Mitochondria-Lysosome Contact Sites: Emerging Players in Cellular Homeostasis and Disease

Abstract

Mitochondria and lysosomes regulate a multitude of biological processes that are essential for the maintenance of nutrient and metabolic homeostasis and overall cell viability. Recent evidence reveals that these pivotal organelles, similarly to others previously studied, communicate through specialized membrane contact sites (MCSs), hereafter referred to as mitochondria-lysosome contacts (or MLCs), which promote their dynamic interaction without involving membrane fusion. Signal integration through MLCs is implicated in key processes, including mitochondrial fission and dynamics, and the exchange of calcium, cholesterol, and amino acids. Impairments in the formation and function of MLCs are increasingly associated with age-related diseases, specifically neurodegenerative disorders and lysosomal storage diseases. However, MLCs may play roles in other pathological contexts where lysosomes and mitochondria are crucial. In this review, we introduce the methodologies used to study MLCs and discuss known molecular players and key factors involved in their regulation in mammalian cells. We also argue other potential regulatory mechanisms depending on the acidic lysosomal pH and their impact on MLC's function. Finally, we explore the emerging implications of dysfunctional mitochondria-lysosome interactions in disease, highlighting their potential as therapeutic targets in cancer.

Introduction

Over the last two decades, our understanding of mitochondria, lysosomes, and their crosstalk, has evolved substantially. Originally defined as the cell's “powerhouses” and “recycling centers”, respectively, mitochondria and lysosomes are now recognized as key regulatory and signaling hubs that can sense and dynamically respond to changes in the metabolic and redox status of the cell by generating adaptive responses that preserve cellular homeostasis (Rizzollo et al., 2021; Settembre and Perera, 2024; Suomalainen and Nunnari, 2024). To properly perform these coordinated functions, mitochondria and lysosomes engage in physical interactions and communicate with each other through specialized membrane contact sites (MCSs) (Wong et al., 2018). MCSs are membrane subdomains where organelles can dynamically and physically engage through the interaction between tethering molecules on the surface of each organelle without undergoing fusion (Voeltz et al., 2024).

While membrane contacts between the endoplasmic reticulum (ER) and the mitochondria or the plasma membrane have been extensively characterized (Bernhard and Rouiller, 1956; Copeland and Dalton, 1959; Dennis and Kennedy, 1972; Rizzuto et al., 1998; Pichler et al., 2001; Peretti et al., 2008), the contact sites involving lysosomes have only recently been discovered. Interactions between the ER and lysosomes were identified first (Pan et al., 2000), while lysosome-mitochondria associations have been described more recently and have become an area of intense research focus (Wong et al., 2018).

Increasing evidence highlights that the crosstalk between these two organelles via the formation of mitochondria-lysosome contacts (MLCs) is essential for maintaining cellular homeostasis, metabolic signaling, and mitochondrial quality control. It is now recognized that MLCs play a pivotal role in several key processes, including – but not limited to – mitochondrial fission and dynamics (Wong et al., 2018), as well as the exchange of metabolites and ions, such as cholesterol (Höglinger et al., 2019; Boutry and Kim, 2021; Juhl et al., 2021), amino acids (Peng et al., 2023), and calcium (Peng et al., 2020; Giamogante et al., 2024).

Importantly, dysregulation of lysosomal and mitochondrial interactions at these membrane interfaces is increasingly linked to ageing and age-related diseases, including neurodegenerative disorders (Deus et al., 2020). In this review, after a brief introduction to the state-of-the-art methodologies to study MLCs, which mostly involve advanced microscopy-based techniques, we discuss the molecular players and key factors involved in their regulation in mammals and the emerging implications of dysfunctional mitochondria-lysosome appositions in diseases.

Imaging of Mitochondria-Lysosome Contact Sites

Characterizing MLCs is particularly challenging due to their dynamics and highly transient nature (Wong et al., 2018; Scorrano et al., 2019; Giamogante et al., 2024). To define an MLC, it is crucial to consider both the distance between the two organelles (≤10 nm) (Wong et al., 2018; Giamogante et al., 2024) and the duration of their proximity (>10 s) (Wong et al., 2018). Several methods recognized as the gold standards for MCS characterization, largely relying on advanced microscopy techniques, have been adapted and applied specifically to study MLCs, offering valuable insights into their structure, dynamics, and function. However, each of these methodologies has its limitations (Scorrano et al., 2019), and employing complementary microscopy techniques can provide the most reliable results.

Electron microscopy (EM) techniques, like Transmission Electron Microscopy (TEM) and Correlative Light and Electron Microscopy (CLEM), are instrumental in defining the ultrastructure of MCSs and allow accurate visualization of the surrounding cellular context. Transmission Electron Microscopy, often used in combination with biochemical staining techniques, including horseradish peroxidase (HRP), genetic tags like APEX, or immunogold labeling, is still considered the gold-standard technique in the field of MCSs, as it allows to investigate the ultrastructural architecture of contact sites and identify specific proteins within these interfaces. However, TEM-based techniques require cell fixation, which may introduce artifacts. CLEM-based approaches, combining imaging of fluorescent markers (both in fixed and live cells) with high-resolution 3D electron microscopy, are particularly powerful techniques for the study of mitochondria-lysosome dynamics (Fermie et al., 2018; Jung and Mun, 2024).

Finally, cryo-EM provides an unprecedented tool to inspect MCSs in their native state at the highest resolution and obtain 3D architectures of protein assemblies on membranes, overcoming some of the limitations of standard electron microscopy techniques. However, beyond its advantages, challenges, e.g., in sample preparation and the need for specialized equipment and expertise, make cryo-EM less accessible in many research contexts.

Conversely, super-resolution microscopy techniques, such as Structured Illumination Microscopy (SIM), Stimulated Emission Depletion Microscopy (STED), and Stochastic Optical Reconstruction Microscopy (STORM), are widely used in this field for visualizing lysosome-mitochondria contacts at a resolution below the diffraction limit (∼10–20 nm) in live cells, enabling the measurement of contact durations. Dual-labeling of lysosomes and mitochondria with specific dyes (e.g., LysoTracker and MitoTracker) or genetically encoded markers (e.g., LAMP1-GFP for endosomes/lysosomes and mito-DsRed for mitochondria) is generally used for live-cell imaging of contact dynamics. These markers, used in conjunction with time-lapse microscopy, allow analysis of the frequency, duration, and mobility of lysosome-mitochondria contacts. While offering high resolution, live-cell imaging using techniques like STORM and SIM, suffers from several drawbacks, including potential phototoxicity and the need for specialized and costly equipment.

Förster Resonance Energy Transfer (FRET) has been used to measure particularly close proximities (1–10 nm) between fluorescently tagged proteins on lysosomes and mitochondria in live cells. By tagging proteins with compatible fluorophores, FRET enables the detection of energy transfer between lysosome- and mitochondria-associated proteins, indicating contact site formation (Wong et al., 2018). Despite its high sensitivity, FRET experiments can be technically challenging, as they require meticulous controls, as well as the equimolar expression of the FRET pair.

Recently, a split-GFP reporter (also called SPLICS) for mitochondria-lysosome proximity has been generated and characterized for studying MLCs in vitro and in vivo (Giamogante et al., 2024). Split-GFP or split-mCherry systems have been widely used to study contacts between two organelles by attaching one-half of a fluorescent protein to proteins on the outer membranes of the two interacting organelles (Cieri et al., 2018; Vallese et al., 2020; Calì and Brini, 2021b, 2021a). Fluorescence reconstitution occurs only when these two proteins are close together, and at a specific distance, enabling the selective visualization of contact sites. While SPLICS reporters have proven valuable in detecting contact site formation, a major limitation is the irreversibility of the GFP reconstitution. Indeed, once the contact is formed and EGFP-marked, the signal remains fixed, preventing the observation of dynamic changes in the formation and dissociation of the MCSs. To address this weakness, newer split-protein tools, such as Contact-FP, have been more recently developed, allowing reversible protein-protein interactions and, consequently, a more dynamic tracking of MCSs (Miner et al., 2024).

In the future, the advent of correlative microscopy techniques that combine high-resolution imaging with functional readouts, such as calcium signaling or metabolic activity, holds significant potential for studying the dynamic interplay between mitochondria and lysosomes. Additionally, the integration of AI-powered image analysis tools will be crucial for facilitating the analysis of large datasets and extracting meaningful information from complex images.

Molecular Mechanisms Regulating Mitochondria-Lysosome Contact Formation and Functioning

MLCs are highly dynamic interfaces established at an average intermembrane distance of approximately 10 nm (Wong et al., 2018), which rapidly form and dissolve under physiological conditions (Cioni et al., 2019; Cantarero et al., 2021; Kim et al., 2021). A key component of the tethering machinery controlling mitochondria-lysosome contact dynamics is the small GTPase Rab7 (Wong et al., 2018) (Figure 1). Rab7 GTP hydrolysis is critical for the maintenance of lysosomal acidification, endosomal trafficking, and for the transition of early to late endosomes (Guerra and Bucci, 2016). In its active GTP-bound state, Rab7 localizes to lysosomes and interacts with potential effector proteins on mitochondria, promoting the formation of MLCs. Upon GTP hydrolysis, Rab7's conversion to its GDP-bound state drives the disassembly of these contacts and mitochondria fission (Nagashima et al., 2020). This process is regulated by the Rab7 GTPase-activating protein (GAP) TBC1D15, which binds to the outer mitochondrial membrane protein Fis1 (Wong et al., 2018). Expression of a constitutively active Rab7 (Q67L-GTP) mutant, which cannot undergo GTP hydrolysis, enhances both the number and stability of these contacts, indicating prolonged tethering (Wong et al., 2018). Similarly, inhibition of Rab7's GTP hydrolysis by a GAP-domain mutant of TBC1D15 (D397A) results in inefficient contact untethering (Wong et al., 2018). The formation and untethering of MLCs governed by the Rab7-TBC1D15 interaction is intimately coupled to lysosome and mitochondria dynamics, particularly mitochondria fission. Lysosomes engaged in MLCs have been shown to mark the sites of mitochondrial fission events (Wong et al., 2018), and inhibition of Rab7 GTP hydrolysis leads to significant alterations in organelle dynamics, highlighting the relevance of this pathway in cellular homeostasis.

Figure 1.

Left: Mechanism regulating Mitochondria-Lysosome Contact (MLC) tethering and untethering. Right: Schematics of transporters, channels, and other proteins involved in the trafficking of calcium (Ca²⁺), amino acids (AA), cholesterol, and iron (Fe²⁺) at the MLCs. In the third panel, the ER is also depicted.

However, MLCs do not appear to be directly involved in the degradation of mitochondria by mitophagy (Pickles et al., 2018) or the lysosomal transfer of mitochondrial-derived vesicles (Sugiura et al., 2014). In line with this, the molecular composition of MLCs differs from that of canonical autophagy or mitophagy. Autophagosome formation or lysosomal engulfment of bulk mitochondria or intermembrane proteins are not associated with these sites (Wong et al., 2018). Moreover, MLC formation is not perturbed by genetic deficiency of different autophagy receptors (Wong et al., 2018). Together these findings suggest that lysosomal degradation of mitochondria and MLCs are regulated by different mechanisms.

However, new evidence indicates a more nuanced relationship between these processes. A recent report, using the SPLICS reporter system for the analysis of MLCs, revealed the existence of functionally distinct short- (∼4 nm) and long-range (∼10 nm) interfaces, suggesting a molecular and functional heterogeneity (Giamogante et al., 2024). Consistent with this assumption, these two types of contacts responded differently to specific stimuli and changes in the levels of tethering and untethering factors, being only the short MLCs dependent on the Rab7/TBC1D15 pathway (Giamogante et al., 2024). Specifically, starvation-induced autophagy or PINK1-Parkin-mediated mitophagy evoked by mitochondrial depolarization, increased the narrow MLC interfaces, whereas the long contacts remained unaltered (Giamogante et al., 2024), suggesting that these MLCs may be co-regulated by autophagy pathways. However, the underlying mechanisms governing the differential regulation of short and long MLCs, as well as their precise roles in mitophagy or other cellular processes, remain to be fully elucidated. Interestingly, in homeostasis, the E3 ubiquitin ligase activity of Parkin (PARK2) plays a critical role in enhancing the recruitment of Rab7 to the lysosome and stabilizing active GTP-bound Rab7 independent of its role in stress-induced mitophagy (Peng et al., 2023). In line, Parkinson's disease (PD)-associated mutations in Parkin abrogate Rab7-dependent MLC formation, leading to lysosomal accumulation of several essential (isoleucine/leucine, methionine, phenylalanine, threonine, homoserine, tryptophan, and valine) and non-essential (arginine and tyrosine) amino acids, with consequent reduction in their mitochondrial levels (Peng et al., 2023) (Figure 1). Together, these studies highlight the critical role of Parkin in the regulation of MLCs both under homeostasis and stress conditions evoking mitophagy.

Future research should identify shared regulatory pathways and assess whether potential therapeutic compounds targeting disease-driven perturbations in MLCs may also impact quality control mechanisms such as mitophagy and their consequences.

Beyond the established role of Rab7, mitochondria-lysosome contacts are regulated by additional tethering proteins and pathways in a context-dependent manner and often involving a broader network of inter-organelle contact sites.

The outer mitochondrial membrane protein Mitofusin 2 (MFN2) has been implicated in the tethering of mitochondria and lysosomes since MFN2 silencing reduced the number of MLCs in erythroid progenitors (Khalil et al., 2017).

Although primarily implicated in the transfer of cholesterol at the ER-lysosome contacts, the late endosome (LE)/Lysosomal transporter StAR-related lipid transfer domain-3 (STARD3) promotes cholesterol delivery to mitochondria (Charman et al., 2010; Höglinger et al., 2019) (Figure 1). Depletion of STARD3 in fibroblasts from Niemann-Pick type C (NPC) patients drastically reduces MLCs, indicating an essential role for STARD3 in tethering these organelles (Höglinger et al., 2019).

Localized to the lysosomal membrane, NPC1 is crucial for exporting cholesterol out of lysosomes (Davies and Ioannou, 2000; Blom, 2003). Also, NPC1 has a role in tethering LE/lysosomes to the ER for cholesterol trafficking between these organelles (Höglinger et al., 2019) (Figure 1). In NPC1-deficient cells, reduced ER-lysosome contacts lead to cholesterol accumulation in the lysosomes, which consequently form more contacts with mitochondria. This likely contributes to the mitochondrial cholesterol accumulation and dysfunction observed in NPC (Höglinger et al., 2019). A recent report using the newly generated SPLICS MLC reporter confirmed this mechanism (Giamogante et al., 2024). Treatment with the NPC1 inhibitor U18666A induced cholesterol accumulation in lysosomes and increased the number of mitochondria-lysosome short contacts (Giamogante et al., 2024).

Mitochondria-lysosome contacts also play a significant role in regulating intracellular calcium dynamics by facilitating the direct transfer of calcium from lysosomes to mitochondria (Peng et al., 2020) (Figure 1). At these contact sites, the lysosomal calcium channel TRPML1 mediates calcium efflux from lysosomes to adjacent mitochondria (Li et al., 2017; Peng et al., 2020). Activating TRPML1 with agonists like ML-SA1, selectively increased calcium levels in mitochondria that were in direct contact with lysosomes (Peng et al., 2020). Conversely, the expression of a dominant-negative TRPML1 mutant resulting in a defective pore reduced mitochondrial calcium influx compared to wild-type TRPML1 (Peng et al., 2020).

Ultimately, mitochondrial cholesterol and calcium flux are regulated by the dynamic status of contact sites between the ER, the key organelle where cholesterol is produced and calcium is stored, and the mitochondria (often referred to as ERMCS or MAMs) (Sassano et al., 2022). The formation of three-way contacts between the ER-mitochondria-lysosomes (Boutry and Kim, 2021) indicates the existence of a complex interplay between these organelles, for the coordination and integration of cellular responses to various needs and stress conditions.

A better understanding of this sophisticated inter-organelle cooperation will be critical for designing novel therapeutic interventions that correct lipid and calcium homeostasis defects in relevant disease conditions (as discussed below).

Finally, a potential role for MLCs in iron transfer has been suggested (Khalil et al., 2017; Rizzollo et al., 2024) (Figure 1). While the exact mechanism for iron exchange between lysosomes and mitochondria has yet to be fully elucidated, a so-called “kiss-and-run” mechanism promotes iron transfer from early endosomes to mitochondria (Sheftel et al., 2007; Das et al., 2016; Hamdi et al., 2016). This finding suggests that a similar pathway could be functionally involved in the exchange of iron between the lysosomes and mitochondria. Because TRPML1 is permeable to iron, this channel may contribute to its transfer to mitochondria via MLCs, although additional mediators remain unidentified.

Recent findings also indicate that the maintenance of acidic lysosomal pH is critical for the transfer of iron to mitochondria (Yambire et al., 2019; Hughes et al., 2020; Weber et al., 2020; Rizzollo et al., 2024). Alterations of the lysosomal pH may impair pH-dependent mechanisms, such as the release of endocytosed iron from transferrin receptors (TfR) and the function of lysosomal transporters responsible for the export of iron. Since these transporters may also serve as intermediaries for iron transfer to mitochondria via MLCs, lysosomal pH dysregulation could broadly impact mitochondria-lysosome communication. It remains unknown, however, whether lysosomal acidity also directly influences the formation of MLCs themselves.

Mitochondria-Lysosome Contact Sites in Disease

Recent studies have identified disruptions in the formation and function of MLCs as contributors to the pathogenesis of several neurodegenerative diseases, including PD (Burbulla et al., 2019; Kim et al., 2021; Peng et al., 2023; Giamogante et al., 2024), Charcot-Marie-Tooth (CMT) disease (Wong et al., 2019; Cantarero et al., 2021), and lysosomal storage disorders, such as Mucolipidosis type IV (MLIV) (Peng et al., 2020) and NPC (Höglinger et al., 2019) (Figure 2). Mutations in genes essential for the formation and function of these contact sites impair critical cellular processes, such as the exchange of metabolites, calcium signaling, and mitochondrial quality control – processes that are vital for neuronal health and function. Consequently, modulation of lysosome-mitochondria contact sites represents a promising therapeutic target, offering potential avenues to mitigate or prevent neurodegenerative disease progression through interventions aimed at restoring or enhancing inter-organelle communication.

Figure 2.

Diseases where a role for MLCs has been already described (blue) versus diseases where MLCs might play a role, but a direct link is still missing.

Although less is known about MLCs in cancer, recent insights suggest they may be important for tumor growth and survival (Weber et al., 2020; Rizzollo et al., 2024) (Figure 2). Cancer cells often rewire metabolic pathways to meet their high energy demands during proliferation but also reprogram their metabolism to support their phenotypic plasticity during processes, such as epithelial-to-mesenchymal transition (EMT), that promote cancer progression and therapy resistance (Finley, 2023; Yang et al., 2024). MLCs, which facilitate the transfer of metabolites and ions, support metabolic flexibility and may contribute to the metabolic shifts needed for rapid cancer cell proliferation and invasion. For instance, calcium transfer from lysosomes to mitochondria at MLCs (Peng et al., 2020) can enhance mitochondrial oxidative phosphorylation, potentially boosting energy production to meet tumor growth demands (Zheng et al., 2023), but also promoting the drug-tolerant persister phenotype (Adiga et al., 2022).

Moreover, lysosomal acidity, which is crucial for cancer cell proliferation (Weber et al., 2020), could also affect MLC function. Acidic lysosomes regulate iron homeostasis, and when acidification is lost, iron becomes sequestered within lysosomes, reducing mitochondrial iron availability and altering mitochondrial function (Yambire et al., 2019; Hughes et al., 2020; Weber et al., 2020; Rizzollo et al., 2024), potentially generating new therapeutically exploitable vulnerabilities. In line, lysosomotropic agents that disrupt lysosomal pH, such as hydroxychloroquine, have shown anticancer efficacy in preclinical and clinical studies (Yang et al., 2011; Amaravadi et al., 2019; Low et al., 2023). Whether lysosomal pH changes during cancer progression, such as EMT and invasion, and how these changes influence MLCs and cellular metabolism warrants further investigation. Notably, loss of proper lysosomal function and autophagic dysfunction are common features of both cancer and other ageing-related diseases.

Notably, cancer cells rely on altered lipid metabolism to support membrane synthesis and signaling molecule production (Szlasa et al., 2020; Xiao et al., 2023). For example, cholesterol is a key component of cell membranes, contributing to their integrity and fluidity (Szlasa et al., 2020; Xiao et al., 2023). Additionally, cholesterol is enriched in lipid rafts, which are plasma membrane subdomains with a role in signal transduction and cellular communication (Simons and Ehehalt, 2002). Finally, cholesterol plays important functions within mitochondria, being involved in mitochondrial membrane stabilization, as well as oxidative phosphorylation, which may support the metabolic needs of proliferating cancer cells (Finley, 2023; Yang et al., 2024). Because of this, in physiological conditions, tight regulation of cholesterol levels and organellar trafficking is fundamental to maintaining mitochondrial function and overall cellular homeostasis (Höglinger et al., 2019; Boutry and Kim, 2021; Juhl et al., 2021). Congruently, abnormal accumulation of cholesterol in mitochondria, in part via dysfunction of MLCs, has been linked to several pathologies, like lysosomal disorders and neurodegeneration (Goicoechea et al., 2023), and may also be crucial for cancer biology and therapy resistance (Garcia-Ruiz et al., 2021).

Conclusion and Future Perspectives

In the last decade, MLCs have become a focus of intense research. This is not only due to the fundamental physiological roles that lysosomes and mitochondria play but also to accumulating evidence indicating that perturbations in the mitochondria-lysosome communication underlie the pathophysiology of several diseases. Important links between the dysregulation of MLCs have been established with neurodegenerative and lysosomal storage disorders (Burbulla et al., 2019; Höglinger et al., 2019; Wong et al., 2019; Peng et al., 2020, 2023; Cantarero et al., 2021; Kim et al., 2021; Giamogante et al., 2024). However, we argue that MLC's dysfunction may also be implicated in a broader range of pathologies associated with lysosomal and mitochondrial abnormalities, including inflammation (Yambire et al., 2019; Mihaylova et al., 2024), autoimmunity (Bonam et al., 2019; Zhang et al., 2021), and kidney disease (Surendran et al., 2014; Jia et al., 2024). From a mechanistic point of view, are disease-linked MLC perturbations associated with or driven by alterations in lysosomal pH, activity, and storage capacity? Answering this question could establish direct links to several diseases, including cancer (Weber et al., 2020).

Furthermore, lysosomes and mitochondria not only interact with each other but can also simultaneously engage in tripartite contact sites with other organelles, such as the ER (Boutry and Kim, 2021). In yeast perturbation of ER-mitochondria (ERMES) tethering increases compensatory vacuole-mitochondria interactions through the tethering protein Lam6, which colocalizes with these three organelles (Elbaz-Alon et al., 2015). However, whether similar inter-organelle compensatory mechanisms operate in mammalian cells and under which conditions remains elusive. This complexity adds to the challenge of studying MLCs. Therefore, advancements in tools to study these contact sites are essential. While EM and super-resolution microscopy remain the gold standards for defining MLCs, the development of more user-friendly tools, such as the SPLICS (Giamogante et al., 2024), holds significant potential due to their ability to provide fast and reliable monitoring. Also, these reporters revealed the existence of short and long MLCs; however, some key questions remain open: what are the molecular effectors of long MLCs? Which type of cellular metabolic demand or stress signals do short and long MLCs specifically respond to?

In addition to reporters for MLC monitoring, there is a need for tools to modulate them. For instance, rapamycin-induced dimerization systems can bring together proteins from different organelles to create contact sites, while the expression of spacers can disrupt these contacts (Scorrano et al., 2019; Tei and Baskin, 2021).

Finally, it is increasingly evident that heterogeneous populations of lysosomes and mitochondria exist within cells (Pekkurnaz and Wang, 2022; Ryu et al., 2024; Bond et al., 2025).This raises some outstanding questions: does this diversity reflect a plasticity in the shape, dynamics, and molecular composition of the MLCs? Is the heterogeneity in MLCs a response to the cell's metabolic demands or the various cellular phenotypes that emerge at different stages of the disease?

We assume that at the pace at which new knowledge and tools to study MLCs are becoming available to the scientific community, these outstanding questions will be solved in the near future, opening new therapeutic possibilities for the modulation of MLCs in different physio-pathological settings.