SPLICS: a split green fluorescent protein-based contact site sensor for narrow and wide heterotypic organelle juxtaposition

SPLICS: a split green fluorescent protein-based contact site sensor for narrow and wide... Contact sites are discrete areas of organelle proximity that coordinate essential physiological processes across membranes, 2+ including Ca signaling, lipid biosynthesis, apoptosis, and autophagy. However, tools to easily image inter-organelle proximity over a range of distances in living cells and in vivo are lacking. Here we report a split-GFP-based contact site sensor (SPLICS) engineered to fluoresce when organelles are in proximity. Two SPLICS versions efficiently measured narrow (8–10 nm) and wide (40–50 nm) juxtapositions between endoplasmic reticulum and mitochondria, documenting the existence of at least two types of contact sites in human cells. Narrow and wide ER–mitochondria contact sites responded differently to starvation, ER stress, mitochondrial shape modifications, and changes in the levels of modulators of ER–mitochondria juxtaposition. SPLICS detected contact sites in soma and axons of D. rerio Rohon Beard (RB) sensory neurons in vivo, extending its use to analyses of organelle juxtaposition in the whole animal. Introduction cellular activities. Indeed, a network of contact sites between membranes of different organelles guarantees their In eukaryotic cells, organelles are often found in close mutual communication by creating microdomains that favor proximity, leading to the generation of heterotypic mem- different signaling and metabolic pathways [1,2]. Due to brane appositions that ensure the coordination of several their central role in many fundamental cell processes, the sites of apposition between mitochondria and the endo- plasmic reticulum (ER), which range from 10 to 100 nm, Edited by N. Chandel are, so far, the best characterized [3–5]. Domenico Cieri, Mattia Vicario and Marta Giacomello contributed Several approaches are currently available to assess equally to this work. ER–mitochondria contact sites. Electron microscopy (EM) Electronic supplementary material The online version of this article allows to calculate contact site distance, but it is time- (https://doi.org/10.1038/s41418-017-0033-z) contains supplementary consuming. The in situ proximity ligation assay is based on material, which is available to authorized users. the use of pairs of primary antibodies against proteins on opposing membranes [6]. It is widely used [7–9] but is not * Marisa Brini marisa.brini@unipd.it devoid of drawbacks: as the EM, it can only be used in fixed * Tito Calì cells and is limited by the availability and the specificity of tito.cali@unipd.it the antibodies. The use of fluorescent proteins (FPs) selectively targeted Department of Biomedical Sciences, University of Padova, to the mitochondrial matrix and the lumen of the ER [10] Padova, Italy has been the golden standard to visualize contact sites in Department of Biology, University of Padova, Padova, Italy living cells for years. However, limited resolution in the distance range below 200 nm, differences in FPs expression Dulbecco-Telethon Institute, Venetian Institute of Molecular Medicine, Padua, Italy levels or alterations in organelle morphology complicated Venetian Institute of Molecular Medicine, Padua, Italy the interpretation of experiments of ER–mitochondria jux- taposition upon ablation of the mitochondria-shaping pro- Department of Biomedical Sciences, Institute of Neuroscience, Italian National Research Council (CNR), Padua, Italy tein Mitofusin 2 (Mfn2) [11–13]. 1234567890 1132 D. Cieri et al. To overcome these limits, FP-based sensors of proximity targeted each moiety on one of the juxtaposed membranes, the were developed: a dimerization-dependent FP (ddGFP) [14] GFP fluorescence would be restored only when the two or Venus FP [2,15,16] and a FRET-based probe coupled to a portions were close enough. We therefore placed the non- rapamycin-binding module (FEMP) [17]. While these two fluorescent GFP moiety on the cytosolic face of the OMM 1–10 probes improved the analysis of ER–mitochondria proximity, (OMM-GFP ). To follow short- (≈8–10 nm) and long- 1–10 they also face some limitations: the ddGFP probe is intrin- range (≈40–50 nm) ER–mitochondria interactions [5], two sically not extremely bright [14]; the FRET probe requires constructs that differ for the length of the spacer placed equimolar expression of the two moieties [12] and its in vivo between the ER targeting sequence and the β fragment were applications are limited by the use of rapamycin, a potent created by considering the distance of 0.36 nm between two inducer of autophagy [18–20], to maximize juxtaposition and alpha-carbons in a peptide chain: a ER-Short β with a 29 aa FRET signal. Moreover, both probes cannot be adapted for spacer and a ER-Long β with a 146 aa spacer (i.e., a the investigation of contact sites potentially placed at dif- maximum of ≈10.4 and 52.5 nm, respectively). These values ferent distances, because their dynamic range must be might clearly be subjected to changes (i.e., reduction) since characterized each time that the linker is changed. Artificial the amino acid sequences might not always be fully extended. GFP-based tethers have proved useful to uncover a novel We reasoned that co-expression of ER-Short β with OMM- ER–mitochondria tethering complex in yeast, but they cannot GFP (SPLICS ) and of ER-Long β with OMM-GFP 1–10 S 11 1–10 been used to monitor changes in the ER–mitochondria con- (SPLICS ) would result in reconstitution of GFP fluorescence tact sites [21]. Therefore, an easy, one-step probe that can (Fig. 1a). Two additional constructs, a β -tagged FP (Kate- dynamically detect ER–mitochondria juxtaposition in cellulo β ) and an untargeted GFP were also generated to verify 11 1–10, and in vivo is lacking. the complementation of the OMM-GFP at the OMM 1–10 To overcome these limitations, we devised a split-GFP- (Fig. 1a, left) and the ER -β at the ER (Fig. 1a, middle), S/L 11 based contact site sensor (SPLICS) that can be easily respectively. Expression of SPLICS and SPLICS will result S L adapted to measure ER–mitochondria contact sites over a in fluorescence emission specifically at the ER–mitochondria range of distances as well as other types of hetero and interface (Fig. 1a, right). homotypic contact sites. Upon expression in human cells, The different versions of the SPLICS probes were first this one-step imaging technique specifically identifies nar- tested for their correct localization and topology. A clear row and wide ER–mitochondria apposition lying in a range mitochondrial network appeared in HeLa cells co- of around 10 and 50 nm [5], i.e, that found between mito- expressing OMM-GFP with Kate-β (Fig. 1b, first 1–10 11 chondria and smooth or rough ER [22]. The narrow SPLICS panels); similarly, the ER network became fluorescent when can also detect ER–mitochondria contact sites in vivo in ER -β and ER -β were co-expressed with a cytosolic S 11 L 11 zebrafish sensory neurons. Pharmacological and genetic non-fluorescent GFP (Fig. 1b, second and third panel 1–10 manipulations indicate that these narrow and wide contact couples). Interestingly, when SPLICS or SPLICS were S L sites respond differentially to ER stress, autophagy, mito- expressed in HeLa (Fig. 1b, fourth and fifth panel couples) phagy, and changes in the levels of modulators of and in HEK293 cells (Supplementary Figure S1), fluor- ER–mitochondria juxtaposition. escent individual foci appeared, likely representing the In conclusion, using SPLICS as a tool to investigate juxtapositions between ER and mitochondria. At a closer ER–mitochondria contact sites, we unravel their hetero- inspection, the SPLICS and SPLICS signals retrieved in S L geneity and provide the community with a sensor that can HeLa cells appeared different for number (see quantification be easily adapted to image other types of heterotypic in 3D rendered z-stack images, Fig. 1c). organelle contact sites in human cells and in whole We therefore verified whether SPLICS really recognized organisms. areas of ER–mitochondria juxtaposition. In HeLa cells expressing SPLICS ,the fluorescent dots co-localized with S/L endogenous markers of mitochondria (mtHSP60) and ER Results (calreticulin) (Fig. 1d). Noteworthy, the mitochondrial and ER networks were not completely engaged in the formation of the Two SPLICS probes for different ranges of ER–mitochondria contacts reported by the SPLICS (see ER–mitochondria juxtaposition merge panels of Fig. 1d), suggesting that SPLICS snapshots the juxtaposition at any given moment even when transiently To generate a modular fluorescence based sensor of organelle formed. Immuno-EM with anti-GFP antibody revealed that proximity, we decided to capitalize on the ability of two non- mitochondria and ER membranes in contact with mitochon- fluorescent portions (the GFP moiety and the GFP β-strand dria were preferentially marked (arrowheads in Supplemen- 1–10 11) of the superfolder GFP variant [23–25] to restore a fully tary Figure S2). Despite the non-complemented and fluorescent GFP upon self-assembly. We reasoned that if we complemented OMM-GFP cannot be distinguished by the 1–10 ER-mitochondria contact sites sensor 1133 Fig. 1 Functional characterization of the SPLICS probes. a Cartoon showing the general approach used to design the SPLICS. The mitochondrial network, the ER network, and the ER–mitochondria contact sites are revealed by co- expression of the β -tagged cytosolic RFP (Kate) and the OMM-GFP (left panel), of 1–10 the ER -β constructs and a S/L 11 cytosolic GFP (middle panel) 1–10 and of the SPLICS (right S/L panel), respectively. b Experimental controls showing the correct targeting of the mitochondrial (OMM-GFP ) 1–10 and the ER (ER -β and ER - S 11 L β ) targeted fragments verified by complementation with Kate- β and GFP , respectively. 11 1–10 Co-transfection of HeLa cells with OMM-GFP and both 1–10 ER -β or ER -β induces the S 11 L 11 appearance of a “dotted” fluorescence. c Quantification of ER–mitochondria contacts in HeLa cells. The SPLICS dots were quantified from the 3D rendering of a complete z-stack. Mean ± SEM: SPLICS 56 ± 4, n = 37 cells; SPLICS 229 ± 12, n = 25 cells. d Co-localization of SPLICS fluorescence with S/L mitochondria (mtHSP60) and ER (CRT: calreticulin) markers. Representative traces e and statistical analysis f of 2+ mitochondrial Ca uptake in HeLa cells transfected with SPLICS or SPLICS along with S L mtAeqmut. Mean ± SEM: Void Vector 75 ± 2, n = 65 wells; SPLICS 77 ± 1, n = 54 wells; SPLICS 71 ± 2, n = 54 wells. Scale bar 15 µm. Data shown are the result of 3–5 independent experiments. anti-GFP antibody, it is evident that a consistent number of To gain further insights on the nature of the reconstituted gold nanoparticles are distributed at the ER–mitochondria SPLICs, we evaluated their stability by checking whether interface (inset in Supplementary Figure S2). the number of SPLICS foci could change after 24, 48, S/L 1134 D. Cieri et al. 2+ and 72 h post transfection. Supplementary Figure S3 shows that it does not artificially increase tethering and Ca that the number of SPLICS is stable during the time transfer between ER and mitochondria. S/L course. The number of fluorescent reconstituted foci was also unaffected by the expression level of the probes Modulation of short- and long-range (Supplementary Figure S4), suggesting that bona fide ER–mitochondria interfaces during ER stress and changes in the SPLICS fluorescent foci likely reflect a autophagy S/L variation in ER–mitochondria contact sites number rather than differences in the stability/expression levels of the We next wished to address if SPLICS could respond to probes. Additionally, the overall morphology of the ER and pathophysiological conditions known to affect the extent of mitochondria in cells expressing the SPLICS remained ER–mitochondria contacts. We therefore measured short- S/L grossly unaltered (Supplementary Figure S5). and long-range ER–mitochondria interactions in conditions To exclude that novel and non-physiological contact sites where increased ER–mitochondria coupling was reported, between ER and mitochondria might be artificially induced such as ER stress and induction of autophagy [5,26,27]. In 2+ by SPLICS expression, ER–mitochondria Ca transfer and HeLa cells treated with the ER stress inducer tunicamycin, 2+ mitochondrial Ca uptake were evaluated in HeLa cells or starved, the number of short-range ER–mitochondria expressing SPLICS by aequorin-based measurements. If contact sites measured by SPLICS were increased S/L S 2+ this was the case, mitochondrial Ca transients generated (Figs. 2a,b), in agreement with previous results [5,26,27]. by stimulation with the InsP -linked agonist histamine The picture in the case of long-range ER–mitochondria should be increased in SPLICS-expressing cells [17]; interactions measured by the SPLICS was more complex: however, they were superimposable to those of control cells while tunicamycin significantly decreased the number of (Fig. 1f and quantification in Fig. 1g). Taken together, these SPLICS dots, starvation did not induce any significant results indicate that SPLICS retains the ability to self- change (Figs. 2c,d). Altogether, these results indicate associate only in specific areas where the two organelles are that short and long ER–mitochondria interactions are dif- found within the distance imposed by the linker region and ferentially modulated in response to different stimuli and Fig. 2 Effect of Tunicamycin and Hbss treatment on ER–mitochondria contacts. Immunofluorescence against mitochondria (Tom20, red) is shown in the panels on the middle. The green channel is the merge of several planes. Scale bar 20 µm. a Representative confocal pictures of HeLa cells expressing the SPLICS probe. b Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 58 ± 3, n = 32 cells; Tunicamycin 84 ± 5, n = 33 cells; Hbss 81 ± 5, n = 25 cells. c Representative confocal pictures of HeLa cells expressing the SPLICS probe. d Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 218 ± 11, n = 27 cells; Tunicamycin 171 ± 9, n = 33 cells; Hbss 204 ± 10, n = 23 cells. Data shown are the result of three independent experiments. **p ≤ 0.01, ***p ≤ 0.001. ER-mitochondria contact sites sensor 1135 suggest that the heterogeneity between the two types of reduction in the number of wide ER–mitochondria inter- contact sites reflects their involvement in specialized cel- actions in cells expressing wt Drp1 (Fig. 3c top panels vs lular pathways. middle panels, and Fig. 3d). Interestingly, forced mito- chondrial elongation induced by Drp1-K38A expression Short- and long-range ER–mitochondria interactions resulted in the labeling of the whole surface of mitochondria are differentially modulated by mitochondrial by SPLICS fluorescence, suggesting a complete engage- morphology ment of the mitochondrial network with the ER (Fig. 3c, top panels vs. bottom panels). Due to the filamentous nature During starvation, inhibition of the mitochondrial fission of the observed SPLICS staining, the number of GTPase Dynamin-related protein 1 (Drp1) results in mito- ER–mitochondria contacts/cell under this condition could chondrial elongation, increasing energy conversion and not be reliably quantified; nevertheless, the GFP signal sparing mitochondria from autophagosomal degradation occupied almost completely (about 85%) the mitochondrial [28,29]. We therefore wished to verify short and long surface as measured by Tom20 staining (Supplementary ER–mitochondria interactions upon Drp1-driven mito- Figure S6). Altogether, these results suggest that unopposed chondrial shape changes. We expressed wt or a dominant- mitochondrial fusion is paralleled by an enhancement of the negative mutant form of Drp1 (Drp1-K38A) to induce ER–mitochondria interface that may ensure the supply of mitochondrial fragmentation or elongation and measured lipids required for the sustained mitochondrial morpholo- the occurrence of short- and long-range ER–mitochondria gical changes [26,28,29,31]. juxtaposition with SPLICS. Mitochondrial fragmentation induced by wt Drp1 expression did not change the number Short- and long-range ER–mitochondria contacts of short-range ER–mitochondria interactions (Figs. 3a and respond differentially to Mfn2 silencing and b, compare top panels vs. middle panels), in agreement with presenilin 2 mutant expression previous data [30]. Conversely, mitochondrial elongation induced by dominant-negative Drp1 expression resulted in a We next wished to verify if SPLICS responded to genetic significant increase in the short-range ER–mitochondria modulation of the ER/mitochondria interaction. To this end, contacts detected by SPLICS (Fig. 3a, top panels vs. lower we decided to monitor SPLICS behavior following S S/L panel, and Fig. 3b). The SPLICS measured a significant ablation of Mitofusin 2 (Mfn2), a pro-fusion mitochondria- Fig. 3 Effects of Drp1 overexpression on ER–mitochondria contacts. Immunofluorescence against mitochondria (Tom20, cyan) and Drp1 (red) is shown in the corresponding panels. The green channel is the merge of several planes. Scale bar 20 µm. a Representative confocal pictures of HeLa cells expressing the SPLICS probe. b Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 59 ± 3, n = 79 cells; Drp1 WT 70 ± 5, n = 32 cells; Drp1-K38A 89 ± 9, n = 28 cells. c Representative confocal pictures of HeLa cells expressing the SPLICS probe. d Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 260 ± 14, n = 24 cells; Drp1 WT 198 ± 14, n = 24 cells. Data shown are the result of 3–4 independent experiments. **p ≤ 0.01. 1136 D. Cieri et al. shaping protein originally identified as a tether between the mitochondria coupling in HeLa cells by organelle-targeted two organelles [11]. However, whether Mfn2 tethers FPs, and in nigral neurons by transmission EM analysis [12,14,32–35] or separates [13,22,36–38] ER and mito- [42,43]. Nevertheless, increased ER–mitochondria juxta- chondria is still a matter of debate. We reasoned that position in patient-derived fibroblasts and in PARK2 SPLICS might contribute to clarify the issue by providing knockout MEFs [44] was also reported. Thus, the exact S/L an estimate of the contact sites over different ranges of function of Parkin at the ER–mitochondria interface under interaction. Acute downregulation of Mfn2 by shRNA in basal conditions and upon mitophagy is unclear. We gen- HeLa cells by three independent shRNA (Supplementary erated a bicistronic vector in which Parkin was cloned Figure S7) increased by ≈40% the number of SPLICS foci upstream of a self-cleaving viral 2A peptide (P2A) [45] (Figs. 4a,b). Conversely, under the same conditions of Mfn2 followed by a plasma membrane-targeted RFP (mCherry- downregulation the SPLICS detected a significant decrease CAAX) to track Parkin-positive cells (Supplementary Fig- by ≈30% in the number of ER–mitochondria interactions ure S9). This construct was co-expressed along with (Figs. 4c,d). Altogether the short- and long-range SPLICS SPLICS in HeLa cells where Parkin is absent or weakly S/L probes not only respond to changes in known modulators of expressed [46,47]. Parkin overexpression increased ER–mitochondria tethering, but they might also prove SPLICS number (Figs. 5a,b), in agreement with our pre- useful to shed light on the observed discrepancies on the vious data [42]. Conversely, Parkin overexpression reduced role of Mfn2 at the ER–mitochondria interface. the SPLICS foci (Figs. 5c,d). Treatment with CCCP Mfn2 and the Familial Alzheimer’s Disease (FAD)- reduced the number of fluorescent foci measured using both related protein Presenilin-2 (PS2) have been reported to act the SPLICS probes, suggesting that activation of PINK1/ in a common route to tune the ER–mitochondria interface Parkin-mediated mitophagy loosens all types of [38,39]. We measured short-range ER–mitochondria inter- ER–mitochondria contacts. actions in human fibroblasts from an FAD-patient carrying the PS2-N141I mutation, previously shown to enhance SPLICS visualizes ER–mitochondria interactions in ER–mitochondria coupling in an Mfn2-dependent manner living zebrafish neurons [38], and a healthy sex- and age-matched control. The SPLICS signal was more than doubled in human FAD-PS2 We finally wished to test if SPLICS can measure fibroblasts compared to controls, thus confirming that ER–mitochondria tethering in an in vivo setting. Imaging of endogenous FAD-PS2 increases ER–mitochondria cou- subcellular structures in living animals, and even more in pling, as already reported, and proving that SPLICS repre- neuronal axons, is limited by the thickness and anatomical sents a useful tool also in patient-derived samples (Figs. 4e, accessibility of tissues. In vivo detection of organelle con- f). Lastly, we tested SPLICS with an additional well- tact sites is still a major challenge because of their dynamic S/L established tethering machinery, i.e., the VAPB/PTPIP51 nature and the lack of appropriate tools. To verify if complex. Interestingly, we detected an increase in the SPLICS could overcome these hurdles, we expressed the SPLICS number, in agreement with previous data [7,8,40] new probes in D. rerio, specifically in Rohon-Beard (RB) (Supplementary Figure S8). The long-range interactions sensory neurons. The correct targeting of the OMM- monitored by SPLICS were instead decreased (Supple- GFP and the ER -β constructs was first verified after L 1–10 S 11 mentary Figure S8): this finding certainly deserves addi- mosaic expression in D. rerio embryos. The OMM-GFP 1–10 tional experiments but again, it might indicate that signal reconstituted by complementation with a β -tagged ER–mitochondria tethering can be heterogeneous and cytosolic protein (DJ-1-β ) fully overlapped with a mito- tightly modulated. chondrial targeted RFP (pTagRFP-mito). Analogously, injection of ER -β and a cytosolic GFP resulted in S 11 1–10 Long- and short-range ER mitochondria contacts fluorescence emission that co-localized with an ER marker reduction during Parkin-mediated mitophagy (pDsRed2-ER) (Supplementary Figure S10), thus demon- strating that the SPLICS fragments are properly expressed, Comforted by the ability of SPLICS to provide insights targeted and self-assembled in living zebrafish embryos. To S/L under pharmacological and genetic manipulation of the allow tissue specific as well as equimolar expression of ER–mitochondria interface, we decided to detect changes in SPLICS, we generated an expression vector where OMM- ER–mitochondria tethering during Parkin-mediated mito- GFP and ER -β are linked by a P2A peptide 1–10 S 11 phagy. In mammalian cells, dysfunctional mitochondria (SPLICS -P2A), an approach suitable also in zebrafish [48]. recruit the E3 ubiquitin ligase Parkin to the OMM through SPLICS -P2A was placed under the control of a bidirec- PINK1 kinase activity, resulting in the recruitment and tional UAS promoter together with a cytosolic DsRed activation of the autophagy machinery [41]. Parkin has been (pT2-DsRed-UAS-SPLICS -P2A) to allow GAL4-driven shown to act as a positive modulator of ER and expression of the UAS promoter (Fig. 6a). The pT2-DsRed- ER-mitochondria contact sites sensor 1137 UAS-SPLICS -P2A vector was then microinjected in the simultaneous, tissue specific expression of DsRed and zebrafish s1102t:GAL4 transgenic line where GAL4 SPLICS (Fig. 6c). By imaging the DsRed-positive neu- expression is restricted to RB neurons (Fig. 6b), yielding rons, we noticed the occurrence of short-range ER 1138 D. Cieri et al. Fig. 4 Effect of Mfn2 knockdown and mutant PS2 on (implying that any reduction observed with SPLICS may S/L ER–mitochondria interface. Immunofluorescence against mitochondria not reflect a dynamic decrease during time). Even if the (Tom20, red) is shown in the panels on the middle. The green channel observation of transient interactions between the ER and is the merge of several planes. Scale bar 20 µm. a Representative motile mitochondria could be limited by the time required confocal pictures of HeLa cells expressing the SPLICS probe. b Quantification of SPLICS contacts by 3D rendering of complete z- to achieve full reconstitution of the SPLICS ,wewere S S/L stacks. Mean ± SEM: SCR shRNA 70 ± 4, n = 76 cells; shRNA Mfn2 able to provide important insights in the biology of this #1 98 ± 7, n = 26 cells; shRNA Mfn2 #3 108 ± 7, n = 28 cells; interface. shRNA Mfn2 #4: 95 ± 7, n = 23 cells. c Representative confocal SPLICS can be also adapted to monitor other types of pictures of HeLa cells expressing the SPLICS probe. d Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± heterotypic organelle contact sites, e.g., ER and plasma SEM: SCR shRNA 238 ± 10, n = 30 cells; shRNA Mfn2 #1 172 ± 9, membrane (PM), mitochondria and PM, or mitochondria n = 27 cells; shRNA Mfn2 #3 176 ± 8, n = 29 cells; shRNA Mfn2 #4 and endosomes/lysosomes, creating a palette of SPLICS to 190 ± 10, n = 27 cells. e Representative confocal pictures of human image inter-organelle interactions. fibroblasts from a patient with the N141I mutation in PS2 (bottom panel) and an age-matched control (upper panel) expressing the The physiological significance of long-range ER– SPLICS probe. The green channel is the merge of several planes. mitochondria contacts has not been completely defined; Scale bar 20 µm. f Quantification of ER–mitochondria short contacts nevertheless, the comparison of the SPLICS signals under S/L by 3D rendering of complete z-stacks. Mean ± SEM: CTRL 50 ± 5, different pathophysiological conditions indicates that n = 20 cells; PS2-N14I: 101 ± 12, n = 21 cells. Data shown are the result of 2–5 independent experiments. **p ≤ 0.01, ***p ≤ 0.001, ER–mitochondria tethering is heterogeneous and tightly ****p ≤ 0.0001. modulated. Both ER stress and starvation increased SPLICS foci while SPLICS dots were decreased in num- S L mitochondria contacts in both cell body and axons ber under ER stress, suggesting a spatial and functional (Figs. 6d–g). We retrieved several SPLICS contacts in the S specialization of different ER–mitochondria contact sites soma of RB neurons; their frequency was comparable to [27]. Changes in mitochondrial shape also affected the that observed in cultured cells. ER–mitochondria contact ER–mitochondria interface differently: Drp1 overexpression sites were also retrieved in RB axons and enriched at axonal reduced SPLICS interactions, whereas forced mitochon- varicosities and branching points, possibly representing drial elongation increased both SPLICS foci number. S/L 2+ axon zones with specialized functions where Fragmentation concomitantly ensures basal Ca -dependent ER–mitochondria crosstalk is important to propagate and homeostatic mitochondrial functions and protects from stress 2+ 2+ regulate Ca signals [49–51] (arrowheads in Fig. 6f). The responses involving ER–mitochondria Ca crosstalk [30]. number of short ER–mitochondria interactions was com- Indeed, potentiation of the ER–mitochondria interface under parable in soma and axons (Fig. 6g), suggesting that these conditions of Drp1-dependent fragmentation can lead to 2+ juxtapositions are regulated by similar mechanisms in the mitochondrial Ca overload and cell death. The reduction two portions of the neuron. in SPLICS number is probably due to the reduction of the interface available for additional contacts and it could be 2+ relevant in the Ca -dependent stress responses. Of note, Discussion Drp1-dependent mitochondrial fission facilitates mitophagy whereas mitochondrial elongation inhibits mitochondrial Here we report a one-step, split-GFP-based method to autophagy [52]. Thus, it is tempting to speculate that mito- assess narrow and wide ER–mitochondria contact sites and chondria are protected from mitophagy when mitochondrial their modulation and demonstrate its suitability to monitor fusion is forced due to increased ER–mitochondria contacts; by confocal microscopy inter-organelle interactions in conversely, mitochondrial fission is accompanied by a human cells and in vivo in zebrafish RB neurons. reduction in SPLICS signal to favor the engulfment of The SPLICS is versatile and sensitive and reveals short mitochondria by the autophagosome. and long-range ER–mitochondria interactions and their The SPLICS described here can also help to understand changes upon pharmacological and genetic manipulations. the role of Mfn2 in ER–mitochondria juxtaposition: we An advantage of SPLICS over the other probes relies observed that Mfn2 silencing led to an increase in SPLICS not only on its unnerving modularity, but also on the and a decrease in SPLICS dots. Similar findings were high brightness, stability and a high threshold over back- observed upon the overexpression of the well-established ground. The first feature was exploited here to generate two tethering machinery VAPB/PTPIP51. Future work might versions of SPLICS, one engineered to detect narrow therefore capitalize on spectral variant of the SPLICS S/L (≈8–10 nm) and a second to image wide (≈40–50 nm) probes to verify if the wide interactions comprise the narrow ER–mitochondria interactions [5,13]. Although the low ones, or if they occur on different areas of the organelles. dissociation rate of the GFP fragments [23] could, upon CCCP treatment in HeLa cells expressing Parkin caused reconstitution, make their association poorly reversible a marked reduction in both SPLICS and SPLICS signal, S L ER-mitochondria contact sites sensor 1139 cells; Parkin CCCP 65 ± 7, n = 26 cells. c Representative confocal Fig. 5 Effects of Parkin on ER–mitochondria contacts. Immuno- pictures of HeLa cells expressing the SPLICS probe along with fluorescence against mitochondria (Tom20, cyan) is shown in the panels Parkin-2A-mCherry-CAAX (bottom panels). d Quantification of the on the middle. The green channel is the merge of several planes. Scale bar SPLICS contacts by 3D rendering of complete z-stacks. Mean ± 20 µm. a Representative confocal pictures of HeLa cells expressing the SEM: CTRL 227 ± 9, n = 44 cells; Parkin 193 ± 11, n = 32 cells; SPLICS probe along with Parkin-2A-mCherry-CAAX (bottom panels). Parkin CCCP 159 ± 8, n = 22 cells. Data shown are the result of 3–8 b Quantification of SPLICS contacts by 3D rendering of complete z- independent experiments. *p ≤ 0.05, **p ≤ 0.01, ****p ≤ 0.0001. stacks. Mean ± SEM: Ctrl 61 ± 3, n= 57 cells; Parkin 94 ± 6, n= 33 supporting a model whereby mitochondria and ER separate in this process. SPLICS is a versatile tool to visualize during mitophagy. Under basal conditions, overexpression ER–mitochondria contacts in vivo both in physiological and of Parkin conversely increases ER–mitochondria tethering pathological conditions and it will be useful to explore how in several cellular models [42,43], suggesting a house- disease-related genes affect neuronal function and survival keeping function at this interface, whereas massive Parkin through the modulation of the ER–mitochondria interface. activation and mitophagy induction mirrors a condition of degradation/deactivation of a handful of molecules involved in ER–mitochondria juxtaposition (e.g., VDAC, Materials and methods Mfn2, TOM20) [53] with consequent ER–mitochondria separation. Cell Lines Despite the efforts in the analysis of the ER–mitochondria contact sites, their properties in neurons and, most impor- HeLa and HEK293 cells (ATCC) were grown in a 5% tantly, in living vertebrates are still poorly characterized due CO atmosphere in Dulbecco’s modified Eagle’smedium to the lack of appropriate methods suitable for in vivo stu- high glucose (DMEM; Euroclone), supplemented with dies. We optimized the SPLICS for in vivo expression by 10% fetal bovine serum (Gibco), 100 U/ml penicillin and generating the pT2-DsRed-UAS-SPLICS -2A construct. 100 mg/ml streptomycin. Where indicated, the cells were SPLICS could report ER–mitochondria contacts in cell body treated 48 h after transfection with 10 µM CCCP (Sigma- and axons of RB neurons in living zebrafish embryos. Aldrich), Hbss (Thermo Fisher) or 10 µg/ml Tunicamycin ER–mitochondria contacts in the axons often localized at (Sigma-Aldrich) for 4 h at 37 °C, in a 5% CO atmo- sites of branching, a process sustained by mitochondrial sphere. Mock cells were maintained in growth medium, 2+ ATP [54] and Ca release from internal stores [55], thus which was changed simultaneously with the beginning of suggesting that ER–mitochondria interplay may have a role treatments. Female FAD-PS2-N141I fibroblasts (Coriell 1140 D. Cieri et al. Fig. 6 Expression of the SPLICS probe in living zebrafish embryos. a Schematic depiction of the bidirectional construct that allows detection of DsRed and SPLICS -P2A in a Gal4-dependent manner. The 2A peptide guarantees the generation of an equimolar amount of the two spGFP fragments. b Experimental setting used to image ER–mitochondria contacts in zebrafish embryos: a schematic drawing of RB neurons is shown on the right. c Representative image of a 1 dpf s1102t:GAL4 embryo injected with the pT2- DsRed-UAS-SPLICS -P2A construct. d Live imaging of short ER–mitochondria contacts in RB neurons of s1102t:GAL4 zebrafish embryos. The picture is the merge of several planes. The 3D rendering of the z-stack is shown on the right. Scale bar: 15 µm. e Quantification of SPLICS contacts in the cell body of RB neurons by 3D rendering of complete z-stacks. Mean ± SEM: 22 ± 1, n = 28 cells from 11 fish. f Live imaging of SPLICS contacts in the axons of RB neurons. The picture is the merge of several planes. The 3D rendering of the complete z-stack is shown on the right. Scale bar: 15 µm. g Quantification of the density of SPLICS contacts in the cell body and the axons of RB neurons. Mean ± SEM: RB soma: 0.17 ± 0.01, n = 28 cells from 11 fish; RB axon: 0.14 ± 0.01, n = 20 cells from 6 fish. Data shown are the result of two independent experiments. Institute, AG09908) and age/sex-matched control fibro- Rohon-Beard neurons (s1102t:GAL4), Et(E1b:GAL4- blasts (Coriell Institute, AG08269) were grown in DMEM VP16)s1102t; Tg(UAS-E1b:Kaede)s1999t fish (ZIRC, containing FCS (15%), L-glutamine (2 mM), penicillin ZL1384) were outcrossed with wild-type (wt) fish [57] and (100 U/ml, Euroclone), and streptomycin (100 µg/ml, the fluorescent offspring was discarded. The remaining fish Euroclone). were genotyped from fin clips with primers specific for GAL4. To perform experiments, both wt and s1102t:GAL4 Zebrafish husbandry and transgenic lines fish were used. All experiments were conducted on 24 h post fertilization (hpf) embryos. All animal experiments were conducted as previously reported [56]. Adult fish were maintained and raised in 5 l Transfection tanks with freshwater at 28 °C with a 12 h light/12 h dark cycle. Zebrafish embryos were obtained from spontaneous Twelve hours before transfection, HeLa cells were seeded spawnings. To obtain fish selectively expressing Gal4 in onto 13 mm glass coverslips and allowed to grow to 50% ER-mitochondria contact sites sensor 1141 confluence. Cells were transfected by calcium phosphate Myc were kindly provided by Professor Christopher C.J. [58]. For co-transfection, the two SPLICS ER and mito- Miller (Department of Neuroscience and Department of chondrial fragments were in a 1.5:2 ratio with the over- Basic and Clinical Neurosciences, King’s College London). expressed protein of interest. Human fibroblasts were Where indicated, mitochondria are labeled with a pTagRFP- electroporated by using a NeonTM transfection system mito construct (Evrogen). To knockdown Mfn2, three dif- (Life Technologies), according to the manufacturer's ferent shRNAs against Mfn2 were used (SureSilencing instruction. shRNA Plasmid, Hygromycin Gene: Mfn2; Refseq Acces- sion #:NM_014874; Clone #1: AGAGGCGGTTCGACT- Cloning and fusion plasmid construction CATCAT; Clone #3: TGATGTG GCCCAACTCTAAGT; Clone #4: CCAGTAGTCCTCAA GGTTTAT; Scramble: Humanized GFP (GFP ) was PCR amplified from GGAATCTCATTCGATGCATA C). The aminoacid 1–10 1–10(h) the GPI-GFP kindly provided by Prof. Fabien Pinaud, sequences of the ER anchored probes are: 1–10(h) Department of Biological Sciences and Chemistry, Uni- MRDHMVLHEYVNAAGITGGDGGSGGGSKLRVF versity of Southern California [59], to insert the Tom20 N33 LALPIIMVVAFSMCIICLLMAGDTWTETLAYVLFWG targeting sequence [60] by using the OMM-GFP For. VASIGTFFIILYNGKDFVDAPRLVQKEKID (the Short) 1–10 (TCGAATTCATGGTGGGCCGGAACAGCGCCATCGC and MRDHMVLHEYVNAAGITGGDGGSGGGSKLMW CGCGGGCGTGTGCGGTGCCCTCTTCATAGGGTACT HEGLEEASRLYF GERNVKGMFEVLEPLHAMMERGP GCATCTACTTTGACCGCAAAAGGCGGAGTGACCC QTLKETSFNQAYGRDLMEAQEWCRKYMKSGNVKD CAACTCCAAAGGAGAAG) and Rev. (AC TTCTCACT LTQAWDLYYHVFRRISKQGSEAAAREAAARGGASG CGAGTTATGTTCCTTTTTCATTTGGATCTTTGCTCA AGAGAGAILNSRVFLALPIIMVVAFSMCIICLLMAGD GG) primers and inserted into the vector pcDNA3 between TWTETLAYVLFWGVASIGTFFIILYNGKDFVDAPRLV the EcoRV and XhoI restriction sites. The chimeric QKEKID (the Long). sequences composed by the minimal Sac1 ER targeting sequence and the Sac1 ER targeting sequence containing Zebrafish constructs additional 267 bp of the helix-FRB fragment (derived from the pEGFP-C3/CFP-HA-FRB-helix-ER plasmid [5]) (ER - To properly express SPLICS in zebrafish, the ER -β and L S 11 β ) were fused with the β tag to generate the ER -β and the OMM-GFP coding sequences were respectively 11 11 s 11 1–10 the ER -β constructs, respectively. The constructs were cloned upstream and downstream of a viral 2A peptide L 11 chemically synthetized (Thermo Fisher) by using the sequence (pSYC-181, Addgene), previously reported to be EcoRI/NotI and the EcoRV/XhoI restriction sites of cleaved within the cell to generate an equimolar amount of pCDNA3.1(+), respectively. The sequence encoding for the two genes [48]. The fragment encoding for the OMM- β , flanked by different multi cloning sites, has been GFP was excised from pcDNA3 with EcoRI and XhoI 11 1–10 inserted into the commercial vectors pCDNA6.5/V5-DEST and ligated into the pSYC-181 vector digested with the (Invitrogen), in order to obtain a backbone that could easily same enzymes. Since these sites cloned the OMM-GFP 1–10 allow fusion of the β fragment with the protein of interest. fragment out of frame, we mutagenized the resulting vector Kate was amplified with the Kate For. (AAAAAAGCT- in order to re-establish the correct frame, using the primer TATGGTGAGCGAGC) and the Kate Rev. (TTTTTG OMM-GFP mut. (CCCTGGACCTAGATCTGAATTC 1–10 GTACCTCATCTGTGCC) primers using as a template ATGGTGGGCC). ER was amplified from pcDNA3 using mKate2-pcDNA3.1 (pEVROGEN) and inserted in pDEST- the primers ER -β 2A For. (ACGCGGATCCGCCAC- S 11 β to generate Kate β . To generate a vector simulta- CATGCGGGACCACATGGTG C) and Rev. (TTCCCCCG 11 11 neously expressing (through a 2A peptide) Parkin and GGGTCGATCTTCTCTTT), which introduce, respectively, membrane mCherry (CAAX-mCherry), the coding a BamHI-Kozak sequence and a SmaI site at the 5′ and 3′ sequence of Parkin [42] was amplified with the primers ends of the ER coding sequence. The PCR product was Parkin For. (ACGCGGATCCGCCACCATGATAGTGTT purified, digested with both BamHI and SmaI and then TGTCAGGTT) and Rev. (TTCCCCCGGGCA CGTCGAA ligated into the pSYC-181-OMM-GFP vector, pre- 1–10 CCAGTGGTCC). The PCR products were purified using viously digested with the same restriction enzymes. We the GenElute Gel Extraction Kit (Sigma), digested with refer to this plasmid as pSYC-SPLICS -P2A. To selectively BamHI and SmaI and then ligated into the pSYC-187 vector express the SPLICS -P2A in zebrafish neurons, we (Addgene) digested with the same restriction enzymes. The exploited the Gateway technology to generate the pT2- constructs codifying for Drp1 WT and K38A [30,61,62] DsRed-UAS-SPLICS -P2A vector. Briefly, the fragment and pmTurquoise2-ER were kindly donated by Diego De encoding for the SPLICS -P2A was excised from the Stefani (Department of Biomedical Sciences, University of pSYC-SPLICS -P2A vector with BamHI and XbaI and Padova). The pCIneo-PTPIP51-HA and the pCIneo-VAPB- cloned into the BamHI-XbaI sites of the pME-MCS vector 1142 D. Cieri et al. (Tol2 kit). The resulting pME-SPLICS -P2A was recom- overnight in 2.3 M sucrose, mounted on aluminum pins, and bined through the LR reaction with the pT2dDESTpADs frozen in liquid nitrogen. Ultrathin cryosections of 70 nm Red.T4E1bUASE1bGWR1R2pA vector (kindly donated by were cut with a Leica FC7 ultramicrotome (Leica Micro- Christian Haas, Ludwig-Maximilians University Munich). systems, Germany), collected on formvar, carbon-coated 150 The resulting pT2-DsRed-UAS-SPLICS -P2A plasmid was mesh copper grids and immunolabeled with anti-GFP injected into 1–2 cells stage s1102t:GAL4 embryos. To (Abcam, Cat#ab290) and 10 nm Protein A-gold (Utrecht label mitochondria and the endoplasmic reticulum, the University, The Netherlands). Grids were then contrasted for pTagRFP-mito (Evrogen) and the pDsRed2-ER (Clontech) 10 min on ice in a solution of 1.8% methylcellusose/0.4% plasmids were, respectively, injected in WT eggs. For uranyl acetate, air dried on wire loops and observed in a injections, all plasmids were diluted in Danieau solution ZEISS Leo912AB (Zeiss, Oberkochen, Germany). Images (58 mM NaCl, 0.7 mM KCl, 0.4 mM MgSO , 0.6 mM Ca were acquired using a 2Kx2K bottom mounted slow-scan (NO ) , 5 mM HEPES pH 7.6) and 0.5% phenol red. Proscan camera (Scheuring, Germany) controlled by the 3 2 EsivisionPro 3.2 (Soft Imaging System, Münster, Germany). Immunocytochemistry Image acquisition and processing To image cells over-expressing a protein of interest, trans- fected cells plated on 13 mm glass coverslips were fixed Cells were generally imaged 48–72 h after transfection with 48–72 h post transfection with 3.7% formaldehyde in a Leica TSC SP5 inverted confocal microscope, using either phosphate-buffered saline (PBS; 140 mM NaCl, 2 mM KCl, a HCX PL APO 63X/numerical aperture 1.40–0.60 or a 1.5 mM KH PO ,8mM Na HPO ,pH 7.4) for 20minand HCX PL APO ×100/numerical aperture 1.4 oil-immersion 2 4 2 4 washed three times with PBS. Cell permeabilization was objective. Images were acquired by using the Leica AS performed by 20 min incubation in 0.1% Triton X-100/PBS software. To count ER–mitochondria contacts, a complete followed by 30 min wash in 1% gelatin/PBS (type IV, from z-stack of the cell was acquired every 0.29 µm. Z-stacks bovine skin, Sigma) and 15 min wash in PBS at room tem- were processed using Fiji [64]: images were first convolved, perature (RT). The coverslips were then incubated for 90 min and then filtered using the Gaussian Blur filter. A 3D at 37 °C with the specific primary antibody diluted 1:20 in reconstruction of the resulting image was obtained using the PBS (Tom20: Santa Cruz Biotech., Cat#sc-11415; Parkin: Volume J plugin (http://bij.isi.uu.nl/vr.htm). A selected face Santa Cruz Biotech., Cat#sc-32282; Drp1: BD Biosciences, of the 3D rendering was then thresholded and used to count Cat#611113; mtHSP60: Abcam, Cat#ab82520; Calreticulin: ER–mitochondria contact sites. Thermo Fisher, Cat#PA3-900; Myc: Millipore, Cat#05-724; HA: Cell Signalling, Cat#3724S). Further washing steps with Zebrafish imaging gelatine and PBS were repeated as mentioned before to remove the excess of primary antibody. Staining was revealed At 24 hpf, embryos were screened for fluorescence, by the incubation with specific AlexaFluor secondary anti- dechorionated and fixed or anesthetised according to the bodies (Thermo Fisher: Goat anti-Rabbit IgG AlexaFluor 405, experiment. To image the co-localization of ER -β and S 11 Cat#A-31556; Goat anti-Rabbit IgG AlexaFluor 594, Cat#A- OMM-GFP with mitochondria and the ER, fish were 1–10 11012; Donkey anti-Goat IgG AlexaFluor 633, Cat#A-21082; fixed for 2 h at RT with 4% PFA, then washed with PBS Goat anti-Mouse IgG AlexaFluor 633, Cat#A-21050; Goat and mounted in low melting agarose (1.3%, Euroclone) on anti-Mouse IgG AlexaFluor 488, Cat#A-11001; Goat anti- glass coverslips. Rabbit IgG AlexaFluor 488, Cat#A-11008) for 45 min at RT For in vivo imaging, fish were anesthetised and mounted (1:50 dilution in PBS; 1:20 only for Goat anti-Rabbit IgG on 35 × 10 mm glass bottom Petri dishes (Ted Pella, INC. AlexaFluor 405). After further washing steps, coverslips were Prod. No. 14023-20) in low melting agarose (1.3%, Euro- mounted using Mowiol 4–88 (Sigma). The coverslips were Clone). Fish water containing tricaine methanesulfonate observed at the SP5 Leica confocal microscope at lasers 0.61 mM (Sigma) was added in the Petri dishes, in order to wavelength of 405, 458, 488, 543, 555, and 633 nm. keep fish anesthetised. Mounted fish were imaged at RT (20–23 °C) using a Leica TSC SP5 inverted confocal Electron microscopy microscope, using either a HCX PL APO ×63/numerical aperture 1.40–0.60 or a HCX PL APO ×100/numerical Cells were immunogold labeled by the Tokuyasu technique aperture 1.4 oil-immersion objective. To count ER-mito [63]. Briefly, cells were fixed in 2% paraformaldehyde, 0.2% contacts, a complete z-stack of the cell was acquired every glutaraldehyde in 0.1 M phosphate buffer (PB) for 1 h at RT. 0.29 µm. To acquire a representative image of a whole fish Next cells were gently scraped in 1% gelatin and embedded expressing pT2-DsRed-UAS-SPLICS -P2A, a ×10 HCPX in 12% gelatin in PB. Gelatin squared blocks were infiltrated PL Fluotar NA 0.3 objective was used. ER-mitochondria contact sites sensor 1143 2+ Ca measurements Statistical analysis 2+ Ca measurements were performed by co-transfecting Results shown are mean values ± SEM. Student’s unpaired HeLa cells in a six-well plate with low-affinity mitochon- two-tailed t-test was used for comparisons involving two drial aequorin (mtAeqmut) with the two SPLICS moieties groups when sample followed a Gaussian distribution, in a 1.5:1.5:1 ratio favouring SPLICS. Forty-eight hours otherwise Mann–Whitney test was used. Differences post transfection, cells were re-plated into a 96-wells plate between groups were considered significant when p ≤ 0.05. (PerkinElmer). mtAeqmut was reconstituted by incubating All statistical analyses were performed using GraphPad cells for1.5hwith5 µM coelenterazine (Santa Cruz Bio- Prism version 6.00 for Mac OS X, GraphPad Software (La tech.) in modified Krebs Ringer Buffer (KRB: 125 mM Jolla, California, USA). The exact values of n and their NaCl, 5 mM KCl, 400 mM KH PO ,1mM MgSO ,20 means are indicated in the figure legends. *p ≤ 0.05, **p ≤ 2 4 4 mM Hepes, pH 7.4) supplemented with 5 mM glucose at 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. 37 °C. Luminescence measurements were carried out using a PerkinElmer EnVision plate reader equipped with two Acknowledgements We are deeply indebted with Prof. F. Pinaud (University of Southern California) for kindly providing the humanized injector units. After reconstitution, cells were placed in 70 GFP construct. We thank Prof. G. Hajnoczky (Thomas Jefferson 1–10 µl of KRB solution and luminescence from each well was University, Philadelphia) for kindly providing the plasmid pEGFP-C3/ measured for 1 min. During the experiment, 100 µM his- CFP-HA-FRB-helix-ER. We thank Prof. C. C.J. Miller (King’s College tamine at the final concentration were first injected to London) for kindly providing the PTPIP51-HA and the VAPB-Myc 2+ 2+ expression vectors. We thank Prof. R. Rizzuto (University of Padova), activate Ca transients, and then a hypotonic, Ca -rich, Dr. D. De Stefani (University of Padova) and Prof. G. Szabadkai digitonin-containing solution was added to discharge the (University College London) for helpful discussions, we also remaining aequorin pool. Output data were analyzed and thank Prof. F. Argenton and the Zebrafish Facility of the Department of calibrated with a custom-made macro-enabled Excel Biology, University of Padova and Dr. A. Raimondi (Advanced Light and Electron Microscopy BioImaging Center ALEMBIC, San Raffaele workbook. Scientific Institute). This research was supported by grants from the Ministry of University and Research (Bando SIR 2014 no. Western blot RBSI14C65Z to T.C.; FIRB RBAP11Z3YA_005 to L.S.), from the University of Padova (Progetto Giovani 2012 no. GRIC128SP0 to T.C., Progetto di Ateneo 2016 no. CALI_SID16_01 to T.C., Progetto di HeLa cells were seeded in a six-well plate and transfected Ateneo 2015 no. CPDA153402 to M.B.), from Cariparo Starting Grant with three different shRNAs against Mfn2. A scramble 2016 AIFbiol to M.G., from the EU Joint Programme in Neurode- shRNA was used as control. At 48 h post transfection, generative Disease (2015–2018, “Cellular Bioenergetics in Neurode- cells were washed with PBS and proteins were extracted generative Diseases: A System-Based Pathway and Target Analysis”)to P.P., from the European Union (ERC FP7-282280, FP7 CIG PCIG13- for 20 min using ice cold lysis buffer (50 mM Tris-HCl pH GA-2013-618697 to L.S.) and from Telethon (GGP02016 to L.S.). 7.4, 150 mM NaCl, 10 mM EGTA, 1% Triton X-100, 1 mM protease inhibitor cocktail (Sigma)). Samples were then centrifuged at 10,000 rpm for 10 min at 4 °C. Protein Compliance with ethical standards concentration was determined by the Bradford assay (Bio- Conflict of interest The authors declare that they have no conflict of Rad). Samples were separated on a 4–15% Mini- interest. PROTEAN TGX™ Precast Protein Gels (Bio-Rad) and blotted using a Immobilon-PSQ PVDF Membrane (Merck Open Access This article is licensed under a Creative Commons Millipore). The membrane was blocked for 1 h at RT using Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as 5% non-fat dried milk in TBST (20 mM Tris-HCl, pH 7.4, long as you give appropriate credit to the original author(s) and the 150 mM NaCl, 0.05% Tween-20) and incubated with pri- source, provide a link to the Creative Commons license, and indicate if mary antibodies (Mfn2: Abcam, Cat#ab50838, 1:1000 in changes were made. The images or other third party material in this TBST, overnight at 4 °C; β-actin: Sigma, Cat#A5441, article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not 1:30,000 in TBST, 2 h at RT; Myc: Millipore, Cat#05-724, included in the article’s Creative Commons license and your intended 1:1000 in TBST, overnight at 4 °C; HA: Cell Signalling, use is not permitted by statutory regulation or exceeds the permitted Cat#3724 S, 1:1000 in TBST, overnight at 4 °C). After use, you will need to obtain permission directly from the copyright three washing steps in TBST, detection was obtained by holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. incubating the membrane with secondary horseradish peroxidase-conjugated antibodies (Santa Cruz Biotech.: Goat anti-Rabbit IgG-HRP, Cat#sc-2004; Goat anti-Mouse References IgG-HRP, Cat#sc-2005; 1:2000 in TBST) for 1 h at RT and byincubationwiththe Luminata HRPsubstrate 1. Prinz WA. Bridging the gap: membrane contact sites in signaling, (Merck Millipore). metabolism, and organelle dynamics. J Cell Biol. 2014;205:759–69. 1144 D. Cieri et al. 2. Eisenberg-Bord M, Shai N, Schuldiner M, Bohnert M. A tether is revealed by a synthetic biology screen. Science. 2009;325: a tether is a tether: tethering at membrane contact sites. Dev 477–81. Cell.2016;39:395–409. 22. Wang PT, Garcin PO, Fu M, Masoudi M, St-Pierre P, Pante N, 3. Murley A, Nunnari J. The emerging network of mitochondria- et al. Distinct mechanisms controlling rough and smooth endo- organelle contacts. Mol Cell.2016;61:648–53. plasmic reticulum contacts with mitochondria. J Cell Sci. 4. Giacomello M, Pellegrini L. The coming of age of the 2015;128:2759–65. mitochondria-ER contact: a matter of thickness. Cell Death Differ. 23. Cabantous S, Terwilliger TC, Waldo GS. Protein tagging and 2016;23:1417–27. detection with engineered self-assembling fragments of green 5. Csordas G, Renken C, Varnai P, Walter L, Weaver D, Buttle KF, fluorescent protein. Nat Biotechnol. 2005;23:102–7. et al. Structural and functional features and significance of the 24. Pedelacq JD, Cabantous S, Tran T, Terwilliger TC, Waldo GS. physical linkage between ER and mitochondria. J Cell Biol. Engineering and characterization of a superfolder green fluor- 2006;174:915–21. escent protein. Nat Biotechnol. 2006;24:79–88. 6. Tubbs E, Rieusset J. Study of endoplasmic reticulum and mito- 25. Cabantous S, Waldo GS. In vivo and in vitro protein solubility chondria interactions by in situ proximity ligation assay in fixed assays using split GFP. Nat Methods. 2006;3:845–54. cells. J Vis Exp. 2016;118. doi: 10.3791/54899. PubMed PMID: 26. Bravo R, Vicencio JM, Parra V, Troncoso R, Munoz JP, Bui M, 28060261. et al. Increased ER-mitochondrial coupling promotes mitochon- 7. Gomez-Suaga P, Paillusson S, Stoica R, Noble W, Hanger DP, drial respiration and bioenergetics during early phases of ER Miller CC. The ER-mitochondria tethering complex VAPB- stress. J Cell Sci.2011;124(Pt 13):2143–52. PTPIP51 regulates autophagy. Curr Biol. 2017;27:371–85. 27. Bravo-Sagua R, Lopez-Crisosto C, Parra V, Rodriguez-Pena M, 8. Stoica R, Paillusson S, Gomez-Suaga P, Mitchell JC, Lau DH, Rothermel BA, Quest AF, et al. mTORC1 inhibitor rapamycin and Gray EH, et al. ALS/FTD-associated FUS activates GSK-3beta to ER stressor tunicamycin induce differential patterns of ER- disrupt the VAPB-PTPIP51 interaction and ER-mitochondria mitochondria coupling. Sci Rep. 2016;6:36394. associations. EMBO Rep. 2016;17:1326–42. 28. Rambold AS, Kostelecky B, Elia N, Lippincott-Schwartz J. 9. Paillusson S, Gomez-Suaga P, Stoica R, Little D, Gissen P, Tubular network formation protects mitochondria from autopha- Devine MJ, et al. alpha-Synuclein binds to the ER-mitochondria gosomal degradation during nutrient starvation. Proc Natl Acad tethering protein VAPB to disrupt Ca2+ homeostasis Sci USA.2011;108:10190–5. and mitochondrial ATP production. Acta Neuropathol. 29. Gomes LC, Di Benedetto G, Scorrano L. During autophagy 2017;134:129–49. mitochondria elongate, are spared from degradation and sustain 10. Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz cell viability. Nat Cell Biol. 2011;13:589–98. LM, et al. Close contacts with the endoplasmic reticulum as 30. Szabadkai G, Simoni AM, Chami M, Wieckowski MR, Youle RJ, determinants of mitochondrial Ca2+ responses. Science. 1998; Rizzuto R. Drp-1-dependent division of the mitochondrial net- 280:1763–6. work blocks intraorganellar Ca2+ waves and protects against Ca2 11. de Brito OM, Scorrano L. Mitofusin 2 tethers endoplasmic reti- + -mediated apoptosis. Mol Cell. 2004;16:59–68. culum to mitochondria. Nature. 2008;456:605–10. 31. Hamasaki M, Furuta N, Matsuda A, Nezu A, Yamamoto A, Fujita 12. Naon D, Zaninello M, Giacomello M, Varanita T, Grespi F, N, et al. Autophagosomes form at ER-mitochondria contact sites. Lakshminaranayan S, et al. Critical reappraisal confirms that Nature. 2013;495:389–93. Mitofusin 2 is an endoplasmic reticulum-mitochondria tether. Proc 32. Sugiura A, Nagashima S, Tokuyama T, Amo T, Matsuki Y, Ishido Natl Acad Sci USA. 2016;113:11249–54. S, et al. MITOL regulates endoplasmic reticulum-mitochondria 13. Filadi R, Greotti E, Turacchio G, Luini A, Pozzan T, Pizzo P. contacts via Mitofusin2. Mol Cell. 2013;51:20–34. Mitofusin 2 ablation increases endoplasmic reticulum- 33. Chen Y, Csordas G, Jowdy C, Schneider TG, Csordas N, Wang mitochondria coupling. Proc Natl Acad Sci USA. 2015;112: W, et al. Mitofusin 2-containing mitochondrial-reticular micro- E2174–81. domains direct rapid cardiomyocyte bioenergetic responses via 14. Alford SC, Ding Y, Simmen T, Campbell RE. Dimerization- interorganelle Ca(2+) crosstalk. Circ Res. 2012;111:863–75. dependent green and yellow fluorescent proteins. ACS Synth Biol. 34. Schneeberger M, Dietrich MO, Sebastian D, Imbernon M, Cas- 2012;1:569–75. tano C, Garcia A, et al. Mitofusin 2 in POMC neurons connects 15. Toulmay A, Prinz WA, A conserved membrane-binding domain ER stress with leptin resistance and energy imbalance. Cell. targets proteins to organelle contact sites. J Cell Sci.2012;125(Pt 2013;155:172–87. 1):49–58. 35. Li D, Li X, Guan Y, Guo X. Mitofusin-2-mediated tethering of 16. MacDonald ML, Lamerdin J, Owens S, Keon BH, Bilter GK, mitochondria and endoplasmic reticulum promotes cell cycle Shang Z, et al. Identifying off-target effects and hidden pheno- arrest of vascular smooth muscle cells in G0/G1 phase. Acta types of drugs in human cells. Nat Chem Biol. 2006;2:329–37. Biochim Biophys Sin (Shanghai). 2015;47:441–50. 17. Csordas G, Varnai P, Golenar T, Roy S, Purkins G, Schneider TG, 36. Cosson P, Marchetti A, Ravazzola M, Orci L. Mitofusin-2 inde- et al. Imaging interorganelle contacts and local calcium dynamics pendent juxtaposition of endoplasmic reticulum and mitochondria: at the ER-mitochondrial interface. Mol Cell. 2010;39:121–32. an ultrastructural study. PLoS ONE 2012;7:e46293. 18. Ravikumar B, Duden R, Rubinsztein DC. Aggregate-prone pro- 37. Leal NS, Schreiner B, Pinho CM, Filadi R, Wiehager B, Karl- teins with polyglutamine and polyalanine expansions are degraded strom H, et al. Mitofusin-2 knockdown increases ER- by autophagy. Hum Mol Genet. 2002;11:1107–17. mitochondria contact and decreases amyloid beta-peptide pro- 19. Ravikumar B, Vacher C, Berger Z, Davies JE, Luo S, Oroz LG, duction. J Cell Mol Med 2016;20:1686–95. et al. Inhibition of mTOR induces autophagy and reduces toxicity 38. Filadi R, Greotti E, Turacchio G, Luini A, Pozzan T, Pizzo P. of polyglutamine expansions in fly and mouse models of Hun- Presenilin 2 modulates endoplasmic reticulum-mitochondria cou- tington disease. Nat Genet. 2004;36:585–95. pling by tuning the antagonistic effect of Mitofusin 2. Cell 20. Noda T, Ohsumi Y. Tor, a phosphatidylinositol kinase homo- Rep.2016;15:2226–38. logue, controls autophagy in yeast. J Biol Chem. 1998;273: 39. Zampese E, Fasolato C, Kipanyula MJ, Bortolozzi M, Pozzan T, 3963–6. Pizzo P. Presenilin 2 modulates endoplasmic reticulum (ER)- 21. Kornmann B, Currie E, Collins SR, Schuldiner M, Nunnari J, mitochondria interactions and Ca2+ cross-talk. Proc Natl Acad Weissman JS, et al. An ER-mitochondria tethering complex Sci USA. 2011;108:2777–82. ER-mitochondria contact sites sensor 1145 40. De Vos KJ, Morotz GM, Stoica R, Tudor EL, Lau KF, Ackerley 52. Twig G, Elorza A, Molina AJ, Mohamed H, Wikstrom JD, Walzer S, et al. VAPB interacts with the mitochondrial protein PTPIP51 G, et al. Fission and selective fusion govern mitochondrial segre- to regulate calcium homeostasis. Hum Mol Genet. 2012;21: gation and elimination by autophagy. EMBO J. 2008;27:433–46. 1299–311. 53. Sarraf SA, Raman M, Guarani-Pereira V, Sowa ME, Huttlin EL, 41. Eiyama A, Okamoto K. PINK1/Parkin-mediated mitophagy in Gygi SP, et al. Landscape of the PARKIN-dependent ubiquity- mammalian cells. Curr Opin Cell Biol. 2015;33:95–101. lome in response to mitochondrial depolarization. Nature. 42. Calì T, Ottolini D, Negro A, Brini M. Enhanced parkin levels 2013;496:372–6. 2+ favour ER-mitochondria crosstalk and guarantee Ca transfer to 54. Spillane M, Ketschek A, Merianda TT, Twiss JL, Gallo G. sustain cell bioenergetics. Biochim Biophys Acta.2013;1832: Mitochondria coordinate sites of axon branching through localized 495–508. intra-axonal protein synthesis. Cell Rep. 2013;5:1564–75. 43. Zheng L, Bernard-Marissal N, Moullan N, D’Amico D, Auwerx J, 55. Tang F, Kalil K. Netrin-1 induces axon branching in developing Moore DJ, et al. Parkin functionally interacts with PGC-1alpha to cortical neurons by frequency-dependent calcium signaling path- preserve mitochondria and protect dopaminergic neurons. Hum ways. J Neurosci. 2005;25:6702–15. Mol Genet. 2017;26:582–98. 56. Bergamin G, Cieri D, Vazza G, Argenton F, Mostacciuolo ML. 44. Gautier CA, Erpapazoglou Z, Mouton-Liger F, Muriel MP, Cor- Zebrafish Tg(hb9: MTS-Kaede): a new in vivo tool for studying mier F, Bigou S, et al. The endoplasmic reticulum-mitochondria the axonal movement of mitochondria. Biochim Biophys Acta. interface is perturbed in PARK2 knockout mice and patients with 2016;1860:1247–55. PARK2 mutations. Hum Mol Genet. 2016;25:2972–84. 57. Pauls S, Zecchin E, Tiso N, Bortolussi M, Argenton F. Function 45. Ryan MD, King AM, Thomas GP, Cleavage of foot-and-mouth and regulation of zebrafishnkx2.2a during development of pan- disease virus polyprotein is mediated by residues located within a creatic islet and ducts. Dev Biol. 2007;304:875–90. 19 amino acid sequence. J Gen Virol. 1991;72(Pt 11):2727–32. 58. Ottolini D, Cali T, Brini M. Methods to measure intracellular Ca(2 46. Denison SR, Wang F, Becker NA, Schule B, Kock N, Phillips LA, +) fluxes with organelle-targeted aequorin-based probes. Methods et al. Alterations in the common fragile site gene Parkin in ovarian Enzymol. 2014;543:21–45. and other cancers. Oncogene. 2003;22:8370–8. 59. Pinaud F, Dahan M. Targeting and imaging single biomolecules in 47. Pawlyk AC, Giasson BI, Sampathu DM, Perez FA, Lim KL, living cells by complementation-activated light microscopy with Dawson VL, et al. Novel monoclonal antibodies demonstrate split-fluorescent proteins. Proc Natl Acad Sci USA. 2011;108: biochemical variation of brain parkin with age. J Biol Chem. E201–10. 2003;278:48120–8. 60. Giacomello M, Drago I, Bortolozzi M, Scorzeto M, Gianelle A, 48. Kim JH, Lee SR, Li LH, Park HJ, Park JH, Lee KY, et al. High Pizzo P, et al. Ca2+ hot spots on the mitochondrial surface are cleavage efficiency of a 2A peptide derived from porcine generated by Ca2+ mobilization from stores, but not by activation teschovirus-1 in human cell lines, zebrafish and mice. PLoS ONE. of store-operated Ca2+ channels. Mol Cell. 2010;38:280–90. 2011;6:e18556. 61. Smirnova E, Shurland DL, Ryazantsev SN, van der Bliek AM. A 49. Bannai H, Inoue T, Nakayama T, Hattori M, Mikoshiba K, human dynamin-related protein controls the distribution of mito- Kinesin dependent, rapid, bi-directional transport of ER sub- chondria. J Cell Biol. 1998;143:351–8. compartment in dendrites of hippocampal neurons. J Cell 62. Frank S, Gaume B, Bergmann-Leitner ES, Leitner WW, Robert Sci.2004;117(Pt 2):163–75. EG, Catez F, et al. The role of dynamin-related protein 1, a 50. Villegas R, Martinez NW, Lillo J, Pihan P, Hernandez D, Twiss mediator of mitochondrial fission, in apoptosis. Dev Cell. JL, et al. Calcium release from intra-axonal endoplasmic reticulum 2001;1:515–25. leads to axon degeneration through mitochondrial dysfunction. J 63. Tokuyasu KT. A technique for ultracryotomy of cell suspensions Neurosci.2014;34:7179–89. and tissues. J Cell Biol. 1973;57:551–65. 51. Cui-Wang T, Hanus C, Cui T, Helton T, Bourne J, Watson D, 64. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, et al. Local zones of endoplasmic reticulum complexity confine Pietzsch T, et al. Fiji: an open-source platform for biological- cargo in neuronal dendrites. Cell. 2012;148:309–21. image analysis. Nat Methods. 2012;9:676–82. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Death & Differentiation Springer Journals

SPLICS: a split green fluorescent protein-based contact site sensor for narrow and wide heterotypic organelle juxtaposition

Free
15 pages
Loading next page...
 
/lp/springer_journal/splics-a-split-green-fluorescent-protein-based-contact-site-sensor-for-z1QMR5LE06
Publisher
Nature Publishing Group UK
Copyright
Copyright © 2017 by ADMC Associazione Differenziamento e Morte Cellulare
Subject
Life Sciences; Life Sciences, general; Biochemistry, general; Cell Biology; Stem Cells; Apoptosis; Cell Cycle Analysis
ISSN
1350-9047
eISSN
1476-5403
D.O.I.
10.1038/s41418-017-0033-z
Publisher site
See Article on Publisher Site

Abstract

Contact sites are discrete areas of organelle proximity that coordinate essential physiological processes across membranes, 2+ including Ca signaling, lipid biosynthesis, apoptosis, and autophagy. However, tools to easily image inter-organelle proximity over a range of distances in living cells and in vivo are lacking. Here we report a split-GFP-based contact site sensor (SPLICS) engineered to fluoresce when organelles are in proximity. Two SPLICS versions efficiently measured narrow (8–10 nm) and wide (40–50 nm) juxtapositions between endoplasmic reticulum and mitochondria, documenting the existence of at least two types of contact sites in human cells. Narrow and wide ER–mitochondria contact sites responded differently to starvation, ER stress, mitochondrial shape modifications, and changes in the levels of modulators of ER–mitochondria juxtaposition. SPLICS detected contact sites in soma and axons of D. rerio Rohon Beard (RB) sensory neurons in vivo, extending its use to analyses of organelle juxtaposition in the whole animal. Introduction cellular activities. Indeed, a network of contact sites between membranes of different organelles guarantees their In eukaryotic cells, organelles are often found in close mutual communication by creating microdomains that favor proximity, leading to the generation of heterotypic mem- different signaling and metabolic pathways [1,2]. Due to brane appositions that ensure the coordination of several their central role in many fundamental cell processes, the sites of apposition between mitochondria and the endo- plasmic reticulum (ER), which range from 10 to 100 nm, Edited by N. Chandel are, so far, the best characterized [3–5]. Domenico Cieri, Mattia Vicario and Marta Giacomello contributed Several approaches are currently available to assess equally to this work. ER–mitochondria contact sites. Electron microscopy (EM) Electronic supplementary material The online version of this article allows to calculate contact site distance, but it is time- (https://doi.org/10.1038/s41418-017-0033-z) contains supplementary consuming. The in situ proximity ligation assay is based on material, which is available to authorized users. the use of pairs of primary antibodies against proteins on opposing membranes [6]. It is widely used [7–9] but is not * Marisa Brini marisa.brini@unipd.it devoid of drawbacks: as the EM, it can only be used in fixed * Tito Calì cells and is limited by the availability and the specificity of tito.cali@unipd.it the antibodies. The use of fluorescent proteins (FPs) selectively targeted Department of Biomedical Sciences, University of Padova, to the mitochondrial matrix and the lumen of the ER [10] Padova, Italy has been the golden standard to visualize contact sites in Department of Biology, University of Padova, Padova, Italy living cells for years. However, limited resolution in the distance range below 200 nm, differences in FPs expression Dulbecco-Telethon Institute, Venetian Institute of Molecular Medicine, Padua, Italy levels or alterations in organelle morphology complicated Venetian Institute of Molecular Medicine, Padua, Italy the interpretation of experiments of ER–mitochondria jux- taposition upon ablation of the mitochondria-shaping pro- Department of Biomedical Sciences, Institute of Neuroscience, Italian National Research Council (CNR), Padua, Italy tein Mitofusin 2 (Mfn2) [11–13]. 1234567890 1132 D. Cieri et al. To overcome these limits, FP-based sensors of proximity targeted each moiety on one of the juxtaposed membranes, the were developed: a dimerization-dependent FP (ddGFP) [14] GFP fluorescence would be restored only when the two or Venus FP [2,15,16] and a FRET-based probe coupled to a portions were close enough. We therefore placed the non- rapamycin-binding module (FEMP) [17]. While these two fluorescent GFP moiety on the cytosolic face of the OMM 1–10 probes improved the analysis of ER–mitochondria proximity, (OMM-GFP ). To follow short- (≈8–10 nm) and long- 1–10 they also face some limitations: the ddGFP probe is intrin- range (≈40–50 nm) ER–mitochondria interactions [5], two sically not extremely bright [14]; the FRET probe requires constructs that differ for the length of the spacer placed equimolar expression of the two moieties [12] and its in vivo between the ER targeting sequence and the β fragment were applications are limited by the use of rapamycin, a potent created by considering the distance of 0.36 nm between two inducer of autophagy [18–20], to maximize juxtaposition and alpha-carbons in a peptide chain: a ER-Short β with a 29 aa FRET signal. Moreover, both probes cannot be adapted for spacer and a ER-Long β with a 146 aa spacer (i.e., a the investigation of contact sites potentially placed at dif- maximum of ≈10.4 and 52.5 nm, respectively). These values ferent distances, because their dynamic range must be might clearly be subjected to changes (i.e., reduction) since characterized each time that the linker is changed. Artificial the amino acid sequences might not always be fully extended. GFP-based tethers have proved useful to uncover a novel We reasoned that co-expression of ER-Short β with OMM- ER–mitochondria tethering complex in yeast, but they cannot GFP (SPLICS ) and of ER-Long β with OMM-GFP 1–10 S 11 1–10 been used to monitor changes in the ER–mitochondria con- (SPLICS ) would result in reconstitution of GFP fluorescence tact sites [21]. Therefore, an easy, one-step probe that can (Fig. 1a). Two additional constructs, a β -tagged FP (Kate- dynamically detect ER–mitochondria juxtaposition in cellulo β ) and an untargeted GFP were also generated to verify 11 1–10, and in vivo is lacking. the complementation of the OMM-GFP at the OMM 1–10 To overcome these limitations, we devised a split-GFP- (Fig. 1a, left) and the ER -β at the ER (Fig. 1a, middle), S/L 11 based contact site sensor (SPLICS) that can be easily respectively. Expression of SPLICS and SPLICS will result S L adapted to measure ER–mitochondria contact sites over a in fluorescence emission specifically at the ER–mitochondria range of distances as well as other types of hetero and interface (Fig. 1a, right). homotypic contact sites. Upon expression in human cells, The different versions of the SPLICS probes were first this one-step imaging technique specifically identifies nar- tested for their correct localization and topology. A clear row and wide ER–mitochondria apposition lying in a range mitochondrial network appeared in HeLa cells co- of around 10 and 50 nm [5], i.e, that found between mito- expressing OMM-GFP with Kate-β (Fig. 1b, first 1–10 11 chondria and smooth or rough ER [22]. The narrow SPLICS panels); similarly, the ER network became fluorescent when can also detect ER–mitochondria contact sites in vivo in ER -β and ER -β were co-expressed with a cytosolic S 11 L 11 zebrafish sensory neurons. Pharmacological and genetic non-fluorescent GFP (Fig. 1b, second and third panel 1–10 manipulations indicate that these narrow and wide contact couples). Interestingly, when SPLICS or SPLICS were S L sites respond differentially to ER stress, autophagy, mito- expressed in HeLa (Fig. 1b, fourth and fifth panel couples) phagy, and changes in the levels of modulators of and in HEK293 cells (Supplementary Figure S1), fluor- ER–mitochondria juxtaposition. escent individual foci appeared, likely representing the In conclusion, using SPLICS as a tool to investigate juxtapositions between ER and mitochondria. At a closer ER–mitochondria contact sites, we unravel their hetero- inspection, the SPLICS and SPLICS signals retrieved in S L geneity and provide the community with a sensor that can HeLa cells appeared different for number (see quantification be easily adapted to image other types of heterotypic in 3D rendered z-stack images, Fig. 1c). organelle contact sites in human cells and in whole We therefore verified whether SPLICS really recognized organisms. areas of ER–mitochondria juxtaposition. In HeLa cells expressing SPLICS ,the fluorescent dots co-localized with S/L endogenous markers of mitochondria (mtHSP60) and ER Results (calreticulin) (Fig. 1d). Noteworthy, the mitochondrial and ER networks were not completely engaged in the formation of the Two SPLICS probes for different ranges of ER–mitochondria contacts reported by the SPLICS (see ER–mitochondria juxtaposition merge panels of Fig. 1d), suggesting that SPLICS snapshots the juxtaposition at any given moment even when transiently To generate a modular fluorescence based sensor of organelle formed. Immuno-EM with anti-GFP antibody revealed that proximity, we decided to capitalize on the ability of two non- mitochondria and ER membranes in contact with mitochon- fluorescent portions (the GFP moiety and the GFP β-strand dria were preferentially marked (arrowheads in Supplemen- 1–10 11) of the superfolder GFP variant [23–25] to restore a fully tary Figure S2). Despite the non-complemented and fluorescent GFP upon self-assembly. We reasoned that if we complemented OMM-GFP cannot be distinguished by the 1–10 ER-mitochondria contact sites sensor 1133 Fig. 1 Functional characterization of the SPLICS probes. a Cartoon showing the general approach used to design the SPLICS. The mitochondrial network, the ER network, and the ER–mitochondria contact sites are revealed by co- expression of the β -tagged cytosolic RFP (Kate) and the OMM-GFP (left panel), of 1–10 the ER -β constructs and a S/L 11 cytosolic GFP (middle panel) 1–10 and of the SPLICS (right S/L panel), respectively. b Experimental controls showing the correct targeting of the mitochondrial (OMM-GFP ) 1–10 and the ER (ER -β and ER - S 11 L β ) targeted fragments verified by complementation with Kate- β and GFP , respectively. 11 1–10 Co-transfection of HeLa cells with OMM-GFP and both 1–10 ER -β or ER -β induces the S 11 L 11 appearance of a “dotted” fluorescence. c Quantification of ER–mitochondria contacts in HeLa cells. The SPLICS dots were quantified from the 3D rendering of a complete z-stack. Mean ± SEM: SPLICS 56 ± 4, n = 37 cells; SPLICS 229 ± 12, n = 25 cells. d Co-localization of SPLICS fluorescence with S/L mitochondria (mtHSP60) and ER (CRT: calreticulin) markers. Representative traces e and statistical analysis f of 2+ mitochondrial Ca uptake in HeLa cells transfected with SPLICS or SPLICS along with S L mtAeqmut. Mean ± SEM: Void Vector 75 ± 2, n = 65 wells; SPLICS 77 ± 1, n = 54 wells; SPLICS 71 ± 2, n = 54 wells. Scale bar 15 µm. Data shown are the result of 3–5 independent experiments. anti-GFP antibody, it is evident that a consistent number of To gain further insights on the nature of the reconstituted gold nanoparticles are distributed at the ER–mitochondria SPLICs, we evaluated their stability by checking whether interface (inset in Supplementary Figure S2). the number of SPLICS foci could change after 24, 48, S/L 1134 D. Cieri et al. 2+ and 72 h post transfection. Supplementary Figure S3 shows that it does not artificially increase tethering and Ca that the number of SPLICS is stable during the time transfer between ER and mitochondria. S/L course. The number of fluorescent reconstituted foci was also unaffected by the expression level of the probes Modulation of short- and long-range (Supplementary Figure S4), suggesting that bona fide ER–mitochondria interfaces during ER stress and changes in the SPLICS fluorescent foci likely reflect a autophagy S/L variation in ER–mitochondria contact sites number rather than differences in the stability/expression levels of the We next wished to address if SPLICS could respond to probes. Additionally, the overall morphology of the ER and pathophysiological conditions known to affect the extent of mitochondria in cells expressing the SPLICS remained ER–mitochondria contacts. We therefore measured short- S/L grossly unaltered (Supplementary Figure S5). and long-range ER–mitochondria interactions in conditions To exclude that novel and non-physiological contact sites where increased ER–mitochondria coupling was reported, between ER and mitochondria might be artificially induced such as ER stress and induction of autophagy [5,26,27]. In 2+ by SPLICS expression, ER–mitochondria Ca transfer and HeLa cells treated with the ER stress inducer tunicamycin, 2+ mitochondrial Ca uptake were evaluated in HeLa cells or starved, the number of short-range ER–mitochondria expressing SPLICS by aequorin-based measurements. If contact sites measured by SPLICS were increased S/L S 2+ this was the case, mitochondrial Ca transients generated (Figs. 2a,b), in agreement with previous results [5,26,27]. by stimulation with the InsP -linked agonist histamine The picture in the case of long-range ER–mitochondria should be increased in SPLICS-expressing cells [17]; interactions measured by the SPLICS was more complex: however, they were superimposable to those of control cells while tunicamycin significantly decreased the number of (Fig. 1f and quantification in Fig. 1g). Taken together, these SPLICS dots, starvation did not induce any significant results indicate that SPLICS retains the ability to self- change (Figs. 2c,d). Altogether, these results indicate associate only in specific areas where the two organelles are that short and long ER–mitochondria interactions are dif- found within the distance imposed by the linker region and ferentially modulated in response to different stimuli and Fig. 2 Effect of Tunicamycin and Hbss treatment on ER–mitochondria contacts. Immunofluorescence against mitochondria (Tom20, red) is shown in the panels on the middle. The green channel is the merge of several planes. Scale bar 20 µm. a Representative confocal pictures of HeLa cells expressing the SPLICS probe. b Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 58 ± 3, n = 32 cells; Tunicamycin 84 ± 5, n = 33 cells; Hbss 81 ± 5, n = 25 cells. c Representative confocal pictures of HeLa cells expressing the SPLICS probe. d Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 218 ± 11, n = 27 cells; Tunicamycin 171 ± 9, n = 33 cells; Hbss 204 ± 10, n = 23 cells. Data shown are the result of three independent experiments. **p ≤ 0.01, ***p ≤ 0.001. ER-mitochondria contact sites sensor 1135 suggest that the heterogeneity between the two types of reduction in the number of wide ER–mitochondria inter- contact sites reflects their involvement in specialized cel- actions in cells expressing wt Drp1 (Fig. 3c top panels vs lular pathways. middle panels, and Fig. 3d). Interestingly, forced mito- chondrial elongation induced by Drp1-K38A expression Short- and long-range ER–mitochondria interactions resulted in the labeling of the whole surface of mitochondria are differentially modulated by mitochondrial by SPLICS fluorescence, suggesting a complete engage- morphology ment of the mitochondrial network with the ER (Fig. 3c, top panels vs. bottom panels). Due to the filamentous nature During starvation, inhibition of the mitochondrial fission of the observed SPLICS staining, the number of GTPase Dynamin-related protein 1 (Drp1) results in mito- ER–mitochondria contacts/cell under this condition could chondrial elongation, increasing energy conversion and not be reliably quantified; nevertheless, the GFP signal sparing mitochondria from autophagosomal degradation occupied almost completely (about 85%) the mitochondrial [28,29]. We therefore wished to verify short and long surface as measured by Tom20 staining (Supplementary ER–mitochondria interactions upon Drp1-driven mito- Figure S6). Altogether, these results suggest that unopposed chondrial shape changes. We expressed wt or a dominant- mitochondrial fusion is paralleled by an enhancement of the negative mutant form of Drp1 (Drp1-K38A) to induce ER–mitochondria interface that may ensure the supply of mitochondrial fragmentation or elongation and measured lipids required for the sustained mitochondrial morpholo- the occurrence of short- and long-range ER–mitochondria gical changes [26,28,29,31]. juxtaposition with SPLICS. Mitochondrial fragmentation induced by wt Drp1 expression did not change the number Short- and long-range ER–mitochondria contacts of short-range ER–mitochondria interactions (Figs. 3a and respond differentially to Mfn2 silencing and b, compare top panels vs. middle panels), in agreement with presenilin 2 mutant expression previous data [30]. Conversely, mitochondrial elongation induced by dominant-negative Drp1 expression resulted in a We next wished to verify if SPLICS responded to genetic significant increase in the short-range ER–mitochondria modulation of the ER/mitochondria interaction. To this end, contacts detected by SPLICS (Fig. 3a, top panels vs. lower we decided to monitor SPLICS behavior following S S/L panel, and Fig. 3b). The SPLICS measured a significant ablation of Mitofusin 2 (Mfn2), a pro-fusion mitochondria- Fig. 3 Effects of Drp1 overexpression on ER–mitochondria contacts. Immunofluorescence against mitochondria (Tom20, cyan) and Drp1 (red) is shown in the corresponding panels. The green channel is the merge of several planes. Scale bar 20 µm. a Representative confocal pictures of HeLa cells expressing the SPLICS probe. b Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 59 ± 3, n = 79 cells; Drp1 WT 70 ± 5, n = 32 cells; Drp1-K38A 89 ± 9, n = 28 cells. c Representative confocal pictures of HeLa cells expressing the SPLICS probe. d Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± SEM: Ctrl 260 ± 14, n = 24 cells; Drp1 WT 198 ± 14, n = 24 cells. Data shown are the result of 3–4 independent experiments. **p ≤ 0.01. 1136 D. Cieri et al. shaping protein originally identified as a tether between the mitochondria coupling in HeLa cells by organelle-targeted two organelles [11]. However, whether Mfn2 tethers FPs, and in nigral neurons by transmission EM analysis [12,14,32–35] or separates [13,22,36–38] ER and mito- [42,43]. Nevertheless, increased ER–mitochondria juxta- chondria is still a matter of debate. We reasoned that position in patient-derived fibroblasts and in PARK2 SPLICS might contribute to clarify the issue by providing knockout MEFs [44] was also reported. Thus, the exact S/L an estimate of the contact sites over different ranges of function of Parkin at the ER–mitochondria interface under interaction. Acute downregulation of Mfn2 by shRNA in basal conditions and upon mitophagy is unclear. We gen- HeLa cells by three independent shRNA (Supplementary erated a bicistronic vector in which Parkin was cloned Figure S7) increased by ≈40% the number of SPLICS foci upstream of a self-cleaving viral 2A peptide (P2A) [45] (Figs. 4a,b). Conversely, under the same conditions of Mfn2 followed by a plasma membrane-targeted RFP (mCherry- downregulation the SPLICS detected a significant decrease CAAX) to track Parkin-positive cells (Supplementary Fig- by ≈30% in the number of ER–mitochondria interactions ure S9). This construct was co-expressed along with (Figs. 4c,d). Altogether the short- and long-range SPLICS SPLICS in HeLa cells where Parkin is absent or weakly S/L probes not only respond to changes in known modulators of expressed [46,47]. Parkin overexpression increased ER–mitochondria tethering, but they might also prove SPLICS number (Figs. 5a,b), in agreement with our pre- useful to shed light on the observed discrepancies on the vious data [42]. Conversely, Parkin overexpression reduced role of Mfn2 at the ER–mitochondria interface. the SPLICS foci (Figs. 5c,d). Treatment with CCCP Mfn2 and the Familial Alzheimer’s Disease (FAD)- reduced the number of fluorescent foci measured using both related protein Presenilin-2 (PS2) have been reported to act the SPLICS probes, suggesting that activation of PINK1/ in a common route to tune the ER–mitochondria interface Parkin-mediated mitophagy loosens all types of [38,39]. We measured short-range ER–mitochondria inter- ER–mitochondria contacts. actions in human fibroblasts from an FAD-patient carrying the PS2-N141I mutation, previously shown to enhance SPLICS visualizes ER–mitochondria interactions in ER–mitochondria coupling in an Mfn2-dependent manner living zebrafish neurons [38], and a healthy sex- and age-matched control. The SPLICS signal was more than doubled in human FAD-PS2 We finally wished to test if SPLICS can measure fibroblasts compared to controls, thus confirming that ER–mitochondria tethering in an in vivo setting. Imaging of endogenous FAD-PS2 increases ER–mitochondria cou- subcellular structures in living animals, and even more in pling, as already reported, and proving that SPLICS repre- neuronal axons, is limited by the thickness and anatomical sents a useful tool also in patient-derived samples (Figs. 4e, accessibility of tissues. In vivo detection of organelle con- f). Lastly, we tested SPLICS with an additional well- tact sites is still a major challenge because of their dynamic S/L established tethering machinery, i.e., the VAPB/PTPIP51 nature and the lack of appropriate tools. To verify if complex. Interestingly, we detected an increase in the SPLICS could overcome these hurdles, we expressed the SPLICS number, in agreement with previous data [7,8,40] new probes in D. rerio, specifically in Rohon-Beard (RB) (Supplementary Figure S8). The long-range interactions sensory neurons. The correct targeting of the OMM- monitored by SPLICS were instead decreased (Supple- GFP and the ER -β constructs was first verified after L 1–10 S 11 mentary Figure S8): this finding certainly deserves addi- mosaic expression in D. rerio embryos. The OMM-GFP 1–10 tional experiments but again, it might indicate that signal reconstituted by complementation with a β -tagged ER–mitochondria tethering can be heterogeneous and cytosolic protein (DJ-1-β ) fully overlapped with a mito- tightly modulated. chondrial targeted RFP (pTagRFP-mito). Analogously, injection of ER -β and a cytosolic GFP resulted in S 11 1–10 Long- and short-range ER mitochondria contacts fluorescence emission that co-localized with an ER marker reduction during Parkin-mediated mitophagy (pDsRed2-ER) (Supplementary Figure S10), thus demon- strating that the SPLICS fragments are properly expressed, Comforted by the ability of SPLICS to provide insights targeted and self-assembled in living zebrafish embryos. To S/L under pharmacological and genetic manipulation of the allow tissue specific as well as equimolar expression of ER–mitochondria interface, we decided to detect changes in SPLICS, we generated an expression vector where OMM- ER–mitochondria tethering during Parkin-mediated mito- GFP and ER -β are linked by a P2A peptide 1–10 S 11 phagy. In mammalian cells, dysfunctional mitochondria (SPLICS -P2A), an approach suitable also in zebrafish [48]. recruit the E3 ubiquitin ligase Parkin to the OMM through SPLICS -P2A was placed under the control of a bidirec- PINK1 kinase activity, resulting in the recruitment and tional UAS promoter together with a cytosolic DsRed activation of the autophagy machinery [41]. Parkin has been (pT2-DsRed-UAS-SPLICS -P2A) to allow GAL4-driven shown to act as a positive modulator of ER and expression of the UAS promoter (Fig. 6a). The pT2-DsRed- ER-mitochondria contact sites sensor 1137 UAS-SPLICS -P2A vector was then microinjected in the simultaneous, tissue specific expression of DsRed and zebrafish s1102t:GAL4 transgenic line where GAL4 SPLICS (Fig. 6c). By imaging the DsRed-positive neu- expression is restricted to RB neurons (Fig. 6b), yielding rons, we noticed the occurrence of short-range ER 1138 D. Cieri et al. Fig. 4 Effect of Mfn2 knockdown and mutant PS2 on (implying that any reduction observed with SPLICS may S/L ER–mitochondria interface. Immunofluorescence against mitochondria not reflect a dynamic decrease during time). Even if the (Tom20, red) is shown in the panels on the middle. The green channel observation of transient interactions between the ER and is the merge of several planes. Scale bar 20 µm. a Representative motile mitochondria could be limited by the time required confocal pictures of HeLa cells expressing the SPLICS probe. b Quantification of SPLICS contacts by 3D rendering of complete z- to achieve full reconstitution of the SPLICS ,wewere S S/L stacks. Mean ± SEM: SCR shRNA 70 ± 4, n = 76 cells; shRNA Mfn2 able to provide important insights in the biology of this #1 98 ± 7, n = 26 cells; shRNA Mfn2 #3 108 ± 7, n = 28 cells; interface. shRNA Mfn2 #4: 95 ± 7, n = 23 cells. c Representative confocal SPLICS can be also adapted to monitor other types of pictures of HeLa cells expressing the SPLICS probe. d Quantification of SPLICS contacts by 3D rendering of complete z-stacks. Mean ± heterotypic organelle contact sites, e.g., ER and plasma SEM: SCR shRNA 238 ± 10, n = 30 cells; shRNA Mfn2 #1 172 ± 9, membrane (PM), mitochondria and PM, or mitochondria n = 27 cells; shRNA Mfn2 #3 176 ± 8, n = 29 cells; shRNA Mfn2 #4 and endosomes/lysosomes, creating a palette of SPLICS to 190 ± 10, n = 27 cells. e Representative confocal pictures of human image inter-organelle interactions. fibroblasts from a patient with the N141I mutation in PS2 (bottom panel) and an age-matched control (upper panel) expressing the The physiological significance of long-range ER– SPLICS probe. The green channel is the merge of several planes. mitochondria contacts has not been completely defined; Scale bar 20 µm. f Quantification of ER–mitochondria short contacts nevertheless, the comparison of the SPLICS signals under S/L by 3D rendering of complete z-stacks. Mean ± SEM: CTRL 50 ± 5, different pathophysiological conditions indicates that n = 20 cells; PS2-N14I: 101 ± 12, n = 21 cells. Data shown are the result of 2–5 independent experiments. **p ≤ 0.01, ***p ≤ 0.001, ER–mitochondria tethering is heterogeneous and tightly ****p ≤ 0.0001. modulated. Both ER stress and starvation increased SPLICS foci while SPLICS dots were decreased in num- S L mitochondria contacts in both cell body and axons ber under ER stress, suggesting a spatial and functional (Figs. 6d–g). We retrieved several SPLICS contacts in the S specialization of different ER–mitochondria contact sites soma of RB neurons; their frequency was comparable to [27]. Changes in mitochondrial shape also affected the that observed in cultured cells. ER–mitochondria contact ER–mitochondria interface differently: Drp1 overexpression sites were also retrieved in RB axons and enriched at axonal reduced SPLICS interactions, whereas forced mitochon- varicosities and branching points, possibly representing drial elongation increased both SPLICS foci number. S/L 2+ axon zones with specialized functions where Fragmentation concomitantly ensures basal Ca -dependent ER–mitochondria crosstalk is important to propagate and homeostatic mitochondrial functions and protects from stress 2+ 2+ regulate Ca signals [49–51] (arrowheads in Fig. 6f). The responses involving ER–mitochondria Ca crosstalk [30]. number of short ER–mitochondria interactions was com- Indeed, potentiation of the ER–mitochondria interface under parable in soma and axons (Fig. 6g), suggesting that these conditions of Drp1-dependent fragmentation can lead to 2+ juxtapositions are regulated by similar mechanisms in the mitochondrial Ca overload and cell death. The reduction two portions of the neuron. in SPLICS number is probably due to the reduction of the interface available for additional contacts and it could be 2+ relevant in the Ca -dependent stress responses. Of note, Discussion Drp1-dependent mitochondrial fission facilitates mitophagy whereas mitochondrial elongation inhibits mitochondrial Here we report a one-step, split-GFP-based method to autophagy [52]. Thus, it is tempting to speculate that mito- assess narrow and wide ER–mitochondria contact sites and chondria are protected from mitophagy when mitochondrial their modulation and demonstrate its suitability to monitor fusion is forced due to increased ER–mitochondria contacts; by confocal microscopy inter-organelle interactions in conversely, mitochondrial fission is accompanied by a human cells and in vivo in zebrafish RB neurons. reduction in SPLICS signal to favor the engulfment of The SPLICS is versatile and sensitive and reveals short mitochondria by the autophagosome. and long-range ER–mitochondria interactions and their The SPLICS described here can also help to understand changes upon pharmacological and genetic manipulations. the role of Mfn2 in ER–mitochondria juxtaposition: we An advantage of SPLICS over the other probes relies observed that Mfn2 silencing led to an increase in SPLICS not only on its unnerving modularity, but also on the and a decrease in SPLICS dots. Similar findings were high brightness, stability and a high threshold over back- observed upon the overexpression of the well-established ground. The first feature was exploited here to generate two tethering machinery VAPB/PTPIP51. Future work might versions of SPLICS, one engineered to detect narrow therefore capitalize on spectral variant of the SPLICS S/L (≈8–10 nm) and a second to image wide (≈40–50 nm) probes to verify if the wide interactions comprise the narrow ER–mitochondria interactions [5,13]. Although the low ones, or if they occur on different areas of the organelles. dissociation rate of the GFP fragments [23] could, upon CCCP treatment in HeLa cells expressing Parkin caused reconstitution, make their association poorly reversible a marked reduction in both SPLICS and SPLICS signal, S L ER-mitochondria contact sites sensor 1139 cells; Parkin CCCP 65 ± 7, n = 26 cells. c Representative confocal Fig. 5 Effects of Parkin on ER–mitochondria contacts. Immuno- pictures of HeLa cells expressing the SPLICS probe along with fluorescence against mitochondria (Tom20, cyan) is shown in the panels Parkin-2A-mCherry-CAAX (bottom panels). d Quantification of the on the middle. The green channel is the merge of several planes. Scale bar SPLICS contacts by 3D rendering of complete z-stacks. Mean ± 20 µm. a Representative confocal pictures of HeLa cells expressing the SEM: CTRL 227 ± 9, n = 44 cells; Parkin 193 ± 11, n = 32 cells; SPLICS probe along with Parkin-2A-mCherry-CAAX (bottom panels). Parkin CCCP 159 ± 8, n = 22 cells. Data shown are the result of 3–8 b Quantification of SPLICS contacts by 3D rendering of complete z- independent experiments. *p ≤ 0.05, **p ≤ 0.01, ****p ≤ 0.0001. stacks. Mean ± SEM: Ctrl 61 ± 3, n= 57 cells; Parkin 94 ± 6, n= 33 supporting a model whereby mitochondria and ER separate in this process. SPLICS is a versatile tool to visualize during mitophagy. Under basal conditions, overexpression ER–mitochondria contacts in vivo both in physiological and of Parkin conversely increases ER–mitochondria tethering pathological conditions and it will be useful to explore how in several cellular models [42,43], suggesting a house- disease-related genes affect neuronal function and survival keeping function at this interface, whereas massive Parkin through the modulation of the ER–mitochondria interface. activation and mitophagy induction mirrors a condition of degradation/deactivation of a handful of molecules involved in ER–mitochondria juxtaposition (e.g., VDAC, Materials and methods Mfn2, TOM20) [53] with consequent ER–mitochondria separation. Cell Lines Despite the efforts in the analysis of the ER–mitochondria contact sites, their properties in neurons and, most impor- HeLa and HEK293 cells (ATCC) were grown in a 5% tantly, in living vertebrates are still poorly characterized due CO atmosphere in Dulbecco’s modified Eagle’smedium to the lack of appropriate methods suitable for in vivo stu- high glucose (DMEM; Euroclone), supplemented with dies. We optimized the SPLICS for in vivo expression by 10% fetal bovine serum (Gibco), 100 U/ml penicillin and generating the pT2-DsRed-UAS-SPLICS -2A construct. 100 mg/ml streptomycin. Where indicated, the cells were SPLICS could report ER–mitochondria contacts in cell body treated 48 h after transfection with 10 µM CCCP (Sigma- and axons of RB neurons in living zebrafish embryos. Aldrich), Hbss (Thermo Fisher) or 10 µg/ml Tunicamycin ER–mitochondria contacts in the axons often localized at (Sigma-Aldrich) for 4 h at 37 °C, in a 5% CO atmo- sites of branching, a process sustained by mitochondrial sphere. Mock cells were maintained in growth medium, 2+ ATP [54] and Ca release from internal stores [55], thus which was changed simultaneously with the beginning of suggesting that ER–mitochondria interplay may have a role treatments. Female FAD-PS2-N141I fibroblasts (Coriell 1140 D. Cieri et al. Fig. 6 Expression of the SPLICS probe in living zebrafish embryos. a Schematic depiction of the bidirectional construct that allows detection of DsRed and SPLICS -P2A in a Gal4-dependent manner. The 2A peptide guarantees the generation of an equimolar amount of the two spGFP fragments. b Experimental setting used to image ER–mitochondria contacts in zebrafish embryos: a schematic drawing of RB neurons is shown on the right. c Representative image of a 1 dpf s1102t:GAL4 embryo injected with the pT2- DsRed-UAS-SPLICS -P2A construct. d Live imaging of short ER–mitochondria contacts in RB neurons of s1102t:GAL4 zebrafish embryos. The picture is the merge of several planes. The 3D rendering of the z-stack is shown on the right. Scale bar: 15 µm. e Quantification of SPLICS contacts in the cell body of RB neurons by 3D rendering of complete z-stacks. Mean ± SEM: 22 ± 1, n = 28 cells from 11 fish. f Live imaging of SPLICS contacts in the axons of RB neurons. The picture is the merge of several planes. The 3D rendering of the complete z-stack is shown on the right. Scale bar: 15 µm. g Quantification of the density of SPLICS contacts in the cell body and the axons of RB neurons. Mean ± SEM: RB soma: 0.17 ± 0.01, n = 28 cells from 11 fish; RB axon: 0.14 ± 0.01, n = 20 cells from 6 fish. Data shown are the result of two independent experiments. Institute, AG09908) and age/sex-matched control fibro- Rohon-Beard neurons (s1102t:GAL4), Et(E1b:GAL4- blasts (Coriell Institute, AG08269) were grown in DMEM VP16)s1102t; Tg(UAS-E1b:Kaede)s1999t fish (ZIRC, containing FCS (15%), L-glutamine (2 mM), penicillin ZL1384) were outcrossed with wild-type (wt) fish [57] and (100 U/ml, Euroclone), and streptomycin (100 µg/ml, the fluorescent offspring was discarded. The remaining fish Euroclone). were genotyped from fin clips with primers specific for GAL4. To perform experiments, both wt and s1102t:GAL4 Zebrafish husbandry and transgenic lines fish were used. All experiments were conducted on 24 h post fertilization (hpf) embryos. All animal experiments were conducted as previously reported [56]. Adult fish were maintained and raised in 5 l Transfection tanks with freshwater at 28 °C with a 12 h light/12 h dark cycle. Zebrafish embryos were obtained from spontaneous Twelve hours before transfection, HeLa cells were seeded spawnings. To obtain fish selectively expressing Gal4 in onto 13 mm glass coverslips and allowed to grow to 50% ER-mitochondria contact sites sensor 1141 confluence. Cells were transfected by calcium phosphate Myc were kindly provided by Professor Christopher C.J. [58]. For co-transfection, the two SPLICS ER and mito- Miller (Department of Neuroscience and Department of chondrial fragments were in a 1.5:2 ratio with the over- Basic and Clinical Neurosciences, King’s College London). expressed protein of interest. Human fibroblasts were Where indicated, mitochondria are labeled with a pTagRFP- electroporated by using a NeonTM transfection system mito construct (Evrogen). To knockdown Mfn2, three dif- (Life Technologies), according to the manufacturer's ferent shRNAs against Mfn2 were used (SureSilencing instruction. shRNA Plasmid, Hygromycin Gene: Mfn2; Refseq Acces- sion #:NM_014874; Clone #1: AGAGGCGGTTCGACT- Cloning and fusion plasmid construction CATCAT; Clone #3: TGATGTG GCCCAACTCTAAGT; Clone #4: CCAGTAGTCCTCAA GGTTTAT; Scramble: Humanized GFP (GFP ) was PCR amplified from GGAATCTCATTCGATGCATA C). The aminoacid 1–10 1–10(h) the GPI-GFP kindly provided by Prof. Fabien Pinaud, sequences of the ER anchored probes are: 1–10(h) Department of Biological Sciences and Chemistry, Uni- MRDHMVLHEYVNAAGITGGDGGSGGGSKLRVF versity of Southern California [59], to insert the Tom20 N33 LALPIIMVVAFSMCIICLLMAGDTWTETLAYVLFWG targeting sequence [60] by using the OMM-GFP For. VASIGTFFIILYNGKDFVDAPRLVQKEKID (the Short) 1–10 (TCGAATTCATGGTGGGCCGGAACAGCGCCATCGC and MRDHMVLHEYVNAAGITGGDGGSGGGSKLMW CGCGGGCGTGTGCGGTGCCCTCTTCATAGGGTACT HEGLEEASRLYF GERNVKGMFEVLEPLHAMMERGP GCATCTACTTTGACCGCAAAAGGCGGAGTGACCC QTLKETSFNQAYGRDLMEAQEWCRKYMKSGNVKD CAACTCCAAAGGAGAAG) and Rev. (AC TTCTCACT LTQAWDLYYHVFRRISKQGSEAAAREAAARGGASG CGAGTTATGTTCCTTTTTCATTTGGATCTTTGCTCA AGAGAGAILNSRVFLALPIIMVVAFSMCIICLLMAGD GG) primers and inserted into the vector pcDNA3 between TWTETLAYVLFWGVASIGTFFIILYNGKDFVDAPRLV the EcoRV and XhoI restriction sites. The chimeric QKEKID (the Long). sequences composed by the minimal Sac1 ER targeting sequence and the Sac1 ER targeting sequence containing Zebrafish constructs additional 267 bp of the helix-FRB fragment (derived from the pEGFP-C3/CFP-HA-FRB-helix-ER plasmid [5]) (ER - To properly express SPLICS in zebrafish, the ER -β and L S 11 β ) were fused with the β tag to generate the ER -β and the OMM-GFP coding sequences were respectively 11 11 s 11 1–10 the ER -β constructs, respectively. The constructs were cloned upstream and downstream of a viral 2A peptide L 11 chemically synthetized (Thermo Fisher) by using the sequence (pSYC-181, Addgene), previously reported to be EcoRI/NotI and the EcoRV/XhoI restriction sites of cleaved within the cell to generate an equimolar amount of pCDNA3.1(+), respectively. The sequence encoding for the two genes [48]. The fragment encoding for the OMM- β , flanked by different multi cloning sites, has been GFP was excised from pcDNA3 with EcoRI and XhoI 11 1–10 inserted into the commercial vectors pCDNA6.5/V5-DEST and ligated into the pSYC-181 vector digested with the (Invitrogen), in order to obtain a backbone that could easily same enzymes. Since these sites cloned the OMM-GFP 1–10 allow fusion of the β fragment with the protein of interest. fragment out of frame, we mutagenized the resulting vector Kate was amplified with the Kate For. (AAAAAAGCT- in order to re-establish the correct frame, using the primer TATGGTGAGCGAGC) and the Kate Rev. (TTTTTG OMM-GFP mut. (CCCTGGACCTAGATCTGAATTC 1–10 GTACCTCATCTGTGCC) primers using as a template ATGGTGGGCC). ER was amplified from pcDNA3 using mKate2-pcDNA3.1 (pEVROGEN) and inserted in pDEST- the primers ER -β 2A For. (ACGCGGATCCGCCAC- S 11 β to generate Kate β . To generate a vector simulta- CATGCGGGACCACATGGTG C) and Rev. (TTCCCCCG 11 11 neously expressing (through a 2A peptide) Parkin and GGGTCGATCTTCTCTTT), which introduce, respectively, membrane mCherry (CAAX-mCherry), the coding a BamHI-Kozak sequence and a SmaI site at the 5′ and 3′ sequence of Parkin [42] was amplified with the primers ends of the ER coding sequence. The PCR product was Parkin For. (ACGCGGATCCGCCACCATGATAGTGTT purified, digested with both BamHI and SmaI and then TGTCAGGTT) and Rev. (TTCCCCCGGGCA CGTCGAA ligated into the pSYC-181-OMM-GFP vector, pre- 1–10 CCAGTGGTCC). The PCR products were purified using viously digested with the same restriction enzymes. We the GenElute Gel Extraction Kit (Sigma), digested with refer to this plasmid as pSYC-SPLICS -P2A. To selectively BamHI and SmaI and then ligated into the pSYC-187 vector express the SPLICS -P2A in zebrafish neurons, we (Addgene) digested with the same restriction enzymes. The exploited the Gateway technology to generate the pT2- constructs codifying for Drp1 WT and K38A [30,61,62] DsRed-UAS-SPLICS -P2A vector. Briefly, the fragment and pmTurquoise2-ER were kindly donated by Diego De encoding for the SPLICS -P2A was excised from the Stefani (Department of Biomedical Sciences, University of pSYC-SPLICS -P2A vector with BamHI and XbaI and Padova). The pCIneo-PTPIP51-HA and the pCIneo-VAPB- cloned into the BamHI-XbaI sites of the pME-MCS vector 1142 D. Cieri et al. (Tol2 kit). The resulting pME-SPLICS -P2A was recom- overnight in 2.3 M sucrose, mounted on aluminum pins, and bined through the LR reaction with the pT2dDESTpADs frozen in liquid nitrogen. Ultrathin cryosections of 70 nm Red.T4E1bUASE1bGWR1R2pA vector (kindly donated by were cut with a Leica FC7 ultramicrotome (Leica Micro- Christian Haas, Ludwig-Maximilians University Munich). systems, Germany), collected on formvar, carbon-coated 150 The resulting pT2-DsRed-UAS-SPLICS -P2A plasmid was mesh copper grids and immunolabeled with anti-GFP injected into 1–2 cells stage s1102t:GAL4 embryos. To (Abcam, Cat#ab290) and 10 nm Protein A-gold (Utrecht label mitochondria and the endoplasmic reticulum, the University, The Netherlands). Grids were then contrasted for pTagRFP-mito (Evrogen) and the pDsRed2-ER (Clontech) 10 min on ice in a solution of 1.8% methylcellusose/0.4% plasmids were, respectively, injected in WT eggs. For uranyl acetate, air dried on wire loops and observed in a injections, all plasmids were diluted in Danieau solution ZEISS Leo912AB (Zeiss, Oberkochen, Germany). Images (58 mM NaCl, 0.7 mM KCl, 0.4 mM MgSO , 0.6 mM Ca were acquired using a 2Kx2K bottom mounted slow-scan (NO ) , 5 mM HEPES pH 7.6) and 0.5% phenol red. Proscan camera (Scheuring, Germany) controlled by the 3 2 EsivisionPro 3.2 (Soft Imaging System, Münster, Germany). Immunocytochemistry Image acquisition and processing To image cells over-expressing a protein of interest, trans- fected cells plated on 13 mm glass coverslips were fixed Cells were generally imaged 48–72 h after transfection with 48–72 h post transfection with 3.7% formaldehyde in a Leica TSC SP5 inverted confocal microscope, using either phosphate-buffered saline (PBS; 140 mM NaCl, 2 mM KCl, a HCX PL APO 63X/numerical aperture 1.40–0.60 or a 1.5 mM KH PO ,8mM Na HPO ,pH 7.4) for 20minand HCX PL APO ×100/numerical aperture 1.4 oil-immersion 2 4 2 4 washed three times with PBS. Cell permeabilization was objective. Images were acquired by using the Leica AS performed by 20 min incubation in 0.1% Triton X-100/PBS software. To count ER–mitochondria contacts, a complete followed by 30 min wash in 1% gelatin/PBS (type IV, from z-stack of the cell was acquired every 0.29 µm. Z-stacks bovine skin, Sigma) and 15 min wash in PBS at room tem- were processed using Fiji [64]: images were first convolved, perature (RT). The coverslips were then incubated for 90 min and then filtered using the Gaussian Blur filter. A 3D at 37 °C with the specific primary antibody diluted 1:20 in reconstruction of the resulting image was obtained using the PBS (Tom20: Santa Cruz Biotech., Cat#sc-11415; Parkin: Volume J plugin (http://bij.isi.uu.nl/vr.htm). A selected face Santa Cruz Biotech., Cat#sc-32282; Drp1: BD Biosciences, of the 3D rendering was then thresholded and used to count Cat#611113; mtHSP60: Abcam, Cat#ab82520; Calreticulin: ER–mitochondria contact sites. Thermo Fisher, Cat#PA3-900; Myc: Millipore, Cat#05-724; HA: Cell Signalling, Cat#3724S). Further washing steps with Zebrafish imaging gelatine and PBS were repeated as mentioned before to remove the excess of primary antibody. Staining was revealed At 24 hpf, embryos were screened for fluorescence, by the incubation with specific AlexaFluor secondary anti- dechorionated and fixed or anesthetised according to the bodies (Thermo Fisher: Goat anti-Rabbit IgG AlexaFluor 405, experiment. To image the co-localization of ER -β and S 11 Cat#A-31556; Goat anti-Rabbit IgG AlexaFluor 594, Cat#A- OMM-GFP with mitochondria and the ER, fish were 1–10 11012; Donkey anti-Goat IgG AlexaFluor 633, Cat#A-21082; fixed for 2 h at RT with 4% PFA, then washed with PBS Goat anti-Mouse IgG AlexaFluor 633, Cat#A-21050; Goat and mounted in low melting agarose (1.3%, Euroclone) on anti-Mouse IgG AlexaFluor 488, Cat#A-11001; Goat anti- glass coverslips. Rabbit IgG AlexaFluor 488, Cat#A-11008) for 45 min at RT For in vivo imaging, fish were anesthetised and mounted (1:50 dilution in PBS; 1:20 only for Goat anti-Rabbit IgG on 35 × 10 mm glass bottom Petri dishes (Ted Pella, INC. AlexaFluor 405). After further washing steps, coverslips were Prod. No. 14023-20) in low melting agarose (1.3%, Euro- mounted using Mowiol 4–88 (Sigma). The coverslips were Clone). Fish water containing tricaine methanesulfonate observed at the SP5 Leica confocal microscope at lasers 0.61 mM (Sigma) was added in the Petri dishes, in order to wavelength of 405, 458, 488, 543, 555, and 633 nm. keep fish anesthetised. Mounted fish were imaged at RT (20–23 °C) using a Leica TSC SP5 inverted confocal Electron microscopy microscope, using either a HCX PL APO ×63/numerical aperture 1.40–0.60 or a HCX PL APO ×100/numerical Cells were immunogold labeled by the Tokuyasu technique aperture 1.4 oil-immersion objective. To count ER-mito [63]. Briefly, cells were fixed in 2% paraformaldehyde, 0.2% contacts, a complete z-stack of the cell was acquired every glutaraldehyde in 0.1 M phosphate buffer (PB) for 1 h at RT. 0.29 µm. To acquire a representative image of a whole fish Next cells were gently scraped in 1% gelatin and embedded expressing pT2-DsRed-UAS-SPLICS -P2A, a ×10 HCPX in 12% gelatin in PB. Gelatin squared blocks were infiltrated PL Fluotar NA 0.3 objective was used. ER-mitochondria contact sites sensor 1143 2+ Ca measurements Statistical analysis 2+ Ca measurements were performed by co-transfecting Results shown are mean values ± SEM. Student’s unpaired HeLa cells in a six-well plate with low-affinity mitochon- two-tailed t-test was used for comparisons involving two drial aequorin (mtAeqmut) with the two SPLICS moieties groups when sample followed a Gaussian distribution, in a 1.5:1.5:1 ratio favouring SPLICS. Forty-eight hours otherwise Mann–Whitney test was used. Differences post transfection, cells were re-plated into a 96-wells plate between groups were considered significant when p ≤ 0.05. (PerkinElmer). mtAeqmut was reconstituted by incubating All statistical analyses were performed using GraphPad cells for1.5hwith5 µM coelenterazine (Santa Cruz Bio- Prism version 6.00 for Mac OS X, GraphPad Software (La tech.) in modified Krebs Ringer Buffer (KRB: 125 mM Jolla, California, USA). The exact values of n and their NaCl, 5 mM KCl, 400 mM KH PO ,1mM MgSO ,20 means are indicated in the figure legends. *p ≤ 0.05, **p ≤ 2 4 4 mM Hepes, pH 7.4) supplemented with 5 mM glucose at 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. 37 °C. Luminescence measurements were carried out using a PerkinElmer EnVision plate reader equipped with two Acknowledgements We are deeply indebted with Prof. F. Pinaud (University of Southern California) for kindly providing the humanized injector units. After reconstitution, cells were placed in 70 GFP construct. We thank Prof. G. Hajnoczky (Thomas Jefferson 1–10 µl of KRB solution and luminescence from each well was University, Philadelphia) for kindly providing the plasmid pEGFP-C3/ measured for 1 min. During the experiment, 100 µM his- CFP-HA-FRB-helix-ER. We thank Prof. C. C.J. Miller (King’s College tamine at the final concentration were first injected to London) for kindly providing the PTPIP51-HA and the VAPB-Myc 2+ 2+ expression vectors. We thank Prof. R. Rizzuto (University of Padova), activate Ca transients, and then a hypotonic, Ca -rich, Dr. D. De Stefani (University of Padova) and Prof. G. Szabadkai digitonin-containing solution was added to discharge the (University College London) for helpful discussions, we also remaining aequorin pool. Output data were analyzed and thank Prof. F. Argenton and the Zebrafish Facility of the Department of calibrated with a custom-made macro-enabled Excel Biology, University of Padova and Dr. A. Raimondi (Advanced Light and Electron Microscopy BioImaging Center ALEMBIC, San Raffaele workbook. Scientific Institute). This research was supported by grants from the Ministry of University and Research (Bando SIR 2014 no. Western blot RBSI14C65Z to T.C.; FIRB RBAP11Z3YA_005 to L.S.), from the University of Padova (Progetto Giovani 2012 no. GRIC128SP0 to T.C., Progetto di Ateneo 2016 no. CALI_SID16_01 to T.C., Progetto di HeLa cells were seeded in a six-well plate and transfected Ateneo 2015 no. CPDA153402 to M.B.), from Cariparo Starting Grant with three different shRNAs against Mfn2. A scramble 2016 AIFbiol to M.G., from the EU Joint Programme in Neurode- shRNA was used as control. At 48 h post transfection, generative Disease (2015–2018, “Cellular Bioenergetics in Neurode- cells were washed with PBS and proteins were extracted generative Diseases: A System-Based Pathway and Target Analysis”)to P.P., from the European Union (ERC FP7-282280, FP7 CIG PCIG13- for 20 min using ice cold lysis buffer (50 mM Tris-HCl pH GA-2013-618697 to L.S.) and from Telethon (GGP02016 to L.S.). 7.4, 150 mM NaCl, 10 mM EGTA, 1% Triton X-100, 1 mM protease inhibitor cocktail (Sigma)). Samples were then centrifuged at 10,000 rpm for 10 min at 4 °C. Protein Compliance with ethical standards concentration was determined by the Bradford assay (Bio- Conflict of interest The authors declare that they have no conflict of Rad). Samples were separated on a 4–15% Mini- interest. PROTEAN TGX™ Precast Protein Gels (Bio-Rad) and blotted using a Immobilon-PSQ PVDF Membrane (Merck Open Access This article is licensed under a Creative Commons Millipore). The membrane was blocked for 1 h at RT using Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as 5% non-fat dried milk in TBST (20 mM Tris-HCl, pH 7.4, long as you give appropriate credit to the original author(s) and the 150 mM NaCl, 0.05% Tween-20) and incubated with pri- source, provide a link to the Creative Commons license, and indicate if mary antibodies (Mfn2: Abcam, Cat#ab50838, 1:1000 in changes were made. The images or other third party material in this TBST, overnight at 4 °C; β-actin: Sigma, Cat#A5441, article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not 1:30,000 in TBST, 2 h at RT; Myc: Millipore, Cat#05-724, included in the article’s Creative Commons license and your intended 1:1000 in TBST, overnight at 4 °C; HA: Cell Signalling, use is not permitted by statutory regulation or exceeds the permitted Cat#3724 S, 1:1000 in TBST, overnight at 4 °C). After use, you will need to obtain permission directly from the copyright three washing steps in TBST, detection was obtained by holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. incubating the membrane with secondary horseradish peroxidase-conjugated antibodies (Santa Cruz Biotech.: Goat anti-Rabbit IgG-HRP, Cat#sc-2004; Goat anti-Mouse References IgG-HRP, Cat#sc-2005; 1:2000 in TBST) for 1 h at RT and byincubationwiththe Luminata HRPsubstrate 1. Prinz WA. Bridging the gap: membrane contact sites in signaling, (Merck Millipore). metabolism, and organelle dynamics. J Cell Biol. 2014;205:759–69. 1144 D. Cieri et al. 2. Eisenberg-Bord M, Shai N, Schuldiner M, Bohnert M. A tether is revealed by a synthetic biology screen. Science. 2009;325: a tether is a tether: tethering at membrane contact sites. Dev 477–81. Cell.2016;39:395–409. 22. Wang PT, Garcin PO, Fu M, Masoudi M, St-Pierre P, Pante N, 3. Murley A, Nunnari J. The emerging network of mitochondria- et al. Distinct mechanisms controlling rough and smooth endo- organelle contacts. Mol Cell.2016;61:648–53. plasmic reticulum contacts with mitochondria. J Cell Sci. 4. Giacomello M, Pellegrini L. The coming of age of the 2015;128:2759–65. mitochondria-ER contact: a matter of thickness. Cell Death Differ. 23. Cabantous S, Terwilliger TC, Waldo GS. Protein tagging and 2016;23:1417–27. detection with engineered self-assembling fragments of green 5. Csordas G, Renken C, Varnai P, Walter L, Weaver D, Buttle KF, fluorescent protein. Nat Biotechnol. 2005;23:102–7. et al. Structural and functional features and significance of the 24. Pedelacq JD, Cabantous S, Tran T, Terwilliger TC, Waldo GS. physical linkage between ER and mitochondria. J Cell Biol. Engineering and characterization of a superfolder green fluor- 2006;174:915–21. escent protein. Nat Biotechnol. 2006;24:79–88. 6. Tubbs E, Rieusset J. Study of endoplasmic reticulum and mito- 25. Cabantous S, Waldo GS. In vivo and in vitro protein solubility chondria interactions by in situ proximity ligation assay in fixed assays using split GFP. Nat Methods. 2006;3:845–54. cells. J Vis Exp. 2016;118. doi: 10.3791/54899. PubMed PMID: 26. Bravo R, Vicencio JM, Parra V, Troncoso R, Munoz JP, Bui M, 28060261. et al. Increased ER-mitochondrial coupling promotes mitochon- 7. Gomez-Suaga P, Paillusson S, Stoica R, Noble W, Hanger DP, drial respiration and bioenergetics during early phases of ER Miller CC. The ER-mitochondria tethering complex VAPB- stress. J Cell Sci.2011;124(Pt 13):2143–52. PTPIP51 regulates autophagy. Curr Biol. 2017;27:371–85. 27. Bravo-Sagua R, Lopez-Crisosto C, Parra V, Rodriguez-Pena M, 8. Stoica R, Paillusson S, Gomez-Suaga P, Mitchell JC, Lau DH, Rothermel BA, Quest AF, et al. mTORC1 inhibitor rapamycin and Gray EH, et al. ALS/FTD-associated FUS activates GSK-3beta to ER stressor tunicamycin induce differential patterns of ER- disrupt the VAPB-PTPIP51 interaction and ER-mitochondria mitochondria coupling. Sci Rep. 2016;6:36394. associations. EMBO Rep. 2016;17:1326–42. 28. Rambold AS, Kostelecky B, Elia N, Lippincott-Schwartz J. 9. Paillusson S, Gomez-Suaga P, Stoica R, Little D, Gissen P, Tubular network formation protects mitochondria from autopha- Devine MJ, et al. alpha-Synuclein binds to the ER-mitochondria gosomal degradation during nutrient starvation. Proc Natl Acad tethering protein VAPB to disrupt Ca2+ homeostasis Sci USA.2011;108:10190–5. and mitochondrial ATP production. Acta Neuropathol. 29. Gomes LC, Di Benedetto G, Scorrano L. During autophagy 2017;134:129–49. mitochondria elongate, are spared from degradation and sustain 10. Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz cell viability. Nat Cell Biol. 2011;13:589–98. LM, et al. Close contacts with the endoplasmic reticulum as 30. Szabadkai G, Simoni AM, Chami M, Wieckowski MR, Youle RJ, determinants of mitochondrial Ca2+ responses. Science. 1998; Rizzuto R. Drp-1-dependent division of the mitochondrial net- 280:1763–6. work blocks intraorganellar Ca2+ waves and protects against Ca2 11. de Brito OM, Scorrano L. Mitofusin 2 tethers endoplasmic reti- + -mediated apoptosis. Mol Cell. 2004;16:59–68. culum to mitochondria. Nature. 2008;456:605–10. 31. Hamasaki M, Furuta N, Matsuda A, Nezu A, Yamamoto A, Fujita 12. Naon D, Zaninello M, Giacomello M, Varanita T, Grespi F, N, et al. Autophagosomes form at ER-mitochondria contact sites. Lakshminaranayan S, et al. Critical reappraisal confirms that Nature. 2013;495:389–93. Mitofusin 2 is an endoplasmic reticulum-mitochondria tether. Proc 32. Sugiura A, Nagashima S, Tokuyama T, Amo T, Matsuki Y, Ishido Natl Acad Sci USA. 2016;113:11249–54. S, et al. MITOL regulates endoplasmic reticulum-mitochondria 13. Filadi R, Greotti E, Turacchio G, Luini A, Pozzan T, Pizzo P. contacts via Mitofusin2. Mol Cell. 2013;51:20–34. Mitofusin 2 ablation increases endoplasmic reticulum- 33. Chen Y, Csordas G, Jowdy C, Schneider TG, Csordas N, Wang mitochondria coupling. Proc Natl Acad Sci USA. 2015;112: W, et al. Mitofusin 2-containing mitochondrial-reticular micro- E2174–81. domains direct rapid cardiomyocyte bioenergetic responses via 14. Alford SC, Ding Y, Simmen T, Campbell RE. Dimerization- interorganelle Ca(2+) crosstalk. Circ Res. 2012;111:863–75. dependent green and yellow fluorescent proteins. ACS Synth Biol. 34. Schneeberger M, Dietrich MO, Sebastian D, Imbernon M, Cas- 2012;1:569–75. tano C, Garcia A, et al. Mitofusin 2 in POMC neurons connects 15. Toulmay A, Prinz WA, A conserved membrane-binding domain ER stress with leptin resistance and energy imbalance. Cell. targets proteins to organelle contact sites. J Cell Sci.2012;125(Pt 2013;155:172–87. 1):49–58. 35. Li D, Li X, Guan Y, Guo X. Mitofusin-2-mediated tethering of 16. MacDonald ML, Lamerdin J, Owens S, Keon BH, Bilter GK, mitochondria and endoplasmic reticulum promotes cell cycle Shang Z, et al. Identifying off-target effects and hidden pheno- arrest of vascular smooth muscle cells in G0/G1 phase. Acta types of drugs in human cells. Nat Chem Biol. 2006;2:329–37. Biochim Biophys Sin (Shanghai). 2015;47:441–50. 17. Csordas G, Varnai P, Golenar T, Roy S, Purkins G, Schneider TG, 36. Cosson P, Marchetti A, Ravazzola M, Orci L. Mitofusin-2 inde- et al. Imaging interorganelle contacts and local calcium dynamics pendent juxtaposition of endoplasmic reticulum and mitochondria: at the ER-mitochondrial interface. Mol Cell. 2010;39:121–32. an ultrastructural study. PLoS ONE 2012;7:e46293. 18. Ravikumar B, Duden R, Rubinsztein DC. Aggregate-prone pro- 37. Leal NS, Schreiner B, Pinho CM, Filadi R, Wiehager B, Karl- teins with polyglutamine and polyalanine expansions are degraded strom H, et al. Mitofusin-2 knockdown increases ER- by autophagy. Hum Mol Genet. 2002;11:1107–17. mitochondria contact and decreases amyloid beta-peptide pro- 19. Ravikumar B, Vacher C, Berger Z, Davies JE, Luo S, Oroz LG, duction. J Cell Mol Med 2016;20:1686–95. et al. Inhibition of mTOR induces autophagy and reduces toxicity 38. Filadi R, Greotti E, Turacchio G, Luini A, Pozzan T, Pizzo P. of polyglutamine expansions in fly and mouse models of Hun- Presenilin 2 modulates endoplasmic reticulum-mitochondria cou- tington disease. Nat Genet. 2004;36:585–95. pling by tuning the antagonistic effect of Mitofusin 2. Cell 20. Noda T, Ohsumi Y. Tor, a phosphatidylinositol kinase homo- Rep.2016;15:2226–38. logue, controls autophagy in yeast. J Biol Chem. 1998;273: 39. Zampese E, Fasolato C, Kipanyula MJ, Bortolozzi M, Pozzan T, 3963–6. Pizzo P. Presenilin 2 modulates endoplasmic reticulum (ER)- 21. Kornmann B, Currie E, Collins SR, Schuldiner M, Nunnari J, mitochondria interactions and Ca2+ cross-talk. Proc Natl Acad Weissman JS, et al. An ER-mitochondria tethering complex Sci USA. 2011;108:2777–82. ER-mitochondria contact sites sensor 1145 40. De Vos KJ, Morotz GM, Stoica R, Tudor EL, Lau KF, Ackerley 52. Twig G, Elorza A, Molina AJ, Mohamed H, Wikstrom JD, Walzer S, et al. VAPB interacts with the mitochondrial protein PTPIP51 G, et al. Fission and selective fusion govern mitochondrial segre- to regulate calcium homeostasis. Hum Mol Genet. 2012;21: gation and elimination by autophagy. EMBO J. 2008;27:433–46. 1299–311. 53. Sarraf SA, Raman M, Guarani-Pereira V, Sowa ME, Huttlin EL, 41. Eiyama A, Okamoto K. PINK1/Parkin-mediated mitophagy in Gygi SP, et al. Landscape of the PARKIN-dependent ubiquity- mammalian cells. Curr Opin Cell Biol. 2015;33:95–101. lome in response to mitochondrial depolarization. Nature. 42. Calì T, Ottolini D, Negro A, Brini M. Enhanced parkin levels 2013;496:372–6. 2+ favour ER-mitochondria crosstalk and guarantee Ca transfer to 54. Spillane M, Ketschek A, Merianda TT, Twiss JL, Gallo G. sustain cell bioenergetics. Biochim Biophys Acta.2013;1832: Mitochondria coordinate sites of axon branching through localized 495–508. intra-axonal protein synthesis. Cell Rep. 2013;5:1564–75. 43. Zheng L, Bernard-Marissal N, Moullan N, D’Amico D, Auwerx J, 55. Tang F, Kalil K. Netrin-1 induces axon branching in developing Moore DJ, et al. Parkin functionally interacts with PGC-1alpha to cortical neurons by frequency-dependent calcium signaling path- preserve mitochondria and protect dopaminergic neurons. Hum ways. J Neurosci. 2005;25:6702–15. Mol Genet. 2017;26:582–98. 56. Bergamin G, Cieri D, Vazza G, Argenton F, Mostacciuolo ML. 44. Gautier CA, Erpapazoglou Z, Mouton-Liger F, Muriel MP, Cor- Zebrafish Tg(hb9: MTS-Kaede): a new in vivo tool for studying mier F, Bigou S, et al. The endoplasmic reticulum-mitochondria the axonal movement of mitochondria. Biochim Biophys Acta. interface is perturbed in PARK2 knockout mice and patients with 2016;1860:1247–55. PARK2 mutations. Hum Mol Genet. 2016;25:2972–84. 57. Pauls S, Zecchin E, Tiso N, Bortolussi M, Argenton F. Function 45. Ryan MD, King AM, Thomas GP, Cleavage of foot-and-mouth and regulation of zebrafishnkx2.2a during development of pan- disease virus polyprotein is mediated by residues located within a creatic islet and ducts. Dev Biol. 2007;304:875–90. 19 amino acid sequence. J Gen Virol. 1991;72(Pt 11):2727–32. 58. Ottolini D, Cali T, Brini M. Methods to measure intracellular Ca(2 46. Denison SR, Wang F, Becker NA, Schule B, Kock N, Phillips LA, +) fluxes with organelle-targeted aequorin-based probes. Methods et al. Alterations in the common fragile site gene Parkin in ovarian Enzymol. 2014;543:21–45. and other cancers. Oncogene. 2003;22:8370–8. 59. Pinaud F, Dahan M. Targeting and imaging single biomolecules in 47. Pawlyk AC, Giasson BI, Sampathu DM, Perez FA, Lim KL, living cells by complementation-activated light microscopy with Dawson VL, et al. Novel monoclonal antibodies demonstrate split-fluorescent proteins. Proc Natl Acad Sci USA. 2011;108: biochemical variation of brain parkin with age. J Biol Chem. E201–10. 2003;278:48120–8. 60. Giacomello M, Drago I, Bortolozzi M, Scorzeto M, Gianelle A, 48. Kim JH, Lee SR, Li LH, Park HJ, Park JH, Lee KY, et al. High Pizzo P, et al. Ca2+ hot spots on the mitochondrial surface are cleavage efficiency of a 2A peptide derived from porcine generated by Ca2+ mobilization from stores, but not by activation teschovirus-1 in human cell lines, zebrafish and mice. PLoS ONE. of store-operated Ca2+ channels. Mol Cell. 2010;38:280–90. 2011;6:e18556. 61. Smirnova E, Shurland DL, Ryazantsev SN, van der Bliek AM. A 49. Bannai H, Inoue T, Nakayama T, Hattori M, Mikoshiba K, human dynamin-related protein controls the distribution of mito- Kinesin dependent, rapid, bi-directional transport of ER sub- chondria. J Cell Biol. 1998;143:351–8. compartment in dendrites of hippocampal neurons. J Cell 62. Frank S, Gaume B, Bergmann-Leitner ES, Leitner WW, Robert Sci.2004;117(Pt 2):163–75. EG, Catez F, et al. The role of dynamin-related protein 1, a 50. Villegas R, Martinez NW, Lillo J, Pihan P, Hernandez D, Twiss mediator of mitochondrial fission, in apoptosis. Dev Cell. JL, et al. Calcium release from intra-axonal endoplasmic reticulum 2001;1:515–25. leads to axon degeneration through mitochondrial dysfunction. J 63. Tokuyasu KT. A technique for ultracryotomy of cell suspensions Neurosci.2014;34:7179–89. and tissues. J Cell Biol. 1973;57:551–65. 51. Cui-Wang T, Hanus C, Cui T, Helton T, Bourne J, Watson D, 64. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, et al. Local zones of endoplasmic reticulum complexity confine Pietzsch T, et al. Fiji: an open-source platform for biological- cargo in neuronal dendrites. Cell. 2012;148:309–21. image analysis. Nat Methods. 2012;9:676–82.

Journal

Cell Death & DifferentiationSpringer Journals

Published: Dec 11, 2017

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off