Proximity labeling and other novel mass spectrometric approaches for spatiotemporal protein dynamics

Abstract

Background: Proteins are highly dynamic and their biological function is controlled by not only temporal abundance changes but also via regulated protein-protein interaction networks, which respond to internal and external perturbations. A wealth of novel analytical reagents and workflows allow studying spatiotemporal protein environments with great granularity while maintaining high throughput and ease of analysis.

Areas covered: We review technology advances for measuring protein-protein proximity interactions with an emphasis on proximity labeling, and briefly summarize other spatiotemporal approaches including protein localization, and their dynamic changes over time, specifically in human cells and mammalian tissues. We focus especially on novel technologies and workflows emerging within the past 5 years. This includes enrichment-based techniques (proximity labeling and crosslinking), separation-based techniques (organelle fractionation and size exclusion chromatography), and finally sorting-based techniques (laser capture microdissection and mass spectrometry imaging).

Expert opinion: Spatiotemporal proteomics is a key step in assessing biological complexity, understanding refined regulatory mechanisms, and forming protein complexes and networks. Studying protein dynamics across space and time holds promise for gaining deep insights into how protein networks may be perturbed during disease and aging processes, and offer potential avenues for therapeutic interventions, drug discovery, and biomarker development.

Keywords: Chemical crosslinking; mass spectrometry; protein interactome; protein networks; protein turnover; proximity labeling.

Figure 1:. The spatiotemporal proteome measured by mass spectrometry.

Top left panel: identifying the individual members of protein complexes or nearby protein partners provides information for protein-protein interactions. Top right panel: proteins may exist in multiple sub- or extracellular locations and can be measured by protein localization approaches. Bottom right panel: measuring the degradation and/or synthesis of proteins over time provides temporal information of protein turnover rates. Bottom left panel: the various post-translational modifications that can be made to a base protein sequence, such as phosphorylation, ubiquitination, or glycosylation, may change the biological function, interactions, and localization of a protein.

Figure 2.. Proximity labeling compared to affinity purification and epitope tagging for mass spectrometry-based protein-protein interaction studies.

Affinity purification methods (left) require an antibody specific to the protein of interest, enriching for the protein by immobilizing it with the antibody for enrichment. Epitope tagging methods (center) require a plasmid construct with the protein of interest fused to a small peptide (epitope), then an anti-epitope antibody is used to enrich for the protein of interest. Proximity labeling methods (right) require a plasmid construct with the protein of interest fused to a proximity labeling enzyme, which after expression can be combined with biotin to covalently modify all nearby proteins for streptavidin enrichment. This figure was originally published by Pino et. al Biochem Soc Trans 30 October 2020; 48 (5): 1953–1966.[81]

Figure 3.. Methods for spatial proteomics directly from tissue samples.

Tissue samples are typically harvested and embedded in paraffin (left). Mass spectrometry imaging approaches sample proteins directly from the surface of the embedded tissue and analyze the proteins directly by mass spectrometry. Data can be analyzed as an image using multimodal image alignment or analyzed as proteomics data with detection and statistical analysis (right). Laser capture microdissection approaches will subsample specific sections of the embedded tissue, and subsamples are then prepared either for protein-centric (top down) or peptide-centric (bottom up) mass spectrometry. Data is analyzed like typical proteomics experiments with detection and statistical analysis.