Recent Advancements in Subcellular Proteomics: Growing Impact of Organellar Protein Niches on the Understanding of Cell Biology

Abstract

The mammalian cell is a complex entity, with membrane-bound and membrane-less organelles playing vital roles in regulating cellular homeostasis. Organellar protein niches drive discrete biological processes and cell functions, thus maintaining cell equilibrium. Cellular processes such as signaling, growth, proliferation, motility, and programmed cell death require dynamic protein movements between cell compartments. Aberrant protein localization is associated with a wide range of diseases. Therefore, analyzing the subcellular proteome of the cell can provide a comprehensive overview of cellular biology. With recent advancements in mass spectrometry, imaging technology, computational tools, and deep machine learning algorithms, studies pertaining to subcellular protein localization and their dynamic distributions are gaining momentum. These studies reveal changing interaction networks because of "moonlighting proteins" and serve as a discovery tool for disease network mechanisms. Consequently, this review aims to provide a comprehensive repository for recent advancements in subcellular proteomics subcontexting methods, challenges, and future perspectives for method developers. In summary, subcellular proteomics is crucial to the understanding of the fundamental cellular mechanisms and the associated diseases.

Keywords: APEX; BioID; computational tools; dynamic organellar maps; imaging technology; machine learning algorithms; mass spectrometry; proximity protein labeling; subcellular localization.

Figure 1.

A schematic workflow of mass spectrometry-based subcellular protein niche identification involves the following steps: (1) Cell lysis and homogenization to obtain membrane-bound and membrane-less organelles, (2) differential and density gradient centrifugation to separate organelles, and (3) data analysis using deep machine learning tools for distinct subcellular niche identification and protein dynamic translocation. Machine learning (ML) tools help to simplify complex data sets.

Figure 2.

Schematic workflow of proximity labeling strategies. (a) Proteins of interest (Bait) are genetically fused with an enzyme such as APEX/APEX2/BirA, BioID2, miniTurbo, or TurboID (1) that biotinylates adjacent proteins upon incubation with biotin (2). Control lines can indicate the labeling enzyme, which is fused to a control bait that is nonspecifically localized, such as GFP. (3). After labeling, proteins are enriched through a streptavidin pull-down, followed by identification through mass spectrometry. These labeled proteins are termed “Prey.” The prey is compared with proteins isolated from control lines to identify high-confidence proximity interactors. (b) Two types of analyses can be used to study subcellular components through proximity labeling, namely bait-centric and prey-centric analyses. Isotopic labeling and bait quantification techniques are used to identify proteins in organelles. Clustering baits and prey-centric studies can reveal proximity interaction networks.

Figure 3.

Schematic workflow of imaging-based subcellular protein niche identification. Imaging analysis can be done through live cell imaging using membrane-permeable dyes and fusion proteins. Time-lapse microscopy studies the protein translocation between subcellular compartments. Alternatively, the immunochemistry-based method uses antibodies, aptamers, and nanobodies against target proteins to study the cell compartment proteome.

Figure 4.

Summary of machine learning-aided spatial proteomics applications in cell biology.

Figure 5.

Following is a schematic workflow of machine learning tools that can be used with MS-based spatial proteome data analysis: (1) Cell homogenization and subcellular fractionation are performed to determine the enrichment of the organelle of interest. (2) Mass spectrometry is used to identify each subcellular component proteome, which provides a large amount of raw data to analyze. (3) The data processing step is crucial, as missing values are imputed, and the data is normalized against the database. (4) Processing large data sets can be challenging, but machine learning tools can assist in data reduction and clustering. (5) Semisupervised clustering is generally used for novelty detection of the cellular compartments. (6) Similarly, supervised clustering can predict the subcellular niche of the protein of interest. (7) Downstream quantitative analysis methods such as cluster analysis algorithms are then used to visualize the data.