Peroxisome loss leads to increased mitochondrial biogenesis and reduced autophagy to preserve mitochondrial function

doi:10.21203/rs.3.rs-2496824/v1

Download PDF

Research Article

Peroxisome loss leads to increased mitochondrial biogenesis and reduced autophagy to preserve mitochondrial function

https://doi.org/10.21203/rs.3.rs-2496824/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 20 Jun, 2023

Read the published version in Cellular and Molecular Life Sciences →

You are reading this latest preprint version

Peroxisomes are essential for mitochondrial health. However, the mechanisms underlying the relationship between these two organelles in hepatic metabolism remains unclear. To address this, we developed a conditional hepatocyte specific Pex16 deficient mouse (Pex16 KO) and subjected these animals to a low protein diet to induce metabolic stress. Loss of PEX16 in hepatocytes led to increased biogenesis of small mitochondria and a reduction in autophagy flux but with preserved capacity for respiration and ATP production. Metabolic stress induced by low protein feeding did lead to mitochondrial dysfunction in Pex16 KO mice and impaired the ability to upregulate its biogenesis. Activation of PPARα partially corrected the mitochondrial disturbances caused by low protein feeding, independent of the presence of peroxisomes. This study shows that peroxisome loss in hepatocytes affects mitochondrial biogenesis and autophagy, thereby preserving mitochondrial function and underscore the relation between peroxisomes and mitochondria in regulating the hepatic metabolic responses to nutritional stressors.

Peroxisome

PEX16

Knock out mice

Mitochondria

Mitophagy

Low protein diet

Fenofibrate

Hepatic peroxisomes are critical for several key lipid anabolic and catabolic processes, including oxidation of very long-chain fatty acids and branched fatty acids, and the synthesis of plasmalogen, bile acid and sterol precursor synthesis. These peroxisomal lipid biochemical reactions are not entirely completed in peroxisomes but take place between several organelles including the ER and mitochondria [1, 2]. Peroxisomes are important for mitochondrial health, as some peroxisome biogenesis disorders (PBD), also known as Zellweger spectrum disorders, genetic disorders resulting in the loss of peroxisomes, are associated with hepatic mitochondrial dysfunction [3, 4]. However, our understanding of underlying mechanisms is limited.

Peroxisome loss in PBD results from defect in one of 14 genes called Peroxins (PEX) responsible for peroxisome biogenesis and homeostasis [1, 5]. Some members of the peroxin family, such as PEX11, are important for peroxisomal growth and division. Others, including PEX5, PEX7, PEX14 and PEX13 regulate peroxisomal matrix protein import, while PEX3, PEX16, and PEX19 are essential in the initiation of peroxisome biogenesis [5]. Loss of the matrix peroxins such as PEX5 results in ghost peroxisomes that contain membranal but are devoid of matrix proteins [6], whereas, the loss of PEX16 completely abolishes peroxisomes [7]. As peroxisomes are known to interact with mitochondria [1, 2], complete abolishment of peroxisomes, i.e. in Pex16 deficiency, may serves as a good model to understand the role of peroxisomes in mitochondrial function.

Mitochondria continuously undergo fission and fusion to maintain dynamic homeostasis. These processes are mediated primarily by Dynamin-related protein 1 (Drp1) and Mitofusin-2 (Mfn2), respectively [8]. Impaired mitochondrial function can lead to reduced ATP production, excessive generation of reactive oxygen species (ROS), altered metabolism and calcium storage and apoptosis. In order to prevent the detrimental effect of accumulating damaged mitochondria, including cellular oxidative stress and eventually cell death, impaired mitochondria are turned over by selective autophagy called mitophagy [9–11]. Peroxisomes and mitochondria are closely linked and share responsibility for various metabolic pathways, including β-oxidation. Studies in various peroxisome deficient mouse models have provided some insight into the effects on mitochondrial homeostasis and function. PEX5 deficiency in mice has been shown to lead to proliferation of mitochondria with altered morphology and function with a reduced respiration activity mainly due to a decrease in complex I [12, 13]. A similar phenotype was observed in PEX2 and PEX13 deficient mice [14, 15]. However, similar mitochondria defects were also observed in metabolic stress conditions resulting in peroxisome loss in non-PEX mutant animals. We recently reported that young mice and rats fed a very low protein diet had dramatically reduced hepatic peroxisomes due to upregulation of peroxisome-specific autophagy, i.e., pexophagy, and this reduction of peroxisomes was associated with impaired mitochondrial function and change in morphology [16, 17]. Interestingly, an increased mitochondrial volume has been consistently observed in PEX deficient animal models and it has been suggested this is related to increased mitochondrial biogenesis [11]. However, the accumulation of damaged and dysfunctional mitochondria may also result from a defect in mitochondria quality control.

In this study we aimed to uncover the role of peroxisomes in mitochondrial homeostasis. To understand the connection between peroxisome loss and mitochondrial function, we generated a liver specific Pex16 knockout mouse and analyzed mitochondrial morphology, dynamics and function at baseline and in response to nutritional stress through low protein diet feeding. We found that the loss of PEX16 leads to increased mitochondrial biogenesis with development of small mitochondria with upregulation of genes controlling mitochondrial fission and fusion. Surprisingly, autophagy flux was reduced in this peroxisome deficient model, potentially associated with a reduction in degradation of damaged mitochondria through mitophagy. A nutritionally-induced metabolic insult by feeding mice a low protein diet led to mitochondrial dysfunction associated with a repression of mitochondrial biogenesis despite upregulation of autophagy. Upregulating mitochondria biogenesis and β-oxidation activity through PPARα activation was able to improve mitochondrial homeostasis and function independent of the presence of peroxisomes.

Mouse generation

To generate the Pex16 targeting vector, DNA fragment containing the murine Pex16 locus was subcloned into pBlueScriptIIKS+, and an oligonucleotide of loxP site was inserted into the intron separating exons 4 and 5. A fragment of containing the loxP site and FRT-flanked neo gene was inserted to intron separating exons 2 and 3. The plasmids was subcloned into pDTA-TK to get final targeting construct. ES cells were transfected, selected, and used to produce chimeric mice. Chimeras were crossed to C57BL / 6J. Pex16 targeted mice were maintained on C57BL / 6J genetic background. To generate hepatocyte specific Pex16 deficient mice (Pex16 KO), male Albumin-Cre mice were crossed bred with female Pex16^f/f mice. Male Pex16 ^f/+;Alb-cre was further crossed with female Pex16 ^f/f to generate homozygous Pex16^f/f;Alb-cre and wild-type control mice. Only male wild-type control mice and male homozygous Pex16^f/f;Alb-cre mice were used in the experiments. Genotyping of mice were confirmed by PCR (Table S1). Procedures followed the Canadian Council for Animal Care guidelines, and were approved by the Animal Care Committee at the Hospital for Sick Children.

Construction Of Lpd Mouse Model And Fenofibrate Treatment By Gavage

Fenofibrate (Sigma, F6020) treatment was delivered by gavage at a dose of 100µg/g body weight. Fenofibrate was dissolved in dimethyl sulfoxide in a concentration of 100ug/mL from 4 until sacrifice at 5 weeks of age. The same volume of DMSO as the fenofibrate group was given as a control vehicle group.

Modulating autophagy flux in Pex16 KO and LPD mouse model

Pex16 KO and wild-type control mice were randomly divided into four groups, two groups were fed the normal 18% protein chow, and the other two groups were fed the LPD using a 1% protein diet. Mice were kept on the different diets for 2 weeks starting from 3 weeks after birth when they weaned from their mother. Chloroquine (Abcam, Cambridge, UK) treatment was delivered by intraperitoneal (IP) injection with a dose of 50mg/kg body weight prior to 8 hours of liver harvesting. The mice were starved for 4 hours before sacrifice after which livers were harvested for western blotting of markers of autophagy.

Histology And Immunofluorescence

Livers were fixed and embedded in paraffin. They were then sectioned and stained with H&E. Frozen sections were cut and stained with Oil Red O or immunofluorescent antibodies. Frozen sections were blocked with 3%BSA and 0.1% Tween-20 in PBS for one hour at room temperature. Appropriate antibodies (Table S2 ) were diluted in 1%BSA and 0.1% Tween-20 in PBS. Slides were incubated in primary antibodies overnight at 4°C, followed by secondary antibodies (Table S2) incubation for one hour at room temperature. DAPI was added together with the secondary antibody for nuclear staining.

Western Blotting

Liver protein extraction, protein concentration and expression levels were measured by standard western blotting. Liver tissues were lysated in tissue extraction buffer containing protease inhibitor, and protein was isolated by centrifugation. Protein concentration was measured by BCA assay. Samples were then denatured in sodium dodecyl sulfate (SDS) and loaded onto SDS-PAGE gels. Proteins were transferred onto polyvinylidene fluoride (PVDF) membranes with Trans-Blot system. Membranes were blocked with 3%BSA and 0.1% Tween-20 in TBS for one hour at room temperature. Appropriate antibodies (Table S2) were diluted in 1%BSA and 0.1% Tween-20 in TBS. Membranes were incubated in primary antibodies overnight at 4°C, followed by secondary antibodies incubation for one hour at room temperature. Membranes were visualized using Pierce enhanced chemiluminescence (ECL).

qPCR

cDNA was synthesized from 500ng of liver total RNA using the SuperScript VILO cDNA Synthesis Kit (Thermo Fisher Scientific, Rockford, IL, USA). For qPCR, 0.1ul of cDNA were used with Advanced qPCR Mastermix with SuperG Dye (Wisent, Bioproducts, Saint-Jean-Baptiste, QC, Canada). For measurement of mtDNA number, 500ng of genomic DNA were used for each qPCR reaction. qPCR was performed using a CFX384 Touch Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA) using the primers listed in Table S3.

Hepatic Triglyceride Assay

Briefly, 50 mg of wet liver tissues was homogenized in cold PBS. Tissue lipids were extracted with methanol-chloroform, dried and dissolved for triglyceride analysis. Liver and serum triglyceride concentrations were determined by a commercially available kit (Randox Laboratories, Crumlin, UK). Photometric absorbances were read at 500 nm. The triglyceride values were normalized to protein concentration from same liver tissue which was measured using the BCA kit (Thermo Fisher Scientific).

Hepatic Atp Assay

Liver tissue (~ 10mg) was homogenized using a dounce homogenizer in perchloric acid. Excess PCA was precipitated by adding ice-cold 2M KOH to neutralize the samples (pH 6.5-8). The deproteinized samples were then ready to assay. ATP concentrations in the liver samples were measured using a commercially available ATP (Colorimetric/Fluorometric) assay kit (Abcam). Fluorescence was measured at Ex/Em = 537/578 nm with Skanlt software by Varioskan Lux Multimode Microplate Reader (Thermo Fisher Scientific). Fluorescence readings from samples were compared to the readings from standards available in the kit in order to calculate the concentration.

Transmission Electron Microscopy (Tem)

Fresh liver samples were fixed in 2% glutaraldehyde in 0.1 M sodium cacodylate buffer, rinsed in buffer, post-fixed in 1% osmium tetroxide in buffer, dehydrated in a graded ethanol series followed by propylene oxide, and embedded in Quetol-Spurr resin. Sections (90 nm thick) were cut on a Leica Ultracut ultramicrotome (Leica Microsystem, Wetzlar, Germany), stained with uranyl acetate and lead citrate and viewed in an FEI Tecnai 20 transmission electron microscope (Field Electron and Ion Company, Hilldboro, Oregon, USA). Mitochondrial content was measured using Image J software.

Central Carbon Metabolism Assay

For quantitation of central carbon metabolism (CCM) metabolites in mouse liver by UPLC-MRM/MS, a Dionex 3400 UHPLC system (Dionex Sunnyvale, CA, USA) coupled to a 4000 QTRAP mass spectrometer (SCIEX, Framingham, MA, USA) was used. Liver metabolites were extracted and TCA cycle carboxylic acids, glucose and selected sugar phosphates, and other phosphate-containing metabolites and nucleotides were quantified (Table S4 and S5). For all the above quantifications, concentrations of the detected metabolites were calculated from their linear-regression calibration curves with internal calibration.

Isolation of liver mitochondria and mitochondrial respiration

Mice were starved for 4 hours before the isolation experiment. Mitochondria from livers were isolated by differential centrifugation as described previously [18]. The livers were dissected in ice-cold mitochondrial isolation buffer that was composed of 100 mM Tris-MOPS, 10 mM EGTA/Tris and 90 mM sucrose, adjust pH to 7.4. The liver was minced into small pieces and rinsed free of blood in above buffer. The suspension was transferred to a glass dounce homogenizer and homogenized. Following centrifugation at different speeds at 4°C, the pellet containing mitochondria was resuspended in an approximate volume in isolation buffer. The total mitochondrial protein was determined using Pierce BCA Protein Assay kit (Thermo Fisher Scientific). Finally, the mitochondrial (5mg/well) preparations were used for the coupling assay by the Agilent seahorse XFe/XF96 analyzer (Agilent Technologies, Santa Clara, CA, USA). The mitochondrial respiration was measured with different substrates (glutamate, malate, succinate and pyruvic acid) present. The injections were as following final concentration: ADP was 1 mM, oligomycin was 2.5mg/ml, FCCP was 2.2 mM/ml and rotenone was 2 mM/ml [19].

Statistics

Data analyses were performed with GraphPad Prism 8.0. All the measurement was performed in Image J. CCM data were processed by mixOmics packages with R software. A supervised Patrial Least Squared discriminant analysis (PLS-DA) was performed to visualize general clustering and trends. PLS-DA scored plots of components one and two, comparing different diet groups were generated. KEGG pathway analysis was conducted in MetaboAnalyst 4.0. Two group comparisons were using a paired two-tailed student’s T-test. The data of more than two groups were analyzed using one-way ANOVA followed by Turkey’s post-hoc t-tests to compare all possible pairs of experimental groups. A value of P < 0.05 was considered significant.

Loss of Pex16 leads to hepatic steatosis

We generated a liver specific Pex16 knockout (Pex16 KO) mouse model by Albumin-Cre-mediated homologous recombination of a floxed allele (Fig. 1a). Western blot analysis of liver tissue showed loss of PEX16 and peroxisomal membrane protein 70 (PMP70) at postnatal week 3 indicating efficient Pex16 deletion and peroxisome loss (Fig. 1b). Decreased immunofluorescence of PMP70 confirmed peroxisome ablation in hepatocytes (Fig. 1c, Fig. S1a). Pex16 KO mice had a decreased body weight compared to wild-type control animals at postnatal week 3 (Fig. 1d, 1e). However, the liver weight of Pex16 KO mice was significantly increased (Fig. 1d, 1f). Although the liver tissue structure appeared normal (Fig. 1g), Oil Red O staining revealed an increased presence of oil droplets in the livers of 3- (Fig. 1h) and 5-week-old (Fig. S1b) Pex16 KO mice. Together, these results show that hepatic loss of PEX16 leads to disappearance of peroxisomes and hepatic steatosis.

Nutritional Stress Induced With A Low Protein Diet Exacerbates Loss Of Peroxisome-induced Hepatic Steatosis

To determine the susceptibility of Pex16 KO mice to nutritional stress-induced metabolic change, we subjected Pex16 KO mice to a low protein diet (LPD) (Fig. S2) for 2 weeks starting at weaning age (3 weeks) (Fig. 2a). The LPD is an isocaloric diet that only differs in its total protein content compared to normal chow (1% for the LPD vs 18% for the control diet). As expected, body weight and food intake were significantly decreased during LPD feeding in Pex16 KO mice as well as in the wild-type controls, but food intake was similar when corrected for body weight (Fig. 2b, 2c). The liver size and weight relative to the body weight were also decreased in mice fed a LPD (Fig. 2d, 2e). Further, the enlargement of the liver seen in the Pex16 KO mice was no longer observed in the Pex16 KO LPD animals (Fig. 1d, 1f, Fig. 2d, 2e). Histological analysis showed more vacuoles in liver sections of Pex16 KO mice which was exacerbated by LPD feeding (Fig. 2f). Oil Red O staining confirmed that such vacuoles correspond to lipid droplets (Fig. 2g). Lipid droplets in the Pex16 KO mice fed a regular diet appeared to be fewer and smaller than in the wild-type mice fed a LPD (Fig. 2g, 2h). In contrast, lipid droplets were more numerous in the LPD-fed Pex16 KOs than wild-type controls fed the LPD. Lipid staining with BODIPY further demonstrated the Pex16 KOs fed a LPD to have the most severe hepatic steatosis (Fig. 2h, 2i), indicating that LPD feeding exacerbates hepatic steatosis in Pex16 KO mice.

Loss Of Peroxisomes Leads To Disturbed Hepatic Energy Homeostasis And Is Exacerbated By Nutritional Stress

To determine the effect of depleting peroxisomes in the metabolic response of the liver during normal conditions and during exposure to nutritional stress we first measured fasting blood glucose levels in Pex16 KOs. During fasting, the liver becomes primarily responsible for gluconeogenesis, an ATP demanding activity. Fasting glucose concentrations were not affected in the Pex16 KOs compared to wild-type mice (Fig. 3a, Fig. S3a). Fasting glucose levels were reduced by a LPD and exacerbated by loss of PEX16, indicating reduced hepatic glucose production. ATP content was similarly decreased after LPD feeding in Pex16 KO and wild-type mice (Fig. 3b). We further analyzed the hepatic intermediate metabolome targeting the central carbon metabolism. A total of 53 metabolites were detected. Pex16 KOs were clearly separated by their metabolic profile from wild-type controls (Fig. S3b). Similarly, LPD feeding strongly affected the overall metabolite profile, but metabolic profiles did not clearly separate Pex16 KOs from wild-type mice fed a LPD. However, the heat map did suggest metabolic differences between these two groups (Fig. 3c). The levels of each metabolite are shown in the heat map and the tables (Fig. 3c, and Table S4, S5). Pathway enrichment analysis revealed that the tricarboxylic acid (TCA) cycle was affected by loss of PEX16 as well as by LPD feeding (Fig. 3d). Specifically, entry to the TCA cycle appeared to be blocked in wild-type mice fed a LPD as shown by an increase in citrate content and a concomitant decrease in acetyl-CoA content. The glycolytic pathway was also altered in Pex16 KOs, exacerbated by LPD feeding, and associated with substantial reductions in NAD⁺. These results indicate that loss of PEX16 affects overall energy homeostasis and demonstrate a metabolic stress pattern that has similarities with that induced by LPD feeding.

Pex16 Ko Mice Show Decreased Mitochondrial β-oxidation

To better understand the etiology of hepatic steatosis by loss of PEX16, we assessed relevant lipid pathways in the liver. We found that the loss of peroxisomes led to increased hepatic triglyceride content, consistent with our histological findings (Fig. 3e). This was further exacerbated by a LPD. As hepatic steatosis can be caused by increased lipogenesis, we examined expression levels of lipogenesis regulatory genes. qPCR of acetyl-CoA carboxylase alpha (ACACA) and fatty acid synthase (FAS) were decreased in the Pex16 KO mice and in both the wild-type and KO mice fed an LPD (Fig. S3c). We explored the possible inhibition of the lipogenic pathway by quantifying the active phosphorylated AMP-activated protein kinase alpha (AMPKα) in liver. Phosphorylated AMPKα protein levels and pAMPKα/ AMPkα were increased by PEX16 deficiency and more modestly by feeding a LPD (Fig. 3f). Consistent with increased AMPKα phosphorylation, the expression of genes regulating mitochondrial β-oxidation, peroxisomal β-oxidation and CoA formation were increased in Pex16 KO mice fed a control diet (Fig. S3d-f). However, these increases in β-oxidation genes did not occur in Pex16 KO mice fed a LPD. The protein level of nuclear hormone receptor peroxisome proliferator-activated receptor alpha (PPARα), an important transcriptional regulator of β-oxidation was reduced in the Pex16 KO animals, but most profoundly decreased in both LPD fed animals (Fig. 3f). The specific metabolic profile changes, together with upregulation of genes controlling β-oxidation despite lower PPARα protein levels suggest altered mitochondrial function in Pex16 KO mice.

Hepatic Mitochondrial Dynamics Is Differentially Altered In Response To Loss Of Peroxisomes And Nutritional Stress

To assess the mitochondria in the Pex16 KO mice, we examined hepatocyte mitochondria morphology during normal and metabolically stressed states. Mitochondria morphology is dynamic, elongating or dividing to meet the metabolic needs of the cell [8]. Further, damaged mitochondria may fuse with healthy mitochondria to repair, or the irreparable portion of mitochondria may undergo fission to be degraded by autophagy. We first examined the effects of the absence of peroxisomes on mitochondrial morphology by immunofluorescence of the mitochondrial heat shock protein HSP60 (HSP60) (Fig. 4a) and transmission electron microscopy (TEM) (Fig. 4b, Fig. S4a). In Pex16 KO mice the number of mitochondria appeared to be increased (Fig. 4a), but smaller compared to wild-type controls (Fig. 4b-c, circles with dashed lines). The increase in mitochondria in the Pex16 KO liver was supported by an increase in the mitochondrial DNA copy number (mtDNA) (Fig. 4d) and higher levels of the outer membrane translocase 20 (TOM20) and matrix protein heat shock protein 60 (HSP60) (Fig. 4e) compared to wild-type control mice, indicating increased cellular mitochondrial content.

When the mice were subjected to metabolic stress with the LPD, both the wild-type and Pex16 KO mice showed an increase in mitochondria size compared to their normal feed counterpart (Fig. 4a-c), but a reduction in the number, protein levels and mtDNA within the hepatocytes. However, the mitochondria of the LPD treated Pex16 KO mice were smaller but more numerous than the wild-type LPD mice based on the quantification of mitochondria structures and mtDNA (Fig. 4a-e). We also examined for changes in peroxisomal proteins in the livers of LPD-fed mice. In the wild-type LPD-fed mice, the level of liver peroxisomal proteins PMP70 and PEX16 was substantially decreased like those shown in livers of amino acid starved rats [17] (Fig. S4b).

We next assessed whether changes in mitochondrial morphometrics were related to effects on mitochondrial fusion and fission by assessing Mitofusin-2 (MFN2) and dynamin-1-like protein (DRP1), respectively. Western blot showed that MFN2, total DRP1, and the fission inhibited phospho-DRP-S637 (pDRP) were found to be increased in the Pex16 KO mice compared to the wild-type control (Fig. 4f). However, as the Pex16 KO hepatocytes have more mitochondria (Fig. 4a-d), we used TOM20 to normalize the protein levels and found that both MFN2 and DRP1 was lower in the Pex16 KO compared to the wild-type livers, but he pDRP/DRP1 ratio were not different. Taken together, our data do not conclusively demonstrate that the smaller mitochondria in Pex16 KO hepatocytes are related to increased mitochondrial fission. Both LPD-fed mice conditions showed an overall decrease in these morphology regulating proteins with respect to wild-type control animals. When combined with the lower pDRP1/DRP1 ratio (Fig. 4f), it suggests a possible mechanistic explanation for the elongated and larger mitochondria morphology of the LPD fed animals compared to their normal chow counterparts. These analyses of hepatic mitochondria also suggest that the mitochondria respond differently to the loss of peroxisomes compared to LPD, in terms of their size and abundance, such that the combined insults result in an intermediate morphological phenotype.

Loss Of Peroxisomes Exacerbates Mitochondrial Dysfunction In Response To Nutritional Stress

To assess whether the mitochondrial morphological changes were associated with changes to mitochondrial function, we next assessed the mitochondrial health by measuring mitochondrial oxygen consumption. Basal and maximal respiration of isolated hepatocyte mitochondria was assessed in the different mouse groups as a measure of mitochondrial function. Baseline cellular OCR was not different between the groups, but maximal respiration was increased in the Pex16 KO mice compared to controls (Fig. 4g). Interestingly, the mitochondria of the Pex16 KO liver showed a gradual decrease in OCR during FCCP treatment that was not observed in the WT mice suggesting that their mitochondria may be more susceptible to damage. The mitochondria for both LPD-fed groups showed very little respiration reserve as their maximal respiration were not higher than the basal. Further, the Pex16 KO LPD animals showed more pronounced loss of OCR with time supporting the increase susceptibility of these mitochondria to damage. These results suggest that hepatic mitochondrial function is not impaired by loss of peroxisomes, consistent with no change in ATP content at basal conditions, but they are more affected by nutritionally-induced metabolic stress compared mitochondria from wild-type animals.

Next, we examined the isolated mitochondria for possible relative differences in their electron transport chain (ETC) complex components by immunoblot analysis. We normalized mitochondria complex proteins to cytochrome c oxidase subunit 4 (COX IV), the terminal enzyme in the respirator chain. The content of mitochondrial complex II (SDHB), III (UQCRC2) and V (ATP5A) when normalized to COX IV were modestly decreased in isolated mitochondria of Pex16 KOs mice, while protein content of complex I (NDUFB8) and IV (MTCO1) was found to be modestly higher compared to wildtype mice (Fig. 4h). Feeding a LPD led to a more substantial decrease in all of the complexes, including in the Pex16 KO mice. For the LPD fed animals, no differences were observed between the wild-type and knockout animals. However, the difference in the complexes small and thus unlikely explain the difference in the maximal mitochondria output between the two groups. In addition, COX IV was not stable and its content appeared to increase in the mitochondria of the Pex16 KO mice, suggesting that per mitochondrial protein content the content of all complexes was increased in the Pex16 KO mice fed the control diet. Together with the mitochondria stress test, these results suggest that mitochondria in the Pex16 KO animals are able to adapt under normal conditions but are more suspectable to damage under metabolic stress.

Pex16 Deficiency Alters Mitochondria Quality Control Related To A Decrease In Autophagy

One indicator of mitochondria damage is the accumulation of PTEN-induced kinase 1 (PINK1), an inner mitochondria membrane targeting proteins that is readily cleavage by inner membrane protease PARL, retro-translocated and degraded by proteasomes in healthy mitochondria, whereas it accumulates in damaged mitochondria [20]. The accumulation of PINK1 initiates PARKIN dependent mitophagy which includes mitochondria fragmentation and the recruitment of autophagic factors on damaged mitochondria [21]. In the Pex16 KO livers, we observed both an increase in fragmented mitochondria (Fig. 4c), an increase in mitochondrial mass as determined by the increased mtDNA (Fig. 4d) and increase in mitochondrial proteins (Fig. 4e) suggesting a possibility that mitochondrial quality control mechanisms such as mitophagy may be negatively affected. To assess whether mitophagy might be impacted in the PEX16 deficient livers, we quantified the protein levels of PINK1. Full length PINK1 was barely detectable in all conditions (Fig. 4i). When normalized to β-actin, we did find an increase in cleaved PINK1 in the PEX16 deficient livers for both normal and LPD fed conditions (Fig. 4f). LPD treatment led to a small but significant decrease in PINK1 compared to normal fed WT-mice. Similar results were observed when cleaved PINK1 was measured in isolated mitochondria and adjusted for mitochondrial content through COX IV (Fig. 4f). It is unclear what the accumulation of the 42 kDa cleaved PINK1 signifies, as it should be retro-translocated from the mitochondria and degraded by proteosomes [22, 23]. However, we did not observe an accumulation of full-length PINK which would be indicative of significantly damaged mitochondria.

We next examined whether autophagy may be downregulated in PEX16 KO livers. We first analyzed the autophagy receptor protein sequestosome 1 (p62) and microtubule-associated protein 1A/1B light chain 3B (LC3B). p62 is turned over by autophagy and thus accumulate in cells when autophagy is defective (ref). LC3B on the other hand increases as its cleaved from LC3B-II during upregulated autophagy, however, its accumulation may be an indicator of a reduction of autophagy flux, the turnover of autophagosomes. We found that both p62 and LC3B-II were strongly increased in the Pex16 KOs compared to wild-type controls fed a regular diet (Fig. 5b). Acute administration of chloroquine to inhibit autophagic flux led to moderate and similar increase in p62 and LC3B-II levels in the livers of the wild-type and Pex16 KO mice (Fig. 5c). Combined, this data suggests that the rate of autophagy flux is retarded in the livers of Pex16 KO mice compared to wild-type animals. This in turn may be contributing to the accumulation of mitochondria in the liver of Pex16 KO mice (Fig. 4a-e).

Given that amino acid starvation can upregulate autophagy, an increase autophagy flux is expected. As such we found that LPD feeding in both wild type and Pex16 KO mice showed reduced p62 as well as LC3 content compared to normal fed animals. The LC3B-II levels were modestly upregulated in the Pex16 KOs fed a LPD compared to the wild type mice fed a LPD (Fig. 5d). Upon chloroquine administration there was a robust increase in LC3B-II content in the LPD fed groups, suggesting that the amino acid starvation may be over-riding the reduction in autophagy flux observed in the Pex16 KOs.

mTORC1 is the main regulator or autophagy, where the inhibition of mTORC1 kinase activity results in the induction of autophagy [24]. To further investigate whether the impact on autophagy by loss of the PEX16 protein was related to changes in mTORC1 activation, we analyzed the phosphorylated p70S6K (p-p70S6K) and phosphorylated 4EBP1 (p-4EBP1), two substrates of mTORC1 kinase activity. p-p70S6K/p70S6K ratios were down-regulated in the Pex16 KO mice consistent with inhibition of mTOR, but without a significant effect on p-4EBP1/4EBP1 ratio (Fig. 5e-f). Both LPD feeding animal groups showed a decrease in both the p-p70S6K/p70S6K and p-4EBP1/4EBP1 ratios supporting the increase in autophagy. Interestingly, loss of peroxisomes increased total protein levels of 4EBP1 and p70S6K, with mRNA levels p70S6K most dramatically upregulated in Pex16 KO mice (Fig. S5a, S5b), suggesting an effect on transcription of this protein. Increased reactive oxygen species (ROS) have been shown to upregulate 4EBP and S6K transcription [25]. One could speculate that increased ROS is present in our model systems thereby affecting concentration of these proteins. This increase in 4EBP1 and p70S6K levels in the knockout mice limits our ability to draw firm conclusions on mTOR activation in these mice but may explain the lack of a difference in the p-4EBP1/4EBP1 ratio in comparison with the wild-type mice. In summary, these set of results indicates that the accumulation of damaged mitochondria in the Pex16 KO mice may be the result of a downstream impairment in overall autophagic flux, despite a suggested inhibition of mTORC1.

Pex16 Ko Mice Demonstrated Increased Mitochondrial Biogenesis

To determine whether changes in mitochondrial content were related to changes in mitochondrial biogenesis, we analyzed the levels of regulators of mitochondrial biogenesis peroxisome-proliferator-activated receptor coactivator-1α (PGC-1α) and its downstream target nuclear respiratory factor 1(NRF1) [26]. PGC-1α and NRF1 protein (Fig. 5d) and Pgc-1α and Nrf1 mRNA (Fig. S5c, S5d) leves were elevated in the liver of Pex16 KO compared with wild-type mice, consistent with the increase in mitochondrial number and mass. In contrast, PGC-1α was decreased to comparable levels in the controls and Pex16 KOs fed a LPD compared with mice fed a regular diet (Fig. 5e, S5c). Our results indicate that increased mitochondrial content in Pex16 KO is not only related to reduced autophagy but also to increased mitochondrial biogenesis. In contrast, the reduced mitochondrial content in mice fed a LPD, appears to be related to reduced mitochondrial biogenesis.

Activation of PPARα improves hepatic steatosis caused by loss of peroxisomes and low protein diet feeding

PPARα is a nuclear receptor that enhances mitochondrial biogenesis and β-oxidation, and more recently been demonstrated to also induce autophagy [17, 27]. Therefore, we next assessed whether PPARα activation was able to improve hepatic metabolism in our Pex16 KO mice in the context of LPD feeding as well as in the single insult models (Fig. 6a). Fenofibrate, a PPARα ligand, did not affect liver weight, food intake, body weight or fasting glucose concentrations in Pex16 KO or wild-type mice fed a control or a LPD (Fig. 6b, S6a-b, S6g-h). However, hepatic steatosis was significantly decreased in Pex16 KO mice fed either a control or LPD that were treated with fenofibrate compared to non-treated mice (Fig. 6c-g, Sf-h). Fenofibrate had a similar effect in wild-type mice fed a LPD. As control, vehicle (DMSO) treatment did not affect hepatic lipid accumulation (Fig. S6c-f, S6h).

Pparα Activation Improves Hepatic Mitochondrial Function Independently Of The Presence Of Peroxisomes

We next examined the effect of activating PPARα on mitochondrial fitness in Pex16 KOs fed a LPD and regular diet. Fenofibrate treatment increased peroxisome numbers in wild-type controls fed a LPD (Fig. 7a) consistently with an increase in PMP70 (Fig. S7e). Peroxisomal structures were not observed in the EM micrographs of the Pex16 KO livers (Fig. 7a), although we did not perform an alkaline 3,3′-diaminobenzidine staining to directly visualize peroxisomes. We therefore further validated our EM findings using immunofluorescence PMP70 staining, which showed an increase in PMP70 protein levels observed with fenofibrate treatment, including small increases in Pex16 KO livers (Fig. S7e). However, the PMP70 appeared to co-localize with endothelial cells as indicated by co-staining with CD31, an endothelial marker (Fig. S7f), indicating that fenofibrate treatment is able to stimulate non-hepatocyte peroxisome biogenesis. Mitochondria were smaller after fenofibrate treatment in Pex16 KOs fed a LPD (Fig. 7a-b), however, their numbers were not affected by fenofibrate (Fig. 7b). In contrast, the Pex16 KOs on regular diet treated with fenofibrate showed no clear change in the size of mitochondria but showed a reduction in number compared to untreated mice (Fig. S7a-c). For the wild-type controls fed a LPD, fenofibrate treatment led to increased numbers of mitochondria that were smaller and less elongated (Fig. 7a-c). Consistently, Tom20 protein levels and mtDNA content were not significantly increased with fenofibrate treatment in the Pex16 KOs fed a LPD, but were increased in wild-type animals (Fig. 7c, 7d). The effects on mitochondrial markers were associated with a fenofibrate-induced decrease in p-AMPKα and increased PPARα protein levels (Fig. 7e). As an additional control, DMSO treatment in wild-type controls on a control diet did not affect the mitochondrial content (Fig. S7a-e).

As expected, fibrate treatment led to an increase in mRNA levels of most genes involved in mitochondrial ${\beta }$-oxidation (except for Acadv and Acada1), peroxisomal ${\beta }$-oxidation and CoA synthesis in Pex16 KO and wild-type mice fed a LPD (Fig. S7g). mRNA levels of lipogenesis markers Acaca and Fasn were further decreased in fenofibrate-treated mice. The effects of fibrate treatment on expression of genes in these pathways were generally lacking in Pex16 KO mice fed a regular diet (Fig. S7h).

Mitochondrial protein complexes in isolated mitochondria were increased by fenofibrate in wild-type fed a LPD mice compared to untreated mice, although the effect appeared less pronounced in the Pex16 KOs (Fig. 7f). No substantial increases in mitochondrial complex proteins were observed in the Pex16 KOs fed the control diet, with the exception of complex IV and even a small decrease in complex II and V (Fig. S7i). This data suggests an improved mitochondrial function after PPARα activation in Pex16 KOs fed a LPD, associated with increases in mitochondrial complex content without an effect on overall mitochondrial mass.

In line with these results, the ATP content was similarly increased in Pex16 KOs fed a LPD after fenofibrate treatment compared to wild-type controls (Fig. 7g). Finally, mitochondrial respirometry showed an improvement in respiration in the Pex16 KOs with or without LPD feeding (Fig. 7h, S7j). These results indicate that PPARα activation improves mitochondrial fitness in mice subjected to LPD-induced nutritional stress despite an absence of functional peroxisomes.

Pparα Activation Stimulates Mitochondrial Biogenesis And Autophagy In Pex16 Ko Mice

We next assessed the effects of PPARα activation on markers of mitochondrial dynamics, biogenesis and autophagy. The MFN2 protein and pDrp1/Drp1 were increased in the fenofibrate-treated Pex16 KOs fed a regular or LPD compared to untreated animals (Fig. 8a, S8a). Both Pex16 KOs as well as the wild-type animals that were fed a LPD and treated with fenofibrate showed an increase in autophagosome formation (ratio LC3B-II/I), and a decrease in p62 (Fig. 8b). The decrease p62 with increased LC3B-II/I suggest an increase in autophagic flux. Similar changes in the autophagic factors also observed in fenofibrate treated Pex16 KO mice fed the control diet (Fig. S8b).

The effects on autophagy were associated with a decrease in p-p70S6K/p70S6K indicating inhibition of mTOR (Fig. 8c). However, fenofibrate treatment increased p-4EBP1/4EBP1 in all mouse group (Fig. 8d). mRNA levels of these factors showed a similar pattern (Fig. S8d). PGC-1α and NRF1 were upregulated in Pex16 KOs as well as wild-type animals after fenofibrate treatment (Fig. 8e, S8e). Together, these data suggest that the improved mitochondrial health observed after PPARα activation in Pex16 KOs fed a LPD are related to increased mitochondrial biogenesis and enhanced autophagy.

In this study, we describe a novel mouse model that are deficient in peroxisomal membrane structures in its liver. We demonstrate that loss of PEX16 results not only in the loss of peroxisomal membrane structures, but leads to disturbed mitochondrial homeostasis which is exacerbated by nutritional stress induced through low protein diet feeding. We show that the change in the mitochondria morphology and numbers in the Pex16 KO mice appears to be able to support ATP production at similar levels as the wild-type animals, suggesting that the hepatocytes are able to compensate for the loss of peroxisomal function. However, not all metabolic pathways were protected as a decrease in the ${\beta }$-oxidation pathway resulted in an increase in lipid droplet with in hepatocytes. Metabolically challenging the hepatocytes with a low protein diet resulted in a more severe mitochondrial dysfunction in the PEX16 KO mice compared to wild-type mice. Activating PPARα, however, improves hepatic mitochondrial and overall metabolic function in the absence of functional peroxisomes by stimulating mitochondrial biogenesis and autophagy. Our work not only gives further support for a role of peroxisomes in maintaining mitochondria health, but it also reveals the ability of hepatocytes to adapt to an absence of peroxisomes to maintain mitochondrial health.

Loss Of Pex16 And Peroxisomes Affect Mitochondrial Homeostasis And Function

PEX16 is essential for peroxisome biogenesis as its absence results in the loss of peroxisomal membrane [28] and is associated with Zellweger syndrome [7] and leukodystrophy [29]. We developed a mouse model with liver-specific Pex16 deficiency as demonstrated by loss of PEX16 and the peroxisome membrane protein, PMP70. These mice develop hepatic steatosis similar to other models of loss of peroxisomes, such as in Pex1 and Pex5 deficiency [12, 13]. Mitochondrial abnormalities are found in patients with Zellweger syndrome [30] and mouse models with deletion of Pex2 and Pex5 [13, 14, 31], underscoring the role of peroxisomes in mitochondrial health. In this study, we show that loss of Pex16 leads to accumulation of small mitochondria, but only minor disruption in mitochondrial function was observed such as a decrease in ${\beta }$-oxidation. Mitochondria morphological changes with no ATP disruption was also report for the PEX5 KO mice [13]. Although, they did observe mitochondria that were visually damaged in their EM micrographs, whereas we did not. The difference could be related to to the age of the animals examined (5-week vs 20-week old mice). However, we did observe that the mitochondria isolated from the Pex16 KO mice were more susceptible to damage. This was evidenced in our mitochondria stress test where FCCP treatment resulted in a rapid loss of oxygen consumption rate (Fig. 4g). Therefore, it is possible that our Pex16 KO mice may also accumulate more damaged mitochondria with age. Further, we did observe an increase in cleaved PINK1. The cause of the accumulation of cleaved PINK1 in the Pex16 KO liver is not known. However, given that PINK1 in health cells is readily cleaved by the inner mitochondria membrane protease PARL, retro-translocated out to the cytosol and degraded by the proteasome in healthy mitochondria [32], an accumulation of the cleaved form may indicate a mild defect in the mitochondria.

The mitochondria respiration analysis of isolated mitochondria also showed that the Pex16 KO mice have a higher maximum respiration capacity compared to the wild-type. although the difference was very short lived as evident from the rapid loss of OCR. Therefore, we propose that the changes in hepatocyte mitochondria in the Pex16 deficient mice may be a compensatory response to maintain mitochondrial function. However, these changes make these mitochondria more susceptible to damage during metabolically stressed conditions.

Decreased autophagy and induction of mitochondria biogenesis contributes to a mitochondria mass increase in Pex16 KO mice.

Our studies suggest that the increase in mitochondrial mass might be caused by two different mechanisms. As AMPK was activated by loss of peroxisomes (Fig. 3), this could induce mitochondrial biogenesis through the AMPK-dependent activation of PGC-1α [33]. The increase in activated pAMPK also inhibits mTORC1 and activates autophagy [34]. We found in the livers of Pex16 KO an increased pAMPK. Another group showed that pAMPK levels were also increased in liver specific Pex5 deficiency [35]. Our autophagy assays suggest that hepatocyte autophagic flux is decreased in Pex16 KO mice. We would speculate that increase mitochondrial biogenesis and reduced autophagic flux are part of the compensatory mechanism allowing mitochondrial function to be preserved by increasing mitochondrial mass. Down regulating the the ß-oxidation pathway could also potentially further prevent further mitochondrial damage by limiting the production of reactive oxygen species.

Loss Of Peroxisomes Exacerbates Low Protein Diet Induced Mitochondrial Changes Through Separate Pathways

Nutritional stress induced by low protein feeding in weaning mice exacerbated the hepatic steatosis caused by Pex16 deficiency. Interestingly, cold-induced stress also exacerbated mitochondrial damage in brown adipocytes in adipose-specific Pex16 KO mice [36]. Our data suggests that the specific loss of peroxisomes play a role in the mitochondrial changes induced by a LPD [17]. However, our data does indicate that low protein diet-induced mitochondrial changes are not exclusively due to the loss of peroxisomes but are partially independent from the loss of peroxisome-induced mitochondrial changes. This is mainly supported by the observation that the loss of PEX16 led to the accumulation of small mitochondria whereas a LPD induced a reduction in mitochondrial mass and the formation of elongated and abnormally shaped mitochondria. The small mitochondria seen in the Pex16 KO mice were not described in the livers of a Pex5 deficient model [12], and might also be related to the fact that Pex16 is involved in the early steps of peroxisome biogenesis, whereas Pex5 is involved in peroxisomal matrix protein import with potential differential effects on mitochondrial homeostasis. Interestingly, low protein feeding was associated with a reduction in LC3B-II and p62 compared to normal chow diet animals. This decrease is likely due to an increase in autophagy flux as seen by the increase in LC3B-II in animals treated with Chloroquine (Fig. 5d) and pAMPK levels. These findings are contrary to what we previously demonstrated in a rat model of low protein diet-induced loss of peroxisomes and hepatic mitochondrial dysfunction where an accumulation of damaged mitochondria was observed in LPD fed animals [17]. This might be related to differences in severity of the two model systems, or a temporal difference as the rats were place in a 4-week low protein diet instead of the 2-weeks for the mice.

Fibrates Improve Mitochondrial Function In Pex16 Ko Low Protein Fed Mice

PPARα induces peroxisome and mitochondrial biogenesis and increase ${\beta }$-oxidation, and has more recently been shown to induce autophagy [27]. Previously, we showed that activation PPARα was able to reverse the peroxisome loss and improve mitochondria fitness in the hepatic of rats subjected to LPD [17]. Here, we examined whether peroxisomes were necessary for the possible rescue of PPARα induction on mitochondrial function. Fibrate treatment was able induce peroxisomes biogenesis genes (Fig. 7a) and induce the expression of peroxisomal and mitochondrial ${\beta }$-oxidation associated with a reduction in lipid content in both low protein-fed Pex16 deficient and wild-type mice, suggesting that this effect is independent on the presence of peroxisomes. Fibrates also increased mitochondrial biogenesis, which is consistent with recent observations [37], likely related to the induction of PPARα. Fibrate treatment led to an induction of autophagy as indicated by decreased mTORC1 activity, an increased LC3B-II and decreased p62 content. Activation of PPARα has been shown to regulate autophagy transcriptionally (20), and could account, in part, for the increase in autophagy. Of interest, fibrate treatment has been tried in a few patients with Zellweger Syndrome without any clear impact on clinical outcome (30, 31) which suggests that mitochondria damage is not the sole pathology of Zellweger syndrome.

In conclusion, this study describes the phenotypical consequences of hepatic loss of Pex16 which are exacerbated by nutritional stress induced by low protein feeding and demonstrates that Pex16 deficiency leads to enhanced production of mitochondria and reduced autophagy resulting in an increase in mitochondria (Fig. 9). These mitochondria do have a high respiration potential but are more susceptible to nutritionally-induced damage. PPARα treatment beneficial effects on mitochondrial health can be established independent of peroxisome function by reversing the reduction in autophagy flux caused by the loss of PEX16 and enhancement of mitochondrial biogenesis. Our finding that mitochondrial dysfunction can be uncoupled from peroxisome loss is of relevance for children with peroxisomal biogenesis disorders and suggest that interventions to restore mitochondrial homeostasis and preventing malnutrition could improve their outcome.

Acaa1a Acetyl-coenzyme A carboxylase carboxyl transferase subunit α

Acaca Acetyl-CoA Carboxylase α

Acadm1 Acyl-CoA Dehydrogenase Medium Chain1

Acadv1 Very long-chain specific acyl-CoA dehydrogenase 1

Acadl Acyl-CoA Dehydrogenase Long Chain

Acca2 Acetyl-CoA carboxylase α subunit

Abcd3 ATP Binding Cassette Subfamily D Member 3

Acot12 Acyl-CoA Thioesterase 12

Acox1 Peroxisomal acyl-coenzyme A oxidase 1

Acsl1 Acyl-CoA Synthetase Long Chain Family Member 1

ADP Adenosine diphosphate

AMP Adenosine monophosphate

AMPK AMP-activated protein kinase

ATP Adenosine triphosphate

CQ Chloroquine

DHAP 3-Deoxy-D-arabinoheptulosonate 7-phosphate

DRP1 Dynamin-1-like protein

DMSO Dimethylsulfoxide

Ech1 Enoyl-CoA Hydratase 1fabp1

ER Endoplasmic reticulum

4EBP1 Eukaryotic translation initiation factor 4E binding protein 1

Fabp1 Fatty Acid Binding Protein 1

Fasn Fatty Acid Synthase

FCCP Carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone

FF Fenofibrate

GMP Guanosine monophosphate

GDP Guanosine diphosphate

GTP Guanosine triphosphate

H&E Haemotoxylin and eosin

HSP60 Heat shock protein 60

IP Intraperitoneal

KO Knockout

LC3B Microtubule-associated protein light chain 3B

LPD Low protein diet

MFM2 Mitofusin-2

mtDNA mitochondrial DNA

NAD Nicotinamide adenine dinucleotide

NADH Nicotinamide adenine dinucleotide hydride

NADPH Nicotinamide adenine dinucleotide phosphate

NRF1 Nuclear respiratory factor 1

OCR Oxygen consumption rat

PBD Peroxisome biogenesis disorders

PEP Prolyl endopeptidase

PINK1 PTEN Induced Kinase 1

PMP70 Peroxisomal membrane protein 70

p62/SQSTM1 Autophagy adaptor protein sequestosome 1

p70S6K Ribosomal protein S6 kinase β-1

Pex Peroxisome

PPARGC1A Peroxisome-proliferator-activated receptor coactivator-1α

PPARα Peroxisome Proliferator Activated Receptor α

Rpl13a Ribosomal protein L13a

TCA Tricarboxylic acid

TEM Transmission electron microscopy

TOM20 Membrane translocase 20

UMP Uridine monophosphate

UDP Uridine diphosphate

UTP Uridine triphosphate

Ethics approval and consent to participate

Procedures followed the Canadian Council for Animal Care guidelines, and were approved by the Animal Care Committee at the Hospital for Sick Children.

Consent for publication

N/A

Availability of data and material

The datasets generated during and/or analysed during the current study are available in the BioStudies repository, https://urldefense.com/v3/__https://www.ebi.ac.uk/biostudies/studies/S-BSST1001?key=e6f1cd62-9f3e-4066-8f30-503991df0aa8__;!!D0zGoin7BXfl!-zYZ4ryBdtkgF9VdWOMLboOXrn38pxiVq2ixIGpAqJ9C0z16eKd_gcSDVGkhhVFaSrXWQNdGpE-nJnSPoGJ_cAQ3_Z4QIA$.

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Funding

This work was supported by Canadian Institutes of Health Research (Grant reference number 156307)

Authors' contributions

Lijun Chi and Dorothy Lee designed the experiments, performed the experiments and wrote the manuscript. Sharon Leung generated and initially characterized the Albumin-Cre Pex16 KO mouse model and edited the manuscript. Guanlan Hu analyzed the Central carbon metabolism assay profiling, helped interpret this data and edited the manuscript. Bijun Wen performed measurements of mitochondrial area from TEM images and edited the manuscript. Paul Delgado provided expertise on interpretation of the mitochondrial data and edited the manuscript. Ren Li and John Brumell performed TEM experiments with Joh Brumell also providing input in interpretation of the autophagy data and editing of the manuscript. Peter Kim and Robert Bandsma conceived the project, designed the experiments and wrote and edited the manuscript.

Acknowledgements

We would like to acknowledge Prof Barbara Bakker for the excellent technical support the mitochondrial function assays. We thank Michelle Huong (Department of Human Biology, University of Toronto) for help the genotype, Dr. Abdalla Ahmed (The Hospital for Sick Children) for feedback and suggestions, Dr. Dave Schibli and Dr. Jun Han (UVic Genome BC Proteomic Centre and The Metabolomics Innovation Centre) for CCM assay, Paul Paroutis and Kimberly Lau (Imaging Facility at The Hospital for Sick Children) for help with confocal microscopy and LAS (Lab Animal Services, The Hospital for Sick Children) for mouse husbandry and car

Farré J, Mahalingam SS, Proietto M, Subramani S (2019) Peroxisome biogenesis, membrane contact sites, and quality control. EMBO Rep 20:e46864. https://doi.org/10.15252/embr.201846864
Fransen M, Lismont C, Walton P (2017) The peroxisome-mitochondria connection: How and why? Int J Mol Sci 18:1126. https://doi.org/10.3390/ijms18061126
Mooi WJ, Dingemans KP, Van Den Bergh Weerman MA, et al (1983) Ultrastructure of the liver in the cerebrohepatorenal syndrome of zellweger. Ultrastruct Pathol 5:135–144. https://doi.org/10.3109/01913128309141833
Trijbels JMF, Berden JA, Monnens LAH, et al (1983) Biochemical studies in the liver and muscle of patients with zellweger syndrome. Pediatr Res 17:514–517. https://doi.org/10.1203/00006450-198306000-00018
Germain K, Kim PK (2020) Pexophagy: A model for selective autophagy. Int J Mol Sci 21:578. https://doi.org/10.3390/ijms21020578
Soliman K, Göttfert F, Rosewich H, et al (2018) Super-resolution imaging reveals the sub-diffraction phenotype of Zellweger Syndrome ghosts and wild-type peroxisomes. Sci Rep 8:. https://doi.org/10.1038/S41598-018-24119-2
Honsho M, Tamura S, Shimozawa N, et al (1998) Mutation in PEX16 is causal in the peroxisome-deficient Zellweger syndrome of complementation group D. Am J Hum Genet 63:1622–30. https://doi.org/10.1086/302161
Chan DC (2020) Mitochondrial Dynamics and Its Involvement in Disease. Annu Rev Pathol Mech Dis 15:235–259. https://doi.org/10.1146/annurev-pathmechdis-012419-032711
Ploumi C, Daskalaki I, Tavernarakis N (2017) Mitochondrial biogenesis and clearance: a balancing act. FEBS J 284:183–195. https://doi.org/10.1111/febs.13820
Kim I, Rodriguez-Enriquez S, Lemasters JJ (2007) Selective degradation of mitochondria by mitophagy. Arch Biochem Biophys 264:245–253. https://doi.org/10.1016/j.abb.2007.03.034
Peeters A, Shinde AB, Dirkx R, et al (2015) Mitochondria in peroxisome-deficient hepatocytes exhibit impaired respiration, depleted DNA, and PGC-1α independent proliferation. Biochim Biophys Acta - Mol Cell Res 1853:285–298. https://doi.org/10.1016/j.bbamcr.2014.11.017
Baumgart E, Vanhorebeek I, Grabenbauer M, et al (2001) Mitochondrial alterations caused by defective peroxisomal biogenesis in a mouse model for Zellweger syndrome (PEX5 knockout mouse). Am J Pathol 159:. https://doi.org/10.1016/S0002-9440(10)62534-5
Dirkx R, Vanhorebeek I, Martens K, et al (2005) Absence of peroxisomes in mouse hepatocytes causes mitochondrial and ER abnormalities. Hepatology 41:868–878. https://doi.org/10.1002/hep.20628
Keane MH, Overmars H, Wikander TM, et al (2007) Bile acid treatment alters hepatic disease and bile acid transport in peroxisome-deficient PEX2 Zellweger mice. Hepatology 45:982–997. https://doi.org/10.1002/hep.21532
Maxwell M, Bjorkman J, Nguyen T, et al (2003) Pex13 Inactivation in the Mouse Disrupts Peroxisome Biogenesis and Leads to a Zellweger Syndrome Phenotype. Mol Cell Biol 23:5947–5957. https://doi.org/10.1128/mcb.23.16.5947-5957.2003
Sargent G, Zutphen T van, Shatseva T, et al (2016) PEX2 is the E3 ubiquitin ligase required for pexophagy during starvation. J Cell Biol 214:677–690. https://doi.org/10.1083/jcb.201511034
van Zutphen T, Ciapaite J, Bloks VW, et al (2016) Malnutrition-associated liver steatosis and ATP depletion is caused by peroxisomal and mitochondrial dysfunction. J Hepatol 65:1198–1208. https://doi.org/10.1016/j.jhep.2016.05.046
Klionsky DJ, Abdelmohsen K, Abe A, et al (2016) Guidelines for the use and interpretation of assays for monitoring autophagy (3rd edition). Autophagy 12:1–222. https://doi.org/10.1080/15548627.2015.1100356
Iuso A, Repp B, Biagosch C, et al (2017) Assessing Mitochondrial Bioenergetics in Isolated Mitochondria from Various Mouse Tissues Using Seahorse XF96 Analyzer. Methods Mol Biol 1567:217–230. https://doi.org/10.1007/978-1-4939-6824-4_13
Deas E, Plun-Favreau H, Gandhi S, et al (2011) PINK1 cleavage at position A103 by the mitochondrial protease PARL. Hum Mol Genet 20:867–879. https://doi.org/10.1093/hmg/ddq526
Matsuda N, Sato S, Shiba K, et al (2010) PINK1 stabilized by mitochondrial depolarization recruits Parkin to damaged mitochondria and activates latent Parkin for mitophagy. J Cell Biol 189:211–221. https://doi.org/10.1083/jcb.200910140
Takatori S, Ito G, Iwatsubo T (2008) Cytoplasmic localization and proteasomal degradation of N-terminally cleaved form of PINK1. Neurosci Lett 430:13–17. https://doi.org/10.1016/J.NEULET.2007.10.019
Sim CH, Gabriel K, Mills RD, et al (2012) Analysis of the regulatory and catalytic domains of PTEN-induced kinase-1 (PINK1). Hum Mutat 33:1408–1422. https://doi.org/10.1002/HUMU.22127
Russell RC, Yuan H-X, Guan K-L (2014) Autophagy regulation by nutrient signaling. Cell Res 24:42–57. https://doi.org/10.1038/cr.2013.166
Toshniwal AG, Gupta S, Mandal L, Mandal S (2019) ROS Inhibits Cell Growth by Regulating 4EBP and S6K, Independent of TOR, during Development. Dev Cell 49:473-489.e9. https://doi.org/10.1016/J.DEVCEL.2019.04.008
Jornayvaz FR, Shulman GI (2010) Regulation of mitochondrial biogenesis. Essays Biochem 67:69–84. https://doi.org/10.1042/BSE0470069
Lee JM, Wagner M, Xiao R, et al (2014) Nutrient-sensing nuclear receptors coordinate autophagy. Nature 516:112–115. https://doi.org/10.1038/nature13961
Honsho M, Hiroshige T, Fujiki Y (2002) The membrane biogenesis peroxin Pex16p: Topogenesis and functional roles in peroxisomal membrane assembly. J Biol Chem 277:44513–24. https://doi.org/10.1074/jbc.M206139200
Ebberink MS, Csanyi B, Chong WK, et al (2010) Identification of an unusual variant peroxisome biogenesis disorder caused by mutations in the PEX16 gene. J Med Genet 47:608–615. https://doi.org/10.1136/jmg.2009.074302
Goldfischer S, Moore CL, Johnson AB, et al (1973) Peroxisomal and mitochondrial defects in the cerebro-hepato-renal syndrome. Science 182:62–4
Baes M, Gressens P, Baumgart E, et al (1997) A mouse model for Zellweger syndrome. Nat Genet 17:49–57. https://doi.org/10.1038/ng0997-49
Sekine S, Youle RJ (2018) PINK1 import regulation; a fine system to convey mitochondrial stress to the cytosol. BMC Biol 16:. https://doi.org/10.1186/S12915-017-0470-7
Zong H, Ren JM, Young LH, et al (2002) AMP kinase is required for mitochondrial biogenesis in skeletal muscle in response to chronic energy deprivation. Proc Natl Acad Sci U S A 99:15983–15987. https://doi.org/10.1073/pnas.252625599
Mihaylova MM, Shaw RJ (2011) The AMPK signalling pathway coordinates cell growth, autophagy and metabolism. Nat Cell Biol 13:1016–1023. https://doi.org/10.1038/ncb2329
Peeters A, Fraisl P, Van Den Berg S, et al (2011) Carbohydrate metabolism is perturbed in peroxisome-deficient hepatocytes due to mitochondrial dysfunction, AMP-activated protein kinase (AMPK) activation, and peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) suppression. J Biol Chem 286:42162–42179. https://doi.org/10.1074/jbc.M111.299727
Park H, He A, Tan M, et al (2019) Peroxisome-derived lipids regulate adipose thermogenesis by mediating cold-induced mitochondrial fission. J Clin Invest 129:694–711. https://doi.org/10.1172/JCI120606
Iershov A, Nemazanyy I, Alkhoury C, et al (2019) The class 3 PI3K coordinates autophagy and mitochondrial lipid catabolism by controlling nuclear receptor PPARα. Nat Commun 10:1566. https://doi.org/10.1038/s41467-019-09598-9
Björkhem I, Blomstrand S, Glaumann H, Strandvik B (1985) Unsuccessful attempts to induce peroxisomes in two cases of zellweger disease by treatment with clofibrate. Pediatr Res. https://doi.org/10.1203/00006450-198506000-00018
Lazarow PB, Black V, Shio H, et al (1985) Zellweger syndrome: Biochemical and morphological studies on two patients treated with cloflbrate. Pediatr Res 19:1356–1364. https://doi.org/10.1203/00006450-198512000-00030

ChiandLeeSI20230118FINAL.docx

Download PDF

Journal Publication

published 20 Jun, 2023

Read the published version in Cellular and Molecular Life Sciences →

Editorial decision: Major Revision
13 Feb, 2023
Reviewers agreed at journal
23 Jan, 2023
Reviewers invited by journal
23 Jan, 2023
Editor assigned by journal
22 Jan, 2023
First submitted to journal
20 Jan, 2023

You are reading this latest preprint version

Peroxisome loss leads to increased mitochondrial biogenesis and reduced autophagy to preserve mitochondrial function

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials And Methods

Mouse generation

Construction Of Lpd Mouse Model And Fenofibrate Treatment By Gavage

Histology And Immunofluorescence

Western Blotting

qPCR

Hepatic Triglyceride Assay

Hepatic Atp Assay

Transmission Electron Microscopy (Tem)

Central Carbon Metabolism Assay

Isolation of liver mitochondria and mitochondrial respiration

Statistics

Results

Loss of Pex16 leads to hepatic steatosis

Nutritional Stress Induced With A Low Protein Diet Exacerbates Loss Of Peroxisome-induced Hepatic Steatosis

Loss Of Peroxisomes Leads To Disturbed Hepatic Energy Homeostasis And Is Exacerbated By Nutritional Stress

Pex16 Ko Mice Show Decreased Mitochondrial β-oxidation

Hepatic Mitochondrial Dynamics Is Differentially Altered In Response To Loss Of Peroxisomes And Nutritional Stress

Loss Of Peroxisomes Exacerbates Mitochondrial Dysfunction In Response To Nutritional Stress

Pex16 Deficiency Alters Mitochondria Quality Control Related To A Decrease In Autophagy

Pex16 Ko Mice Demonstrated Increased Mitochondrial Biogenesis

Pparα Activation Improves Hepatic Mitochondrial Function Independently Of The Presence Of Peroxisomes

Pparα Activation Stimulates Mitochondrial Biogenesis And Autophagy In Pex16 Ko Mice

Discussion

Loss Of Pex16 And Peroxisomes Affect Mitochondrial Homeostasis And Function

Loss Of Peroxisomes Exacerbates Low Protein Diet Induced Mitochondrial Changes Through Separate Pathways

Fibrates Improve Mitochondrial Function In Pex16 Ko Low Protein Fed Mice

Abbreviations

Declarations

References

Supplementary Files

Status:

Journal Publication

Version 1