Protein-bound uremic toxins impaired mitochondrial dynamics and functions

Protein-bound uremic toxins, indoxyl sulfate and p-cresol sulfate, increase oxidative stress and adversely affect chronic kidney disease progression and cardiovascular complications. In this study, we examined whether mitochondria are the target of indoxyl sulfate and p-cresol sulfate intoxication in vivo and in vitro. The kidneys of 10-week-old male B-6 mice with ½-nephrectomy treated with indoxyl sulfate and p-cresol sulfate were used for the animal study. Cultured human renal tubular cells were used for the in vitro study. Our results indicated that indoxyl sulfate and p-cresol sulfate impaired aerobic and anaerobic metabolism in vivo and in vitro. Indoxyl sulfate and p-cresol sulfate caused mitochondrial fission by modulating the expression of mitochondrial fission–fusion proteins. Mitochondrial dysfunction and impaired biogenesis could be protected by treatment with antioxidants. The in vitro study also demonstrated that indoxyl sulfate and p-cresol sulfate reduced mitochondrial mass by activating autophagic machinery. In summary, our study suggests that mitochondrial injury is one of the major pathological mechanisms for uremic intoxication, which is related to chronic kidney disease and its complications.


INTRODUCTION
Impairment of mitochondrial function has been associated with numerous pathological conditions and aging [1,2]. Mitochondria and oxidative stress were important contributors in the renal inflammatory process. Deregulation of the mitochondrial respiratory mechanism has been described in patients with chronic renal disease associated with increased oxidative stress [3]. Mitochondria are highly dynamic organelles that constantly undergo fission and fusion. The balance of mitochondrial dynamics is crucial for controlling mitochondrial quality and cellular metabolism [4]. In diabetic nephropathy, high glucose levels can cause the fragmentation of mitochondria and induce the excess production of reactive oxygen species (ROS). In addition, electron transport chain defects, which result in electron leakage and superoxide radical formation, have been established in diabetic nephropathy [5,6].
Mitophagy, in which damaged mitochondria are selectively removed, is considered an adaptive response to stress that promotes cell survival under stress. Mice with www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No. 44), pp: 77722-77733 Research Paper impaired mitophagic turnover in podocytes and the tubular epithelium exhibit renal histological changes similar to human idiopathic focal segmental glomerulosclerosis [7]. Increased oxidative stress, mitochondrial fragmentation, and dysfunctional mitochondria accumulation are observed in the tubules of diabetic mice [8]. Transforming growth factor (TGF)-β is a major profibrogenic cytokine for renal cell injury in chronic kidney disease (CKD) [9]. TGF-β induces both renal cell apoptosis and renal fibrosis. TGF-β is also associated with mitochondrial dysfunction and fragmentation in multiple renal cells [10]. Mitochondria are the main organelles producing ROS, which play key roles in the initiation and modulation of cell death [11]. Disrupting mitochondria-derived ROS production may attenuate TGF-β-induced fibrosis [12].
Protein-bound uremic toxins, indoxyl sulfate (IS) and p-cresol sulfate (PCS), adversely affect CKD progression and its complications [13]. Cross talk between mitochondria and reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidases play a key role in the maintenance of the cellular redox status [14]. IS and PCS stimulate intracellular and extracellular ROS production through a pathway involving an NADPH oxidase or an NADPH-like oxidase [15][16][17]. In addition to oxidative stress, IS and PCS induce inflammatory reactions and activate the renin-angiotension-aldosterone system (RAAS), which is associated with renal fibrosis [18,19]. Inflammation causes mitochondrial stress by inducing drastic metabolic changes associated with alterations in mitochondrial dynamics that shift the balance between aerobic glycolysis and oxidative phosphorylation [20]. RAAS activation not only alters cellular metabolism but also induces mitochondrial production of ROS, which function as pathological signals [21,22]. Recent evidence has indicated that IS can impair mitochondrial function and biogenesis by increasing mitochondrial depolarization and decreasing mitochondrial mass in vitro [23]. Treatment with a uremic toxin binder can normalize citrate synthetase activity, mitochondrial biogenesis, and superoxide production in the skeletal muscle of CKD mice [24]. Here, we report that IS and PCS deregulate mitochondrial metabolism, mass, and dynamics.

Identification of deregulated mitochondrial dynamics by indoxyl sulfate and p-cresol sulfate in study mice
To identify the possible mitochondrial dysfunction induced by IS and PCS in vivo, mice with ½-nephrectomy that were treated with IS or PCS were evaluated in this study. Mitochondrial glutaminase and glutamate dehydrogenase are the key enzymes in nitrogen metabolism and the acid-base balance [25]. Our results showed that IS and PCS treatment significantly reduced glutaminase and glutamate dehydrogenase protein levels in study mice ( Figure  1A). Western blotting results also demonstrated that IS and PCS treatment significantly reduced nicotinamide nucleotide transhydrogenase (NNT) protein expression ( Figure 1A). NNT is a mitochondrial inner membrane marker that catalyzes the hydride transfer of reducing equivalent between NAD(H) and NADP(+) in a reaction coupled to proton translocation across the inner mitochondrial membrane; NNT plays a crucial role in mitochondrial antioxidant defense [26]. These results suggest that IS and PCS impair renal mitochondrial function in vivo.
To investigate the putative impact of uremic toxins on mitochondrial dynamics, Western blotting of the regulators of mitochondrial morphology, namely mitofusin1 (Mfn1), Mfn2, and mitochondrial fission 1(Fis1), was conducted. IS and PCS significantly reduced Mfn1 and Mfn2 but increased Fis1 protein expression in the kidneys of study mice ( Figure 1B). In addition, increased modified microtubule-associated protein 1 light chain 3 II (LC3-II), which is a regulator of autophagy, was noted in the kidneys of study mice treated with IS and PCS ( Figure 1C). Cisd2, another regulator of autophagy, was also downregulated by IS and PCS in vivo ( Figure 1C). Immunofleurocent staining results that the intensity of NNT decreased significantly in the glomerulus of the kidneys of mice treated with IS and PCS ( Figure 1D). Cisd2 plays key roles in maintaining mitochondrial integrity [27]. Immunohistological staining showed that Cisd2 expression was high in the renal tubules of the control mice. The staining intensity of Cisd2 decreased significantly in the renal tubules of study mice treated with IS and PCS ( Figure 1E). The positive staining for the lysosome marker lysosomalassociated membrane protein 1 (LAMP1) was also increased in the renal tubules of mice treated with IS and PCS ( Figure 1F). These results suggest that IS and PCS impair mitochondrial dynamics and induce autophagy in vivo.

Cellular metabolic changes in cells treated with indoxyl sulfate and p-cresol sulfate
To study the metabolic impact of IS and PCS on cellular metabolism, cultured HK2 cells treated with IS and PCS were analyzed. The oxygen and glucose consumption rates were calculated by detecting the decrease in oxygen and glucose concentrations in medium. Our results showed that the oxygen consumption rates increased significantly after 24h treatment with IS or PCS, but this finding was not persistently noted after 48h treatment with IS or PCS. By contrast, the significant increase in glucose consumption rates was not observed until after 48h treatment with IS or PCS (Figure 2A, 2B). Metabolite analysis demonstrated that the cellular levels of oxidized metabolites, NAD and oxidized glutathione, significantly increased after 48h treatment with IS and PCS ( Figure 2C, 2D). After 24h treatment with IS and PCS, periodic acid-Schiff (PAS) staining showed lower PAS staining intensity in cells treated with IS and PCS than in control cells ( Figure 2E). This finding indicated that glycogen storage decreased in cells treated with IS and PCS.

Indoxyl sulfate and p-cresol sulfate attenuated cellular energy pool
Flow cytometry results showed that HK2 cells treated with IS and PCS for 24h exhibited significantly increased red to green fluorescence among cells with JC-1 staining. This result indicated that HK2 cells treated with IS and PCS had higher mitochondrial membrane potential than control cells ( Figure 3A). The cytochrome complex IV activity assay was performed with isolated mitochondria. The complex IV activity of cells treated with IS and PCS at concentrations of 1 and 5 mg/L for 24h did not significantly differ from that of control cells; however, treatment with a high concentration of IS and PCS (50 mg/L) attenuated complex IV activity ( Figure  3B) However, the complex IV activity of cells treated with IS and PCS at concentrations of 1 and 5 mg/L for 48h was significantly higher than that of control cells ( Figure 3B).
IS and PCS increased glucose and oxygen consumption, mitochondrial membrane potential, and complex IV activity. These results suggest that IS and PCS stimulate aerobic respiration in residual mitochondria. Although aerobic respiration was increased in residual mitochondria, the cellular adenosine diphosphate and adenosine triphosphate levels were significantly lower than those in control cells ( Figure 3C). This result suggests that IS and PCS attenuates the cellular energy pool by increasing energy consumption.

Indoxyl sulfate and p-cresol sulfate enhanced mitochondrial fission in vitro
To study the putative effects of IS and PCS on mitochondrial dynamics, immunofluorescent staining of NNT was conducted. The results showed that IS and PCS significantly increased the globular form of mitochondria ( Figure 4A). Pretreatment with NAC could reverse the mitochondrial dynamic changes caused by IS and PCS ( Figure 4B). Western blotting results showed that IS and PCS downregulated Mfn1 and Mfn2, which regulate mitochondrial fusion, in a dose-dependent manner. By contrast, IS and PCS increased the levels of phosphorylated Drp1, which mediates mitochondrial fission ( Figure 4C). Fis1 is also a signal protein involved in mitochondrial fission. This study showed that low IS and PCS doses (1 and 5 mg/L) upregulated Fis1 levels in vitro. However, high IS and PCS doses downregulated Fis1 levels in cultured HK2 cells ( Figure 4C). Pretreatment with NAC reversed the downregulated Mfn1 and Mfn2 levels induced by IS and PCS ( Figure 4D). Costaining with phosphorylated Drp1 and NNT demonstrated that IS and PCS treatment not only increased the phosphorylated Drp1 level but also recruited the phosphorylated Drp1 to mitochondria ( Figure 4E).

Indoxyl sulfate and p-cresol sulfate induced mitophagy in cultured renal tubular cells
Autophagy was a key protective mechanism for cell survival under stress. [28] Western blotting results revealed that NNT was downregulated in HK2 cells after 24h treatment with IS and PCS ( Figure 5A). This finding suggests that IS and PCS reduces mitochondrial mass in vitro. This study also showed that LC3-II, LAMP1 and Parkin were upregulated in cells treated with IS and PCS ( Figure 5A, 5B). Parkin, an E3 ligase, mediates the selective removal of damaged mitochondria through mitophagy [29].The immunostaining results for LC3 revealed that IS and PCS treatment significantly increased LC3 puncta formation in HK2 cells ( Figure  5C). In addition, overlapping of NNT and LC3 signals increased in cells treated with IS and PCS ( Figure  5D). Western blotting with protein extracts of isolated mitochondria from cells treated with IS and PCS were performed to define the mitochondrial NNT, Parkin and LC3 expression. The results indicated that IS-and PCStreated cells had increased mitochondrial Parkin and LC3-II, but decreased mitochondrial NNT expression ( Figure 5E). To detect the mitochondrial injury, we also performed the immunofleurocent staining with the autophagy markers, LAMP1 and Rab11 with cultured renal tubular cells treated with IS and PCS. The results indicated that mitochondrial localization of LAMP1 and Rab11 increased in cells treated with the indoxyl sulfate and p-cresol sulfate (Supplementary Figure 1). These aforementioned results suggest that IS and PCS can reduce mitochondrial mass through autophagic machinery.
For IS and PCS are protein-bound uremic toxins, the in vitro study in the presence of 4% albumin were  performed. The Western blotting results showed that IS and PCS treatment could activate LC3 ( Figure 6A, 6B) and attenuate NNT level ( Figure 6C) in the culture condition with albumin. The immunostaining results also revealed that IS and PCS significantly increased the co-localization of NNT and LC3 in the presence of albumin ( Figure 6D).

DISCUSSION
The main findings of this study indicate that IS and PCS induce mitochondrial toxicity. This study found that IS and PCS not only affected aerobic respiration but also anaerobic respiration. Our study indicated that IS and PCS caused mitochondrial injury by inducing oxidative and metabolic stress. IS and PCS significantly increased the expression of mitofission proteins but reduced the expression of mitofusion proteins. In addition, IS and PCS treatment significantly increased the factors of mitophagy, activated LC3, Parkin, and LAMP1, in vitro and in vivo. This result suggests that IS and PCS induces mitochondrial dynamic changes by causing metabolic stress, which can further decrease mitochondrial mass through mitophagy.
Mitochondrial oxidative stress is a major cause of mitochondrial fragmentation. It has been demonstrated that IS and PCS induces oxidative stress by activating NADPH oxidase [16,17]. Our study showed that decreasing oxidative stress caused by IS and PCS by NAC significantly reduced the expression of the mitofission proteins Drp1 and Fis1, suppressing mitofission. A recent study in human umbilical vein endothelial cells (HUVECs) also showed similar results. In HUVECs, treatment with antioxidants, vitamin C and NAC, could prevent mitochondrial fragmentation induced by IS [23].
Mitophagy is crucial for controlling mitochondrial quality and biogenesis [30]. It has been reported that IS can reduce mitochondrial mass in vitro [23]. The current study demonstrated that IS and PCS might activate mitophagy, which selectively eliminates mitochondria. Our results indicated that IS and PCS activated autophagic mechanisms. Our study showed that autophagic machinery (LC3 puncta) colocalized with a mitochondrial marker (NNT) under the cytotoxic stress caused by IS and PCS. Parkin-mediated selective mitophagy pathway has been suggested to function as an inducible stress-response mechanism. Recruitment of Parkin to the mitochondria is critical for subsequent mitophagy [30,31]. This study found that the expression of Parkin increased as the IS and PCS concentration increased in vitro. Although mitophagy eliminates damaged mitochondria and mitochondrial biogenesis generates new mitochondria, mitochondrial mass decreased when the damage beyond cellular compensation.
Cisd2 is the causative gene for the degenerative disease of Wolfram syndrome 2 [32]. Cisd2 protein exerts an antiaging function by preventing mitochondrial degeneration [33,34]. Cisd2 depletion causes mitochondrial breakdown and dysfunction accompanied by autophagy. Cisd2 can antagonize BECN1-mediated cellular autophagy by modulating calcium homeostasis at the endoplasmic reticulum [35,36]. In this study, IS and PCS toxicity decreased Cisd2 expression in vivo. However, the protective roles of Cisd2 in the kidney injury induced by IS and PCS were still not defined.
In conclusion, IS and PCS might increase mitochondrial oxidative stress, which causes mitochondrial fission by modulating the expression of mitochondrial fission-fusion proteins and activating autophagic machinery. Decompensated mitophagy can reduce mitochondrial mass and lead to deregulated cellular metabolism. Our study suggests that mitochondrial injury is one of the major pathological mechanisms for uremic intoxication, which is related to CKD and its complications.

Study approval
All animal experiments were approved by the Institutional Animal Care and Use Committee of Chang Gung Memorial Hospital (IACUC number: 2008101703, 2012010903). Our animal center is an AAALAC certified center and this study was performed in accordance with all the relevant guidelines and regulations.

Animal model
The animal model was modified from our previous study [18]. In brief, 10-week-old male B-6 mice with ½-nephrectomy were used in this study. The study animals were divided into control and experimental groups. The control mice (n = 8) received daily phosphate-buffered saline (PBS) injections at the same volume as the experimental mice for 1 week. The experimental mice received intraperitoneal injection with IS (Sigma-Aldrich, St. Louis, MO) (n = 8) or PCS (Kureha Corporation, Tokyo, Japan) (n = 8) at a dose of 100 mg/kg/day for 1 week. The kidneys of study animal were harvested for following study.

Cell culture
Human renal tubular cells (HK2) were cultured in the DMEM based medium as described previously.
HK2 cell cultures at approximately 70% confluence were synchronized under serum-free conditions for 48 h. These cells were then treated with IS or PCS at concentrations of 0, 1, 5, and 50 mg/L under the serum-free condition in the presence or absence of 4 % albumin (Sigma-Aldrich). For the antioxidation study, the HK2 cells were pretreated with N-acetylcysteine (NAC) (Sigma-Aldrich) for 2 h before IS and PCS treatment under the serum-free condition. The concentrations of IS, PCS, and NAC and the treatment duration are illustrated in figure legends.

Western blotting
Total protein was extracted using a commercial kit according to the manufacturer's instructions (Protein Extraction Kit, Millipore, Billerica, MA). Thirty micrograms of protein from each sample was mixed with sample loading buffer and loaded onto separate lanes on a 12%sodium dodecyl sulfate-polyacrylamide gel. Proteins were electrotransferred onto polyvinylidene fluoride membranes (0.2 μm: Immun-Blot, Bio-Rad, Hercules, CA) and then immunoblotted with primary antibodies. The intensity of each band was quantified using NIH Image software (Bethesda, MD), and the densitometric intensity corresponding to each band was normalized against β-actin expression.

Immunostaining and periodic acid-Schiff stain
Paraffin tissue sections were cut, mounted, deparaffinized, rehydrated, and stained with hematoxylin-eosin by using standard histological techniques. For immunohistochemical staining, the Ventana Benchmark automated staining system and Ventana reagents were used (Ventana Medical Systems, Tucson, AZ), and primary antibodies. The tissue sections were observed using light microscopy (Nikon Eclipse Ti, Tokyo, Japan). For immunofluorescence staining, cells were incubated with a primary antibody, followed by incubation with a fluorescent secondary antibody. The sections were counterstained with 4′,6-diamidino-2phenylindole (1:500 dilution; Sigma-Aldrich) to identify cellular nuclei. The stained samples were observed under a confocal microscope (Leica Microsystems, Bannockburn Ill, Wetzlar, Germany). The mitochondrial morphology was analyzed by MicroP software with the default setting [37]. The antibodies used for Western blot and immunostaining were listed in Supplementary Table  1. For detecting cellular glycogen storage, HK2 cells were stained with a PAS staining system (Sigma-Aldrich) according to the product instructions. The cells were fixed in formalin-ethanol fixative solution, followed by staining with periodic acid solution and Schiff's reagent. The air dried slides were examined microscopically (400×) (Nikon Eclipse Ti).

Mitochondrial isolation and complex IV enzyme activity
Mitochondria were isolated from HK-2 cells by using the Qproteome mitochondrial isolation kit (Qiagen, Venlo, Netherlands) according to the standard protocol. In brief, washed cells were first suspended in lysis buffer; subsequently, the cells were centrifuged to obtain a pellet consisting of compartmentalized organelles. The resulting pellet was resuspended in disruption buffer, repeatedly passed through a narrowgauge needle, and centrifuged to obtain a pellet consisting of nuclei, cell debris, and unbroken cells. The supernatant, which contained mitochondria and the microsomal fraction, was re-centrifuged to obtain a pellet consisting of mitochondria. After removal of the supernatant, mitochondria were washed and resuspended in mitochondria storage buffer.
Freshly isolated mitochondria were used for assaying complex IV enzyme activity. The concentrations of mitochondrial proteins were determined using the Pierce TM BCA protein assay (Pierce Biotechnology, Waltham, MA). Freshly isolated mitochondria were maintained on ice until use in subsequent assays. Mitochondrial complex IV enzyme activity was determined using a mitochondrial complex IV enzyme activity assay kit (Abcam, Cambridge, MA) according to the manufacturer's protocol. The sample at a concentration of 5.0 mg/mL was analyzed. Complex IV was immunocaptured within the wells, and its activity was determined calorimetrically at OD 550 nm after the oxidation of reduced cytochrome C. To determine the activity in the sample, the slope was calculated using the formula: Rate = (Absorbance 1 − Absorbance 2)/Time (min). The relative complex IV activity was obtained by comparing the sample rate with the rate of the control (normal) sample and with the rate of the null (background). In this study, each reaction was repeated in triplet. The relative ratios versus control were plotted.

Oxygen consumption
The oxygen consumption of HK2 cells was determined using a Clark-type polarographic electrode (Mitocell Respirometery SystemMT200/MT200A, Strathkelvin Instruments, North Lanarkshire, Scotland) according to the manufacturer's protocol. After treatment with IS and PCS, each sample (5×10 6 cells) was analyzed during incubation in a magnetically stirred chamber over a period of 5 min at constant temperature (37 °C). The signals were detected and analyzed using built-in software from Strathkelvin Instruments. The rate of oxygen consumption was normalized to the number of living cells, which was determined using by the Cellometer Vision CBA image cytometer (Nexcelon, Lawrence, MA).Three wells were utilized per condition in any given experiment.

Mitochondrial membrane potential
The mitochondrial membrane potential was examined using the JC-1mitochondrial potential sensor (Invitrogen, Carlsbad, CA). After treatment with IS and PCS, cells were incubated with the JC-1 dye (10 μg/mL) at 37°C for 15 min, followed by analysis with the Gallios flow cytometer (Beckman Coulter, Brea, CA) for quantifying 488-nmexcited fluorescence signals at 585/42 nm (FL2; red) and 525/50 nm (FL1; green). JC-1 monomers emit at 530 ±15 nm (FL1 channel), and J-aggregates emit at 590 ± 17.5 nm (FL2 channel). Cytometry settings were optimized for green (FL1) and red (FL2) fluorescence, and the data were analyzed with the Kaluza Flow Cytometry analysis software V1.2 (Beckman Coulter). The relative red to green fluorescence ratio of cells was calculated.

Glucose uptake and metabolite analysis
The glucose concentration of the culture medium was detected using a glucose meter (Abbott, Chicago, IL). Glucose uptake was calculated by detecting the decrease in the glucose concentrations of the culture medium after IS and PCS treatment. Sample processing and metabolite analysis were performed as previously described [38,39].
For extracting hydrophilic metabolites, the cells were washed twice with PBS and scraped on ice in 1mL of 80% methanol. The cell lysate was centrifuged at 14,000×gfor 15min at 4°C, and the supernatant was collected for rapid resolution liquid chromatographytime-of-flight-mass spectrometry analysis. A modified Folch's method was employed for extracting lipids. Liquid chromatographic separation for processed cell samples was achieved on a 100× 2.1 mm Acquity 1.7μm C8 column (Waters Corp., Milford, MA) by using an ACQUITY TM UPLC system (Waters Corp.). The eluent was analyzed using high-definition, time-of-flight mass spectrometry (SYNAPT G1, Waters Corp.) operating in the electrospray ionization-positive ion mode. Raw mass spectrometric data were processed using Mass Lynx V4.1 and Marker Lynx software (Waters Corp.). The multivariate data matrix was analyzed using SIMCA-P software (version 13.0, Umetrics AB, Umea, Sweden).

Statistical analyses
All data are expressed as mean ± standard error. One-way analysis of variance with Bonferroni corrections was performed for analyzing the data of the cell culture study. The data of the different animal groups were compared using the Wilcoxon-Mann-Whitney test. P-values of <0.05 were considered statistically significant.

Abbreviations
CKD, chronic kidney disease; ROS, reactive oxygen species; TGF, transforming growth factor; PCS, p-cresol sulfate; NADPH, nicotinamide adenine dinucleotide phosphate hydrogen; RAAS, renin-angiotensionaldosterone system; DNA, deoxyribonucleic acid; Cisd2, CDGSH iron sulfur domain 2; NNT, nicotinamide nucleotide transhydrogenase; Mfn, mitofusin; LAMP1, lysosomal-associated membrane protein 1; NAC, Nacetylcysteine; PAS, periodic acid-Schiff; HUVECs, human umbilical vein endothelial cells; PBS, phosphatebuffered saline Author contributions CYS, MLC, and HCP contributed to the conception, design, and interpretation of data. CYS, MLC, and JHL performed the experiments and drafted the manuscript. CCL, HCP, and staff of the Community Medicine Research Center provided intellectual content for data analyzation, results interpretation, figures preparation and were involved in editing and revising the manuscript. All authors discussed, contributed to, and approved the final version of the manuscript.