Dissecting mitochondria–nucleus contact sites using SPLICS reveals a distributed and dynamic tethering network Authors: Adamantia Deligiannopoulou, Lucia Barazzuol, Caterina Peggion, Elena Poggio, Michelangelo Campanella, Marisa Brini, Tito Calì Published: 2026 ABSTRACT AbstractMitochondria-nucleus contact sites (also known as nucleus-associated mitochondria, NAMs) are emerging as important platforms for inter-organelle communication; however, their molecular organisation and regulation remain poorly understood. Here, we use split-GFP-based contact site sensors (SPLICSS-P2ANU−MT) to quantitatively dissect the contribution of candidate tethering proteins to the formation of short-range mitochondria–nucleus contacts in HeLa cells. Through systematic overexpression and downregulation approaches, we identify TOM70, MFN2, AKAP95, and the catalytic subunit of PKA as positive modulators of contact formation when individually overexpressed. In contrast, loss-of-function experiments reveal a selective role for TOM70, whose downregulation significantly induces contact site abundance, whereas depletion of other candidates has limited effects. These results suggest that mitochondria-nucleus contacts are not maintained by single dominant tethers but instead rely on the coordinated contribution of multiple proteins. To further investigate cooperative mechanisms, we performed co-expression experiments. Among the combinations tested, TSPO in conjunction with either the regulatory or catalytic subunits of PKA significantly increased contact site formation, indicating that specific protein partnerships are required to establish functional tethering. Overall, our data support a model in which mitochondria–nucleus contact sites are heterogeneous and regulated by a network of interacting factors rather than a single structural component. These findings highlight the complexity of inter-organelle communication and underscore the importance of dissecting both individual and combinatorial roles of contact site components to understand their functional relevance.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s13062-026-00870-9.Keywords: NAM, Organelle communication, SPLICSS, TOM70, Protein kinase A (PKA) and TSPO Introduction Organelle contact sites, also referred to as membrane contact sites (MCSs), are specialized regions where the membranes of distinct organelles come into close proximity without undergoing fusion. These structures have emerged as central hubs for inter-organelle communication, enabling the exchange of lipids, metabolites, and ions, and coordinating key cellular processes such as mitochondrial dynamics, calcium signaling, autophagy, metabolic regulation [1]. While contacts between the endoplasmic reticulum and mitochondria have been extensively characterized, our understanding of the broader landscape of inter-organelle interfaces remains incomplete [2, 3]. Mitochondria-nucleus communication is essential for cellular homeostasis, as it coordinates nuclear gene expression with mitochondrial function, stress responses, and metabolic adaptation. Interestingly, such contacts were observed from yeast to mammals, and in pathogens like Plasmodium falciparum and Toxoplasma gondii [4–6]. The crosstalk between the two organelles is critical for lipid transfer and gene expression via physical contacts, particularly by trafficking excess cholesterol to the nucleus, thereby enhancing the transcriptional activity of the NF-κB signalling pathway, or by stimulating histone acetylation through the interaction of the pyruvate dehydrogenase complex (PDC) with the nuclear lamina protein Lamin A [7–9] thereby determining cellular epigenome [10]. So far, how these are organized, and the different players playing a role remain open questions. A major obstacle in addressing this issue has been methodological. The small size and dynamic nature of contact sites make them difficult to detect, and conventional imaging approaches often fail to distinguish true contacts from mere proximity. The development of genetically encoded proximity reporters, such as SPLICS (split-GFP-based contact site sensors), provides a way to overcome these limitations by enabling the detection of organelle contacts in living cells with defined spatial resolution [11–14]. SPLICSNU-MT reporters allow the selective visualisation of short-range contacts between mitochondria and the nuclear envelope [15], offering a quantitative readout of contact site dynamics. Despite the availability of these tools, the identity and role of the molecular players involved in mitochondria-nucleus contacts remain poorly characterized. Several candidates, including TOM70, MFN2, AKAP95, TSPO, and components of the PKA signaling pathway, have been implicated in organelle positioning, signaling, or membrane organization [5–7, 9, 16], but their contribution to this specific interface has not been systematically tested. More in detail, Tom70 has attracted attention not only for its canonical role as a mitochondrial outer membrane import receptor, but also for its ability to interact with signaling proteins and chaperones, potentially helping to organize communication hubs at the mitochondrial surface [5, 16]. Similarly, AKAP95 has been suggested to participate in mitochondria–nucleus coupling through its scaffolding activity, particularly by anchoring signaling enzymes such as PKA in spatially restricted microdomains [7]. In this context, PKA signaling appears particularly relevant. Anchoring of PKA by AKAP95 proteins could facilitate localized phosphorylation events at mitochondria-nucleus contact regions, thereby coordinating transcriptional responses with mitochondrial metabolic status. TSPO, a mitochondrial outer membrane protein involved in cholesterol transport and stress responses, has also been proposed as a component of mitochondria-associated signaling interfaces with the nucleus [7]. Through interactions with cytoskeletal and signaling proteins, TSPO may contribute to the dynamic positioning of mitochondria relative to the nucleus and to retrograde signaling pathways. Among other mitochondrial proteins that may be involved in the dynamic of nuclear-mitochondria juxtaposition, MNF2 was also demonstrated to be involved in the mitochondria-nucleus tethering, as required for the PDC nuclear import [9]. In addition to the unclear roles of the single candidate tethers, it is not known whether these proteins act independently, redundantly, or cooperatively to establish functional contacts. Here, we use SPLICS NU-MT reporters to directly address these questions. By systematically perturbing candidate proteins through overexpression, downregulation, and co-expression approaches in HeLa cells, we dissect their contribution to mitochondria-nucleus contact formation. Our results challenge the idea of a single dominant tether and instead support a model in which these contact sites emerge from the combined action of multiple, partially redundant players. This suggests that mitochondria-nucleus contacts are more complex and heterogeneous than previously anticipated. Materials and methods Plasmid cloning The SPLICSNU−MT probe in pcDNA3.1 + was generated by subcloning the SPLICSNU−MT sequence in the pRK5 vector (Addgene, Cat#210129), already available in the laboratory, into an empty pcDNA3.1 + plasmid (ThermoFisher, Cat#V79020). Briefly, both vectors were digested with BamHI (NEB, Cat#R0136S) and XbaI (NEB, Cat#R0145S), followed by ligation using T4 DNA Ligase (NEB, Cat#M0202S). The ACBD3 construct tagged at the N-terminus with a FLAG sequence was generated by PCR amplification from the purchased plasmid (Addgene, Cat#80633) using the following primers: FW, 5’-TTACAAGGATGACGACGATAAGCTCAGATCTACCATGGCGGCG-3’; REV, 5’-TCGTCATCCTTGTAATCCATGGTGGCGACCGGTGGGTCTCC-3’. The resulting amplicon was subsequently cloned into the pcDNA3.1 + vector through NheI (NEB, Cat#R3131) and EcoRI (NEB, Cat#R3101) restriction digestion followed by ligation. Cell culture maintenance The HeLa (ATCC) cell lines were cultured in High Glucose Dulbecco’s Modified Eagle Medium (DMEM HG) (Gibco, Cat#41966-029) supplemented with 10% Fetal Bovine Serum (FBS) (Euroclone, Cat#ECS0180L) and 1% Penicillin-Streptomycin (Euroclone, Cat#ECB3001D). Cells were maintained in a humidified incubator at 37 °C with an atmosphere of 5% CO2. Overexpression of mitochondria-nucleus tethers HeLa cells were seeded on 13 mm glass coverslips in 24-well plate at 60% confluency the day prior to transfection. Cells were transfected in a ratio of 1:2 of plasmids encoding SPLICSNU−MT in pRK5 (Addgene, Cat#210129) and tether using a modified calcium phosphate precipitation method as reported by [17]. 8 h after the addition of the transfection mix, cells were washed 3 times with Dulbecco’s Phosphate Buffered Saline (PBS) (EuroClone, Cat#ECB4004L) and fresh medium was added to each well. Plasmid used for overexpression experiments: TSPO, HG17109-UT (Sinobiological); TOM70, Already available ([18]); AKAP95, 138,254 (Addgene); ACBD3, 180,633 (Addgene); Mitofusin 2 ([19]); PKA reg (kind gift of Dr Giulietta Di Benedetto); PKA cat (kind gift of Dr Giulietta Di Benedetto). Downregulation of mitochondria-nucleus tethers HeLa cells (ATCC) were seeded in 6-well plates at high confluency (80–90%), transfected with Lipofectamine 2000 reagent (Invitrogen, Cat#11668027) with 1 µg of DNA SPLICSNU−MT in pcDNA3.1 + and 50 pmol of siRNAs against the tethers investigated, as described above, according to manufactures’ instructions. 6 h post transfection, cells were washed three times with PBS and reseeded with a fresh medium with no antibiotics: 40,000 cells were reseeded onto 13 mm diameter glass coverslips for immunofluorescence analysis, while the remaining cells in 6-well plate for Western blotting (WB) analysis. After 48 h, cells were either collected for WB or fixed for immunostaining. List of the siRNA used: TSPO, GS706 (QIAGEN), pool; TOM70, GS9868 (QIAGEN), n. 1; AKAP95, GS10270 (QIAGEN), pool; ACBD3, GS64746 (QIAGEN), n. 6; Mitofusin 2, GS9927 (QIAGN), pool. Immunostaining When needed, before fixation cells were incubated for 30 min with 100nM MitoTracker™ Red CMXRos (Thermo Fisher Scientific; M7512) at 37 °C and 5% of CO2. MitoTracker was diluted in pre-warmed HBSS (Gibco, Cat#14025). Cells were then fixed for 20 min with 3.7% formaldehyde (Sigma-Aldrich, Cat#F8775) in PBS at room temperature (RT) followed by three washes of 5 min with PBS. Fixed cells were permeabilized with 0.3% Triton X-100 (PanReac AppliChem, Cat#A1388)/PBS for 10 min, followed by three washes of 10 min with 1% gelatin/ PBS (type IV, from bovine skin, (Sigma-Aldrich, Cat#G9382) and three 5 min- washes with PBS. Coverslips were incubated for 2 h with the primary antibody diluted 1:50 in PBS at RT (anti TOM20: Proteintech, Cat#11802-1-AP; anti HA: Proteintech, Cat#51064-2-AP; anti-Flag: Proteintech, Cat#20543-1-AP, anti-myc: Merck Millipore, Cat#05-724). After the incubation, three washes of 10 min with gelatin were followed by three washes of 5 min with PBS. Cells were then incubated for 45 min with the secondary antibody diluted 1:100 in PBS at RT and in the dark (Thermo Fisher: Goat anti-Rabbit IgG AlexaFluor 647;, Cat# A-21244; Goat anti-Mouse IgG AlexaFluor 633, Cat#A-21050). Then, three washes with gelatin and three washes with PBS. Nuclei were stained with Hoechst (Thermo Fischer Scientific, Cat#H3570, 1:10,000), with a 10-minutes incubation followed by two PBS washes. Lastly, coverslips were mounted with Mowiol 40–88 (Sigma-Aldrich, Cat#81386) and imaged using SP5 or MICA Leica confocal microscope using a HCX PL APO 100X/numerical aperture 1.40–0.60 or a Plan-Apochromat 63x, Oil DIC M27 (oil immersion objective). Images were acquired at lasers wavelength of 405, 488, 568, 633–647 nm. Protein extraction and western blot 48 h after transfection, HeLa cells were collected, resuspended in RIPA Lysis Buffer (50 mM Tris-HCl pH 7.4, NaCl 150 mM, 1% Triton X-100, 0.5% sodium deoxycholate, 10 mM EDTA, 0.1% SDS, 1 mM DTT) completed with phosphatase and protease inhibitor cocktails (Sigma-Aldritch, Cat#8340, and incubated on ice for 30 min. Lysates were cleared by centrifugation at 12,000 x g for 30 min at 4 °C, and protein concentration was determined using Bradford assay (Bio-Rad, Cat#500 − 0205). Proteins were then separated by SDS-PAGE and transferred onto PVDF membranes (BioRad, Cat#162–0177)). After incubation with a blocking solution (BS) (5% (w/v) non-fat dry milk (Euroclone; ERM180500) in Tris Buffered solution (TBS) added with 0.1% (w/v) Tween-20 (TBS-T)) (1 h, RT), membranes were probed (overnight, 4 °C) with the required primary antibodies (ab) in TBS-T. The following primary monoclonal ab were used (dilutions are in parentheses): anti-AKAP95 (1:1,000); Abclonal, Cat#A3049; anti-ACBD3 (1:1,000); Abclonal, Cat#A16568; anti-TOM70 (1:1,000); Abclonal, Cat#A4349; anti-TSPO (1:1,000); Abclonal, Cat#A4881, anti-MFN2 (1:1,000); Abnova, Cat# H00009927-M03; anti-PKARI α/β (1:1,000); Santa Cruz, Cat#271125; antiPKA Cat α (1:1,000); Santa Cruz, Cat#28315; anti-actin (1:2,000); Proteintech, Cat#20536-1-AP; anti-β-Tubulin (1:5,000); Proteintech, Cat#66240-1-Ig). The following day, membranes were washed three times (10 min each) with TBS-T and incubated (1 h, 25 °C) with horseradish peroxidase (HRP)-conjugated anti-mouse or anti-rabbit IgG, depending on the primary antibody host specification (anti-rabbit HRP: Sigma-Aldritch, Cat#A6154, 1:5000; anti-mouse HRP: Sigma Aldrich, Cat#A5278, 1:3000) for 1 h at room temperature. After a final set of three washes in TBS-T, membranes were developed using Immobilon® Forte substrate (Millipore, Cat#WBLUF0100). Chemiluminescent signals were captured using the Uvitec Imaging System. Bands intensities were analysed with imageJ [15]. Briefly, to determine the relative protein abundance, densitometric analyses were performed by normalizing the intensity of each immunoreactive band to the to the corresponding signal of the housekeeping protein (i.e., β-Tubulin or actin). Statistical and data analysis Z-stack of the cells were acquired with a z-step of 0.29 μm, then processed using ImageJ (National Institutes of Health (NIH)) following the protocol published by [11]. Briefly, single transfected cells were selected by freehand selection of ImageJ in the drawing/selection polygon tool and then processed using the “Quantification 1” plugin (https://github.com/titocali1/Quantification-Plugins). A 3D reconstruction of the resulting image was obtained using the VolumeJ plugin (https://github.com/titocali1/Quantification-Plugins). A selected face of the 3D rendering was then thresholded and used to count contact sites through the “Quantification 2” plugin (https://github.com/titocali1/Quantification-Plugins). All statistical analyses were performed using GraphPad Prism version 8.00 for Mac OS X (La Jolla, California, USA). Depending on data normality, the appropriate test (ANOVA or Kruskal-Wallis) was applied. Statistical significance threshold was set at p < 0.05. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. The value of n is indicated in the figure legends. Results Overexpression of putative players at the mitochondria-nucleus contact To investigate the contribution of candidate proteins to mitochondria-nucleus contact site formation, we performed overexpression experiments in HeLa cells using SPLICSS-P2ANU−MT reporter, which generates equimolar amounts of the two organelle-targeted SPLICS moieties and selectively detect short-range contacts (~ 10 nm) (Fig. 1a). Cells were co-transfected with SPLICSS-P2ANU−MT and plasmids encoding TOM70, MFN2, TSPO, AKAP95, ACBD3, and the catalytic (PKA Cα) or regulatory (PKA RIα) subunits of PKA, with the latter expressed as a chimera fused to the RFP protein. Contact sites were quantified as SPLICS-positive puncta per cell. Efficient overexpression was confirmed by WB analysis (Fig. 1b). Overexpression of nucleus-mitochondria tethers TOM70, MFN2, AKAP95, and PKA Cα alone increase contact sites. (a) Cartoon of SPLICS P2A short NU-MT. (b) Representative WB and corresponding quantification histogram confirming successful overexpression relative to the mock condition. (c) Representative z-projection images of HeLa cells expressing SPLICS NU-MT and either the empty vector (mock) or one nucleus-mitochondria tether. (d) Quantification of contact sites upon tether overexpression. Kruskal-Wallis test and multiple comparison Dunn’s test were performed (mock n = 134, SD = 19.16, TOM70 n = 45,SD = 37.13, MFN2 n = 67, SD = 34.89, TSPO n = 35, SD = 29.58, AKAP95 n = 94, SD = 29.58, ACBD3 n = 39, SD = 23.34, PKA Cα n = 38, SD = 36.93, PKA RIα n = 18, SD = 15.85) P-value: P < 0.05*, P < 0.01**, and P < 0.001***. The scale bar is 10 μm Representative z-projection images (Fig. 1c) show the distribution of SPLICS signals together with mitochondrial (MitoTracker) or the regulatory subunit of PKA (RFP) and nuclear (Hoechst) markers. Quantification of contact sites on 3D rendered z-stacks (Fig. 1d) revealed that overexpression of TOM70, MFN2, AKAP95, and PKA Cα, whose overexpression was validated by immunostaining (Supplementary Fig. S1), significantly increased the number of mitochondria-nucleus contacts compared to mock-transfected cells. In contrast, TSPO, ACBD3, and PKA RIα did not induce a consistent increase of the SPLICS dots when expressed individually (n° of contacts/cell ± SEM: Mock: 56.09 ± 1.655; TOM70: 79.04 ± 5.535; MFN2: 81.40 ± 4.263; TSPO: 69.43 ± 5.000; AKAP95: 68.90 ± 3.177; ACBD3: 53.05 ± 3.737; PKA CATα: 91.13 ± 5.990; PKA RIa: 52.11 ± 3.737). Although a tendency for an increase in the number of contacts was observed while overexpressing TSPO, no statistically significant increase was observed. These data indicate that multiple proteins with distinct cellular functions can enhance contact site formation when overexpressed, suggesting that mitochondria-nucleus contacts are sensitive to perturbations in both mitochondrial- and nuclear-associated factors. However, the lack of a uniform effect across all candidates already points to a non-redundant and possibly context-dependent contribution of individual players. Downregulation of endogenous players at the mitochondria-nucleus contact To assess the role of endogenous proteins, we performed siRNA-mediated knockdown of TSPO, AKAP95, TOM70, MFN2, and ACBD3 in HeLa cells expressing SPLICSS-P2ANU−MT reporters. Representative images and quantification of 3D-rendered z-stacks from SPLICS-positive puncta (Fig. 2a and b) revealed a striking and selective effect of TOM70 depletion. Specifically, siRNA-mediated knockdown of TOM70 resulted in a significant increase in mitochondria-nucleus contact sites compared to scramble controls. In contrast, downregulation of TSPO, AKAP95, MFN2, or ACBD3 did not lead to significant changes (n° of contacts/cell ± SEM: siRNA scramble: 48.95 ± 4.355; siRNA TSPO: 51.24 ± 5.09; siRNA AKAP: 60.58 ± 5.473; siRNA TOM70: 87.31 ± 9.695. Downregulation of nucleus-mitochondria tethers: downregulation of TOM70 increases the number of contact sites. (a) Representative z-projection images of HeLa cells expressing SPLICS NU-MT and transfected either with scramble siRNA (mock condition) or with siRNAs targeting the indicated nucleus-mitochondria tethers. (b) Quantification of contact sites following tethers downregulation. (c) WB and corresponding histogram confirming successful downregulation., Kruskal-Wallis test followed by multiple comparison Dunn test was performed (siRNA scramble n = 21, SD = 19.96, siRNA TSPO n = 19, SD = 22.19, siRNA AKAP95 n = 19, SD = 23.85, siRNA TOM70 n = 16, SD = 38.78, siRNA MFN2 n = 17, SD = 33.2, siRNA ACBD3 n = 20, SD = 16.95) P-value: P < 0.05*, P < 0.01**, and P < 0.001***. The scale bar is 10 μm siRNA MFN2: 67.21 ± 8.051; siRNA ACBD3: 59.88 ± 3.790). Efficient downregulation was confirmed by Western blot analysis (Fig. 2c). This result is particularly intriguing considering the overexpression data, where TOM70 promoted contact formation. The increase in contacts upon TOM70 depletion suggests a more complex regulatory role for TOM70 in inter-organelle communication, in which physiological levels maintain proper contact homeostasis, while reduced levels may promote the recruitment of alternative tethering factors and excessive levels may directly enhance organelle interactions. More broadly, the absence of strong phenotypes upon depletion of most candidates supports the idea that mitochondria-nucleus contacts are not maintained by a single essential protein but rather by a distributed network of contributors. Co-expression of putative players at the mitochondria-nucleus contact Given the limited and sometimes opposing effects observed in single perturbations, we next explored whether cooperative interactions between proteins are required for efficient tethering. To this end, we co-expressed TSPO with additional candidate players, including ACBD3, AKAP95, and both the catalytic and regulatory subunits of PKA. Representative images (Fig. 3a) show the localization of SPLICS signals in cells co-expressing these combinations. Quantitative analysis (Fig. 3b) revealed that only the co-expression of TSPO with PKA RIα or TSPO with PKA Cα significantly increased the number of mitochondria-nucleus contacts. In contrast, TSPO co-expression with ACBD3 (albeit with a trend towards the increase) or AKAP95 did not produce a significant effect (n° of contacts/cell ± SEM: Mock: 63.81 ± 3.105. Overexpression of TSPO and PKA RIα or TSPO and PKA Cα increase the number of contact sites. (a) Representative z-projection images of HeLa cells expressing SPLICS NU-MT and TSPO together with another nucleus-mitochondria player (PKA RIα, PKA Cα, ACBD3, AKAP95). The immunostaining with anti-TOM20 or MitoTracker helps to observe the mitochondrial network, immunostaining with anti-FLAG, anti-HA observed at 633–647 nm for the overexpressed tagged constructs (anti-HA for AKAP95 and anti-FLAG for ACBD3). Fluorescence at 555 nm is observed for the tagged construct PKA RIα with tagRFP. Hoechst for nuclei staining is observed at 405 nm. (b) Quantification of nucleus-mitochondria contact sites in each condition. Brown-Forsythe and Welch ANOVA multiple comparisons Dunnett T3 were performed (mock n = 32, SD = 18.89, TSPO&AKAP95 n = 18, SD = 21.57, TSPO&ACBD3 n = 28, SD = 24.43, TSPO&PKA RIα n = 26, SD = 30.58, TSPO&PKA Cα n = 32, SD = 42.17). P- value: P < 0.05*, P < 0.01**, and P < 0.001**. The scale bar is 10 μm TSPO and AKAP95: 61.61 ± 5.085; TSPO and ACBD3: 71.98 ± 4.616; TSPO and PKA RIa: 82.87 ± 5.997; TSPO and PKA Cα: 90.30 ± 7.455), however, considering the positive effect of AKAP95 overexpression on contact site number, one might speculate on a possible role of TSPO in counteracting or dampening the effect of AKAP95. These findings indicate that TSPO alone is not sufficient to drive contact formation but requires specific partners, particularly components of the PKA signaling pathway. The specificity of this effect suggests that functional tethering may depend on defined protein complexes rather than simple additive contributions of individual proteins. Mitochondria-nucleus contacts as multi-component and heterogeneous interfaces A schematic summary of the proposed model is shown in Fig. 4. Individual overexpression experiments (Fig. 4b-e) demonstrate that multiple proteins, including MFN2, TOM70, AKAP95, and PKA Cα, can independently enhance contact formation, although the identity of their interacting partners remains unclear (indicated by question marks). In contrast, co-expression experiments (Fig. 4f-g) highlight a specific cooperative interaction between TSPO and PKA subunits, suggesting that defined protein pairs can work synergistically to influence mitochondria-nucleus communication. The increase in contact sites upon TOM70 downregulation (Figs. 2 and 4b) might suggest that these interfaces are dynamically regulated and subject to compensatory mechanisms. Taken together, our data support a model in which mitochondria-nucleus contact sites are heterogeneous, multi-component structures arising from the coordinated action of several proteins rather than a single dedicated tether. This complexity likely reflects the integration of multiple cellular pathways at this interface and suggests that distinct subsets of contact sites may serve different functional roles. Overview of the effect of a single or multiple tethers on the increase in nucleus-mitochondria contact sites. (a) All the potential players of NAMs. (b) Single (left) or combined (right) tether overexpression leading to an increase in nucleus-mitochondria contact sites: from left to right MFN2 overexpression, TOM70 overexpression, AKAP95 overexpression, PKA Cα overexpression; TSPO and PKA Cα co-overexpression, and TSPO and PKA RIα co-overexpression. The question mark indicates the unknown interacting partner localized either at the mitochondrial or nuclear compartment Discussion Mitochondria-nucleus communication has long been interpreted as a predominantly signaling-driven process mediated by transcriptional and metabolic feedback loops [7, 15, 16, 20]. This form of communication also relies on direct physical contact [7, 15]; however, the lack of appropriate tools to detect such interactions has limited the investigation of these contacts under both physiological and pathological conditions, leaving them largely unexplored. By using the SPLICSS-P2ANU−MT reporter [15], we provide direct evidence that mitochondria-nucleus contacts are dynamically regulated in living cells. More importantly, our data challenge the idea that these contacts are defined by a single or dominant molecular tether. A key observation emerging from this study is the discrepancy between gain- and loss-of-function approaches. Several proteins, including TOM70, MFN2, AKAP95, and PKA Cα, increase contact sites when overexpressed, yet their depletion does not produce a corresponding decrease. This asymmetry is inconsistent with a classical tethering model, in which specific components are both necessary and sufficient to establish inter-organelle contacts. Instead, our findings support a model in which mitochondria–nucleus contacts arise from a distributed system of partially redundant and context-dependent interactions. The behavior of TOM70 exemplifies this complexity. While its overexpression increases contact sites, its downregulation also results in an increase, suggesting a non-linear and potentially regulatory role. One possible explanation is that TOM70 participates in a broader network involved in the organization of mitochondria–nucleus contacts, such that its removal may trigger compensatory responses or alter the distribution of other factors contributing to contact formation [16]. However, this remains speculative, as the potential effects of TOM70 modulation on other candidate contact-site components were not directly investigated in the present study. More generally, these observations are compatible with the view that contact sites may behave as dynamic assemblies capable of adapting to cellular perturbations, although additional experimental evidence will be required to establish the underlying mechanisms. Furthermore, it remains unclear whether manipulation of individual components triggers compensatory changes in other candidate regulators of mitochondria–nucleus contacts. Addressing this possibility will be important for understanding the regulatory hierarchy and molecular plasticity of these contact sites. Co-expression experiments further reveal that not all protein combinations are functionally equivalent. The selective effect observed for TSPO when co-expressed with PKA RIα or AKAP95 suggests that contact formation depends on specific protein partnerships rather than simply increased local protein abundance. This finding supports a model in which signaling components actively contribute to the establishment or stabilization of contacts. In this context, the involvement of PKA is particularly notable, as it raises the possibility that mitochondria–nucleus contacts are regulated by phosphorylation-dependent mechanisms [21]. Such a framework implies that distinct subsets of contact sites may differ in composition and function, potentially reflecting specific cellular states or signaling environments. This heterogeneity may also explain why perturbation of individual components often results in modest or unexpected phenotypes. Future studies should combine proximity measurements with functional readouts—such as calcium flux, metabolic exchange, or transcriptional responses—to define the physiological relevance of these contacts [22]. More broadly, our findings raise a conceptual issue that extends beyond mitochondria-nucleus interactions. If contact sites are indeed emergent properties of multiple, partially redundant interactions, then the search for a single “tether” may be fundamentally misguided. Instead, it may be more appropriate to view these interfaces as dynamic platforms assembled through combinatorial interactions between structural and signaling proteins. In this framework, perturbations would not simply disrupt a single complex but would reshape the overall architecture of the contact site network. Our data address the basal contribution of several candidate components, which may vary under stress conditions such as staurosporin treatment or in other cellular contexts [7]. At the same time, they point to an intrinsic heterogeneity of these contacts that is likely cell type- and context-dependent. Dissecting how these pathways intersect with organelle positioning and membrane dynamics will be essential to understanding the functional significance of mitochondria–nucleus contacts. Ultimately, this will require integrative approaches capable of linking structure, dynamics, and function. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Author contributions Writing original draft: T.C.; Conceptualization: T.C. and MB.; Investigation: A.D., E.P., L.B., C.P.; Writing, review and editing: T.C., M.B., E.P., L.B., C.P., M.C.; Methodology: A.D., E.P., L.B., C.P.; Resources: T.C. and M.B.; Data curation: A.D.; Supervision: T.C. and M.B.; Formal analysis: A.D.; Project administration: T.C. and M.B.; Funding acquisition: T.C., M.C. and M.B. Funding The work is supported by grants from the Ministry of University and Research (PRIN2017 no 2017E5L5P3 and PRIN2022 no 2022XKAAZ7 to T.C. and PRIN2022 no 20223ABZ82 to M.B.), from the Università degli Studi di Padova (STARS Consolidator Grant 2019 to T.C., Progetto di Ateneo 2023 no. CALI_BIRD23_01 to T.C.), NIH Award 5R01HL168178-02 Subaward to T.C., European Union—Next-Generation EU PNRR – CN3 National Center for Gene Therapy and Drugs based on RNA Technology to M.B. and T.C., n. CN00000041 (2022-26), from CARIPARO foundation (Excellence Project 2023 to M.B, 2024–2027). The research activities led by M.C. are supported by the following funders, who are gratefully acknowledged: Biotechnology and Biological Sciences Research Council [grant numbers BB/M010384/; BB/N007042/1; BB/Z516417/1;] The European Research Council COG 2018-819600_FIRM; Fondation ARC pour la Recherche sur le Cancer ARCLEADER2022020004901; AIRC-MFAG 21903, Rotary Foundation, MIUR (PRIN-2022 Prot. 202252ZLSX) and The Barts Charity (Research Grant/G-002762). Data availability No datasets were generated or analysed during the current study. Declarations Ethical approval Not applicable. Consent for publication All authors have read and agreed to the published version of the manuscript. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Marisa Brini, Email: marisa.brini@unipd.it. Tito Calì, Email: tito.cali@unipd.it. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement No datasets were generated or analysed during the current study.