important source of these alterations and one of the most common PTMs is phosphorylation-the naturally occurring addition of a phosphoryl group to an amino acid on a protein. There are hundreds of different types of PTMs, and they often co-occur on the same protein; other PTMs involve the addition of sugars (glycosylation), lipids (lipidation) and acetyl groups (acetylation). Due in large part to the complexity arising from PTMs, the field of proteomics - which focuses on identifying and quantifying proteins - has so far struggled in its efforts to fully describe how proteins in a given cell function and work together. Rather, two fundamental limitations to current proteomics strategies have emerged. The first is a reliance on costly and complex technology. The second is the insensitivity of the technology to the complexity and combinatorics of many PTMs, including some forms of phosphorylation. Still, the ability to distinguish different proteoforms and understand the effects of PTMs remains essential. Even if difficult to detect, PTMs affect nearly all proteins, and proteomics is incomplete without them. We propose to transform the capabilities of proteomics by developing a next-generation approach that overcomes the above limitations. Instead of relying on mass spectrometry, the dominant proteomics technology, we will bring together three complementary new technologies. The first, nanopore technology, can be used to infer a protein's amino acid sequence. The second, electrometry, measures electrical charge. The third, mass photometry, measures mass. We will combine these measurements with microfluidics so that we can analyse the protein content of single cells. Our hypothesis is that bringing together the three types of measurements of a given protein (along with existing data about the proteins in the types of cells under study and applying machine learning) will enable identification of individual proteins and detection of their PTMs. We have three main objectives. The first two centre on developing, validating and refining our platform. The third is to apply our approach in bacteria, where the most common forms of phosphorylation tend to be more unstable and difficult to detect using existing proteomics methods. There are critical gaps in our knowledge of phosphorylation in the complex protein networks fundamental for bacterial life. We will study ubiquitous regulatory systems (known as two-component systems) in the opportunistic, disease-causing and increasingly multi-drug resistant bacterial pathogen P. aeruginosa. Our experienced and accomplished team includes the inventors of the three nanometric technologies, who are all based at the University of Oxford's Department of Chemistry. Other team members, based in Oxford, at the University of Liverpool and at the Wellcome Sanger Institute in Cambridge, bring expertise in microfluidics, machine learning, bioinformatics, biochemistry and microbiology, while the input from two supportive companies aligned with our vision will also be welcome. Our platform will make it possible to capture the PTMs that enable the rich complexity of protein function but are currently effectively invisible. This ambitious work will give rise to numerous valuable insights-both during development and once the platform is established. It will transform proteomics research across the life and environmental sciences, may bring economic impacts through commercialisation of the technology, and enhance our knowledge of PTMs' roles in disease, and bacterial virulence and drug resistance.