Mitochondrial adventures at the organelle society.

Mitochondria are constantly communicating with the rest of the cell. Defects in mitochondria underlie severe pathologies, whose mechanisms remain poorly understood. It is becoming increasingly evident that mitochondrial malfunction resonates in other organelles, perturbing their function and their biogenesis. In this manuscript, we review the current knowledge on the cross-talk between mitochondria and other organelles, particularly lysosomes, peroxisomes and the endoplasmic reticulum. Several organelle interactions are mediated by transcriptional programs, and other signaling mechanisms are likely mediating organelle dysfunction downstream of mitochondrial impairments. Many of these organelle crosstalk pathways are likely to have a role in pathological processes.

Introduction

Mitochondria are the energy factory of the cell. This sentence was repeated, in many stylistic variations, in schoolbooks, reference books and articles for decades. While undoubtedly correct, the designation “energy factory” for mitochondria is simplistic and incomplete. In addition, mitochondria are major Ca2+ stores, assembly lines for Fe-S and participants in the synthesis of many other cellular components, and are nowadays recognized as key signaling platforms impacting fundamental processes such as cell division, differentiation, anti-viral signaling, autophagy and death [1], [2], [3], [4], [5], [6]. The understanding of mitochondria as a signaling organelle started with the discovery that under certain challenges mitochondria release their Ca2+ content, and was further solidified by the identification of mutations in mitochondrial DNA (mtDNA) as cause of disease [7], [8], and with the understanding that mitochondria communicate with the cellular signaling environment, regulating gene expression programs [9]. While the term “mitochondrial signaling” was originally referring to the pathways employed by mitochondria to affect gene expression, it is now clear that mitochondrial signaling impacts more than just gene expression [9], [10], [11], [12], [13]. While several pathways capable of relaying mitochondrial signaling, both in physiological and pathological conditions, were identified in the last two decades, many fundamental aspects remain unclear. A comprehensive understanding of mitochondrial signaling requires several key factors, such as where the signal originates, what is its molecular identity, and how is it sensed outside mitochondria [10]. A more detailed discussion on the framework of mitochondrial signaling is beyond the scope of this review, and has been subject of attention by many researchers [10], [14], [15].

One of the aspects of mitochondrial signaling that is less well understood pertains to the response of other organelles to mitochondrial malfunction and, reciprocally, to the effect of other organelles on mitochondrial function. This review will focus on how mitochondria communicate with other organelles, and how this communication is involved in the pathology of mitochondrial diseases.

Cross-talk between mitochondria and other organelles

The traditional approach to study mitochondrial signaling has been focused on mitochondria and on the signaling pathways triggered by mitochondrial stress that eventually affect nuclear gene expression. However, there is a wealth of evidence that the picture is significantly more complicated than mitochondria → signaling cascades → gene expression. Mitochondria are constantly interacting with other organelles via signaling pathways, and in some occasions even through physical contact sites [16]. Of these, the contact sites between mitochondria and endoplasmic reticulum (ER) are pivotal for the regulation of many cellular functions, such as Ca2+ homeostasis, and have been implicated in autophagy initiation and in marking the sites for mitochondrial division [17], [18], [19]. But the mitochondrial “social organelle network” doesn't stop here: the peroxisomes receive lipid and protein components from ER, and share DRP1 (dynamin-related protein 1, a key fission regulator), Fis1 and other proteins with mitochondria [20]. Some pathways (e.g., PGC1α and PPARγ) promote the biogenesis of both mitochondria and peroxisomes [20]. Recently, peroxisomal biogenesis was shown to involve components from both ER and mitochondria [21]. Damaged mitochondria and damaged or excess peroxisomes are removed by selective autophagy, which is dependent on lysosomal function [22]. Destabilization of the lysosomal membrane generates a cross-talk between lysosomes and mitochondria which promotes apoptosis [23]. In addition to the vesicle traffic released from mitochondria to lysosomes and peroxisomes, discussed above, it remains to be determined if the contact sites between mitochondria-vacuoles in yeast also exist in higher eukaryotes.

The relevance of the interactions between mitochondria and other organelles is not a mere academic curiosity: genetic defects in mitochondrial proteins cause a group of diseases referred to as “mitochondrial diseases”, in which lysosomes and peroxisomes are known to be often affected structurally and functionally. Furthermore, many peroxisomal and lysosomal diseases present secondary mitochondrial perturbations. For example, many lysosomal storage diseases have perturbed peroxisomal metabolism and mitochondrial function [24], [25], [26], [27]. Peroxisomal diseases (e.g., Zellweger syndrome) often lead to perturbations in mitochondrial structure, redox balance and metabolism [20], [28]. Reciprocally, saturation of the lysosomal capacity is often observed in mitochondrial diseases, with accumulation of dysfunctional lysosomes and autophagosomes [29], [30]. Disorders of mitochondrial β-oxidation can result in the stimulation of peroxisomal biogenesis [31], highlighting that defects in one organelle can induce the biogenesis of another. (See Fig. 1)

Fig. 1

Interactions between mitochondria and other organelles. (1) Contact sites between mitochondria and the endoplasmic reticulum (ER). (2) Mitochondria release mitochondria-derived vesicles (MDVs) to lysosomes and peroxisomes. (3) Peroxisome-targeted MDVs cooperate with ER-derived pre-peroxisome to generate new peroxisomes. (4) Mitochondria and peroxisomes share fission mediators, biogenesis pathways and form contact sites. (5) Lysosomes are required for the degradation of damaged mitochondria and excess peroxisomes. For further interactions and details, please see the text.

In order to have an complete understanding of organelle stress signaling, and of mitochondrial stress signaling in particular, it is therefore important to step out of the traditional reductionist view and take into account how different organelles respond to and contribute towards mitochondrial stress [11], [14], [32]. The next sections discuss the interactions between mitochondria, lysosomes and the autophagic system, between mitochondria and the ER, and between mitochondria and peroxisomes).

Mitochondrial malfunction affects lysosomes

For decades, lysosomes were seen as little more than the endpoint of the endocytic and autophagic pathways [33], [34]. Currently, it is known that lysosomes have an instrumental role in other cellular processes like amino acid sensing, exocytosis, plasma membrane repair, transcriptional regulation and also act as a reservoir of amino acids, metabolites and ions [33], [34], [35]. The deep involvement of lysosomes in so many metabolic functions raises the question of coordination with other organelles deeply involved in metabolism, particularly mitochondria. The cross-talk between mitochondria and lysosomes is complex, since lysosomes are also at the end of the mitochondrial life cycle: damaged mitochondria are degraded via autophagy. Furthermore, the release of one population of mitochondria-derived vesicles directed at the lysosomes raises another communication pathway from mitochondria to lysosomes [36]. Recently, some light was shed on signaling mechanisms that mediate the communication between the two organelles, particularly in the context of mitochondrial malfunction [11].

Malfunctioning or damaged mitochondria are removed from the cytoplasm by a process of selective autophagy, or mitophagy [37]. Given the lysosomal endpoint for mitophagy, this process the most conspicuous interaction between mitochondria and lysosomes. Interestingly, the process of mitophagy triggers a signaling cascade that culminates in the activation of lysosomal biogenesis. Upon mitophagy induction by a 10-h treatment with antimycin A (complex III inhibitor) or with the uncoupler valinomycin, Parkin-expressing HeLa cells show increased nuclear localization of the transcription factor TFEB, a known regulator of lysosomal biogenesis, suggestive of increased TFEB activity [38]. Notably, the mechanisms leading to TFEB activation by mitophagy or by starvation are different, implying a specific effect of mitochondrial malfunction on lysosomal biogenesis [38]. Indeed, activation of TFEB by mitophagy is dependent on Parkin, a E3-ubiquitin ligase involved in targeting damaged mitochondria for autophagy, as well as on canonical autophagy proteins Atg9A and Atg5 [38]. Nezich and colleagues also observed a strong redundancy between TFEB and other transcription factors of the microphtalmy family, namely MITF and TFE3. Mitochondrial malfunction caused by inhibition of respiratory chain complex I with rotenone was also found to trigger TFEB-related lysosomal biogenesis in neuroblastoma cells [39].

Two other studies observed that absence of key mitochondrial proteins leads to lysosomal impairment. Germline deletion of AIF, OPA1 or PINK1 in mice result in impaired lysosomal activity and lysosomal enlargement in a ROS-dependent mechanism [30]. Genetic ablation of TFAM was also used as a model of mitochondrial malfunction. TFAM is an essential protein for mitochondrial DNA (mtDNA) replication and transcription [40]. Loss of TFAM results in decreased mtDNA, decreased transcript levels of mtDNA-encoded genes and the absence of a functional respiratory chain [41]. Upon activation of TFAM−/- T-lymphocytes, mitochondrial function was severely compromised. Furthermore, there was an increase in the amount of lysosomes, but their function was impaired, as estimated by an increase in lysosomal pH (less acidic), cellular accumulation of sphingomyelin and autophagy intermediates [42]. The mechanisms linking the mitochondrial malfunction to the lysosomal impairments were not addressed in this study, but it was found that the increased lysosomal number correlated with activation of the transcription factor TFEB, a mediator of lysosomal biogenesis. Nevertheless, it remains unclear what is the mechanism that drove TFEB activation and if this was indeed the causal link of increased lysosomal number. Most importantly, the study by Baixauli and colleagues raises a key question: why the increase in lysosomal biogenesis is not resulting in functional lysosomes. These observations compound many other studies in which, independently of lysosomal biogenesis, defects in lysosomal function or the autophagic pathway were found to be a secondary consequence of mitochondrial malfunction [11].

Overall, the studies published so far show that in proliferating cells, both in vivo and in culture, acute mitochondrial malfunction triggers TFEB signaling and promotes lysosomal biogenesis. However, it seems that different mechanisms are in play during chronic mitochondrial stress and in post-mitotic tissues. A mouse model of amyotrophic lateral sclerosis (ALS) provided some insights into this question [43]. One of the pathological hallmarks of ALS is the accumulation of damaged mitochondrial in motor neurons [44]. It was recently reported that the mitochondrial dysfunction in that mouse model is compounded by lower lysosomal mass and lower autophagic capacity [43]. This would suggest that lysosomal biogenesis may be decreased under chronic mitochondrial dysfunction, and thus that chronic mitochondrial stress affects lysosomal biogenesis in a different way than acute mitochondrial stress. Indeed, it was recently revealed by another study [45] that lysosomal biogenesis is repressed at transcript level in the heart of a mouse model with mitochondrial impairments due to a defect in ubiquinone biosynthesis (Coq9R239X) [46]. The same study used complex I inhibitor rotenone or uncoupler CCCP to test how the duration of the mitochondrial impairment affects lysosomal biogenesis. In agreement with previous studies, the authors reported that acute treatments with rotenone or CCCP triggered lysosome biogenesis [45]. However, the persistence of the mitochondrial impairment over time leads to a repression of TFEB activity and lysosomal biogenesis [45]. This biphasic effect is in part explained by the activation of AMPK under acute mitochondrial malfunction, which is necessary for the activation of TFEB- or MITF-dependent lysosomal biogenesis [45]. Under chronic stress, AMPK is no longer active, and thus TFEB/MITF lack their activation stimulus. However, lysosomal biogenesis is actively repressed, and not just at baseline, in the heart of the Coq9R239X mice [45]. Thus, another mechanism that actively represses TFEB/MITF is likely in play under chronic mitochondrial malfunction.

Altogether, what is currently known about the effect of mitochondrial malfunction on lysosomes converges into three main points (see Fig. 2): (1) acute mitochondrial stress triggers lysosomal biogenesis, in an AMPK- and TFEB/MITF-dependent manner; (2) chronic mitochondrial stress represses lysosomal biogenesis; (3) chronic mitochondrial stress impairs lysosomal function.

Fig. 2

Interplay between mitochondria and lysosomes. Acute mitochondrial stress stimulates AMPK leading to TFEB-dependent lysosome biogenesis, but chronic mitochondrial stress represses lysosomal biogenesis. Lysosomal malfunction represses mitochondrial function.

The mechanisms underlying the impairment of lysosomal function during chronic mitochondrial deficiency remain unknown. One possible link is that reactive oxygen species released from mitochondria inhibit key proteins of the lysosome [30], but the exact targets remain unidentified. The identification of a mechanism showing how mitochondrial deficiency results in lysosomal impairments will clarify the contribution of lysosomal malfunction to mitochondrial pathology.

Lysosomal malfunction affects mitochondria

Lysosomal storage diseases (LSDs) are caused by mutations in genes encoding for lysosomal proteins, resulting in the accumulation (storage) of different molecules inside the lysosomes and the consequent inability of the organelle to function [47]. Recent data from lysosomal storage diseases shows that besides the impact on lysosomes, other cellular organelles are also affected, suggesting a potential crosstalk between organelles as a component of LSD pathogenesis [26]. For example, in Pompe's disease infants, the activities of respiratory chain complexes I, II and III are reduced [25]. In neuronal ceroild lipofuscinosis, often referred to as Batten disease, mitochondrial function and structure are severely impaired [48]. Ultrastructural alterations in mitochondrial morphology, decreased mitochondrial potential and impaired mitochondrial Ca2+ homeostasis have been described in several lysosomal diseases [11], [26], [49]. Dysfunctional lysosomes have lower capacity to process incoming autophagy traffic, thus limiting the clearance of damaged mitochondria from the cytoplasm, leading to the accumulation of less optimal mitochondria that release pathological signaling. However, it is important to distinguish what effects on mitochondria are specifically due to the lysosomes from those that are rather caused by impaired autophagy. Deficiencies in autophagy may involve inability to form autophagosomes or inability to fuse them with lysosomes. In both cases, there is likely accumulation of cytoplasmic components, including organelles, which are no longer in optimal function. For example, deletion of autophagy genes in yeast results in the accumulation of damaged mitochondria [50], while autophagy inhibition in mammalian fibroblasts results in increased mitochondrial mass due to the inability to remove the damaged organelles [51]. Autophagy inhibition by genetic ablation of Atg7, a necessary protein for autophagosome formation, results in decreased oxygen consumption in the skeletal muscle and in impaired reticulocyte maturation due to the inability to degrade mitochondria [52], [53].

The compounded lysosomal-mitochondrial malfunction is evident in a mouse model of Gaucher's disease, the most common LSD. These mice are knockout for glucocerebrosidase (GBA−/−) and present parkinsonism-related phenotypes [54]. In the neurons of these mice, accumulation of fragmented mitochondria with respiratory chain defects was detected [27]. The dysfunction in mitochondria contributed to lower alpha synuclein turnover, resulting in its increased oligomerization and deposition, a common downstream observation in Parkinson's disease.

In patient cells of another lysosomal disease, mucolipidosis II and III, the mitochondrial defects were not associated with defective autophagy like in most LSDs [55], suggesting that direct cross-talk between lysosomes and mitochondria, independent of autophagy, may exist. Furthermore, in mucolipidosis IV, mitochondria are also known to be fragmented and impaired [56]. The impact of lysosomal dysfunction on mitochondria is still far from understood in a systematic way, but the data so far implies that impaired lysosomes result in dysfunctional mitochondria, and that this cannot be explained solely by accumulation of damaged mitochondria due to defective autophagy.

Mitochondria and endoplasmic reticulum

The endoplasmic reticulum (ER) is a dynamic and complex organelle which is responsible for protein folding and Ca2+ storage, as well as metabolism of carbohydrates and lipids [57], [58]. The ER forms contact sites with several other organelles [58]. The contact sites between ER and mitochondria have important roles in the regulation of Ca2+ homeostasis and lipid transfer. Furthermore, the ER also wraps around mitochondria to mark the mitochondrial fission sites [59], [60]. Given the complexity of the ER-mitochondria contact sites, and their yet uncharacterized role in organelle stress response, this section is focused on the signaling cross-talk between mitochondria and ER rather than the physical contact sites.

Perturbations in Ca2+ homeostasis, redox imbalance and also defects in protein folding promote the accumulation of unfolded or misfolded proteins in the lumen of the ER, causing ER stress. In order to protect from damage caused by ER stress, cells have an integrated signaling system that reestablishes the equilibrium and normal ER function, comprising several pathways like the unfolded protein response (UPR), ER-associated degradation (ERAD), autophagy and even mitochondrial biogenesis [61]. The orchestrated activity of these pathways determines if the cell reestablishes homeostasis or activates cell death [61].

Since the respiratory chain complexes contain subunits encoded both by mitochondrial DNA and by the nuclear genome, the assembly of the complexes requires coordination between the two genomes [62], [63]. Consequently, imbalanced synthesis of these subunits can promote accumulation and aggregation of unassembled protein complexes. This accumulation is able to activate the mitochondrial unfolded protein response in order to reestablish mitochondrial equilibrium by the regulation of the expression of mitochondrial proteases and chaperones, in the same fashion as in the ER stress response [64].

Interestingly, there are several connections between the UPR and the regulation of mitochondrial function, chiefly that ATF4, one of the transcription factors which execute the transcriptional component of the UPR, can affect mitochondrial homeostasis through the controlled expression of the ubiquitin ligase Parkin [65]. This protein is activated by ATF4, and is involved in mitochondrial dynamics [66], bioenergetics [67] and, as mentioned above, mitophagy [37]. Importantly, the upregulation of Parkin by the UPR seems to be unrelated with its role in mitophagy [57]. The overexpression of parkin enhances ER-mitochondria coupling, favoring in this way the transfer of calcium between these organelles. The silencing of parkin, performed with siRNA, promoted mitochondrial fragmentation, impaired mitochondrial Ca2+ handling and reduced the tethering between ER and mitochondria [67].

Reciprocally, mitochondrial malfunction can trigger ER stress and lead to up-regulation of Parkin [65]. Thus, ER stress can cause mitochondrial stress but via ATF4 it up-regulates Parkin to avoid the latter. Under ER stress, there is also suppression of the transcriptional regulators C/EBPα, PPARα, and PGC-1α before lipid accumulation in mouse liver of mice lacking the ER stress sensor ATF6α [68]. Thus, as described for the mitochondria-lysosome interaction, the effect of ER stress on mitochondria also involves the regulation of mitochondrial biogenesis. Interestingly, both ER stress and lysosomal malfunction trigger the repression of mitochondrial biogenesis.

Several other mechanisms link the UPR with ER stress and mitochondrial function, such as the regulation of mitochondrial fusion and fission. Under ER stress conditions, Mfn2-deficient cells exhibit massive ER expansion and excessive activation of UPR, as well as reduced activation of apoptosis and autophagy [69]. Upon silencing of PERK, a kinase that mediates certain aspects of ER stress signaling, there is an increase in the apoptosis of Mfn2−/−cells when exposed to ER stress. The same study also shows that loss-of-function of XBP-1, another transcription factor mediating the UPR, ameliorates autophagic activity of these cells upon ER stress. Surprisingly, PERK silencing in these cells reduced ROS production, normalized mitochondrial calcium, and also improved mitochondrial morphology [70]. This study highlights PERK as an important upstream regulator of mitochondrial function, and provides another signaling pathway mediating ER-mitochondria cross-talk.

Due to its particular oxidizing environment, ER is considered one of the most unique and versatile components of the cell. Several studies demonstrate that under ER-associated stress conditions, redox-signaling mediators have pivotal roles in ROS generation and also that mitochondria are major contributors in the synthesis of ROS. Nevertheless, further studies are necessary in order to completely understand the mechanisms and pathways underlying their interaction under physiological and pathological conditions.

Mitochondria and peroxisomes

Peroxisomes are single-membrane- enclosed organelles which participate in key pathways of cellular metabolism such as amino acid catabolism, β-oxidation of fatty acids and also detoxification of reactive oxygen species. Despite being mainly regarded as spherical compartments, peroxisomes can be tubular and even found connected in a reticulum [71]. Moreover, peroxisomes are highly dynamic and responsive organelles, which is illustrated by the fact that their size and numbers, as well as the enzymatic repertoire, is cell type-specific and highly responsive to external stimuli [72].

The crosstalk between mitochondria and peroxisomes is essential for several metabolic processes and one of the best known examples of cooperation between these two organelles is the β-oxidation of fatty acids [73]. The biochemical steps of this pathway in both organelles are similar but each organelle owns its specific set of substrate specificity of enzymes. β-oxidation in peroxisomes is able to generate only chain-shortened fatty acids and, contrary to mitochondria, it is not able to fully degrade fatty acids. In this way, the medium chain fatty acids originated in peroxisomes, as well as acetyl-CoA, are then guided to mitochondria where further oxidation and ATP production in the tricarboxylic acid (TCA) cycle takes place [74]. The mechanism by which molecules are exchanged between peroxisomes and mitochondria are still poorly understood, but are likely to involve shuttle mechanisms such as carnitine system and membrane pores [75] and could also be promoted by contact sites between the two organelles [76]. These contact sites were extensively characterized in yeast by Schuldiner and colleagues, and seem to exhibit an intricate geographical organization of peroxisomes in subdomains of mitochondria that are also coincident with foci of the pyruvate dehydrogenase (PDH) complex and ER [76]. Importantly, the mitochondria-peroxisome contact sites have also been reported in mammalian systems, specifically in Leydig cells, where the peroxisome-mitochondria contact site is thought to enhance the efficiency of cholesterol transport from peroxisomes to mitochondria during steroid hormone synthesis [77].

Peroxisomes and mitochondria also share transcriptional programs of biogenesis, namely those triggered by PPAR-gamma and its coactivator PGC1a [20]. In addition, peroxisomal and mitochondrial membranes share the same division machinery, including the large GTPase Dynamin-related protein 1 (Drp1) [66]. While mitochondria have Fis1 and MFF as DRP1 receptors, peroxisomes have also their own specific DRP1 receptors, namely Peroxin 11 (Pex11) [78]. Curiously, this Pex11 was proposed to modulate the interactions between peroxisomes and mitochondria, by working as a potential molecular tether [79]. Furthermore, mitochondria and peroxisomes are both hosts for BAK, an apoptotic protein for a long time associated with the OMM, which was recently detected also in the peroxisomal membrane, where it regulates its permeability [80]. Future studies will be needed to both elucidate in more detail the tethering of these organelles and how BAK influences peroxisomal functions. Finally, peroxisomes and mitochondria share an immunity role, as the antiviral adaptor protein MAVS is dually-targeted to both organelles, where it can trigger an antiviral signaling cascade response upon viral detection [81].

Besides the cooperation in lipid metabolism, mitochondria and peroxisomes communicate and influence signaling pathways. In fact, the peroxisomal and mitochondrial metabolism are intimately linked with the generation of reactive oxygen and nitrogen species (ROS and RNS) and are also well-equipped with antioxidant systems, thereby being active regulators of the cellular redox state.

Peroxisomes and mitochondria also show interdependency. For example, when a key mediator of protein translocation to the peroxisomal matrix, Pex5, is knocked-out in mouse hepatocytes, thus yielding peroxisomal “ghosts” akin to those observed in Zellweger disease, the respiratory chain was severely perturbed, with decreased activity of complexes I and III and incomplete assembly of complex V, and a consequent impairment in oxygen consumption and increased oxidative stress [28]. Another example of this interdependency between the two organelles is provided by the mitochondrial status following deletion of ATP-binding cassete transporter (ABCD1) gene in B12 oligodendrocytes and U87 astrocytes. ABCD1 gene encodes peroxisomal ABC transporter adrenoleukodystrophy protein (ALDP) and is mutated/deleted in X-linked Adrenoleukodystrophy (X-ALD), an inherited peroxisomal metabolic neurodegenerative disorder. The silencing of peroxisomal protein ABCD1 produces structural and functional perturbations in mitochondria, including decreased activities of electron chain and citric acid cycle [82].

All these findings confirm and reinforce the observations which suggest that the disturbances in peroxisomal redox control and metabolism can sensitize cells to oxidative stress. Moreover, they provide strong support to the ideas that peroxisomes and mitochondria share a redox-sensitive relationship and that the redox communication between these two organelles may not only be mediated by diffusion of reactive oxygen species from one compartment to the other.

Finally, a novel cross-talk pathway between mitochondria and peroxisomes was recently discovered. Mitochondria release small vesicles, called mitochondrial-derived vesicles (MDVs), some of which target a small subset of peroxisomes [83], [84]. The cargo of these vesicles is different than those that are targeted to the lysosomes, which are involved in mitochondrial quality control [36]. The role of the peroxisome-targeted MDVs remained unclear until recently. Earlier this year, insightful evidences from McBride and colleagues showed that some peroxisome-targeted MDVs bud off from the mitochondria with incorporated peroxisomal biogenesis proteins, peroxins, and fuse with other preperoxisomal vesicles originated from the ER, in turn carrying other subsets of peroxins, to give rise to mature peroxisomes [21]. This finding completely revolutionized the current understanding of peroxisome biogenesis, and revealed a novel platform of crosstalk between different mitochondria and peroxisomes.

Altogether, these findings show that mitochondria may act as dynamic receivers, integrators, and transmitters of peroxisome-derived mediators of oxidative stress, while at the same time participating in the biogenesis of peroxisomes.

Perspectives

The way mitochondria communicate with the rest of the cell is often a matter of life or death. The mechanisms by which mitochondria communicate with the cellular signaling networks has been subject of attention for many years, and while there are some patterns clear at this stage, it remains to be understood how mitochondrial signaling explains known facts such as the tissue-specificity of mitochondrial diseases. In particular, the mechanisms by which mitochondria interact with other organelles, both in physiological and pathological circumstances, are pivotal to establish how the other organelles contribute to the pathology of mitochondrial diseases. So far, it is clear that there are several transcriptional programs mediating organelle communication, and that other mechanisms must be in place to mediate interdependency of organelle function, be those mediated by signaling networks, metabolites or physical contact sites. Given the increasing attention that organelle interactions are receiving, some of the answers to these questions are bound to be revealed very soon.