Imaging the proteome at the nanoscale

ended

The key principle of biological imaging is specific labeling. The protein of interest is revealed by tagging with fluorophores, either by genetic encoding, using green-fluorescent-protein (GFP) variants, or by affinity labeling, using antibodies. This procedure has been successful for several decades, but has the great disadvantage that each protein needs to be tagged individually: specific antibodies are needed for each and every protein. This limitation stops imaging from becoming a high-throughput “omics” approach. We propose to change this here, through an imageomics approach based on a combination of probe development and nanoscale imaging. We will develop affinity probes that bind with high specificity not to specific proteins, but to amino acid sequences (peptides) that are present in more than one protein. We will choose 20-40 such peptides, in a fashion that ensures that virtually every protein in the human proteome contains a specific subset of the peptides. We will then develop nanobodies that bind to each of these peptides. We prefer nanobodies to antibodies, based on their small size and optimal penetration into biological samples. We will then use the nanobodies to label biological samples, and we will image them at the nanoscale, with a resolution that is sufficient to reveal single proteins. By applying the nanobodies in a combinatorial fashion, we will “read” the sequence of each protein in the preparation, which will result in an image of its whole proteome. We will start by applying this approach to 2-dimensional samples, such as fluids adsorbed to coverslips. This will lay the foundation for future diagnostic studies for a variety of human diseases, based on human fluids such as plasma or cerebrospinal fluid. In a later stage, we will proceed to analyze cells and tissues, by generating 3-dimensional proteomic images. This approach will make antibody-based imaging, blotting and diagnostics obsolete, and has therfore an immense potential.