Aging-dependent mitochondrial bioenergetic impairment in the skeletal muscle of NNT-deficient mice

, we aimed to investigate how muscle bioenergetic function and motor performance are affected by NNT expression and aging. Mice were subjected to the wire-hang test to assess locomotor performance, while mitochondrial bioenergetics was evaluated in fiber bundles from the soleus, vastus lateralis and plantaris muscles. An age-related decrease in the average wire-hang score was observed in middle-aged and older Nnt (cid:0) / (cid:0) mice compared to age-matched controls. Although respiratory rates in the soleus, vastus lateralis and plantaris muscles did not significantly differ between the genotypes in young mice, the rates of oxygen consumption did decrease in the soleus and vastus lateralis muscles of middle-aged and older Nnt (cid:0) / (cid:0) mice. Notably, the soleus, which exhibited the highest NNT expression level, was the muscle most affected by aging, and NNT loss. Additionally, histology of the soleus fibers revealed increased numbers of centralized nuclei in older Nnt (cid:0) / (cid:0) mice, indicating abnormal morphology. In summary, our findings suggest that NNT expression deficiency causes locomotor impairments and muscle dysfunction during aging in mice.


Introduction
The metabolic and motor functions of skeletal muscle undergo several changes with advancing age, often resulting in declines in muscle mass, strength, and power and impaired whole-body metabolic homeostasis (Clegg et al., 2013;Tieland et al., 2018;Wilkinson et al., 2018).These aging-related losses in muscle function stand out as determinants of individual autonomy, disease risk and overall health (Demontis et al., 2013;Strasser et al., 2018).
The molecular mechanisms underlying muscle remodeling during aging seem to involve mitochondria.In particular, the impairment of several mitochondrial functions, including bioenergetics, redox balance and signaling, Ca 2+ handling, and dynamics, has been shown to play pivotal roles in age-related muscle loss (Scudese et al., 2023;Xu and Wen, 2023).In this context, a recent study from our group revealed motor behavior deficits in aged mice carrying an Nnt mutation (Francisco et al., 2020), which is known to adversely affect the mitochondrial redox balance (Dogar et al., 2019;Francisco et al., 2022;Ronchi et al., 2013Ronchi et al., , 2016)).Although abnormal motor behavior in Nnt-mutant mice has been linked to mood changes and serotoninergic pathways (Francisco et al., 2020), it remains unclear whether these mice also exhibit peripheral muscular abnormalities.
The Nnt gene encodes the proton-translocating NAD(P) + transhydrogenase, also known as nicotinamide nucleotide transhydrogenase (NNT;EC 7.1.1.1).This enzyme is located in the inner mitochondrial membrane and plays a crucial role in supplying NADPH within the mitochondrial matrix (Francisco et al., 2022;Rydström, 2006).NADPH serves as a reducing agent that supports various antioxidant and biosynthetic pathways (Lewis et al., 2008;Zhen et al., 2023).Notably, the absence of NNT activity was confirmed in various tissues from Nnt − / − mice, including the brain and skeletal muscle, in contrast to Nnt wildtype controls (Figueira et al., 2021;Francisco et al., 2020;Huang et al., 2006;Nickel et al., 2015).The discovery of variants in the NNT gene in humans nearly a decade ago as a cause of primary adrenal insufficiency in some patients (Meimaridou et al., 2012), was followed by the identification of >50 different disease-causing NNT variants in human patients (Buonocore et al., 2021;Guran et al., 2016;Metherell et al., 2016;Roucher-Boulez et al., 2016).These findings highlighted the importance of studies expanding the understanding of the pathophysiological consequences of NNT deficiency.
Although the significance of NNT in regenerating NADPH and maintaining redox balance has been recently established under certain conditions, the specific role of NNT in the context of age-related mitochondrial bioenergetic dysfunction in muscle remains to be explored.Notably, we recently described specific respiratory conditions in which NNT plays a key role in redox balance in mitochondria isolated from skeletal muscle (Figueira et al., 2021).In this context, in the present study we aimed to elucidate the effects of NNT deficiency on muscle function using a suitable murine model.Our findings indicated that NNT deficiency interacts with the effects of aging to aggravate motor dysfunction and cause mitochondrial respiration impairment in oxidative muscle.These results may suggest that assessing muscle integrity and function in patients carrying pathogenic NNT variants is worthwhile.

Mice and ethical considerations
Control C57BL/6JUnib male mice with the wild-type Nnt gene (Nnt +/ + ) and congenic male mice homozygous for the mutated Nnt C57BL/6J allele (C57BL/6JUnib Nnt − /− ) were obtained from the Multidisciplinary Center for Biological Research (CEMIB/UNICAMP, Brazil) (Francisco et al., 2018).The mice were kept under standard laboratory conditions (22-24 • C and a 12-h light/dark cycle) with access to NuvilabCR1 chow (Nuvital, Colombo, PR, Brazil) and water ad libitum until they reached 3-4 months of age (young, YNG), 12-13 months of age (middle-aged, MID), and 22 months of age (older, OLD).Upon reaching the target ages, the mice underwent behavioral tests and were euthanized at least one week later using isoflurane (5 %), followed by cervical dislocation.Skeletal muscle dissection from the hind limb was performed to obtain muscle samples for further analyses.After removing the white superficial parts of the muscle, a sample was extracted from the deep middlethird portion of the vastus lateralis, which is redder in color.The soleus and plantaris muscles were initially obtained intact.The soleus, vastus lateralis, and plantaris muscles were weighed immediately after dissection.
All the experimental protocols were approved by the Ethics Committee on Animal Use of the University of Campinas (CEUA-UNICAMP, Campinas, Brazil, approval number 5693-1/2021) and complied with the ethical rules of the Brazilian National Council for the Control of Animal Experimentation (CONCEA).

Histology
Mouse soleus muscles were fixed in a solution of 4 % paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for at least 24 h.Subsequently, the samples were dehydrated, embedded in paraffin, and then sliced into 5 μm-thick transverse sections.Sections were stained with hematoxylin and eosin for morphological analysis.Images were obtained on a Nikon 80i microscope.All transverse fibers present in each cross-sectional image were counted, and among the fibers, those exhibiting centralized nuclei were specifically counted.The crosssectional area of the fibers with and without central nuclei was also measured.

Hanging wire test
Muscle strength was assessed in mice of different genotypes and ages by the falls-and-reaches method of the hanging wire test, as previously described (Putten, 2011).Mice were suspended on the middle of a metallic wire rod (55 cm long and 2 mm in diameter), which was horizontally positioned 36 cm above a soft bedding of wood shavings.The number of falls during the test (180 s) was recorded.Measurements were performed in duplicate with an interval of 2 h between tests for each mouse.The results were quantified using a scoring system: the fall score was set at 10 at the beginning of the test and was decreased by 1 each time a mouse fell from the wire; meanwhile, the reach score started at 0 and was increased by 1 each time a mouse reached one of the extremities of the rod.

Western blotting
The soleus muscle, redder portion of the vastus lateralis muscle and plantaris muscle of mice were homogenized in a glass-on-glass homogenizer in extraction medium (1 % Tween 20, 150 mM NaCl, 1 mM EGTA, 1 mM Na 3 VO 4 , 1 mM NaF, 1 % protease inhibitor cocktail, 50 mM Tris-HCl, pH 7.4).The protein content in the samples was measured by the Bradford reagent.The samples were then diluted 1:1 in Laemmli solution and electrophoresed (5 or 20 μg of protein per lane) in 10 % polyacrylamide gels by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).The resolved proteins were then transferred to a nitrocellulose membrane (0.45 μm).Reversible Ponceau staining was applied to evaluate the total protein levels in the blot lanes as a protein loading control (Romero-Calvo et al., 2010).The membranes were blocked with 5 % nonfat milk in phosphate buffered saline (0.9% NaCl, 0.05 M phosphate, pH 7.4) supplemented with 0.05 % Tween 20 and incubated with either a mouse anti-NNT antibody (Santa Cruz Biotechnology, 1:500) or a rabbit anti-VDAC antibody (Thermo Fisher Scientific, 1:500) overnight at 4 • C. The labeling was detected using an HRP-conjugated goat anti-mouse secondary antibody (1:10,000) or HRP-conjugated goat anti-rabbit secondary antibody (1:10,000) and SuperSignal West Pico chemiluminescent substrate on a digital instrument (UVITEC, Cambridge, UK).The labeling intensity was quantified using ImageJ software (https://imagej.nih.gov/ij/).mice labeled with a primary antibody against NNT.(D) Relative NNT expression in homogenates from the soleus muscles of YNG, MID and OLD Nnt +/+ mice.Samples from Nnt − /− mice were used as negative controls.The data are presented as the means ± SDs, and statistical analysis was performed by one-way ANOVA and post hoc Tukey's test.

Data analysis
The data are presented as representative traces or as the means ± standard deviations (SDs), with individual data points representing each independent sample depicted alongside the corresponding bar graphs.Quantitative data were analyzed using GraphPad Prism 8 software (GraphPad Software, Inc.).The data distribution was initially evaluated by the Shapiro-Wilk test, Brown-Forsythe test or F test.Since the assumptions of normality were validated, the data were evaluated using an unpaired t-test, one-way ANOVA with Tukey's HSD test, or two-way ANOVA with Sidak's multiple comparisons test, depending on the specific case.Outliers were identified and removed using the robust regression and outlier removal (ROUT) test (Q = 1 %) in GraphPad Prism.A minimum significance level (α) of 0.05 was set.

NNT is more expressed in the soleus muscle
Early studies on skeletal muscle suggested metabolic and mitochondrial specialization among the different fiber types, including a fiber type-specific pattern of NNT expression (Howlett and Willis, 1998;Murgia et al., 2015Murgia et al., , 2021)).Here, to characterize our mouse model for studying NNT expression in muscles, we analyzed the expression of NNT in the soleus, plantaris and red portion of the vastus lateralis muscles from Nnt +/+ mice, which exhibit distinct oxidative phenotypes: the soleus has nearly 80 % oxidative fibers (types I and IIa); the red portion of the vastus lateralis features a balanced mixture of oxidative and nonoxidative fibers; and the plantaris exhibits a high percentage of glycolytic oxidative and glycolytic fibers (types IIx and IIb; lacks type I fibers) (Armstrong and Phelps, 1984;Bloemberg and Quadrilatero, 2012;Cornachione et al., 2011;Holloszy et al., 1991;Luginbuhl et al., 1984;Schuenke et al., 2008).
Fig. 1A and B shows that NNT expression was >10-fold greater in the soleus muscle than in the plantaris.Among the analyzed muscles, the vastus lateralis presented intermediate levels of NNT expression, with great variability among samples, probably due to the mixed nature of its fiber type composition.In these analyses, 5 μg of protein had to be loaded for soleus samples, while 20 μg of protein from vastus lateralis and plantaris muscles was loaded into the respective lanes due to the large differences in NNT protein expression.These muscles exhibited comparable levels of VDAC expression, indicating similar mitochondrial contents despite the differences in NNT expression (Fig. S1).The two bands at approximately 30 kDa were identified as VDAC protein.This pattern has been previously reported and may be due to differences in VDAC isoforms expressed in muscle samples (Messina et al., 2012).When NNT expression was examined in the soleus muscles across different ages of mice (Fig. 1C and D), there was a nonsignificant trend toward higher NNT expression in middle-aged mice than in young and older mice.

NNT deficiency exacerbates age-related impairments in muscle function
The association between muscle strength and the Nnt genotype across different ages was explored by using motor tests.The "falls and reaches" method of the hanging wire test was used to assess muscular function and strength.Fig. 2A shows the Kaplan-Meier-like curve of the falls as a plot of the time course of the mean of the fall scores in the hanging wire test.These curves show that young mice exhibited fewer falls during the test, with no significant differences between genotypes at that age.Interestingly, an increase in falls occurred with age, as indicated by the steeper slope of the curves of the middle-aged and older mouse cohorts within a given genotype.Moreover, the curves for agematched Nnt − /− mice were steeper than those for Nnt +/+ mice, suggesting that the Nnt genotype exacerbates age-related decline in muscular strength.
Statistical comparisons were performed with the data observed at 90 s into the hanging wire test (Fig. 2B).At this time point, there was a significant decrease in the fall score in middle-aged and older Nnt − /− mice compared to that in age-matched wild-type counterparts.Two-way ANOVA revealed statistically significant effects of both age and genotype (age: F (2,101) = 42.15,p < 0.0001; genotype: F (1,101) = 19.58,p < 0.0001), as well as an interaction between these two independent variables (F (2, 101) = 5.744, p = 0.0043).Importantly, no differences in body mass were observed between Nnt +/+ and Nnt − /− mice in the middleaged and older cohorts (Fig. S2), suggesting that the observed differences cannot be attributed to differences in workload imposed by body mass.
In contrast, the reach score from the hanging wire test was not significantly different between age-matched Nnt +/+ and Nnt − /− mice, with only a significant effect of age detected at 90 s into the test (F (2,102) = 11.20,p < 0.001) (Fig. 2C -D).

NNT deficiency is linked to a smaller soleus muscle in older mice
The hanging wire score is likely dependent on relative muscle strength which correlated with muscle mass.As depicted in Fig. 3A, there was a significant decrease in soleus wet mass of Nnt − /− mice at older age.Analysis of soleus muscle wet mass also revealed that age (F (2,37) = 27.57,p < 0.001) and the interaction between age and genotype (F (2,37) = 10.46,p = 0.003) were significant sources of variation for the data.No changes were observed between genotypes for other age points or muscle analyzed.However, a significant effect of age was detected for plantaris wet mass data (Fig. 3C, F(2, 38) = 3.524, p = 0.0395).(B) mice were permeabilized with saponin and incubated with MiR05 supplemented with 10 mM pyruvate, 5 mM malate and 10 mM glutamate.Where indicated, 400 μM ADP, 1 μg/mL oligomycin, 0.2 μM FCCP, 1 μM rotenone, 5 mM succinate, 1 μM antimycin A and 1 mM ascorbate plus 100 μM TMPD were added to evaluate mitochondrial bioenergetics under different respiratory states.The shaded rectangles represent the curved regions where OCRs were measured to assess respiratory states: oxidative phosphorylation (OXPHOS); H + leak; maximum flux through the electron transport system (Max ETS); mitochondrial complex II substrate-driven respiration (Rotenone/Succinate); and mitochondrial complex IV substrate-driven respiration (Antimycin/Asc/TMPD).

NNT deficiency impairs mitochondrial bioenergetics in the soleus and vastus lateralis muscles of middle-aged and older mice
To investigate the effects of the absence of NNT expression on skeletal muscle mitochondrial bioenergetics, permeabilized fiber bundles from the soleus, vastus lateralis and plantaris muscles from mice of different ages and genotypes were evaluated via high-resolution respirometry.Fig. 4A and B show representative traces of the OCR in the soleus muscles of middle-aged Nnt +/+ and Nnt − /− mice, respectively.The OCR, supported by pyruvate, malate and glutamate substrates, was evaluated during sequential additions of ADP, oligomycin, and FCCP.Afterward, rotenone plus succinate, and antimycin A, ascorbate plus TMPD were added.This experimental protocol allowed for the evaluation of mitochondrial respiratory rates during oxidative phosphorylation (OXPHOS) supported by electron entry via complex I via NADH-linked substrate oxidation, a nonphosphorylating state (i.e., the OCR not sensitive to the ATP synthase inhibitor oligomycin, frequently referred to as H + leak respiration), the maximum flux of the electron transport system when the protonmotive force is disrupted by FCCP (Max ETS), and alternative electron delivery to complex II (Rotenone/Succinate), or to complex IV (Antimycin/Asc/TMPD).
In the vastus lateralis (Fig. 6), the OCR was approximately 40 % lower in middle-aged and older Nnt − /− mice than in age-matched Nnt +/ + mice, but only during oxidative phosphorylation supported by complex I electron entry (Fig. 6A).The Nnt genotype was a significant factor (F (1, 50) = 12.50, p = 0.0009) only for the oxidative phosphorylation data (Fig. 6A), and there were no other statistically significant differences in the vastus lateralis respiratory states between the Nnt +/+ and Nnt − /− mouse genotypes or between the different ages (Fig. 6B-E).
In the plantaris muscle (Fig. 7), no significant differences were observed between Nnt +/+ and Nnt − /− mice in the evaluated respiratory states at any age.Only age was a significant factor according to ANOVA for the maximum flux of the electron transport system (Fig. 7C, F (2, 58) = 6.248, p = 0.035).
All muscle evaluated here presented similar OCR values at young age, irrespective of fiber type composition.Although counterintuitive, similar results were obtained in previous studies using isolated muscle fibers from rodents (Gaglianone et al., 2019;Warren et al., 2014).
To account for the potential influence of mitochondrial content on the OCR measured in bundles of permeabilized muscles, citrate synthase activity was quantified as a marker of mitochondrial abundance in soleus samples (Figueiredo et al., 2008;Larsen et al., 2012) since this muscle exhibited the most differences in OCR between the two genotypes.Fig. S3 shows that there was no difference in citrate synthase activity between Nnt +/+ and Nnt − /− mice in soleus samples that were recovered from the oxygraph chamber after the OCR measurements.A control experiment for tissue processing for the citrate synthase assay was also conducted using samples of soleus muscle extracted from young Nnt +/+ and Nnt − /− mice, with 4 samples for each condition and genotype.The analysis showed no significant difference in citrate synthase activity between samples that were frozen immediately upon retrieval from the euthanized animal and those frozen after the experiment in the Oroboros equipment (275.7 ± 20.2 and 256.9 ± 6.1 mU/mg for Nnt +/+ ; 276.3 ± 16.2 and 246.0 ± 29.1 mU/mg for Nnt − /− , respectively).

NNT deficiency is associated with an increased frequency of centralized nuclei in soleus fibers
As major impairments in mitochondrial bioenergetics and changes in mass were observed in the soleus of Nnt − /− mice beyond young age, we analyzed histological sections of soleus muscles from young and older Nnt +/+ and Nnt − /− mice to better understand the interaction between NNT deficiency and aging on muscle function (Fig. 8).Centrally nucleated muscle fibers are indicative of muscle damage and the extent of degeneration and regeneration processes (Cadot et al., 2015;Joseph et al., 2019).Soleus muscle from older Nnt − /− mice presented 1.7-fold more centrally nucleated muscle fibers than did that from agematched controls (Fig. 8B), thereby demonstrating a clear genotype effect (F (1, 19) = 5.379, p = 0.0317).Furthermore, two-way ANOVA revealed a statistically significant effect of age on the number of centrally nucleated fibers (F (1, 19) = 6.935, p = 0.0164).No significant difference in muscle fiber area was observed between fibers with peripheral nuclei and centrally located nuclei in aged Nnt − /− mice (576.5 ± 114.4 μm 2 and 655.0 ± 180.5 μm 2 , respectively).This suggests that the process of fiber regeneration is not compromised in aged Nnt − /− mice, despite an increase in fiber degeneration.

Discussion
Mitochondrial dysfunction plays a significant role in the age-related decline in muscle function.The bioenergetic impairment observed in aging muscles results from complex interplay involving redox imbalance, compromised mitochondrial quality control, and damage to mitochondrial DNA (Brunk and Terman, 2002;Marzetti et al., 2013;Srivastava, 2017).NNT is a key player in mitochondrial redox regulation and significantly contributes to the redox balance in skeletal muscle (Figueira et al., 2021;Howlett and Willis, 1998).Thus, a lack of NNT expression could exacerbate age-related changes in skeletal muscle.In this study, we utilized a mouse model carrying the Nnt C57BL/6J mutation and congenic controls with the wild-type Nnt gene at three different ages to assess the contribution of NNT to skeletal muscle function and bioenergetics during aging.Our findings indicated that the absence of NNT exacerbated age-related muscle dysfunction and induced bioenergetic mitochondrial dysfunction, particularly in muscles with a greater content of oxidative fibers.
In line with the findings of previous studies on rabbits and humans (Howlett and Willis, 1998;Murgia et al., 2015Murgia et al., , 2021)), our results revealed greater NNT expression in the soleus muscle than in other muscles, which has the highest proportion of oxidative type I fibers among the muscles evaluated.The vastus lateralis showed intermediate levels of NNT expression and high interindividual differences, a finding compatible with its mixed composition of glycolytic and oxidative fibers and with a heterogeneous distribution within the muscle belly (Armstrong and Phelps, 1984;Bloemberg and Quadrilatero, 2012;Holloszy et al., 1991;Luginbuhl et al., 1984).These differences in NNT expression among muscle types may be related to variations in reactive oxygen species metabolism among fiber types.It is unclear whether the antioxidant power of glycolytic fibers relies more upon other sources of NADPH in the mitochondria and the cytoplasm.
In the middle-age and older mice, the comparison among the five respiratory conditions shown in Figs.5-7 suggested that stimulated respiration (OXPHOS and Max ETS) supported by complex I-linked energy substrates was most affected by NNT deficiency in the soleus.Bypassing electron entry via complex I (i.e., the rotenone/succinate condition) led mitochondrial OCR of both the soleus and vastus lateralis to be similar between mouse genotypes, providing additional support for depressed complex I activity under these conditions.Additionally, the analysis of respiration supported by ascorbate plus TMPD suggested that decreased complex IV activity may be another contributor to overall respiratory dysfunction in the soleus of middle-aged Nnt − /− mice.Notably, a 30 % lower expression of the mitochondrial marker VDAC was observed in the soleus of middle-aged mice.Although this reduction was not statistically significant, it may suggest that the lower OCR observed in the soleus of middle-aged Nnt − /− mice could also be a consequence of a lower mitochondrial content in these tissues.Importantly, mitochondrial dysfunction was detected in the soleus muscle at middle-age, whereas a decrease in muscle mass was only observed in older mice.This observation suggests that the absence of NNT leads to functional impairment noticeable earlier in mitochondrial function than in changes to muscle mass.Considering the muscle-type dependent NNT expression levels and mitochondrial dysfunction, it seems that muscles with a more oxidative phenotype rely more strongly on NNT function.Convincing evidence indicates that NNT operates forwardly in skeletal muscle mitochondria and, along with NADP-dependent isocitrate dehydrogenase, regenerates most of the NADPH required for peroxide removal in the mitochondrial matrix (Figueira et al., 2021).According to an established biochemical perspective, the absence of NNT-supplied NADPH compromises antioxidant systems and reductive biosynthetic pathways that depend on NADPH as an electron donor (Dogar et al., 2019;Francisco et al., 2022;Ronchi et al., 2013Ronchi et al., , 2016)).Therefore, considering that NNT deficiency imposes an oxidative burden and that redox imbalance is implicated in various muscle dysfunctions (Bonnard et al., 2008;Moylan and Reid, 2007;Scicchitano et al., 2018;Wesson et al., 2023), the observed interaction between aging and NNT deficiency in skeletal muscle can be understood in the context of redox imbalance.
The profile of mitochondrial respiratory dysfunction reported here suggests that complex I is likely the most affected among the mitochondrial respiratory structures in the soleus and vastus lateralis muscles of middle-aged and older mice devoid of NNT.Mitochondrial complex I is highly sensitive to oxidative stress and is frequently the respiratory component that is compromised under various conditions (Brown and Borutaite, 2004;Lin et al., 2002;Mailloux et al., 2014;Schapira et al., 1989).Although previous data (Lenaz et al., 1997;Michelini et al., 2015) have suggested an age-related decrease in complex I activity, our results revealed a decrease only when aging occurred in association with NNT deficiency.
The mitochondrial bioenergetic dysfunction observed in the soleus muscle of middle-aged and older mice devoid of NNT is also associated with a greater percentage of centrally located nuclei, a marker of structural degeneration (Cadot et al., 2015;Joseph et al., 2019).Accordingly, the soleus muscle also exhibited a lower wet mass in older mice lacking NNT, indicating that the absence of NNT, combined with aging, may lead to exacerbated muscle loss in more oxidative muscles.These muscular abnormalities are consistent with the impaired locomotor behavior observed in middle-aged and older Nnt − /− mice during the hanging wire test (Fig. 2) and rotarod test (Francisco et al., 2020).The hanging wire test, acknowledged for its sensitivity in determining muscular impairments in mice (Oliván et al., 2014), was performed here using the "falls and reaches" method.Beyond a young age, the NNTdeficient mice displayed worse falling scores than their age-matched controls.However, the reaching score from the hanging wire test did not differ between the genotypes.While the fall score is more directly related to muscle strength, the reach score may be more related to ability and exploratory drive (Agosti et al., 2020;Younger et al., 2022).
Concerning contractile and highly metabolically active cells, the pathophysiological consequences of the absence of NNT expression during aging have been previously investigated only in cardiac muscle (Williams et al., 2020(Williams et al., , 2021)), but these studies showed no dysfunction in mice of up to 18 months of age.The heart has the highest NNT expression level among all tissues evaluated to date (Fagerberg et al., 2014;Francisco et al., 2022;Nickel et al., 2015).While contractile activity in cardiac muscle constantly occurs at the frequency of heartbeats (at least 8 times per s in mice), the soleus appears to be activated only between 5 and 8 h per day, despite being the most frequently activated muscle in the rodent hind limb (Hennig and Lømo, 1985).Previous studies from our group have clearly indicated that the mitochondrial redox balance relies more on NNT function when respiratory activity and Krebs cycle flux are low (Figueira et al., 2021;Francisco et al., 2022;Ronchi et al., 2016).Therefore, a highly metabolically active tissue, such as contracting cardiac muscle, may be differentially protected from NADPH shortage due to NNT deficiency.In this scenario, the fast mitochondrial respiratory rate and Krebs cycle flux in cardiac muscle conceivably favors NADPH supply from NNT concurrent sources, mainly via isocitrate dehydrogenase and malic enzymes (Ronchi et al., 2016).
Clinical interest in NNT arose when pathogenic variants of NNT were identified as rare causes of primary adrenal insufficiency (Meimaridou et al., 2012).Like what is observed in mice bearing the Nnt C56BL/6J mutation, NNT pathogenic variants in humans lead to a total lack of NNT activity (Francisco et al., 2024).In this context, our results may suggest that these patients could be more susceptible to muscle function loss with aging.This aspect should be considered in the clinical follow-up of individuals with pathogenic variants in NNT.
This study, however, has some limitations.Our findings indicate that the absence of NNT over aging is associated with impaired mitochondrial bioenergetics, which could ultimately impact muscle endurance.Therefore, the lack of specific muscle endurance tests represents a limitation of the present study.Assessing endurance in aged mice without functional NNT remains an objective for future investigations.In addition, a more comprehensive study on NNT in skeletal muscle in the aging context should also include the assessment of markers of muscle injury and repair, along with direct evaluation of mitochondrial redox homeostasis to provide a better understanding of the role of NNT in muscle function during aging.

Conclusion
Our study revealed that NNT deficiency aggravated age-related locomotor impairments and muscle dysfunction in mice.Notably, the highly oxidative soleus muscle exhibited the highest NNT expression and the most pronounced decrease in mitochondrial respiratory function in the absence of NNT.Additionally, this metabolic dysfunction in the soleus was associated with fibers with centralized nuclei in older NNTdeficient mice, indicating morphological abnormalities.Our findings suggest a crucial role for NNT in promoting healthy aging in this mouse model of NNT deficiency.Consequently, a comprehensive assessment of neuromotor function in elderly patients carrying pathogenic variants in NNT appears to be warranted for future clinical studies.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that have or could be perceived to have influenced the work reported in this article.

Fig. 1 .
Fig. 1.NNT is more expressed in the soleus muscle.(A) Representative Western blot in homogenates from the soleus (SOL; 5 μg), vastus lateralis (VL; 20 μg) and plantaris muscles (PL; 20 μg) of middle-aged Nnt +/+ mice labeled with a primary antibody against NNT (~113 kDa).This representative membrane image was obtained through superexposure to demonstrate NNT labeling in the VL and PL, but the quantification was performed using an unsaturated image.The total protein content determined by Ponceau staining was used for data normalization.(B) Relative NNT expression in muscles from middle-aged Nnt +/+ mice determined via Western blot analysis.(C) Representative Western blot in homogenates from the soleus muscles (20 μg) of young (YNG), middle-aged (MID) and older (OLD) Nnt +/+

Fig. 2 .
Fig. 2. NNT deficiency exacerbates the age-related decrease in muscle strength in mice.(A) Kaplan-Meier-like curve representing the average fall score over time in the hanging wire test in young (YNG), middle-aged (MID) and older (OLD) Nnt +/+ and Nnt − /− mice.(B) Average fall scores in the hanging wire test at 90 s. (C) Kaplan-Meier-like curve representing the average reach score over time in the hanging wire.(D) Average reach scores in the hanging wire test at 90 s.The data are presented as the means ± SDs, and statistical analysis was performed by two-way ANOVA and post hoc Sidak's multiple comparisons test.

Fig. 5 .
Fig. 5. NNT deficiency impairs mitochondrial bioenergetic function in the soleus muscles of middle-aged and older mice.Antimycin-sensitive OCR measured in soleus fiber bundles from young (YNG), middle-aged (MID) and older (OLD) Nnt +/+ and Nnt − /− mice during (A) OXPHOS; (B) H + leak; (C) Max ETS; (D) mitochondrial complex II substrate-driven respiration; and (E) mitochondrial complex IV substrate-driven respiration, according to the experimental protocol in Fig. 4. The data are presented as the means ± SDs, and statistical analysis was performed by two-way ANOVA and post hoc Sidak's multiple comparisons test.

Fig. 6 .
Fig. 6.NNT deficiency impairs oxidative phosphorylation in the vastus lateralis of middle-aged and older mice.Antimycin-sensitive OCR measured in vastus lateralis fiber bundles from young (YNG), middle-aged (MID) and older (OLD) Nnt +/+ and Nnt − /− mice during (A) OXPHOS; (B) H + leak; (C) Max ETS; (D) mitochondrial complex II substrate-driven respiration; and (E) mitochondrial complex IV substrate-driven respiration, according to the experimental protocol in Fig. 4. The data are presented as the means ± SDs, and statistical analysis was performed by two-way ANOVA and post hoc Sidak's multiple comparisons test.

Fig. 7 .
Fig. 7. NNT deficiency in mice does not interfere with mitochondrial bioenergetic function in the plantaris muscle, irrespective of age.Antimycin-sensitive OCR measured in plantaris fiber bundles from young (YNG), middle-aged (MID) and older (OLD) Nnt +/+ and Nnt − /− mice during (A) OXPHOS; (B) H + leak; (C) Max ETS; (D) mitochondrial complex II substrate-driven respiration; and (E) mitochondrial complex IV substrate-driven respiration, according to the experimental protocol in Fig. 4. The data are presented as the means ± SDs, and statistical analysis was performed by two-way ANOVA and post hoc Sidak's multiple comparisons test.

Fig. 8 .
Fig. 8. NNT deficiency is associated with an increase in centralized nuclei in soleus muscle fibers from older mice.(A) Representative hematoxylin and eosin staining images of soleus muscles from older Nnt − /− mice.Arrows indicate centralized nuclei in muscle fibers from older Nnt − /− mice.Scale bar: 50 μm.(B) Percentage of centralized nuclei in the soleus muscle fibers of young (YNG) and older (OLD) Nnt +/+ and Nnt − /− mice.The data are presented as the means ± SDs, and statistical analysis was performed by two-way ANOVA and post hoc Sidak's multiple comparisons test.