INTRODUCTION
Mitochondria play a role in various essential cellular functions, including bioenergetics, metabolism, ion homeostasis, and apoptosis. Mitochondrial dysfunction has been linked to many common neurodegenerative disorders that are disabling and are often fatal. However, the cause of mitochondrial dysfunction remains largely undefined. Mitochondria interact physically and functionally with endoplasmic reticulum (ER) through mitochondria-associated membranes (MAMs) (
1–
3) to influence cellular physiology and viability (
4,
5). Accordingly, impaired MAM function is increasingly recognized as a contributor to neurodegeneration (
6,
7).
Loss of function of the ER protein WFS1 leads to a neurodegenerative disease called Wolfram syndrome that is associated with diabetes, optic atrophy, and deafness (
8). Wolfram syndrome was originally described as a mitochondriopathy due to a clinical phenotype that resembled other mitochondrial disorders (
9). However, this hypothesis has been challenged because WFS1 localized to the ER (
10). A relationship between cytosolic Ca
2+ disturbance, impaired mitochondrial dynamics, and delayed neuronal development in WFS1-deficient neurons has been suggested (
11). Here, we report the identification of a WFS1 binding partner, neuronal calcium sensor 1 (NCS1), a cytoplasmic Ca
2+ binding protein that is abundant in neurons. Our data suggest that NCS1 promoted ER-mitochondria contact and Ca
2+ transfer, subsequently controlling mitochondrial bioenergetics. This study shows that WFS1 and NCS1 may participate in the tethering of ER to mitochondria and the lack of this interaction may contribute to neurodegeneration.
DISCUSSION
Our study provides new insights into mechanisms underlying the mitochondrial dysfunction in WFS1-deficient cells involving NCS1, a WFS1 interacting protein. We demonstrated that WFS1 interacted with the Ca
2+ sensor NCS1 (
39) and that NCS1 protein abundance was substantially decreased in WFS1-deficient cells. Addition of NCS1 to planar lipid bilayers or overexpression of NCS1 in PC12 cells results in increased IP
3R channel activity (
30–
32), which should lead to increased Ca
2+ release from the ER. In fibroblasts from our Wolfram syndrome patients, a decrease in Ca
2+ flux from the ER was observed only under histamine application, whereas ER Ca
2+ load was unchanged, suggesting a role of NCS1 in regulating IP
3R activity. In addition, there was a concomitant decrease in the [Ca
2+]
m uptake. We further showed that ER-mitochondria structural interactions were reduced in patient cells, as evidenced by reduced interactions of two pairs of proteins at the MAM interface (VDAC1-IP
3R1 and GRP75-IP
3R1). This reduced tethering likely contributed to reduced Ca
2+ transfer from ER to mitochondria in these patient cells, as has been previously described in MFN2-deficient cells that show a decrease in MAM tethering (
28,
40). Dysregulation of Ca
2+ homeostasis and impaired MAM formation have been previously described in WFS2-deficient cells (
41,
42), suggesting a common pathophysiological mechanism. In many neurodegenerative diseases, the contact between ER and mitochondria is decreased (
43), similar to what we observed in WFS1 patients. Specifically, in Alzheimer’s disease, the number of contacts is increased (
44), similar to what is observed in Wolfram syndrome patients. Moreover, in amyotrophic lateral sclerosis type 8 (ALS8), the number of ER-mitochondria contacts is increased, and in ALS16 (
45), the number is decreased. These two observations are similar to what is observed in Wolfram syndrome. It seems that the number of ER-mitochondria contacts depends on the protein affected. MAM structure and function are tightly regulated to ensure proper functioning of the mitochondria. Any modification in the structure and/or function will lead to a dysfunction of mitochondrial respiration that will have repercussions on cellular physiology.
Although the reduction of [Ca
2+]
m uptake was not attributed to a change in Δψ
m, it may have resulted from two factors: the decrease in the ER Ca
2+ release and disorganization of the MAMs. One limitation of the study was the small number of patients from whom fibroblasts were analyzed because of the extreme rarity of the pathology. Rescue experiments for WFS1 failed to express the protein; however, down-regulation or overexpression of NCS1 in cells confirms the results, reinforcing the robustness of analyses of patient cells and suggesting that NCS1 is involved in Ca
2+ and MAM dysregulation in Wolfram syndrome. The reduced [Ca
2+]
m uptake impairs the function of the mitochondria, because some dehydrogenases of the tricarboxylic acid (TCA) cycle require Ca
2+ to maintain complex I and complex III and V function. There is no evidence that Ca
2+ controls the abundance of complex II. Nakamura
et al. (
46) have found that the loss of NCS1 in mouse cardiomyocytes alters the proper functioning of the mitochondria by decreasing ATP levels and the respiratory rate, which is associated with a decrease in complex I, III, and V protein abundance. Consistent with this work, cells from our Wolfram syndrome patients showed a substantial decrease in mitochondrial respiratory chain activity. The decrease in complex II–driven respiration is probably linked to the decrease in complex II expression.
The clinical hallmark of Wolfram syndrome is the association of diabetes mellitus and optic atrophy, due respectively to degeneration of pancreatic β cells and retinal ganglion cells (
47), two cell types in which NCS1 is abundant (
48,
49). In β cells, NCS1 localizes mainly in the cytosol, ER, and secretory granules. NCS1 promotes the priming of the secretory granules for release and increases the number of granules residing in the readily releasable pool. This action is mediated by an activation of the phosphatidylinositol 4-kinase β (
48). In MIN6 pancreatic β cells, WFS1 knockdown reduces insulin secretion, which is associated with a decrease in cytosolic Ca
2+ (
50). It is tempting to speculate that this dysfunction could be due to the possible loss of NCS1. In retinal ganglion cells, NCS1 may be involved in synaptogenesis during development and in synaptic transmission upon maturation (
49). These two processes depend on energy supplied by mitochondria transported along axons to the synaptic terminals (
51,
52) notably to maintain Ca
2+ homeostasis (
53,
54). Efficient axonal transport of mitochondria depends on their own energy production, because axonal transport is ATP dependent and, on cytoskeletal-mitochondrial interaction, because axonal transport is microtubule based (
55). Our data, combined with the work of Cagalinec
et al. (
11), suggest that altered Ca
2+ homeostasis due to decreased NCS1 protein abundance may be responsible for the impaired mitochondrial dynamics. Our data provide a potential explanation for the defective transport of mitochondria along the retinal ganglion cell axons because when NCS1 is diminished, as would be expected in WFS1-deficient retinal ganglion cells, mitochondrial function is impaired. This defect may be responsible for the retinal ganglion cell degeneration observed in Wolfram syndrome patients.
Wolfram syndrome patients may develop psychiatric disorders (
56), such as bipolar disorder, schizophrenia, anxiety, and depression (
57–
59). Surprisingly, changes in NCS1 abundance provoke similar symptoms in human and mice (
60,
61). It is tempting to speculate that the neurological and psychiatric disorders are linked to destabilization of NCS1 expression levels. In line with this, Schlecker
et al. (
31) have demonstrated that lithium promotes the interaction of IP
3R and NCS1 and in bipolar disorder.
In this study, we demonstrated that WFS1, NCS1, and IP
3R may be part of a complex associated with ER-mitochondria contact sites. In healthy cells, WFS1 interacts with NCS1 and may prevent its degradation. Our studies of the interactions between IP
3R and NCS1 have used coimmunoprecipitation (
31) or PLA (
30), approaches that do not allow us to conclude whether the interaction is direct. It is possible that another protein interacts with either NCS1 or IP
3R to bridge the two proteins. Nevertheless, the modulation of IP
3R function by NCS1 favors a direct interaction. When the WFS1/NCS1/IP
3R complex is functional, Ca
2+ can be properly transferred from the ER to mitochondria and activate the TCA cycle and the mitochondrial respiratory chain. When WFS1 is lost, the WFS1/NCS1/IP
3R complex is disrupted and NCS1 is partially degraded or destabilized. Consequently, altered ER-[Ca
2+]
m transfer leads to mitochondrial bioenergetic dysfunction that may result in the activation of cell death in the tissues affected by Wolfram syndrome (
Fig. 8). We show that NCS1 overexpression in Wolfram syndrome patient fibroblasts reverses most of the observed defects. As a consequence, NCS1 may be a drug target for the treatment of Wolfram syndrome.
MATERIALS AND METHODS
Cell cultures
Fibroblasts were cultured from skin biopsies taken after obtaining informed consent from three individuals without diabetes mellitus and four affected patients carrying mutations in
WFS1 gene, as previously described (
62). Human embryonic kidney (HEK) 293T cells were cultured in MegaCell Dulbecco’s Modified Eagle’s Medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (Sigma-Aldrich) and 20 mM penicillin/streptomycin (Sigma-Aldrich).
Cell transfection
Control fibroblasts were transfected with siRNA directed against WFS1 (ON-TARGETplus Human WFS1 siRNA SMARTpool, Thermo Fisher Scientific Biosciences, Waltham, MA, USA) and/or with siRNA directed against NCS1 (ON-TARGETplus Human NCS1 siRNA SMARTpool, Thermo Fisher Scientific Biosciences) using Lipofectamine 2000 (Invitrogen, Carlsbad, USA) according to the manufacturer’s instructions and processed 72 or 96 hours later for Western blot or oxygraphy analyses. Transfection with ON-TARGET Non-Targeting Pool (siScr) was used as controls.
To overexpress NCS1-Flag or pericam-mt vectors, fibroblasts were transfected by nucleofection. Nucleofection was performed according to an optimized protocol provided by the manufacturer (Amaxa Biosystems, Cologne, Germany). Cells were suspended in 100 μl of nucleofactor solution (Amaxa Biosystems); mixed with pericam-mt, Flag, or NCS1-Flag construct; and subjected to nucleofection, after which cells were transferred into prewarmed fresh medium. Cells were analyzed 24 hours after nucleofection.
Real-time RT-PCR
Real-time PCR was performed on total RNA extracted from cells using the RNeasy Mini Kit (Qiagen, Venlo, Netherlands) and reverse transcribed with the SuperScript III First-Strand Kit (Invitrogen) according to the manufacturer’s instructions.
Measurements of enzymatic activity and protein content
The activity of the mitochondrial respiratory chain complexes and respiratory rates was measured on cell homogenates, as described previously (
62). The activity of the mitochondrial respiratory chain complexes was measured with cell homogenates in a cell buffer [250 mM saccharose, 20 mM tris(hydroxymethyl)aminomethane, 2 mM EGTA, and bovine serum albumin (BSA; 1 mg/ml) (pH 7.2)] at 37°C using a UV-SAFAS spectrophotometer (SAFAS). Cellular protein content was determined with the BCA (bicinchoninic assay) Kit (Pierce) using BSA as a standard. Complex I (NADH ubiquinone reductase, EC 1.6.5.3) activity was measured according to a procedure described elsewhere (
63) and adapted using 2,6-dichloroindophenol (DCPIP) to avoid the inhibition of complex I activity by decylubiquinol (
64). Cells were disrupted by two freezing-thawing cycles, washed, centrifuged for 1 min at 16,000
g, and resuspended in cell buffer (50 μl per 10
6 cells). Complex II (succinate ubiquinone reductase) activity was measured according to James and colleagues (
65). Specific enzymatic activities of complexes I and II were expressed in mIU (nanomoles of DCPIP per minute per milligram of protein). Complex IV (cytochrome c oxidase) activity was recorded according to a method that Rustin and colleagues (
66) adapted in a 50 mM KH
2PO
4 buffer, using 15 μM reduced cytochrome c. Specific enzymatic activity was expressed in mIU (nanomoles of cytochrome c per minute per milligram of protein). CS activity was assayed by a standard procedure (
67). Specific enzymatic activity was expressed in mIU [nanomoles of 5-5′-dithiobis(2-nitrobenzoic acid) per minute per milligram of protein].
Oxygen consumption
The respiratory rate was measured using cells resuspended in respiratory buffer [0.5 mM EGTA, 3 mM MgCl2⋅6H2O, 60 mM K-lactobionate, 20 mM taurine, 10 mM KH2PO4, 20 mM Hepes, 110 mM sucrose, and BSA (1 mg/ml) (pH 7.1)] and permeabilized by incubation with digitonin (15 μg per 106 cells). The respiratory rates of 3 × 106 to 5 × 106 cells were recorded at 37°C in 2-ml glass chambers using a high-resolution Oxygraph respirometer (Oroboros). Malate (5 mM) and pyruvate (5 mM) were added to provide NADH to complex I. The decarboxylation of pyruvate is catalyzed by pyruvate dehydrogenase (PDH), producing acetyl coenzyme A. Malate dehydrogenase (MDH) located in the mitochondrial matrix induces the oxidation of malate to oxaloacetate. The condensation of oxaloacetate and acetyl coenzyme A induces the formation of citrate by CS. Thus, the addition of malate and pyruvate results in the production of NADH by four enzymes: MDH, PDH, isocitrate dehydrogenase, and α-ketoglutarate dehydrogenase. Because of the activation of different malate shuttles, complex II is not activated. Citrate and α-ketoglutarate are depleted by antiport transport with malate. Thus, the addition of malate and pyruvate defines the respiration driven by complex I used to compensate the physiological proton leak. Activation of ATP synthesis was induced by the addition of 1.5 mM adenosine 5′-diphosphate (ADP). Thus, we obtained the respiration driven by complex I coupled to the ATP synthesis. The addition of succinate (10 mM) allowed reconstitution of TCA cycle function with the activation of succinate dehydrogenase. Thus, maximal respiration of the mitochondrial chain induced by the competition of the respiration driven by complexes I and II was determined. Addition of rotenone (10 μM) inhibited the electron transfer from complex I to coenzyme Q and allowed the measurement of respiration driven by complex II. Oligomycin supplementation (8 μg/ml) inhibited the ATP synthesis, and respiration uncoupled driven by complex II was determined. Last, FCCP (1 μM) was added to test the quality of the permeabilization of the plasma membrane by digitonin.
Western blot analysis
Protein abundance was detected by immunoblot using commercially available antibodies and revealed using chemiluminescence. Frozen fibroblast pellets were lysed by a hypoosmotic shock (10 μl of H2O per 106 cells). For coimmunoprecipitation, HEK293T pellets were mixed with 500 μl of lysis buffer consisting of 0.1 M NaCl, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride (PMSF; Fluka), 1× SIGMAFAST protease inhibitor cocktail (Sigma-Aldrich), 5 mM EDTA, and 20 mM tris and rotated overnight at 4°C. The cellular protein content was determined with the BCA Kit.
Thirty micrograms of total fibroblasts or HEK293T-transfected protein was separated on SDS–polyacrylamide gel electrophoresis (PAGE) and transferred onto polyvinylidene difluoride membranes (Bio-Rad). Membranes were saturated with 5% nonfat milk dissolved in 0.1% tris-buffered saline (TBS)–Tween (pH 7.4, 137 mM NaCl, 2.7 mM KCl, 23 mM tris, and 0.1% Tween) for 1 hour at room temperature and incubated overnight at 4°C with polyclonal rabbit anti-WFS1 (1:500; Cell Signaling, USA), anti-MCU (1:1000; Abcam, UK), anti-MFN2 (1:1000; Cell Signaling), polyclonal chicken anti-NCS1 (1:1000; United States Biological, USA), monoclonal mouse anti-NDUFA13 (1:1000; Abcam), anti-SDHA (1:1000; Abcam), anti-UQCRC2 (1:1000; Abcam), anti-MTCO1 (1:1000; Abcam), anti–ATP synthase (1:1000; Invitrogen), anti-VDAC1 (1:1000; Abcam), anti-GAPDH (1:1000; Abcam), and anti-GRP75 (1:1000; Abcam). Anti-IP
3R antibody was a gift from J. Parys (Leuven, Belgium). Membranes were then washed three times in 0.1% TBS-Tween and incubated with anti-rabbit immunoglobulin G (IgG), anti-chicken IgG, or anti-mouse IgG horseradish peroxidase–linked antibody (1:10,000; Sigma-Aldrich) for 1 hour at room temperature. Immunoreactive proteins were visualized with enhanced chemiluminescence (ECL+ Western Blotting Detection Reagents, Amersham Biosciences, UK). Band intensities were quantified with ImageJ [National Institutes of Health (NIH), USA;
http://rsb.info.nih.gov/ij].
In situ PLA
ER-mitochondria interactions were analyzed using an optimized in situ PLA targeting the IP
3R/GRP75/VDAC1 at the MAM interface, as previously described (
27,
68).
Yeast two-hybrid screening
Yeast two-hybrid screening was performed by Hybrigenics Services SAS, Paris, France (
www.hybrigenics-services.com). The coding sequence for
Mus musculus–Wfs1 (amino acids 1 to 311; GenBank accession number GI:28422738) was PCR amplified and cloned into pB29 as a C-terminal fusion to LexA (N-Wfs1-LexA-C). The construct was checked by sequencing the entire insert and used as a bait to screen a random-primed mouse inner ear cDNA library constructed into pP6. pB29 and pP6 derive from the original pBTM116 (
69,
70) and pGADGH (
71) plasmids, respectively. Clones (54.3 million) (fivefold the complexity of the library) were screened using a mating approach with YHGX13 (Y187 ade2-101:loxP-kanMX-loxP, matα) and L40ΔGal4 (matα) yeast strains, as previously described (
72). Twenty-nine His+ colonies were selected on a medium lacking tryptophan, leucine, and histidine. The prey fragments of the positive clones were amplified by PCR and sequenced at their 5′ and 3′ junctions. The resulting sequences were used to identify the corresponding interacting proteins in the GenBank database (National Center for Biotechnology Information) using a fully automated procedure. A confidence score [Predicted Biological Score (PBS)] was attributed to each interaction, as previously described (
73).
The PBS relies on two different levels of analysis. First, a local score takes into account the redundancy and independency of prey fragments, as well as the distribution of reading frames and stop codons in overlapping fragments. Second, a global score takes into account the interactions found in all the screens performed at Hybrigenics using the same library. This global score represents the probability of an interaction being nonspecific. For practical purposes, the scores were divided into four categories, from A (highest confidence) to D (lowest confidence). A fifth category (E) specifically flags interactions involving highly connected prey domains previously found several times in screens performed on libraries derived from the same organism. Last, several of these highly connected domains have been confirmed as false positives of the technique and are now tagged as F. The PBS scores have been shown to positively correlate with the biological relevance of the interactions (
74,
75).
Confocal imaging
All confocal experiments were performed on fibroblasts placed on the stage of a Zeiss LSM 510 inverted confocal microscope (Zeiss, Le Pecq, France). Pericam-mt/pcDNA was used to measure [Ca
2+]
m. To measure cytosolic Ca
2+, fibroblasts were loaded with Fluo-4 AM (5 μM; Molecular Probes), Fluo-4 signal was obtained by excitation at 488 nm, and emitted light was collected at 515 nm. For pericam-mt, the excitation wavelength upon Ca
2+ binding switches from 415 to 488 nm. Thus, the transfected cells were excited at 488, and images were collected at 515 nm. All the collected images were processed with ImageJ (NIH, USA;
http://rsb.info.nih.gov/ij/). All confocal experiments were performed on fibroblasts placed on the stage of a Zeiss LSM 510 inverted confocal microscope (Zeiss) equipped with a 63× lens [oil immersion; numerical aperture (NA), 1.2].
ER Ca2+ release
Luminal Ca
2+ dynamics were measured using the Ca
2+-sensitive FRET-based cameleon protein D1-ER (
76). Cells were cultured on 24-mm coverslips and transfected with D1-ER. Cells were imaged on a Zeiss Axiovert 200M microscope, equipped with a 40× oil objective (NA, 1.3). Emission ratio imaging of D1-ER was accomplished by using a 436DF20 excitation filter, a 450-nm dichroic mirror, and two emission filters [475/40 for enhanced cyan fluorescent protein (ECFP) and 535/25 for citrine]. Exposure times were 100 ms, and images were taken every 1 to 2 s. The FRET signal (yellow fluorescent protein/CFP) was normalized to CFP emission intensity, and changes in ER Ca
2+ were expressed as the ratio of the emissions at 535 and 470 nm. To induce Ca
2+ release from ER, the cells were challenged with an agonist that, through interaction with G protein–coupled receptors, evokes a rapid discharge from IP
3Rs.
Ca2+ measurements with aequorin constructs
For mt-AEQ measurements, coverslips with control and mutated fibroblasts overexpressing MCU were incubated for 2 hours in KRB (Krebs-Ringer modified buffer) (135 mM NaCl, 5 mM KCl, 0.4 mM KH2PO4, 1 mM MgSO4, 5.5 mM glucose, and 20 mM Hepes (pH 7.4)] at 37°C supplemented with wild-type coelenterazine (5 M) and then transferred to the perfusion chamber. In the experiments with permeabilized cells, cells were perfused in intracellular buffer [140 mM KCl, 10 mM NaCl, 1 mM KH2PO4, 5.5 mM glucose, 2 mM MgSO4, 1 mM ATP, 2 mM Na+ succinate, and 20 mM Hepes (pH 7.05)] at 37°C. Cells were permeabilized through a 1-min perfusion with 100 μM digitonin during luminescence measurements. After a 2-min washing (to remove digitonin), [Ca2+]m uptake was estimated by mt-AEQ after cell perfusion with the same intracellular solution without EGTA and containing Ca2+ at different concentrations (1 and 4 μM). Calcium uptake speed was calculated as the first derivative by using the first derivative function smoothed for three time points.
To reconstitute erAEQ with high efficiency, luminal [Ca2+] of the ER was first reduced by incubating the cells for 45 min at 4°C in KRB supplemented with 5 μM coelenterazine, the Ca2+ ionophore ionomycin, and 600 μM EGTA. After incubation, the cells were extensively washed with KRB supplemented with 2% BSA and 2 mM EGTA before the luminescence measurement was initiated. Aequorin signals were measured in KRB supplemented with either 1 mM CaCl2 or 100 μM EGTA, using a purpose-built luminometer. The agonist (bradykinin at 100 μM) was added to the same medium.
Mitochondrial membrane potential
To measure Δψ
m, fibroblasts were loaded with 10 nM TMRM (Life Technologies) for 20 min at 37°C, followed by washout. TMRM dyes were excited at 565 nm, and emitted light was collected at 585 nm. At the end of each experiment using TMRM dyes, cells were exposed to the mitochondrial uncoupler FCCP (1 μM) to fully dissipate (Δψ
m) and to determine the dynamic range of the dye. TMRM fluorescence was normalized to the fluorescence signals obtained after FCCP application. After loading, cells were placed on the stage of a Zeiss LSM 510 inverted confocal microscope (Zeiss) equipped with a 63× lens (oil immersion; NA, 1.2). Collected images were processed with ImageJ (NIH, USA;
http://rsb.info.nih.gov/ij/). Detection of Δψ
m was also performed by loading cells with a JC-1 staining kit (Thermo Fisher Scientific). JC-1 emissions were collected using a Zeiss LSM 510 confocal microscope equipped with a 40× oil objective (NA, 1.30). The green fluorescent/red fluorescent intensity ratios (540/595 nm) were calculated to determine Δψ
m levels.
Coimmunoprecipitation
pcDNA-IP
3R1 was provided by K. Mikoshiba (Okasaki, Japan) (
77). The plasmid expressing Wfs1 was obtained by subcloning Wfs1 into myc-tagged pCS2 + MT using Hind III and Bam H1. Ncs1 was subcloned into C-terminal p3XFLAG-CMV between Eco R1 and Bam H1. To assess interactions between IP
3R1 and Ncs1-Flag and between Wfs1-myc and Ncs1-Flag, 100 μg of proteins from each transfection reaction was mixed with 2 μl of mouse anti-Flag M2 antibody (Sigma-Aldrich) containing 1 mM PMSF and 2% BSA in DWBa buffer [100 mM NaCl, 20 mM tris, 1 mM EDTA (pH 7.4), 1% Triton X-100]. For IP
3R1 and Wfs1-myc antibody, 100 μg of proteins from each transfection reaction was mixed with 2 μg of a rabbit anti-myc (Sigma-Aldrich). The mixtures were vortexed and incubated at 4°C overnight. Thirty microliters of Sepharose A and 10 μl of Sepharose G (Immunoprecipitation Starter Pack, GE Healthcare) were used per sample and prepared following supplier instruction. Sepharose suspension was added to the tubes containing antibody/protein samples and mixed on a rotating wheel at 4°C for 1 hour. The samples were then centrifuged at 2000 rpm for 1 min and washed three times with 1 ml of DWBa and two times with 1 ml of DWBb [100 mM NaCl, 20 mM tris, and 1 mM EDTA (pH 7.4)] each. The pellets obtained after the last wash were resuspended in 25 μl of Laemmli sample buffer (Bio-Rad). Last, all the samples were boiled at 95°C for 5 min and centrifuged at 13,200 rpm for 1 min. The supernatants were loaded on an SDS-PAGE gel for Western blotting and analyzed with an anti-myc antibody for Wfs1-myc and Ncs1-Flag coimmunoprecipitation, with the rabbit Rbt476 pan anti-IP
3R antibody (a gift from J. Parys, Leuven, Belgium) (
78) for Ncs1-Flag and IP
3R1 coimmunoprecipitation and Wfs1-myc and IP
3R1 coimmunoprecipitation. Thirty micrograms of protein from Ncs1-Flag and Wfs1-myc transfection reactions was used as controls.
Transmission electron microscopy
Cells were immersed in a solution of 2.5% glutaraldehyde in PHEM buffer (1×, pH 7.4) overnight at 4°C, rinsed in PHEM buffer, and postfixed in a 0.5% osmic acid for 2 hours in the dark and at room temperature. After two rinses in PHEM buffer, cells were dehydrated in a graded series of ethanol solutions (30 to 100%) and were embedded in EMbed 812 using an Automated Microwave Tissue Processor for Electronic Microscopy (Leica). Ultrathin sections (70 nm; Leica-Reichert Ultracut E) were collected at different levels of each block. These sections were stained with uranyl acetate and lead citrate before examination in a Tecnai F20 TEM at 200 kV in the CoMET Montpellier Rio Imaging (MRI) facilities, Institute for Neurosciences of Montpellier, France. To evaluate the frequency of contact between the mitochondria and the ER in the different fibroblasts, we randomly took 20 images of cytoplasmic area from 20 different cells in each sample with a JEOL 1400 TEM at ×15,000 magnification. In each image, we counted the number of mitochondria (403 in control cells and 649 in patient cells for a total of 1052 mitochondria) and calculated the proportion of mitochondria in close contact with ER (<30 nm). Moreover, the perimeter of each mitochondria and the proportion of the mitochondrial surface closely associated with ER were calculated. For experiments with knockdown of NCS1, 257 mitochondria were analyzed for siScr and 239 mitochondria were analyzed for siRNA NCS1.
Statistical analysis
The Mann-Whitney was used to compare the fibroblasts from WFS1 patients and controls. The Kruskal-Wallis test was used to compare the effect of the overexpression of NCS1 or vector alone. Differences were considered significant at *P < 0.05, **P < 0.01, and ***P < 0.001. Proportions of ER-mitochondrial contacts between controls and patient fibroblasts were compared with their 95% confidence intervals.