Energy Problems in Cells with Cystinosis: The Central Role of Mitochondria

The OXPHOS includes mitochondrial RC complexes and the F₁F_O-ATP synthase. The RC consists of NADH-ubiquinone oxidoreductase (complex I), succinate-ubiquinone oxidoreductase (complex II), ubiquinone-cytochrome c oxidoreductase (complex III), and cytochrome c oxidase (complex IV). The proteins of the OXPHOS are encoded by nuclear and mitochondrial DNAs [62,63].

The first observation of altered mitochondrial RC dysfunction revealed a reduction in mitochondrial ATP levels when using a complex I substrate (malate-glutamate) to fuel the RC in human renal tubular cystinotic epithelial cells [51]. This did not occur with the complex II substrate (succinate), indicating a possible complex I deficiency. This finding was reproduced in a Ctns knockout mouse model, which showed a reduction in complex I substrate-dependent respiration, associated with a proton leak and decreased ATP production [29]. Confirmation of complex I activity impairment, as measured by V(max) activity, was demonstrated in conditionally immortalized proximal tubular epithelial cells from two patients with different CTNS gene mutations and in human kidney-2 (HK-2) cells silenced for the CTNS gene [53]. Altered mitochondrial complex I activity in cystinosis may be associated with the emerging "dynamic assembly" theory of complex I [64]. This hypothetical mechanism suggests that, once imported into mitochondria, several peripheral subunits of complex I replace pre-existing, oxidized, aged copies that are already assembled into the complex, thereby modulating complex I activity [65]. Regarding this theory, a specific downregulation of the complex I subunits involved in dynamic assembly has been found in cystinotic cells [53], suggesting a possible inhibition of the "rejuvenation" of complex I. A reduced mitochondrial membrane potential has been found alongside a reduction in mitochondrial complex I activity in the BRIN-BD11 rat clonal pancreatic β-cell line silenced for the Ctns gene [66], in the Ctns-knockout mouse model [29], in the HK-2 cell line silenced for the CTNS gene, and in epithelial proximal tubular cells from patients [53]. Additionally, a reduction in complex V F₁F_O-ATP synthase activity has been observed in cystinotic proximal tubular epithelial cells, which is associated with the downregulation of specific subunits such as "d" and OSCP that are involved in the dimerization and oligomerization of the complex [67]. This could be another cause of the reduced cellular ATP level. Interestingly, UCP-1 and UCP-2 protein levels were augmented in the gastrocnemius muscles and adipose tissues of Ctns^−/− mice compared to wild-type mice, suggesting increased energy expenditure associated with decreased ATP content, which contributes to muscle wasting and adipose tissue browning, respectively [68].

Another aspect of mitochondrial bioenergetics is nitric oxide (NO). NO plays an important role in renal pathophysiology; it is involved in fluid and electrolyte reabsorption by the proximal tubule. Most in vivo studies indicate that NO inhibits fluid and electrolyte reabsorption in the proximal tubule by inhibiting both the Na⁺/H⁺ exchange and the Na⁺, K⁺-ATPase activities. This condition is linked to renal Fanconi syndrome. Furthermore, NO has been found to be a potent inhibitor of cytochrome c oxidase of the mitochondrial RC [69]. Studies of the Ctns null PTECs line, which was isolated from Ctns knockout SV40T transgenic mice, revealed increased inducible nitric oxide synthase (iNOS) expression. This enzyme produces NO and nitrite/nitrate. It also reduces the expression and activity of Na⁺, K⁺-ATPase. These cells display depolarized mitochondria, a lower ATP content, altered nutrient metabolism, and are more sensitive to apoptosis due to their inhibitory effect on cytochrome c oxidase. Treating Ctns null PTECs with the iNOS-specific inhibitor 1400W recovered these parameters, which are associated with the mitigation of apoptosis. This explains, at least in part, the etiological factors of Fanconi syndrome in cystinosis. In a recent study, Berlingero et al. demonstrated that podocytes lacking CTNS exhibit impaired TCA cycle function and disrupted energy metabolism. This, in turn, resulted in reduced respiratory activity and elevated mitochondrial reactive oxygen species (ROS) levels [55] (Figure 2).

The loss of cystinosin is associated with alterations in the endolysosomal compartment and proximal tubular cell dysfunction [70,71]. Endolysosomes capture and degrade worn-out intracellular components, including mitochondria, through a process called autophagy [72]. Starting with the observation that autophagy is required to protect proximal tubular cells from acute tubular injury by removing damaged mitochondria [73], an increase in altered mitochondria and SQSTM1/p62 receptors has been observed in urinary cells and kidney biopsies from cystinotic patients [74]. Furthermore, reduced basal autophagy flux has been found in various cystinotic cell lines, resulting in an increased number of mitochondria containing autophagosomes [33,75,76] associated with elevated LC3-II protein levels. Recently, studies in primary proximal tubular cells from Ctns-knockout mice and a zebrafish model demonstrated that lysosomal dysfunction leads to defects in the clearance of damaged mitochondria by autophagy [29]. Interestingly, as with mitochondrial RC activity, defects in autophagy have not been found in Ctns^−/− fibroblasts [33] likely because they rely on glycolysis rather than mitochondrial oxidative phosphorylation for energy. This aspect also highlights the tissue-specific role of cystinosin [8].

It has also been found that the cystinosin protein can physically interact with the molecular partners of the mammalian target of rapamycin complex 1 (mTORC1) [77,78]. The mTORC1 pathway modulates cellular metabolism, survival, proliferation, and autophagy [79,80]. Inhibiting the mTORC1 pathway with everolimus, a clinically used anticancer and immunosuppressant agent, activates autophagy [81,82] and abolishes the block of autophagic flux [83].

Mitochondria are dynamic structures that undergo a process of dynamic structural changes, including fission and fusion, forming either fragmented or networked mitochondria, respectively, also depending on the energy status of the cell [84].

The process of mitochondrial fission plays a pivotal role in the formation of new mitochondria and the segregation of damaged mitochondria. In the event of impaired mitochondrial dynamics, defective mitochondria undergo selective elimination through a specialized form of autophagy, termed mitophagy. This process is initiated by signals that are either ubiquitin-independent or ubiquitin-dependent [85]. The process of fission is subject to regulation by the GTPase dynamin-related protein 1 (Drp-1), which is recruited from the cytosol to the mitochondrial outer membrane. In this location, Drp-1 binds to its protein receptors, including mitochondrial fission 1 (Fis-1), mitochondrial fission factor (MFF), and mitochondrial dynamics proteins, MID49 and MID51 [86,87]. According to recent research, CTNS^−/− PTECs exhibits a greater incidence of mitochondrial fragmentation, which is concomitant with diminished mitochondrial potential [54]. The condition is meticulously regulated and intimately linked to the quality control systems of the cell, such as the ubiquitin protease system (UPS) and the intra-mitochondrial proteolytic systems [88]. Despite the absence of alterations in Drp-1 protein levels and its PKA-dependent phosphorylation at Ser-637, as indicated by mitochondrial fragmentation [53], an increase in the expression of Fis-1 and the E3 ubiquitin ligase parkin, along with the ubiquitination of Mfn2, was observed in CTNS^−/− PTECs [54]. The results of the increase in parkin protein expression in cystinotic cells have become more significant considering the altered mitophagy observed in other studies [30,51,75]. In fact, parkin, a component of the PINK1/Parkin pathway, has been shown to promote mitophagy, a process that facilitates the elimination of damaged mitochondria. This, in turn, contributes to enhanced mitochondrial quality control, a critical aspect of cellular health and function. However, in the context of cystinosis, excessive mitophagy can be a cellular response to overwhelming damage. In cases where the lysosomal system is unable to adequately perform its function, this process can become counterproductive, resulting in a net loss of mitochondria.

The process of mitochondrial fusion is governed by a minimum of three regulatory fusion proteins, including the dynamin-related GTPase optic atrophy 1 (OPA1) and the dynamin-related GTPase mitofusins, Mfn1 and Mfn2 [89]. OPA1, an inner mitochondrial membrane protein, functions beyond mitochondrial fusion to regulate the stability of the mitochondrial RC complexes, the release of cytochrome c from mitochondrial cristae, and the maintenance of the architecture of mitochondrial cristae [90]. OPA1 activity is subject to rigorous regulation by post-translational modifications, including proteolytic processing and acetylation [91,92]. In fact, OPA1 undergoes a constitutive proteolytic process, during which the uncleaved long OPA1 (L-OPA) is converted into the cleaved shorter form of OPA1 (S-OPA). A variety of stress conditions, including apoptotic stimulation, have been shown to trigger the complete conversion of L-OPA1 into S-OPA1 [93]. The primary mitochondrial protease that regulates the proteolytic processing of OPA1 is OMA1 [94,95]. The total protein expression of OPA1 remained unchanged in cystinotic PTECs compared to control PTECs. However, an increase in S-OPA1, associated with an increase in the protease OMA1 activity, has been observed in CTNS^−/− PTECs [54]. It is noteworthy that, in addition to its function as a fusion protein, OPA1 modulates the restructuring of mitochondrial cristae. Specifically, OPA1 has been shown to form oligomers within the inner mitochondrial membrane, thereby maintaining the structural integrity of the cristae junctions [89]. During the process of apoptosis, the oligomers undergo a state of destabilization, thereby inducing the opening of the cristae and the subsequent dissipation of cytochrome c from the mitochondria. OPA1 oligomers were found to decrease in cystinotic cells. The association between the defect and the increase in cristae junction width was determined through TEM ultrastructural analyses [54].

Furthermore, ultrastructural analysis of mitochondria in PTECs, utilizing transmission electron microscopy (TEM), revealed the presence of smaller mitochondria in CTNS^−/− PTECs compared to CTNS^+/+ PTECs. This observation is accompanied by a significant reduction in the number of cristae per mitochondrial section and an increase in cristae lumen width [54]. It is noteworthy that the mitochondrial ultrastructure observed in cystinotic cells is closely associated with the decrease in mitochondrial RC activity and the increased susceptibility to apoptosis [89,96].

The process of mitochondrial motility is dependent on the function of microtubule-based motors, specifically dynein and kinesin, and is critical for maintaining a balance between fusion and fission [97]. Recent findings have indicated a decrease in the expression of the motor protein subunit dynein light intermediate chain DYNC1LI2 in the kidneys and fibroblasts of Ctns^−/− mice, as well as in the CTNS knockout human proximal tubule cell line [98]. The mechanistic role of DYNC1LI2 in mitochondrial transport remains to be fully elucidated. However, its reconstitution has been shown to restore vesicle/ER-lysosome trafficking, alleviate ER stress, and significantly mitigate mitochondrial fragmentation, membrane potential, and ER-mitochondrial associations in cystinotic cells [98]. The altered mitochondrial morphology has not been found in CTNS-silenced human induced pluripotent stem cells (iPSCs) [99], but this may be due to the more glycolytic metabolism of iPSCs than oxidative [100] (Figure 3).

Mitochondria have been demonstrated to play a pivotal role in the progression of inflammation in cystinosis, thereby establishing a link between lysosomal cystine accumulation and subsequent immune activation. In the context of nephropathic cystinosis, mitochondrial dysfunction has been demonstrated to promote the generation of excess reactive oxygen species (ROS), which in turn amplifies inflammatory signaling pathways, thereby contributing to tissue injury. The upregulation of inflammatory mediators—including cytokines, inflammasome components, and macrophage-recruiting signals—interacts with mitochondrial stress, abnormal autophagy, and enhanced cell death, forming a self-reinforcing pathogenic loop.

This loop is now understood to be reinforced at the epigenetic level. Recent genome-wide analyses indicate that the cystinotic environment triggers extensive DNA hypermethylation, particularly in genes essential for proximal tubule physiology, effectively "locking" the cells into a dysfunctional state that can be partially reversed by demethylating agents [101]. Beyond the established NLRP3 pathway, the NLRP2 protein has emerged as a critical in vivo mediator of this process. The deletion of NLRP2 in Ctns^−/− models has been demonstrated to markedly delay the onset of Fanconi syndrome and reduce the expression of profibrotic markers and chemokines such as Cxcl1 and Saa1, thereby identifying it as a primary driver of the early inflammatory phase [102].

Furthermore, studies demonstrate that impaired cystinosin function alters lysosomal–mitochondrial crosstalk, leading to increased expression of inflammatory molecules such as galectin 3 and monocyte chemoattractant protein 1. These molecules drive macrophage infiltration and exacerbate renal inflammation [103].

Additionally, cystine crystals are potent activators of the NLRP3 inflammasome in macrophages, acting as a chronic, sterile inflammatory trigger. Recent evidence suggests this inflammatory state induces 'trained immunity', a form of innate immune memory characterized by epigenetic and metabolic reprogramming [104]. In the cystinotic kidney, resident macrophages functioning as crystal "custodians" may acquire a maladaptive memory phenotype. Once reprogrammed, these cells could perpetuate inflammation and drive fibrosis autonomously, independent of the concurrent crystal burden. Furthermore, because trained immunity necessitates a metabolic shift toward aerobic glycolysis, this process likely exacerbates the intrinsic energy metabolism defects associated with CTNS deficiency, creating a compounding metabolic insult.

Collectively, these findings underscore the pivotal role of mitochondria in integrating metabolic stress and inflammatory responses in cystinosis, thereby suggesting that targeting mitochondrial dysfunction, in conjunction with epigenetic modulators or NLRP2 inhibitors, may prove efficacious in mitigating chronic inflammation and decelerating kidney disease progression.

Cysteamine exerts significant effects on cellular energy metabolism, primarily through modulation of mitochondrial function. In a pre-clinical study, cysteamine bitartrate was used to treat mitochondrial RC disease in worms, zebrafish, and human fibroblasts. High doses caused toxicity via hydrogen peroxide, while lower doses showed limited benefits. In worms, treatment reduced oxidative stress and improved mitochondrial function, though it did not improve survival. Human fibroblasts exhibited broad metabolic recovery. Pre-treatment with cysteamine in zebrafish prevented brain death and neuromuscular defects caused by RC inhibitors. This highlights cysteamine's neuroprotective potential through enhancement of the oxidative stress response in RC disorders such as Leigh syndrome and MELAS.

At low micromolar concentrations, cysteamine has been shown to enhance ATP levels, restore oxidative phosphorylation, improve mitochondrial membrane potential, and maintain mitochondrial structure in cystinotic proximal tubular cell cultures [52,53,54]. Similarly, improved membrane potential and oxidative phosphorylation have been observed in C. elegans complex I mutants (NDUFS2 orthologue) and human fibroblasts of Leigh syndrome patients without significant alterations in glutathione levels, indicating the presence of alternative antioxidant or metabolic pathways [112]. Cysteamine has also been demonstrated to influence metabolite flux, notably increasing aspartate availability, and reducing ROS under stress conditions. However, its therapeutic window is narrow: while low doses confer bioenergetic and antioxidant benefits, millimolar concentrations can induce oxidative stress and cellular toxicity. Cysteamine demonstrated an inability to improve alterations in the tricarboxylic acid (TCA) cycle or energy metabolites, as well as an absence of oxidative stress, in podocytes from CTNS^−/− patients [55]. In addition, the increased cristae junction width observed in cystinotic renal cell cultures showed a non-reversible response to cysteamine treatment, paralleling the reduction in OPA1 oligomerization [54]. Finally, the administration of cysteamine resulted in the restoration of mitochondrial cAMP levels in PTECs lacking CTNS, concomitant with enhanced SIRT3 levels [53]. Mitochondrial cAMP is produced by soluble adenylyl cyclase and regulates localized signaling independent of plasma membrane GPCRs. Specifically, mitochondrial cAMP activates PKA within the mitochondrial matrix, leading to the phosphorylation of components of the electron transport chain, thereby enhancing ATP production [113]. SIRT3 has been identified as a pivotal mitochondrial deacetylase, involved in crucial cellular processes such as enhancing electron transport, TCA flux, antioxidant defenses, and mitophagy [114].

Flavonoids, a diverse group of plant-derived polyphenols, have attracted significant attention for their ability to modulate mitochondrial function and protect against cellular stress [115]. Two studies investigated the molecular mechanisms of flavonoid compounds in cystinosis models through complementary experimental approaches. Luteolin was identified through high-throughput screening and subsequently tested in the Ctns^−/− mouse model. The results demonstrated an improvement in the autophagy-lysosome degradative pathway, as evidenced by reduced p62/SQSTM1 levels measured by in-cell ELISA. Additionally, enhanced megalin expression was observed, along with antioxidant and antiapoptotic properties [31,110]. Genistein, which exhibited efficacy in a 14-month preclinical trial conducted in Ctns^−/− mice, demonstrated a reduction in kidney cystine concentrations, cystine crystal formation, and LAMP1-positive structures while preserving parenchymal architecture. It has been demonstrated that both compounds act through cysteamine-independent mechanisms, targeting cellular pathways beyond lysosomal cystine accumulation [111]. It is noteworthy that nutraceutical compounds, such as genistein and luteolin, have been observed to enhance cAMP levels [116,117]. Flavonoids have been reported to modulate the endogenous levels of antioxidants, including superoxide dismutase, catalase, and glutathione peroxidase, as well as the activity of enzymes responsible for glutathione synthesis, such as glutathione reductase [102,103]. In addition, flavonoids have been shown to stimulate mitochondrial biogenesis, which is an adaptive cellular response to various stressors. This stimulation occurs through the upregulation or activation of PGC-1α [118]. Furthermore, flavonoids have been demonstrated to stimulate a series of transcription factors, including NRF1 and NRF2, which play important roles in regulating metabolism-related genes [105,106]. Specifically, genistein has been observed to enhance the activating phosphorylation of Src (Tyr416) and the phosphorylation of acetyl-CoA carboxylase (ACC), a substrate of AMPK and a marker of AMPK activation [119]. Analogously, luteolin induces AMPK phosphorylation, which increases the activity of SIRT1, which in turn activates PGC1α by deacetylating it and increasing the phosphorylation of ACC.

The mitochondria-targeted antioxidants mitoquinone (MitoQ) and mitoTEMPO are, respectively, the derivatives of mitochondrial quinolone and a superoxide dismutase mimetic that has superoxide and alkyl radical scavenging properties. MitoQ treatment has been shown to increase PGC1α expression and prevent mitochondrial membrane depolarization, reduce free radicals, and increase cytochrome oxidase activity [120]. MitoTEMPO treatment has been shown to attenuate mitochondrial oxidative damage by specifically reducing mitochondrial superoxide anion levels [121].

Recent findings indicate that the presence of swan-neck lesions in the renal proximal tubule associated with cystinosis is indicative of an adaptation to oxidative stress. These lesions have been observed to be associated with the loss of mitochondria and an increase in mitochondrial anion superoxide production. The administration of MitoQ to Ctns^−/− mice at one month of age or three months of age, continuing until six months of age, resulted in the preservation of the integrity of the glomerulotubular junctions and the deferral of the onset of the so-called swan-neck lesions. However, this intervention did not lead to the rescue of low-molecular-weight proteinuria [122].

Treatment of Ctns^−/− mouse PTECs for 24 h with mitoTEMPO has been demonstrated to promote the integrity of tight junctions and the differentiation and endocytic uptake capacity of these cells [22]. In preclinical studies investigating mitoTEMPO in immortalized patient-derived podocytes and zebrafish larvae models, the combination therapy with cysteamine has been shown to improve clinical outcomes. Indeed, cysteamine, when administered as a standalone agent, demonstrated no significant impact. However, when administered concomitantly with the mitochondrial O₂^•− scavenger MitoTEMPO, there was a substantial reduction in the ROS-induced lipid peroxidation, improved cell adhesion and reduced podocyte permeability and a significant reduction in proteinuria in zebrafish [55].

A total of five compounds were identified as effective in reducing cystine content and apoptosis in human CTNS^−/− PTECs. This identification was made through a combination of high-throughput and high-content drug screenings [123]. The subsequent computational analysis of these selected drugs, namely alexidine dihydrochloride, beta-escin, digoxin, disulfiram, and fluspirilene, demonstrated a shared mechanism of action in cystinotic cells, predominantly involving the regulation of metabolic processes, response to stress, RTK signaling pathway, cell–cell adhesion molecules, and mitoribosomes [123]. However, subsequent studies performed on mouse and zebrafish cystinosis models revealed that prolonged exposure to the selected drug disulfiram, even if was effective in cystine reduction, resulted in toxicity in both animal models [124]. The pleiotropic effects of disulfiram on mitochondria are of particular interest. Indeed, it has been demonstrated that this compound can rescue mitochondrial defects in disease models such as Barth syndrome or MELAS, by enhancing oxidative phosphorylation, improving ATP synthase, respiratory complexes III/IV, cardiolipin remodeling, and protein synthesis inside mitochondria [125]. However, disulfiram can also trigger mitochondrial dysfunction by forming a complex with copper ions (DSF-Cu) within cells. This complex leads to a cascade of events, including the accumulation of ROS and the disruption of key mitochondrial functions [126]. Consequently, the outcomes observed in animal models of cystinosis were contingent upon the cellular context, copper levels, and redox status, which could have contributed to the toxic effect of disulfiram.

An increase in the protein content of UCPs was observed in the adipose tissues and muscles of Ctns^−/− mice compared with the control group. Conversely, a decrease in ATP content was observed in adipose tissues and muscles of Ctns^−/− mice in comparison with the control group. The presence of vitamin D deficiency has been demonstrated to have a detrimental effect on adipose tissue and muscle metabolism. As demonstrated by Cheung et al., the repletion of 25(OH)D₃ and 1,25(OH)₂D₃ was found to attenuate the perturbations of uncoupling protein (UCP) and adenosine triphosphate (ATP) contents in adipose tissues and muscles in Ctns^−/− mice [127]. It is noteworthy that vitamin D has been observed to modulate the expression of mitofusin-1/2 (Mfn1/2), OPA1, and DRP1, as well as the shape of mitochondrial cristae [128,129].

Another molecule, designated 1400W, has been shown to reduce nitric oxide production by inhibiting iNOS. This molecule has demonstrated a remarkable efficacy in counteracting the effects of Fanconi syndrome. It is imperative to acknowledge that NO, which has been demonstrated to impede both Na⁺/H⁺ exchange and Na⁺, K⁺-ATPase activity, concurrently functions as a robust inhibitor of cytochrome c oxidase within the mitochondrial RC [130]. This phenomenon has the potential to result in depolarized mitochondria and diminished ATP levels, which, in turn, could lead to an escalation in apoptosis. In a study of Ctns-null PTECs, the administration of 1400W was found to restore normal mitochondrial function, improve ATP content, normalize nutrient metabolism, and reduce apoptosis [131].

Recent findings indicate that ketogenic diets (KD), which shift energy metabolism from glucose to ketone bodies, significantly preserve kidney function in mouse models of nephropathic cystinosis. This preservation occurs even when the diets are initiated after the onset of the disease. The KD prevents Fanconi syndrome-related proteinuria, glycosuria, polyuria, inflammation, and fibrosis [109,132]. To date, the tissue-specific effects of KD have been demonstrated by its ability to enhance mitochondrial function and biogenesis in certain tissues, thereby supporting its therapeutic application in neurological and metabolic disorders. Indeed, KD has been observed to upregulate key regulators of mitochondrial biogenesis and function, including PGC1α, SIRT3, and UCP2. This upregulation results in increased mitochondrial mass and enhanced bioenergetics, particularly in neurons and skeletal muscle [133,134].

As illustrated in Table 1, a comparative analysis is presented, encompassing the mechanisms of action and the effect on mitochondrial parameters.