Current and Emerging Approaches for Studying Inter-Organelle Membrane Contact Sites

Abstract

Inter-organelle membrane contact sites (MCSs) are classically defined as areas of close proximity between heterologous membranes and established by specific proteins (termed tethers). The interest on MCSs has rapidly increased in the last years, since MCSs play a crucial role in the transfer of cellular components between different organelles and have been involved in important cellular functions such as apoptosis, organelle division and biogenesis, and cell growth. Recently, an unprecedented depth and breadth in insights into the details of MCSs have been uncovered. On one hand, extensive MCSs (organelles interactome) are revealed by comprehensive analysis of organelle network with high temporal-spatial resolution at the system level. On the other hand, more and more tethers involving in MCSs are identified and further works are focusing on addressing the role of these tethers in regulating the function of MCSs at the molecular level. These enormous progresses largely depend on the powerful approaches, including several different types of microscopies and various biochemical techniques. These approaches have greatly accelerated recent advances in MCSs at the system and molecular level. In this review, we summarize the current and emerging approaches for studying MCSs, such as various microscopies, proximity-driven fluorescent signal generation and proximity-dependent biotinylation. In addition, we highlight the advantages and disadvantages of the techniques to provide a general guidance for the study of MCSs.

Keywords: APEX; BioID; FRET; bimolecular fluorescence complementation; electron microscopy; membrane contact sites; proximity ligation assay; super-resolution microscopy.

FIGURE 1

Extensive membrane contact sites (MCSs) in mammalian cells. MCSs exist widely in cell and regulate exchanges signaling and regulates cellular functions. The common strategy for studying MCSs consists of various microscopies and biochemical approaches. Multiple canonical protein complexes have been reported to regulate MCSs, such as VDAC-IP3R-Grp 75 complex at ER-mitochondria MCS. Some MCSs, such as mitochondria-endosome MCS, Golgi-autophagosome MCS and lysosome-peroxisome MCS, are not shown due to limited space in this figure.

FIGURE 2

Overview of various approaches for studying MCSs. Various approaches have greatly accelerated recent advances in MCSs at the system and molecular level. Before 2000, EM and confocal microscopy were most accessible tools. EM provided first evidence of intracellular membrane contact at nanoscale resolution, while confocal microscopy has been employed to reconstruct the 3D imaging of ER-mitochondria juxtaposition as sites of Ca²⁺ transfer between both organelles. Since 2010, super resolution microscopy (SRM) was increasingly used to visualize MCSs. Furthermore, proximity-driven fluorescent signal generation approaches, such as FRET, PLA, BiFC and ddFP were exploited to identify the potential MCSs. Notably, by combination with proteomic analysis, proximity-dependent biotinylation approaches mediated by BioID and APEX provide a promising strategy for global mapping MCSs and identification of tethers involving in MCSs, termed as “tether-omics.” BRET, bioluminescence resonance energy transfer; BioID, proximity-dependent biotin identification; APEX, ascorbate peroxidase; MERLIN, mitochondria-ER length indicator nanosensor; FIB-SEM, focused ion beam-scanning EM; PALM, photoactivation localization microscopy; LLSM, lattice light-sheet microscopy; SR-FACT, super-resolution fluorescence-assisted diffraction computational tomography.

FIGURE 3

Proximity-driven fluorescent signal generation approaches. (A) Proximity ligation assay (PLA). Single-stranded oligonucleotides are conjugated to antibody of proteins. Membrane proximity serves to template the hybridization of circularization oligonucleotides, which is then joined by ligation into a circular DNA molecule and followed by amplification of the signal by rolling-circle amplification (RCA). (B) Fluorescence resonance energy transfer (FRET). The photon energy transfers from excited donor to acceptor since the accessible proximity. (C) Bimolecular fluorescence complementation (BiFC). Accessible proximity made GFP1-10 and GFP11 remodel into a mature fluorescent protein. (D) Dimerization-dependent fluorescent proteins (ddFP). Two GFP monomer (ddGFP-A and ddGFP-B) with weak signal form a dimer with fluorescent strong signal since the accessible proximity.

FIGURE 4

Proximity-dependent biotinylation approaches. (A) BioID- or APEX-based proximity labeling. BirA* (Upper panel) or APEX (Lower panel) were fused to mitochondria membrane protein (bait). The potential target that be thought to interact with bait within the accessible proximity (10 nm for BirA*, 20 nm for APEX) would be biotinylated. For BirA*, endogenous biotin was accessible, instead APEX required the exogenous biotin-phenol and H₂O₂. (B) Cell expressing BirA*-bait or APEX-bait were lysed, then the biotinylated targets were captured by streptavidin-beads and identified by WESTERN BLOT and MS. (C) Reconstitution of split-APEX at ER- mitochondria MCS. By AP fused to TOM20^{1– 34} for targeting OMM and EX fused to Cb5^{100– 134} for targeting ERM, respectively, the split APEX enzyme activity was reconstituted and ER-mitochondria MCS was visualized at higher resolution by EM through the combination of the DAB staining.

FIGURE 5

Considerations of various approaches for studying MCSs. (A) Comparison of various microscopies. Imaging of EM, Cryo-ET and FIB-SEM are fluorescence independent with high resolution (1.8–4 nm), rather than suitable for the dynamics of MCSs in living cell. Comparison to confocal (<0.5 Hz, ∼200 nm) (Schulz et al., 2013), super resolution fluorescence microscopy like STEDM (Schneider et al., 2015), PALM (Betzig et al., 2006; Gould et al., 2009) and STORM (<10 Hz, ∼20 nm) (Rust et al., 2006), GI-SIM (266 Hz and 97 nm) (Guo Y. et al., 2018) as well as combination TIRF with SIM (TIRF-SIM: ∼11 Hz, 100 nm) (Kner et al., 2009) showed extreme high temporal and spatial resolution. (B) The availability of various biochemical techniques. Generally, 12–24 h was required for biotinylation by BioID, while biotinylation based on APEX2 required ∼30 min. BirA* enable to effectively label proteins at distance ∼10 nm, and APEX at ∼20 nm.