Inter-organelle contact sites: molecular modulation and generation of new probes to study their dynamics and associated signaling

Intracellular organelles contact each other by establishing functional membrane contact sites (MCSs), involved in the exchange of signals and metabolites. Dysregulations at the interface between endoplasmic reticulum (ER) and mitochondria (mit) were reported in different pathological conditions, including Alzheimer’s disease (AD). focused my attention on Presenilin 2 (PS2), one of the three proteins that are mutated in familial AD (FAD). PS2 (especially its FAD mutants) is enriched at the level of the ER-mit MCSs, where it increases ER-mit tethering (as revealed by a split-GFP probe) by interacting with Mitofusin 2 (MFN2), thus altering ER-mit calcium (Ca2+) exchange and lipid synthesis/metabolism. I found that the domain of PS2 involved in the binding with MFN2 is its big cytosolic loop and surprisingly, by targeting it on the outer mitochondrial membrane (Mit-PS2-LOOP), I observed an opposite regulation of ER-mit interface respect to the one mediated by both PS2 and FAD-PS2. Indeed, the expression of the Mit-PS2-LOOP in FAD-PS2-N141I patient-derived fibroblasts was able to rectify the increased ER-mit tethering, as well as the higher number and size of lipid droplets, observed in these cells. Next, since the need for new probes to follow MCS dynamics appears urgent, I worked on the generation of a reversible fluorescent reporter to mark MCSs, based on the recently introduced splitFAST reversible system. I engineered the system to allow the fully reversible self-complementation of the whole FAST protein only in those sites where ER and mitochondria are sufficiently close. In this way, ER-mit contacts can be visualized as a green, red or far-red fluorescent signal, by the addition of different fluorogenic substrates, which become fluorescent only when bound to the fully reconstituted FAST protein. The expression of the probe revealed a dot-like fluorescence corresponding to the sites of proximity between the two organelles. Correlative light electron microscopy (CLEM) analysis reported that the expression of this probe does not force the physiological extent of MCSs. Moreover, by different targeting sequences, I engineered the splitFAST constructs to visualize ER-plasma membrane (PM) and mit-PM interactions, or both ER-mit and ER-PM contacts in the very same cell. Taking advantage of the probe reversibility, I followed the dynamics of ER-mit MCSs over time, upon different treatments, in living inducible HeLa clones. As an example, I detected that upon ER stress, there is a short-term increase in the formation of ER-mit MCSs, followed, at later times by a strong reduction. Moreover, the removal of the stressor from the cell media fully restored basal, physiological organelle MCSs. In addition, I observed an increase in ER-mit MCSs in FAD-PS2-N141I patient-derived fibroblasts and astrocytes from a FAD mouse model, compared to respective controls. Considering the highly dynamic complementation of the splitFAST systems, I collected 4D videos of ER-mit MCSs in living HeLa and COS-7 cells. I found that they are highly dynamic, undergoing continuous rearrangements, occasionally marking sites of mitochondria or ER tubule remodeling. Finally, integrating our reporters with Ca2+-binding domains, I was able to detect transient Ca2+ rises at both ER-mit and ER-PM interfaces, while visualizing the contacts, by single reporters. Overall, these results show the multiple potentialities of the new split-FAST probes for the study of MCS dynamics and their role in different physiological and pathological conditions.