An Introduction to Spatial Proteomics

The human body consists of around 30 trillion cells, coordinating across tissues and organs to maintain life. Critical for their function are processes that occur on a subcellular scale, mediated by biological pathways that are tightly coupled to the position of proteins within the cell. While scientists have been observing cells under microscopes for almost 400 years, only relatively recently has it been possible to precisely locate proteins within them. Through spatial proteomics, we have a new view of protein dynamics at this minute scale. We can now fill gaps in our biological understanding, explore the molecular underpinning of disease, and reduce the uncertainty of the drug discovery process, by illuminating this previously invisible complexity.

Why does subcellular location matter?

Each cell consists of multiple compartments with specialised environments. These range from membrane-bound organelles, such as the nucleus (the primary site of DNA synthesis), to local substructures such as protein complexes. After synthesis, proteins are trafficked to specific locations, often determined by variations in DNA sequence, RNA splicing, or post-translational modifications. Subcellular localisation is highly dynamic: more than half of proteins exist in multiple compartments, and translocation between locations often regulates protein function and interactions, due to changes in the chemical environment and binding partners. In addition, many proteins are now known to have more than one distinct function depending on the subcellular context; such proteins are disproportionately likely to be involved in disease mechanisms and targeted by drugs. Spatial proteomics is therefore essential for understanding the complete picture of cellular behaviour.

How do we capture spatial information?

Methods for spatial proteomics generally fall into two categories: mass spectroscopy, and fluorescence imaging.

Mass spectroscopy (MS) is a sensitive method for identifying proteins and quantifying their abundance within a sample. To map the spatial distribution of proteins, MS is applied to cellular compartments that have been separated by gradient centrifugation, or, more recently, by laser microdissection of samples. As this method does not target specific proteins, it can achieve an unbiased, high-coverage view of spatial distribution at the proteoform level. However, it is challenging to perform comparative analyses across multiple experiments and varied conditions, and single-cell resolution has only recently been achieved.

Imaging-based methods visualise the location of proteins within their native cell context using fluorescent dyes, antibody labels, or genetically engineered tags. Multiple proteins can be labelled simultaneously in the same cell, known as multiplexing. These labels are protein-specific, and can be time-consuming to produce and validate. However, in contrast to MS-based methods, imaging allows direct visualisation of protein location, and it is possible to perform high-throughput comparisons across many time points, conditions, and cell types. This method therefore is well-suited to answering specific biological questions, or systematically profiling the cellular responses to different perturbations using representative protein panels.

Protein interactions at nanoscale resolution

At Micrographia Bio, we combine state-of-the-art microscopy with unique deconvolution technology and computer vision to achieve highly multiplexed images at nanoscale resolution. This allows the precise quantification of protein abundance and distribution, as well as direct observation of protein-protein interactions. Combined with advanced automation, we are pushing protein coverage and experimental throughput to new limits. 

We believe that multiplexed, nanoscale-resolution, spatial proteomics will transform drug discovery. We can directly and quantitatively observe cell morphology, protein interaction networks, and mechanistic behaviour in response to perturbations. 

Drug discovery is a challenging process, and many failures are caused by an incomplete understanding of pathological processes, the mechanism of action of candidate drugs, and the pathways underlying drug toxicity. Through this technology, we can now shed new light on these questions in unprecedented detail.