Cisd2 ensures adequate ER-mitochondrial coupling, thereby critically supporting mitochondrial function in neurons

Cisd2 ensures adequate ER-mitochondrial coupling, thereby critically supporting mitochondrial function in neurons | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Cisd2 ensures adequate ER-mitochondrial coupling, thereby critically supporting mitochondrial function in neurons Jens Loncke, Ian de Ridder, Rita La Rovere, Annika Vaarmann, Guizhen Fan, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6298090/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Loss of CISD2, an iron-sulfur cluster transfer protein, results in type 2 Wolfram syndrome (WFS2), a disorder associated with severe impacts on pancreatic beta cell and neuronal functions. CISD2 has been implicated in Ca2+ signaling but the molecular basis and cellular consequences remain poorly understood. In this work, we demonstrate that Cisd2 intersects with intracellular Ca2+ dynamics at different levels, including as an interactor of IP3Rs and as a protein contributing to ER-mitochondrial tethering. As such, loss of CISD2 in HeLa cells results in reduced ER-mitochondrial Ca2+ transfer without majorly impact cytosolic Ca2+ signaling. In these cells, CISD2 deficiency promotes autophagic flux, yet has minimal impact mitochondrial function. However, studying the impact of CISD2 deficiency in iPSC-derived cortical neurons, relevant for WFS2, revealed a severe loss of glutamate-evoked Ca2+ responses in cytosol and mitochondria and loss of ER-mitochondrial contact. Correlating with the profound changes in cellular Ca2+ handling, mitochondrial function (oxygen consumption rate, ATP production, mitochondrial potential maintenance) was severely declined, while autophagic flux was increased. Overall, these deficiencies further impact the resilience of CISD2-deficient cortical neurons against cell stress as CISD2-KO neurons were highly susceptible to staurosporine, a cell death inducer. Overall, this work is one of the first to decipher the impact of CISD2 on ER-mitochondrial Ca2+ handling in disease-relevant cell models, thereby revealing a unique dependence of neurons on CISD2 for their mitochondrial health and cell stress resilience. Cell Survival and Cell Death Cell Communication and Signaling General Cell Biology & Physiology Stem Cell & Developmental Cell Biology Cisd2 calcium signaling Wolfram syndrome MAMs neurodegeneration Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Introduction CDGSH iron sulfur domain 2 (Cisd2) is an Fe-S-cluster domain-binding protein, important for cellular homeostasis ( 1, 2 ). Cisd2 together with Wolframin (Wfs1) are the two genes that upon acquiring loss-of-function mutations result in Wolfram syndrome (WS). This rare genetic disease starts during childhood with diabetes mellitus due to loss of functional pancreatic β cells , progressing during adolescence towards neurodegeneration in central and peripheral nervous system ( 3 ). Physiologically, Cisd2-protein levels have been implicated in longevity and the adequate function of other physiological systems besides β cells and neurons, such as skeletal and cardiac muscles ( 4-6 ), liver ( 7-10 ), corneal epithelia ( 11 ), since a reduction of Cisd2 expression (e.g. in heterozygous Cisd2 knockout mice) evokes early aging and organ dysfunctions ( 4 ). At the pathophysiological level, Cisd2 is upregulated in breast cancer, whereby inhibition of Cisd2 results in breast cancer cell death ( 12, 13 ). In hepatocellular carcinoma, however, haplosufficiency of Cisd2 results in hepatocarcinogenesis ( 10, 14 ), underpinning the complexity of Cisd2 in cancer cell biology. Fascinatingly, although WFS1 and Cisd2 belong to two completely different gene families resulting in two proteins with unrelated protein domains and cellular function, loss-of-function of either WFS1 or CISD2 results in Wolfram syndrome ( 3 ). Therefore, we have postulated that Wfs1 and Cisd2 could also have converging roles, such as in Ca 2+ signaling( 3 ). In virtually all cells, ER-originating Ca 2+ signals are mediated by the opening of and Ca 2+ -flux through inositol (1, 4, 5)-trisphosphate (IP 3 ) receptors (IP 3 Rs), intracellular Ca 2+ -release channels ( 15 ). These Ca 2+ signals do not only occur in the cytosol but are also transmitted to the mitochondrial matrix via mitochondria ER contact sites (MERCS) through interorganellar IP 3 R-GRP75-VDAC1-protein complexes ( 16 ). Two important clues hinted towards the involvement of Cisd2 in intracellular Ca 2+ signaling arising from the ER. First, both Wfs1 and Cisd2 have been reported to form protein complexes with IP 3 Rs ( 17, 18 ). Wfs1 was found that to control IP 3 R function by operating in a triparental complex involving neuronal calcium sensor-1 protein (NCS-1) ( 17 ). Instead, Cisd2 (alias: Naf-1) appears to form a complex with IP 3 Rs involving anti-apoptotic Bcl-2 . Fascinatingly, both NCS-1 as well as Bcl-2 are well-established direct interactors of IP 3 Rs ( 19-21 ). Furthermore, Bcl-2 and Cisd2 also directly interact with each other ( 22-24 ). More recently, it was found that Cisd2 alleviated Bcl-2’s ability to reduce the number and extent of MERCS ( 24 ). Second, both Wfs1 and Cisd2 have been found to reside at MERCS ( 24-29 ). Biochemical isolation of the ER membrane fractions involved in MERCS, so-called mitochondrial ER-associated membranes (MAMs), are enriched in both Wfs1 and Cisd2 ( 3, 24, 25 ). Interestingly, in a proteome-wide, proximity biotinylation-based approach (ContactID), Cisd2 was among the top hits of proteins resident at MERCS ( 26 ). Loss of Wfs1 resulted in a reduced MERCS integrity and impaired ER-mitochondrial Ca 2+ transfers, thereby impairing mitochondrial functions such mitochondrial bio-energetics, particularly impacting neurons ( 30 ). MERCS disassembly and dysfunction is also found in cellular and animal models carrying pathogenic Wfs1 mutations identified in Wolfram syndrome patients ( 17, 31 ). Moreover, strategies that restored MERCS organization and Ca 2+ -signaling function such as NCS-1 overexpression ( 17, 32 ) or pharmacological agonists of Sigma-1 receptor, a MERCS-resident chaperone , also improved mitochondrial function ( 31 ). While our understanding of MERCS control by Wfs1 has been majorly advanced the last 5 years, the functional impact of loss of Cisd2 on MERCS remains poorly understood. There has been one report linking patient-linked Cisd2 mutations to an increase of MERCS in patient-derived fibroblasts that also displayed an increased in agonist-induced ER-mitochondrial Ca 2+ transfers, compared to control fibroblasts obtained from unaffected family members ( 33 ). Nevertheless, the impact of wild-type Cisd2 on MERCS, particularly in disease-relevant cells remains elusive, thus exploring its function in neurons is critical. Furthermore, pathogenic mutations of Cisd2 may also introduce gain-of-function aspects and thus may not reflect the biological functions of wild-type Cisd2. Finally, MERCS may differ across individuals irrespective of the Cisd2 gene status, thus it’s instrumental to study in Cisd2 using isogenic conditions. To address this knowledge gap, we set out to study MERCS organization and Ca 2+ -signaling function in cell models in which Cisd2 was knocked out using CRISPR/Cas9. To explore its cell biological function in controlling MERCS, we used HeLa cells, anexcellent cell models to study Ca 2+ signaling and particularly MERCS and its function in ER-mitochondrial Ca 2+ transfer due to the efficient coupling of ER and mitochondria in these cells ( 34 ). In addition to this, to explore the cell physiological impact of loss of Cisd2, we developed Cisd2-KO iPSCs that were differentiated into cortical neurons. In both cell systems, we found that loss of Cisd2 resulted in loss of MERCS and impaired ER-mitochondrial Ca 2+ transfer, revealing an important role of Cisd2 in MERCS integrity irrespective of the cellular context. Yet, while HeLa cells were rather resilient towards loss of Cisd2 for their survival, mitochondrial metabolism and cell death susceptibility, iPSC-derived cortical neurons lacking Cisd2 were particularly susceptible to cell stress inducers such as staurosporine. Furthermore, neurons lacking Cisd2 displayed impaired mitochondrial health, including reduced oxphos and ATP-production capacity. Of note, while loss of Cisd2 had miminal impact on agonist-evoked cytosolic Ca 2+ signals in HeLa cells, it severely impaired such cytosolic Ca 2+ signals in cortical neurons. This unique feature may explain why neurons are highly susceptible to lowered Cisd2-protein levels, thereby accounting for Wolfram syndrome symptoms related to Cisd2 dysfunction. Results Cisd2 Directly Interacts with IP 3 R1 Given the emerging link between CISD2 and intracellular Ca 2+ signaling ( 1, 3, 10, 24, 28, 30, 33 ) and evidence of Cisd2 being engaged in a macrocomplex with IP 3 R1 ( 22 ), we examined whether Cisd2 could directly interact with IP 3 Rs, the main intracellular Ca 2+ -release channels. First, by performing immunoprecipitation (IP) with an anti-IP 3 R1 antibody (Rbt03) and using lysates from HEK293 cells triple KO for all three isoform of IP 3 R(TKO) cells and re-expressing IP 3 R1, we found that endogenous Cisd2 co-IP’ed with IP 3 R1, demonstrating that Cisd2 exists in a complex with IP 3 R1 ( Figure 1A ). To assess whether Cisd2 can directly interact with IP 3 R1, we purified the cytosolic region of the Cisd2 protein (Cisd2 CYT ; Figure 1B for schematic representation) and assessed its binding to purified fully functional IP 3 R1 channel reconstituted into lipid nanodiscs and purified fragments of the IP 3 R1 ( Figure 1C for schematic representation). To acquire quantitative insights in the IP 3 R1/CIS2 CYT complex, we performed microscale thermophoresis (MST) experiments using fluorescently labeled Cisd2 CYT in the presence of increasing concentrations of full-channel IP 3 R1 (10 nM- 5 mM). Representative MST traces are shown in Figure 1D for Cisd2 CYT in absence and presence of full-channel IP 3 R1 (5 mM). The presence of full-channel IP 3 R1 resulted in a thermophoretic shift of Cisd2 CYT , indicating both proteins can directly interact with each other in solution. Fluorescence was normalized to the average baseline values obtained in the blue shaded area, and differences in normalized fluorescence were determined over the range indicated as a red shaded area. Figure 1E shows the averaged shift in thermophoretic shift of Cisd2 in function of [full-channel IP 3 R1] over 5-6s after excitation. Due to the nature of the purified full-channel IP 3 R1 samples (low [full-channel IP 3 R1], nanodisc reconstitution, great size of the full-channel IP 3 R1), the quality of fitting in the binding curve was imperfect. From this fit, we could estimate and determine a binding affinity (K D ) value for its interaction with Cisd2 of about 1.8 µM. To substantiate these findings and to map the Cisd2-interacting IP 3 R region, we performed MST experiments with Cisd2 CYT and several purified fragments of IP 3 R1. Figure 1F shows representative MST traces of fluorescently labeled Cisd2 CYT with and without addition of GST-Fragment 3 of IP 3 R1 (20 µM), showing a shift in thermophoretic mobility of CISD2 CYT in the presence of GST-Fragment 3 of IP 3 R1. Figure 1G shows average shifts in thermophoretic shift of Cisd2 CYT in function of fragment concentrations curves of all screened GST-Fragments of IP 3 R1. By fitting binding curves, we could determine the K D s of Cisd2 CYT /GST-Fragment interactions. We observe the highest affinity for the interaction between Cisd2 CYT and GST-Fragment 3 (K D of ~800 nM), marking aa 923-1581 of IP 3 R1 to harbor a likely binding site for Cisd2 CYT . We also found a lower affinity interaction between Cisd2 and the ligand binding domain (LBD, aa 1-604) and the ligand binding core (LBC, aa 226-604), both with a K D around 2 µM. Cisd2 CYT to SD, fragment 5, fragment 6 of IP3R1 displayed very high K D values of 7.3 µM, 16 µM, and 6.6 µM, thus likely too low affinity for meaningful contribution, while for Fragment 4, no adequate binding curve could fitted. Overall, these interaction studies indicate that in cells Cisd2 forms a complex with IP 3 R1, at least in part via a direct binding of Cisd2 to IP 3 R1, involving the central, modulatory domain and potentially the LBC. Loss of Cisd2 Does Not Alter Basal ER and cytosolic Ca 2+ Levels To assess function effects of Cisd2 on IP 3 R related to Ca 2+ signaling, we generated Cisd2 KO HeLa cell populations using a CRISPR/Cas9 approach in combination with a gRNA targeting Cisd2 (gRNA Cisd2 ), while expressing Cas9 in absence of gRNA Cisd2 served as control cells (CTRL), as described in ( 24 ). Figure 2A and 2B indicate that Cisd2-protein levels are almost absent in Cisd2-KO cells. Additionally, immunofluorescent staining with a monoclonal Cisd2 antibody in HeLa CTRL and Cisd2 KO cells further underpins the absence of endogenous CISD2 proteins in CISD2-KO cells ( Figure 2C ). First, we evaluated the impact of loss of Cisd2 on resting cytosolic Ca 2+ levels by comparing the values of the ratiometric Fura-2 Ca 2+ sensor, loaded as Fura-2-AM in HeLa CTRL and Cisd2 KO cells. Fura-2 enables the measurement of resting cytosolic Ca 2+ levels in a bleaching and loading-independent manner. Fura-2 ratios were recorded using a fluorescence microscope for 30 seconds in single cells ( Figure 2D for representative fluorescence acquisitions). Quantificationof basal Fura-2 ratios during a 30 second recording revealed no major difference in basal [Ca 2+ ] cyt ( Figure 2E ). Furthermore, ER Ca 2+ store content was evaluated by addition of thapsigargin, a high-affinity and irreversible inhibitor of SERCA2, in extracellular Ca 2+ chelating conditions. Averaged traces of Cal520-loaded cell populations of HeLa CTRL and Cisd2 KO cells are shown in Figure 2F . Quantification of area under the curve (AUC) of the thapsigargin-releasable Ca 2+ indicated that absence of Cisd2 did not affect the ER Ca 2+ -store content. Moreover, Ca 2+ responses to ionomycin were recorded to obtain a measure of total intracellular Ca 2+ content ( Figure 2H ). Similarly, total intracellular Ca 2+ content was not affected by absence of Cisd2 ( Figure 2H and 2I ). These results convey that, at least in HeLa cells, Cisd2 is not required for the maintenance of resting Ca 2+ homeostasis. Cisd2 Ensures IP 3 R-Mediated ER-Mitochondrial Ca 2+ Transfer Next, we evaluated the impact of Cisd2 deficiency on IP 3 R-mediated Ca 2+ dynamics and ER-mitochondrial Ca 2+ transfer. Ca 2+ responses to 5 µM ATP, a submaximal agonist concentration, and subsequently 100 µM ATP, a supramaximal agonist concentration, in both the cytosol and mitochondria were simultaneously recorded in single cells and compared between CTRL and Cisd2 KO HeLa cells ( Figure 3A and 3C) . For this, the cytosolic Fluo-4 probe was multiplexed with the genetically encoded red mitochondrial CEPIA3 (R-mtCEPIA3) ( 35 ). In response to 5 µM ATP, both CTRL and Cisd2 KO cells exhibited a robust cytosolic Ca 2+ . Loss of Cisd2 tended to result in a slightly, yet not significant, decreased AUC for 5 mM ATP while the response to 100 µM ATP remain completely unaltered ( Figure 3B ). In contrast, at the level of mitochondria, the effects of Cisd2 were much more pronounced, as the mitochondrial [Ca 2+ ] increases were severely blunted in Cisd2-KO cells compared to CTRL cells. Cisd2 KO cells displayed a blunted Ca 2+ response to 5 µM, thereby reducing the ATP-evoked mitochondrial Ca 2+ uptake ( Figure 3D ). To demonstrate that this change in mitochondrial Ca 2+ dynamics in Cisd2-KO cells was due to an on-target of effect of loss of Cisd2, we performed a rescue experiment, by re-expressing 3x-FLAG tagged Cisd2 in Cisd2 KO HeLa ( Figure 3E ). Similarly, cytosolic and mitochondrial Ca 2+ dynamics were recorded with Fluo-4 and R-mtCEPIA3, respectively ( Figure 3F and 3G ). AUC of cytosolic responses to 5 µM ATP was not significantly altered upon re-expression of Cisd2 ( Figure 3H ). However, at the level of mitochondrial Ca 2+ , Cisd2 enhanced the AUC of mitochondrial Ca 2+ uptake. ( Figure 3I ). These findings indicate that Cisd2 is important for efficient IP 3 R-mediated Ca 2+ transfer from the ER to mitochondria. Cisd2 is Enriched in MAMs independently of IP 3 R In concordance with our previous work ( 24 ), we validated through biochemical MAM isolation that Cisd2 is enriched in the MAM fraction of HeLa CTRL cells. Cisd2-KO cells are used as a reference control ( Figure 4A ). Furthermore, Cisd2 was still enriched in MAM fractions of HeLa-TKO cells that lack all three IP 3 R isoforms ( Figure 4B ), indicating that the localization of Cisd2 to the MAMs is not strictly dependent on the presence of IP 3 Rs. Moreover, we evaluated the protein levels of different MAM-resident proteins via immunoblotting of HeLa CTRL and Cisd2 lysates ( Figure 4C ). Densitometric quantification of key proteins IP 3 R, mitofusin 2 (Mfn2), Vdac1, vesicle-associated protein B (VapB) and Grp75 ( Figure 4D, E, F, G & H ) showed no changed protein levels in absence of Cisd2. Cisd2 Maintains ER-Mitochondrial Contact Site Integrity Next, in the search for a mechanistic explanation why loss of Cisd2 decreased ER-mitochondrial Ca 2+ transfer, we sought to explore the effect of loss of Cisd2 to extent of ER-mitochondrial contact. The short ER-mitochondria SPLICS (SPLICS S ) sensor was transfected in HeLa CTRL and Cisd2 KO cells. The detection of SPLICS S signal is based on a split green fluorescent protein (GFP), that assembles into a functional fluorescent protein only when ER membranes and the OMM are within a distance smaller than 10 nm ( 36 ). Z-stacks of HeLa CTRL and Cisd2 KO cells were aquired using confocal fluorescence microscopy and assembled into volumetric projections. Representative acquisitions of SPLICS S fluorescence, unfiltered emission light serving as a brightfield replacement and volume renders are shown in Figure 5A . Distinct ER-mitochondrial contact sites per cell ( Figure 5B ) and total volume of ER-mitochondrial contact per cell ( Figure 5C ) were significantly decreased in Cisd2 KO cells. To obtain an independent validation of impaired ER-mitochondrial contact formation, we used MAMtracker Green. Contrarily to SPLICS S , MAMtracker Green fluorescence is dependent on reversible dimerization of a dimerization-dependent GFP ( 37 ). MAMtracker Green was co-expressed with Cisd2-p2a-mCherry, or p2a-mCherry control ( Figure 5D ). Correlating with the SPLICS S results, MAMtracker Green fluorescence was significantly decreased in Cisd2 KO HeLa cells compared to CTRL cells. Of note, Cisd2 overexpression fully rescued loss of MAMtracker Green fluorescence, suggesting a recovery in ER-mitochondrial contact ( Figure 5E ). Mitochondrial Health is Unaffected by Cisd2 Deficiency in HeLa Cells To investigate whether Cisd2 loss affects mitochondrial health, mitochondrial volume, morphology and function was assessed. HeLa CTRL and Cisd2 KO cells were co-stained with Mitotracker Green and Cellmask Orange stainings ( Figure 6A ) Morphometric analysis revealed no significant difference in mitochondrial volume, surface area, or branching ( Figures 6B , 6C and 6D ). Furthermore, mitochondrial inner membrane potential, assessed by JC-1 staining, was similar between CTRL and Cisd2 KO cells ( Figure 6E , 6F ). Mitochondrial respiration, measured by a Seahorse Mito Stress Test assay, was not affected by loss of Cisd2 in HeLa cells ( Figures 6G and 6H ). In brief, these results indicate that mitochondrial morphology and metabolism are largely intact despite the loss of Cisd2 in HeLa cells. Cisd2 Deficiency Increases Autophagic Flux Since Chang et al. reported that Cisd2 is required for Bcl-2’s ability to inhibit Beclin 1-dependent autophagy in H1299 cells ( 22 ), we further studied autophagic flux. We found that Cisd2-KO cells displayed an increased autophagic flux, as indicated by an elevation of detected LC3-II levels in response to blocking autophagosomal-lysosomal fusion with bafilomycin A1 ( Figure 7A and 7B ). Additionally, when transfecting HeLa CTRL and Cisd2 KO cells with GFP-LC3, the number of detected GFP-LC3 punctae was significantly higher in bafilomycin A1 treated Cisd2 KO cells, suggesting enhanced autophagosome formation ( Figure 7D ). These findings highlight a potential link between Cisd2 and the regulation of autophagy. Autophagy is a process that can precede apoptosis ( 38 ). To evaluate a potentiating effect to apoptotic stimuli in absence of Cisd2, HeLa CTRL and Cisd2 KO cells were subjected to several concentrations of staurosporine, a potent inducer of apoptosis. Apoptosis was monitored by poly (ADP-ribose) polymerase (PARP) cleavage detecting through immunoblotting, an established ratiometric readout for apoptosis. A representative immunoblot is shown in Figure 7E . No heightened sensitivity to staurosporine, nor a significantly elevated level of apoptosis was observed in Cisd2 KO HeLa cells ( Figure 7F ). iPSC-derived Cisd2-Deficient Cortical Neurons Display Diminished IP 3 R-Mediated Ca 2+ Release As loss of Cisd2 in HeLa cells impacted ER-mitochondrial Ca 2+ transfer but without major impacts on cell function, we decided to study Cisd2 in a WS disease-relevant cellular context by using control and Cisd2-KO iPSCs into cortical neurons. The same approach as in HeLa cells was used to generate the knockout of Cisd2 in iPSCs. After 72h of selection with puromycin, complete loss of detectable Cisd2 was attained ( Figure 8A ). iPSCs still express purinergic receptors, enabling to study IP 3 R-mediated Ca 2+ release through addition of ATP in Cal520 loaded iPSCs ( Figure 8B ). Similarly to HeLa cells, loss of Cisd2 did not result in a significantly decreased sensitivity to ATP-evoked IP 3 R-mediated Ca 2+ releases ( Figure 8C ). Figure 8D displays the timeline of differentiation from iPSC to cortical neurons. After 54 days of differentiation, differentiated cortical neurons were used for experiments. Of note, differentiated cortical neurons expressed SATB2 and MAP2 markers, as shown using immunostaining ( Figure 8E ). Figure 8F shows a proof-of-concept averaged trace of cortical neurons loaded with Cal520, responding to glutamate as an agonist and 1:10 glycine as a co-agonist. Extracellular Ca 2+ was chelated, and cortical neurons were pretreated with 30 µM, effectively abolishing any Ca 2+ flux through RyRs. This demonstrates the observed Ca 2+ response to be bona fide IP 3 R-mediated, corresponding with work of others ( 39 ). Cal520-loaded cortical neurons were recorded responding to 10 mM glutamate, supplemented with 1 mM glycine, and 100 mM glutamate, supplemented with 10 mM glycine, to generate a supramaximal response ( Figure 8G ). Responses to both concentrations of glutamate were severely diminished in Cisd2 KO cortical neurons ( Figure 8H ). Conversely, thapsigargin-releasable Ca 2+ was significantly elevated in Cisd2 KO cortical neurons. Thus, in the disease relevant context of cortical neurons, loss of CISD2 also impairs Ca2+ release in the cytosol, which was not due to a depletion of the ER Ca 2+ store content. Cisd2 Is Essential for ER-Mitochondrial Contact Integrity and Mitochondrial Function in Cortical Neurons The observed decreased IP 3 R-mediated Ca 2+ release in Cisd2-deficient cortical neurons might hold repercussions for fluxed Ca 2+ to the mitochondria. Green mtCEPIA2 ( 40 ) was transfected in CTRL and Cisd2 cortical neurons and cells were stimulated with 100 mM glutamate / 10 mM glycine as a supramaximal agonist of the IP 3 R (Figure 9A ). In correspondence with the detected responses in the cytosol, AUC, amplitude and rate of rise of mitochondrial Ca 2+ responses were significantly diminished in absence of Cisd2 ( Figure 9B ). By expressing SPLICS S in cortical neurons, we evaluated the decreased mitochondrial Ca2+ uptake was due to a decreased Ca 2+ efflux from the ER or can also be due to a decrease in ER-mitochondrial contact ( Figure 9C ). The number of ER-mitochondrial contact sites per cell were significantly decreased in cortical neurons deficient for Cisd2 ( Figure 9D ). In addition, MAMtracker Green was expressed, together with Cisd2-p2a-mCherry or p2a-mCherry EV as a control ( Figure 9E ). Normalized MAMtracker Green fluorescence was significantly decreased in absence of Cisd2, while re-expressing Cisd2 in Cisd2 KO cortical neurons effectively restored MAMtracker Green fluorescence. Loss of CISD2 impairs mitochondrial function in iPSC-derived cortical neurons Given the more profound impacts of loss of CISD2 on intracellular Ca2+ dynamics in iPSC-derived cortical neurons, we wondered whether mitochondrial function was more severely perturbed by CISD2 deficiency in these cell systems. We therefore assessed mitochondrial health using Seahorse Mito Stress Test assays in control and CISD2-KO iPSC-derived cortical neurons ( Figure 9G-J). In contrast to HeLa cells,in iPSC-derived cortical neurons,mitochondrial health was severely affected by loss of Cisd2, evidenced by significantly decreased basal OCR ( Figure 9H ), lowered ATP-linked respiration ( Figure 9I ) and a decreased maximal capacity ( Figure 9J ) in mitochondria of Cisd2-deficient cortical neurons. Furthermore, CTRL and Cisd2 KO cortical neurons were stained with JC-1 to evaluate maintenance of IMM potential ( Figure 9K ). Cisd2-deficiency led to a significant loss of IMM potential, as normalized JC-1 ratios were significantly decreased compared to CTRL neurons ( Figure 9L ). To expand these findings, we also determined the neuronal ATP context in rat neonatal cortical neurons using PercevalHR, an established sensor for intracellular ATP / ADP ratio ( 41 ). In these cells, we performed shRNA-mediated knockdown of CISD2, which evoked a decrease in the PercevalHR ratio ( Figure 9M ), indicative for a lower intracellular ATP content in cells lacking CISD2. (7). Re-expressing Cisd2 in these shRNA-CISD2-treated cells rescued the loss of ATP content. Loss of CISD2 Increases Autophagic Flux and Sensitizes Neurons to Apoptosis Finally, we assessed whether autophagy flux and apoptosis sensitivity were affected by loss of CISD2 in iPSC-derived cortical neurons. To determine autophagic flux, we used the GFP-RFP-LC3, whereby GFP is effectively quenched in lysosomal pH conditions, while RFP fluorescence is unaffected ( Figure 10A ). As such, the GFP / RFP autophagosomal punctae ratio can be used as a readout for autophagic flux. GFP / RFP punctae ratios was significantly decreased in Cisd2 KO cortical neurons compared to CTRL neurons, indicating more autophagic turnover by the lysosomes. This decrease in GFP/RFP ratio observed in CISD2-KO cells was abrogated by the addition of bafilomycin A1, a lysosomal inhibitor effectively elevated ( Figure 10B ). Finally, we evaluated whether the impacts on mitochondrial health could sensitize the cortical neurons to cell stress treatments such as staurosporine. Therefore, CTRL and Cisd2 KO cortical neurons were subjected to different doses of staurosporine and PARP ratios were determined through immunoblotting ( Figure 10C ). Volumetric quantification of cleaved PARP over total PARP indicated that Cisd2-KO iPSC-derived cortical neurons were much more sensitive to staurosporine than CTRL iPsC-derived cortical neurons in two ways: first, the threshold for inducing apoptosis in Cisd2-KO neurons is lower than that of control neurons (cfr 100 nM condition); second, at higher staurosporine concentrations (300 nM-1microM), Cisd2-KO neurons display more apoptosis than control neurons ( Figure 10D ). Discussion The main findings of this work are that Cisd2 can directly interact with IP 3 R channels, intracellular Ca 2+ -release channels not only delivering Ca 2+ towards cytosol but also towards mitochondria. Loss of Cisd2 impaired MERCS organization with subsequent lowered ER-mitochondrial Ca 2+ transfer, both in HeLa and in neurons. Yet, only neurons – not HeLa cells - displayed severe perturbations in their cellular functions upon loss of Cisd2, showing decreased mitochondrial potential, decreased oxidative phosphorylation and ATP production, accompanied with increased autophagic flux. As a consequence, Cisd2-lacking neurons became susceptible to chemical cell stress inducers such as staurosporine, while Cisd2-lacking HeLa cells were rather unaffected. Consistently, loss of Cisd2 had more profound impacts on Ca 2+ dynamics in neurons compared to HeLa cells, heavily suppressing cytosolic Ca 2+ signals in neurons but not in HeLa cells exposed to physiological agonists. Of note, autophagy flux appeared increased in both HeLa and in neurons lacking Cisd2, thus indicating that this process may be (in part) unrelated to impaired metabolism. This hints towards other functions for Cisd2 in control of autophagy. In fact, Cisd2 has been reported to function as an important co-factor for Bcl-2, which operates as an anti-autophagic protein by scaffolding Beclin 1 ( 22, 42 ). Hence, in absence of Cisd2, it’s possible that Bcl-2, endogenously present in HeLa and neurons, is impaired in suppressing autophagy, thereby explaining the increased autophagic flux in both HeLa cells and neurons. Though other mechanisms might be at play, since Cisd2 has been reported to prevent starvation-induced autophagy by inhibiting AMPK activity, at least in cardiomyocytes ( 43 ). A recent unbiased proteome-wide analysis of MAMs using Contact-ID, a Bio-ID-based labeling approach of proteins proximal to ER-mitochondrial contact sites revealed 115 MAM-specific proteins of which Cisd2 was ranked among the top candidates ( 26 ). Validating this screen, our findings indicate that, consistent with other reports ( 24, 26, 28 ), that Cisd2 is a critical component for the adequate organization of MERCS in cells. Being one of the two Wolfram-linked genes besides WFS1 , it appears that Wfs1 and Cisd2 execute converging functions at MERCS, as loss of either Wfs1 or Cisd2 results in decreased MERCS and impaired ER-mitochondrial Ca 2+ transfer. Moreover, recent work indicated that Wfs1 and Cisd2 can interact with each other and that impaired ER Ca 2+ release by loss of Wfs1 could be compensated by overexpression of Cisd2 ( 30 ). Furthermore, it’s fascinating that Cisd2 via its cytosolic region can directly interact with IP 3 Rs, yet this interaction does not seem to functionally control IP 3 R function as addition of the purified cytosolic Cisd2 fragment did not change the open probability of single IP 3 R channels ( 24 ). This correlates with our findings in HeLa cells that displayed comparable agonist-evoked Ca 2+ signals in the cytosol irrespective of the presence or absence of Cisd2 ( 24 ). Of note, in those conditions, we have used intermediated [agonist], so that in principle both sensitizing and inhibitory effects should have been detectable. Instead in neurons, loss of Cisd2 did severely hamper cytosolic Ca 2+ rises evoked by the physiological agonist glutamate. Moreover, these experiments were performed in absence of extracellular Ca 2+ , thus ensuring the Ca 2+ release originated from the ER through IP 3 Rs. Of note, ryanodine addition, thereby blocking ryanodine receptor channels, did not majorly lower glutamate-evoked Ca 2+ rises further underpinning that those signals are mainly mediated through IP 3 Rs, which is consistent with previous observations reported by other teams ( 39 ). The diverging impact of Cisd2 on intracellular Ca 2+ dynamics dependent on the cell type is fascinating and requires further study. Hence, it will be important to unravel the mechanisms responsible for this effect. It’s tempting to speculate that in neurons Cisd2 cooperates with other factors/proteins to support the adequate function of IP 3 Rs in mediating Ca 2+ release. For instance, the impact of Wfs1 on MAMs and ER-mitochondrial Ca 2+ transfers is executed via Neuronal Calcium Sensor-1 (NCS-1). Thus, it’s possible that NCS-1, which is an IP 3 R-accessory protein that enhances IP 3 R function ( 19 ), too cooperates with Cisd2, whereby loss of Cisd2 could affect the presence of NCS-1 in complex with IP 3 Rs, thereby limiting its function. Moreover, the importance of the IP 3 R-Cisd2 complexes and such potentially additional proteins for Cisd2’s impact on ER-mitochondrial integrity requires further scrutiny. Therefore, it will be of interest to assess whether Cisd2 requires the presence of IP 3 Rs to sustain ER-mitochondrial contact sites. Furthermore, putting our findings in context with previous study performed in patient fibroblasts carrying Cisd2 mutations, it was found that ER-mitochondrial Ca 2+ was increased ( 33 ). In that study, it was found that Cisd2-mutated fibroblasts displayed increased ER-mitochondrial Ca 2+ transfer compared to the age-matched controls and that ER-mitochondrial contact sites (in terms of contact ER/mito / mm 2 , number of mitochondria with ER contact sites and # mitochondria length adjacent to ER) are increased. Of interest, also in these patient fibroblasts, that represent a cell type likely not affected in WS2, mitochondrial morphology and oxidative phosphorylation function was only limitedly affected, thereby resembling our findings obtained in HeLa cells at least for mitochondrial function. Therefore, further work is needed to assess how Cisd2 mutations impact ER-mitochondrial contact sites and ER-mitochondrial contact site formation. The differences with our findings are striking but could be due to different reasons. First, in our work, we have used isogenic cell models thereby allowing a careful comparison between wild-type and KO models, while Rouzier et al used cells derived from different individuals, though being close relatives ( 33 ). Second, in our study, we have focused on KO/knockdown approaches thus lowering overall Cisd2-protein levels, while Rouzier et al have studied the impact of Cisd2 patient mutations associated with aberrant Cisd2 function. Hence, it’s certainly possible that loss of Cisd2 impairs ER-mitochondrial contacts while patient-associated Cisd2 mutations display gain-of-function properties for certain Cisd2-related functions. Therefore, further work will be needed to explore the impact of these functions on cell function. Third, our work also reveals the critical importance of the cell type, and thus the impact of Cisd2 mutations in the context of neurons, a cell type affected in WS2, could certainly be different than in fibroblasts, a cell type not affected in WS2. Finally, it will be of interest to explore whether strategies that improve ER-mitochondrial Ca 2+ transfer can restore mitochondrial function and improve cell stress resilience in neurons. While genetic linkers could offer proof-of-concept evidence, such strategies are less tangible to translate into therapeutically applicable interventions. An attractive target is Sigma-1 receptor, an ER-resident chaperone important of ER-mitochondrial contact sites and Ca 2+ transfer. Recent developments in the pharmacology of Sigma-1 receptors have resulted in small molecule Sigma-1 receptor agonists such as PRE-084 ( 31 ). Excitingly, PRE-084 has recently been applied in preclinical models of WS1 including cellular models and zebrafish, thereby alleviating several symptoms of disease. Hence, the application of PRE-084 in cells lacking Cisd2 to rescue mitochondrial deficits could be of interest. In conclusion, our study highlights the critical role of Cisd2 in ER-mitochondrial communication and its impact on cellular function, particularly in neurons. We demonstrate that Cisd2 directly interacts with IP 3 Rs and is essential for maintaining MERCS integrity and ER-mitochondrial Ca 2+ transfer. Notably, the loss of Cisd2 in neurons leads to a greater extent of metabolic and Ca 2+ signaling defects than in HeLa cells, underscoring the cell-type-specific functions of Cisd2. These findings raise important questions regarding the molecular mechanisms by which Cisd2 modulates intracellular Ca 2+ dynamics, potentially involving interactions with other regulatory proteins such as NCS-1. Furthermore, the contrasting effects observed in patient-derived fibroblasts by others ( 33 ) suggest that distinct Cisd2 mutations may differentially affect ER-mitochondrial contact sites and function. Future studies should focus on elucidating these mechanisms and exploring therapeutic strategies to restore ER-mitochondrial Ca 2+ transfer. The potential of Sigma-1 receptor agonists, such as PRE-084, to rescue mitochondrial deficits presents an exciting avenue for further investigation in Cisd2-related pathologies. Materials and Methods Chemicals and consumables Unless specifically stated, all chemicals and consumables were obtained from Thermo Fischer (Merelbeke, Belgium). Antibodies Rabbit anti-IP 3 R1 (alias: Rbt03, 1:1000, homemade) ( 44 ), rabbit polyclonal anti-CISD2 (1:1000, ABClonal, A5231), mouse monoclonal anti-IP 3 R1 E-8 (5 µg per coIP, Santa cruz, sc-271197), mouse aspecific IgG (5 µg per coIP, sc-2025), mouse monoclonal anti-GAPDH (1:1000, Merck, G8795), mouse monoclonal anti-Cisd2 (1:100, Proteintech, 66082-1-Ig), goat anti mouse (1:1000, A11017), mouse monoclonal anti-vinculin (1:10000, Merck, V9131), rabbit monoclonal anti-PERK (1:1000, Cell Signaling Technology, 3192), rabbit monoclonal anti-calnexin (1:1000, Cell Signaling Technology, 2679S), mouse monoclonal anti-VDAC1 (1:1000, Abcam, ab14734), rabbit polyclonal anti-Cytochrome C (1:1000, Cell Signaling Technology, 4272), rabbit pan anti-IP 3 R1 (alias: Rbt475, 1:1000, homemade), rabbit polyclonal anti-mitofusin 2 (1:1000, Abcam, ab50838), mouse monoclonal anti-α Tubulin (1:500, Thermo Fischer, A11126), rabbit polyclonal anti-VABP (1:1000, Invitrogen, PA5-53023), rabbit polyclonal anti-LC3 (1:500, Cell Signaling Technology, 4108S), rabbit monoclonal anti-PARP (1:1000, Cell Signaling, 9532S). Plasmids and constructs For the purification of Cisd2 CYT the cDNA sequence of the cytosolic part of the protein was cloned as a gBlock™ (Integrated DNA Technologies, Leuven, Belgium) in a pET21b(+) plasmid for purification using standard His-purification protocols, as previously performed ( 24, 45 ). gBlock™ sequence: 5’-TACTTCTTTCTTCTTCAGTATTAGTGGACCCACATTATCTCCTGTCAATTCATTGTGTTTATTATGTGAACCATCGCAGG CAGGAAACGTTTTAGAACGCCAACACCTACAATAAGCTGCTTTAGTAAGACACAAATCTTCAATGTTTATTTCATTCACTAC TTTCGGATTTTCCTTTTGTATTTTAAGATTAATCAAGCTATCC TTCTGTTGTTTCTTCTTCGGGAGGAATGGACGCAT-3’. pCMV26-Cisd2 was generated by cloning the cDNA sequence of Cisd2 using HindIII and EcoRI as described before ( 24 ). pCMV24-Cisd2-p2a-mCherry was subcloned as follows: a gBlock™ containing the cDNA for Cisd2 (Integrated DNA Technologies, Leuven, Belgium) was restriction-ligated with HindIII and EcoRI into an empty pCMV24-p2a-mCherry, created as described in ( 46 ). gBlock sequence: 5’-GCCAAGCTTGTGCTGGAGAGCGTGGCCCGTATCGTGAAGGTGCAGCTCCCTGCATATCTGA AGCGGCTCCCAGTCCCTGAAAGCATTACCGGGTTCGCTAGGCTCACAGTTTCAGAATGGCTTCGGTTATTGCC TTTCCTTGGTGTACTCGCACTTCTTGGCTACCTTGCAGTTCGTCCATTCCTCCCG AAGAAGAAACAACAGAAGGATAGCTTGATTAATCTTAAAATACAAAAGGAAAATCCGAAAGTAGTGAATGAAATA AACATTGAAGATTTGTGTCTTACTAAAGCAGCTTATTGTAGGTGTTGGCGTTCTAAAACGTTTCCTGCCTGCGATGGTTC ACATAATAAACACAATGAATTGACAGGAGATAATGTGGGTCCACTAATACTGAAGTACCCATACGATGTTCCAGATTACGCT AAGAAAGAAGTAGAATTCGCCGC-3’. pSpCas9(BB)-2A-Puro (PX459) was a gift from Feng Zhang (Addgene plasmid # 48139; http://n2t.net/addgene:48139; RRID:Addgene_48139) ( 47 ). pCMV R-CEPIA3mt was a gift from Masamitsu Iino (Addgene plasmid # 140464 ; http://n2t.net/addgene:140464 ; RRID:Addgene_140464). SPLICS Mt-ER Short P2A was a gift from Marisa Brini & Tito Calì (Addgene plasmid # 164108 ; http://n2t.net/addgene:164108 ; RRID:Addgene_164108). GW1-PercevalHR was a gift from Gary Yellen (Addgene plasmid # 49082 ; http://n2t.net/addgene:49082 ; RRID:Addgene_49082). pCMV CEPIA2mt was a gift from Masamitsu Iino (Addgene plasmid # 58218 ; http://n2t.net/addgene:58218 ; RRID:Addgene_58218). Cell culture and transfections HEK293 IP 3 R TKO cells expressing rat IP 3 R1 were kindly gifted by Dr. Yule and were cultured at 37 °C, 10 % CO 2 in Dulbecco’s Modified Eagle’s Medium (DMEM), supplemented with 10 % fetal calf serum, 100 IU/mL penicillin and 100 μg/mL streptomycin, 2 mg/mL geneticin. HeLa cells were cultured at 37 °C, 5 % CO 2 in Dulbecco's Modified Eagle's Medium (DMEM), supplemented with 5 % fetal calf serum, 100 IU/mL penicillin and 100 μg/mL streptomycin, 2 mM Glutamax. HeLa cells were seeded 24 h before transfection and were transfected using Mirus TransIT-X2 transfection reagent (Mirus Bio, WI, USA) with a 2:1 transfection reagent in μL per μg DNA. Cells were routinely checked for the absence of mycoplasma infection. hiPSCs were seeded on human Matrigel coated plates (Corning, Lasne, Belgium, 354277) in mTESR (StemCell Technologies, Cambridge, UK, 8580) with 1:100 Revitacell (Life Technologies, Bleiswijk, The Netherlands, A2644501). Maintenance of cultures was performed in E8 flex medium (E8 basal medium complemented with E8 supplement Flex (A2858501) and 5 U/mL penicillin-streptomycin) and splitting twide a week with 0.5 mM EDTA. Differentiation from hiPSCs to cortical neurons was performed as extensively described in ( 48 ). All experiments were performed on neurons differentiated for between 52 and 59 days. Cortical neurons were transfected using Mrius TransIT-X2 with a 3:1 transfection reagent in µL per µg DNA. Subsequently, 6 h after transfection, neuronal maintenance medium was replaced to neuronal maintenance medium supplemented with 0.5 mM kynurenic acid. Generation of Cisd2 KO and Cas9 CTRL HeLa cells and iPSCs CTRL and Cisd2 KO HeLa cells were generated as described before ( 24 ). In brief, HeLa cells were transfected with pSpCas9(BB)-2A-Puro vectors and selected for 48 h in presence of 3 µg/mL puromycin. CTRL and Cisd2 KO iPSCs were generated by seeding iPSCs on 6 well plates coated with 1 % Geltrex™ (A1413301). 24 h after seeding, medium was changed to mTESR and supplemented with 1:500 ROCK inhibitor (Tocris, Bristol, UK, Y-27632). 2 h after, iPSCs were nucleofected (Lonza) with 3 µg pSpCas9(BB)-2A-Puro vectors. Cells were selected with 0.4 µg/mL puromycin for 72 h. Guide RNA sequence (GGAGCTGCACCTTCACGATA) was obtained from Synthego (Redwood City, CA, USA). Protein purification BL21 (DE3) E. coli bacteria were transformed with the plasmids pGEX6p2 inserted with the following fragments of the ratIP3R1: suppressor domain (SD), ligand binding region (LBR), ligand binding region (LBR), fragment 3, fragment 4, fragment 5, and fragment 6. After growing at 37 °C to an A600 of 0.2, the culture was incubated at 14 °C for 20 h with 100 μM isopropyl-d-1-thiogalactopyranoside for protein expression. Benzamidine and Fenylmethylsulfonylfluoride (PMSF) were added to a final concentration of 0.23 mM and 0.83 mM respectively. The bacteria were harvested via centrifugation at 5000 ×g for 10 min and resuspended in phosphate-buffered saline (PBS) containing 1% Triton X-100. Samples were sonicated 15 times for 10 seconds at 20 kHz using a MSE Ltd. (Westminster, Great Britain) probe sonicator with a 10 min pause after each 5 sonication rounds. The resulting bacterial lysate was centrifuged at 10,000 rpm for 20 min using a Sorvall SS-34 rotor at 4°C. Supernatants were collected and incubated with glutathione-Sepharose 4B beads for 2 hours at 4 °C. Subsequently, the beads were washed two times using 6 mL of the following buffers supplemented 0.83 mM benzamidine and 0.23 mM PMSF respectively: 1% Triton X-100 in PBS, PBS, and 50 mM Tris-HCl (pH 8.0). The GST-tagged proteins were eluted with 50 mM Tris-HCl containing 10 mM glutathione (pH 8.0) and dialyzed with PBS. The resulting protein purity was evaluated via western blotting and Coomassie blue staining (Imperial Protein Stain, Thermo Fisher, Cat. 24615). Images of the gel were taken using the ChemiDoc™ MP imaging system and Image Lab 6.1 software (Bio-Rad, Hercules, CA, USA). Co-immunoprecipitation assays CTRL and Cisd2 KO HEK IP 3 R TKO overexpressing IP 3 R1 were seeded 72 h before harvesting. Cells were harvested by scraping in ice cold PBS and lysed with a 3-((3-cholamidopropyl) dimethylammonio)-1-propanesulfonate (CHAPS) based buffer (50 mM Tris, 100 mM NaCl, 2 mM Ethylenediaminetetraacetic acid, 50 mM NaF, 1 mM Na3VO4, 1 % CHAPS and protease inhibitor tablets (Roche, Basel, Switzerland) according to manufacturer’s instructions). 5 µg of anti IP 3 R1 E-8 antibody or mouse aspecific IgG antibody was incubated with 20 µL of protein G coated Dynabeads™ (ThermoFischer Scientific, Bleiswijk, the Netherlands) at room temperature for 30 minutes. Next, beads were incubated with 1000 µg of cell lysate overnight at 4 °C in CHAPS lysis buffer. Subsequently, beads were washed with ice cold PBS. Proteins were eluted from the beads using SDS elution buffer (0.2 % SDS + 0.01 % Tween pH 8). Immunoblotting NuPAGE™ LDS Sample Buffer was added to all immunoblot samples which were boiled and ran on NuPAGE™ 4–12 % Bis-Tris gels. Subsequently, proteins were transferred on a polyvinylidene fluoride membrane. Membranes were blocked with Tris-buffered saline (TBS) containing 5 % milk powder and 0.1 % Tween and incubated with primary antibody overnight. The next day, membranes were incubated for 1 h with secondary horseradish peroxidase-linked antibodies in TBS 0.1 % Tween. Pierce™ ECL chemiluminescent western blot reagent was used for detection in a Chemidoc imaging system (Bio-Rad, CA, USA). Microscale thermophoresis Using the Monolith His-Tag Labeling Kit RED-tris-NTA 2nd Generation (Nano Temper Technologies, Munich, Germany) purified 6xHis-tagged CISD2 CYT protein was fluorescently labeled. The fluorescently labeled Cisd2 CYT was used to determine binding affinities for purified full-channel rIP3R1 (obtained in collaboration with the lab of Irina Serysheva) as well as purified GST-tagged fragments of rIP3R1 through microscale thermophoresis using a Monolith NT automated instrument (Nano Temper Technologies), as previously published [PMID: 39370046]. The concentration of labeled CISD2 CYT was maintain at a 10 nM. The concentrations of the full-channel IP3R1 and IP3R1 fragments vary depending on the biological replicate and were optimized to ensure datapoints at higher protein concentrations, thus enabling more accurate prediction of any possible interaction. Measurements were performed with a pico-red laser channel at 20 % excitation and 40 % MST power in steady-state conditions using premium capillaries. Fluorescence was normalized to the average baseline values obtained in the cold region (blue shaded area, -1 s until 0 s) and differences in normalized fluorescence (F / F0) were determined over the hot region (red shaded area, 5 s until 6 s). For our analysis of the interaction between Cisd2 and full-channel IP3R1 an alternative hot region of 1.5 s – 2.5 s was selected to ensure proper evaluation of the thermophoretic shift in each of the evaluated conditions. All experiments were repeated at least 2 times for each condition using freshly thawed proteins with at least 2 technical replicates for each biological replicate. Immunofluorescent staining and imaging HeLa cells were seeded on coverslips 48 h before fixation with 4 % paraformaldehyde (room temperature, 15 minutes), after which cells were washed 3 times with PBS. Cells were permeabilized with 0.1 % Triton-X-100 in PBS (room temperature, 10 minutes) and were washed 3 times with PBS. Permeabilized cells were blocked with 4 % bovine serum albumin and 0.1 % Triton-X-100 in PBS (room temperature, 1 h). Cells were stained with 1:100 mouse monoclonal anti-Cisd2 in blocking solution (4 °C, overnight). The next day, cells were washed 2 times with PBS and stained with Alexa 488 linked goat anti-mouse (1:1000 in blocking solution) (room temperature, 2 h). Cells were washed 3 times and kept in PBS for imaging on a fluorescence confocal microscope (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4). Excitation: 488 nm, emission: 530 nm. Single cell Fura-2 Ca 2+ imaging For Fura-2 measurements, a Nikon TI2-E inverted microscope equipped with a 20 × 0.5 NA Plan Fluor DIC N2 air objective and a pco.edge 4.2bi sCMOS camera was used. Fura-2 was alternatingly excited at a 2 s interval using a CoolLED pE-300 ultra/pE-340 lamp set at 340 nm and 380 nm (CoolLED, Andover, UK), and using a dichroic mirror FF02-409/LP-25 (Semrock, New York, USA) and a bandpass emission filter 515/30 (Semrock, Rochester, USA). mCherry was visualized using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 550 nm, and using the cubical excitation filterset FF01–378/474/554/635 and a bandpass emission filter 595/31 (Semrock, Rochester, USA). All cells were seeded in 4 chamber slides with coverslips (IBL, Gerasdorf bei Wien, Austria). HeLa CTRL and Cisd2 KO cells, were loaded for 30 min, at room temperature, with 1 μM FURA-2-AM (AnaSpec, Fremont, CA, USA) in modified Krebs buffer (135 mM NaCl, 6.2 mM KCl, 1.2 mM MgCl 2 , 12 mM HEPES, pH 7.3, 11.5 mM glucose and 1.5 mM CaCl 2 ). Cells were washed once with modified Krebs and the Fura-2-AM was allowed to de-esterify during 30 min at room temperature. After measuring for about 30 s, 3 mM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA) in Krebs without Ca 2+ was added to chelate extracellular Ca 2+ . Basal FURA-2 ratios were recorded for 30 s. a Nikon TI2-E inverted microscope equipped with a 20 × 0.5 NA Plan Fluor DIC N2 air objective and a pco.edge 4.2bi sCMOS camera was used. FURA-2 was alternatingly excited at a 2 s interval using a CoolLED pE-300 ultra/pE-340 lamp set at 340 nm and 380 nm (CoolLED, Andover, UK), and using a dichroic mirror FF02-409/LP-25 (Semrock, New York, USA) and a bandpass emission filter 515/30 (Semrock, Rochester, USA). mCherry was visualized using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 550 nm, and using the cubical excitation filterset FF01–378/474/554/635 and a bandpass emission filter 595/31 (Semrock, Rochester, USA). All cells were seeded in 4 chamber slides with coverslips (IBL, Gerasdorf bei Wien, Austria). Data was quantified using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/FURA Ca 2+ imaging by camera-based multiplate imaging on cell populations HeLa CTRL and Cisd2 KO cells were seeded on black 96 well plates 48 h before acquisition. Cells were loaded with 3 µM Cal520 + 0.04 % pluronic acid in modified Krebs and incubated for 30 minutes at room temperature. Cells were washed twice and left for de-esterification for 30 minutes at room temperature. Basal fluorescence was measured for 30 seconds, then EGTA was added. Thapsigargin/ionomycin was added at 90 seconds, and acquisition was continued for 10 min at a one second acquisition interval. CTRL and Cisd2 KO iPSCs were seeded 72h before acquisition on Geltrex-coated black 96 well plates. Cells were loaded with 3 µM Cal520 + 0.04 % pluronic acid in E8 Flex medium and incubated for 30 minutes at 37 °C 5% CO 2 . Cells were washed twice and left for de-esterification for 30 minutes. Basal fluorescence was measured for 30 seconds, then various ATP concentrations were added. Cortical neurons were seeded 52-59 days before acquisition. Cells were loaded with 3 µM Cal520 + 0.04 % pluronic acid in neuronal maintenance medium and incubated for 30 minutes at 37 °C 5% CO 2 . Cells were washed twice and left for de-esterification for 30 minutes. Basal fluorescence was measured for 30 seconds, then EGTA was added. Various concentrations of glutamate/glycine were added at 120 seconds, followed by 2.5 µM ionomycin at 300 seconds. Fluorescence was acquired every second. Plates were imaged using an FDSS/µCell kinetic plate imager C13299 (Hamamatsu). Data was further processed using a custom Python script: https://github.com/jensloncke/LMCS-python-scripts/tree/main/FDSS%20single%20response Multiplexed single cell cytosolic and mitochondrial Ca 2+ imaging For these measurements, a Nikon TI2-E inverted microscope equipped with a 40 × 1.3 NA Plan Fluor DIC H N2 Oil objective and a pco.edge 4.2bi sCMOS camera was used. Fluo-4 was excited at a 2 s interval using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 488 nm, the cubical excitation filterset FF01–378/474/554/635, and using a bandpass emission filter 515/30 (Semrock, Rochester, USA). R-mtCEPIA3 was visualized using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 550 nm, and using the cubical excitation filterset FF01–378/474/554/635 and a bandpass emission filter 595/31 (Semrock, Rochester, USA). HeLa cells were seeded in 4 chamber slides with coverslips and transfected with 200 ng of R-mtCEPIA3 ( 35 ). Two days after transfection, cells were loaded for 30 min, at room temperature, with 1 μM Fluo-4-AM (F14201) in modified Krebs buffer. Cells were washed once with modified Krebs and the Fluo-4-AM was left was allowed to de-esterify during 30 min at room temperature. Cells were washed twice after de-esterification. After measuring for about 30 s, EGTA was added in Krebs without Ca 2+ to chelate extracellular Ca 2+ . After 90 s, responses of mtCEPIA transfected cells to 5 µM ATP were recorded. 180 s after 5 µM ATP addition, cells were stimulated supramaximally with 100 µM ATP. Data was quantified using a custom ImageJ script, which can be found on: https://github.com/jensloncke/ImageJ_macros/tree/master/Fluo-4%20transfected%20cells. Relevant response parameters, such as area under curve, were extracted using a custom Python script: https://github.com/jensloncke/LMCS-python-scripts/tree/main/Quantify%20agonist%20response. Cells that detached, or did not respond to either ATP concentration were excluded from analysis. Biochemical purification of HeLa MAM fractions Subcellular fractions were purified from CTRL and Cisd2-KO HeLa cells as previously described ( 24 ). Concisely, HeLa cells were homogenized and after several centrifugation steps, a crude mitochondrial fraction is obtained separately from the ER and cytosolic fractions. After ultracentrifugation of the crude mitochondrial fraction in a Percoll® (Santa Cruz Biotechnology Inc., Dallas, USA) gradient, MAM and pure mitochondrial fractions were obtained. Imaging of HeLa ER-mitochondrial contact sites using Mt-ER short SPLICS probe ER-mitochondrial contact sites were specifically detected using the SPLICS Mt-ER Short P2A probe. SPLICS Mt-ER Short P2A was a gift from Marisa Brini & Tito Calì (Addgene plasmid # 164108; http://n2t.net/addgene:164108; RRID:Addgene_164108) ( 36 ) on a fluorescence confocal microscope (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4). Excitation: 488 nm, emission: 530 nm. All cells were seeded in 4 chamber slides with coverslips. HeLa cells were transfected with 500 ng SPLICS Two days after transfection, cells were fixed with 4 % paraformaldehyde and imaged immediately. 3D Z-stacks were acquired SPLICS fluorescence and unfiltered light was captured in another channel to serve as a brightfield image replacement. Data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/splics%203D. Imaging of HeLa and cortical neuron ER-mitochondrial contact sites using MAMtracker Green HeLa cells were seeded 24 h before transfection in 4 chamber slides with coverslip. 48 h before acquisition, cells were transfected with 200 ng MAMtracker Green and 200 ng pCMV24-Cisd2-p2a-mCherry or pCMV24-p2a-mCherry. Two days after transfection, cells were stained with Hoechst and imaged in modified Krebs buffer. Cortical neurons were seeded 50 days before transfection. Cortical neurons were transfected with pCMV24-Cisd2-p2a-mCherry or pCMV24-p2a-mCherry. 2D images were simultaneously acquired of Hoechst stains, MAMtracker Green signal, mCherry signal and brightfield images. The p-MAMtracker Green vector was a kind gift from Koji Yamanaka (Nagoya University, Japan) ( 37 ). A Nikon TI2-E inverted microscope equipped with a 40 × 1.3 NA Plan Fluor DIC H N2 Oil objective and a pco.edge 4.2bi sCMOS camera was used. MAMtracker Green was excited at 470 nm, using a CoolLed pR-4000 lamp and using the cubical excitation filterset FF01–378/474/554/635 and bandpass emission filters of 515/30. mCherry was excited at 550 nm, using a CoolLed pR-4000 lamp, and using the cubical excitation filterset FF01–378/474/554/635 and a bandpass emission filter 595/31. Hoechst was excited at 350 nm using a CoolLed pR-4000 lamp, and using the cubical excitation filterset FF01-378/474/554/635. Data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/MAMtracker Assessment of mitochondrial morphology HeLa cells were seeded 48 h before acquisition on 4 chamber slides with coverslips. Cells were stained with 100 nM Mitotracker Green (M7514) and 1:1000 CelMask™ Orange (C10045). Fluorescence was acquired on a fluorescence confocal microscope (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4). Mitotracker Green: Excitation: 488 nm, emission: 530 nm. CellMask™ Orange: excitation: 543 nm emission: 633 nm. 3D Z-stacks were acquired and data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/PM%20and%20mito%203D and the Mitochondria analyzer plugin ( 49 ). Relative quantification of inner mitochondrial membrane potential using JC-1 HeLa cells were seeded 48 h before acquisition on 4 chamber slides with coverslips. Cells were stained with 2 µM JC-1 (Invitrogen, T3168), diluted in modified Krebs. After incubation, the cells were washed twice with the modified Krebs solution for live-cell imaging. Images were acquired using a (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4), an 488 nm Argon laser equipped with a BP 505-530 filter, and a 543 nm laser equipped with an LP 560 filter. Data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/Hoechst%20and%20JC-1 Acquisition of mitochondrial oxygen consumption rate The OCR and ECAR were measured using an XFp extracellular analyzer (Seahorse Bioscience, North Billerica, MA, USA). HeLa cells were plated on XFp Cell culture miniplates (Seahorse Bioscience, Cat. 103022-100) at 12 000 cells per well. Cortical neurons were plated 45 days before acquisition. The next day, the medium was changed to XF Base Medium (Seahorse Bioscience, 103334-100) containing 10 mM of glucose, and the cells were incubated at 37 °C in a low-CO 2 incubator for 1 h. The cells were sequentially exposed to oligomycin (1 µM), FCCP (0.5 µM), and a mixture of rotenone (0.5 µM) and antimycin A (0.5 µM). Data was analyzed using R Analysis of autophagic flux via western blotting HeLa cells CTRL and Cisd2 KO cells were seeded on 6 well plates and treated 48 h after seeding. CTRL and Cisd2 cortical neurons were seeded on Matrigel-coated 6 well plates 52-59 days before treatment. 2 h before treatment, medium was refreshed. Cells were treated with 100 nM bafilomycin A1 or DMSO. Cells were harvested by scraping in ice cold PBS lysed and immunoblotted. Densitometric quantification was performed using the gel analyzer plugin in ImageJ. Single-cell quantification of autophagic flux through the GFP(-RFP)-LC3 probe CTRL and Cisd2 KO HeLa cells were seeded on 4 chamber slides with coverslips. 24 h afterwards, cells were transfected with 200 ng GFP-LC3. Medium was changed 24 h post-transfection. Cortical neurons were seeded on Matrigel-coated 4 chamber slides with coverslips 52-59 days before treatment. Cortical neurons were transfected with 500 ng p4-neo-mRFP-GFP-LC3 probe. 48 h after transfection, medium was changed. 2 h after, cells were treated with 100 nM bafilomycin A1 or DMSO. 15 min before acquisition, cells were stained with Hoechst 33342. Before imaging, cells were washed once with modified Krebs. Fluorescence was acquired using the Nikon TI-2E setup described earlier. Autophagic puncti were quantified using the Cell Counter plugin in ImageJ. Staurosporine-induced apoptosis experiments HeLa CTRL or Cisd2 KO cells were seeded 48 h before transfection, cortical neurons were seeded 52-59 days before treatment and maintained in neuronal maintenance medium. Cells were treated with various staurosporine concentrations or with DMSO for 6 h. After treatment, cells were harvested by scraping on ice, and lysed with CHAPS buffer. Samples were analyzed via immunoblotting. Densitometric quantification was performed using the gel analyzer plugin in ImageJ. Mitochondrial Ca 2+ imaging in cortical neurons 52-59 days before acquisition, cortical neurons were seeded on Matrigel-coated 4 chamber slides with coverslips. 72 h before imaging, cortical neurons were transfected with 600 ng G-mtCEPIA2 ( 40 ). Before imaging, cortical neurons were washed once with modified Krebs. During measurements, G-mtCEPIA2 was recorded at a 2 s interval. 100 mM glutamate/10 mM glycine was added after 60 s of starting acquisitions. Fluorescence was acquired using the Nikon TI-2E setup described earlier. Data was processed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/Cortical%20neurons%20mtCEPIA . Relevant response parameters, such as area under curve, were extracted using a custom Python script: https://github.com/jensloncke/LMCS-python-scripts/tree/main/Quantify%20agonist%20response. Imaging of cortical neurons ER-mitochondrial contact sites using Mt-ER short SPLICS probe 52-59 days before acquisition, cortical neurons were seeded on Matrigel-coated 4 chamber slides with coverslips. 72 h before imaging, cortical neurons were transfected with 200 ng SPLICS. Cells were stained with Hoechst 33342 for 15 minutes. Before imaging, cortical neurons were washed once with modified Krebs. Fluorescence was acquired using the Nikon TI-2E setup described earlier. Data was quantified using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/splics%202D Single cell quantification of relative cellular ATP:ADP ratios shCTRL, shCisd2 and shCisd2 + Cisd2 were cultured, transfected and acquired as described in ( 30 ). Data analysis Quantification of immunoblots was performed using the FIJI software ( 50 ). MST data was analyzed as described in ( 24 ). All other plots were created using the ggplot2 package in the R programming language. Normality of residuals of fit was tested via Shapiro-Wilk normality testing and assessing equality of variances with the Levene test. Where appropriate, data was square-root or log10 transformed. In the event of both of the assumptions being fulfilled, a two-way ANOVA was performed, followed by fdr-corrected pairwise t-tests. In case of unequal variances within conditions, ANOVA was corrected for heteroskedasticity. When assumption of normal distribution was not fulfilled, a non-parametric Kruskal-Wallis was performed, followed by pairwise Mann-Whitney U tests. Microscopy data was acquired using the NIS elements software and numerically extracted in FIJI using custom macros. Data was further processed in Python using the numpy, pandas and plotly packages. Declarations Conflict of interest The authors declare no conflict of interest. Acknowledgements Research supported was by Eye Hope Foundation and Research Foundation—Flanders (FWO) to G.B. (G081821N), the KU Leuven Research Council (C14/19/099 and AKUL/19/34) and the Central European Leuven Strategic Alliance (CELSA/23/031 and CELSA/23/032). G.B., J.B.P., I.S. and A.K. are partners of the FWO Scientific Research Network CaSign (W0.014.22N). I.D.R. is supported by a PhD fellowship from the FWO (1131322N|1131324N). T.V was supported by a Post-doctoral fellowship from the FWO (12ZG121N). We thank Anja Florizoone for the excellent technical support. References Z.-Q. Shen, Y.-L. Huang, Y.-C. Teng, T.-W. Wang, C.-H. Kao, C.-H. Yeh, T.-F. Tsai, CISD2 maintains cellular homeostasis. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1868 , 118954 (2021). S. Amr, C. Heisey, M. Zhang, X.-J. Xia, K. H. Shows, K. Ajlouni, A. Pandya, L. S. Satin, H. El-Shanti, R. Shiang, A Homozygous Mutation in a Novel Zinc-Finger Protein, ERIS, Is Responsible for Wolfram Syndrome 2. The American Journal of Human Genetics 81 , 673-683 (2007). J. Loncke, T. Vervliet, J. B. Parys, A. Kaasik, G. Bultynck, Uniting the divergent Wolfram syndrome–linked proteins WFS1 and CISD2 as modulators of Ca2+ signaling. Science Signaling 14 , eabc6165 (2021). Y.-F. Chen, C.-H. Kao, Y.-T. Chen, C.-H. Wang, C.-Y. Wu, C.-Y. Tsai, F.-C. Liu, C.-W. Yang, Y.-H. Wei, M.-T. Hsu, S.-F. Tsai, T.-F. Tsai, Cisd2 deficiency drives premature aging and causes mitochondria-mediated defects in mice. Genes & Development 23 , 1183-1194 (2009). C.-H. Yeh, Z.-Q. Shen, S.-Y. Hsiung, P.-C. Wu, Y.-C. Teng, Y.-J. Chou, S.-W. Fang, C.-F. Chen, Y.-T. Yan, L.-S. Kao, C.-H. Kao, T.-F. Tsai, Cisd2 is essential to delaying cardiac aging and to maintaining heart functions. PLOS Biology 17 , e3000508 (2019). Y.-C. Teng, J.-Y. Wang, Y.-H. Chi, T.-F. Tsai, Exercise and the Cisd2 Prolongevity Gene: Two Promising Strategies to Delay the Aging of Skeletal Muscle. International Journal of Molecular Sciences . 2020. C.-H. Huang, Y.-L. Huang, Z.-Q. Shen, C.-H. Lin, T.-F. Tsai, Cisd2 Preserves the Youthful Pattern of the Liver Proteome during Natural Aging of Mice. Biomedicines . 2021. Y.-L. Huang, Z.-Q. Shen, C.-H. Huang, C.-H. Lin, T.-F. Tsai, Cisd2 slows down liver aging and attenuates age-related metabolic dysfunction in male mice. Aging Cell 20 , e13523 (2021). Y.-L. Huang, Z.-Q. Shen, C.-H. Huang, Y.-C. Teng, C.-H. Lin, T.-F. Tsai, Cisd2 Protects the Liver from Oxidative Stress and Ameliorates Western Diet-Induced Nonalcoholic Fatty Liver Disease. Antioxidants . 2021. Z.-Q. Shen, Y.-F. Chen, J.-R. Chen, Y.-S. Jou, P.-C. Wu, C.-H. Kao, C.-H. Wang, Y.-L. Huang, C.-F. Chen, T.-S. Huang, Y.-C. Shyu, S.-F. Tsai, L.-S. Kao, T.-F. Tsai, CISD2 Haploinsufficiency Disrupts Calcium Homeostasis, Causes Nonalcoholic Fatty Liver Disease, and Promotes Hepatocellular Carcinoma. Cell Reports 21 , 2198-2211 (2017). C.-C. Sun, S.-Y. Lee, C.-H. Kao, L.-H. Chen, Z.-Q. Shen, C.-H. Lai, T.-Y. Tzeng, J.-H. S. Pang, W.-T. Chiu, T.-F. Tsai, Cisd2 plays an essential role in corneal epithelial regeneration. EBioMedicine 73 , 103654 (2021). O. Karmi, Y.-S. Sohn, S. I. Zandalinas, L. Rowland, S. D. King, R. Nechushtai, R. Mittler, Disrupting CISD2 function in cancer cells primarily impacts mitochondrial labile iron levels and triggers TXNIP expression. Free Radical Biology and Medicine 176 , 92-104 (2021). S. H. Holt, M. Darash-Yahana, Y. S. Sohn, L. Song, O. Karmi, S. Tamir, D. Michaeli, Y. Luo, M. L. Paddock, P. A. Jennings, J. N. Onuchic, R. K. Azad, E. Pikarsky, I. Z. Cabantchik, R. Nechushtai, R. Mittler, Activation of apoptosis in NAF-1-deficient human epithelial breast cancer cells. Journal of Cell Science 129 , 155-165 (2016). Z.-Q. Shen, H. Yi-Long, T.-F. and Tsai, Cisd2 haploinsufficiency: A driving force for hepatocellular carcinoma. Molecular & Cellular Oncology 5 , e1441627 (2018). M. D. Bootman, G. Bultynck, Fundamentals of Cellular Calcium Signaling: A Primer. Cold Spring Harbor Perspectives in Biology 12 , a038802 (2020). B. Cartes-Saavedra, A. Ghosh, G. Hajnóczky, The roles of mitochondria in global and local intracellular calcium signalling. Nature Reviews Molecular Cell Biology , (2025). C. Angebault, J. Fauconnier, S. Patergnani, J. Rieusset, A. Danese, C. A. Affortit, J. Jagodzinska, C. Mégy, M. Quiles, C. Cazevieille, J. Korchagina, D. Bonnet-Wersinger, D. Milea, C. Hamel, P. Pinton, M. Thiry, A. Lacampagne, B. Delprat, C. Delettre, ER-mitochondria cross-talk is regulated by the Ca2+ sensor NCS1 and is impaired in Wolfram syndrome. Science Signaling 11 , eaaq1380 (2018). N. C. Chang, M. Nguyen, M. Germain, G. C. Shore, Antagonism of Beclin 1‐dependent autophagy by BCL‐2 at the endoplasmic reticulum requires NAF‐1. The EMBO Journal 29 , 606-618-618 (2010). L. D. Nguyen, E. T. Petri, L. K. Huynh, B. E. Ehrlich, Characterization of NCS1–InsP3R1 interaction and its functional significance. Journal of Biological Chemistry 294 , 18923-18933 (2019). G. Monaco, M. Beckers, H. Ivanova, L. Missiaen, J. B. Parys, H. De Smedt, G. Bultynck, Profiling of the Bcl-2/Bcl-XL-binding sites on type 1 IP3 receptor. Biochemical and Biophysical Research Communications 428 , 31-35 (2012). H. Ivanova, L. E. Wagner, A. Tanimura, E. Vandermarliere, T. Luyten, K. Welkenhuyzen, K. J. Alzayady, L. Wang, K. Hamada, K. Mikoshiba, H. De Smedt, L. Martens, D. I. Yule, J. B. Parys, G. Bultynck, Bcl-2 and IP3 compete for the ligand-binding domain of IP3Rs modulating Ca2+ signaling output. Cellular and Molecular Life Sciences 76 , 3843-3859 (2019). N. C. Chang, M. Nguyen, M. Germain, G. C. Shore, Antagonism of Beclin 1‐dependent autophagy by BCL‐2 at the endoplasmic reticulum requires NAF‐1. The EMBO Journal 29 , 606-618 (2010). S. Tamir, S. Rotem-Bamberger, C. Katz, F. Morcos, K. L. Hailey, J. A. Zuris, C. Wang, A. R. Conlan, C. H. Lipper, M. L. Paddock, R. Mittler, J. N. Onuchic, P. A. Jennings, A. Friedler, R. Nechushtai, Integrated strategy reveals the protein interface between cancer targets Bcl-2 and NAF-1. Proceedings of the National Academy of Sciences 111 , 5177-5182 (2014). J. Loncke, I. de Ridder, J. Kale, L. Wagner, A. Kaasik, J. B. Parys, M. Kerkhofs, D. W. Andrews, D. Yule, T. Vervliet, G. Bultynck, CISD2 counteracts the inhibition of ER-mitochondrial calcium transfer by anti-apoptotic BCL-2. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1872 , 119857 (2025). B. Delprat, T. Maurice, C. Delettre, Wolfram syndrome: MAMs’ connection? Cell Death & Disease 9 , 364 (2018). C. Kwak, S. Shin, J.-S. Park, M. Jung, T. T. M. Nhung, M.-G. Kang, C. Lee, T.-H. Kwon, S. K. Park, J. Y. Mun, J.-S. Kim, H.-W. Rhee, Contact-ID, a tool for profiling organelle contact sites, reveals regulatory proteins of mitochondrial-associated membrane formation. Proceedings of the National Academy of Sciences 117 , 12109-12120 (2020). S. E. Wiley, A. Y. Andreyev, A. S. Divakaruni, R. Karisch, G. Perkins, E. A. Wall, P. van der Geer, Y. F. Chen, T. F. Tsai, M. I. Simon, B. G. Neel, J. E. Dixon, A. N. Murphy, Wolfram Syndrome protein, Miner1, regulates sulphydryl redox status, the unfolded protein response, and Ca2+ homeostasis. EMBO Molecular Medicine 5 , 904-918-918 (2013). C.-H. Wang, Y.-F. Chen, C.-Y. Wu, P.-C. Wu, Y.-L. Huang, C.-H. Kao, C.-H. Lin, L.-S. Kao, T.-F. Tsai, Y.-H. Wei, Cisd2 modulates the differentiation and functioning of adipocytes by regulating intracellular Ca2+ homeostasis. Human Molecular Genetics 23 , 4770-4785 (2014). A. Zhang, C. D. Williamson, D. S. Wong, M. D. Bullough, K. J. Brown, Y. Hathout, A. M. Colberg-Poley, Quantitative Proteomic Analyses of Human Cytomegalovirus-Induced Restructuring of Endoplasmic Reticulum-Mitochondrial Contacts at Late Times of Infection*. Molecular & Cellular Proteomics 10 , M111.009936 (2011). M. Liiv, A. Vaarmann, D. Safiulina, V. Choubey, R. Gupta, M. Kuum, L. Janickova, Z. Hodurova, M. Cagalinec, A. Zeb, M. A. Hickey, Y.-L. Huang, N. Gogichaishvili, M. Mandel, M. Plaas, E. Vasar, J. Loncke, T. Vervliet, T.-F. Tsai, G. Bultynck, V. Veksler, A. Kaasik, ER calcium depletion as a key driver for impaired ER-to-mitochondria calcium transfer and mitochondrial dysfunction in Wolfram syndrome. Nature Communications 15 , 6143 (2024). L. Crouzier, A. Danese, Y. Yasui, E. M. Richard, J.-C. Liévens, S. Patergnani, S. Couly, C. Diez, M. Denus, N. Cubedo, M. Rossel, M. Thiry, T.-P. Su, P. Pinton, T. Maurice, B. Delprat, Activation of the sigma-1 receptor chaperone alleviates symptoms of Wolfram syndrome in preclinical models. Science Translational Medicine 14 , eabh3763 (2022). L. D. Nguyen, T. T. Fischer, D. Abreu, A. Arroyo, F. Urano, B. E. Ehrlich, Calpain inhibitor and ibudilast rescue β cell functions in a cellular model of Wolfram syndrome. Proceedings of the National Academy of Sciences 117 , 17389-17398 (2020). C. Rouzier, D. Moore, C. Delorme, S. Lacas-Gervais, S. Ait-El-Mkadem, K. Fragaki, F. Burté, V. Serre, S. Bannwarth, A. Chaussenot, M. Catala, P. Yu-Wai-Man, V. Paquis-Flucklinger, A novel CISD2 mutation associated with a classical Wolfram syndrome phenotype alters Ca2+ homeostasis and ER-mitochondria interactions. Human Molecular Genetics 26 , 1599-1611 (2017). M. Katona, Á. Bartók, Z. Nichtova, G. Csordás, E. Berezhnaya, D. Weaver, A. Ghosh, P. Várnai, D. I. Yule, G. Hajnóczky, Capture at the ER-mitochondrial contacts licenses IP3 receptors to stimulate local Ca2+ transfer and oxidative metabolism. Nature Communications 13 , 6779 (2022). K. Kanemaru, J. Suzuki, I. Taiko, M. Iino, Red fluorescent CEPIA indicators for visualization of Ca2+ dynamics in mitochondria. Scientific Reports 10 , 2835 (2020). F. Vallese, C. Catoni, D. Cieri, L. Barazzuol, O. Ramirez, V. Calore, M. Bonora, F. Giamogante, P. Pinton, M. Brini, T. Calì, An expanded palette of improved SPLICS reporters detects multiple organelle contacts in vitro and in vivo. Nature Communications 11 , 6069 (2020). S. Sakai, S. Watanabe, O. Komine, A. Sobue, K. Yamanaka, Novel reporters of mitochondria-associated membranes (MAM), MAMtrackers, demonstrate MAM disruption as a common pathological feature in amyotrophic lateral sclerosis. The FASEB Journal 35 , e21688 (2021). G. Mariño, M. Niso-Santano, E. H. Baehrecke, G. Kroemer, Self-consumption: the interplay of autophagy and apoptosis. Nature Reviews Molecular Cell Biology 15 , 81-94 (2014). P. A. Egorova, K. S. Marinina, I. B. Bezprozvanny, Chronic suppression of STIM1-mediated calcium signaling in Purkinje cells rescues the cerebellar pathology in spinocerebellar ataxia type 2. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1870 , 119466 (2023). J. Suzuki, K. Kanemaru, K. Ishii, M. Ohkura, Y. Okubo, M. Iino, Imaging intraorganellar Ca2+ at subcellular resolution using CEPIA. Nature Communications 5 , 4153 (2014). M. Tantama, J. R. Martínez-François, R. Mongeon, G. Yellen, Imaging energy status in live cells with a fluorescent biosensor of the intracellular ATP-to-ADP ratio. Nature Communications 4 , 2550 (2013). N. C. Chang, M. Nguyen, J. Bourdon, P.-A. Risse, J. Martin, G. Danialou, R. Rizzuto, B. J. Petrof, G. C. Shore, Bcl-2-associated autophagy regulator Naf-1 required for maintenance of skeletal muscle. Human Molecular Genetics 21 , 2277-2287 (2012). X. Du, R. Xiao, F. Xiao, Y. Chen, F. Hua, S. Yu, G. Xu, NAF-1 antagonizes starvation-induced autophagy through AMPK signaling pathway in cardiomyocytes. Cell Biology International 39 , 816-823 (2015). J. B. Parys, H. De Smedt, L. Missiaen, M. D. Bootman, I. Sienaert, R. Casteels, Rat basophiliv leukemia cells as model system for inositol 1,4,5-trisphosphate receptor IV, a receptor of the type II family: functional comparison and immunological detection. Cell Calcium 17 , 239-249 (1995). T. Vervloessem, B. K. Sasi, E. Xerxa, S. Karamanou, J. Kale, R. M. La Rovere, S. Chakraborty, F. Sneyers, M. Vogler, A. Economou, L. Laurenti, D. W. Andrews, D. G. Efremov, G. Bultynck, BDA-366, a putative Bcl-2 BH4 domain antagonist, induces apoptosis independently of Bcl-2 in a variety of cancer cell models. Cell Death & Disease 11 , 769 (2020). T. Vervliet, I. Lemmens, K. Welkenhuyzen, J. Tavernier, J. B. Parys, G. Bultynck, Regulation of the ryanodine receptor by anti-apoptotic Bcl-2 is independent of its BH3-domain-binding properties. Biochemical and Biophysical Research Communications 463 , 174-179 (2015). F. A. Ran, P. D. Hsu, J. Wright, V. Agarwala, D. A. Scott, F. Zhang, Genome engineering using the CRISPR-Cas9 system. Nature Protocols 8 , 2281-2308 (2013). J. L. T. Vervliet, M. Sever, K. Ahuja, C. Van den Haute, C. Verfaillie, T. Tomasic, G. Bultynck, Inactive ryanodine receptors sustain lysosomal availability for autophagy by promoting ER-lysosomal contact site formation. Nature Communications Under Review , (2024). A. Chaudhry, R. Shi, D. S. Luciani, A pipeline for multidimensional confocal analysis of mitochondrial morphology, function, and dynamics in pancreatic β-cells. American Journal of Physiology-Endocrinology and Metabolism 318 , E87-E101 (2019). J. Schindelin, I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, J.-Y. Tinevez, D. J. White, V. Hartenstein, K. Eliceiri, P. Tomancak, A. Cardona, Fiji: an open-source platform for biological-image analysis. Nature Methods 9 , 676-682 (2012). Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6298090","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":433404965,"identity":"3f8887f6-f78d-4ae7-b67a-7937567ff429","order_by":0,"name":"Jens Loncke","email":"","orcid":"https://orcid.org/0000-0002-8703-9009","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Jens","middleName":"","lastName":"Loncke","suffix":""},{"id":433405356,"identity":"d8c598f3-8418-4940-bbc6-3bfe30b1d354","order_by":1,"name":"Ian de Ridder","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Ian","middleName":"","lastName":"de Ridder","suffix":""},{"id":433405357,"identity":"413e5c79-d7b2-4242-ad76-a9a9e05642fe","order_by":2,"name":"Rita La Rovere","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Rita","middleName":"La","lastName":"Rovere","suffix":""},{"id":433405358,"identity":"280e19ac-912d-4b86-a7d1-2cc28b01bc25","order_by":3,"name":"Annika Vaarmann","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Annika","middleName":"","lastName":"Vaarmann","suffix":""},{"id":433405359,"identity":"fc34ead7-1e84-4054-b14e-c82f17e33724","order_by":4,"name":"Guizhen Fan","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Guizhen","middleName":"","lastName":"Fan","suffix":""},{"id":433405360,"identity":"952c249c-dcfb-405c-9f65-089023837625","order_by":5,"name":"Karan Ahuja","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Karan","middleName":"","lastName":"Ahuja","suffix":""},{"id":433405361,"identity":"4a03216f-e2af-4810-93b3-ee34d1ec1e21","order_by":6,"name":"Irina Serysheva","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Irina","middleName":"","lastName":"Serysheva","suffix":""},{"id":433405362,"identity":"58fe88f2-f0ff-4a28-be82-d91180ac262c","order_by":7,"name":"Catherine Verfaillie","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Catherine","middleName":"","lastName":"Verfaillie","suffix":""},{"id":433405363,"identity":"3a6702c3-913e-4047-a2e3-7f268f689860","order_by":8,"name":"Martijn Kerkhofs","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Martijn","middleName":"","lastName":"Kerkhofs","suffix":""},{"id":433405364,"identity":"9f8981bd-008a-45a7-8ff8-ef96b973c23b","order_by":9,"name":"Jan B. Parys","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"B.","lastName":"Parys","suffix":""},{"id":433405365,"identity":"f1030f42-2d0d-4c0d-bdc8-cd1c5d2ca264","order_by":10,"name":"Allen Kaasik","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Allen","middleName":"","lastName":"Kaasik","suffix":""},{"id":433405366,"identity":"99fc489b-d8fd-4e5c-a01b-a0bf1f558620","order_by":11,"name":"Tim Vervliet","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Tim","middleName":"","lastName":"Vervliet","suffix":""},{"id":433405367,"identity":"86f22868-5614-42be-a943-3b6f73bce03c","order_by":12,"name":"Geert Bultynck","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABE0lEQVRIie2RMUvDQBTHHwQuDoeuCUHzFS4cJBaKq1+jpZDJWTI4BApxOXRtJ7+F88GDczl0DTRgRchqpCCditcoUuFQ3Bzut9zx4Mf7/3kADsf/RcL+xztk4H/O2M+CBEL6X87A+6OCvyvx6VStOmjGVSzS5Vo/cIaeimjRQOaXViXRajKT0I4rQrNE1IuUIckjqlsYCGlXZmfcxMKtkkbQLYYMaRrOKwRWj+zKzctqV7k3ysFbON8Y5XFp7xJQb0eppQlGSfBabrfY6zOac9Cs5RXJz0OhJzxEwo87hXQg7MHiS3yGomgOrz28DdbqJLm6mz7Vows8ynx7fdaP+xPsfT8EtccyW8qvr29v63A4HI537r9ibVuP6KgAAAAASUVORK5CYII=","orcid":"","institution":"KU Leuven","correspondingAuthor":true,"prefix":"","firstName":"Geert","middleName":"","lastName":"Bultynck","suffix":""}],"badges":[],"createdAt":"2025-03-24 20:01:59","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6298090/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6298090/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":79231910,"identity":"be5c1dd6-b236-460e-9ee2-ff021b35314a","added_by":"auto","created_at":"2025-03-26 02:59:26","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1069647,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCisd2 is a direct interactor of the IP\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eR1 Ca\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e2+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e channel. A:\u003c/strong\u003e Representative immunoblot (N = 5) of co-immunoprecipitation experiment in HEK cells knock-out for all three isoforms of IP\u003csub\u003e3\u003c/sub\u003eR and overexpressing solely IP\u003csub\u003e3\u003c/sub\u003eR1. IP\u003csub\u003e3\u003c/sub\u003eR1 was immunoprecipitated using an anti IP\u003csub\u003e3\u003c/sub\u003eR1 antibody. IP: immunoprecipitation. \u003cstrong\u003eB:\u003c/strong\u003e Protein domain structure of Cisd2. Black line indicates the purified cytosolic fragment that was used in microscale thermophoresis (MST) experiments. TM: transmembrane domain; CDGSH: CDGSH domain. \u003cstrong\u003eC: \u003c/strong\u003eProtein domain structure of IP\u003csub\u003e3\u003c/sub\u003eR1. Limited trypsin digestion of full length IP\u003csub\u003e3\u003c/sub\u003eR1 yields six proteolytic fragments, which are indicated by numbers 1-6. These six fragments were purified for use in MST experiments. SD: suppressor domain; LBC: ligand binding core; CCD: central coupling domain; CD: channel domain; CT: C-terminal tail. \u003cstrong\u003eD:\u003c/strong\u003e Example trace of MST experiment with FITC-labeled 6xHis-Cisd2\u003csup\u003eCYT\u003c/sup\u003e without (black) and with (red) addition of full-length IP\u003csub\u003e3\u003c/sub\u003eR1. FITC fluorescence was normalized to mean fluorescence in the blue shaded area. Change in thermophoresis was quantified in the time range indicated by the red shaded area. FL-IP\u003csub\u003e3\u003c/sub\u003eR1: full-length IP\u003csub\u003e3\u003c/sub\u003eR1. \u003cstrong\u003eE: \u003c/strong\u003eBinding fit curve of N = 4 MST experiments displaying observed thermophoretic shift of FITC-labeled 6xHis-Cisd2\u003csup\u003eCYT\u003c/sup\u003e over full-length IP\u003csub\u003e3\u003c/sub\u003eR1 concentration in molar ([IP\u003csub\u003e3\u003c/sub\u003eR1] (M)). K\u003csub\u003eD\u003c/sub\u003e: dissociation-constant. Vertical bars indicate standard deviations. \u003cstrong\u003eF:\u003c/strong\u003e Example trace of MST experiment with FITC-labeled 6xHis-Cisd2\u003csup\u003eCYT\u003c/sup\u003e without (black) and with (red) addition of purified GST-tagged fragment 3 of IP\u003csub\u003e3\u003c/sub\u003eR1. FITC fluorescence was normalized to mean fluorescence in the blue shaded area. Change in thermophoresis was quantified in the time range indicated by the red shaded area. \u003cstrong\u003eG: \u003c/strong\u003eBinding fit curve of N = 4 MST experiments displaying observed thermophoretic shift of of FITC-labeled 6xHis-Cisd2\u003csup\u003eCYT\u003c/sup\u003e over all GST-tagged fragments in molar ([GST-FragX\u003csup\u003eIP3R1\u003c/sup\u003e] (M)).\u003c/p\u003e","description":"","filename":"Fig1page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/569707e4abd7b040d296a62b.jpg"},{"id":79231912,"identity":"0f8294d8-51aa-4102-9873-eefa4bed894f","added_by":"auto","created_at":"2025-03-26 02:59:26","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1992248,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCisd2 deficient HeLa cells display unaltered resting cytosolic, ER and total Ca\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e2+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e levels. A: \u003c/strong\u003eRepresentative immunoblot of control (CTRL) and Cisd2 KO HeLa cells. \u003cstrong\u003eB: \u003c/strong\u003eDensitometric quantification of Cisd2 levels relative to loading control of N = 5 immunoblots. a.u.: arbitrary units. \u003cstrong\u003eC: \u003c/strong\u003eImmunofluorescent staining of Cisd2 using a monoclonal mouse anti-Cisd2 antibody in control (CTRL) and Cisd2 KO HeLa cells. \u0026nbsp;\u003cstrong\u003eD:\u003c/strong\u003e Representative acquisition of FURA-2 loaded CTRL and Cisd2 HeLa cells with fluorescence at excitation wavelength 340 nm (Ex\u003csub\u003e340 nm\u003c/sub\u003e) in green, fluorescence at excitation wavelength 380 nm (Ex\u003csub\u003e380 nm\u003c/sub\u003e) in cyan and FURA-2 ratio (Ex\u003csub\u003e340 nm\u003c/sub\u003e / Ex\u003csub\u003e380 nm\u003c/sub\u003e). \u003cstrong\u003eE: \u003c/strong\u003eQuantification of basal FURA-2 ratio in CTRL and Cisd2 KO HeLa cells (N = 9). \u0026nbsp;\u003cstrong\u003eF: \u003c/strong\u003eAveraged normalized Cal520 fluorescence traces (F / F\u003csub\u003e0\u003c/sub\u003e) of thapsigargin responses of CTRL and Cisd2 KO HeLa cells. Shaded areas indicate 95 % confidence intervals (95 % CI). \u003cstrong\u003eG: \u003c/strong\u003eQuantification of area under curve (AUC) of thapsigargin responses shown in \u003cstrong\u003eF \u003c/strong\u003e(N = 9). \u003cstrong\u003eH: \u003c/strong\u003eAveraged normalized Cal520 fluorescence traces (F / F\u003csub\u003e0\u003c/sub\u003e) of ionomycin responses of CTRL and Cisd2 KO HeLa cells. Shaded areas indicate 95 % CI. \u003cstrong\u003eI: \u003c/strong\u003eQuantification of AUC of ionomycin responses shown in \u003cstrong\u003eH \u003c/strong\u003e(N = 9).\u003c/p\u003e","description":"","filename":"Fig2page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/0314992078a81ab551573459.jpg"},{"id":79231911,"identity":"e0ca3daf-00bb-4b87-b9a7-3306ad9560c5","added_by":"auto","created_at":"2025-03-26 02:59:26","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1799425,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCisd2 ensures IP\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eR-mediated ER-mitochondrial Ca\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e2+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e transfer. A: \u003c/strong\u003eAveraged traces of normalized fluo-4 fluorescence (F / F\u003csub\u003e0\u003c/sub\u003e) of control (CTRL) and Cisd2 KO HeLa cells. Shaded areas indicate 95 % confidence intervals (95 % CI). \u003cstrong\u003eB: \u003c/strong\u003eQuantification of area under curve (AUC) of cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e responses to 5 µM ATP and residual responses to 100 µM ATP shown in \u003cstrong\u003eA \u003c/strong\u003e(N = at least 15, n = at least 249 cells). \u003cstrong\u003eC:\u003c/strong\u003e Averaged traces of normalized R-mtCEPIA3 fluorescence of CTRL and Cisd2 KO HeLa cells. Shaded areas indicate 95 % CI. \u003cstrong\u003eD:\u003c/strong\u003e Quantification of logarithmically transformed AUC of mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e responses to 5 µM ATP and residual responses to 100 µM ATP shown in \u003cstrong\u003eC \u003c/strong\u003e(N = at least 15, n = at least 249 cells). \u003cstrong\u003eE: \u003c/strong\u003eImmunoblot displaying overexpression of triple-FLAG tagged Cisd2 (FLAG\u003csub\u003e3\u003c/sub\u003e-Cisd2). \u003cstrong\u003eF: \u003c/strong\u003eAveraged traces of normalized fluo-4 fluorescence of Cisd2 KO HeLa cells transfected with empty vector control (EV) or pCMV26-FLAG\u003csub\u003e3\u003c/sub\u003e-Cisd2. Shaded areas indicate 95 % CI. \u003cstrong\u003eG: \u003c/strong\u003eAveraged traces of normalized R-mtCEPIA3 fluorescence of Cisd2 KO HeLa cells transfected with empty vector control (EV) or pCMV26-FLAG\u003csub\u003e3\u003c/sub\u003e-Cisd2. Shaded areas indicate 95 % CI. \u003cstrong\u003eH: \u003c/strong\u003eQuantification of AUC of cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e responses to 5 µM ATP shown in \u003cstrong\u003eF \u003c/strong\u003e(N = at least 14, n = at least 368 cells). \u003cstrong\u003eI:\u003c/strong\u003e Quantification of logarithmically transformed AUC of mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e responses to 5 µM ATP shown in \u003cstrong\u003eG\u003c/strong\u003e. (N = at least 14, n = at least 368 cells).\u003c/p\u003e","description":"","filename":"Fig3page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/79cfc3c09122695b441fbcc4.jpg"},{"id":79232130,"identity":"e25d2e52-a838-4177-aa14-e3ebb4280bd6","added_by":"auto","created_at":"2025-03-26 03:07:26","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1193150,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCisd2 is enriched in MAMs but does not influence the levels of common MAM-associated proteins. A: \u003c/strong\u003eImmunoblot of subcellular fractionation experiment of control (CTRL) and CISD2 KO HeLa cells. Tot: total protein lysate; ER: endoplasmic reticulum; MAM: mitochondria-associated ER membranes; cM: crude mitochondria; pM: pure mitochondria. \u003cstrong\u003eB: \u003c/strong\u003eImmunoblot of subcellular fractionation experiment of CTRL and IP\u003csub\u003e3\u003c/sub\u003eR knock-out for all three isoforms (IP\u003csub\u003e3\u003c/sub\u003eR TKO) HeLa cells. Triangles indicate appropriate height of respective proteins in multi-band areas. \u003cstrong\u003eC:\u003c/strong\u003e Immunoblot of CTRL and CISD2 KO cells stained for common MAM-associated proteins. \u003cstrong\u003eD: \u003c/strong\u003eDensitometric quantification of IP\u003csub\u003e3\u003c/sub\u003eR levels relative to loading control (N = 4), normalized to CTRL value. a.u.: arbitrary units. \u003cstrong\u003eE: \u003c/strong\u003eDensitometric quantification of Mfn2 levels relative to loading control (N = 4), normalized to CTRL value. \u003cstrong\u003eF: \u003c/strong\u003eDensitometric quantification of Vdac1 levels relative to loading control (N = 4), normalized to CTRL value, \u003cstrong\u003eG: \u003c/strong\u003eDensitometric quantification of VapB levels relative to loading control (N = 4), normalized to CTRL value, \u003cstrong\u003eH: \u003c/strong\u003eDensitometric quantification of Grp75 levels relative to loading control (N = 3), normalized to CTRL value.\u003c/p\u003e","description":"","filename":"Fig4compressedpage0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/617fec551254e2049a2483c0.jpg"},{"id":79231917,"identity":"bf40a414-1794-4be7-83d2-3f5ec252b28e","added_by":"auto","created_at":"2025-03-26 02:59:26","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1192769,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCisd2 maintains ER-mitochondrial contact site integrity. A: \u003c/strong\u003eRepresentative acquisitions of unfiltered emission light, short SPLICS (SPLICS\u003csub\u003eS\u003c/sub\u003e) signal and volume render of Z-stacks of control (CTRL) and Cisd2 HeLa cells. \u003cstrong\u003eB: \u003c/strong\u003eQuantification of dinstinct ER-mitochondrial contact site count per cell of N = 12 SPLICS\u003csub\u003eS\u003c/sub\u003e experiments. \u003cstrong\u003eC: \u003c/strong\u003eQuantification of ER-mitochondrial contact site volume per cell of N = 12 SPLICS\u003csub\u003eS\u003c/sub\u003e experiments. \u003cstrong\u003eD: \u003c/strong\u003eRepresentative acquisitions of Hoechst, MAMtracker Green and mCherry fluorescence of CTRL HeLa cells, expressing p2a-mCherry and Cisd2 KO HeLa cells expressing either p2a-mCherry or Cisd2-p2a-mCherry. \u003cstrong\u003eE: \u003c/strong\u003eQuantification of mean MAMtracker Green fluorescence per cell, normalized to CTRL per experimental day of N = 8 experiments.\u003c/p\u003e","description":"","filename":"Fig5page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/65236b7dc8beb4e55722fda3.jpg"},{"id":79231921,"identity":"4fe1adb6-fcfb-40b2-83fc-219d810267de","added_by":"auto","created_at":"2025-03-26 02:59:26","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1960685,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMitochondrial health is unaffected by loss of Cisd2 in HeLa cells. A: \u003c/strong\u003eRepresentative acquisitions of control (CTRL) and Cisd2 KO HeLa cells, stained with Mitotracker Green and CellMask Orange.\u003cstrong\u003e B: \u003c/strong\u003eQuantification of fractional mitochondrial volume over total cell volume. Violin plot displays the distribution of individual cells, datapoints indicate acquisition averages whereby different point shape corresponds to different repeats. Black crossbar corresponds to mean. n = at least 32 cells, N = 4. \u003cstrong\u003eC: \u003c/strong\u003eQuantification of surface to volume ratios of individual mitochondria in CTRL and Cisd2 KO HeLa cells.\u003cstrong\u003e \u003c/strong\u003en = at least 5651 mitochondria, N = 4. \u003cstrong\u003eD: \u003c/strong\u003eQuantification of branches per mitochondria, normalized over mitochondrial volume in CTRL and Cisd2 KO HeLa cells. \u003cstrong\u003eE: \u003c/strong\u003eRepresentative acquisitions of CTRL and Cisd2 KO HeLa cells, stained with JC-1. \u003cstrong\u003eF:\u003c/strong\u003eQuantification of JC-1 ratio of CTRL and Cisd2 KO HeLa cells. Errorbars indicate 95 % confidence intervals (95 % CI). n = at least 87 cells, N = at least 13. \u003cstrong\u003eG: \u003c/strong\u003eAveraged oxygen consumption rate (OCR), normalized over protein concentration plotted over time. Timepoints of compound additions are indicated with black arrows. Shaded areas indicate 95 % CI. \u003cstrong\u003eH: \u003c/strong\u003eQuantification of normalized basal OCR of N = 4 Mito Stress Test Seahorse experiments shown in \u003cstrong\u003eG\u003c/strong\u003e. Errorbars correspond to 95 % CI.\u003c/p\u003e","description":"","filename":"Fig6page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/139530b3679328866cef2bcf.jpg"},{"id":79231922,"identity":"990272a3-77aa-4319-bcac-f20614c4234a","added_by":"auto","created_at":"2025-03-26 02:59:26","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1275979,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAbsence of Cisd2 increases autophagic flux in HeLa cells. A:\u003c/strong\u003e Representative immunoblot of HeLa control (CTRL) and Cisd2 KO cells treated with 100 nM Bafilomycin a1 or DMSO vehicle control. Migration heights of both LC3 isoforms are indicated with a triangle. \u003cstrong\u003eB: \u003c/strong\u003eDosimetric quantification of LC3-II levels (square root transformed) over loading control, normalized to CTRL vehicle levels (N = 3 experiments). sqrt: square root; Baf A1: bafilomycin a1. \u003cstrong\u003eC: \u003c/strong\u003eRepresentative fluorescence acquisitions of CTRL and Cisd2 KO HeLa cells expression GFP-LC3 treated with DMSO or Baf a1. \u003cstrong\u003eD: \u003c/strong\u003eQuantification of GFP-LC3 punctae per cell (n = at least 520 cells, N = 27). \u003cstrong\u003eE: \u003c/strong\u003eRepresentative immunoblot of staurosporine-induced apoptosis experiment in HeLa CTRL and Cisd2 KO cells. UT: untreated; VEH: vehicle. \u003cstrong\u003eF: \u003c/strong\u003eQuantification of PARP ratio of N = 3 apoptosis experiments shown in \u003cstrong\u003eE\u003c/strong\u003e.\u003c/p\u003e","description":"","filename":"Fig7page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/c43674394b6bde5fab846bba.jpg"},{"id":79231924,"identity":"d2f845cc-7db6-46c2-ac6f-358e25d4b41b","added_by":"auto","created_at":"2025-03-26 02:59:27","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":847418,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCisd2-deficient cortical neurons differentiated from iPSCs display diminished IP\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eR-mediated Ca\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e2+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e release evoked by glutamate. A: \u003c/strong\u003e\u0026nbsp;Immunoblot of control and Cisd2 KO iPSCs. \u003cstrong\u003eB:\u003c/strong\u003e Averaged traces of CTRL and Cisd2 KO iPSCs responding to 30 µM ATP. Shaded areas indicate 95 % confidence intervals (CI). \u003cstrong\u003eC: \u003c/strong\u003eDose-response fit of CTRL and Cisd2 KO cells responding to ATP. \u003cstrong\u003eD: \u003c/strong\u003eTimeline of differentiation of CTRL and Cisd2 KO cortical neurons from iPSCs. FGF-2: fibroblast growth factor 2. \u003cstrong\u003eE: \u003c/strong\u003eRepresentative fluorescence acquisition of Cisd2 KO cortical neurons immunofluorescently stained for SATB2 and MAP2. \u003cstrong\u003eF: \u003c/strong\u003eAveraged trace of CTRL cortical neurons responding to 6 mM glutamate (+ 0.6 mM glycine) in absence of extracellular Ca\u003csup\u003e2+\u003c/sup\u003e after 30 minutes pretreatment with 30 µM ryanodine. \u003cstrong\u003eG: \u003c/strong\u003eAveraged traces of CTRL and Cisd2 KO cortical neurons responding to 10 mM glutamate (+ 1 mM glycine) (dashed lines) or 100 mM glutamate (+ 10 mM glycine) solid lines. Shaded areas correspond to 95 % CI. \u003cstrong\u003eH\u003c/strong\u003e: Quantification of response amplitudes of CTRL and Cisd2 KO cortical neurons responding to glutamate (N = 3). \u003cstrong\u003eI: \u003c/strong\u003eQuantification of area under curve of thapsigargin-releasable Ca\u003csup\u003e2+\u003c/sup\u003e in CTRL and Cisd2 KO cortical neurons. AUC: area under curve; TG: thapsigargin.\u003c/p\u003e","description":"","filename":"Fig8page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/5e63e108618b6adf15e99114.jpg"},{"id":79232131,"identity":"9eb5d78c-62fb-47a7-ae03-13aadc6265f5","added_by":"auto","created_at":"2025-03-26 03:07:26","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":2350942,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCisd2 is essential for ER-mitochondrial contact integrity and mitochondrial homeostasis in cortical neurons. A: \u003c/strong\u003eAveraged traces of G-mtCEPIA2 fluorescence over time in control (CTRL) and Cisd2 KO cortical neurons responding to 100 mM glutamate (+ 10 mM glycine). Shaded areas correspond to 95 % confidence intervals (CI). \u003cstrong\u003eB: \u003c/strong\u003eQuantification of area under curve of mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e responses shown in \u003cstrong\u003eA\u003c/strong\u003e. n = at least 94 neurons. \u003cstrong\u003eC: \u003c/strong\u003eRepresentative acquisitions of cortical neurons transfected with SPLICS\u003csub\u003es\u003c/sub\u003e. \u003cstrong\u003eD: \u003c/strong\u003eQuantification of detectable ER-mitochondrial contact sites per CTRL or Cisd2 KO neuron (N = at least 37 neurons). \u003cstrong\u003eE: \u003c/strong\u003eRepresentative fluorescence acquisitions of CTRL and Cisd2 KOcortical neurons transfected with MAMtracker Green and p2a-mCherry / Cisd2-p2a-mCherry. \u003cstrong\u003eF: \u003c/strong\u003eQuantification of MAMtracker green fluorescence intensity (normalized to CTRL p2a-mCherry). a.u.: arbitrary units. \u003cstrong\u003eG: \u003c/strong\u003eAveraged traces of MitoStress Test Seahorse X\u003csub\u003eF \u003c/sub\u003eassays displaying oxygen consumption rate (normalized over protein concentration) over time. Shaded areas correspond to 95 % CI. \u003cstrong\u003eH: \u003c/strong\u003eQuantification of basal respiration of N = 6 experiments. \u003cstrong\u003eI: \u003c/strong\u003eQuantification of ATP-linked respiration of N = 6 experiments. \u003cstrong\u003eJ: \u003c/strong\u003eQuantification of maximal respiratory capacity of N = 6 experiments. \u003cstrong\u003eK: \u003c/strong\u003eRepresentative fluorescence acquisitions of CTRL and Cisd2 KO cortical neurons stained with JC-1. \u003cstrong\u003eL: \u003c/strong\u003eQuantification of JC-1 ratio of CTRL and Cisd2 KO cortical neurons normalized to CTRL. (N = 12). \u003cstrong\u003eM: \u003c/strong\u003eBasal ATP / ADP ratio recorded in axonal endings of CTRL and Cisd2 knockdown neonatal cortical rat neurons expressing PercevalHR.\u003c/p\u003e","description":"","filename":"Fig9compressedpage0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/55913c6322c580b5e0e212dd.jpg"},{"id":79231919,"identity":"443272ff-d838-4299-be12-6158dd46930f","added_by":"auto","created_at":"2025-03-26 02:59:26","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":1113043,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAbsence of Cisd2 increases autophagic flux and sensitizes cortical neurons to apoptotic stimuli. A: \u003c/strong\u003eRepresentative fluorescence acquisitions of control (CTRL) and Cisd2 KO cells expressing LC3-GFP-RFP. \u003cstrong\u003eB: \u003c/strong\u003eQuantification of GFP / RFP punctae per cell of CTRL and Cisd2 KO cortical neurons \u003cstrong\u003eC: \u003c/strong\u003eRepresentative immunoblot of staurosporine-induced apoptosis experiment in CTRL and Cisd2 KO cortical neurons.\u003cstrong\u003e D: \u003c/strong\u003eQuantification of PARP ratio of N = 3 apoptosis experiments shown in \u003cstrong\u003eC\u003c/strong\u003e.\u003c/p\u003e","description":"","filename":"Fig10page0001.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/5c806d9bdc85a9f3320b76b6.jpg"},{"id":79232919,"identity":"90108d53-57f8-492e-b585-3c137ce321b6","added_by":"auto","created_at":"2025-03-26 03:15:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":17054154,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6298090/v1/fcf811c3-5211-4eef-9cb5-40dece4d4e13.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eCisd2 ensures adequate ER-mitochondrial coupling, thereby critically supporting mitochondrial function in neurons\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCDGSH iron sulfur domain 2 (Cisd2) is an Fe-S-cluster domain-binding protein, important for cellular homeostasis (\u003cem\u003e1, 2\u003c/em\u003e). Cisd2 together with \u003cem\u003eWolframin\u003c/em\u003e (Wfs1) are the two genes that upon acquiring loss-of-function mutations result in Wolfram syndrome (WS). This rare genetic disease starts during childhood with diabetes mellitus due to loss of functional pancreatic β cells , progressing during adolescence towards neurodegeneration in central and peripheral nervous system (\u003cem\u003e3\u003c/em\u003e). Physiologically, Cisd2-protein levels have been implicated in longevity and the adequate function of other physiological systems besides β cells and neurons, such as skeletal and cardiac muscles (\u003cem\u003e4-6\u003c/em\u003e), liver (\u003cem\u003e7-10\u003c/em\u003e), corneal epithelia (\u003cem\u003e11\u003c/em\u003e), since a reduction of Cisd2 expression (e.g. in heterozygous Cisd2 knockout mice) evokes early aging and organ dysfunctions (\u003cem\u003e4\u003c/em\u003e). At the pathophysiological level, Cisd2 is upregulated in breast cancer, whereby inhibition of Cisd2 results in breast cancer cell death (\u003cem\u003e12, 13\u003c/em\u003e). In hepatocellular carcinoma, however, haplosufficiency of Cisd2 results in hepatocarcinogenesis (\u003cem\u003e10, 14\u003c/em\u003e), underpinning the complexity of Cisd2 in cancer cell biology.\u003c/p\u003e\n\u003cp\u003eFascinatingly, although WFS1 and Cisd2 belong to two completely different gene families resulting in two proteins with unrelated protein domains and cellular function, loss-of-function of either WFS1 or CISD2 results in Wolfram syndrome (\u003cem\u003e3\u003c/em\u003e). Therefore, we have postulated that Wfs1 and Cisd2 could also have converging roles, such as in Ca\u003csup\u003e2+\u003c/sup\u003e signaling(\u003cem\u003e3\u003c/em\u003e). In virtually all cells, ER-originating Ca\u003csup\u003e2+\u003c/sup\u003e signals are mediated by the opening of and Ca\u003csup\u003e2+\u003c/sup\u003e-flux through inositol (1, 4, 5)-trisphosphate (IP\u003csub\u003e3\u003c/sub\u003e) receptors (IP\u003csub\u003e3\u003c/sub\u003eRs), intracellular Ca\u003csup\u003e2+\u003c/sup\u003e-release channels (\u003cem\u003e15\u003c/em\u003e). These Ca\u003csub\u003e2+\u003c/sub\u003e signals do not only occur in the cytosol but are also transmitted to the mitochondrial matrix via mitochondria ER contact sites (MERCS) through interorganellar IP\u003csub\u003e3\u003c/sub\u003eR-GRP75-VDAC1-protein complexes (\u003cem\u003e16\u003c/em\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTwo important clues hinted towards the involvement of Cisd2 in intracellular Ca\u003csup\u003e2+\u003c/sup\u003e signaling arising from the ER. First, both Wfs1 and Cisd2 have been reported to form protein complexes with IP\u003csub\u003e3\u003c/sub\u003eRs (\u003cem\u003e17, 18\u003c/em\u003e). Wfs1 was found that to control IP\u003csub\u003e3\u003c/sub\u003eR function by operating in a triparental complex involving neuronal calcium sensor-1 protein (NCS-1) (\u003cem\u003e17\u003c/em\u003e). Instead, Cisd2 (alias: Naf-1) appears to form a complex with IP\u003csub\u003e3\u003c/sub\u003eRs involving anti-apoptotic Bcl-2 . Fascinatingly, both NCS-1 as well as Bcl-2 are well-established direct interactors of IP\u003csub\u003e3\u003c/sub\u003eRs (\u003cem\u003e19-21\u003c/em\u003e). Furthermore, Bcl-2 and Cisd2 also directly interact with each other (\u003cem\u003e22-24\u003c/em\u003e). More recently, it was found that Cisd2 alleviated Bcl-2’s ability to reduce the number and extent of MERCS (\u003cem\u003e24\u003c/em\u003e). Second, both Wfs1 and Cisd2 have been found to reside at MERCS (\u003cem\u003e24-29\u003c/em\u003e). Biochemical isolation of the ER membrane fractions involved in MERCS, so-called mitochondrial ER-associated membranes (MAMs), are enriched in both Wfs1 and Cisd2 (\u003cem\u003e3, 24, 25\u003c/em\u003e). Interestingly, in a proteome-wide, proximity biotinylation-based approach (ContactID), Cisd2 was among the top hits of proteins resident at MERCS (\u003cem\u003e26\u003c/em\u003e). Loss of Wfs1 resulted in a reduced MERCS integrity and impaired ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfers, thereby impairing mitochondrial functions such mitochondrial bio-energetics, particularly impacting neurons (\u003cem\u003e30\u003c/em\u003e). MERCS disassembly and dysfunction is also found in cellular and animal models carrying pathogenic Wfs1 mutations identified in Wolfram syndrome patients (\u003cem\u003e17, 31\u003c/em\u003e). Moreover, strategies that restored MERCS organization and Ca\u003csup\u003e2+\u003c/sup\u003e-signaling function such as NCS-1 overexpression (\u003cem\u003e17, 32\u003c/em\u003e) or pharmacological agonists of Sigma-1 receptor, a MERCS-resident chaperone , also improved mitochondrial function (\u003cem\u003e31\u003c/em\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile our understanding of MERCS control by Wfs1 has been majorly advanced the last 5 years, the functional impact of loss of Cisd2 on MERCS remains poorly understood. There has been one report linking patient-linked Cisd2 mutations to an increase of MERCS in patient-derived fibroblasts that also displayed an increased in agonist-induced ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfers, compared to control fibroblasts obtained from unaffected family members (\u003cem\u003e33\u003c/em\u003e). Nevertheless, the impact of wild-type Cisd2 on MERCS, particularly in disease-relevant cells remains elusive, thus exploring its function in neurons is critical. Furthermore, pathogenic mutations of Cisd2 may also introduce gain-of-function aspects and thus may not reflect the biological functions of wild-type Cisd2. Finally, MERCS may differ across individuals irrespective of the Cisd2 gene status, thus it’s instrumental to study in Cisd2 using isogenic conditions.\u003c/p\u003e\n\u003cp\u003eTo address this knowledge gap, we set out to study MERCS organization and Ca\u003csup\u003e2+\u003c/sup\u003e-signaling function in cell models in which Cisd2 was knocked out using CRISPR/Cas9. To explore its cell biological function in controlling MERCS, we used HeLa cells, anexcellent cell models to study Ca\u003csup\u003e2+\u0026nbsp;\u003c/sup\u003esignaling and particularly MERCS and its function in ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer due to the efficient coupling of ER and mitochondria in these cells (\u003cem\u003e34\u003c/em\u003e). In addition to this, to explore the cell physiological impact of loss of Cisd2, we developed Cisd2-KO iPSCs that were differentiated into cortical neurons. In both cell systems, we found that loss of Cisd2 resulted in loss of MERCS and impaired ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer, revealing an important role of Cisd2 in MERCS integrity irrespective of the cellular context. Yet, while HeLa cells were rather resilient towards loss of Cisd2 for their survival, mitochondrial metabolism and cell death susceptibility, iPSC-derived cortical neurons lacking Cisd2 were particularly susceptible to cell stress inducers such as staurosporine. Furthermore, neurons lacking Cisd2 displayed impaired mitochondrial health, including reduced oxphos and ATP-production capacity. Of note, while loss of Cisd2 had miminal impact on agonist-evoked cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e signals in HeLa cells, it severely impaired such cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e signals in cortical neurons. This unique feature may explain why neurons are highly susceptible to lowered Cisd2-protein levels, thereby accounting for Wolfram syndrome symptoms related to Cisd2 dysfunction.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eCisd2 Directly Interacts with IP\u003csub\u003e3\u003c/sub\u003eR1\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the emerging link between CISD2 and intracellular Ca\u003csup\u003e2+\u003c/sup\u003e signaling (\u003cem\u003e1, 3, 10, 24, 28, 30, 33\u003c/em\u003e) and evidence of Cisd2 being engaged in a macrocomplex with IP\u003csub\u003e3\u003c/sub\u003eR1 (\u003cem\u003e22\u003c/em\u003e), we examined whether Cisd2 could directly interact with IP\u003csub\u003e3\u003c/sub\u003eRs, the main intracellular Ca\u003csup\u003e2+\u003c/sup\u003e-release channels. First, by performing immunoprecipitation (IP) with an anti-IP\u003csub\u003e3\u003c/sub\u003eR1 antibody (Rbt03) and using lysates from HEK293 cells triple KO for all three isoform of IP\u003csub\u003e3\u003c/sub\u003eR(TKO) cells and re-expressing IP\u003csub\u003e3\u003c/sub\u003eR1, we found that endogenous Cisd2 co-IP’ed with IP\u003csub\u003e3\u003c/sub\u003eR1, demonstrating that Cisd2 exists in a complex with IP\u003csub\u003e3\u003c/sub\u003eR1 (\u003cstrong\u003eFigure 1A\u003c/strong\u003e). To assess whether Cisd2 can directly interact with IP\u003csub\u003e3\u003c/sub\u003eR1, we purified the cytosolic region of the Cisd2 protein (Cisd2\u003csup\u003eCYT\u003c/sup\u003e; \u003cstrong\u003e\u0026nbsp;Figure 1B\u0026nbsp;\u003c/strong\u003efor schematic representation) and assessed its binding to purified fully functional IP\u003csub\u003e3\u003c/sub\u003eR1 channel reconstituted into lipid nanodiscs and purified fragments of the IP\u003csub\u003e3\u003c/sub\u003eR1 (\u003cstrong\u003eFigure 1C\u0026nbsp;\u003c/strong\u003efor schematic representation). To acquire quantitative insights in the IP\u003csub\u003e3\u003c/sub\u003eR1/CIS2\u003csup\u003eCYT\u003c/sup\u003e complex, we performed microscale thermophoresis (MST) experiments using fluorescently labeled Cisd2\u003csup\u003eCYT\u003c/sup\u003e in the presence of increasing concentrations of full-channel IP\u003csub\u003e3\u003c/sub\u003eR1 (10 nM- 5\u0026nbsp;mM). Representative MST traces are shown in \u003cstrong\u003eFigure 1D\u0026nbsp;\u003c/strong\u003efor\u003cstrong\u003e\u0026nbsp;Cisd2\u003csup\u003eCYT\u003c/sup\u003e\u0026nbsp;\u003c/strong\u003ein absence and presence of full-channel IP\u003csub\u003e3\u003c/sub\u003eR1 (5\u0026nbsp;mM). The presence of full-channel IP\u003csub\u003e3\u003c/sub\u003eR1 resulted in a thermophoretic shift of Cisd2\u003csup\u003eCYT\u003c/sup\u003e, indicating both proteins can directly interact with each other in solution. Fluorescence was normalized to the average baseline values obtained in the blue shaded area, and differences in normalized fluorescence were determined over the range indicated as a red shaded area. \u003cstrong\u003eFigure 1E\u003c/strong\u003e shows the averaged shift in thermophoretic shift of Cisd2 in function of [full-channel IP\u003csub\u003e3\u003c/sub\u003eR1] over 5-6s after excitation. Due to the nature of the purified full-channel IP\u003csub\u003e3\u003c/sub\u003eR1 samples (low [full-channel IP\u003csub\u003e3\u003c/sub\u003eR1], nanodisc reconstitution, great size of the full-channel IP\u003csub\u003e3\u003c/sub\u003eR1), the quality of fitting in the binding curve was imperfect. From this fit, we could estimate and determine a binding affinity (K\u003csub\u003eD\u003c/sub\u003e) value for its interaction with Cisd2 of about 1.8 µM. To substantiate these findings and to map the Cisd2-interacting IP\u003csub\u003e3\u003c/sub\u003eR region, \u0026nbsp;we performed MST experiments with Cisd2\u003csup\u003eCYT\u0026nbsp;\u003c/sup\u003eand several purified fragments of IP\u003csub\u003e3\u003c/sub\u003eR1. \u003cstrong\u003eFigure 1F\u003c/strong\u003e shows representative MST traces of fluorescently labeled Cisd2\u003csup\u003eCYT\u0026nbsp;\u003c/sup\u003ewith and without addition of GST-Fragment 3 of IP\u003csub\u003e3\u003c/sub\u003eR1 (20 µM), showing a shift in thermophoretic mobility of CISD2\u003csup\u003eCYT\u003c/sup\u003e in the presence of GST-Fragment 3 of IP\u003csub\u003e3\u003c/sub\u003eR1. \u003cstrong\u003eFigure 1G\u0026nbsp;\u003c/strong\u003eshows average shifts in thermophoretic shift of Cisd2\u003csup\u003eCYT\u003c/sup\u003e in function of fragment concentrations curves of all screened GST-Fragments of IP\u003csub\u003e3\u003c/sub\u003eR1. By fitting binding curves, we could determine the K\u003csub\u003eD\u003c/sub\u003es of Cisd2\u003csup\u003eCYT\u003c/sup\u003e/GST-Fragment interactions. We observe the highest affinity for the interaction between Cisd2\u003csup\u003eCYT\u003c/sup\u003e and GST-Fragment 3 (K\u003csub\u003eD\u003c/sub\u003e of ~800 nM), marking aa 923-1581 of IP\u003csub\u003e3\u003c/sub\u003eR1 to harbor a likely binding site for Cisd2\u003csup\u003eCYT\u003c/sup\u003e. We also found a lower affinity interaction between Cisd2 and the ligand binding domain (LBD, aa 1-604) and the ligand binding core (LBC, aa 226-604), both with a K\u003csub\u003eD\u003c/sub\u003e around 2 µM. Cisd2\u003csup\u003eCYT\u003c/sup\u003e to SD, fragment 5, fragment 6 of IP3R1 displayed very high K\u003csub\u003eD\u003c/sub\u003e values of 7.3 µM, 16 µM, and 6.6 µM, thus likely too low affinity for \u0026nbsp;meaningful contribution, while for Fragment 4, no adequate binding curve could fitted. Overall, these interaction studies indicate that in cells Cisd2 forms a complex with IP\u003csub\u003e3\u003c/sub\u003eR1, at least in part via a direct binding of Cisd2 to IP\u003csub\u003e3\u003c/sub\u003eR1, involving the central, modulatory domain and potentially the LBC.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLoss of Cisd2 Does Not Alter Basal ER and cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e Levels\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo assess function effects of Cisd2 on IP\u003csub\u003e3\u003c/sub\u003eR related to Ca\u003csup\u003e2+\u003c/sup\u003e signaling, we generated Cisd2 KO HeLa cell populations using a CRISPR/Cas9 approach in combination with a gRNA targeting Cisd2 (gRNA\u003csup\u003eCisd2\u003c/sup\u003e), while expressing Cas9 in absence of gRNA\u003csup\u003eCisd2\u003c/sup\u003e served as control cells (CTRL), as described in (\u003cem\u003e24\u003c/em\u003e). \u003cstrong\u003eFigure 2A\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003e2B\u0026nbsp;\u003c/strong\u003eindicate that Cisd2-protein levels are almost absent in Cisd2-KO cells. Additionally, immunofluorescent staining with a monoclonal Cisd2 antibody in HeLa CTRL and Cisd2 KO cells further underpins the absence of endogenous CISD2 proteins in CISD2-KO cells (\u003cstrong\u003eFigure 2C\u003c/strong\u003e). First, we evaluated the impact of loss of Cisd2 on resting cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e levels by comparing the values of the ratiometric Fura-2 Ca\u003csup\u003e2+\u003c/sup\u003e sensor, loaded as Fura-2-AM in HeLa CTRL and Cisd2 KO cells. Fura-2 enables the measurement of resting cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e levels in a bleaching and loading-independent manner. Fura-2 ratios were recorded using a fluorescence microscope for 30 seconds in single cells (\u003cstrong\u003eFigure 2D\u003c/strong\u003e for representative fluorescence acquisitions). \u0026nbsp;Quantificationof basal Fura-2 ratios during a 30 second recording revealed no major difference in basal [Ca\u003csup\u003e2+\u003c/sup\u003e]\u003csub\u003ecyt\u003c/sub\u003e (\u003cstrong\u003eFigure 2E\u003c/strong\u003e). Furthermore, ER Ca\u003csup\u003e2+\u003c/sup\u003e store content was evaluated by addition of thapsigargin, a high-affinity and irreversible inhibitor of SERCA2, in extracellular Ca\u003csup\u003e2+\u0026nbsp;\u003c/sup\u003echelating conditions. Averaged traces of Cal520-loaded cell populations of HeLa CTRL and Cisd2 KO cells are shown in \u003cstrong\u003eFigure 2F\u003c/strong\u003e. Quantification of area under the curve (AUC) of the thapsigargin-releasable Ca\u003csup\u003e2+\u003c/sup\u003e indicated that absence of Cisd2 did not affect the ER Ca\u003csup\u003e2+\u003c/sup\u003e-store content. Moreover, Ca\u003csup\u003e2+\u003c/sup\u003e responses to ionomycin were recorded to obtain a measure of total intracellular Ca\u003csup\u003e2+\u003c/sup\u003e content (\u003cstrong\u003eFigure 2H\u003c/strong\u003e). Similarly, total intracellular Ca\u003csup\u003e2+\u003c/sup\u003e content was not affected by absence of Cisd2 (\u003cstrong\u003eFigure 2H\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003e2I\u003c/strong\u003e). These results convey that, at least in HeLa cells, Cisd2 is not required for the maintenance of resting Ca\u003csup\u003e2+\u003c/sup\u003e homeostasis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCisd2 Ensures IP\u003csub\u003e3\u003c/sub\u003eR-Mediated ER-Mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e Transfer\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNext, we evaluated the impact of Cisd2 deficiency on IP\u003csub\u003e3\u003c/sub\u003eR-mediated Ca\u003csup\u003e2+\u003c/sup\u003e dynamics and ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer. Ca\u003csup\u003e2+\u003c/sup\u003e responses to 5 µM ATP, a submaximal agonist concentration, and subsequently 100 µM ATP, a supramaximal agonist concentration, in both the cytosol and mitochondria were simultaneously recorded in single cells and compared between CTRL and Cisd2 KO HeLa cells (\u003cstrong\u003eFigure 3A and 3C)\u003c/strong\u003e. For this, the cytosolic Fluo-4 probe was multiplexed with the genetically encoded red mitochondrial CEPIA3 (R-mtCEPIA3) (\u003cem\u003e35\u003c/em\u003e). In response to 5 µM ATP, both CTRL and Cisd2 KO cells exhibited a robust cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e. Loss of Cisd2 tended to result in a slightly, yet not significant, decreased AUC for 5 mM ATP while the response to 100 µM ATP remain completely unaltered (\u003cstrong\u003eFigure 3B\u003c/strong\u003e). In contrast, at the level of mitochondria, the effects of Cisd2 were much more pronounced, as the mitochondrial [Ca\u003csup\u003e2+\u003c/sup\u003e] increases were severely blunted in Cisd2-KO cells compared to CTRL cells. \u0026nbsp;Cisd2 KO cells displayed a blunted Ca\u003csup\u003e2+\u003c/sup\u003e response to 5 µM, thereby reducing the ATP-evoked mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e uptake (\u003cstrong\u003eFigure 3D\u003c/strong\u003e). \u0026nbsp;To demonstrate that this change in mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e dynamics in Cisd2-KO cells was due to an on-target of effect of loss of Cisd2, we performed a rescue experiment, by re-expressing 3x-FLAG tagged Cisd2 in Cisd2 KO HeLa (\u003cstrong\u003eFigure 3E\u003c/strong\u003e). Similarly, cytosolic and mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e dynamics were recorded with Fluo-4 and R-mtCEPIA3, respectively (\u003cstrong\u003eFigure 3F\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003e3G\u003c/strong\u003e). AUC of cytosolic responses to 5 µM ATP was not significantly altered upon re-expression of Cisd2 (\u003cstrong\u003eFigure 3H\u003c/strong\u003e). However, at the level of mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e, Cisd2 enhanced the AUC of mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e uptake. (\u003cstrong\u003eFigure 3I\u003c/strong\u003e). These findings indicate that Cisd2 is important for efficient IP\u003csub\u003e3\u003c/sub\u003eR-mediated Ca\u003csup\u003e2+\u003c/sup\u003e transfer from the ER to mitochondria.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCisd2 is Enriched in MAMs independently of IP\u003csub\u003e3\u003c/sub\u003eR\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn concordance with our previous work (\u003cem\u003e24\u003c/em\u003e), we validated through biochemical MAM isolation that Cisd2 is enriched in the MAM fraction of HeLa CTRL cells. Cisd2-KO cells are used as a reference control (\u003cstrong\u003eFigure 4A\u003c/strong\u003e). Furthermore, Cisd2 was still enriched in MAM fractions of HeLa-TKO cells that lack all three IP\u003csub\u003e3\u003c/sub\u003eR isoforms (\u003cstrong\u003eFigure 4B\u003c/strong\u003e), indicating that the localization of Cisd2 to the MAMs is not strictly dependent on the presence of IP\u003csub\u003e3\u003c/sub\u003eRs. Moreover, we evaluated the protein levels of different MAM-resident proteins via immunoblotting of HeLa CTRL and Cisd2 lysates (\u003cstrong\u003eFigure 4C\u003c/strong\u003e). Densitometric quantification of key proteins IP\u003csub\u003e3\u003c/sub\u003eR, mitofusin 2 (Mfn2), Vdac1, vesicle-associated protein B (VapB) and Grp75 (\u003cstrong\u003eFigure 4D, E, F, G\u0026nbsp;\u003c/strong\u003e\u0026amp; \u003cstrong\u003eH\u003c/strong\u003e) showed no changed protein levels in absence of Cisd2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCisd2 Maintains ER-Mitochondrial Contact Site Integrity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNext, in the search for a mechanistic explanation why loss of Cisd2 decreased ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer, we sought to explore the effect of loss of Cisd2 to extent of ER-mitochondrial contact. The short ER-mitochondria SPLICS (SPLICS\u003csub\u003eS\u003c/sub\u003e) sensor was transfected in HeLa CTRL and Cisd2 KO cells. The detection of SPLICS\u003csub\u003eS\u003c/sub\u003e signal is based on a split green fluorescent protein (GFP), that assembles into a functional fluorescent protein only when ER membranes and the OMM are within a distance smaller than 10 nm (\u003cem\u003e36\u003c/em\u003e). Z-stacks of HeLa CTRL and Cisd2 KO cells were aquired using confocal fluorescence microscopy and assembled into volumetric projections. Representative acquisitions of SPLICS\u003csub\u003eS\u003c/sub\u003e fluorescence, unfiltered emission light serving as a brightfield replacement and volume renders are shown in \u003cstrong\u003eFigure 5A\u003c/strong\u003e. Distinct ER-mitochondrial contact sites per cell (\u003cstrong\u003eFigure 5B\u003c/strong\u003e) and total volume of ER-mitochondrial contact per cell (\u003cstrong\u003eFigure 5C\u003c/strong\u003e) were significantly decreased in Cisd2 KO cells. To obtain an independent validation of impaired ER-mitochondrial contact formation, we used MAMtracker Green. Contrarily to SPLICS\u003csub\u003eS\u003c/sub\u003e, MAMtracker Green fluorescence is dependent on reversible dimerization of a dimerization-dependent GFP (\u003cem\u003e37\u003c/em\u003e). MAMtracker Green was co-expressed with Cisd2-p2a-mCherry, or p2a-mCherry control (\u003cstrong\u003eFigure 5D\u003c/strong\u003e). Correlating with the SPLICS\u003csub\u003eS\u003c/sub\u003e results, MAMtracker Green fluorescence was significantly decreased in Cisd2 KO HeLa cells compared to CTRL cells. Of note, Cisd2 overexpression fully rescued loss of MAMtracker Green fluorescence, suggesting a recovery in ER-mitochondrial contact (\u003cstrong\u003eFigure 5E\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMitochondrial Health is Unaffected by Cisd2 Deficiency in HeLa Cells\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate whether Cisd2 loss affects mitochondrial health, mitochondrial volume, morphology and function was assessed. HeLa CTRL and Cisd2 KO cells were co-stained with Mitotracker Green and Cellmask Orange stainings (\u003cstrong\u003eFigure 6A\u003c/strong\u003e) Morphometric analysis revealed no significant difference in mitochondrial volume, surface area, or branching (\u003cstrong\u003eFigures 6B\u003c/strong\u003e, \u003cstrong\u003e6C\u003c/strong\u003e and \u003cstrong\u003e6D\u003c/strong\u003e). Furthermore, mitochondrial inner membrane potential, assessed by JC-1 staining, was similar between CTRL and Cisd2 KO cells (\u003cstrong\u003eFigure 6E\u003c/strong\u003e, \u003cstrong\u003e6F\u003c/strong\u003e). Mitochondrial respiration, measured by a Seahorse Mito Stress Test assay, was not affected by loss of Cisd2 in HeLa cells (\u003cstrong\u003eFigures 6G\u003c/strong\u003e and \u003cstrong\u003e6H\u003c/strong\u003e). In brief, these results indicate that mitochondrial morphology and metabolism are largely intact despite the loss of Cisd2 in HeLa cells.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCisd2 Deficiency Increases Autophagic Flux\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSince Chang et al. reported that Cisd2 is required for Bcl-2’s ability to inhibit Beclin 1-dependent autophagy in H1299 cells (\u003cem\u003e22\u003c/em\u003e), we further studied autophagic flux. We found that Cisd2-KO cells displayed an increased autophagic flux, as indicated by an elevation of detected LC3-II levels in response to blocking autophagosomal-lysosomal fusion with bafilomycin A1 (\u003cstrong\u003eFigure 7A\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003e7B\u003c/strong\u003e). Additionally, when transfecting HeLa CTRL and Cisd2 KO cells with GFP-LC3, the number of detected GFP-LC3 punctae was significantly higher in bafilomycin A1 treated Cisd2 KO cells, suggesting enhanced autophagosome formation (\u003cstrong\u003eFigure 7D\u003c/strong\u003e). These findings highlight a potential link between Cisd2 and the regulation of autophagy. Autophagy is a process that can precede apoptosis (\u003cem\u003e38\u003c/em\u003e). To evaluate a potentiating effect to apoptotic stimuli in absence of Cisd2, HeLa CTRL and Cisd2 KO cells were subjected to several concentrations of staurosporine, a potent inducer of apoptosis. Apoptosis was monitored by poly (ADP-ribose) polymerase (PARP) cleavage detecting through immunoblotting, an established ratiometric readout for apoptosis. A representative immunoblot is shown in \u003cstrong\u003eFigure 7E\u003c/strong\u003e. No heightened sensitivity to staurosporine, nor a significantly elevated level of apoptosis was observed in Cisd2 KO HeLa cells (\u003cstrong\u003eFigure 7F\u003c/strong\u003e). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eiPSC-derived Cisd2-Deficient Cortical Neurons Display Diminished IP\u003csub\u003e3\u003c/sub\u003eR-Mediated Ca\u003csup\u003e2+\u003c/sup\u003e Release\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs loss of Cisd2 in HeLa cells impacted ER-mitochondrial Ca\u003csup\u003e2+\u0026nbsp;\u003c/sup\u003etransfer but without major impacts on cell function, we decided to study Cisd2 in a WS disease-relevant cellular context by using control and Cisd2-KO iPSCs into cortical neurons. The same approach as in HeLa cells was used to generate the knockout of Cisd2 in iPSCs. After 72h of selection with puromycin, complete loss of detectable Cisd2 was attained (\u003cstrong\u003eFigure 8A\u003c/strong\u003e). iPSCs still express purinergic receptors, enabling to study IP\u003csub\u003e3\u003c/sub\u003eR-mediated Ca\u003csup\u003e2+\u003c/sup\u003e release through addition of ATP in Cal520 loaded iPSCs (\u003cstrong\u003eFigure 8B\u003c/strong\u003e). Similarly to HeLa cells, loss of Cisd2 did not result in a significantly decreased sensitivity to ATP-evoked IP\u003csub\u003e3\u003c/sub\u003eR-mediated Ca\u003csup\u003e2+\u003c/sup\u003e releases (\u003cstrong\u003eFigure 8C\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 8D\u0026nbsp;\u003c/strong\u003edisplays the timeline of differentiation from iPSC to cortical neurons. After 54 days of differentiation, differentiated cortical neurons were used for experiments. Of note, differentiated cortical neurons expressed SATB2 and MAP2 markers, as shown using immunostaining (\u003cstrong\u003eFigure 8E\u003c/strong\u003e). \u003cstrong\u003eFigure 8F\u003c/strong\u003e shows a proof-of-concept averaged trace of cortical neurons loaded with Cal520, responding to glutamate as an agonist and 1:10 glycine as a co-agonist. Extracellular Ca\u003csup\u003e2+\u003c/sup\u003e was chelated, and cortical neurons were pretreated with 30 µM, effectively abolishing any Ca\u003csup\u003e2+\u003c/sup\u003e flux through RyRs. This demonstrates the observed Ca\u003csup\u003e2+\u0026nbsp;\u003c/sup\u003eresponse to be \u003cem\u003ebona fide\u0026nbsp;\u003c/em\u003eIP\u003csub\u003e3\u003c/sub\u003eR-mediated, corresponding with work of others (\u003cem\u003e39\u003c/em\u003e). Cal520-loaded cortical neurons were recorded responding to 10 mM glutamate, supplemented with 1 mM glycine, and 100 mM glutamate, supplemented with 10 mM glycine, to generate a supramaximal response (\u003cstrong\u003eFigure 8G\u003c/strong\u003e). Responses to both concentrations of glutamate were severely diminished in Cisd2 KO cortical neurons (\u003cstrong\u003eFigure 8H\u003c/strong\u003e). Conversely, thapsigargin-releasable Ca\u003csup\u003e2+\u003c/sup\u003e was significantly elevated in Cisd2 KO cortical neurons. Thus, in the disease relevant context of cortical neurons, loss of CISD2 also impairs Ca2+ release in the cytosol, which was not due to a depletion of the ER Ca\u003csup\u003e2+\u0026nbsp;\u003c/sup\u003estore content.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCisd2 Is Essential for ER-Mitochondrial Contact Integrity and Mitochondrial Function in Cortical Neurons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe observed decreased IP\u003csub\u003e3\u003c/sub\u003eR-mediated Ca\u003csup\u003e2+\u003c/sup\u003e release in Cisd2-deficient cortical neurons might hold repercussions for fluxed Ca\u003csup\u003e2+\u003c/sup\u003e to the mitochondria. Green mtCEPIA2 (\u003cem\u003e40\u003c/em\u003e) was transfected in CTRL and Cisd2 cortical neurons and cells were stimulated with 100 mM glutamate / 10 mM glycine as a supramaximal agonist of the IP\u003csub\u003e3\u003c/sub\u003eR\u003cstrong\u003e\u0026nbsp;(Figure 9A\u003c/strong\u003e). In correspondence with the detected responses in the cytosol, AUC, amplitude and rate of rise of mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e responses were significantly diminished in absence of Cisd2 (\u003cstrong\u003eFigure 9B\u003c/strong\u003e). By expressing SPLICS\u003csub\u003eS\u003c/sub\u003e in cortical neurons, we evaluated the decreased mitochondrial Ca2+ uptake was due to a decreased Ca\u003csup\u003e2+\u003c/sup\u003e efflux from the ER or can also be due to a decrease in ER-mitochondrial contact (\u003cstrong\u003eFigure 9C\u003c/strong\u003e). The number of ER-mitochondrial contact sites per cell were significantly decreased in cortical neurons deficient for Cisd2 (\u003cstrong\u003eFigure 9D\u003c/strong\u003e). In addition, MAMtracker Green was expressed, together with Cisd2-p2a-mCherry or p2a-mCherry EV as a control (\u003cstrong\u003eFigure 9E\u003c/strong\u003e). Normalized MAMtracker Green fluorescence was significantly decreased in absence of Cisd2, while re-expressing Cisd2 in Cisd2 KO cortical neurons effectively restored MAMtracker Green fluorescence.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLoss of CISD2 impairs mitochondrial function in iPSC-derived cortical neurons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the more profound impacts of loss of CISD2 on intracellular Ca2+ dynamics in iPSC-derived cortical neurons, we wondered whether mitochondrial function was more severely perturbed by CISD2 deficiency in these cell systems. We therefore assessed mitochondrial health using Seahorse Mito Stress Test assays in control and CISD2-KO iPSC-derived cortical neurons (\u003cstrong\u003eFigure 9G-J).\u0026nbsp;\u003c/strong\u003eIn contrast to HeLa cells,in iPSC-derived cortical neurons,mitochondrial health was severely affected by loss of Cisd2, evidenced by significantly decreased basal OCR (\u003cstrong\u003eFigure 9H\u003c/strong\u003e), lowered ATP-linked respiration (\u003cstrong\u003eFigure 9I\u003c/strong\u003e) and a decreased maximal capacity (\u003cstrong\u003eFigure 9J\u003c/strong\u003e) in mitochondria of Cisd2-deficient cortical neurons. Furthermore, CTRL and Cisd2 KO cortical neurons were stained with JC-1 to evaluate maintenance of IMM potential (\u003cstrong\u003eFigure 9K\u003c/strong\u003e). Cisd2-deficiency led to a significant loss of IMM potential, as normalized JC-1 ratios were significantly decreased compared to CTRL neurons (\u003cstrong\u003eFigure 9L\u003c/strong\u003e). To expand these findings, we also determined the neuronal ATP context in rat neonatal cortical neurons using PercevalHR, an established sensor for intracellular ATP / ADP ratio (\u003cem\u003e41\u003c/em\u003e). In these cells, we performed shRNA-mediated knockdown of CISD2, which evoked a decrease in the PercevalHR ratio (\u003cstrong\u003eFigure 9M\u003c/strong\u003e), indicative for a lower intracellular ATP content in cells lacking CISD2. (7). Re-expressing Cisd2 in these shRNA-CISD2-treated cells rescued the loss of ATP content.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLoss of CISD2 Increases Autophagic Flux and Sensitizes Neurons to Apoptosis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFinally, we assessed whether autophagy flux and apoptosis sensitivity were affected by loss of CISD2 in iPSC-derived cortical neurons. To determine autophagic flux, we used the GFP-RFP-LC3, whereby GFP is effectively quenched in lysosomal pH conditions, while RFP fluorescence is unaffected (\u003cstrong\u003eFigure 10A\u003c/strong\u003e). As such, the GFP / RFP autophagosomal punctae ratio can be used as a readout for autophagic flux. GFP / RFP punctae ratios was significantly decreased in Cisd2 KO cortical neurons compared to CTRL neurons, indicating more autophagic turnover by the lysosomes. This decrease in GFP/RFP ratio observed in CISD2-KO cells was abrogated by the addition of bafilomycin A1, a lysosomal inhibitor effectively elevated (\u003cstrong\u003eFigure 10B\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFinally, we evaluated whether the impacts on mitochondrial health could sensitize the cortical neurons to cell stress treatments such as staurosporine. Therefore, CTRL and Cisd2 KO cortical neurons were subjected to different doses of staurosporine and PARP ratios were determined through immunoblotting (\u003cstrong\u003eFigure 10C\u003c/strong\u003e). Volumetric quantification of cleaved PARP over total PARP indicated that Cisd2-KO iPSC-derived cortical neurons were much more sensitive to staurosporine than CTRL iPsC-derived cortical neurons in two ways: first, the threshold for inducing apoptosis in Cisd2-KO neurons is lower than that of control neurons (cfr 100 nM condition); second, at higher staurosporine concentrations (300 nM-1microM), Cisd2-KO neurons display more apoptosis than control neurons (\u003cstrong\u003eFigure 10D\u003c/strong\u003e).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe main findings of this work are that Cisd2 can directly interact with IP\u003csub\u003e3\u003c/sub\u003eR channels, intracellular Ca\u003csup\u003e2+\u003c/sup\u003e-release channels not only delivering Ca\u003csup\u003e2+\u003c/sup\u003e towards cytosol but also towards mitochondria. Loss of Cisd2 impaired MERCS organization with subsequent lowered ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer, both in HeLa and in neurons. Yet, only neurons – not HeLa cells - displayed severe perturbations in their cellular functions upon loss of Cisd2, showing decreased mitochondrial potential, decreased oxidative phosphorylation and ATP production, accompanied with increased autophagic flux. As a consequence, Cisd2-lacking neurons became susceptible to chemical cell stress inducers such as staurosporine, while Cisd2-lacking HeLa cells were rather unaffected. Consistently, loss of Cisd2 had more profound impacts on Ca\u003csup\u003e2+\u003c/sup\u003e dynamics in neurons compared to HeLa cells, heavily suppressing cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e signals in neurons but not in HeLa cells exposed to physiological agonists.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOf note, autophagy flux appeared increased in both HeLa and in neurons lacking Cisd2, thus indicating that this process may be (in part) unrelated to impaired metabolism. This hints towards other functions for Cisd2 in control of autophagy. In fact, Cisd2 has been reported to function as an important co-factor for Bcl-2, which operates as an anti-autophagic protein by scaffolding Beclin 1 (\u003cem\u003e22, 42\u003c/em\u003e). Hence, in absence of Cisd2, it’s possible that Bcl-2, endogenously present in HeLa and neurons, is impaired in suppressing autophagy, thereby explaining the increased autophagic flux in both HeLa cells and neurons. Though other mechanisms might be at play, since Cisd2 has been reported to prevent starvation-induced autophagy by inhibiting AMPK activity, at least in cardiomyocytes (\u003cem\u003e43\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eA recent unbiased proteome-wide analysis of MAMs using Contact-ID, a Bio-ID-based labeling approach of proteins proximal to ER-mitochondrial contact sites revealed 115 MAM-specific proteins of which Cisd2 was ranked among the top candidates (\u003cem\u003e26\u003c/em\u003e). Validating this screen, our findings indicate that, consistent with other reports (\u003cem\u003e24, 26, 28\u003c/em\u003e), that Cisd2 is a critical component for the adequate organization of MERCS in cells. Being one of the two Wolfram-linked genes besides \u003cem\u003eWFS1\u003c/em\u003e, it appears that Wfs1 and Cisd2 execute converging functions at MERCS, as loss of either Wfs1 or Cisd2 results in decreased MERCS and impaired ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer. Moreover, recent work indicated that Wfs1 and Cisd2 can interact with each other and that impaired ER Ca\u003csup\u003e2+\u003c/sup\u003e release by loss of Wfs1 could be compensated by overexpression of Cisd2 (\u003cem\u003e30\u003c/em\u003e). Furthermore, it’s fascinating that Cisd2 via its cytosolic region can directly interact with IP\u003csub\u003e3\u003c/sub\u003eRs, yet this interaction does not seem to functionally control IP\u003csub\u003e3\u003c/sub\u003eR function as addition of the purified cytosolic Cisd2 fragment did not change the open probability of single IP\u003csub\u003e3\u003c/sub\u003eR channels (\u003cem\u003e24\u003c/em\u003e). This correlates with our findings in HeLa cells that displayed comparable agonist-evoked Ca\u003csup\u003e2+\u003c/sup\u003e signals in the cytosol irrespective of the presence or absence of Cisd2 (\u003cem\u003e24\u003c/em\u003e). Of note, in those conditions, we have used intermediated [agonist], so that in principle both sensitizing and inhibitory effects should have been detectable. Instead in neurons, loss of Cisd2 did severely hamper cytosolic Ca\u003csup\u003e2+\u003c/sup\u003e rises evoked by the physiological agonist glutamate. Moreover, these experiments were performed in absence of extracellular Ca\u003csup\u003e2+\u003c/sup\u003e, thus ensuring the Ca\u003csup\u003e2+\u003c/sup\u003e release originated from the ER through IP\u003csub\u003e3\u003c/sub\u003eRs. Of note, ryanodine addition, thereby blocking ryanodine receptor channels, did not majorly lower glutamate-evoked Ca\u003csup\u003e2+\u003c/sup\u003e rises further underpinning that those signals are mainly mediated through IP\u003csub\u003e3\u003c/sub\u003eRs, which is consistent with previous observations reported by other teams (\u003cem\u003e39\u003c/em\u003e). The diverging impact of Cisd2 on intracellular Ca\u003csup\u003e2+\u003c/sup\u003e dynamics dependent on the cell type is fascinating and requires further study. Hence, it will be important to unravel the mechanisms responsible for this effect. It’s tempting to speculate that in neurons Cisd2 cooperates with other factors/proteins to support the adequate function of IP\u003csub\u003e3\u003c/sub\u003eRs in mediating Ca\u003csup\u003e2+\u003c/sup\u003e release. For instance, the impact of Wfs1 on MAMs and ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfers is executed via Neuronal Calcium Sensor-1 (NCS-1). Thus, it’s possible that NCS-1, which is an IP\u003csub\u003e3\u003c/sub\u003eR-accessory protein that enhances IP\u003csub\u003e3\u003c/sub\u003eR function (\u003cem\u003e19\u003c/em\u003e), too cooperates with Cisd2, whereby loss of Cisd2 could affect the presence of NCS-1 in complex with IP\u003csub\u003e3\u003c/sub\u003eRs, thereby limiting its function. Moreover, the importance of the IP\u003csub\u003e3\u003c/sub\u003eR-Cisd2 complexes and such potentially additional proteins for Cisd2’s impact on ER-mitochondrial integrity requires further scrutiny. Therefore, it will be of interest to assess whether Cisd2 requires the presence of IP\u003csub\u003e3\u003c/sub\u003eRs to sustain ER-mitochondrial contact sites.\u003c/p\u003e\n\u003cp\u003eFurthermore, putting our findings in context with previous study performed in patient fibroblasts carrying Cisd2 mutations, it was found that ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e was increased (\u003cem\u003e33\u003c/em\u003e). In that study, it was found that Cisd2-mutated fibroblasts displayed increased ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer compared to the age-matched controls and that ER-mitochondrial contact sites (in terms of contact ER/mito / mm\u003csup\u003e2\u003c/sup\u003e, number of mitochondria with ER contact sites and # mitochondria length adjacent to ER) are increased. Of interest, also in these patient fibroblasts, that represent a cell type likely not affected in WS2, mitochondrial morphology and oxidative phosphorylation function was only limitedly affected, thereby resembling our findings obtained in HeLa cells at least for mitochondrial function. Therefore, further work is needed to assess how Cisd2 mutations impact ER-mitochondrial contact sites and ER-mitochondrial contact site formation. The differences with our findings are striking but could be due to different reasons. First, in our work, we have used isogenic cell models thereby allowing a careful comparison between wild-type and KO models, while Rouzier et al used cells derived from different individuals, though being close relatives (\u003cem\u003e33\u003c/em\u003e). Second, in our study, we have focused on KO/knockdown approaches thus lowering overall Cisd2-protein levels, while Rouzier et al have studied the impact of Cisd2 patient mutations associated with aberrant Cisd2 function. Hence, it’s certainly possible that loss of Cisd2 impairs ER-mitochondrial contacts while patient-associated Cisd2 mutations display gain-of-function properties for certain Cisd2-related functions. Therefore, further work will be needed to explore the impact of these functions on cell function. Third, our work also reveals the critical importance of the cell type, and thus the impact of Cisd2 mutations in the context of neurons, a cell type affected in WS2, could certainly be different than in fibroblasts, a cell type not affected in WS2.\u003c/p\u003e\n\u003cp\u003eFinally, it will be of interest to explore whether strategies that improve ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer can restore mitochondrial function and improve cell stress resilience in neurons. While genetic linkers could offer proof-of-concept evidence, such strategies are less tangible to translate into therapeutically applicable interventions. An attractive target is Sigma-1 receptor, an ER-resident chaperone important of ER-mitochondrial contact sites and Ca\u003csup\u003e2+\u003c/sup\u003e transfer. Recent developments in the pharmacology of Sigma-1 receptors have resulted in small molecule Sigma-1 receptor agonists such as PRE-084 (\u003cem\u003e31\u003c/em\u003e). Excitingly, PRE-084 has recently been applied in preclinical models of WS1 including cellular models and zebrafish, thereby alleviating several symptoms of disease. Hence, the application of PRE-084 in cells lacking Cisd2 to rescue mitochondrial deficits could be of interest.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn conclusion, our study highlights the critical role of Cisd2 in ER-mitochondrial communication and its impact on cellular function, particularly in neurons. We demonstrate that Cisd2 directly interacts with IP\u003csub\u003e3\u003c/sub\u003eRs and is essential for maintaining MERCS integrity and ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer. Notably, the loss of Cisd2 in neurons leads to a greater extent of metabolic and Ca\u003csup\u003e2+\u003c/sup\u003e signaling defects than in HeLa cells, underscoring the cell-type-specific functions of Cisd2. These findings raise important questions regarding the molecular mechanisms by which Cisd2 modulates intracellular Ca\u003csup\u003e2+\u003c/sup\u003e dynamics, potentially involving interactions with other regulatory proteins such as NCS-1. Furthermore, the contrasting effects observed in patient-derived fibroblasts by others (\u003cem\u003e33\u003c/em\u003e) suggest that distinct Cisd2 mutations may differentially affect ER-mitochondrial contact sites and function. Future studies should focus on elucidating these mechanisms and exploring therapeutic strategies to restore ER-mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e transfer. The potential of Sigma-1 receptor agonists, such as PRE-084, to rescue mitochondrial deficits presents an exciting avenue for further investigation in Cisd2-related pathologies.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e\u003cstrong\u003eChemicals and consumables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnless specifically stated, all chemicals and consumables were obtained from Thermo Fischer (Merelbeke, Belgium).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAntibodies\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRabbit anti-IP\u003csub\u003e3\u003c/sub\u003eR1 (alias: Rbt03, 1:1000, homemade) (\u003cem\u003e44\u003c/em\u003e), rabbit polyclonal anti-CISD2 (1:1000, ABClonal, A5231), mouse monoclonal anti-IP\u003csub\u003e3\u003c/sub\u003eR1 E-8 (5 \u0026micro;g per coIP, Santa cruz, sc-271197), mouse aspecific IgG (5 \u0026micro;g per coIP, sc-2025), mouse monoclonal anti-GAPDH (1:1000, Merck, G8795), mouse monoclonal anti-Cisd2 (1:100, Proteintech, 66082-1-Ig), goat anti mouse (1:1000, A11017), mouse monoclonal anti-vinculin (1:10000, Merck, V9131), rabbit monoclonal anti-PERK (1:1000, Cell Signaling Technology, 3192), rabbit monoclonal anti-calnexin (1:1000, Cell Signaling Technology, 2679S), mouse monoclonal anti-VDAC1 (1:1000, Abcam, ab14734), rabbit polyclonal anti-Cytochrome C (1:1000, Cell Signaling Technology, 4272), rabbit pan anti-IP\u003csub\u003e3\u003c/sub\u003eR1 (alias: Rbt475, 1:1000, homemade), rabbit polyclonal anti-mitofusin 2 (1:1000, Abcam, ab50838), mouse monoclonal anti-\u0026alpha; Tubulin (1:500, Thermo Fischer, A11126), rabbit polyclonal anti-VABP (1:1000, Invitrogen, PA5-53023), rabbit polyclonal anti-LC3 (1:500, Cell Signaling Technology, 4108S), rabbit monoclonal anti-PARP (1:1000, Cell Signaling, 9532S).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePlasmids and constructs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the purification of Cisd2\u003csup\u003eCYT\u003c/sup\u003e the cDNA sequence of the cytosolic part of the protein was cloned as a gBlock\u0026trade; (Integrated DNA Technologies, Leuven, Belgium) in a pET21b(+) plasmid for purification using standard His-purification protocols, as previously performed (\u003cem\u003e24, 45\u003c/em\u003e).\u003cbr\u003egBlock\u0026trade; sequence:\u003cbr\u003e5\u0026rsquo;-TACTTCTTTCTTCTTCAGTATTAGTGGACCCACATTATCTCCTGTCAATTCATTGTGTTTATTATGTGAACCATCGCAGG\u003cbr\u003eCAGGAAACGTTTTAGAACGCCAACACCTACAATAAGCTGCTTTAGTAAGACACAAATCTTCAATGTTTATTTCATTCACTAC\u003cbr\u003eTTTCGGATTTTCCTTTTGTATTTTAAGATTAATCAAGCTATCC\u003cbr\u003eTTCTGTTGTTTCTTCTTCGGGAGGAATGGACGCAT-3\u0026rsquo;. pCMV26-Cisd2 was generated by cloning the cDNA sequence of Cisd2 using HindIII and EcoRI as described before (\u003cem\u003e24\u003c/em\u003e). pCMV24-Cisd2-p2a-mCherry was subcloned as follows: a gBlock\u0026trade; containing the cDNA for Cisd2 (Integrated DNA Technologies, Leuven, Belgium) was restriction-ligated with HindIII and EcoRI into an empty pCMV24-p2a-mCherry, created as described in (\u003cem\u003e46\u003c/em\u003e).\u003cbr\u003egBlock sequence: 5\u0026rsquo;-GCCAAGCTTGTGCTGGAGAGCGTGGCCCGTATCGTGAAGGTGCAGCTCCCTGCATATCTGA\u003cbr\u003eAGCGGCTCCCAGTCCCTGAAAGCATTACCGGGTTCGCTAGGCTCACAGTTTCAGAATGGCTTCGGTTATTGCC\u003cbr\u003eTTTCCTTGGTGTACTCGCACTTCTTGGCTACCTTGCAGTTCGTCCATTCCTCCCG\u003cbr\u003eAAGAAGAAACAACAGAAGGATAGCTTGATTAATCTTAAAATACAAAAGGAAAATCCGAAAGTAGTGAATGAAATA\u003cbr\u003eAACATTGAAGATTTGTGTCTTACTAAAGCAGCTTATTGTAGGTGTTGGCGTTCTAAAACGTTTCCTGCCTGCGATGGTTC\u003cbr\u003eACATAATAAACACAATGAATTGACAGGAGATAATGTGGGTCCACTAATACTGAAGTACCCATACGATGTTCCAGATTACGCT\u003cbr\u003eAAGAAAGAAGTAGAATTCGCCGC-3\u0026rsquo;. pSpCas9(BB)-2A-Puro (PX459) was a gift from Feng Zhang (Addgene plasmid # 48139; http://n2t.net/addgene:48139; RRID:Addgene_48139) (\u003cem\u003e47\u003c/em\u003e). pCMV R-CEPIA3mt was a gift from Masamitsu Iino (Addgene plasmid # 140464 ; http://n2t.net/addgene:140464 ; RRID:Addgene_140464). SPLICS Mt-ER Short P2A was a gift from Marisa Brini \u0026amp; Tito Cal\u0026igrave; (Addgene plasmid # 164108 ; http://n2t.net/addgene:164108 ; RRID:Addgene_164108). GW1-PercevalHR was a gift from Gary Yellen (Addgene plasmid # 49082 ; http://n2t.net/addgene:49082 ; RRID:Addgene_49082). pCMV CEPIA2mt was a gift from Masamitsu Iino (Addgene plasmid # 58218 ; http://n2t.net/addgene:58218 ; RRID:Addgene_58218).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell culture and transfections\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHEK293 IP\u003csub\u003e3\u003c/sub\u003eR TKO cells expressing rat IP\u003csub\u003e3\u003c/sub\u003eR1 were kindly gifted by Dr. Yule and were cultured at 37 \u0026deg;C, 10 % CO\u003csub\u003e2\u003c/sub\u003e in Dulbecco\u0026rsquo;s Modified Eagle\u0026rsquo;s Medium (DMEM), supplemented with 10 % fetal calf serum, 100 IU/mL penicillin and 100 \u0026mu;g/mL streptomycin, 2 mg/mL geneticin. HeLa cells were cultured at 37 \u0026deg;C, 5 % CO\u003csub\u003e2\u003c/sub\u003e in Dulbecco\u0026apos;s Modified Eagle\u0026apos;s Medium (DMEM), supplemented with 5 % fetal calf serum, 100 IU/mL penicillin and 100 \u0026mu;g/mL streptomycin, 2 mM Glutamax. HeLa cells were seeded 24 h before transfection and were transfected using Mirus TransIT-X2 transfection reagent (Mirus Bio, WI, USA) with a 2:1 transfection reagent in \u0026mu;L per \u0026mu;g DNA. Cells were routinely checked for the absence of mycoplasma infection. hiPSCs were seeded on human Matrigel coated plates (Corning, Lasne, Belgium, 354277) in mTESR (StemCell Technologies, Cambridge, UK, 8580) with 1:100 Revitacell (Life Technologies, Bleiswijk, The Netherlands, A2644501). Maintenance of cultures was performed in E8 flex medium (E8 basal medium complemented with E8 supplement Flex (A2858501) and 5 U/mL penicillin-streptomycin) and splitting twide a week with 0.5 mM EDTA. Differentiation from hiPSCs to cortical neurons was performed as extensively described in (\u003cem\u003e48\u003c/em\u003e). All experiments were performed on neurons differentiated for between 52 and 59 days. Cortical neurons were transfected using Mrius TransIT-X2 with a 3:1 transfection reagent in \u0026micro;L per \u0026micro;g DNA. Subsequently, 6 h after transfection, neuronal maintenance medium was replaced to neuronal maintenance medium supplemented with 0.5 mM kynurenic acid.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGeneration of Cisd2 KO and Cas9 CTRL HeLa cells and iPSCs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCTRL and Cisd2 KO HeLa cells were generated as described before (\u003cem\u003e24\u003c/em\u003e). In brief, HeLa cells were transfected with pSpCas9(BB)-2A-Puro vectors and selected for 48 h in presence of 3 \u0026micro;g/mL puromycin. CTRL and Cisd2 KO iPSCs were generated by seeding iPSCs on 6 well plates coated with 1 % Geltrex\u0026trade; (A1413301). 24 h after seeding, medium was changed to mTESR and supplemented with 1:500 ROCK inhibitor (Tocris, Bristol, UK, Y-27632). 2 h after, iPSCs were nucleofected (Lonza) with 3 \u0026micro;g pSpCas9(BB)-2A-Puro vectors. Cells were selected with 0.4 \u0026micro;g/mL puromycin for 72 h. Guide RNA sequence (GGAGCTGCACCTTCACGATA) was obtained from Synthego (Redwood City, CA, USA).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProtein purification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBL21 (DE3) E. coli bacteria were transformed with the plasmids pGEX6p2 inserted with the following fragments of the ratIP3R1: suppressor domain (SD), ligand binding region (LBR), ligand binding region (LBR), fragment 3, fragment 4, fragment 5, and fragment 6. After growing at 37 \u0026deg;C to an A600 of 0.2, the culture was incubated at 14 \u0026deg;C for 20 h with 100 \u0026mu;M isopropyl-d-1-thiogalactopyranoside for protein expression. Benzamidine and Fenylmethylsulfonylfluoride (PMSF) were added to a final concentration of 0.23 mM and 0.83 mM respectively. The bacteria were harvested via centrifugation at 5000 \u0026times;g for 10 min and resuspended in phosphate-buffered saline (PBS) containing 1% Triton X-100. Samples were sonicated 15 times for 10 seconds at 20 kHz using a MSE Ltd. (Westminster, Great Britain) probe sonicator with a 10 min pause after each 5 sonication rounds. The resulting bacterial lysate was centrifuged at 10,000 rpm for 20 min using a Sorvall SS-34 rotor at 4\u0026deg;C. Supernatants were collected and incubated with glutathione-Sepharose 4B beads for 2 hours at 4 \u0026deg;C. Subsequently, the beads were washed two times using 6 mL of the following buffers supplemented 0.83 mM benzamidine and 0.23 mM PMSF respectively: 1% Triton X-100 in PBS, PBS, and 50 mM Tris-HCl (pH 8.0). The GST-tagged proteins were eluted with 50 mM Tris-HCl containing 10 mM glutathione (pH 8.0) and dialyzed with PBS. The resulting protein purity was evaluated via western blotting and Coomassie blue staining (Imperial Protein Stain, Thermo Fisher, Cat. 24615). Images of the gel were taken using the ChemiDoc\u0026trade; MP imaging system and Image Lab 6.1 software (Bio-Rad, Hercules, CA, USA).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCo-immunoprecipitation assays\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCTRL and Cisd2 KO HEK IP\u003csub\u003e3\u003c/sub\u003eR TKO overexpressing IP\u003csub\u003e3\u003c/sub\u003eR1 were seeded 72 h before harvesting. Cells were harvested by scraping in ice cold PBS and lysed with a 3-((3-cholamidopropyl) dimethylammonio)-1-propanesulfonate (CHAPS) based buffer (50 mM Tris, 100 mM NaCl, 2 mM Ethylenediaminetetraacetic acid, 50 mM NaF, 1 mM Na3VO4, 1 % CHAPS and protease inhibitor tablets (Roche, Basel, Switzerland) according to manufacturer\u0026rsquo;s instructions). 5 \u0026micro;g of anti IP\u003csub\u003e3\u003c/sub\u003eR1 E-8 antibody or mouse aspecific IgG antibody was incubated with 20 \u0026micro;L of protein G coated Dynabeads\u0026trade; (ThermoFischer Scientific, Bleiswijk, the Netherlands) at room temperature for 30 minutes. Next, beads were incubated with 1000 \u0026micro;g of cell lysate overnight at 4 \u0026deg;C in CHAPS lysis buffer. Subsequently, beads were washed with ice cold PBS. Proteins were eluted from the beads using SDS elution buffer (0.2 % SDS + 0.01 % Tween pH 8).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunoblotting\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNuPAGE\u0026trade; LDS Sample Buffer was added to all immunoblot samples which were boiled and ran on NuPAGE\u0026trade; 4\u0026ndash;12 % Bis-Tris gels. Subsequently, proteins were transferred on a polyvinylidene fluoride membrane. Membranes were blocked with Tris-buffered saline (TBS) containing 5 % milk powder and 0.1 % Tween and incubated with primary antibody overnight. The next day, membranes were incubated for 1 h with secondary horseradish peroxidase-linked antibodies in TBS 0.1 % Tween. Pierce\u0026trade; ECL chemiluminescent western blot reagent was used for detection in a Chemidoc imaging system (Bio-Rad, CA, USA).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMicroscale thermophoresis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUsing the Monolith His-Tag Labeling Kit RED-tris-NTA 2nd Generation (Nano Temper Technologies, Munich, Germany) purified 6xHis-tagged CISD2\u003csup\u003eCYT\u003c/sup\u003e protein was fluorescently labeled. The fluorescently labeled Cisd2\u003csup\u003eCYT\u003c/sup\u003e was used to determine binding affinities for purified full-channel rIP3R1 (obtained in collaboration with the lab of Irina Serysheva) as well as purified GST-tagged fragments of rIP3R1 through microscale thermophoresis using a Monolith NT automated instrument (Nano Temper Technologies), as previously published [PMID: 39370046]. The concentration of labeled CISD2\u003csup\u003eCYT\u003c/sup\u003e was maintain at a 10\u0026thinsp;nM. The concentrations of the full-channel IP3R1 and IP3R1 fragments vary depending on the biological replicate and were optimized to ensure datapoints at higher protein concentrations, thus enabling more accurate prediction of any possible interaction. Measurements were performed with a pico-red laser channel at 20 % excitation and 40 % MST power in steady-state conditions using premium capillaries. Fluorescence was normalized to the average baseline values obtained in the cold region (blue shaded area, -1 s until 0 s) and differences in normalized fluorescence (F / F0) were determined over the hot region (red shaded area, 5 s until 6 s). For our analysis of the interaction between Cisd2 and full-channel IP3R1 an alternative hot region of 1.5 s \u0026ndash; 2.5 s was selected to ensure proper evaluation of the thermophoretic shift in each of the evaluated conditions. All experiments were repeated at least 2 times for each condition using freshly thawed proteins with at least 2 technical replicates for each biological replicate.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunofluorescent staining and imaging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeLa cells were seeded on coverslips 48 h before fixation with 4 % paraformaldehyde (room temperature, 15 minutes), after which cells were washed 3 times with PBS. Cells were permeabilized with 0.1 % Triton-X-100 in PBS (room temperature, 10 minutes) and were washed 3 times with PBS. Permeabilized cells were blocked with 4 % bovine serum albumin and 0.1 % Triton-X-100 in PBS (room temperature, 1 h). Cells were stained with 1:100 mouse monoclonal anti-Cisd2 in blocking solution (4 \u0026deg;C, overnight). The next day, cells were washed 2 times with PBS and stained with Alexa 488 linked goat anti-mouse (1:1000 in blocking solution) (room temperature, 2 h). Cells were washed 3 times and kept in PBS for imaging on a fluorescence confocal microscope (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4). Excitation: 488 nm, emission: 530 nm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle cell Fura-2 Ca\u003csup\u003e2+\u003c/sup\u003e imaging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor Fura-2 measurements, a Nikon TI2-E inverted microscope equipped with a 20 \u0026times; 0.5 NA Plan Fluor DIC N2 air objective and a pco.edge 4.2bi sCMOS camera was used. Fura-2 was alternatingly excited at a 2 s interval using a CoolLED pE-300 ultra/pE-340 lamp set at 340 nm and 380 nm (CoolLED, Andover, UK), and using a dichroic mirror FF02-409/LP-25 (Semrock, New York, USA) and a bandpass emission filter 515/30 (Semrock, Rochester, USA). mCherry was visualized using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 550 nm, and using the cubical excitation filterset FF01\u0026ndash;378/474/554/635 and a bandpass emission filter 595/31 (Semrock, Rochester, USA). All cells were seeded in 4 chamber slides with coverslips (IBL, Gerasdorf bei Wien, Austria). HeLa CTRL and Cisd2 KO cells, were loaded for 30 min, at room temperature, with 1 \u0026mu;M FURA-2-AM (AnaSpec, Fremont, CA, USA) in modified Krebs buffer (135 mM NaCl, 6.2 mM KCl, 1.2 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 12 mM HEPES, pH 7.3, 11.5 mM glucose and 1.5 mM CaCl\u003csub\u003e2\u003c/sub\u003e). Cells were washed once with modified Krebs and the Fura-2-AM was allowed to de-esterify during 30 min at room temperature. After measuring for about 30 s, 3 mM ethylene glycol-bis(\u0026beta;-aminoethyl ether)-N,N,N\u0026prime;,N\u0026prime;-tetraacetic acid (EGTA) in Krebs without Ca\u003csup\u003e2+\u003c/sup\u003e was added to chelate extracellular Ca\u003csup\u003e2+\u003c/sup\u003e. Basal FURA-2 ratios were recorded for 30 s. a Nikon TI2-E inverted microscope equipped with a 20 \u0026times; 0.5 NA Plan Fluor DIC N2 air objective and a pco.edge 4.2bi sCMOS camera was used. FURA-2 was alternatingly excited at a 2 s interval using a CoolLED pE-300 ultra/pE-340 lamp set at 340 nm and 380 nm (CoolLED, Andover, UK), and using a dichroic mirror FF02-409/LP-25 (Semrock, New York, USA) and a bandpass emission filter 515/30 (Semrock, Rochester, USA). mCherry was visualized using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 550 nm, and using the cubical excitation filterset FF01\u0026ndash;378/474/554/635 and a bandpass emission filter 595/31 (Semrock, Rochester, USA). All cells were seeded in 4 chamber slides with coverslips (IBL, Gerasdorf bei Wien, Austria). Data was quantified using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/FURA\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCa\u003csup\u003e2+\u003c/sup\u003e imaging by camera-based multiplate imaging on cell populations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeLa CTRL and Cisd2 KO cells were seeded on black 96 well plates 48 h before acquisition. Cells were loaded with 3 \u0026micro;M Cal520 + 0.04 % pluronic acid in modified Krebs and incubated for 30 minutes at room temperature. Cells were washed twice and left for de-esterification for 30 minutes at room temperature. Basal fluorescence was measured for 30 seconds, then EGTA was added. Thapsigargin/ionomycin was added at 90 seconds, and acquisition was continued for 10 min at a one second acquisition interval. CTRL and Cisd2 KO iPSCs were seeded 72h before acquisition on Geltrex-coated black 96 well plates. Cells were loaded with 3 \u0026micro;M Cal520 + 0.04 % pluronic acid in E8 Flex medium and incubated for 30 minutes at 37 \u0026deg;C 5% CO\u003csub\u003e2\u003c/sub\u003e. Cells were washed twice and left for de-esterification for 30 minutes. Basal fluorescence was measured for 30 seconds, then various ATP concentrations were added. Cortical neurons were seeded 52-59 days before acquisition. Cells were loaded with 3 \u0026micro;M Cal520 + 0.04 % pluronic acid in neuronal maintenance medium and incubated for 30 minutes at 37 \u0026deg;C 5% CO\u003csub\u003e2\u003c/sub\u003e. Cells were washed twice and left for de-esterification for 30 minutes. Basal fluorescence was measured for 30 seconds, then EGTA was added. Various concentrations of glutamate/glycine were added at 120 seconds, followed by 2.5 \u0026micro;M ionomycin at 300 seconds. Fluorescence was acquired every second. Plates were imaged using an FDSS/\u0026micro;Cell kinetic plate imager C13299 (Hamamatsu). Data was further processed using a custom Python script: https://github.com/jensloncke/LMCS-python-scripts/tree/main/FDSS%20single%20response\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMultiplexed single cell cytosolic and mitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e imaging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor these measurements, a Nikon TI2-E inverted microscope equipped with a 40 \u0026times; 1.3 NA Plan Fluor DIC H N2 Oil objective and a pco.edge 4.2bi sCMOS camera was used. Fluo-4 was excited at a 2 s interval using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 488 nm, the cubical excitation filterset FF01\u0026ndash;378/474/554/635, and using a bandpass emission filter 515/30 (Semrock, Rochester, USA). R-mtCEPIA3 was visualized using a CoolLed pR-4000 lamp (CoolLED, Andover, UK) at 550 nm, and using the cubical excitation filterset FF01\u0026ndash;378/474/554/635 and a bandpass emission filter 595/31 (Semrock, Rochester, USA). HeLa cells were seeded in 4 chamber slides with coverslips and transfected with 200 ng of R-mtCEPIA3 (\u003cem\u003e35\u003c/em\u003e). Two days after transfection, cells were loaded for 30 min, at room temperature, with 1 \u0026mu;M Fluo-4-AM (F14201) in modified Krebs buffer. Cells were washed once with modified Krebs and the Fluo-4-AM was left was allowed to de-esterify during 30 min at room temperature. Cells were washed twice after de-esterification. After measuring for about 30 s, EGTA was added in Krebs without Ca\u003csup\u003e2+\u003c/sup\u003e to chelate extracellular Ca\u003csup\u003e2+\u003c/sup\u003e. After 90 s, responses of mtCEPIA transfected cells to 5 \u0026micro;M ATP were recorded. 180 s after 5 \u0026micro;M ATP addition, cells were stimulated supramaximally with 100 \u0026micro;M ATP. Data was quantified using a custom ImageJ script, which can be found on: https://github.com/jensloncke/ImageJ_macros/tree/master/Fluo-4%20transfected%20cells. Relevant response parameters, such as area under curve, were extracted using a custom Python script: https://github.com/jensloncke/LMCS-python-scripts/tree/main/Quantify%20agonist%20response. Cells that detached, or did not respond to either ATP concentration were excluded from analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBiochemical purification of HeLa MAM fractions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSubcellular fractions were purified from CTRL and Cisd2-KO HeLa cells as previously described (\u003cem\u003e24\u003c/em\u003e). Concisely, HeLa cells were homogenized and after several centrifugation steps, a crude mitochondrial fraction is obtained separately from the ER and cytosolic fractions. After ultracentrifugation of the crude mitochondrial fraction in a Percoll\u0026reg; (Santa Cruz Biotechnology Inc., Dallas, USA) gradient, MAM and pure mitochondrial fractions were obtained.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImaging of HeLa ER-mitochondrial contact sites using Mt-ER short SPLICS probe\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eER-mitochondrial contact sites were specifically detected using the SPLICS Mt-ER Short P2A probe. SPLICS Mt-ER Short P2A was a gift from Marisa Brini \u0026amp; Tito Cal\u0026igrave; (Addgene plasmid # 164108; http://n2t.net/addgene:164108; RRID:Addgene_164108) (\u003cem\u003e36\u003c/em\u003e) on a fluorescence confocal microscope (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4). Excitation: 488 nm, emission: 530 nm. All cells were seeded in 4 chamber slides with coverslips. HeLa cells were transfected with 500 ng SPLICS Two days after transfection, cells were fixed with 4 % paraformaldehyde and imaged immediately. 3D Z-stacks were acquired SPLICS fluorescence and unfiltered light was captured in another channel to serve as a brightfield image replacement. Data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/splics%203D.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImaging of HeLa and cortical neuron ER-mitochondrial contact sites using MAMtracker Green\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeLa cells were seeded 24 h before transfection in 4 chamber slides with coverslip. 48 h before acquisition, cells were transfected with 200 ng MAMtracker Green and 200 ng pCMV24-Cisd2-p2a-mCherry or pCMV24-p2a-mCherry. Two days after transfection, cells were stained with Hoechst and imaged in modified Krebs buffer. Cortical neurons were seeded 50 days before transfection. Cortical neurons were transfected with pCMV24-Cisd2-p2a-mCherry or pCMV24-p2a-mCherry. 2D images were simultaneously acquired of Hoechst stains, MAMtracker Green signal, mCherry signal and brightfield images. The p-MAMtracker Green vector was a kind gift from Koji Yamanaka (Nagoya University, Japan) (\u003cem\u003e37\u003c/em\u003e). A Nikon TI2-E inverted microscope equipped with a 40 \u0026times; 1.3 NA Plan Fluor DIC H N2 Oil objective and a pco.edge 4.2bi sCMOS camera was used. MAMtracker Green was excited at 470 nm, using a CoolLed pR-4000 lamp and using the cubical excitation filterset FF01\u0026ndash;378/474/554/635 and bandpass emission filters of 515/30. mCherry was excited at 550 nm, using a CoolLed pR-4000 lamp, and using the cubical excitation filterset FF01\u0026ndash;378/474/554/635 and a bandpass emission filter 595/31. Hoechst was excited at 350 nm using a CoolLed pR-4000 lamp, and using the cubical excitation filterset FF01-378/474/554/635. Data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/MAMtracker\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAssessment of mitochondrial morphology\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeLa cells were seeded 48 h before acquisition on 4 chamber slides with coverslips. Cells were stained with 100 nM Mitotracker Green (M7514) and 1:1000 CelMask\u0026trade; Orange (C10045). Fluorescence was acquired on a fluorescence confocal microscope (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4). Mitotracker Green: Excitation: 488 nm, emission: 530 nm. CellMask\u0026trade; Orange: excitation: 543 nm emission: 633 nm. 3D Z-stacks were acquired and data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/PM%20and%20mito%203D and the Mitochondria analyzer plugin (\u003cem\u003e49\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRelative quantification of inner mitochondrial membrane potential using JC-1\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeLa cells were seeded 48 h before acquisition on 4 chamber slides with coverslips. Cells were stained with 2 \u0026micro;M JC-1 (Invitrogen, T3168), diluted in modified Krebs. After incubation, the cells were washed twice with the modified Krebs solution for live-cell imaging. Images were acquired using a (Zeiss LSM510) using a 63X oil objective (numerical aperture 1.4), an 488 nm Argon laser equipped with a BP 505-530 filter, and a 543 nm laser equipped with an LP 560 filter. Data was analyzed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/Hoechst%20and%20JC-1\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcquisition of mitochondrial oxygen consumption rate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe OCR and ECAR were measured using an XFp extracellular analyzer (Seahorse Bioscience, North Billerica, MA, USA). HeLa cells were plated on XFp Cell culture miniplates (Seahorse Bioscience, Cat. 103022-100) at 12 000 cells per well. Cortical neurons were plated 45 days before acquisition. The next day, the medium was changed to XF Base Medium (Seahorse Bioscience, 103334-100) containing 10 mM of glucose, and the cells were incubated at 37 \u0026deg;C in a low-CO\u003csub\u003e2\u003c/sub\u003e incubator for 1 h. The cells were sequentially exposed to oligomycin (1 \u0026micro;M), FCCP (0.5 \u0026micro;M), and a mixture of rotenone (0.5 \u0026micro;M) and antimycin A (0.5 \u0026micro;M). Data was analyzed using R\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalysis of autophagic flux via western blotting\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeLa cells CTRL and Cisd2 KO cells were seeded on 6 well plates and treated 48 h after seeding. CTRL and Cisd2 cortical neurons were seeded on Matrigel-coated 6 well plates 52-59 days before treatment. 2 h before treatment, medium was refreshed. Cells were treated with 100 nM bafilomycin A1 or DMSO. Cells were harvested by scraping in ice cold PBS lysed and immunoblotted. Densitometric quantification was performed using the gel analyzer plugin in ImageJ.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle-cell quantification of autophagic flux through the GFP(-RFP)-LC3 probe\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCTRL and Cisd2 KO HeLa cells were seeded on 4 chamber slides with coverslips. 24 h afterwards, cells were transfected with 200 ng GFP-LC3. Medium was changed 24 h post-transfection. Cortical neurons were seeded on Matrigel-coated 4 chamber slides with coverslips 52-59 days before treatment. Cortical neurons were transfected with 500 ng p4-neo-mRFP-GFP-LC3 probe. 48 h after transfection, medium was changed. 2 h after, cells were treated with 100 nM bafilomycin A1 or DMSO. 15 min before acquisition, cells were stained with Hoechst 33342. Before imaging, cells were washed once with modified Krebs. Fluorescence was acquired using the Nikon TI-2E setup described earlier. Autophagic puncti were quantified using the Cell Counter plugin in ImageJ.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStaurosporine-induced apoptosis experiments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeLa CTRL or Cisd2 KO cells were seeded 48 h before transfection, cortical neurons were seeded 52-59 days before treatment and maintained in neuronal maintenance medium. Cells were treated with various staurosporine concentrations or with DMSO for 6 h. After treatment, cells were harvested by scraping on ice, and lysed with CHAPS buffer. Samples were analyzed via immunoblotting. Densitometric quantification was performed using the gel analyzer plugin in ImageJ.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMitochondrial Ca\u003csup\u003e2+\u003c/sup\u003e imaging in cortical neurons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e52-59 days before acquisition, cortical neurons were seeded on Matrigel-coated 4 chamber slides with coverslips. 72 h before imaging, cortical neurons were transfected with 600 ng G-mtCEPIA2 (\u003cem\u003e40\u003c/em\u003e). Before imaging, cortical neurons were washed once with modified Krebs. During measurements, G-mtCEPIA2 was recorded at a 2 s interval. 100 mM glutamate/10 mM glycine was added after 60 s of starting acquisitions. Fluorescence was acquired using the Nikon TI-2E setup described earlier. Data was processed using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/Cortical%20neurons%20mtCEPIA . Relevant response parameters, such as area under curve, were extracted using a custom Python script: https://github.com/jensloncke/LMCS-python-scripts/tree/main/Quantify%20agonist%20response.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImaging of cortical neurons ER-mitochondrial contact sites using Mt-ER short SPLICS probe\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e52-59 days before acquisition, cortical neurons were seeded on Matrigel-coated 4 chamber slides with coverslips. 72 h before imaging, cortical neurons were transfected with 200 ng SPLICS. Cells were stained with Hoechst 33342 for 15 minutes. Before imaging, cortical neurons were washed once with modified Krebs. Fluorescence was acquired using the Nikon TI-2E setup described earlier. Data was quantified using a custom ImageJ macro: https://github.com/jensloncke/ImageJ_macros/tree/master/splics%202D\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle cell quantification of relative cellular ATP:ADP ratios\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eshCTRL, shCisd2 and shCisd2 + Cisd2 were cultured, transfected and acquired as described in (\u003cem\u003e30\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eQuantification of immunoblots was performed using the FIJI software (\u003cem\u003e50\u003c/em\u003e). MST data was analyzed as described in (\u003cem\u003e24\u003c/em\u003e). All other plots were created using the ggplot2 package in the R programming language. Normality of residuals of fit was tested via Shapiro-Wilk normality testing and assessing equality of variances with the Levene test. Where appropriate, data was square-root or log10 transformed. In the event of both of the assumptions being fulfilled, a two-way ANOVA was performed, followed by fdr-corrected pairwise t-tests. In case of unequal variances within conditions, ANOVA was corrected for heteroskedasticity. When assumption of normal distribution was not fulfilled, a non-parametric Kruskal-Wallis was performed, followed by pairwise Mann-Whitney U tests. Microscopy data was acquired using the NIS elements software and numerically extracted in FIJI using custom macros. Data was further processed in Python using the numpy, pandas and plotly packages.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflict of interest\u003c/h2\u003e \u003cp\u003eThe authors declare no conflict of interest.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eResearch supported was by Eye Hope Foundation and Research Foundation\u0026mdash;Flanders (FWO) to G.B. (G081821N), the KU Leuven Research Council (C14/19/099 and AKUL/19/34) and the Central European Leuven Strategic Alliance (CELSA/23/031 and CELSA/23/032). G.B., J.B.P., I.S. and A.K. are partners of the FWO Scientific Research Network CaSign (W0.014.22N). I.D.R. is supported by a PhD fellowship from the FWO (1131322N|1131324N). T.V was supported by a Post-doctoral fellowship from the FWO (12ZG121N). We thank Anja Florizoone for the excellent technical support.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eZ.-Q. Shen, Y.-L. Huang, Y.-C. Teng, T.-W. Wang, C.-H. Kao, C.-H. Yeh, T.-F. Tsai, CISD2 maintains cellular homeostasis. \u003cem\u003eBiochimica et Biophysica Acta (BBA) - Molecular Cell Research\u003c/em\u003e \u003cstrong\u003e1868\u003c/strong\u003e, 118954 (2021).\u003c/li\u003e\n\u003cli\u003eS. Amr, C. Heisey, M. Zhang, X.-J. Xia, K. H. Shows, K. Ajlouni, A. Pandya, L. S. Satin, H. El-Shanti, R. Shiang, A Homozygous Mutation in a Novel Zinc-Finger Protein, ERIS, Is Responsible for Wolfram Syndrome 2. \u003cem\u003eThe American Journal of Human Genetics\u003c/em\u003e \u003cstrong\u003e81\u003c/strong\u003e, 673-683 (2007).\u003c/li\u003e\n\u003cli\u003eJ. Loncke, T. Vervliet, J. B. Parys, A. Kaasik, G. Bultynck, Uniting the divergent Wolfram syndrome\u0026amp;#x2013;linked proteins WFS1 and CISD2 as modulators of Ca\u0026lt;sup\u0026gt;2+\u0026lt;/sup\u0026gt; signaling. \u003cem\u003eScience Signaling\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, eabc6165 (2021).\u003c/li\u003e\n\u003cli\u003eY.-F. Chen, C.-H. Kao, Y.-T. Chen, C.-H. Wang, C.-Y. Wu, C.-Y. Tsai, F.-C. Liu, C.-W. Yang, Y.-H. Wei, M.-T. Hsu, S.-F. Tsai, T.-F. Tsai, Cisd2 deficiency drives premature aging and causes mitochondria-mediated defects in mice. \u003cem\u003eGenes \u0026amp; Development\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 1183-1194 (2009).\u003c/li\u003e\n\u003cli\u003eC.-H. Yeh, Z.-Q. Shen, S.-Y. Hsiung, P.-C. Wu, Y.-C. Teng, Y.-J. Chou, S.-W. Fang, C.-F. Chen, Y.-T. Yan, L.-S. Kao, C.-H. Kao, T.-F. Tsai, Cisd2 is essential to delaying cardiac aging and to maintaining heart functions. \u003cem\u003ePLOS Biology\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, e3000508 (2019).\u003c/li\u003e\n\u003cli\u003eY.-C. Teng, J.-Y. Wang, Y.-H. Chi, T.-F. Tsai, Exercise and the Cisd2 Prolongevity Gene: Two Promising Strategies to Delay the Aging of Skeletal Muscle. \u003cem\u003eInternational Journal of Molecular Sciences\u003c/em\u003e. 2020.\u003c/li\u003e\n\u003cli\u003eC.-H. Huang, Y.-L. Huang, Z.-Q. Shen, C.-H. Lin, T.-F. Tsai, Cisd2 Preserves the Youthful Pattern of the Liver Proteome during Natural Aging of Mice. \u003cem\u003eBiomedicines\u003c/em\u003e. 2021.\u003c/li\u003e\n\u003cli\u003eY.-L. Huang, Z.-Q. Shen, C.-H. Huang, C.-H. Lin, T.-F. Tsai, Cisd2 slows down liver aging and attenuates age-related metabolic dysfunction in male mice. \u003cem\u003eAging Cell\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, e13523 (2021).\u003c/li\u003e\n\u003cli\u003eY.-L. Huang, Z.-Q. Shen, C.-H. Huang, Y.-C. Teng, C.-H. Lin, T.-F. Tsai, Cisd2 Protects the Liver from Oxidative Stress and Ameliorates Western Diet-Induced Nonalcoholic Fatty Liver Disease. \u003cem\u003eAntioxidants\u003c/em\u003e. 2021.\u003c/li\u003e\n\u003cli\u003eZ.-Q. Shen, Y.-F. Chen, J.-R. Chen, Y.-S. Jou, P.-C. Wu, C.-H. Kao, C.-H. Wang, Y.-L. Huang, C.-F. Chen, T.-S. Huang, Y.-C. Shyu, S.-F. Tsai, L.-S. Kao, T.-F. Tsai, CISD2 Haploinsufficiency Disrupts Calcium Homeostasis, Causes Nonalcoholic Fatty Liver Disease, and Promotes Hepatocellular Carcinoma. \u003cem\u003eCell Reports\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 2198-2211 (2017).\u003c/li\u003e\n\u003cli\u003eC.-C. Sun, S.-Y. Lee, C.-H. Kao, L.-H. Chen, Z.-Q. Shen, C.-H. Lai, T.-Y. Tzeng, J.-H. S. Pang, W.-T. Chiu, T.-F. Tsai, Cisd2 plays an essential role in corneal epithelial regeneration. \u003cem\u003eEBioMedicine\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, 103654 (2021).\u003c/li\u003e\n\u003cli\u003eO. Karmi, Y.-S. Sohn, S. I. Zandalinas, L. Rowland, S. D. King, R. Nechushtai, R. Mittler, Disrupting CISD2 function in cancer cells primarily impacts mitochondrial labile iron levels and triggers TXNIP expression. \u003cem\u003eFree Radical Biology and Medicine\u003c/em\u003e \u003cstrong\u003e176\u003c/strong\u003e, 92-104 (2021).\u003c/li\u003e\n\u003cli\u003eS. H. Holt, M. Darash-Yahana, Y. S. Sohn, L. Song, O. Karmi, S. Tamir, D. Michaeli, Y. Luo, M. L. Paddock, P. A. Jennings, J. N. Onuchic, R. K. Azad, E. Pikarsky, I. Z. Cabantchik, R. Nechushtai, R. Mittler, Activation of apoptosis in NAF-1-deficient human epithelial breast cancer cells. \u003cem\u003eJournal of Cell Science\u003c/em\u003e \u003cstrong\u003e129\u003c/strong\u003e, 155-165 (2016).\u003c/li\u003e\n\u003cli\u003eZ.-Q. Shen, H. Yi-Long, T.-F. and Tsai, Cisd2 haploinsufficiency: A driving force for hepatocellular carcinoma. \u003cem\u003eMolecular \u0026amp; Cellular Oncology\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e1441627 (2018).\u003c/li\u003e\n\u003cli\u003eM. D. Bootman, G. Bultynck, Fundamentals of Cellular Calcium Signaling: A Primer. \u003cem\u003eCold Spring Harbor Perspectives in Biology\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, a038802 (2020).\u003c/li\u003e\n\u003cli\u003eB. Cartes-Saavedra, A. Ghosh, G. Hajn\u0026oacute;czky, The roles of mitochondria in global and local intracellular calcium signalling. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e, (2025).\u003c/li\u003e\n\u003cli\u003eC. Angebault, J. Fauconnier, S. Patergnani, J. Rieusset, A. Danese, C. A. Affortit, J. Jagodzinska, C. M\u0026eacute;gy, M. Quiles, C. Cazevieille, J. Korchagina, D. Bonnet-Wersinger, D. Milea, C. Hamel, P. Pinton, M. Thiry, A. Lacampagne, B. Delprat, C. Delettre, ER-mitochondria cross-talk is regulated by the Ca2+ sensor NCS1 and is impaired in Wolfram syndrome. \u003cem\u003eScience Signaling\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, eaaq1380 (2018).\u003c/li\u003e\n\u003cli\u003eN. C. Chang, M. Nguyen, M. Germain, G. C. Shore, Antagonism of Beclin 1‐dependent autophagy by BCL‐2 at the endoplasmic reticulum requires NAF‐1. \u003cem\u003eThe EMBO Journal\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 606-618-618 (2010).\u003c/li\u003e\n\u003cli\u003eL. D. Nguyen, E. T. Petri, L. K. Huynh, B. E. Ehrlich, Characterization of NCS1\u0026ndash;InsP3R1 interaction and its functional significance. \u003cem\u003eJournal of Biological Chemistry\u003c/em\u003e \u003cstrong\u003e294\u003c/strong\u003e, 18923-18933 (2019).\u003c/li\u003e\n\u003cli\u003eG. Monaco, M. Beckers, H. Ivanova, L. Missiaen, J. B. Parys, H. De Smedt, G. Bultynck, Profiling of the Bcl-2/Bcl-XL-binding sites on type 1 IP3 receptor. \u003cem\u003eBiochemical and Biophysical Research Communications\u003c/em\u003e \u003cstrong\u003e428\u003c/strong\u003e, 31-35 (2012).\u003c/li\u003e\n\u003cli\u003eH. Ivanova, L. E. Wagner, A. Tanimura, E. Vandermarliere, T. Luyten, K. Welkenhuyzen, K. J. Alzayady, L. Wang, K. Hamada, K. Mikoshiba, H. De Smedt, L. Martens, D. I. Yule, J. B. Parys, G. Bultynck, Bcl-2 and IP3 compete for the ligand-binding domain of IP3Rs modulating Ca2+ signaling output. \u003cem\u003eCellular and Molecular Life Sciences\u003c/em\u003e \u003cstrong\u003e76\u003c/strong\u003e, 3843-3859 (2019).\u003c/li\u003e\n\u003cli\u003eN. C. Chang, M. Nguyen, M. Germain, G. C. Shore, Antagonism of Beclin 1‐dependent autophagy by BCL‐2 at the endoplasmic reticulum requires NAF‐1. \u003cem\u003eThe EMBO Journal\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 606-618 (2010).\u003c/li\u003e\n\u003cli\u003eS. Tamir, S. Rotem-Bamberger, C. Katz, F. Morcos, K. L. Hailey, J. A. Zuris, C. Wang, A. R. Conlan, C. H. Lipper, M. L. Paddock, R. Mittler, J. N. Onuchic, P. A. Jennings, A. Friedler, R. Nechushtai, Integrated strategy reveals the protein interface between cancer targets Bcl-2 and NAF-1. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 5177-5182 (2014).\u003c/li\u003e\n\u003cli\u003eJ. Loncke, I. de Ridder, J. Kale, L. Wagner, A. Kaasik, J. B. Parys, M. Kerkhofs, D. W. Andrews, D. Yule, T. Vervliet, G. Bultynck, CISD2 counteracts the inhibition of ER-mitochondrial calcium transfer by anti-apoptotic BCL-2. \u003cem\u003eBiochimica et Biophysica Acta (BBA) - Molecular Cell Research\u003c/em\u003e \u003cstrong\u003e1872\u003c/strong\u003e, 119857 (2025).\u003c/li\u003e\n\u003cli\u003eB. Delprat, T. Maurice, C. Delettre, Wolfram syndrome: MAMs\u0026rsquo; connection? \u003cem\u003eCell Death \u0026amp; Disease\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 364 (2018).\u003c/li\u003e\n\u003cli\u003eC. Kwak, S. Shin, J.-S. Park, M. Jung, T. T. M. Nhung, M.-G. Kang, C. Lee, T.-H. Kwon, S. K. Park, J. Y. Mun, J.-S. Kim, H.-W. Rhee, Contact-ID, a tool for profiling organelle contact sites, reveals regulatory proteins of mitochondrial-associated membrane formation. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 12109-12120 (2020).\u003c/li\u003e\n\u003cli\u003eS. E. Wiley, A. Y. Andreyev, A. S. Divakaruni, R. Karisch, G. Perkins, E. A. Wall, P. van der Geer, Y. F. Chen, T. F. Tsai, M. I. Simon, B. G. Neel, J. E. Dixon, A. N. Murphy, Wolfram Syndrome protein, Miner1, regulates sulphydryl redox status, the unfolded protein response, and Ca\u0026lt;sup\u0026gt;2+\u0026lt;/sup\u0026gt; homeostasis. \u003cem\u003eEMBO Molecular Medicine\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 904-918-918 (2013).\u003c/li\u003e\n\u003cli\u003eC.-H. Wang, Y.-F. Chen, C.-Y. Wu, P.-C. Wu, Y.-L. Huang, C.-H. Kao, C.-H. Lin, L.-S. Kao, T.-F. Tsai, Y.-H. Wei, Cisd2 modulates the differentiation and functioning of adipocytes by regulating intracellular Ca2+ homeostasis. \u003cem\u003eHuman Molecular Genetics\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 4770-4785 (2014).\u003c/li\u003e\n\u003cli\u003eA. Zhang, C. D. Williamson, D. S. Wong, M. D. Bullough, K. J. Brown, Y. Hathout, A. M. Colberg-Poley, Quantitative Proteomic Analyses of Human Cytomegalovirus-Induced Restructuring of Endoplasmic Reticulum-Mitochondrial Contacts at Late Times of Infection*. \u003cem\u003eMolecular \u0026amp; Cellular Proteomics\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, M111.009936 (2011).\u003c/li\u003e\n\u003cli\u003eM. Liiv, A. Vaarmann, D. Safiulina, V. Choubey, R. Gupta, M. Kuum, L. Janickova, Z. Hodurova, M. Cagalinec, A. Zeb, M. A. Hickey, Y.-L. Huang, N. Gogichaishvili, M. Mandel, M. Plaas, E. Vasar, J. Loncke, T. Vervliet, T.-F. Tsai, G. Bultynck, V. Veksler, A. Kaasik, ER calcium depletion as a key driver for impaired ER-to-mitochondria calcium transfer and mitochondrial dysfunction in Wolfram syndrome. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 6143 (2024).\u003c/li\u003e\n\u003cli\u003eL. Crouzier, A. Danese, Y. Yasui, E. M. Richard, J.-C. Li\u0026eacute;vens, S. Patergnani, S. Couly, C. Diez, M. Denus, N. Cubedo, M. Rossel, M. Thiry, T.-P. Su, P. Pinton, T. Maurice, B. Delprat, Activation of the sigma-1 receptor chaperone alleviates symptoms of Wolfram syndrome in preclinical models. \u003cem\u003eScience Translational Medicine\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, eabh3763 (2022).\u003c/li\u003e\n\u003cli\u003eL. D. Nguyen, T. T. Fischer, D. Abreu, A. Arroyo, F. Urano, B. E. Ehrlich, Calpain inhibitor and ibudilast rescue \u0026beta; cell functions in a cellular model of Wolfram syndrome. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 17389-17398 (2020).\u003c/li\u003e\n\u003cli\u003eC. Rouzier, D. Moore, C. Delorme, S. Lacas-Gervais, S. Ait-El-Mkadem, K. Fragaki, F. Burt\u0026eacute;, V. Serre, S. Bannwarth, A. Chaussenot, M. Catala, P. Yu-Wai-Man, V. Paquis-Flucklinger, A novel CISD2 mutation associated with a classical Wolfram syndrome phenotype alters Ca2+ homeostasis and ER-mitochondria interactions. \u003cem\u003eHuman Molecular Genetics\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 1599-1611 (2017).\u003c/li\u003e\n\u003cli\u003eM. Katona, \u0026Aacute;. Bart\u0026oacute;k, Z. Nichtova, G. Csord\u0026aacute;s, E. Berezhnaya, D. Weaver, A. Ghosh, P. V\u0026aacute;rnai, D. I. Yule, G. Hajn\u0026oacute;czky, Capture at the ER-mitochondrial contacts licenses IP3 receptors to stimulate local Ca2+ transfer and oxidative metabolism. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 6779 (2022).\u003c/li\u003e\n\u003cli\u003eK. Kanemaru, J. Suzuki, I. Taiko, M. Iino, Red fluorescent CEPIA indicators for visualization of Ca2+ dynamics in mitochondria. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 2835 (2020).\u003c/li\u003e\n\u003cli\u003eF. Vallese, C. Catoni, D. Cieri, L. Barazzuol, O. Ramirez, V. Calore, M. Bonora, F. Giamogante, P. Pinton, M. Brini, T. Cal\u0026igrave;, An expanded palette of improved SPLICS reporters detects multiple organelle contacts in vitro and in vivo. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 6069 (2020).\u003c/li\u003e\n\u003cli\u003eS. Sakai, S. Watanabe, O. Komine, A. Sobue, K. Yamanaka, Novel reporters of mitochondria-associated membranes (MAM), MAMtrackers, demonstrate MAM disruption as a common pathological feature in amyotrophic lateral sclerosis. \u003cem\u003eThe FASEB Journal\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, e21688 (2021).\u003c/li\u003e\n\u003cli\u003eG. Mari\u0026ntilde;o, M. Niso-Santano, E. H. Baehrecke, G. Kroemer, Self-consumption: the interplay of autophagy and apoptosis. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 81-94 (2014).\u003c/li\u003e\n\u003cli\u003eP. A. Egorova, K. S. Marinina, I. B. Bezprozvanny, Chronic suppression of STIM1-mediated calcium signaling in Purkinje cells rescues the cerebellar pathology in spinocerebellar ataxia type 2. \u003cem\u003eBiochimica et Biophysica Acta (BBA) - Molecular Cell Research\u003c/em\u003e \u003cstrong\u003e1870\u003c/strong\u003e, 119466 (2023).\u003c/li\u003e\n\u003cli\u003eJ. Suzuki, K. Kanemaru, K. Ishii, M. Ohkura, Y. Okubo, M. Iino, Imaging intraorganellar Ca2+ at subcellular resolution using CEPIA. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 4153 (2014).\u003c/li\u003e\n\u003cli\u003eM. Tantama, J. R. Mart\u0026iacute;nez-Fran\u0026ccedil;ois, R. Mongeon, G. Yellen, Imaging energy status in live cells with a fluorescent biosensor of the intracellular ATP-to-ADP ratio. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 2550 (2013).\u003c/li\u003e\n\u003cli\u003eN. C. Chang, M. Nguyen, J. Bourdon, P.-A. Risse, J. Martin, G. Danialou, R. Rizzuto, B. J. Petrof, G. C. Shore, Bcl-2-associated autophagy regulator Naf-1 required for maintenance of skeletal muscle. \u003cem\u003eHuman Molecular Genetics\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 2277-2287 (2012).\u003c/li\u003e\n\u003cli\u003eX. Du, R. Xiao, F. Xiao, Y. Chen, F. Hua, S. Yu, G. Xu, NAF-1 antagonizes starvation-induced autophagy through AMPK signaling pathway in cardiomyocytes. \u003cem\u003eCell Biology International\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 816-823 (2015).\u003c/li\u003e\n\u003cli\u003eJ. B. Parys, H. De Smedt, L. Missiaen, M. D. Bootman, I. Sienaert, R. Casteels, Rat basophiliv leukemia cells as model system for inositol 1,4,5-trisphosphate receptor IV, a receptor of the type II family: functional comparison and immunological detection. \u003cem\u003eCell Calcium\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 239-249 (1995).\u003c/li\u003e\n\u003cli\u003eT. Vervloessem, B. K. Sasi, E. Xerxa, S. Karamanou, J. Kale, R. M. La Rovere, S. Chakraborty, F. Sneyers, M. Vogler, A. Economou, L. Laurenti, D. W. Andrews, D. G. Efremov, G. Bultynck, BDA-366, a putative Bcl-2 BH4 domain antagonist, induces apoptosis independently of Bcl-2 in a variety of cancer cell models. \u003cem\u003eCell Death \u0026amp; Disease\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 769 (2020).\u003c/li\u003e\n\u003cli\u003eT. Vervliet, I. Lemmens, K. Welkenhuyzen, J. Tavernier, J. B. Parys, G. Bultynck, Regulation of the ryanodine receptor by anti-apoptotic Bcl-2 is independent of its BH3-domain-binding properties. \u003cem\u003eBiochemical and Biophysical Research Communications\u003c/em\u003e \u003cstrong\u003e463\u003c/strong\u003e, 174-179 (2015).\u003c/li\u003e\n\u003cli\u003eF. A. Ran, P. D. Hsu, J. Wright, V. Agarwala, D. A. Scott, F. Zhang, Genome engineering using the CRISPR-Cas9 system. \u003cem\u003eNature Protocols\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 2281-2308 (2013).\u003c/li\u003e\n\u003cli\u003eJ. L. T. Vervliet, M. Sever, K. Ahuja, C. Van den Haute, C. Verfaillie, T. Tomasic, G. Bultynck, Inactive ryanodine receptors sustain lysosomal availability for autophagy by promoting ER-lysosomal contact site formation. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003eUnder Review\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eA. Chaudhry, R. Shi, D. S. Luciani, A pipeline for multidimensional confocal analysis of mitochondrial morphology, function, and dynamics in pancreatic \u0026beta;-cells. \u003cem\u003eAmerican Journal of Physiology-Endocrinology and Metabolism\u003c/em\u003e \u003cstrong\u003e318\u003c/strong\u003e, E87-E101 (2019).\u003c/li\u003e\n\u003cli\u003eJ. Schindelin, I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, J.-Y. Tinevez, D. J. White, V. Hartenstein, K. Eliceiri, P. Tomancak, A. Cardona, Fiji: an open-source platform for biological-image analysis. \u003cem\u003eNature Methods\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 676-682 (2012).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"826e6126-8cc0-4c62-8bab-7bc5d91f54cd","identifier":"10.13039/501100003130","name":"Fonds Wetenschappelijk Onderzoek","awardNumber":"G081821N","order_by":0}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"KU Leuven","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Cisd2, calcium signaling, Wolfram syndrome, MAMs, neurodegeneration","lastPublishedDoi":"10.21203/rs.3.rs-6298090/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6298090/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLoss of CISD2, an iron-sulfur cluster transfer protein, results in type 2 Wolfram syndrome (WFS2), a disorder associated with severe impacts on pancreatic beta cell and neuronal functions. CISD2 has been implicated in Ca2+ signaling but the molecular basis and cellular consequences remain poorly understood. In this work, we demonstrate that Cisd2 intersects with intracellular Ca2+ dynamics at different levels, including as an interactor of IP3Rs and as a protein contributing to ER-mitochondrial tethering. \u0026nbsp;As such, loss of CISD2 in HeLa cells results in reduced ER-mitochondrial Ca2+ transfer without majorly impact cytosolic Ca2+ signaling. In these cells, CISD2 deficiency promotes autophagic flux, yet has minimal impact mitochondrial function. However, studying the impact of CISD2 deficiency in iPSC-derived cortical neurons, relevant for WFS2, revealed a severe loss of glutamate-evoked Ca2+ responses in cytosol and mitochondria and loss of ER-mitochondrial contact. Correlating with the profound changes in cellular Ca2+ handling, mitochondrial function (oxygen consumption rate, ATP production, mitochondrial potential maintenance) was severely declined, while autophagic flux was increased. Overall, these deficiencies further impact the resilience of CISD2-deficient cortical neurons against cell stress as CISD2-KO neurons were highly susceptible to staurosporine, a cell death inducer. Overall, this work is one of the first to decipher the impact of CISD2 on ER-mitochondrial Ca2+ handling in disease-relevant cell models, thereby revealing a unique dependence of neurons on CISD2 for their mitochondrial health and cell stress resilience.\u003c/p\u003e","manuscriptTitle":"Cisd2 ensures adequate ER-mitochondrial coupling, thereby critically supporting mitochondrial function in neurons","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-26 02:59:21","doi":"10.21203/rs.3.rs-6298090/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4581de40-0af1-4c44-bbf3-d56cd0a4c694","owner":[],"postedDate":"March 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":46149105,"name":"Cell Survival and Cell Death"},{"id":46149106,"name":"Cell Communication and Signaling"},{"id":46149107,"name":"General Cell Biology \u0026 Physiology"},{"id":46149108,"name":"Stem Cell \u0026 Developmental Cell Biology"}],"tags":[],"updatedAt":"2025-03-26T02:59:21+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-26 02:59:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6298090","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6298090","identity":"rs-6298090","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}