BASIC RESEARCH www.jasn.org

Disruption of Renal Tubular Mitochondrial Quality Control by Myo-Inositol Oxygenase in Diabetic Kidney Disease

Ming Zhan, Irtaza M. Usman, Lin Sun, and Yashpal S. Kanwar

Departments of Pathology and Medicine, Northwestern University, Chicago, Illinois

ABSTRACT Diabetic kidney disease (DKD) is associated with oxidative stress and mitochondrial injury. Myo-inositol oxy- genase (MIOX), a tubular-specific enzyme, modulates redox imbalance and in tubular cells in di- abetes, but these mechanisms remain unclear. We investigated the role of MIOX in perturbation of mitochondrial quality control, including mitochondrial dynamics and /mitophagy, under high- glucose (HG) ambience or a diabetic state. HK-2 or LLC-PK1 cells subjected to HG exhibited an upregu- lation of MIOX accompanied by mitochondrial fragmentation and depolarization, inhibition of autophagy/ mitophagy, and altered expression of mitochondrial dynamic and mitophagic proteins. Furthermore, dys- functional mitochondria accumulated in the cytoplasm, which coincided with increased reactive oxygen species generation, Bax activation, cytochrome C release, and apoptosis. Overexpression of MIOX in LLC-PK1 cells enhanced the effects of HG, whereas MIOX siRNA or D-glucarate, an inhibitor of MIOX, partially reversed these perturbations. Moreover, decreasing the expression of MIOX under HG ambience increased PTEN-induced putative kinase 1 expression and the dependent mitofusin-2– interaction. In tubules of diabetic mice, increased MIOX expression and mitochondrial fragmentation and defective autophagy were observed. Dietary supplementation of D-glucarate in diabetic mice decreased MIOX ex- pression, attenuated tubular damage, and improved renal functions. Notably, D-glucarate administration also partially attenuated mitochondrial fragmentation, oxidative stress, and apoptosis and restored autoph- agy/mitophagy in the tubular cells of these mice. These results suggest a novel mechanism linking MIOX to impaired mitochondrial quality control during tubular injury in the pathogenesis of DKD and suggest D-glucarate as a potential therapeutic agent for the amelioration of DKD.

J Am Soc Nephrol 26: 1304–1321, 2015. doi: 10.1681/ASN.2014050457

The incidence of diabetes-related complications has is regarded as secondary to the changes in the glo- declined in the United States in recent years. merular compartment, although at times, it sub- However, among them, ESRD resulting from di- stantially contributes to the acceleration of disease abetic kidney disease (DKD) has the smallest re- progression and deterioration of renal functions.3,4 duction, and it remains a heavy burden on the health In fact, it is well described in the literature that the care system.1 DKD is a microvascular complication, in which hyperglycemia induces dysfunction and at times, irreversible injury in various cell types of the Received May 9, 2014. Accepted July 23, 2014. kidney. Typically, one sees accumulation of Published online ahead of print. Publication date available at extracellular matrix in glomerular and tubulointer- www.jasn.org. stitial compartments with thickening and hyalin- Correspondence: Dr. Yashpal S. Kanwar, Departments of Pa- 2 ization of intrarenal vasculature. Among these thology and Medicine–Nephrology, FSM, Northwestern Univer- changes, tubular injury, including tubular hyper- sity, 300 East Superior Street, Room Tarry 12-713, Chicago, IL trophy and thickening of the tubular basement 60611. Email: [email protected] membrane, is readily seen. Usually, tubular injury Copyright © 2015 by the American Society of Nephrology

1304 ISSN : 1046-6673/2606-1304 JAmSocNephrol26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH extent of tubulointerstitial injury correlates with the degree of while, inhibits autophagic removal of damaged mitochondria. compromise in renal functions.5,6 Although the pathogenesis The surveillance of quality control was noted to be disrupted, of DKD still remains unclear, recent studies in animals and and accumulated dysfunctional organelles generated excessive humans have led to an emerging concept that mitochondrial ROS and initiated apoptotic cascade, leading to tubular injury. dysfunctions, which are pivotal events, lead to activation of Importantly, inhibition of MIOX by D-glucarate normalized various cellular processes and aberrant signaling in different regulation of the mitochondrial lifecycle, and thus, it could cell types of the kidney.7–9 serve as a potential therapeutic agent for the amelioration of Mitochondria are dynamic organelles that constantly un- tubulopathy in DKD. dergo fusion and fission cycles to maintain a balance of cellular reticular network and mitochondrial turnover.10 Their dy- namics are regulated by profission proteins (Drp1 and Fis1) RESULTS and profusion mediators (Mfn1/2 and OPA1). During this process, the stress-induced dysfunctional mitochondria un- Alterations in Mitochondrial Dynamics, Autophagy, dergo fragmentation and membrane depolarization, and ROS Production, and Apoptosis in Renal Proximal then, they are engulfed and degraded by the autophago- Tubular Cells Subjected to High-Glucose Ambience somes—a process known as mitophagy.11 In the mitochon- Confocal microscopy delineated mitochondrial morphology drial lifecycle, mitochondrial fusion, mitochondrial fission, in HK2 cells after Mitotracker red staining. Under low- and mitophagy form an integrated quality control axis that glucose (LG; 5 mM) ambience, mitochondria exhibited a prevents damaged mitochondria from accumulating within filamentous shape, whereas they transformed into short rods the cell, generation of reactive oxygen species (ROS), buildup after high-glucose (HG; 30 mM) treatment for 24 hours of mutated mtDNA, and release of apoptogenic proteins (i.e., (Figure 1A, a). Immunoblot analyses revealed upregulation Cytochrome C). Thus, the cell avoids undergoing oxidative of the mitochondrial profission proteins Drp1 and Fis1, stress, mitochondrial outer membrane (MOM) permeabiliza- whereas the MOM profusion protein Mfn2 was downregu- tion, and apoptosis, and such quality control maintains a pop- lated. Interestingly, expression of OPA1, a protein regulating ulation of healthy mitochondria.12,13 In kidneys, because of fusion of the mitochondrial inner membrane, was not ob- high energy demand in the mitochondrial-enriched tubular viously altered under HG ambience (Figure 1B). LC3 was compartment, it is likely that mitochondrial dysfunctions in used as an autophagy marker and the immunofluorescence the diabetic state could contribute to cell injury, proteinuria, performed by using LC3 antibody yielded punctate green and loss of renal functions.4,14 However, the mechanism of fluorescence demarcating the autophagosomes (Figure 1A, mitochondrial damage and the dynamic regulation in states of b). After HG treatment, the number of autophagosomes as hyperglycemia in tubular pathobiology are not clearly defined well as their intensity of fluorescence were reduced (Figure and remain to be investigated. 1A, b). During autophagy, LC3-I is known to convert to LC3- Myo-inositol oxygenase (MIOX) is a tubular-specificen- II.21 Expression of converted LC3-II was decreased under zyme that catabolizes myo-inositol to D-glucuronate, which HG ambience by immunoblot analyses, but no significant enters into the pentose phosphate pathway.15,16 Other than its change was seen in Autophagy-Related Gene 5 (Atg5) (Fig- role to channel glucose intermediaries, studies have indicated ure 1C). In addition, expression of mitophagy-related pro- upregulation of MIOX and its relevance in the pathogenesis of teins PTEN-induced putative kinase 1 (Pink1) and Parkin tubular injury in DKD.15,16 Previously, we had shown that the was decreased (Figure 1C). Cytosolic and mitochondrial upregulation of MIOX during hyperglycemia could modulate ROS generations were increased notably by HG treatment, redox imbalance by transcriptional mechanisms and also, and they are indicated by H2-DCFDA (Figure 1A, green) and conceivably, induce mitochondrial-dependent apoptosis, MitoSox (Figure 1A, red) staining, respectively (Figure 1A, which inevitably may lead to tubulointerstitial injury and fi- c and d). An increase in apoptosis was also observed in brosis.16–18 With respect to the amelioration of such an injury, cells subjected to HG, which was assessed by the ter- D-glucaric acid, an anticancer drug, may serve as a potential minal deoxynucleotidyl transferase-mediated digoxigenin- pharmacologic inhibitor of MIOX.19 Interestingly, a deriva- deoxyuridine nick end-labeling (TUNEL) procedure (Figure tive of this drug is believed to have a renoprotective effect in 1A, e). The apoptosis was accompanied with upregulation of DKD by a yet to be defined mechanism.20 Also, the mecha- Bax and Cytochrome C (Figure 1D). Notably, the level of the nism by which MIOX induces mitochondrial dysfunctions tubular-specific enzyme MIOX was increased under HG am- during tubular injury and whether its pharmacologic inhibi- bience (Figure 1, A, f and D). Quantitative analyses con- tion ameliorates DKD are unknown. firmed that HG induced mitochondrial fragmentation in a In this study, we showed the effect of MIOX on mitochon- time-dependent manner, concurring with increased intra- drial quality control in tubules in DKD and highlighted the cellular ROS production and apoptosis after HG treatment mechanism by which MIOX perturbs the mitochondrial (Figure1,E,G,andH).Autophagywasmarginallyincreased homeostasis. Activated MIOX during hyperglycemia induces initially, but it was reduced after 6 hours under HG ambience mitochondrial fragmentation and depolarization but mean- (Figure 1F).

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1305 BASIC RESEARCH www.jasn.org

Figure 1. Effect of HG ambience on mitochondrial dynamics, autophagy, ROS production, and apoptosis in renal proximal tubular cells. (A, a) Under LG (5 mM) ambience, mitochondria exhibited a filamentous network, whereas they transformed into short rods after HG (30 mM) treatment for 24 hours. (B) Western blot analyses revealed upregulation of mitochondrial profission proteins Drp1 and Fis1; profusion protein Mfn2 was downregulated, whereas another fusion mediator OPA1 was not obviously altered. (A, b) By immunofluorescence microscopy, LC3, an autophagy marker, yielded an intense punctate green fluorescence highlighting the cellular autophagosomes, which was reduced under HG ambience. (C) By immunoblotting, expression of converted LC3-II was found to be decreased under HG ambience, whereas that of Atg5 was unaltered. Expression of mitophagy-related proteins, Pink1 and Parkin, was decreased. (A, c and d) Intracellular and mitochondrial ROS generation was increased by HG treatment, which was detected by H2-DCFDA (green) and MitoSox (red) staining. (A, e) By TUNEL procedure, an increase in apoptosis was observed in HK2 cells subjected to HG treatment. (D) The apoptosis was associated with upreg- ulation of Bax and Cytochrome C. (A, f and D) The expression of MIOX, a tubular-specific enzyme, was also concomitantly increased. (E–H)

1306 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

Overexpression of MIOX Accentuates Altered especially under HG ambience, whereas Atg5 expression was Mitochondrial Dynamics and Associated Protein largely unaffected (Figure 3B). In addition, expression of mi- Expression and Reduces Mitochondrial Membrane tophagy-related proteins, Pink1 and Parkin, was reduced in a Potential under HG Ambience dose-dependent manner under HG ambience. Notably, their To determine whether MIOX upregulation has any effect on expression was not significantly changed by MIOX under LG mitochondrial fusion and fission processes, we used an MIOX but was inhibited by MIOX overexpression after HG treatment stably transfected LLC-PK1 cell line that is available in our (Figure 3, A, d and B). laboratory. The cells overexpressing MIOX have higher percen- tages of fragmented mitochondria under LG or HG ambience Overexpression of MIOX Induces ROS Production, (Figure 2, A, a and b and C). This was accompanied with dose- Cytochrome C Release, Bax Mitochondrial dependent increased expressions of Drp1 and Fis1 under HG Translocation, and Apoptosis ambience, whereas the levels of Mfn2 were reduced compared Failure of mitochondrial quality control and accumulation of with control cells under similar conditions (Figure 2, A, c and damaged mitochondria may inevitably lead to oxidative stress d and B). Cells transfected with MIOX and subjected to LG had and cell death. In view of this information, the effect of MIOX Drp1 and Mfn2 fluorescence intensities similar to that of control on ROS production and mitochondrial-dependent apoptosis cells treated with HG (Figure 2A, c and d). Notably, activated wasinvestigated. Boththe intracellular andmitochondrial ROS Drp1 was translocated to mitochondria in a dose-dependent levels were increased under HG ambience (Figure 4A). The manner, and this process was further accentuated in cells over- cells overexpressing MIOX exhibited relatively higher inten- expressing MIOX (Figure 2B). In addition, a loss of mitochon- sity of staining with both H2-DCFDA and MitoSox probes, drial voltage potential (DCm) was seen under HG ambiance, and statistics attested to these observations (Figure 4, A and B). which was assessed by TMRE staining, and it was further atten- For assessment of apoptogenic proteins Cytochrome C and uated by MIOX overexpression (Figure 2, A, e and D). Bax, double immunofluorescence was performed. Under LG ambience, cells with MIOX overexpression had higher degrees Overexpression of MIOX Further Contributes to the of fluorescence and altered distribution of both Cytochrome Inhibition of Autophagy, Mitophagy, and Their C (Figure 4C, green) and Bax (Figure 4C, red) as well as an Associated Protein Expression under HG Ambience increased number of apoptotic nuclei (Figure 4C, blue) (49,6- During fission, the daughter mitochondria may undergo diamidino-2-phenylindole [DAPI]), suggesting that MIOX uneven division, and the ones with significant loss of mem- alone may induce Cytochrome C and Bax redistribution that brane potential (DCm) are likely to be segregated and elimi- leads to subsequent apoptosis (Figure 4C, columns 1 and 2). nated by mitophagy, whereas healthy daughter mitochondria While under HG ambience, MIOX further enhanced their re- with increased voltage potential can undergo fusion to main- distribution and apoptosis (Figure 4C, columns 3 and 4). tain mitochondrial homeostasis. These dynamic events lead to Western blot analyses clearly delineated the correlative the maintenance of a purified mitochondrial pool and efficient changes that cytosolic Bax activated and translocated to mi- cellular processes.22 In view of this dynamic interplay, the tochondria after HG treatment, whereas intramitochondrial effect of MIOX overexpression on both autophagy and mi- Cytochrome C was released into cytosol. These changes were tophagy was assessed. LC3 staining was used to mark auto- further enhanced in cells overexpressing MIOX, leading to , whereas a combination of LC3 and Mitotracker accentuated apoptosis (Figure 4, D and E). red staining was used to delineate mitophagy. A reduction of autophagy was noted in cells overexpressing MIOX under LG Inhibition of MIOX Attenuates Mitochondrial or HG ambience compared with the control cells (Figure 3A, Fragmentation, ROS, and Apoptosis but Increases a). This was accompanied with increased mitochondrial fis- Autophagy and Mitophagy through Its Modulation of sion, and the merged photographs indicated an associated re- Pink1 and ROS Levels duction of mitophagy (Figure 3A, b and c). The decrease of MIOX expression was inhibited in HK2 and LLC-PK1 cells autophagy was confirmed after enumeration of cells with using MIOX siRNA and an inhibitor of MIOX, D-glucaric acid. punctate LC3 appearance (Figure 3C). Western blot analyses The drug has been reported to suppress MIOX in tubular also indicated a reduced expression of autophagy-associated cells.19 Under HG conditions, cells transfected with MIOX proteins (LC3-II and Atg5), and this decrease was dose de- siRNA or incubated with 10 mM D-glucaric acid had increased pendent in cells subjected to HG treatment. The LC3-II autophagy and mitophagy, which was evidenced by the in- expression was further attenuated by MIOX overexpression, creased number of punctate LC3 dots and Pink1 upregulation.

Quantitative analyses confirmed HG-induced mitochondrial fragmentation in HK2 cells in a time-dependent manner, concurring with increased intracellular ROS production and apoptosis after HG treatment. Interestingly, autophagy was reduced only after 6 hours after the HG treatment. D-Glu, D-glucose. *P,0.05 compared with the LG (5 mM) group.

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1307 BASIC RESEARCH www.jasn.org

Figure 2. Accelerating effect of MIOX on the altered mitochondrial dynamics, associated proteins, and mitochondrial membrane potential under HG ambience. (A, a and b and C) MIOX overexpressing LLC-PK1 cells had higher percentages of fragmented mitochondria under LG or HG ambience. (A, c and d) Under LG ambience, MIOX-transfected cells had Drp1 and Mfn2 fluorescence intensity similar to the control cells treated with HG. (B) A dose-dependent increase in the expression of Drp1 and Fis1 was seen under HG ambience, whereas the expression of Mfn2 was slightly reduced. Translocation of Drp1 to mitochondria was also seen in a dose-dependent manner, and it was accentuated in MIOX overexpressing cells. (A, e and D) This was associated with a decrease in TMRE staining, indicating loss of mitochondrial voltage potential (DCm). Con, control.

At the same time, there was an attenuation of mitochondrial and induces cell injury, we investigated two conceivable path- fragmentation and altered expression of mitochondrial- ways that may relate to Pink1 regulation and MIOX-induced shaping proteins (Figure 5, A, a–c, B, C, and F). These changes oxidative stress. As shown in Figure 5, A–F, cells transfected were accompanied by reduced mitochondrial ROS production, with Pink1 pcDNA plasmid and subjected to HG had reduced decreased expression of Bax and Cytochrome C, and apoptosis mitochondrial fragmentation and Drp1 levels but increased compared with cells transfected with empty vector (Figure 5, Mfn2 expression together with increased autophagy and mi- A, d, D, and F). These observations indicated that MIOX in- tophagy and decreased ROS generation and apoptosis. These hibition could normalize the mitochondrial quality surveil- cellular and biochemical changes are similar to those seen with lance by attenuating mitochondrial fragmentation, generation MIOX knockdown. Interestingly, Pink1 overexpression had little of ROS, and apoptosis while accentuating mitophagy under effect on MIOX expression, whereas MIOX inhibition upregu- HG ambience. To further explore the possible mechanisms by lated Pink1 expression compared with cells treated with HG which MIOX disrupts the mitochondrial quality control axis (Figure 5F), suggesting that MIOX may serve as an upstream

1308 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

Figure 3. Accelerating effect of MIOX on autophagy, mitophagy, and their associated protein expression under HG ambience. (A, a–c) Fluorescent punctate LC3 staining (green) was reduced in MIOX overexpressing cells in both LG and HG ambience, and this was associated with altered Mitored staining (red) and decreased LC3 and Mitored merged dots (yellow), indicating increased mito- chondrial fission and decreased autophagy and mitophagy. (B) Western blot analyses revealed a reduced expression of LC3-II and Atg5 in a dose-dependent manner in cells subjected to HG treatment, and their expression was further attenuated by MIOX overexpression, whereas Atg5 expression was not obviously affected. (C) Quantification also indicated reduced cell percentage with punctate LC3 staining. (A, d and B) The expression of mitophagy-related proteins, Pink1 and Parkin, was also reduced under HG ambience, and their expression was further suppressed by MIOX overexpression when subjected to HG. Con, control. factor that negatively modulates Pink1 activity. In addition, cells suggested that the negative effect of MIOX on mitophagy under pretreated with the ROS scavenger N-acetylcysteine (NAC) and HG ambience may be through the downregulation of Pink1 and then subjected to HG showed similar recovery of mitochondrial consequentially, the reduced Parkin–Mfn2 interaction. These in integrity and reduced apoptosis. Because MIOX has been shown vitro data indicated a key role of mitochondrial quality control in to induce redox imbalance under HG ambience,18 this would the maintenance of their homeostasis and cellular integrity, indicate another conceivable pathway through which MIOX where MIOX may be critically involved in such pathobiologic may disrupt mitochondrial quality surveillance. Furthermore, processes. In view of this contention, we examined if inhibition Pink1 has been recently reported to facilitate and enhance of MIOX in vivo has a therapeutic potential in amelioration of the protein–protein interaction between Parkin and Mfn2 on tubulopathy in diabetic mice. the MOM, which could induce activation of mitophagy.23 The coimmunoprecipitation experiments of this study indicated Effect of D-Glucarate on Renal Functional and that, in tubular cells, HG inhibited the Parkin–Mfn2 interaction Morphologic Characteristics and MIOX Expression in compared with LG (Figure 5G). Knockdown of MIOX under Streptozotocin-Induced Diabetes in Mice HG ambience upregulated Pink1 expression and increased the D-glucarate, the calcium salt of D-glucaric acid, was added to Parkin–Mfn2 interaction as well, whereas overexpression of the mouse dietary chow (1% or 2% [wt/wt]) to inhibit MIOX. Pink1 had similar and more prominent effects on such an in- Four groups of mice (n=8 per group) were assigned: control teraction, which thereby enhance the process of mitophagy. This CD1, streptozotocin (STZ)-induced diabetic, and STZ+1% or

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1309 BASIC RESEARCH www.jasn.org

Figure 4. Accelerating effect of MIOX on ROS production, Cytochrome C release, Bax mitochondrial translocation, and apoptosis under HG ambience. (A and B) Both intracellular and mitochondrial ROS levels were increased under HG ambience, which was measured by H2-DCFDA and MitoSox staining, respectively. The MIOX overexpressing cells exhibited relatively strong staining with both H2-DCFDA and MitoSox. *P,0.05 compared with the H2-DCFDA-stained LG group. #P,0.05 compared with the MitoSox-stained LG group. (C) The MIOX overexpressing cells had (columns 1 and 2) a higher degree of fluorescence, altered distribution of Cytochrome C (green) and Bax (red), and an increased number of apoptotic nuclei (blue; DAPI), whereas they were enhanced (columns 3 and 4) after HG treatment. (D and E) Western blot analyses indicated translocation of Bax into the mitochondria and that of Cytochrome C into the cytosol in a dose-dependent manner in cells subjected to HG treatment, which is exacerbated by MIOX overexpression, leading to increased apoptosis. Con, control; Cyt.C, Cytochrome C; cyto, cytosol; mito, mitochondria.

STZ+2% D-glucarate dietary supplementation groups. decreased serum creatinine and urinary ACR compared with D-glucarate treatment was initiated after 1 week of STZ injection the STZ group (Figure 6, A–D). Interestingly, mice with 2% and continued for another 8 weeks. During this time frame, D-glucarate treatment had better preservation of renal function body weight, blood glucose, serum creatinine, and urinary than those receiving 1% D-glucarate. Morphology studies re- albumin-to-creatinine ratio (ACR) were monitored. Com- vealed increased periodic acid–Schiff-positive tubular brush pared with controls, the STZ group had higher blood glucose border staining, which is indicative of protein excretion, no- levels and increased in serum creatinine and urinary ACR val- table tubular epithelial disruption, hypertrophy of glomeruli, ues but reduced body weight (Figure 6, A–D). The diabetic and increase of mesangial matrices in the STZ group. The di- mice receiving D-glucarate had a smaller reduction in body abetic mice with drug treatment showed notable improvement weight and slightly reduced blood glucose as well as notably largely in the tubular compartment with less cellular disruption

1310 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

Figure 5. Attenuation of mitochondrial fragmentation, ROS, and apoptosis but increased autophagy and mitophagy after inhibition of MIOX under HG ambience. (A, a–c, B, C, and F) Under HG, cells transfected with MIOX siRNA or incubated with D-glucaric acid had increased expression of punctate LC3 and Pink1, indicative of increased autophagy/mitophagy, as well as decreased mitochondrial fragmentation and altered expression of mitochondrial shaping proteins. (A, d and D–F) This was accompanied with decreased mi- tochondrial ROS production, reduced expression of Bax and Cytochrome C, and apoptosis compared with empty vector-transfected cells treated with HG. MIOX may modulate mitochondrial quality control through two different mechanisms (i.e., Pink1 pathway and MIOX-induced oxidative stress). (A, columns 5 and 6 and F) Pink1 pcDNA-transfected cells had reduced mitochondrial fragmentation and Drp1 levels but increased Mfn2 expression together with increased autophagy and mitophagy and decreased ROS generation and apoptosis. Similarly, the cells pretreated with ROS scavenger N-acetylcysteine (NAC) and subjected to HG showed partial recovery of mitochondrial integrity and reduced apoptosis. (E and F) These changes were similar to those seen with MIOX inhibition, whereas MIOX has been shown to induce ROS and inhibit Pink1 expression. (G–I) The coimmunoprecipitation experiments indicated that, in tubular cells, HG inhibited the Parkin–Mfn2 interaction, whereas knockdown of MIOX upregulated Pink1 expression and increased the Parkin–Mfn2 interaction. The overexpression of Pink1 had similar and more prominent effects on such interactions. Cyt.C, Cytochrome C; D-Glu, D-glucose; IP, immunoprecipitation; WB, Western blot. *P,0.05 compared with HG plus the empty vector group.

(Figure 6E). Similar tubular alterations as well as changes in scattered in tubules as well in their lumens in the STZ group. mitochondria morphology were observed after Toluidine blue There was partial recovery in mitochondrial morphology along staining of EPON-embedded sections (Figure 6F). In the control with improvement in tubular cell integrity in the STZ and group, the mitochondria appeared cylindrical, elongated, and D-glucarate groups. Quantitative analyses with damage scores well organized. However, the organelles were fragmented and ranging from zero to three indicated that both the STZ+1%

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1311 BASIC RESEARCH www.jasn.org

Figure 6. Ameliorative effect of D-glucarate on renal functional and morphologic characteristics by modulation of MIOX expression in STZ-induced diabetes in mice. (A–D) The STZ-treated group had higher blood glucose levels, an increase in serum creatinine and urinary ACR values, and reduced body weight compared with the controls. In general, the diabetic mice receiving 2% D-glucarate had better preservation of renal functions than those receiving 1% D-glucarate. (E) Increased periodic acid–Schiff (PAS)-positive tubular

1312 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

and STZ+2% D-glucarate groups had slightly reduced glomeru- mice and the effect of MIOX inhibitor on these pathobiologic lar damage but notable decrease of tubular damage compared events. Frozen kidney sections from four groups were stained with the STZ group (Figure 6, H and I). Concurrently, the MIOX with LC3. A basal degree of autophagy, which was indicated by expression and activity (Figure 6, G and J) were markedly in- punctate LC3 appearance in renal tubules, was seen in the creased in proximal tubules in the STZ group compared with control group. The LC3 staining was greatly reduced in tubules controls and notably attenuated in STZ-treated mice adminis- from the STZ group, and the formation of autophagosomes tered D-glucarate. was partially restored or enhanced after D-glucarate treat- ment (Figure 8A). The morphometric analyses showed that D-Glucarate Restores Altered Mitochondrial the number of autophagosomes per tubule in the STZ and Morphology in Tubules of STZ-Induced Diabetic Mice D-glucarate groups was much higher than that in the STZ To confirm alterations of mitochondrial morphology as observed group and comparable with or higher than that in the control by Toluidine blue staining, expression of mitochondrial-shaping group (Figure 8B). Meanwhile, Atg5 and Pink1 showed par- proteins and organelle ultrastructural changes were assessed. The allel changes in the tubules of mice from four groups, and their mitochondrial profission proteins Drp1 and Fis1, expressed in situ expression was reduced in the STZ group but partially mainly in renal tubules by immunohistochemistry, were upregu- restored after D-glucarate treatment (Figure 8C). The status of lated in the STZ group, but their in situ expression was reduced to a autophagic vacuoles, including autophagosome and autolyso- certainextentinSTZ+1%orSTZ+2%D-glucarate groups (Figure some, was evaluated by electron microscopy. Autophagic 7A, a and b). Conversely, Mfn2 levels were notably decreased in vacuoles were reduced in tubular cells in the STZ group com- tubules of the STZ group but partially restored in some of the pared with the controls, whereas they were partially restored in proximal tubules after D-glucarate treatment (Figure 7A, c). These tubules from mice treated with 1% D-glucarate and even en- results suggested a restoration of disrupted mitochondrial dynam- hanced in the 2% drug-treated group (Figure 8, D and H). ics by D-glucarate in tubules of diabetic mice. Next, status of mi- High-magnification electron micrographs depicted a typical tochondrial morphology was assessed by electron microscopy. In morphology of autophagosome containing fragmented cellu- the control group, the majority of the tubular cells had elongated lar organelles (Figure 8E). An autophagosome-engulfing cylindrical-shaped mitochondria (marked by the asterisks in Fig- mitochondria, indicative of mitophagy (Figure 8F), and an ure 7, B and F). In the STZ group, .50% of the tubular cells autolysosome (Figure 8G) were readily seen in tubular cells exhibited fragmented mitochondria; the mitochondrial morphol- of D-glucarate–treated groups, suggesting restoration of au- ogy was partially restored by D-glucarate supplementation, and tophagy and mitophagy. the percentage of tubular cells having fragmented mitochondria was reduced. High-magnification electron microscopy (Figure 7, D-Glucarate Reduces Tubular Expression of C–E) revealed elongated mitochondria with organized cristae in Apoptogenic Proteins, ROS Production, and Apoptosis the control group (Figure 7C). However, in the STZ group, most in STZ-Induced Diabetic Mice mitochondria exhibited rod or sphere shapes and had cristolysis Expression of one key apoptogenic protein, Bax, was found to with focal disruption of the inner mitochondrial membranes (Fig- be increased in tubules of the STZ group, and it was partially ure 7D). These changes in the diabetic group were partly restored inhibited by D-glucarate treatment in diabetic mice (Figure after administration of D-glucarate, and they showed the morpho- 9A). Similarly, Cytochrome C, which is normally confined logic features of the mitochondrial fusion–fission process (the to the mitochondrial matrix in the control group, was seen fragmented mitochondria in the hyperglycemic state underwent released into the cytosol in the STZ group with concomitant refusion to form elongated mitochondria after D-glucarate treat- upregulation of MIOX (Figure 9B). However, lesser degrees of ment) (Figure 7, E and F). Cytochrome C redistribution together with the suppressed expression of MIOX were noted in tubules of the STZ mice D-Glucarate Normalizes Reduced Autophagy and treated with D-glucarate. The renal tissues of the STZ group Mitophagy in Tubules of STZ-Induced Diabetic Mice subjected to DHE or TUNEL staining showed concomitant Various morphologic studies were performed to investigate the increases of oxidative stress and apoptosis, both of which were changes of autophagy and mitophagy in STZ-induced diabetic largely reduced with D-glucarate treatment, especially in tubules

brush border staining, tubular epithelial disruption, hypertrophy of glomeruli, and mesangial matrix numbers were observed in the STZ group. Treatment of diabetic mice with D-glucarate had notable improvement (largely in the tubular compartment) with less cellular disruption. (F) In the tubules of the control group after Toluidine blue staining, the mitochondria appeared cylindrical, elongated, and well organized. The mitochondria were fragmented and scattered in the STZ group. A partial recovery in mitochondrial morphology along with improvement in tubular cell integrity was seen after 1% or 2% D-glucarate administration. (H and I) Quantitative analyses with damage scores ranging from zero to three indicated that D-glucarate treatment resulted in amelioration of changes in glomeruli and tubules after STZ induction of diabetes. (G and J) Concurrently, MIOX expression and activity were decreased in diabetic mice receiving D-glucarate. # DG, D-glucarate. *P,0.05 compared with the STZ group; P,0.05 compared with the control group.

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1313 BASIC RESEARCH www.jasn.org

Figure 7. Restoration of altered tubular mitochondrial morphology in diabetic mice after administration of D-glucarate. (A) The mi- tochondrial profission proteins Drp1 and Fis1, expressed mainly in renal tubules, were upregulated in the STZ group, but their expression was reduced after administration of D-glucarate. Conversely, expression of mitochondrial profusion protein Mfn2 was de- creased in the STZ group but partially restored in some of the proximal tubules after drug treatment. (B and F) By electron microscopy, the majority of the tubular cells had mitochondria with elongated cylindrical shape (yellow asterisks) in the control group, whereas in the STZ group, about 50% of the tubular cells had fragmented mitochondria, and their morphology was partially restored by D-glucarate treatment. (C) Higher magnification electron micrographs showed elongated mitochondria with organized cristae in the control group. (D) In the STZ group, mitochondria showed a spherical shape and had undergone cristolysis (matrix vacuoles). (E) These changes were negated after D-glucarate treatment, and mitochondria undergoing fusion are seen. (F) The mitochondrial changes were quantified and # included in bar graphs. DG, D-glucarate; EM, electron microscopy. *P,0.05 compared with the STZ group; P,0.05 compared with the control group.

1314 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

Figure 8. Restorative effect of D-glucarate on autophagy and mitophagy in STZ-induced diabetic mice. (A) The basal level of au- tophagy, indicated by punctate LC3 staining, was seen in the renal tubules of the control group, which disappeared in the tubules of the STZ group, whereas they were restored after D-glucarate treatment. (B) Morphometric analyses confirmed these observations. (C) The expressions of Atg5 and Pink1 exhibited parallel reduction and restoration in the tubules of mice from four groups. (D and H) By electron microscopy, autophagic vacuoles seen in the control group were greatly reduced in proximal tubular cells in the STZ group, whereas they were restored in tubules from mice treated with D-glucarate. (E) Higher magnification electron micrographs revealed a typical morphology of an autophagosome containing fragmented cellular organelles, (F) a mitophagosome, and (G) an autolysosome # in tubular cells after D-glucarate treatment. DG, D-glucarate; EM, electron microscopy. *P,0.05 compared with the STZ group; P,0.05 compared with the control group.

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1315 BASIC RESEARCH www.jasn.org

(Figure 9C). These observations suggest that D-glucarate, by in- the basal rate of autophagy may increase initially but because of hibiting MIOX, exerts antioxidant and antiapoptotic effects in its relative decrease, subsequently, the podocytes and tubular the renal tubular compartment of diabetic mice. cells get readily damaged.33–36 In addition, elimination of dam- aged mitochondria by mitophagy was reported to be relatively deficient or decreased in the cardiac musculature of patients with DISCUSSION diabetes.37 Similarly, two mitophagy regulators, Pink1 and Parkin (the genes that are relevant in the pathogenesis of Parkinson Glomerular lesions have been the major focus of investigation in disease), were found to be downregulated in cardiac and skeletal DKD; however, inrecent years,the importance ofspecific tubular muscles in diabetic states.37,38 In this study, we observed that changes has also been emphasized. The features of tubulointer- most of the mitochondria within the injured tubules in STZ- stitialhistopathologyincludeinflammationandapoptosis,which induced diabetic mice were fragmented and had a randomly are not necessarily secondary to glomerular lesions.7,24–26 In disorganized cellular distribution (Figures 6F and 7B). At the addition, tubular injury alone is believed to be closely correlate same time, the cellular autophagic vacuoles and mitophagy with clinical manifestations of DKD, including the extent albu- were reduced in tubules, indicating less efficient removal of min excretion and its degradation in tubules. The latter may damaged mitochondria and a compromised surveillance mech- serve as a primary key site for the development of DKD, because anism. Importantly, increased mitochondrial fragmentation and tubular injury has been described to be a better predictor in the reduced autophagy or mitophagy under HG ambience were ob- progression of the renal disease.3,4 Thus, the dominant glomer- served in tubular cells in both our in vitro and in vivo studies ulocentric view of DKD has partly shifted to tubules, and the (Figures 2, 3, 7, and 8). These two cellular events together in- injury has been described as diabetic tubulopathy.3 In this study, dicated an impaired mitochondrial quality control, which may notable changes in the tubules were also observed in STZ-induced inevitably result in excessive ROS generation and initiation of the diabetic mice (Figures 6 and 7). This brings up an important mitochondrial apoptotic cascade (Figures 4 and 9). question of whether attenuation of tubular damage by thera- Next, the question needs to be addressed of what caused the peutic approaches can delay or reverse DKD progression? Our mitochondrial quality control failure to rescue the damaged results (Figure 6) indicate that D-glucarate, an MIOX inhibitor, mitochondria. It is conceivable that MIOX may have interfered not only ameliorates diabetic tubular injury but also, improves with the mitochondrial self-rescuing process. MIOX, a tubular the renal functions, which are reflected by decreased serum enzyme, is upregulated during HG ambience, and its critical – creatinine levels and urinary ACR. Interestingly, D-glucarate also involvement in the pathogenesis of DKD has been shown.15 18 attenuated glomerular damage to some extent, suggesting During hyperglycemia, MIOX induces redox imbalance and mi- a possible link between tubular injury and glomerular burden tochondrial dysfunction that could result in tubular oxidative during the development of DKD. Similarly, recent studies by injury and apoptosis.17,18 In view of this contention, the under- Lin et al.26,27 showed that inhibition of TLR4 in the tubules of lying mechanisms by which MIOX may cause mitochondrial diabetic mice led to reduced albuminuria, BUN, glomerular dysfunction and tubular cell injury were delineated. As shown hypertrophy, and glomerulosclerosis in addition to attenuation in Figure 10, under HG ambience, MIOX is upregulated in tu- of tubulointerstitial injury. In this regard, the approach targeting bular cells,16–18 and then, it modulates mitochondrial dynamics the tubular compartment with respect to mitochondrial homeo- through two conceivable pathways. In the first pathway, MIOX stasis can be beneficial for the overall amelioration of DKD. increases intracellular ROS generation through transcriptional Mitochondria are abundantly present in tubular cells to meet mechanisms,18 which would then activate cytosolic Drp1 and high metabolic energy demand, such that they play a pivotal role promote its mitochondrial translocation, leading to mitochon- in the pathogenesis of acute or chronic tubular injury.10,28,29 drial fragmentation.10,39 In the second pathway, elevated MIOX Recent studies have delineated an integrated regulation of the levels would suppress Pink1 expression, which has a negative mitochondrial lifecycle.11,22 The dynamic nature of mitochon- regulatory effect on Drp1 activity, such that the Pink1 deficiency dria enables them to constantly undergo fusion and fission would increase Drp1-dependent mitochondrial fragmenta- cycles, whereas the selective autophagic removal of damaged tion.40 In other words, excessive ROS generation and paucity organelles (mitophagy) purifies the mitochondrial pool, all of of Pink1, both induced by MIOX, lead to Drp1 activation and which form an integrated quality control axis to maintain mi- translocation to mitochondria. Subsequently, Drp1 interacts tochondrial homeostasis. This concept of mitochondrial quality with Fis1 on the MOM and causing mitochondrial fragmenta- control is intricately related to mitochondrial dynamics and mi- tion. Excessive fragmented mitochondria are depolarized and tophagy, and it has been investigated in various diseases,12,30–32 lose viability for the fusion process, but the quality control of but studies on mitochondrial quality control in kidney diseases, mitochondria allows these injured organelles to be recycled by including DKD, are somewhat limited. A few reports have autophagosomes through mitophagy. However, under HG am- indicated a pivotal role of mitochondrial dynamics in the ho- bience, activated MIOX inhibited Pink1 and conceivably, pre- meostasis of different kidney cells during DKD and AKI.8,28 vented its downstream Parkin translocation to mitochondria, However, autophagy is believed to exert a cytoprotective effect thus blocking subsequent interaction between Parkin and during kidney injury, such as in AKI, DKD, and obesity, where Mfn2 on MOM. Eventually, these events led to a compromise

1316 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

Figure 9. Reduction in the expression of apoptogenic proteins, ROS production, and apoptosis in STZ-induced diabetic mice after D-glucarate administration. (A) Expression of Bax was found to be increased in tubules of the STZ group, which were partially inhibited by D-glucarate treatment in diabetic mice. (B) Similarly, Cytochrome C, normally confined to the mitochondria, was seen diffusely distributed in the cytosol in the STZ group with concomitant upregulation of MIOX, and this redistribution was suppressed with the D-glucarate treatment. Their concomitant changes were readily seen in the merged (MIOX and Cyt.C) images. (C) An increase in DHE

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1317 BASIC RESEARCH www.jasn.org

in the process of mitophagy. As a result, the accumulation of 5–30 mM D-glucose. For overexpression of MIOX and Pink1, cells were fragmented and damaged mitochondria triggered ROS overpro- transfected with pcDNA3.1 or pcDNA-DEST47 plasmids. For gene duction as well as the mitochondrial apoptotic pathway—apro- disruption, cells were treated with MIOX siRNA using Lipofectamine cess by which the apoptogenic Bax translocates and inserts into 2000 reagent (Life Technologies). To pharmacologically inhibit MIOX, mitochondria, whereas Cytochrome C is released into the cyto- D-glucaric acid (Sigma-Aldrich) was added to the culture media and sol, leading to Caspase 9 activation and eventually, apoptosis. cotreated with various concentrations of glucose. N-acetylcysteine Conversely, inhibition of MIOX by D-glucarate restores mito- (Sigma-Aldrich) was added for 2 hours before glucose treatment to chondrial quality surveillance by preventing mitochondrial scavenge the ROS. damage and cellular apoptosis, thereby attenuating diabetes- induced tubular damage. Generation of MIOX Stably Transfected Cell Lines Of note, D-glucarate is a natural compound, and it is present MIOX plasmid construct and MIOX overexpressing cell lines were in substantial amounts in certain fruits and vegetables. Daily generated in our laboratory.16,17 A porcine MIOX cDNA was gener- supplementation of D-glucarate has some key beneficial effects ated by RT-PCR and cloned into pcDNA3.1 (Life Technologies). The (i.e.,detoxification of noxious substances and anticancer regi- plasmid constructs and empty vector (control) were then transfected – mens).41 43 The primary mechanisms of action of D-glucarate into LLC-PK1 cells, and stable transfectants were selected by growing may be through its ability to enhance detoxification of toxins and cells in the presence of Geneticin and maintained. carcinogens as well as its antioxidative potential,44,45 which may be linked to the inhibition of MIOX. In addition, a derivative of Western Blotting and Coimmunoprecipitation Studies D-glucaric acid, D-saccharic acid 1, 4-lactone, has been recently For Western blot analyses, whole-cell lysate, cytoplasmic, or mito- reported to be protective in both diabetes and its renovascular chondrial extracts of treated cells were subjected to 5%–15% SDS- complications in rats.20,46 Likewise, we elucidated an important PAGE. The fractionated proteins were transferred onto nitrocellulose therapeutic effect of D-glucarate, which was through the inhibi- membranes, which were then incubated with various primary anti- tion of MIOX and thus, normalization of mitochondrial quality bodies. The anti-MIOX was previously generated in our laboratory.18 control, reduction of oxidative stress, and apoptosis in tubular Other antibodies included anti-Drp1,anti-Fis1,anti-Bax,anti- cells with improvement in renal functions. Despite our extensive Cytochrome C, anti-Pink1, anti-Parkin (Santa Cruz Biotechnology), investigatory efforts, there are still certain issues that remain un- anti-LC3B (Cell Signaling Technology), anti-OPA1 and anti-Mfn2 resolved. For instance, the mechanism by which MIOX modu- (Abcam, Inc.), and anti-Atg5 (Gene Tex). After overnight incubation, lates Pink1 and downstream Parkin remains to be delineated. the membranes were immersed in a solution containing individual Also, whether MIOX plays a direct role in the regulation of mi- secondary antibodies (Santa Cruz Biotechnology). Membrane blots tochondrial dynamics and functions is worthy of future studies. were then used for developing autoradiograms using the ECL system In conclusion, this study shows the effect of MIOX on (Amersham Biosciences). mitochondrial quality control disruption during DKD using in For coimmunoprecipitation, cell lysates from various experiments vitro and in vivo models. MIOX upregulation under HG ambi- were used. They were incubated with respective rabbit polyclonal ence induces mitochondrial fragmentation and depolarization; antibodies followed by the addition of protein A-Sepharose CL-4B, meanwhile, it inhibits selective autophagic removal of damaged and incubation was extended for 12 hours. The antigen–Ab–protein A mitochondria through Pink1-dependent Mfn2–Parkin interac- complex was pelleted and washed with RIPA buffer. The pellet was tion. The net effect of these events would be disruption in the suspended in lysis buffer and subjected to SDS-PAGE for Western blot surveillance of mitochondrial quality control, accumulation of analyses. dysfunctional organelles, generation of ROS, and apoptosis, leading to tubular injury and DKD progression. Interestingly, Immunofluorescence Studies and TUNEL Assay the fact that D-glucarate could normalize mitochondrial ho- Cells were fixed with paraformaldehyde, permeabilized, blocked with moeostasis by inhibition of MIOX would suggest a therapeutic BSA, and incubated with primary antibodies. After washing with PBS, potential of D-glucarate in the amelioration of DKD. they were reincubated with secondary antibodies conjugated with Alexa Fluor 488 or 594 (Life Technologies). Cells were counterstained with DAPI to delineate the nuclei and examined by confocal CONCISE METHODS microscopy. For kidney tissue immunofluorescence, 4-mm-thick cryostat sec- Cell Culture Studies tions were prepared, mounted on glass slides, air dried, and rehydra- Renal proximal tubular epithelial cell lines HK-2 (human) and LLC-PK1 ted with PBS. Similarly, 3-mm-thick paraffin-embedded sections (porcine) were used.17,18 They were maintained in media containing were prepared, dewaxed, and rehydrated. After blocking with BSA,

and TUNEL staining, indicative of increased oxidative stress and apoptosis, respectively, was observed in the STZ group, and both were greatly reduced with the D-glucarate treatment, especially in tubules. Con, control; Cyt.C, Cytochrome C; DG, D-glucarate; G, glomerulus. *P,0.05 compared with the STZ group; #P,0.05 compared with the control group.

1318 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

fields of cells (approximately 200 cells per group) were counted to determine the percentage of cells undergoing apoptosis.

Assessment of Mitochondrial Morphology, Membrane Voltage Potential (DCm), Autophagy, and Mitophagy In Vitro: For Delineation of Mitochondrial Morphology, Cells Were Plated on Coverslips After various treatments, the live cells were in- cubated with Mitotracker dye (Life Technologies) and examined by confocal microscopy. Also, their morphology was assessed after paraformaldehyde fixation and detergent permeabilization. For each treated group, 10 fields of cells were randomly selected (.100 cells/group) to delineate the shape of mitochondria. To assess perturbations in the mitochondrial membrane potential (DCm), the cells were stained with TMRE (Life Technologies) and examined by confocal microscopy.7 Average intensity of fluores- cence from 10 randomly selected fields was mea- sured using ImageJ software (National Institutes of Health, Bethesda, MD) to evaluate DCm. For autophagy/mitophagy assessment, cells in- cubated with Mitotracker red dye were fixed and Figure 10. Schematic diagram depicting the conceivable cellular events by which MIOX permeabilized. They were successively incubated disrupts mitochondrial integrity under HG ambience, which results in renal tubular cell injury. with rabbit polyclonal LC3 antibody and secondary Under HG ambience, MIOX is upregulated in tubular cells, which then modulate mito- antibodies conjugated with Alexa Fluor. After chondrial dynamics through two conceivable pathways. (1) MIOX increases intracellular ROS counterstaining with DAPI, cells were examined. generation through transcriptional mechanisms, which promote Drp1 activation and mi- Ten random fields of cells (.100 cells/group) were tochondrial translocation.10,18,39 (2) Upregulated MIOX reduces Pink1 expression, which has evaluated to calculate the percentage of cells with a negative regulatory effect on Drp1 activity, such that Pink1 inhibition would increase punctate LC3. Drp1-dependent mitochondrial fragmentation.40 In this regard, Drp1 is activated and translocates to MOM through these possible MIOX-mediated pathways, where it interacts Detection of ROS with Fis1, which would lead to mitochondrial fragmentation. Excessively fragmented mi- To assess intracellular and mitochondrial ROS tochondria are depolarized and lose viability for the fusion process, but the surveillance of production, cells were incubated with H2-DCFDA mitochondrial quality control allows these injured organelles to be recycled by mitophagy. However, under HG, both autophagy and mitophagy are suppressed, partly because of the or MitoSOX (Life Technologies), respectively, and activation of MIOX, which inhibited Pink1 and conceivably, prevented subsequent Parkin examined by confocal microscopy. Intracellular translocation into mitochondria, thus blocking the interaction between Parkin and Mfn2 on ROSproductioninkidney tissueswasassessedusing MOM that eventually resulted in a compromise in the process of mitophagy. As a result of 4-mm-thick cryostat sections from various experi- insufficient autophagic removal of mitochondria, the accumulation of fragmented and mental groups and stained with DHE (Sigma- damaged mitochondria triggered ROS overproduction as well as the mitochondrial apo- Aldrich).47 For both cells and tissue sections, 20 ptotic pathway, by which the apoptogenic Bax translocates and inserts into mitochondria, randomly selected fields were photographed, and whereas Cytochrome C is released into the cytosol, leading to apoptosis. Interestingly, mean fluorescence intensity was calculated. pharmacologic inhibition of MIOX by D-glucarate restores mitochondrial quality control machinery by preventing mitochondrial damage and cell apoptosis, thereby attenuating hyperglycemia-induced tubular damage. Cyt.C, Cytochrome C. Animal Model System Eight-week-old ICR male mice (Harlan Co.) were used and divided into four groups (n=8): control, sections were incubated with primary antibodies and then, Alexa STZ-induced diabetic, and diabetic with 1% or 2% D-glucarate sup- Fluor-conjugated secondary antibodies. plement. D-glucarate (1% or 2% [wt/wt]) was mixed with dietary To assess the extent of apoptosis, TUNEL staining was performed on chow.42 A diabetic state was induced by daily intraperitoneal injection cells plated on coverslips and paraffin-embedded kidney sections using an of STZ (50 mg/kg body wt; Sigma-Aldrich) for 5 days. One week after In Situ Cell Death Detection Kit (Roche Applied Science).7 Ten random the injection, mice with blood glucose levels.250 mg/dl were selected

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1319 BASIC RESEARCH www.jasn.org for dietary supplementation and euthanized after 2 months. All ani- ACKNOWLEDGMENTS mal procedures used in this study were approved by the Animal Care and Use Committee of Northwestern University. This study was supported by National Institutes of Health Grant DK60635. Assessment of Physiologic Features and Renal Functions Bimonthly body weight, blood, serum, and urine collections were DISCLOSURES carried out. Blood glucose was measured using the Roche Accu-Chek None. Advantage Meter. Serum and urine creatinine and urine albumin were assessed using the QuantiChrom Creatinine and BCG Albumin Assay Kits (Bioassay Systems). Urinary ACR (milligrams per gram) was REFERENCES calculated as follows: urine albumin (milligrams per deciliter)/urine creatinine (grams per deciliter). 1. Gregg EW, Li Y, Wang J, Burrows NR, Ali MK, Rolka D, Williams DE, Geiss L: Changes in diabetes-related complications in the United Assessment of Renal Tissue Morphology States, 1990-2010. NEnglJMed370: 1514–1523, 2014 Three-micrometer-thick paraffin-embedded sections of kidneys from 2. Kanwar YS, Sun L, Xie P, Liu FY, Chen S: A glimpse of various pathogenetic – four mice groups were dewaxed in xylene and rehydrated. After mechanisms of diabetic nephropathy. Annu Rev Pathol 6: 395 423, 2011 3. Tang SC, Lai KN: The pathogenic role of the renal proximal tubular cell staining with periodic acid–Schiff, morphologic changes were as- in diabetic nephropathy. Nephrol Dial Transplant 27: 3049–3056, 2012 sessed in 20 kidney sections. Glomerular or tubular damage was an- 4. Bonventre JV: Can we target tubular damage to prevent renal function alyzed using a semiquantitative scoring system as range=0–3: 0, no decline in diabetes? Semin Nephrol 32: 452–462, 2012 glomerular/tubular damage; 1, ,25% of the glomerular/tubular area 5. Nath KA: Tubulointerstitial changes as a major determinant in the – affected; 2, 25%–50% of the glomerular/tubular area affected; 3, progression of renal damage. AmJKidneyDis20: 1 17, 1992 6. Nath KA: The tubulointerstitium in progressive renal disease. Kidney Int .50% of the glomerular/tubular area affected.48,49 54: 992–994, 1998 7. Sun L, Xie P, Wada J, Kashihara N, Liu FY, Zhao Y, Kumar D, Chugh SS, Immunohistochemistry Danesh FR, Kanwar YS: Rap1b GTPase ameliorates glucose-induced Three-micrometer-thick paraffin-embedded kidney sections from mitochondrial dysfunction. JAmSocNephrol19: 2293–2301, 2008 four groups of mice were dewaxed in xylene and rehydrated, and then, 8. Wang W, Wang Y, Long J, Wang J, Haudek SB, Overbeek P, Chang BH, Schumacker PT, Danesh FR: Mitochondrial fission triggered by hyper- they were successively incubated with various primary antibodies and glycemia is mediated by ROCK1 activation in podocytes and endo- anti-rabbit IgG conjugated with peroxidase (ImmPRESS kit; Vector thelial cells. Cell Metab 15: 186–200, 2012 Labs). Sections were treated with diaminobenzidine or NovaRed 9. Sharma K, Karl B, Mathew AV, Gangoiti JA, Wassel CL, Saito R, Pu M, substrate (Vector Labs) followed by counterstaining with Mayer Sharma S, You YH, Wang L, Diamond-Stanic M, Lindenmeyer MT, hematoxylin and examined. Forsblom C, Wu W, Ix JH, Ideker T, Kopp JB, Nigam SK, Cohen CD, Groop PH, Barshop BA, Natarajan L, Nyhan WL, Naviaux RK: Meta- bolomics reveals signature of mitochondrial dysfunction in diabetic MIOX Activity Assay kidney disease. J Am Soc Nephrol 24: 1901–1912, 2013 MIOX activity in kidney tissues was determined as previously 10. Zhan M, Brooks C, Liu F, Sun L, Dong Z: Mitochondrial dynamics: described.16,18 Briefly, cortices were homogenized in a buffer Regulatory mechanisms and emerging role in renal pathophysiology. Kidney Int 83: 568–581, 2013 (20 mM sodium acetate [pH 6.0], 2 mM L-cysteine, 1 mM ferrous 11. Youle RJ, van der Bliek AM: Mitochondrial fission, fusion, and stress. ammonium sulfate, 1 mM glutathione, and 1 mM PMSF). The homog- Science 337: 1062–1065, 2012 enate was centrifuged, and the supernatants were saved for measuring 12. Tatsuta T, Langer T: Quality control of mitochondria: Protection against protein concentration and MIOX activity by the Orcinol method. neurodegeneration and ageing. EMBO J 27: 306–314, 2008 13. Westermann B: Mitochondrial fusion and fission in cell life and death. Nat Rev Mol Cell Biol 11: 872–884, 2010 Assessment of Cellular Morphology by Electron 14. Xiao L, Zhu X, Yang S, Liu F, Zhou Z, Zhan M, Xie P, Zhang D, Li J, Song Microscopy P, Kanwar YS, Sun L: Rap1 ameliorates renal tubular injury in diabetic 3 Renal cortices were minced into 1-mm pieces and processed for elec- nephropathy. Diabetes 63: 1366–1380, 2014 tron microscopy.7 After surveying the Toluidine blue-stained EPON- 15. Yang Q, Dixit B, Wada J, Tian Y, Wallner EI, Srivastva SK, Kanwar YS: embedded sections, thin sections were prepared for electron microscopy Identification of a renal-specific oxido-reductase in newborn diabetic – to delineate the extent of mitochondrial fragmentation, autophagy, and mice. Proc Natl Acad Sci U S A 97: 9896 9901, 2000 16. Nayak B, Xie P, Akagi S, Yang Q, Sun L, Wada J, Thakur A, Danesh FR, mitophagy in tubules.28,50 Mitochondria having length.2 mm were Chugh SS, Kanwar YS: Modulation of renal-specific oxidoreductase/ fi , m considered lamentous, and those with length 1 m and spherical myo-inositol oxygenase by high-glucose ambience. Proc Natl Acad Sci configuration were designated as fragmented. USA102: 17952–17957, 2005 17. Xie P, Sun L, Oates PJ, Srivastava SK, Kanwar YS: Pathobiology of renal- specific oxidoreductase/myo-inositol oxygenase in diabetic nephrop- Statistical Analyses fi 6 athy: Its implications in tubulointerstitial brosis. Am J Physiol Renal The data were expressed as means SEMs. Comparison of the results Physiol 298: F1393–F1404, 2010 P , between two groups was evaluated by one-way ANOVA. value 0.05 18. Nayak B, Kondeti VK, Xie P, Lin S, Viswakarma N, Raparia K, Kanwar YS: was considered statistically significant. Transcriptional and post-translational modulation of myo-inositol oxygenase

1320 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 1304–1321, 2015 www.jasn.org BASIC RESEARCH

by high glucose and related pathobiological stresses. J Biol Chem 286: Maegawa H, Uzu T: Obesity-mediated autophagy insufficiency ex- 27594–27611, 2011 acerbates proteinuria-induced tubulointerstitial lesions. JAmSoc 19. Reddy CC, Pierzchala PA, Hamilton GA: myo-Inositol oxygenase from Nephrol 24: 1769–1781, 2013 hog kidney. II. Catalytic properties of the homogeneous enzyme. JBiol 36. Yamahara K, Yasuda M, Kume S, Koya D, Maegawa H, Uzu T: The role of Chem 256: 8519–8524, 1981 autophagy in the pathogenesis of diabetic nephropathy. J Diabetes 20. Bhattacharya S, Manna P, Gachhui R, Sil PC: D-saccharic acid 1,4-lactone Res 2013: 193757, 2013 protects diabetic rat kidney by ameliorating hyperglycemia-mediated 37. Xu X, Kobayashi S, Chen K, Timm D, Volden P, Huang Y, Gulick J, Yue Z, oxidative stress and renal inflammatory cytokines via NF-kB and PKC Robbins J, Epstein PN, Liang Q: Diminished autophagy limits cardiac signaling. Toxicol Appl Pharmacol 267: 16–29, 2013 injury in mouse models of type 1 diabetes. J Biol Chem 288: 18077– 21. Kabeya Y, Mizushima N, Ueno T, Yamamoto A, Kirisako T, Noda T, 18092, 2013 Kominami E, Ohsumi Y, Yoshimori T: LC3, a mammalian homologue of 38. Scheele C, Nielsen AR, Walden TB, Sewell DA, Fischer CP, Brogan RJ, yeast Apg8p, is localized in autophagosome membranes after pro- Petrovic N, Larsson O, Tesch PA, Wennmalm K, Hutchinson DS, cessing. EMBO J 19: 5720–5728, 2000 Cannon B, Wahlestedt C, Pedersen BK, Timmons JA: Altered regula- 22. Twig G, Elorza A, Molina AJ, Mohamed H, Wikstrom JD, Walzer G, tion of the PINK1 locus: A link between type 2 diabetes and neuro- Stiles L, Haigh SE, Katz S, Las G, Alroy J, Wu M, Py BF, Yuan J, Deeney degeneration? FASEB J 21: 3653–3665, 2007 JT, Corkey BE, Shirihai OS: Fission and selective fusion govern mito- 39. Makino A, Scott BT, Dillmann WH: Mitochondrial fragmentation and chondrial segregation and elimination by autophagy. EMBO J 27: 433– superoxide anion production in coronary endothelial cells from a 446, 2008 mouse model of type 1 diabetes. Diabetologia 53: 1783–1794, 2010 23. Chen Y, Dorn GW 2nd: PINK1-phosphorylated mitofusin 2 is a Parkin re- 40. Lutz AK, Exner N, Fett ME, Schlehe JS, Kloos K, Lämmermann K, ceptor for culling damaged mitochondria. Science 340: 471–475, 2013 Brunner B, Kurz-Drexler A, Vogel F, Reichert AS, Bouman L, Vogt- 24. Kumar D, Robertson S, Burns KD: Evidence of apoptosis in human di- Weisenhorn D, Wurst W, Tatzelt J, Haass C, Winklhofer KF: Loss of abetic kidney. Mol Cell Biochem 259: 67–70, 2004 parkin or PINK1 function increases Drp1-dependent mitochondrial 25. Brezniceanu ML, Liu F, Wei CC, Tran S, Sachetelli S, Zhang SL, Guo DF, fragmentation. JBiolChem284: 22938–22951, 2009 Filep JG, Ingelfinger JR, Chan JS: Catalase overexpression attenuates 41. Walaszek Z, Hanausek M, Narog M, Raich PC, Slaga TJ: Mechanisms of angiotensinogen expression and apoptosis in diabetic mice. Kidney Int lung cancer chemoprevention by D-glucarate. Chest 125[Suppl]: 149S– 71: 912–923, 2007 150S, 2004 26. Lin M, Yiu WH, Wu HJ, Chan LY, Leung JC, Au WS, Chan KW, Lai KN, 42. Zoltaszek R, Kowalczyk P, Kowalczyk MC, Hanausek M, Kilianska ZM, Tang SC: Toll-like receptor 4 promotes tubular inflammation in diabetic Slaga TJ, Walaszek Z: Dietary D-glucarate effects on the biomarkers of nephropathy. JAmSocNephrol23: 86–102, 2012 inflammation during early post-initiation stages of benzo[a]pyrene- 27. Lin M, Yiu WH, Li RX, Wu HJ, Wong DW, Chan LY, Leung JC, Lai KN, induced lung tumorigenesis in A/J mice. Oncol Lett 2: 145–154, 2011 Tang SC: The TLR4 antagonist CRX-526 protects against advanced 43. Yoshimi N, Walaszek Z, Mori H, Hanausek M, Szemraj J, Slaga TJ: In- diabetic nephropathy. Kidney Int 83: 887–900, 2013 hibition of azoxymethane-induced rat colon carcinogenesis by potas- 28. Brooks C, Wei Q, Cho SG, Dong Z: Regulation of mitochondrial dy- sium hydrogen D-glucarate. Int J Oncol 16: 43–48, 2000 namics in acute kidney injury in cell culture and rodent models. JClin 44. Hanausek M, Walaszek Z, Slaga TJ: Detoxifying cancer causing agents Invest 119: 1275–1285, 2009 to prevent cancer. Integr Cancer Ther 2: 139–144, 2003 29. Che R, Yuan Y, Huang S, Zhang A: Mitochondrial dysfunction in the 45. Saluk-Juszczak J: A comparative study of antioxidative activity of pathophysiology of renal diseases. Am J Physiol Renal Physiol 306: calcium-D-glucarate, sodium-D-gluconate and D-glucono-1,4-lactone F367–F378, 2014 in a human blood platelet model. Platelets 21: 632–640, 2010 30. Motori E, Puyal J, Toni N, Ghanem A, Angeloni C, Malaguti M, Cantelli- 46. Bhattacharya S, Manna P, Gachhui R, Sil PC: D-saccharic acid-1,4- Forti G, Berninger B, Conzelmann KK, Götz M, Winklhofer KF, Hrelia S, lactone ameliorates alloxan-induced diabetes mellitus and oxidative Bergami M: Inflammation-induced alteration of mitochon- stress in rats through inhibiting pancreatic b-cells from apoptosis via drial dynamics requires autophagy for mitochondrial network mainte- mitochondrial dependent pathway. Toxicol Appl Pharmacol 257: nance. Cell Metab 18: 844–859, 2013 272–283, 2011 31. Hoshino A, Mita Y, Okawa Y, Ariyoshi M, Iwai-Kanai E, Ueyama T, Ikeda 47. Xie P, Sun L, Nayak B, Haruna Y, Liu FY, Kashihara N, Kanwar YS: C/EBP- K, Ogata T, Matoba S: Cytosolic p53 inhibits Parkin-mediated mi- beta modulates transcription of tubulointerstitial nephritis antigen in tophagy and promotes mitochondrial dysfunction in the mouse heart. obstructive uropathy. J Am Soc Nephrol 20: 807–819, 2009 Nat Commun 4: 2308, 2013 48. Sun L, Zhang D, Liu F, Xiang X, Ling G, Xiao L, Liu Y, Zhu X, Zhan M, Yang 32. Green DR, Galluzzi L, Kroemer G: Mitochondria and the autophagy- Y, Kondeti VK, Kanwar YS: Low-dose paclitaxel ameliorates fibrosis in inflammation-cell death axis in organismal aging. Science 333: 1109– the remnant kidney model by down-regulating miR-192. J Pathol 225: 1112, 2011 364–377, 2011 33. Periyasamy-Thandavan S, Jiang M, Wei Q, Smith R, Yin XM, Dong Z: 49. Tervaert TW, Mooyaart AL, Amann K, Cohen AH, Cook HT, Autophagy is cytoprotective during cisplatin injury of renal proximal Drachenberg CB, Ferrario F, Fogo AB, Haas M, de Heer E, Joh K, Noël tubular cells. Kidney Int 74: 631–640, 2008 LH, Radhakrishnan J, Seshan SV, Bajema IM, Bruijn JA; Renal Pathology 34. Fang L, Zhou Y, Cao H, Wen P, Jiang L, He W, Dai C, Yang J: Autophagy Society: Pathologic classification of diabetic nephropathy. JAmSoc attenuates diabetic glomerular damage through protection of hyper- Nephrol 21: 556–563, 2010 glycemia-induced podocyte injury. PLoS ONE 8: e60546, 2013 50. Jiang M, Liu K, Luo J, Dong Z: Autophagy is a renoprotective mecha- 35. Yamahara K, Kume S, Koya D, Tanaka Y, Morita Y, Chin-Kanasaki M, nism during in vitro hypoxia and in vivo ischemia-reperfusion injury. Am Araki H, Isshiki K, Araki S, Haneda M, Matsusaka T, Kashiwagi A, J Pathol 176: 1181–1192, 2010

J Am Soc Nephrol 26: 1304–1321, 2015 MIOX Disrupts Mitochondria in DKD 1321