Introduction
Acute kidney injury (AKI) is an important clinical syndrome with increased short- and long-term morbidity and mortality, which is often caused by ischemia–reperfusion (IR), nephrotoxic drugs, sepsis and urinary tract obstruction [
1]. At present, effective therapies for AKI are limited, partly due to incompletely understand of the pathogenesis and molecular mechanism of renal damage. Therefore, it is urgent to find new targets for the intervention to inhibit its progression.
The proximal renal tubules are rich in mitochondria which generate sufficient quantities of ATP through the process of oxidative phosphorylation (OXPHOS) to maintain mitochondrial homeostasis, which is essential for normal kidney function [
2]. Mitochondria are the major intracellular sources which are extremely vulnerable to damage during different cellular energetic conditions. Mounting evidences have demonstrated that mitochondrial function is vital for AKI and mitochondrial biogenesis plays a beneficial role in kidney injury and repair the function of kidney after AKI [
3,
4]. It has been reported that the amounts of mitochondria were greatly reduced in rat kidney after I/R accompanied with mitochondria structural changes [
5]. Besides, several clinical trials show that the kidney cortical mtDNA copy number is decreased and urinary mitochondrial DNA (UmtDNA) levels which correlated with renal injury in sepsis-induced AKI patients are considered to be a valuable biomarker for the occurrence of AKI [
6]. As an important regulation mode of mitochondrial homeostasis, the decline of mitochondrial biogenesis inevitably leads to the reduction of mitochondrial quantity and inhibition of mitochondrial respiration, which may further induce mitochondrial damage and lead to renal intrinsic cell death, kidney injury and possible organ failure [
7]. Therefore, seeking intervening measures for promoting mitochondrial biogenesis is a significant way to prevent AKI.
Mitochondrial transcription factor A (TFAM) is the most abundant mtDNA packaging protein that is transported from the cell nucleus to mitochondria and is required for mtDNA maintenance, transcription, and replication [
8]. Whole deletion of TFAM is embryonically lethal, but tissue-specific lack of TFAM disrupts respiratory chain function and generates a variety of alterations that recapitulate important phenotypes of human mitochondrial diseases [
9,
10]. Several of posttranslational modifications of TFAM have been reported. Ubiquitination and phosphorylation of TFAM impaired the transcriptional activity [
11,
12]. Lysine acetylation of TFAM is also shown by Graeme A King and colleagues [
13]. However, which is responsible for the acetylation of TFAM is unclear.
General control of amino acid synthesis 5 like-1 (GCN5L1), a recently identified acetyltransferase which sequence homology to the nuclear acetyltransferase GCN5. GCN5L1 first aroused interest in understanding its role in lysosomes biogenesis and endosome–lysosome trafficking in the cytoplasm [
14]. Subsequent studies have explored that GCN5L1 could also be located in mitochondria and modulate mitochondrial protein acetylation to regulate multiple mitochondrial biological functions. Moreover, GCN5L1 could also be used as a cell energy sensor to mediate a variety of cell biological reactions including mitochondrial metabolism by responding to different energy states of cells [
15]. Our previous study found that mitochondrial GCN5L1 drives mitochondrial ROS and fatty acid oxidation [
16,
17]. Herein, we further explore the critical role of GCN5L1 in mitochondrial biogenesis in AKI. We first reported that GCN5L1 was significantly elevated both in vivo in human and mice AKI kidney and in vitro renal tubular epithelial cells (TECs) treated with hypoxia/reoxygenation, while reducing GCN5L1 expression could effectively attenuate kidney damage in AKI. Furthermore, GCN5L1 could acetylate TFAM at its K76 site, and then impair its binding with the translocase of outer mitochondrial membrane 70 (TOMM70), which affected the import of TFAM into mitochondrial and inhibited its DNA-binding capabilities, thereby impairing mitochondrial biogenesis. These results suggested that GCN5L1-mediated TFAM acetylation might be a key regulator and energetic sensor of mitochondrial biogenesis.
Methods
Human subjects
Tissue sections were obtained from patients with acute kidney injury confirmed by renal biopsy from department of Pathology, Cheeloo College of Medicine, Shandong University. Control samples were collected from normal kidney tissues of patients undergoing nephrectomy of renal carcinoma without other renal diseases. All procedures were approved by the institutional review committee of Cheeloo College of Medicine, Shandong University (ECSBMSSDU2018-1-045) after written informed consent obtained from the patients.
Animal experiments
C57BL/6J male mice (6–8 weeks old, 6 mice in each group) were pursued from Shandong University Experimental Animal Centre and were housed under standard conditions and cared for according to the institutional guidelines for animal care. All animal experiments were approved by Institutional Animal Care and Ethics Committee of Shandong University (No. S077).
Intrarenal adeno-associated virus delivery
AAV2/9-HU6-shGCN5L1 (titer: 1 × 1012 vg/ml) and AAV2/9-HU6-Scramble (titer: 1 × 1012 vg/ml) were purchased from Genomeditech (Shanghai, China). The target sequence of shGCN5L1 was 5′-GAAGAGGAGGAGAGAAGCTAT-3′. The sequence of negative control was 5′-TTCTCCGAACGTGTCACGT-3′. Mice were anesthetized by intraperitoneal injection of 1% pentobarbital sodium (35–40 mg/kg), then the kidneys were exposed via a back incision. After diluted in 30 μl sterile 0.9% sodium chloride solution, AAV9-HU6-shGCN5L1 or AAV9-HU6-Scramble (100 μl/kidney) was injected into the kidney parenchyma in four or five sites using a 30G needle.
Mice models
Mice were treated with renal ischemia 1 month after AAV injection. 1% pentobarbital sodium (35–40 mg/kg) was used to intraperitoneal anesthesia, and mice were then put in a prone position to fix on a 37 °C heating plate. Acute kidney injury was induced by bilateral renal pedicles clamping for 30 min. Sham mice underwent surgery without renal pedicle clamp. Mice were sacrificed at 48 h after reperfusion, and the kidney tissues, blood, and urine were collected for further analysis.
Renal function measurement
Blood samples were centrifuged at 3000 rpm for 5 min at 4 °C to collect serum. Serum creatinine and urea nitrogen were measured by commercial kits based on manufacturer’s instructions respectively (Nanjing Jiancheng Bioengineering Institute, C011-2-1, C013-2-1).
Human renal tubular epithelial cell line
Human renal proximal tubular epithelial cells (TECs) were purchased from the American Type Culture Collection (ATCC, Manassas, VA) and cultured in low glucose Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% FBS and 1% penicillin–streptomycin at 37 °C in a 5% CO
2 incubator. For the H/R model, TECs were firstly cultured in medium without nutrients (glucose-free, serum-free) for 12 h in a humidified hypoxic incubator with 1% O2, 5% CO
2 and 94% N2. After hypoxic treatment, cells were transferred into a normal incubator (5% CO
2 and 95% air) with regular culture medium and cultured for 12 h [
18]. Control cells were incubated in complete culture medium in a regular incubator.
Transfection
Cells were plated at a density of 6 × 10
4 cells/well in a 6-well plate. After 12–24 h, cells were transfected with specific siRNAs targeting GCN5L1 or TFAM according to the manufacturer’s protocol of RiboBio (Guangzhou, CN). Lipofectamine 3000 (Invitrogen) was used to transfect the GCN5L1 or TFAM overexpression plasmid when the cells reached 70–80% confluence. pDsRed2-Mito was applied for red fluorescence labeling of mitochondria which could then be transfected into cells using Lipofectamine 3000 (Invitrogen), as described previously [
17].
Western blot analyses
Proteins from lysed cells were fractionated by 10% or 12% SDS-PAGE and transferred to 0.45 μm PVDF nitrocellulose membrane. Then nonspecific binding sites were blocked with 5% skim milk in TBST for 2 h at room temperature (Millipore, USA). The membranes were incubated with primary antibodies overnight at 4 °C (Antibodies were described in Additional file
1: Table S1). The next day, all membranes were washed three times with TBST and incubated with an HRP-conjugated secondary antibody. Finally, proteins bands were visualized by Western Chemiluminescence (Millipore, USA) on a Storm 860 imager (Molecular Devices).
Development of anti-TFAM (acetyl K76) antibody
The polyclonal antibody specific for the acetylated TFAM at K76 (anti-acetylated K76-TFAM) was produced in Chinapeptides company. Rabbits were immunized with the acetylated human TFAM peptide (QNPDAK(Ac)TTELIRC) where aceK76 represents the acetylated K76. Antisera from the immunized rabbits were first depleted with the unacetylated peptide (QNPDAKTTELIRC) and then affinity-purified using the acetylated peptide.
Quantitative real-time PCR
Total RNA was extracted using Trizol Reagent (Invitrogen). cDNA was reversed using PrimeScript™ RT reagent Kit (TAKARA) according to the manufacturer’s instructions. Quantitative real-time PCR was performed with the TB Green Premix Ex Taq™ II (TAKARA, No. RR820L) on SYBR Green I/HRM Dye PCR System (Roche 480II). Expression levels of the detected genes were normalized to β-actin expression levels (Primers are listed in Additional file
1: Table S2).
Chromatin immunoprecipitation assay (ChIP)
Cells were cross-linked with 1% formaldehyde at 37 °C for 10 min. Glycine was added to stop the reaction. Then, cells were resuspended in lysis buffer containing protease inhibitors. The chromatin was disintegrated into small fragments by sonication and then incubated with antibody against TFAM and Protein A/G Magnetic beads at 4 °C overnight. The next day, magnetic beads were separated using magnetic frame. Immunoprecipitated DNA was retrieved from the beads with elution buffer. Expression of the mitochondrial DNA was quantified by qRT-PCR using specific primers (F-AAGAACCCTAACACCAGCCTAAC; R-AAGAACCCTAACACCAGCCTAAC’). Rabbit IgG was used as the negative antibody control.
Immunoprecipitation and co-immunoprecipitation
The immunoprecipitation and co-Immunoprecipitation were performed using Crosslink IP kit (Thermo Scientific Pierce, 26147) and Co-IP kit (Thermo Scientific Pierce, 26149) according to the manufacturer’s instructions respectively. For the immunoprecipitation assay, the cells were lysed with lysis buffer and cell lysates were incubated with 10 μg acetylation antibody and cross-linked resin on a rotator overnight at 4 °C with mouse-IgG antibody as a negative control. The next day, protein complexes bound to the antibody were eluted by elution buffer and subjected to western blot assay using mouse TFAM antibody. For the co-immunoprecipitation assay, 10 μg TFAM antibody was immobilized with Amino Link Plus coupling resin for 2 h and then incubated with cell lysates at 4 °C overnight. After incubation, the resin was washed with elution buffer and eluted proteins were analyzed by western blot assay using rabbit HSP70 antibody, rabbit TOM70 antibody rabbit, TOM40 antibody, rabbit TOM20 antibody, rabbit TIMM44 antibody and rabbit TIMM17A antibody.
Histopathology and immunohistochemistry
Kidney tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 5 μm intervals. After being deparaffinized and rehydrated, selected sections for H/E staining were stained with hematoxylin and eosin. For immunohistochemistry, sections were placed in sodium citrate buffer to retrieval antigens. Then slides were incubated with the primary antibodies diluted in PBS at 4 °C overnight. After three 5 min washes with PBS, sections were incubated with HRP-conjugated secondary antibody at 37 °C for 1 h and then visualized with diaminobenzidine, counterstained with hematoxylin, dehydrated, and sealed with neutral gum. Finally, images were obtained under a Nikon microscope imaging system (Nikon, Tokyo, Japan).
Immunofluorescence staining
After washing with PBS, the cells were fixed in 4% paraformaldehyde for 15 min at room temperature and permeabilized with 0.1% Triton X-100 diluent in PBS for another 15 min at room temperature. Then, 1% goat serum (Solarbio, SL038) was used to block nonspecific sites for 1 h at room temperature followed by overnight incubation with primary antibodies diluted in 1% goat serum at 4 °C. Antibody staining was visualized with Alexa 594 goat anti-rabbit or Alexa Fluor 488 goat anti-mouse. Finally, DAPI (Solarbio, C0060) was added to stain the cell nuclei. Images were obtained under a Nikon microscope (Nikon, Tokyo, Japan) or LSM 700 Laser scanning confocal microscope (Zeiss, Germany).
Mitochondrial DNA quantification
Total DNA was harvested from treated cells or mice kidneys using FastPure Cell/Tissue DNA Isolation Mini Kit (Vazyme, DC102-01). The mtDNA copy number was expressed as the mitochondrial DNA (mtDNA) /nuclear DNA (nDNA) ratio. Mitochondrial gene D-Loop 2 (mtDNA), mitochondrial gene cytochrome c oxidase I (mtDNA), nuclear gene G6PC (nDNA) and nuclear gene β-actin (nDNA) were determined by quantitative real-time PCR-based method as described previously [
17]. Primers are listed in Additional file
1: Table S2.
Transmission electron microscopy
Treated cells or mice kidney were gently collected and fixed with ice-cold 2.5% glutaraldehyde and 1% osmium tetroxide. After being dehydrated and embedded, samples were sliced into ultrathin sections, which were then evaluated under JEM-100sX electron microscopy to observe mitochondrial abundance and morphology. The number of mitochondria was estimated per square micron of the field was estimated using Image-Pro plus 6.0.
Protein-binding site prediction
The experimental structure of TFAM was directly obtained from RCSB PDB database (
http://www.rcsb.org/) with the accession of 3TMM. The protein structure of GCN5L1, which was predicted by AlphaFold, a highly accurate artificial intelligence-based computational structure modeling method, was retrieved from the UniProt database (
http://www.uniprot.org/). Then the structures of these proteins were submitted to the PRISM tool (
http://cosbi.ku.edu.tr/prism) to predict their potential interaction interface. Finally, the prediction results were visualized by the PyMol tool (
http://pymol.org).
Measurement of bioenergetic profile
The treated cells were plated in XF96 cell culture microplates (Seahorse Bioscience). XF96 extracellular flux analyzer (Seahorse Biosciences, North Billerica, MA, USA) were employed to measure the oxygen consumption rate of cells. After measurement of basal OCR, cells were treated with oligomycin (2 μM), FCCP (1 μM), and rotenone/antimycin A (5 μM) to generate a bioenergetic profile, as described previously [
19].
Measurement of oxygen consumption
Mitochondrial oxygen consumption of mice renal cortex was measured by Clark-type oxygen electrode (Hansatech Instruments, UK). Mitochondria were isolated from freshly kidney tissues and suspended in respiration buffer to obtain a final concentration of 0.5 mg/ml. The respiratory substrates and inhibitors were added to detect oxygen consumption of the respiratory chain complexes including Complex I (5 mM Glutamate plus 5 mM Malate, 2 mM Rotenone), Complex II + III (5 mM Succinate, 0.1 mM Antimycin A) and Complex IV (1.2 mM TMPD, 6 mM KCN).
Proximity ligation assay
Cells were grown on glass slides in 24-well plates. Then coverslips were fixed in 4% paraformaldehyde and permeabilized with 0.1% Triton X-100. The interaction of TFAM with acetyl-lysine, HSP70, TOM70, TOM40, TOM20, TIMM44 and TIMM17A was detected by using Duolink™ In Situ proximity ligation assay kit (Duolink, Sigma) to detected as described by the manufacturer. Green fluorescence reactions in cells were captured by Nikon microscope (Nikon, Tokyo, Japan) or TCS SP8 Laser scanning confocal microscope (Leica, Germany) and analyzed by Image J.
Statistical analysis
All the statistical analyses were performed using GraphPad Prism 7 software. Two-tailed unpaired Student’s t-test was used to detect the difference between the two groups. Multiple groups were compared using one-way ANOVA test. P < 0.05 was considered to be significant.
Discussion
Acetylation has emerged as the most prevalent PTM manner for mitochondrial function regulation [
22] And evaluated about 35% mitochondrial proteins contain acetylation sites, and these proteins cover almost every aspect of mitochondrial biology, such as TCA cycle, oxidative phosphorylation, fatty acid oxidation, amino acid metabolism, carbohydrate metabolism, nucleotide metabolism and the urea cycle [
23]. Compared with the well elucidation of mitochondrial deacetylase enzymes including SIRT3, SIRT4 and SIRT5, mitochondrial specific acetyltransferase was not identified and mitochondrial protein acetylation was supposed to be a nonenzymatic process, until the recent identification that GCN5L1 could interact with and acetylate several lines of mitochondrial proteins [
24]. As a 15 kDa protein with sequence homologous to the nuclear acetyltransferase GCN5, GCN5L1 has a cellular distribution of both cytosol and mitochondria [
25]. Cytoplasmic GCN5L1 mainly participates in the process of endosome–lysosome biogenesis, while mitochondria-localized GCN5L1 was evidenced to be capable of modulating multiple mitochondrial functions, such as fatty acid oxidation, gluconeogenesis and mitophagy [
26‐
28]. A recent study also suggested that GCN5L1 might have impacts on mitochondrial biogenesis. In cultured mouse embryonic fibroblasts cells, Scott I and colleagues demonstrated that GCN5L1 deletion could induce the expression of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), and accordingly initiate mitochondrial biogenesis, while the concrete intermediate mechanism underlying the effects of GCN5L1 on PGC-1α remains to be determined [
29]. Our results here also support the notion that GCN5L1 could negatively control mitochondrial biogenesis, as GCN5L1 deletion led to a significant elevation in mtDNA copies and cellular mitochondrial abundance, along with enhanced OXPHOS capability, while GCN5L1 overexpression exerted opposite effects. Furthermore, our present study demonstrated a direct mechanism accounting for the effects of GCN5L1 on mitochondrial biogenesis, as is GCN5L1 could acetylate TFAM at its K76 site, thereby inhibiting its binding to TOM70 and subsequent transporting into mitochondria, consequently reducing its mitochondrial accumulation (Fig.
4). These findings not only supplemented the current knowledge about the biology of GCN5L1, but also provided a novel modulating mechanism for intracellular transport of TFAM as well.
The acetylating modification of TFAM has been reported previously, while the detailed functional elucidation of this modulation manner is only beginning to emerge [
30]. A previous study reported that TFAM could be acetylated at its K154 site within mitochondria, and this modulation could reduce its binding affinity to mtDNA, accordingly inhibiting its biological effects on mitochondrial genome [
31]. In the present study, we demonstrated that GCN5L1-induced TFAM acetylation acted at another site, K76 site (Fig.
2). Furthermore, this modification occurred in the cytoplasm while not within mitochondria, and mainly affected the intracellular transport machinery of TFAM. Thus, it seemed that acetylating modification might be a crucial regulating manner for TFAM’s biology and acetylating at different TFAM molecular sites might exert different effects. Besides, the identification of GCN5L1 as a specific acetyltransferase for TFAM K76 site acetylation indicated that this modulation manner of the intracellular transport of TFAM is not a passive nonenzymatic process secondary to fluctuations of cellular acetyl-coenzyme A contents, but an enzyme-controlled active one. In addition, it should be mentioned that via using bioinformatics analysis, we screened a total of seven potential acetylating sites within TFAM. Besides K154 and K76, whether the remaining five sites could also be acetylated, along with their separate biological effects, are of interests deserving further exploration.
As a nuclear DNA-encoded protein, TFAM should be transported from the cytoplasm into mitochondria to execute its biological functions [
32]. However, the intracellular transporting machinery for TFAM is of little known so far. A recent study reported that HSP70 might act as its chaperone facilitating its translocation to mitochondria [
32]. We employed bioinformatics to search the potential interacting proteins of TFAM and screened a total of six candidate transporting proteins, including the previously reported HSP70, translocase of the outer membrane TOM70, TOM40, TOM20, and translocase of the inner membrane TIMM44 and TIMM17A. Our following co-immunoprecipitation studies revealed all these 6 proteins could bind to TFAM, thus outlining the intracellular transport machinery of TFAM (Fig.
4). Besides, the results showed that GCN5L1 obviously impaired the binding of TFAM to TOM70 without affecting its binding to HSP70, TOM40, TOM20, TIMM44 and TIMM17A, indicating TFAM K76 site acetylation might selectively affect the process of TFAM translocating through OMM. These results also coincided with the findings above that TFAM K76 site acetylation occurred in the cytosol and was mediated by cytoplasmic GCN5L1, although GCN5L1 is also known to have a mitochondrial distribution.
For cellular homeostasis maintenance, mitochondrial biogenesis must be finely tuned to match the different cellular energetic states. The process of mitochondrial biogenesis requires the expression of ample genes from both nuclear and mitochondrial genome, and tightly controlled by coordination between the two genomes as well [
33]. Due to its nuclear DNA-origin property and multifaceted effects on mtDNA, TFAM was proposed as a major player in the adaptive process of mitochondrial biogenesis to environmental stresses. However, the concrete energetic-sensing mechanism of TFAM has not been fully elucidated. Current notion suggested that nuclear respiratory factor 1(Nrf1) and PGC-1α might act as the major regulators for TFAM expression and be involved in the energetic stresses-induced response of mitochondrial biogenesis [
34]. Our results indicated that H/R-induced reduction of mitochondrial abundance was accompanied by a significant elevation of GCN5L1 expression and acetylated TFAM levels, and a concurrent decrease of mitochondrial TFAM contents, thereby suggesting that GCN5L1-mediated TFAM acetylation and TFAM trafficking modulation might be a novel sensing mechanism of mitochondrial biogenesis. Furthermore, such findings also coincided with the emerging role of GCN5L1 as the energetic sensor in modulating mitochondrial glucose and fatty acid oxidation, in response to nutrients challenges including both scarce and oversupply states of nutrients [
16,
27].
The results that both in vitro hypoxia/reoxygenation and in vivo ischemia/reperfusion led to alterations of GCN5L1 expression indicated GCN5L1 might be one active player underlying the pathogenesis of ischemic AKI, and a potential intervening target as well. AKI is one of the most common kidney pathologies leading to high morbidity and mortality, with mitochondrial dysfunction serving as the major fundamental pathogenic event [
35]. Experimental studies have revealed that five to ten minutes of ischemic attack could lead to significant inhibition of cellular respiration of TECs, with subsequent cellular ATP exhaustion and apoptotic or necrotic cell death [
36]. On the other hand, rapid restoration of mitochondrial respiration, for example via promoting mitochondrial biogenesis, was proposed as the key intervening method for both preventing the onset of AKI and facilitating its recovery [
37]. Currently, PGC-1α, a co-transcriptional regulation factor capable of inducing mitochondrial biogenesis by activating TFAM expression, is proposed as the one pharmacologic target, and several lines of agents aiming at PGC-1α including PPAR-γ agonist, β2-adrenergic receptor agonist and 5-HT2 agonists are under clinical evaluation or development for treating AKI [
38]. Our results that kidney-specific knockdown of GCN5L1 was effective in ameliorating bilateral renal pedicle clamping induced kidney morphological and functional injuries, suggested that reducing GCN5L1 could serve as another intervening target, and agents specific acting on GCN5L1 deserve further exploration. Besides, the intermediate mechanisms responsible for I/R-induced GCN5L1 expression are also of significance for future intervening AKI.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.