European Journal of Experimental Biology Open Access

Keywords

ATAD3; Mitochondria; Exercise; Diet

Introduction

ATAD3 (ATPase family AAA Domain-containing protein 3) is a 68 KDa mitochondrial inner membrane ATPase which appeared phylogenetically in the pluricellular organism [1]. Despite a still undiscovered function, ATAD3 is known as vital, as essential for mitochondrial biogenesis [2-4] and for early development of animals like nematodes [5,6] drosophila [7] and mouse [8]. As others AAA+ ATPases, ATAD3 is expected to be active in multimeric states like forming a ring-shaped P-loop exerting its activity through energy-dependent remodeling and translocation of macromolecules [9,10]. Among the cellular phenotypes induced by ATAD3 miss-of-function is the impact on the mitochondrial biogenesis [4,11] and network formation [4,12] as on mtDNA maintenance [2]. ATAD3 is therefore a new and major actor of the mitochondrial physiology, becoming a candidate potentially involved in some unclassified mitochondrial-associated diseases. Either, ATAD3 expression was also tightly linked to cancer appearance and progression [1,13].

Today, the strongest putative function for ATAD3 is thought to be a direct contribution in cholesterol import and/or more likely in lipid rafts transport of nuclear-encoded mitochondrial proteins, from the reticulum to the mitochondria [2,4,14]. ATAD3 is proposed therefore to act as a molecular motor inducing morphological modification of membranes and their translocation. Recent works showed indeed that ATAD3 could interact in complex with VDAC (Voltage-Dependent Anion Channel) and Translocator Protein (TSPO) [14] or also with COX assembly factor [15]. In the side, ATAD3 interacts with Mitochondrial Associated Membranes (MAM) [16,17]. This would support ATAD3 as a potential partner in sterol transport and especially in cholesterol transport. In addition, it is known from others and us that ATAD3 functionally and structurally interacts with Mfn2/Drp1 fusion/fission machinery [12,16]. Then, both observations lead to think that ATAD3 could be involved in contact sites between mitochondria and the endoplasmic reticulum and contributing there in lipidprotein exports to mitochondria. More recently, ATAD3 was also shown to regulate mitochondrial iron homeostasis [18].

Again, and without knowing with its exact function, ATAD3 dysfunction can be potentially involved in mitochondria-linked diseases, especially in yet unclassified ones, among which are myopathies, neuropathies, or metabolic diseases and some cancers. Indeed, three recent studies have shown in a set of family patients, that ATAD3 mutation, mono-or bi-allelic, were linked to severe neuro-myopathies [19-21]. Also, the causality was clearly shown by the skeletal muscle-specific knockout of ATAD3 in mice that induces muscle weakness and mitochondrial abnormalities [22].

Therefore, to address the question of knowing deeper about the direct impact of ATAD3 loss of function at the body level, we went to ATAD3+/-mice model to screen for basic muscular behaviors and metabolic characteristics [8]. Because ATAD3+/-mice do not present any obvious phenotype (normal fertility and sex ratio), we enterprise to watch the mice along periods requiring significant mitochondrial biogenesis, as muscle training and adipose tissue fat uptake [4,23-25]. Mitochondrial biogenesis was investigated by training male mice for running 7 times a week for 6 weeks, as males react higher to physical training, and adipose tissue physiology was tested by submitting female cohorts to high-fat/ high glucose diet for 4 weeks, as females react higher to highcalorie diets.

We found that ATAD3 hemizygosity inhibits running training capacities and mitochondrial biogenesis and impacts body weight increase under high-calorie diet, both phenotypes involving a major increase of ACC phosphorylation in the concerned tissues.

ATAD3 gene does not therefore present allelic compensations, or haplo-sufficiency, under physical training and high-calorie dieting phases in mice and therefore ATAD3 hemizygosity is potentially involved in pathologies such as obesity and not severe myopathies.

Materials and Methods

Mice-breeding

ATAD3+/-male C57bl/6j mice were obtained from Pr. Kolanus laboratory, at the Life and Medical Sciences Institute (LIMES) in Bonn [8] and crossed with C57bl/6j female mice. Progenies were genotyped by PCR (Figure 1A) using a distal finger tarsus cut performed for numbering newborn.

All experiments followed the European Union recommendations concerning the care and use of laboratory animals for experimental and scientific purposes. All animal work was approved by the local board of ethics for animal experimentation and notified to the research animal facility of our laboratory (authorization ComEth Grenoble n°244-04427.04). The performed research was in compliance with Animal Research: Reporting of in vivo Experiments (ARRIVE guidelines) on animal research.

Running training, limit length/speedrunning and gripping tests

Ten male mice of each genotype (WT and ATAD3+/-) were housed each together and trained daily for running on a treadmill (15 cm/ sec) for 30 minutes, 5/7 days, after a week of habituation. For limit length/speedrunning tests, mice ran at increasing speeds: 5 min at 10 cm/sec, 5 min at 20 cm/sec, 5 min at 30 cm/sec and 35 cm/ sec until the limit for abandon (five consecutive electrifications at 10 mV).

Gripping tests were performed by returning mice on a grid, sixty centimeters over a pillow, and measuring the time before falling.

All these experiments were performed according to animal care and under ethical watching (ComEth Grenoble n°244).

Activimetry

Ten male mice of each genotype (WT and ATAD3+/-) were video-recorded in individual cages over a 12 h night period and the spontaneous physical activity was calculated as previously described [26].

High glucose-high fat diet

Ten female mice of each genotype (WT and ATAD3+/-) were housed each together and fed with a high sucrose/high-fat regime (31.8% proteins, 50% lipids (46% milk fat-4% oil) and 23% sucrose) over a two-week period. Immediately, the same cohorts were fed with a high-fat diet (31.5% proteins, 54% lipids (50% lard, 4% soya-bean oil w/w), and 7% cellulose) over another two weeks period. The energy from fat in this diet typically represents more than 50% of total calories as in an average Western diet. All groups were fed ad libitum with free access to water and their body weight and food intake were measured regularly.

Genotyping PCR

Animal genotyping of gene-trapped and wild type (WT) ATAD3 genes was performed on newborn mice the first tarsus, according to the two-steps PCR method of Goller et al. [8]. Biopsies were lysed overnight at 56°C with proteinase K in Viagen buffer™, inactivated at 85°C one hour, and used directly for PCR (Titanium™ PCR kit). PCR products were analyzed by electrophoresis on a 2% agarose gel and imaged by Sybergreen™ staining.

Western-blot

Proteins were extracted from tissues with potter by lysis in Laemmli buffer containing 10 mM DTT and 5% v/v β-mercapto-ethanol. Samples were dosed for protein concentration (Biorad) and the same protein amounts (100 μg to 600 μg) were separated by 10% SDS-PAGE and electro-blotted onto a nitrocellulose membrane. The nitrocellulose membranes were incubated for 1 h at room temperature with a blocking buffer containing TBST (Tris-Buffered Saline (25 mM Tris), 0.1% Tween 20) complemented with 3% nonfat dry milk powder. They were further incubated at 4°C overnight with primary antibodies in TBST, washed three times for 10 min each with TBST at room temperature, and finally incubated at room temperature for 1 h with secondary antibodies in TBST. After washing three times for 10 min each in TBST, the detection was performed by enhanced chemo-luminescence (ecl plus, ge healthcare). Primary rabbit polyclonal antibodies (all from cell signaling technology and used at 1:1000 dilutions unless stated otherwise) were directed against Drp1, OPA1, 4EBP1, Ser333- phospho-ACC, ACC and β-actin. ATAD3 was detected using rabbit polyclonal antibodies raised against a specific sequence of ATAD3 (Anti-Nter: R40PAPKDKWSNFDPTG53 in ATAD3As; produced by Eurogenetec, Belgium) and used as purified immune-globulins. Secondary antibodies were anti-mouse or anti-rabbit IgG-horse radish peroxidase conjugates (GE Healthcare). The membranes were occasionally incubated with TBST again with sodium azide for irreversible inactivation of HRP. Imaged were obtain by chemoluminescence imager and signals were quantified using ImageJ.

Citrate synthase activity measurements

The reduction of 5, 5-dithiobis (2-nitrobenzoic acid) (DNTB) by Citrate Synthase (CS) at 412 nm (extinction coefficient 13.6 mM−1cm−1) was followed in a coupled reaction with coenzyme A and oxaloacetate. A reaction mixture of 150 mM Tris-HCl, pH 8.0, 5 mM acetyl-coenzyme A, 3 mM DTNB, and 20 μl of cells lysates were incubated at 37°C for 5 min. The reaction was initiated by the addition of 0.5 mM oxaloacetate and the absorbance changes were monitored for 5 min. All assays were normalized to protein amounts.

MtDNA/GH genomic PCR

DNA from soleus biopsies was extracted by overnight digestion at 56°C with proteinase K in Viagen Buffer™ and inactivated at 85°C for one hour. DNA purification was performed using Promega™ Wizard cartridges. After spectrophotometric dosage, 5 ng of each DNA samples were used directly for genomic-PCR amplification of the growth hormone gene (reverse primer: TCCCCTACCCCATAGTTTCC; forward primer: CTGGCTGCTGACACCTACAA). After a 400 times dilution of the DNA samples, 12 pg of DNA was used for the amplification of mtDNA (reverse primer: GGGCTAGGATTAGTTCAGAGTG; forward primer: CCGTCACCCTCCTCAAATTA). For both two-steps PCR, cycling parameters were 95°C 1 min at the first step and then 95°C for 30 sec/68°C for 1.5 min for 25 cycles. DNA samples were run on 2% agarose gel and imaged by SYBR green staining. Linear responses of the PCR were checked using 5 ng, 10 ng, 20 ng, and 50 ng, or 5 pg, 10 pg, 20 pg, or 50 pg of each DNA templates respectively (GH and mtDNA).

Statistics

Experiments were repeated at least three times and mostly analyzed in different parallels. Resulting data are given as means +/- standard deviation. Differences between data were analyzed by two-tailed, two-sample, unequal-variance student’s T-test using XLSTAT software with P-levels indicated according to * for P<0,1; ** for P<0,5 and *** for P<0,05.

Results

ATAD3 +/-mice breeding

Because constitutive invalidation of ATAD3 in mice is early lethal for embryo development, at gastrula and implantation stage 8, we used ATAD3+/-and+/+ mice to search for haplo-deficiency and differential phenotypes under physical training and high-calorie dieting. Breeding of ATAD3+/- and +/+ mice were therefore undergone as follows: C57Bl6j (ATAD3+/+) females were crossed with ATAD3+/-C57Bl6j males and the progenies were genotyped 5 days after birth (Figure 1A). Among the 84 animals obtained at F1 and F2 generations, 39 were WT and 45 were ATAD3+/-, showing no impact of ATAD3 deletion on embryonic viability. Among WT mice were 19 females and 20 males, and among ATAD3+/-mice were 23 females and 22 males, showing no impact of ATAD3+/-genotype on the sex ratio (Figure 1B). No impact of ATAD3 hemizygosity was observed on the female body weights at 2 and 5 months, under standard breeding (Figure 1B). However, at 5 months old, ATAD3+/-males were significantly lighter than the wild-type males (-10%). As a second characterization step, we checked by Western-blot for ATAD3 expression levels in different organs. As shown in Figure 1C, ATAD3 expression was significantly reduced, even more than twice, in ATAD3+/-animals. This is true in the skeletal muscle, the heart, the brain, the liver, and the adipose tissue. In adipose tissue, we observed that the expression of ATAD3 is almost 3-5 times more in females than in males, for both genotypes (data not shown).

Figure 1: ATAD3+/-mice breeding. A: ATAD3+/-mice breeding plan. WT ATAD3+/+C57Bl6j females were crossed with ATAD3+/- males of identical genetic background. More than 84 animals were obtained from F1 and F2 generations and were PCR-genotyped for ATAD3 allele’s structures (+ for WT allele, -for gene-trapped mutated one); B: The sex ratio of progenies (left panel) and the body weights of cohorts (n=10) at 2 and 5 months (middle and right panel) are represented by histograms; C: Western-blot analysis of ATAD3 expression levels was performed on muscles, hearts, brains, livers, and adipose tissues of ATAD3 +/- and +/+ mice, males, and females (n=4 for each), 4E-BP1 and actin were used as a loading control.

ATAD3+/-male mice have reduced running training performances, lower weights, higher gripping performances, and disturbed circadian activity

To reach our task in looking for ATAD3+/-phenotypes under a solicitation for mitochondrial biogenesis, we started to train males cohorts (5 months old, n=10) for daily running on a running platform, for a 6 weeks period (after one week of habituation). Running performances were regularly measured along the training period by performing limit length/speedrunning tests (see materials and methods). As can be seen in Figure 2A, and at the opposite of ATAD3+/+C57Bl6j males who improved normally their running performances (200% increase in limit length/ speedrunning tests), ATAD3+/-males were unable to improve their running performances, all over the 6 weeks of training. Watching at mice body weights along this period was to observe that both males lost weight during the training period (5% for both), while ATAD3+/-males with a weight is always 10% less than the WT males (Figure 2B).

Figure 2: ATAD3+/-male mice in physical activity. A: Males (n=10) were tested regularly along the training period for time/length limit running. Meters ran are represented on the curve along the training period; B: The body weights of animals (n=10) over the training period are represented on the curve; C: Gripping tests were performed with males (n=10) after the training period (number of experiments=5).

Our second approach to appreciate the muscle efficiency of ATAD3+/-and WT mice were to perform a gripping test (Figure 2C). As depicted in this figure, the gripping force was higher for ATAD3+/-animals.

Finally, at the term of training experiments, and before sacrifice, we measured the activity of the males along with a 12 h night period by movement video recording, in individual cages. As can be seen in Figure 3A and Figure 3B, we observed a significant difference in the distribution of activity periods, even if the overall activity remains the same (Figure 3B). While WT mice have a peak of activity during the night (22 h to 3 h), the ATAD3+/- mice did not present this peak and are more active in the evening and early morning. In addition, measured as a control, both mice were presenting almost the same overall inactivity even if ATAD3+/-mice were seen slightly more inactive (Figure 3C). This result shows that ATAD3 hemi-deletion can affect the circadian system.

Figure 3: ATAD3+/-male mice circadian activity. A: Activimetry was measured individually by video recording of 9 WT and 9 ATAD3+/- males during a night period. Movements of each mouse were quantified by computer analysis and restituted as an activimetric curve; B: Movements of each mice restituted as accumulative activimetric curve; C: Cumulated inactivity (min/h) is represented as a histogram.

ATAD3+/-male mice leg muscles have lower mitochondrial mass and present ACC hyperphosphorylation

To analyze the biological characteristics of ATAD3+/-males’ muscles after the training period, mice were sacrificed and dissected for the following analysis. At first, calf muscles, hearts, and epididymal adipose tissues were weighted (Figure 4A). No major change was observed for muscle weights, excepted that the gastrocnemius is slightly smaller in ATAD3+/-, and the plantaris and soleus a little bigger. In addition, the hearts of ATAD3+/-mice were smaller, even very slightly. Concerning the adipose tissue (epididymal fat), a very significant difference was observed in ATAD3+/-males (-40%=-0.5 g/1.325 g), explaining, but only in part, the lower whole body mass of ATAD3+/-mice (-4 g). Finally, these differences were observed the same on untrained animals.

To look for a cellular ATAD3+/-phenotype, we evaluated the mitochondrial mass levels of the different calf muscles by the citrate synthase assay (Figure 4B), and mtDNA PCR-based measurements (Figure 4C). Citrate synthase activities were clearly lowered in trained ATAD3+/-mice muscles compared to WT, especially in the plantaris and soleus muscles, as expected for a mitochondrial miss-of-function. Also, the mtDNA content was found to decrease in ATAD3+/-soleus to 42% compared to WT. Checking for the levels of few mitochondrial-associated proteins (OPA1, Drp1, and NRF1) in the soleus of trained +/+ and +/-mice, we found no main difference in any of the proteins studied excepted a slight down-regulation of NRF1, indicating a possible genetic down-regulation of mitochondrial biogenesis.

Figure 4: Biological characteristics of ATAD3+/-male mice muscles. A: Left panel: Weights of calf muscles (gastrocnemius, plantaris, and soleus) were measured after the sacrifice of males (n=10) and represented by a histogram. Middle panel: Weights of hearts were measured after the sacrifice of males (n=10) and represented by a histogram. Right panel: Weights of epididymal fat tissues of trained males were measured after sacrifice (n=10) and represented by a histogram; B: Citrate synthase activities were measured on the soleus of trained males (n=10) and are represented by a histogram; C: mtDNA PCR amplification was performed using Growth Hormone (GH) as a genomic gene control (see M and M). Quantified data are represented by a histogram; D: Westernblot analysis of mitochondrial proteins was performed on soleus protein extracts of trained males (n=6).

We looked finally at ACC phosphorylation status, an enzymelinked to mitochondrial-based lipogenesis. As can be seen in Figure 4D, we observed an increased level of pACC in ATAD3 +/- animals, and a very strong down-expression of the whole ACC amount, both showing a very important increase of the phosphorylation ratio of ACC (pACC/ACC). This result indicates a very significant inhibition of lipogenesis, even if the sampling is too low for statistics.

All the results obtained here let us think that in males, ATAD3 level can impact the circadian activity, the fat deposition level (most probably by ACC inactivation), and the muscle training capacity, by modulation of the muscle mitochondrial mass.

ATAD3 is then described as haplo-insufficient in males under training conditions, this emphasizing why ATAD3 gene status has been studied as a potential cause of unclassified myo- and neuropathies cases.

ATAD3+/-female mice have a higher propensity for obesity under high-calorie diet

Along with the term of our project, we bred female cohorts in parallel (n=10) with high-energetic diets, typical hypercaloric diets requiring as inducing mitochondrial biogenesis [25,27]. A high-glucose/high-fat diet was applied at the start for 3 weeks, followed by a simple high-fat diet for 2 other weeks (Figure 5A). At the beginning of the experimentation, both ATAD3+/- and WT females (5 months old) were of similar weights. Under high highfat/ high glucose regime, ATAD3+/-mice were with increasing weights along the 20 days period, over 15% (+3.6 g), while WT mice only increased their weights of 2% (+0.7 g). Switching then to a simple high-fat hyper-caloric diet (14 days), we found that ATAD3+/-mice started to decrease in weight (-4%), while WT mice were continuing with the weight increase. At the term of the experiment, we measured epididymal adipose tissue weights of both cohorts (Figure 5B) and we found that ATAD3+/-mice had a 50% weight increase of their epididymal fat tissue (0.75 g/0.5 g), this representing however only 10% of the whole body weight difference (Figure 5A). This phenotypic difference at the epididymal fat tissue level was not observed on female mice under standard diet. An important parameter of this diet experiment to mention is that both mice were exactly with the same food intake, given in excess, at the term of the experiment. Also, importantly, WT mice were looking anxious during the diet, more than ATAD3+/-mice.

Figure 5: ATAD3+/-female mice have a higher propensity for obesity under a high-calorie diet. A: The weights of female cohorts (n=10) were measured before and along with the diets and represented on the curve. High sucrose/high fat was a two weeks diet and high fat was, immediately after, a two weeks diet; B: The weights of epididymal fat tissues were measured after the sacrifice of females (n=10) and represented by a histogram; C: Western-blot analysis of pACC/ACC was performed on adipose tissue extracts of trained females (n=4). Actin and 4EBP1 were used as loading controls.

To address better the question of this specific fat increase in ATAD3+/-mice, we analyzed the phosphorylation status of ACC in adipose tissue samples since ACC is mainly involved in lipogenesis/ lipolysis regulation. As can be seen in Figure 5C, we observed that the phosphorylated ACC form increased in ATAD3+/-mice, while the whole ACC amount decreased very drastically, even if the sampling is too low for the statistic. Both phenomenon evoke therefore a strongly decreased in the lipogenesis potential.

Therefore, ATAD3 hemizygosity induces a weight increase capacity upon a high-calorie diet by affecting the adipose tissue mass associated with the inactivation of ACC activity. All of our results obtained here to let us think that ATAD3 is involved in whole-body metabolism and that ATAD3 loss of function can participate in a tendency for obesity in females.

Discussion

From the earliest studies on ATAD3, up today, it has been shown a strong link between ATAD3, lipid, and mitochondrial mass and metabolism, as observed here at the animal level. Knowing that ATAD3 is essential for mitochondrial biogenesis and for downstream processes like lipogenesis, steroidogenesis as proliferation and differentiation, the results obtained in this study are not surprising [2-4,8,14,15,18,28,29]. In addition, many works have focused on the role of ATAD3 in cancers [1,13]. Despite a still unknown precise function in mitochondrial biogenesis, all researches have converged today to the hypothesis that ATAD3 is involved in the transit of MAM-derived cholesterol-proteins rafts to mitochondria. This may furnish mitochondria with both lipids (for the synthesis of mitochondrial membranes, growth, renewing, and other syntheses like for steroids) and imported/ neo-synthetized nuclear-encoded mitochondrial proteins essential for these processes [2,14,15,29].

To approach ATAD3 function better we addressed here the question of understanding its role at the animal level in mouse.

The knockout of ATAD3 in mice is early lethal, as at day 4-5 of embryonic development [8], just at trophoblast growth and embryo implantation. This time corresponds to the first zygotic mitochondrial biogenesis [30]. Using a gene-trapped mouse model, we investigate the physiology of hemizygous ATAD3+/- mice, versus their WT counterparts.

Because ATAD3+/-mice do not present any obvious phenotype under regular breeding, the choice was to test the animals for a haplo-sufficiency during periods involving significant mitochondrial biogenesis, in muscle along with training and in adipose tissue under high-calorie diet. Even testing only males for muscular improvement and females for high-calorie diet we succeeded to observe strong phenotypic differences in both sexes.

For males, we observed that the standard weights of ATAD3+/- mice were 10% lower (not in females). We showed that ATAD3+/- males, trained for running, are unable to improve their running performances. This phenotype is associated with a lower citrate synthase activity in calf muscles after training as a lower mtDNA level, both indicating a lower mitochondrial mass. Nonetheless, ATAD3+/-mice have a higher gripping force and a modified circadian rhythm.

Such a result about the complete inability for ATAD3+/-mice to improve their performances was even not expected the so, but this is clearly understandable since ATAD3 is firstly involved in mitochondrial biogenesis, even being a limiting factor [3,4,11]. Also, if it was also a surprise, the increased gripping force of ATAD3+/-mice could be understood as a miss of mitochondria, provoked by ATAD3 depletion, would normally reduce aerobic muscle fibers which could, in turn, be compensated by the production of glycolytic fibers that will ameliorate gridding strength. The potential plasticity of mammalian muscles can explain this phenotype [31,32]. In addition, this phenotype was observed more significantly in aerobic fibers (soleus) than in glycolytic ones (gastrocnemius). For the few mitochondrial proteins studied, we did not observed the difference between ATAD3+/-and WT mice, excepted for NRF1 which expression is decreased in ATA3+/+ mice. This can be considered as a feedback regulation to adapt the mitochondrial biogenesis at the genetic level.

Looking to Acetyl-CoA Carboxylase activity (pACC/ACC ratio) we found it inhibited in ATAD3+/-male soleus (increased inhibitory phosphorylation ratio), indicating a lower lipogenesis in this tissue. This phenomenon may contribute to the inability of ATAD3+/- males to improve their running performance by reducing their lipid storage and reserves.

The other phenotype observed in ATAD3+/-males is the loss of the night-active peak period, normally observed in WT mice [33]. In spite of a normal active period in the night (22 h-3 h), ATAD3+/- mice were active more homogeneously along the night, without a night activity peak but an evening and early morning ones. However, the overall night activity was the same compared to WT mice. One could think that hormone levels, steroids particularly, are deregulated in ATAD3+/-mice [14], especially considering the decrease of fertility that we have observed for ATAD3+/-males.

To explain why ATAD3+/-males are lighter, before and after training, is to underline that ATAD3+/-males have a smaller epididymal adipose tissue (30% less). This can be due to a modified ACC activity in the adipose tissue as the one observed in the muscle. It appears then that the level of ATAD3 expression controls male body weight, probably because of lower steroid levels.

To conclude about the phenotypes observed in the males ATAD3 hemi-deletion is not haplo-sufficient under training solicitation. This observation may not be only linked to muscle status since ATAD3 is ubiquitously expressed and at a high level in the brain [4]. It is possible that brain activities of ATAD3+/-mice are altered like could be for the resistance to the pain, their motivation, or stress resistance. Steroids (cortisol/corticosterone, testosterone) levels can also be responsible for the observed phenotypes since ATAD3 controls steroid production [2]. Again, the decreased fertility observed in males that would fit with the other observations and the hypothesis that ATAD3 deletion is associated with a lower mitochondrial mass and steroid levels. Finally, the combination of both, like the change in ACC activity, can potentially have a strong impact on behaviors and running potential, in addition to a major role of the differential mitochondrial mass observed.

For females, we observed that the standard weights for ATAD3+/- mice are similar to WT counterpart, contrary to the males, indicating thus a clear impact of the sex on ATAD3 hemi-deletion phenotypes. While in males we detected weight differences we find here in females that it exists a full compensation under normal breeding conditions. Concerning body weight, we can state therefore that it exists a male-linked ATAD3 phenotype. However, we showed that ATAD3+/-female mice are much more prompt to increase their weight under a high-fat/high-glucose diet compare to WT. This phenotype (+3 g of weight increase after a 3 weeks diet) is due in part to an increase of epididymal adipose tissue (+0.3 g) but this plays only for 10% of the weight difference. Like for male’s weight under training, we will have to find which of other organs' weight are impacted during the diet, like could be the liver or the other adipose tissues. Such a result about the propensity of ATAD3+/-mice for obesity was not expected, as in vitro ATAD3 down-regulation reduces adipogenesis and lipogenesis [4]. Then, we expected that in vivo, ATAD3+/-mice would have limited lipogenesis and then a limited propensity for weight increase. However, we observed that ACC was inactivated in female adipose tissue-like observed in male muscles. Then, the increased weight under high diet might be linked somehow and finally to a reduced lipogenesis. In addition, at the animal level, it can be expected opposite responses, for compensation or other reasons. Then, we can hypothesize that ATAD3+/-mice can become more fat under high-calorie diet because all the other organs are not able to use so much of the calories (all mice on the same diet), because of less mitochondrial mass and energy use. We can hypothesize that ACC inactivation observed at sacrifice is the end-consequence of increased lipogenesis during the diet, as observed at bodyweight level. Calorimetric studies are the main clue to be explored, like to study mitochondrial mass in adipose tissue. As for males, this phenotype could also be due to different behaviors, the different activities of the mice, as ATAD3 hemi-deletion could affect brain activity. Therefore it will be interesting to study activimetry of females too, like training potential and other behaviors like cognitive studies. While performing the two-steps diet, with an only high-fat regime in the second period, we observed that ATAD3+/-mice were starting to lose weight, while WT mice continued to increase in weight. This observation allows us to think that ATAD3+/- obesity is not only due to fat consumption but could be linked to either highsugar consumption or its combination with fat. This will have to be discriminated against by further experiments. At this point is precise that ATAD3 is 4-5 times more expressed in adipose tissues of females versus males (data not shown). This would mean that females have either more mitochondria in adipose cells than males or that females contain ATAD3-enriched mitochondria. Deeper studies are needed to address this question. Then, the differences in ATAD3+/-phenotypes between sexes may be also due to hormonal differences.

To conclude about the phenotypes observed in females, it is clear that ATAD3 hemi-deletion in females induces a phenotype of propensity for obesity under a high-fat/high-glucose regime. This phenomenon, like in male muscles, is linked to a diminished ACC activity.

Conclusion

The search for ATAD3+/-phenotypes, even not observed in standard breeding conditions, was with good chances to be found because ATAD3 is playing a major role in mitochondrial biogenesis. Considering all the data on ATAD3, ATAD3 may likely contribute to the transport of lipid rafts of cholesterol and neosynthetized nuclear-encoded mitochondrial proteins, linked to Drp1-driven fission processes, from the MAM (Mitochondrial Associated Membranes), part of the endoplasmic reticulum, to the mitochondria. This process may drive mitochondrial biogenesis, which occurs during cellular growth and differentiation steps, as for simple multiplication of mitochondria, or for differentiationlinked proteomic specification, and mitochondrial renewal.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Acknowledgment

D.R. is very grateful to all the members of the LFBA team for their consciousness and friendship at work.

References

1. Li S, Rousseau D (2012) ATAD3, a vital membrane bound mitochondrial ATPase involved in tumor progression. J Bioenerget Biomembr 44: 189-197.
2. Gerhold JM, Cansiz-Arda Ş, Lõhmus M, Engberg O, Reyes A, et al. (2015) Human mitochondrial DNA-protein complexes attach to a cholesterol-rich membrane structure. Sci Rep 5: 15292.
3. Hoffmann M, Bellance N, Rossignol R, Koopman WJ, Willems PH, et al. (2009) C. elegans ATAD-3 is essential for mitochondrial activity and development. PloS One 4: e7644.
4. Li S, Lamarche F, Charton R, Delphin C, Gires O, et al. (2014) Expression analysis of ATAD3 isoforms in rodent and human cell lines and tissues. Gene 535: 60-69.
5. Piano F, Schetter AJ, Morton DG, Gunsalus KC, Reinke V, et al. (2002) Gene clustering based on RNAi phenotypes of ovary-enriched genes in C. elegans. Curr Biol 12: 1959-1964.
6. Sönnichsen B, Koski LB, Walsh A, Marschall P, Neumann B, et al. (2005) Full-genome RNAi profiling of early embryogenesis in Caenorhabditis elegans. Nature 434: 462-469.
7. Guertin DA, Guntur KV, Bell GW, Thoreen CC, Sabatini DM (2006) Functional genomics identifies TOR-regulated genes that control growth and division. Curr Biol 16: 958-970.
8. Goller T, Seibold UK, Kremmer E, Voos W,  Kolanus W (2013) ATAD3 function is essential for early post-implantation development in the mouse. PloS One 8: e54799.
9. Bogenhagen DF, Rousseau D, Burke S (2008) The layered structure of human mitochondrial DNA nucleoids. J Biolog Chem 283: 3665-3675.
10. Hubstenberger A, Merle N, Charton R, Brandolin G, Rousseau D (2010) Topological analysis of ATAD3A insertion in purified human mitochondria. J Bioenerg Biomembr 42: 143-150.
11. He J, Cooper HM, Reyes A, Di Re M, Sembongi H, et al. (2012) Mitochondrial nucleoid interacting proteins support mitochondrial protein synthesis. Nucleic Acids Res 40: 6109-6121.
12. Gilquin B, Cannon BR, Hubstenberger A, Moulouel B, Falk E, et al. (2010) The calcium-dependent interaction between S100B and the mitochondrial AAA ATPase ATAD3A and the role of this complex in the cytoplasmic processing of ATAD3A. Mol Cell Biol 30: 2724-2736.
13. Hubstenberger A, Labourdette G, Baudier J, Rousseau D (2008) ATAD 3A and ATAD 3B are distal 1p-located genes differentially expressed in human glioma cell lines and present in vitro anti-oncogenic and chemoresistant properties. Exper Cell Res 314: 2870-2883.
14. Rone MB, Midzak AS, Issop L, Rammouz G, Jagannathan S, et al. (2012) Identification of a dynamic mitochondrial protein complex driving cholesterol import, trafficking, and metabolism to steroid hormones. Mol Endocrinol 26: 1868-1882.
15. Ban-Ishihara R, Tomohiro-Takamiya S, Tani M, Baudier J, Ishihara N, et al. (2015) COX assembly factor ccdc56 regulates mitochondrial morphology by affecting mitochondrial recruitment of Drp1. FEBS Lett 589: 3126-3132.
16. Chiang SF, Huang CY, Lin TY, Chiou SH, Chow KC (2012) An alternative import pathway of AIF to the mitochondria. Internat J Mol Med 29: 365-372.
17. De Brito OM, Scorrano L (2010) An intimate liaison: spatial organization of the endoplasmic reticulum-mitochondria relationship. EMBO J 29: 2715-2723.
18. Van den Ecker D, Hoffmann M, Müting G, Maglioni S, Herebian D, et al. (2015) Caenorhabditis elegans ATAD-3 modulates mitochondrial iron and heme homeostasis. Biochem Biophys Res Commun 467: 389-394.
19. Cooper HM, Yang Y, Ylikallio E, Khairullin R, Woldegebriel R, et al. (2017) ATPase-deficient mitochondrial inner membrane protein ATAD3A disturbs mitochondrial dynamics in dominant hereditary spastic paraplegia. Human Mol Genet 26: 1432-1443.
20. Eldomery MK, Coban-Akdemir Z, Harel T, Rosenfeld JA, Gambin T, et al. (2017) Lessons learned from additional research analyses of unsolved clinical exome cases. Genome Med 9: 1-15.
21. Harel T, Yoon WH, Garone C, Gu S, Coban-Akdemir Z, et al. (2016) Recurrent de novo and biallelic variation of ATAD3A, encoding a mitochondrial membrane protein, results in distinct neurological syndromes. Amer J Human Genet 99: 831-845.
22. Peralta S, Goffart S, Williams SL, Diaz F, Garcia S, et al. (2018) ATAD3 controls mitochondrial cristae structure in mouse muscle, influencing mtDNA replication and cholesterol levels. J Cell Sci 131: 217075.
23. Ceddia RB (2013) The role of AMP-activated protein kinase in regulating white adipose tissue metabolism. Mol Cell Endocrinol 366: 194-203.
24. Davies KJ, Packer L, Brooks GA (1981) Biochemical adaptation of mitochondria, muscle, and whole-animal respiration to endurance training. Arch Biochem Biophys 209: 539-554.
25. Wilson-Fritch L, Burkart A, Bell G, Mendelson K, Leszyk J, et al. (2003). Mitochondrial biogenesis and remodeling during adipogenesis and in response to the insulin sensitizer rosiglitazone. Mol Cell Biol 23: 1085-1094.
26. Chabert C, Bottelin P, Pison C, Dubouchaud H (2016) A low-cost system to easily measure spontaneous physical activity in rodents. J Applied Physiol 120: 1097-1103.
27. Jain SS, Paglialunga S, Vigna C, Ludzki A, Herbst EA, et al. (2014) High-fat diet-induced mitochondrial biogenesis is regulated by mitochondrial-derived reactive oxygen species activation of CaMKII. Diabetes 63: 1907-1913.
28. Issop L, Fan J, Lee S, Rone MB, Basu K, et al. (2015) Mitochondria-associated membrane formation in hormone-stimulated leydig cell steroidogenesis: role of ATAD3. Endocrinol 156: 334-345.
29. Zhao Y, Sun X, Hu D, Prosdocimo DA, Hoppel C, et al. (2019) ATAD3A oligomerization causes neurodegeneration by coupling mitochondrial fragmentation and bioenergetics defects. Nature Commun 10: 1-20.
30. Wanet A, Arnould T, Najimi M, Renard P (2015) Connecting mitochondria, metabolism, and stem cell fate. Stem Cells Develop 24: 1957-1971.
31. Bassel-Duby R, Olson EN (2006) Signaling pathways in skeletal muscle remodeling. Annu Rev Biochem 75: 19-37.
32. Schiaffino S (2010) Fibre types in skeletal muscle: a personal account. Acta Physiologica 199: 451-463.
33. Mount LE, Willmott JV (1967) The relation between spontaneous activity, metabolic rate, and the 24-hour cycle in mice at different environmental temperatures. J Physiol 190: 371-380.