The Impact of Stem Cells on Electron Fluxes,Proton Translocation,and ATP Synthesis in Kidney Mitochondria after Ischemia/Reperfusion

Abstract

Tissue damage by ischemia/reperfusion (I/R) results from a temporary cessation of blood flow followed by the restoration of circulation. The injury depresses mitochondrial respiration, increases the production of reactive oxygen species (ROS), decreases the mitochondrial transmembrane potential, and stimulates invasion by inflammatory cells. The primary objective of this work was to address the potential use of bone marrow stem cells (BMSCs) to preserve and restore mitochondrial function in the kidney after I/R. Mitochondria from renal proximal tubule cells were isolated by differential centrifugation from rat kidneys subjected to I/R (clamping of renal arteries followed by release of circulation after 30 min), without or with subcapsular administration of BMSCs. Respiration starting from mitochondrial complex II was strongly affected following I/R. However, when BMSCs were injected before ischemia or together with reperfusion, normal electron fluxes, electrochemical gradient for protons, and ATP synthesis were almost completely preserved, and mitochondrial ROS formation occurred at a low rate. In homogenates from cultured renal cells transiently treated with antimycin A, the coculture with BMSCs induced a remarkable increase in protein S-nitrosylation that was similar to that found in mitochondria isolated from I/R rats, evidence that BMSCs protected against both superoxide anion and peroxynitrite. Labeled BMSCs migrated to damaged tubules, suggesting that the injury functions as a signal to attract and host the injected BMSCs. Structural correlates of BMSC injection in kidney tissue included stimulus of tubule cell proliferation, inhibition of apoptosis, and decreased inflammatory response. Histopathological analysis demonstrated a score of complete preservation of tubular structures by BMSCs, associated with normal plasma creatinine and urinary osmolality. These key findings shed light on the mechanisms that explain, at the mitochondrial level, how stem cells prevent damage by I/R. The action of BMSCs on mitochondrial functions raises the possibility that autologous BMSCs may help prevent I/R injuries associated with transplantation and acute renal diseases.

Keywords

Kidney mitochondria Bone marrow stem cells (BMSCs)Ischemia/reperfusion (IR)Mitochondrial respiration ATP synthesis

Introduction

Ischemia is a crucial event during intrinsic acute kidney injury (AKI), and kidneys of donors are inadvertently exposed to ischemia before transplantation, especially when cardiac arrest precedes removal of the organ (16,34). After recovery of renal blood flow, reperfusion injury is superimposed on the previous insult from ischemia. Generation of reactive oxygen species (ROS) and the onset of apoptosis are the major precursors of the process known as ischemia/reperfusion (I/R) lesion (3,16). In this injury, organs like the kidney that have a high metabolic (aerobic) demand are especially affected. When O₂ is restored, electrons flowing from oxidizing reactions encounter the components of mitochondrial respiration in a reduced state. Therefore, the electron transfer side reactions to O₂ to generate ROS are favored, and a cascade that leads to cell death is also activated (19,23). In these conditions, the long-term success of AKI treatment and of transplantation must rely on preservation of mitochon-drial function or its recovery.

The potential of stem cells in the repair or prevention of I/R injury has been recognized in several studies during the last decade (12,28,41,42). However, the molecular interactions underlying the benefits of cell therapy at mitochondrial and submitochondrial levels in the kidneys and other organs are not known. The present work investigates the actions of subcapsularly injected isogenic bone marrow-derived stem cells (BMSCs), before ischemia or at the moment of re perfusion, on four aspects of renal mitochondrial function: electron fluxes, generation of the electrochemical potential gradient for H+, adenosine-5′-triphosphate (ATP) synthesis, and ROS generation. We also investigated the impact of BMSCs on protein S-nitrosylation in (i) cultured kidney cells (LLC-PK₁ lineage) after transient respiration blockade with antimycin A, washing, and reexposure to O2 and (ii) in mitochondria isolated from renal cortex. The mitochondrial functional results are correlated with the BMSC-induced cell proliferation, antiapoptotic effects, reduced inflammatory response, recovery of tubular lesions, and restoration of creatinine plasma levels and urine concentrating capacity observed after reperfusion.

Materials and Methods

Animal Care

Male Wistar rats were purchased from Bio Campo 2000 Biological Products Ltd. (Bom Jardin, RJ, Brazil), and the animals' health was certified by a licensed veterinarian. They were maintained under constant temperature (23 + 2°C) and kept in a 12-h light/dark cycle and fed standard rat diet (Labina, Purina Agribrands, Paulinia, SP, Brazil) and filtered water ad libitum. The rats were anesthetized with ether (Vetec-Sigma Aldrich, Duque de Caxias, RJ, Brazil; preparation of BMSCs) or by intraperitoneal injection of xylazine (Bayer S.A., São Paulo, SP, Brazil; 5 mg/kg) and ketamine (Cristália, Itapira, SP, Brazil; 50 mg/kg) in the I/R experiments. All experimental procedures were approved by the Committee for Ethics in Animal Experimentation (Federal University of Rio de Janeiro, protocol IBCCF 087) and were carried out in accordance with the Committee's guidelines, which follow the requirements for manuscripts submitted to biomedical journals.

Ischemia/Reperfusion of Rat Kidneys and Mitochondrial Isolation

The ischemia/reperfusion (I/R) model was that used by Benítez-Bribiesca et al. (3) except that rats were Wistar (male, aged 2 months, weighing 180-200 g) and renal arterial clamping (30 min) was bilateral. In the treated group, BMSCs (107 cells in saline) were administered subcapsularly in both kidneys, before ischemia (BMSCs + I/R group) or at the beginning of reperfusion (I/R + BMSCs group), as described by Cavaglieri et al. (6), whereas the other groups (control CTR and I/R) received an equal volume of saline. After 24-h reperfusion, the kidneys were removed, and mitochondria were isolated by differential centrifugation (40) from the external cortical region (cortex corticis) where more than 90% of the cell population corresponds to proximal tubules (43). Briefly, except when otherwise indicated, kidneys were removed 24 h after the beginning of reper-fusion (in the I/R, BMSCs + I/R, and I/R + BMSCs groups) or after the simulation of surgical manipulation (in sham-operated rats), collected on ice, and immediately immersed in a solution containing 250 mM sucrose (Vetec-Sigma Aldrich), 10 mM HEPES-KOH (pH 7.4; Sigma-Aldrich, Saint Louis, MO, USA), 2 mM EGTA (Sigma-Aldrich), and 0.15 mg/ml trypsin inhibitor (Sigma-Aldrich) (solution A). Kidneys were cut into thin slices with the aid of a Stadie-Riggs microtome (Thomas Scientific, Swedesboro, NJ, USA), and the cortex corticis was carefully dissected using iridectomy scissors (Insight, Ribeirão Preto, S P, Brazil). The slices were suspended in 10 ml of solution A and manually homogenized using a glass homogenizer provided with a Teflon pestle (both Sigma-Aldrich). The homogenate was centrifuged for 10 min at 600× g (4°C) to sediment cell debris and unbroken cells, and the supernatant was centrifuged again at 12,000 × g for 10 min. The sediment was resuspended in 10 ml of solution A, gently homogenized, and centrifuged again at 12,000× g for 10 min. The sedimented mitochondria were resuspended in 0.3 ml of solution A and used immediately.

BMSC Preparation and Characterization

Bone marrow was obtained by flushing with saline the femurs and tibias of male Wistar rats, aged 2 months and weighing 150-300 g. BMSCs were isolated using a Ficoll gradient (Science Pro, São Caetano do Sul, SP, Brazil) as described elsewhere (1), resuspended in low glucose and serum-free Dulbecco's modified Eagle's medium (DMEM; Gibco-Life Technologies, Itapevi, SP, Brazil)), counted in a Neubauer chamber (HBG, Giessen, Germany), and used immediately. Their viability was monitored by trypan blue (Vetec-Sigma Aldrich) exclusion, and control phe-notypes were evaluated by flow cytometry using a FACS Aria apparatus (BD Biosciences, Franklin Lakes, NJ, USA) as described previously (33). For immunopheno-typing, rat BMSCs (3× 10⁶ cells) were fixed with Transfix (Immunostep, Salamanca, Spain) for at least 1 day. Before immunoassay, the samples were washed in phosphate-buffered saline (PBS; Na⁺-phosphate buffer and NaCl from Vetec-Sigma Aldrich) and incubated for 10 min with 3 μl FcR blocking buffer (BD Biosciences). Approximately 3× 1 0⁵ cells were used in different tubes to characterize each subpopulation: (i) T helper lymphocytes [cluster of differentiation 45 (CD45)/CD4], (ii) T cytotoxic lymphocytes (CD45/CD8), (iii) monocytes (CD45/CD11b/c/ CD29) and granulocytes (CD45/CD11b/c/CD29), and (iv) precursors (CD45/CD34/CD90.1). The samples were incubated for 20 min at room temperature with the following monoclonal antibodies: pure CD45 (0.25 μg) or an anti-rat biotinylated CD45 (0.5 μg), CD29–phycoeyrthrin-cyanine 7 (PE-Cy7; 0.25 μg) (BD Pharmingen, San Diego, CA, USA); CD4–fluorescein isothiocyanate (FITC; 0.5 μg), CD8–FITC (0.5 μg), CD11b/c–FITC (0.5 μg), CD90.1– FITC (0.5 μg) (Caltag Laboratories, Bangkok, Thailand); CD34–PE (1.0 μg) (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA). The samples were washed with PBS, and a secondary antibody [goat anti-mouse IgG–Cy5 (0.2 μg) (Invitrogen, Grand Island, N Y, USA)] was used to detect CD45⁺ cells when pure CD45 was used as a primary antibody. With the use of the other protocol, CD45⁺ cells were detected after incubation of the samples with the anti-rat CD45 biotinylated antibody for 20 min, followed by washing with PBS and incubation with streptavidin–allophyco-cyanine (APC) (R&D Systems, Minneapolis, MN, USA) for 20 min at 4°C. This population was then tested for the subpopulations shown in Figure 1. After incubation, 1 ml of lysing solution (BD FACS lysing solution; BD Biosciences) was added to samples in order to eliminate red cells, and the remaining cells were washed with PBS before acquisition of the events. DAPI (4′,6-diamidino-2-phenylindole, dihy-drochloride; Invitrogen) was used to distinguish small cells from debris. A total of 50,000 events were acquired, and their analysis was performed using the FlowJo v.7.6.3 flow cytometry analyzer software (Ashland, OR, USA). The representative flow cytometry analysis depicted in Figure 1 shows the phenotype of the BMSC population used, which contained 95% CD45⁺ cells, 54% CD11⁺ CD29^–, 13% CD11^+low CD29⁺, and 15% CD34^– CD90.1⁺, with very low levels of cells having the other surface antigen markers.

Figure 1.

Phenotype characterization of the subcapsularly injected bone marrow stem cells (BMSCs). (A) Representative flow cytometry analysis showing the cluster of differentiation 45 (CD45⁺) cell population (using biotinylated CD45–APC) from which the other cells were characterized. (B) T helper lymphocytes. (C) T cytotoxic lymphocytes. (D) Granulocytes and monocytes. (E) Hematopoietic stem cells. (F) Immature precursors. The corresponding percentage values and the antigens are indicated in the panels within the delimited areas (see also Materials and Methods). SSC, side scatter; APC, allophycocyanine; FITC, fluorescein isothiocyanate; PE, phycoerythrin.

BMSC Tracing After Subcapsular Injection

BMSCs were labeled and fixed using the amine-reactive CellTraceTM Far Red DDAO-succinimidyl ester (SE) (Invitrogen), following the manufacturer's instructions. Briefly, BMSCs were incubated with CellTraceTM (2.5 ng/ μl in ~10⁶ cells) in DMEM without serum for 40 min (37°C, 5% CO₂). After this period, the cells were washed three times with fresh medium to ensure that no free CellTraceTM would be injected. The labeled cells (10⁷) were injected subcapsularly 1 h before the 30-min period of ischemia, the rats were killed 24 h after the beginning of reperfusion, and the kidneys were removed and treated as previously described (1). For the 3D reconstructions, 40-65 Z-stack images per section were collected using AxioVision 4.8.2 software in an ApoTome microscope (ApoTome Axion Imager. M2, Carl Zeiss Inc., Jena, Germany), which was used to visualize Far Red and the nuclei labeled with DAPI.

Mitochondrial Respiration Measurements

Mitochondrial functions were examined using succi-nate, which after oxidation to fumarate generates reduced flavin adenine dinucleotide (FADH₂), the electron donor for respiratory complex II (succinate dehydrogenase). This complex controls the ubiquinone reduction state— and therefore the downstream electron fluxes—also playing a crucial role in ROS generation and handling (19,23). Oxygen consumption was measured using high-resolution respirometry (Oroboros Oxygraph-O2K, Innsbruck, Austria) (11). Respiration was assayed at 37°C by incubating a mitochondrial suspension (0.2 mg/ml of protein) in a medium containing 320 mM mannitol (Vetec-Sigma Aldrich), 4 mM MgCl₂ (Merck, Darmstadt, Germany), 0.08 mM EDTA (GE Healthcare Life Sciences, São Paulo, SP, Brazil), 10 mM Tris-HCl (pH 7.4; Sigma-Aldrich), 8 mM phosphate-Tris (Na+-phosphate and Tris from Vetec-Sigma Aldrich), 10 mM succinate (substrate for complex II; Sigma-Aldrich), 0.001 mM rotenone (Sigma-Aldrich), and 0.1% (w/v) fatty acid-free bovine serum albumin (BSA; Sigma-Aldrich). When required, the respiratory medium was supplemented with 150 μM adenosine diphosphate (ADP; Sigma-Aldrich; to stimulate respiration in a condition in which ATP synthesis occurs) or 1 mM of the oxidative phosphorylation uncoupler carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP; Sigma-Aldrich) to measure respiration in a condition where an H⁺ electrochemical gradient is not formed and electron fluxes are faster and uncoupled.

Recording of Mitochondrial Membrane Potential

The mitochondrial membrane potential was monitored by measuring the fluorescence quenching of safranine O (7 μM; Sigma-Aldrich) (17), at excitation and emission wavelengths of 495 and 586 nm, respectively, using a Hitachi F-4500 fluorometer (Hitachi Ltd., Tokyo, Japan). The assays contained the same solution used for respiration measurements and were carried out at 37°C. When indicated, pulses of ADP were added to investigate whether the potential was utilized for ATP synthesis, and FCCP was added at the end of recordings to visualize the complete collapse of the mitochondrial membrane potential.

Evaluation of Reactive Oxygen Species

Mitochondrial ROS production was evaluated fluoro-metrically (18) by following the oxidation of Amplex® Red (25 μM; Invitrogen) to resorufin in the presence of horseradish peroxidase. The wavelengths used were 563 nm (excitation) and 587 nm (emission), according to the manufacturer's instructions. The assay components and temperature were as in the previous section. The medium was supplemented with 3 μM superoxide dismutase (Sigma-Aldrich) to catalyze the dismutation of anion superoxide in O₂ and H₂O₂ and 10 U/ml peroxidase (Sigma-Aldrich) to catalyze the oxidation of Amplex Red coupled with simultaneous reduction of H₂O₂ in a 1:1 stoichiometry.

S-Nitrosylation Assays in Renal Cells Subjected to Respiration Blockade and Reexposure to Oxygen

Analysis of S-nitrosothiol formation was conducted in lysates from immortalized kidney proximal tubule cells of 3- to 4-week-old male porcine origin (LLC-PK1) and in mitochondria isolated from the proximal tubules that comprise the majority of the cortex corticis cell population (43). The cells (from the American Type Culture Collection, Manassas, VA, USA) were a gift from Dr. Celso Caruso-Neves, and the mitochondria were isolated as described above. The cells were cultured in DMEM supplemented with fetal bovine serum (10%; Gibco) using a plate with six wells (TPP, Trasadingen, Switzerland; 2 ×10₇ cells per well). They formed a firmly attached monolayer and after 24 h were separated into three groups. The control group (CTR) remained under standard conditions in the same medium for an additional period of 24 h, in an atmosphere of 95% air plus 5% CO₂. The second group (HYP) was subjected to chemical hypoxia for 30 min by adding 1 μM antimycin A (Sigma-Aldrich), thus blocking mitochondrial respiration at the level of respiratory complex III, impairing utilization of most endogenous substrate, and leading to ATP depletion (8,22). After removal of antimycin and washing with PBS (saline buffered with 1.7 mM phosphate; pH 7.5), the cells were cultured in DMEM plus fetal bovine serum for an additional period of 24 h in the same gaseous atmosphere. The third group was also subjected to chemical hypoxia for 30 min (1 μM antimycin A), washed, and placed again in DMEM plus serum and BMSCs (10⁶ cells in 2 ml of medium). Both types of cells (LLC-PK₁ and BMSCs) were cocultured for an additional period of 24 h in the gaseous atmosphere above, using a two-compartment miniwell system (Millipore Co., Billerica, MA, USA) that prevents physical contact between the two cell populations. However, possible mutual actions of secreted soluble factors (22) are allowed through a porous membrane (0.4 mm pore diameter) that separates the two compartments of the system. The culture medium was carefully removed, and 2 ml of PBS was added. After gentle manual shaking, the PBS solution was also carefully removed by aspiration. The still firmly attached kidney cells were trypsinized for 2 min by adding 100 μl of a trypsin solution (Gibco Life Technologies) into each well after removal of the upper chamber (containing BMSCs in the third experimental group). The trypsin solution contained 5 g trypsin/L, 5.4 mM EDTA (disodium salt; GE Healthcare Life Sciences), and 145 mM NaCl, adjusted to pH 7.4 with NaOH (both Vetec-Sigma Aldrich). This short exposure to trypsin allowed a complete detachment without damage of the cells, which were immediately suspended in 2 ml of cold DMEM with serum to stop trypsinization. The mitochondria were manually homogenized in a glass homogenizer using a Teflon pestle and lysed with the trypsin solution described above. The lysed suspensions of cells from cultures and the mitochondria isolated from proximal tubule cells were then transferred to conical tubes (Jet Biofil, Guangzhou, China) and centrifuged at 160×g for 2 min in a clinical centrifuge. The final sediments were recovered and mixed with 1 ml of lysis buffer (Sigma-Aldrich) for processing as previously described (32) using the Griess-Saville method and separating high- and low-molecular weight S-nitrosylated species (36).

Cell Proliferation, Apoptosis, and Macrophage Surface Density

These cellular parameters were studied in cortical fragments obtained from kidneys of sham-operated, I/R, and BMSCs + I/R rats. These groups were as described above for respiration studies. Kidneys were removed, fixed, and analyzed for cell proliferation and macrophage infiltration using the antibodies and the procedures described elsewhere and briefly below (1,22). Apoptotic terminal dUTP nick-end labeling (TUNEL)-positive cells were detected using the kit ApopTag (Chemicon International, Temecula, CA, USA) according to the manufacturer's instructions. Paraffin-embedded sections of kidneys were used for immunodetection of macrophages using a mouse monoclonal antibody against rat ED1 (CD68; AbD Serotec, Raleigh, NC, USA) and of proliferating cell nuclear antigen (PCNA) using a monoclonal antibody from Dako (Carpinteria, CA, USA). Antibodies were visualized using diaminobenzidine (Sigma-Aldrich) and the Dako LSAB® 2 system HRP kit (Dako). Images from 30 fields in cortical slices from each group were captured randomly, and the number of cells that were positive for PCNA, ED1, and TUNEL was obtained by manual counting using the Image-Pro Plus program (Media Cybernetics, São Paulo, SP, Brazil).

Histological Score of Tubular Lesions

The kidneys were fixed, dehydrated, and embedded in paraffin as described previously (1). Cortical slices (7 μm thick) were stained with hematoxylin and eosin (Sigma-Aldrich) and observed under light microscopy. Images from 30-35 fields from each group were captured randomly, and an average of 20 tubules were analyzed for tubular dilation, apical cytoplasm vacuolization, cell detachment, brush border integrity, denuded basement membrane, and tubular necrosis, which characterizes proximal tubular injury. A lesion score (mean±SE) was assigned in a blinded manner by a single observer (HJVB) using a 5-point scale, which is a modification from that described in Jung et al. (14). The score and the parameters were as follows: 1, without alterations; 2, mild alterations affecting 10%—or fewer—tubules; 3, alterations affecting 25% tubules; 4, alterations affecting 50% tubules; and 5, alterations in 75% or more tubules.

Creatinine Plasma Levels and Urine Osmolality

Rats from each group were placed in metabolic cages for 24 h (photoperiod of 12 h) at 22-23°C and were given a commercial chow (Labina) and water ad libitum. After this adaptive period, the rats were operated to obtain three of the four groups described above (CTR, I/R, and BMSCs + I/R) and returned to the cages. After another period of 24 h, plasma and urine samples were collected. Plasma creatinine was determined spectropho-tometrically by the alkaline picrate method using a laboratory kit (Analisa, Gold Analisa Diagnostics Ltd., Belo Horizonte, MG, Brazil). Urine osmolality was assessed using a cryoscopic osmometer (Osmomat 030, Gonotec, Berlin, Germany).

Statistical Analysis

Except when otherwise indicated, data are presented as mean±SE. One-way ANOVA followed by Tukey's multiple comparison test, Kruskal-Wallis test followed by Dunn's multiple comparison test, or linear regression was used for the statistical analysis of the data, as detailed in the corresponding figure legends. Differences were considered significant at p < 0.05.

Results

Localization of Tracked BMSCs in the Cortex

Twenty-four hours after the reperfusion release, the distribution of Far Red-labeled BMSCs in the cortex was as shown in Figure 2. In Figure 2A it can be seen, at lower resolution, that there are labeled regions that correspond to injected BMSCs and unlabeled regions that correspond to the original tubular cell population. Nuclei labeled with DAPI (blue) are prominent in both regions. At a higher magnification (Fig. 2B) it can be seen that some tubular structures are decorated by fluorescent BMSCs (red) and others are not. The tubular shape of these structures was confirmed when the images were tridimensionally captured and visualized from the x, y, and z axes (Fig. 2B).

Figure 2.

Tracing BMSCs in cortical sections. (A) Panoramic view of cortical sections indicating the presence of BMSCs in some tubules. White arrow indicates a region that corresponds to labeled injected BMSCs; yellow arrow indicates an unlabeled region that corresponds to the original cell population. (B) S-stack view showing labeled cells that are present along the tubule (☆ labeled BMSCs; # unlabeled epithelial tubule cells).

Defining the Effective Window for the Beneficial Impact of BMSC Administration in I/R

The experiments shown in Figure 3 aimed to define the optimal timing available for producing a possible beneficial influence on renal mitochondrial functions by administering BMSCs before ischemia. Figure 3A presents the respiratory rate without ADP in sham-operated rats (CTR) and after a 30-min ischemic period followed by 1 h reperfusion (I/R), where the pronounced inhibition promoted by the injury can be seen. Figure 3B demonstrates the effectiveness of BMSC injection at different times before ischemia, all evaluated 24 h after the beginning of reperfusion. Protection became evident with infusion of cells 2 h before ischemia, reached its maximum (preservation of the control levels) at 1 h, and disappeared if the BMSCs were given only 30 min before clamping of the arteries. Interestingly, a complete preservation of respiration was also encountered when cells were administered simultaneously with restoration of the renal blood flow (Fig. 3C).

Figure 3.

Critical window for an effective administration of BMSCs before ischemia. Oxygen consumption (QO₂). Mitochondria were isolated 1 h after the beginning of reperfusion in (A). In (B) and (C), mitochondria were isolated after 24 h of reperfusion. (A) Mitochondrial respiration in sham-operated rats (CTR) and in rats subjected to ischemia/reperfusion (I/R). (B) Mitochondrial respiration in rats that received subcapsular BMSCs before ischemia at the times indicated on the abscissa. (C) Mitochondrial respiration in rats that received BMSCs together with restoration of circulation after a 30-min ischemia. Data are mean ± SE (n = 6). Different lowercase letters above the bars indicate statistical differences (p < 0.05; one-way ANOVA followed by Tukey's multiple comparison test).

Slow Respiration Recovery Following BMSC Administration

The full recovery of mitochondrial respiration following reperfusion was the culmination of a slow process. When BMSCs were administered 1 h before ischemia (Fig. 4A) or at the beginning of reperfusion (Fig. 4B), respiration remained depressed by the I/R injury after 30 min and 1 h of reperfusion and only recovered 24 h after circulation was restored (as shown previously in Fig. 3C).

Figure 4.

Maximal recovery of mitochondrial respiration 24 h after BMSC administration is not an immediate process. QO₂ was measured using mitochondria isolated from kidneys of rats that received BMSCs 1 h before ischemia (A) or at the beginning of reperfusion (B). CTR, sham-operated rats; I/R, respiration measured 1 h after the beginning of reperfusion without BMSC injection. Other bars: mitochondrial respiration measurements were carried out at the times after reperfusion indicated on the abscissae. Data are mean ± SE (n = 6). Different lowercase letters above the bars indicate statistical difference (p < 0.05; one-way ANOVA followed by Tukey's multiple comparison test).

BMSCs Preserve Mitochondrial Respiration in Different States After I/R

Figure 5A depicts respiration of mitochondria isolated from rat kidney proximal tubule cells in control conditions (sham-operated animals), now assayed in nonphos-phorylating and phosphorylating conditions (without and with ADP to allow ATP synthesis, respectively), and when electron fluxes were uncoupled by addition of FCCP. The respiration of mitochondria isolated from renal tissue 24 h after a 30-min occlusion of both renal arteries followed by complete restoration of blood flow decreased to a very low value, and there was no response to ADP or to the uncou-pler FCCP (Fig. 5B). The profile of a crippled respiration in the three conditions was reversed completely if BMSCs were injected into kidneys 1 h before the injury (Fig. 5C) or at the moment of reperfusion (Fig. 5D). Thus, with BMSCs, electron fluxes were maintained at control levels despite the injury, that is, regardless of whether respiration was coupled or not to ATP synthesis or fully uncoupled by FCCP. The influence of BMSCs on the response of QO₂ to ADP was measured by the respiration control ratio, RCR (RCR = QO₂ in the presence of ADP/QO₂ in the absence of ADP). The values (mean ± SE) were 1.48 ± 0.11 (control), 1.14 ± 0.04 (I/R), 1.49 ± 0.10 (BMSCs before I/R), and 1.48 ± 0.10 (BMSCs at the moment of reperfusion). The value obtained in the I/R condition was statistically different (p < 0.05) from the RCR values found in the other three conditions (one-way ANOVA followed by Tukey's multiple comparison test).

Figure 5.

BMSC administration preserves normal QO₂ by kidney mitochondria following I/R injury in the absence or presence of ADP and in the presence of the uncoupler FCCP. QO₂ was measured using mitochondria isolated 24 h after initiation of reperfusion. White bars: respiratory state after adenosine diphosphate (ADP) is totally converted to adenosine triphos-phate (ATP); gray bars: initial rate of respiration during ATP synthesis after an ADP pulse of 150 mM; black bars: uncoupled state of mitochondrial respiration [1 μM carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP)]. (A) QO₂ by mitochondria of sham-operated rats (control). (B) QO₂ strongly inhibited by I/R with no differences among the different respiration conditions (nonphosphorylating, phosphorylating, and uncoupled). (C) Administration of BMSCs by subcapsular injection 1 h before ischemia led to preservation of the normal QO₂ after 24 h in all respiratory conditions. (D) The same pattern of preservation was observed when BMSCs were administered at the moment of reperfusion. Data are mean ± SE (n = 4 in all conditions). Different lowercase letters above the bars indicate statistically different means (p < 0.05), as assessed by one-way ANOVA followed by Tukey's multiple comparison test.

Preservation of Mitochondrial Bioenergetic Parameters by BMSCs

The electrochemical gradient for H⁺, required for ATP synthesis (13,20), was almost completely abolished by I/R, and the response to small additions of ADP was suppressed (Fig. 6A, upper dashed trace; compare with control, bottom continuous trace). The H⁺ electrochemical potential and its utilization for ATP synthesis after ADP additions were also preserved to a great extent in the presence of BMSCs (Fig. 6A, middle dotted trace). Preservation of respiratory coupling by BMSCs was confirmed in Figure 6B. The P/O ratio (the ATP synthesized per oxygen atom reduced by the respiratory chain) was maintained in mitochondria isolated from the BMSC-treated kidneys and from controls. The observed ratio (~1.7) is typical for succinate as substrate (13). Since respiration in mitochondria from the I/R group was not stimulated by ADP (Fig. 5B), the P/O ratio could not be determined in this condition (ND) (Fig. 6B).

Figure 6.

BMSCs preserve the transmembrane electrochemical H⁺ gradient and the ATP synthesis capacity following I/R. (A) Evaluation of the transmembrane electrical potential by measurement of safranine O fluorescence quenching after addition of a mito-chondrial suspension “M” and 10 mM succinate, successive pulses of ADP at the micromolar concentrations shown above the arrows and, finally, 1 μM FCCP. The upper trace (dashed) corresponds to a measurement in the I/R condition, showing the slow formation of a practically collapsed H⁺ gradient upon addition of succinate. The middle (dotted) trace was obtained using mitochondria isolated from kidneys treated with BMSCs 1 h before ischemia; the magnitude of the H⁺ gradient is comparable to that found in mitochondria from sham-operated rats (lower continuous trace). The rate of gradient formation and its partial transient collapse after ADP (when the gradient is used to energize ATP synthesis) are similar in the BMSC and sham-operated groups. Each trace is representative of six independent experiments. (B) The efficiency of phosphorylation capacity (P/O ratio) (10) is almost totally suppressed in the I/R group (ND) since the response to ADP was abolished (Fig. 5B) and preserved by BMSCs administered 1 h before ischemia (gray bar) to the same level as that found in the control (CTR) sham-operated rats (empty bar). Results are mean ± SE (n = 4 in all conditions); *** above the bars indicates statistical difference with respect to the I/R group (p < 0.05; one-way ANOVA followed by Tukey's multiple comparison test). NS, no difference between the BMSCs + I/R and CTR groups.

BMSCs Decrease Reactive Oxygen Species and Stimulate Protein S-Nitrosothiol Formation

ROS are generated during reperfusion (3,16,19,23). To investigate whether BMSC administration was able to reduce ROS levels, we measured H₂O₂ formation after dis-mutation of O₂⁻. by addition of superoxide dismutase to the assay medium. Figure 7A, B shows the superfluorescence signal of Amplex Red responding to increased ROS 24 h after I/R after oxidation of succinate was initiated (gray circles in Fig. 7A; gray bar in Fig. 7B). BMSC administration reduced O₂⁻ ^. availability to control levels (compare yellow circles with black squares in Fig. 7A, where they superimpose and yellow bar with black bar in Fig. 7B).

Figure 7.

BMSCs decrease ROS generation by mitochondria from rat kidneys and strongly enhance S-nitrosothiol formation in proteins from mitochondria isolated from kidney cortex and in whole LLC-PK₁ cells extracts. (A) Reactive oxygen species (ROS) levels were assayed by evaluating the Amplex Red fluorescence increase after successive additions of mitochondria (“M,” to 0.2 mg/ ml) and succinate (“S,” to 10 mM) in 2 ml. A representative experiment (n = 5) is shown for three conditions in which fluorescence signals were acquired at 2-s intervals in five experiments carried out with different mitochondrial preparations. Gray circles, mitochondria from kidneys subjected to I/R; black squares, mitochondria from sham-operated rats (CTR); yellow circles, mitochondria from kidneys subjected to I/R and receiving BMSCs 1 h before ischemia. Regression lines were adjusted to the experimental points starting 80 s after addition of succinate to allow stabilization of the traces (r = 0.94 for the I/R condition; r = 0.70 for control; r = 0.72 for BMSCs + I/R; the lines for the last two conditions overlap). (B) Graphic representation of the rate of ROS generation in pmol H₂O₂/min in each condition. The rates of H₂O₂ formation in each condition were calculated from the slopes of the straight lines in (A). Results are mean ± SE (n = 5 in each condition); ***p < 0.05 with respect to the I/R group. NS, no difference between the BMSCs + I/R and the CTR groups (one-way ANOVA followed by Tukey's multiple comparison test). (C) S-Nitrosylations in mitochondria isolated from kidney cortex. (D) S-Nitrosylations in immortalized kidney proximal tubule cells of porcine origin (LLC-PK₁ cells). These cells were cocultured or not with BMSCs after a transient poisoning of mitochondrial respiration at complex III (antimycin A), followed by washing and exposure to normal incubation conditions. Assays for S-nitrosothiols (SNO) were carried out in the conditions as follows. Using mitochondria isolated from cortex and cells not exposed to antimycin A (CTR), mitochondria isolated after 30 min of ischemia and 1 h of reperfusion (I/R) and in cells transiently subjected to hypoxia and reincubated without BMSCs in normoxia for 24 h (HYP), in mitochondria isolated from kidneys that received BMSCs before the I/R injury (BMSCs + I/R), and in cells subjected to hypoxia and then reincubated with BMSCs in normoxia (HYP + BMSCs). Dark gray, white, and light gray bars correspond to total, low-molecular mass, and high-molecular mass S-nitrosothiols, respectively. The Kruskal–Wallis test followed by Dunn's multiple comparison test was used to analyze the high- and low-molecular mass S-nitrosothiol data, after detection of their nonnormal distribution using the Kolmogorov–Smirnov test. The box plots show median values and whiskers at 25th and 75th percentiles (n = 4). ***p < 0.05 with respect to CTR and I/R (mitochondria, C), or with respect to CTR and HYP (LLC-PK₁ cells, D). No significant difference was found among the low-molecular mass fractions of the three groups from mitochondria and LLC-PK₁ cells (p > 0.05).

Mitochondria are also an important source of nitric oxide (NO) formation and metabolization (10,27). More important, NO reacts with cysteine thiols in a process that decreases its availability for reacting with superox-ide (O₂⁻.) to form the toxic anion peroxynitrite (ONOO⁻), thus avoiding N-nitration (10,24,27), which is considered a trigger of cell death (39). BMSCs greatly stimulated S-nitrosylation in mitochondria isolated from tubule cells subjected to I/R injury (two- to threefold; Fig. 7C) and in LLC-PK₁ cells, a stable lineage derived from proximal tubules (Fig. 7D). When the cells were first transiently poisoned with antimycin A to block respiration at the level of mitochondrial complex III—becoming depleted of ATP (8,22)—and then cocultured with BMSCs (after antimy-cin removal), we observed a five- to eightfold increase in high-molecular mass S-nitrosothiols when compared with those found in control cells (CTR) and the antimy-cin A-poisoned untreated cells. No statistically significant differences were found among the low-molecular mass fractions in the three groups from either mitochondria or LLC-PK₁ cells.

BMSCs Stimulate Proliferation and Decrease Apoptosis and Inflammatory Response in Tubule Cells 24 h After I/R

The experiments depicted in Figure 8 demonstrate that BMSC administration was able to avoid early damage in proximal tubule cells that can jeopardize the long-term success after I/R in the case of transplants as well as structural/functional recovery in the case of AKI. Figure 8A-D shows that I/R-induced proliferation in the kidney cortex, measured by the percentage of cells that were positive for PCNA, increased from 8% in controls to 25% in the I/R group and to more than 40% per field in the BMSCs + I/R group, respectively. The number of apoptotic cells (Fig. 8E-H) more than doubled after I/R but decreased even below the control value as a result of BMSC infusion. An additional tissue modification was one that involved the number of inflammatory cells, studied here by measuring surface density of ED1 antigen: it also doubled after I/R and returned to control values if BMSCs were administered (Fig. 8I-L).

Figure 8.

BMSCs enhance proximal tubule cell proliferation (PCNA-positive cells), decrease apoptotic cells (TUNEL-positive cells), and decrease inflammatory response (ED1 surface density) after I/R. Images were captured from sections of cortex corticis (400×). (A-D) proliferating cell nuclear antigen (PCNA)-positive cells. (E-H) Terminal dUTP nick-end labeling (TUNEL)-positive cells. (I-L) Surface density of ED1 (CD68)-positive macrophages. Tissues obtained from sham-operated rats: (A), (E), and (I). Tissues from I/R rats: (B), (F), and (J). Tissues from rats subjected to I/R that were given BMSCs 1 h before ischemia: (C), (G), and (K). Yellow arrows indicate examples of positive reactions for PCNA, TUNEL, and ED1. Bar graphs show the quantifications corresponding to the groups sham-operated (CTR, empty bars), I/R (black bars), and I/R treated with BMSCs (gray bars). Different lowercase letters above the bars in D, H, and L indicate statistically different results (at least p < 0.05; one-way ANOVA followed by Tukey's multiple comparison test within each experimental determination). At least 30 images were acquired from tissues of each of 18 rats (six for each experimental condition).

BMSCs Shift the Tubular Histological Profile From a Pattern of Lesion to Another of Repair and Preserve Key Markers of Normal Kidney Function

Figure 9 shows representative cortical images that allowed evaluation of tubular dilatation, apical cytoplasm vacuolization, cell detachment, brush border integrity, denuded basement membrane, and tubular necrosis in the CTR, I/R, and BMSCs + I/R groups, and Table 1 presents the average score for each parameter. All the intense tubular lesions provoked by the I/R injury were prevented if BMSCs were injected 1 h before bilateral arterial clamping. In terms of renal functional parameters, the elevated plasma creatinine (Fig. 10A) and the decreased urine osmolality (Fig. 10B) returned to control values in the BMSCs + I/R rats.

Table 1.

Histological Score of Rat Kidneys After Ischemia/ Reperfusion Injury

	CTR	I/R	BMSCs + I/R
Tubular dilatation	1	4	1
Apical cytoplasm vacuolization	1	4	1
Cell detachment	1	5	1
Brush border integrity	1.5	5	2
Denuded basement membrane	1	5	2
Tubular necrosis	1	4	2
Score	1.1 ± 0.08^a	4 4.5 ± 0.22^b	2 1.5 ± 0.22^a

Scores are mean ± SE from five rats. Different lowercase letters indicate statistical difference (p < 0.05; one-way ANOVA followed by Tukey's multiple comparison test). CTR, control; I/R, ischemia/reperfusion; BMSCs, bone marrow-derived stem cells.

Figure 9.

BMSCs preserve the normal proximal tubule morphometry after I/R. Representative hematoxylin and eosin (H&E) images (200×) of cortex corticis. (A) Sham-operated rat. (B) I/R rat. (C) BMSCs + I/R rat. For the histological score analysis, see Table 1.

Figure 10.

BMSCs preserve normal creatinine plasma levels and urinary osmolality after I/R. (A) Creatinine plasma levels. (B) Urinary osmolality. The experimental groups were those indicated on the abscissae. Different lowercase letters above the bars indicate statistically different results (p < 0.05; one-way ANOVA followed by Tukey's multiple comparison test).

Discussion

The full recovery of mitochondrial respiration when BMSCs are administered before ischemia or simultaneously with deocclusion of renal arteries highlights the crucial role that BMSCs can play in preventing mito-chondrial dysfunction in medical events such as renal transplantation and AKI. Most important, our results define critical moments for a beneficial impact of BMSC administration in I/R injuries. The short window for effective administration before ischemia (Fig. 3) is compatible with a mechanism in which there is a rapid and transient release of protective soluble factors from BMSCs reaching the epithelium, followed by their binding to mitochondria in a way that averts damage during ischemia. Since there was no protective effect when the cells were administered 30 min before ischemia, it is likely that these factors need to be present at the moment of the ische-mic injury. BMSCs are equally effective when injected at the moment in which the clamps were removed and the circulation restored, so it is reasonable to hypothesize that the injury functions as a signal to attract and host the injected BMSCs, as proposed in other models of ische-mic injury (42). This view is reinforced by our previous observation that BMSC- and mesenchymal stem cell-conditioned media are beneficial against hypoxia only if they have had an indirect contact through the porous membrane of a minicell system (22). The tracing experiments (Fig. 2) are indicative that BMSCs reach a denuded tubular epithelium (Fig. 9B, Table 1) and help to promote dedifferentiation of surviving epithelial cells followed by their differentiation and proliferation to finally recover the normal tubular architecture (Fig. 9C).

From the slow recovery seen in Figure 4, it seems clear that these still not totally known factors induce upregulation of protective signaling pathways—and possibly expression of key proteins—which are able, among other actions, to prevent the disruption of the mitochon-drial respiratory chain during ischemia, thus allowing full recovery of mitochondrial functions 24 h after circulation was restarted. Recently, it has been demonstrated that the expression of a protein able to augment liver regeneration occurs in the kidney after I/R enhances tubule regeneration in a process in which B-cell CLL/lymphoma 2 (bcl-2) (among other proteins) is upregulated (21). The fact that, besides electron fluxes, the response to ADP and the full capacity for ATP synthesis were preserved when BMSCs were preventively administered (Figs. 5 and 6B) supports the idea of a tight structural preservation of mitochon-drial complexes, a process where bcl-2 plays a key role (17,21,44).

Associated with inhibition of electron fluxes by I/R, the collapse of the transmembrane electrochemical gradient for H⁺ (Fig. 6A) clearly means that functioning of H⁺ trans-location mechanisms across the internal mitochondrial membrane (20) was also blocked, as expected from the very low electron fluxes. Preservation of the gradient by BMSCs (Fig. 6A) can be considered the result of a beneficial influence on the structures responsible for H⁺ translocation across the inner mitochondrial membrane. The recovery of the normal P/O ratio (13) seen in Figure 6B demonstrates that utilization of the gradient through the F_oF₁-ATP synthase was properly coupled to ATP synthesis.

The return of mitochondrial ROS production to normal levels by BMSCs, after an approximately 100% increase by I/R (Fig. 7A, B), means that electron delivery to the cytochrome c oxidase, the mitochondrial complex where the final electron transfer to O₂ occurs (13,27), becomes so efficient in the presence of BMSCs that premature electron transfer to O₂ is held strictly at the low, physiological levels. The remarkable increase in S-nitrosylated high-molecular mass proteins in mitochondria isolated from the cortex of kidneys subjected to I/R (Fig. 7C) and in antimycin A-poisoned LLC-PK₁ cells (Fig. 7D) promoted by BMSCs—in a process that can be considered in physiological synchrony with the inhibition of ROS described above—could have two beneficial effects: (i) scavenging NO through transnitrosylation/transfer reactions would avoid its reaction with O₂⁻. to form ONOO⁻and (ii) bioactive nitrosothiols could stimulate protective signaling pathways, as recently proposed (2,25,30,37,39). It may be that the slow recovery process induced by BMSCs (Fig. 4) involves a cascade of recovery processes where S-nitrosylation could play a central protective role, as recently proposed for other cells and tissues (2,15,29). It is interesting that the level of S-nitrosylated proteins (in terms of SNO groups formed per mg of total protein) is more than four times higher in cell extracts than in mitochondria (compare Fig. 7C and D). This difference could indicate an amplification of the nitrosothiol-mediated signals delivered from mitochondria (10,27) to other targets in different cell compartments.

Proliferation of proximal tubule cells normally occurs to different degrees after a great variety of acute injuries. This proliferation is the result of a self-restoration capacity of renal epithelial cells (9), which is clearly stimulated by BMSCs (Fig. 8A-D). BMSCs also promote a decrease in TUNEL-positive apoptotic cells to levels below those encountered in control conditions (Fig. 8E-H). It is noteworthy that the alterations characterized by DNA damage were not accompanied by modifications in the cortical levels of activated caspase-3, as demonstrated by immunohistochemical analysis (not shown). These data are indicative that I/R induces apoptosis—at least in the present model—through a caspase-independent mechanism starting from a damaged mitochondrion, as those recently revisited (4,5,7,31,38); for a recent review, see Pradelli et al. (35).

At a tissular level, blunting of the inflammatory response by BMSCs (Fig. 8I-L) may also have an unanticipated benefit for the kidney. There is growing evidence that a lower stimulation of immune response (16,26) reduces the long-term organ damage or, in other words, the risk of final rejection of a transplanted organ or progressive advance of renal disease (45). In this regard, the influence of BMSCs on the inflammatory response is in line with the renoprotective actions demonstrated by stimulus of proliferation and inhibition of apoptosis with the possible participation of S-nitrosylation (29) and upregulation of different antiapoptotic proteins and factors (21).

The profile of restoration is confirmed by the histo-logical analysis shown in Figure 9 and quantified in Table 1. This ensemble of structural information (14) comes up as an additional evidence for the beneficial impact of BMSCs against the tissue damage that the I/R injury can provoke (41,42) and may result from the interaction of the injected cells with the lesioned epithelium, as suggested from the images obtained with traced BMSCs (Fig. 2). On the other hand, the complete restoration of the normal plasma creatinine levels and urine osmolality (Fig. 10) constitutes two key functional correlates of the preserved tubular architecture and, in the case of plasma creatinine, an indicative of glomerular protection as well (42).

What emerges from the results described here is the possibility of an intervention that will minimize short-and long-term impairment of kidney structure and function after transplantation. Regarding I/R injury in kidney transplantation, its prevention must ideally begin with donor pretreatment (16). Possibly, autogenic administration of stem cells obtained from the donor may achieve this goal, since they are able to promote complete restoration of respiration and ATP synthesis as well as attenuating a wide spectrum of structural damage. Prevention in this way is clearly not feasible in the case of deceased (non-heart-beating) donors (16), but it is noteworthy that full recovery of coupled mitochondrial respiration is also achieved when (isogenic) BMSCs are given at the moment of reperfusion release as now demonstrated.

In conclusion, the results presented here demonstrate that renal mitochondria are a pivotal target for BMSCs to prevent damage resulting from an acute I/R insult. The quenching of ROS generation by BMSCs, thus avoiding toxic side reactions involving ROS (such as formation of ONOO⁻), seems to be one of the main protective mechanisms against cell damage. The enhancement of S-nitrosylation as well as the ability of BMSCs to stimulate tubule cell proliferation, to inhibit apoptosis, and to attenuate the inflammatory response are in line with the proposal that a global benefit—as a therapeutic strategy at the level of mitochondrial respiration—can be achieved with BMSCs in an organ with the highest respiratory rate in humans.

Footnotes

Acknowledgments

This work was supported by the Carlos Chagas Filho Research Foundation of the State of Rio de Janeiro (FAPERJ), the Brazilian National Research Council (CNPq), the Brazilian Federal Agency for Support and Evaluation of Graduate Education (CAPES), and the National Institutes of Science and Technology (INCT), Brazil. C.F-F. holds a graduate fellowship from CAPES; H.J.V.B. and N.C.M. hold fellowships from CNPq. The technical assistance by Glória Costa-Sarmento and Alexandre Abrantes is acknowledged. The authors declare no conflicts of interest.

References

Barreira

A. L.

; Takiya

C. M.

; Castiglione

R. C.

; Maron-Gutierrez

; Barbosa

C. M.

; Ornellas

D. S.

; Verdoorn

K. S.

; Pascarelli

B. M.

; Borojevic

; Einicker-Lamas

; Leite

Jr. ; Morales

M. M

; Vieyra

Bone marrow mononuclear cells attenuate interstitial fibrosis and stimulate the repair of tubular epithelial cells after unilateral ureteral obstruction. Cell. Physiol. Biochem. 24:585–594; 2009.

Benhar

; Forrester

M. T.

; Hess

D. T.

; Stamler

J. S.

Regulated protein denitrosylation by cytosolic and mito-chondrial thioredoxins. Science 320:1050–1054; 2008.

Benitez-Bribiesca

; Gomez-Camarillo

; Castellanos-Juarez

; Mravko

; Sanchez-Suarez

Morphologic, biochemical and molecular mitochondrial changes during reperfusion phase following brief renal ischemia. Ann. N Y Acad. Sci. 926:165–179; 2000.

Brown

; Yu

B. D.

; Joza

; Benit

; Meneses

; Firpo

; Rustin

; Penninger

J. M.

; Martin

G. R.

Loss of Aif function causes cell death in the mouse embryo, but the temporal progression of patterning is normal. Proc. Natl. Acad. Sci. U S A 103:9918–9923; 2006.

Cande

; Vahsen

; Kouranti

; Schmitt

; Daugas

; Spahr

; Luban

; Kroemer

R. T.

; Giordanetto

; Garrido

; Penninger

J. M.

; Kroemer

AIF and cyclo-philin A cooperate in apoptosis-associated chromatinolysis. Oncogene 23:1514–1521; 2004.

Cavaglieri

R. C.

; Martini

; Sogayar

M. C.

; Noronha

I. L.

Mesenchymal stem cells delivered at the subcapsule of the kidney ameliorate renal disease in the rat remnant kidney model. Transplant. Proc. 41:947–951; 2009.

Cheung

E. C.

; Joza

; Steenaart

N. A.

; McClellan

K. A.

; Neuspiel

; McNamara

; MacLaurin

J. G.

; Rippstein

; Park

D. S.

; Shore

G. C.

; McBride

H. M.

; Penninger

J. M.

; Slack

R. S.

Dissociating the dual roles of apoptosis-inducing factor in maintaining mitochondrial structure and apoptosis. EMBO J. 25:4061–4073; 2006.

Dagher

P. C.

Modeling ischemia in vitro: Selective depletion of adenine and guanine nucleotide pools. Am. J. Physiol. Cell Physiol. 279:C1270-1277; 2000.

Duffield

J. S.

; Park

K. M.

; Hsiao

L. L.

; Kelley

V. R.

; Scadden

D. T.

; Ichimura

; Bonventre

J. V.

Restoration of tubular epithelial cells during repair of the postischemic kidney occurs independently of bone marrow-derived stem cells. J. Clin. Invest. 115:1743–1755; 2005.

10.

Foster

M. W.

; Stamler

J. S.

New insights into protein S-nitrosylation. Mitochondria as a model system. J. Biol. Chem. 279:25891–25897; 2004.

11.

Gnaiger

Bioenergetics at low oxygen: Dependence of respiration and phosphorylation on oxygen and adenosine diphosphate supply. Respir. Physiol. 128:277–297; 2001.

12.

Guo

J. K.

; Cantley

L. G.

Cellular maintenance and repair of the kidney. Annu. Rev. Physiol. 72:357–376; 2010.

13.

Hinkle

P. C.

P/O ratios of mitochondrial oxidative phos-phorylation. Biochim. Biophys. Acta 1706:1–11; 2005.

14.

Jung

; Sola

; Hughes

; Kluth

D. C.

; Vinuesa

; Vinas

J. L.

; Perez-Ladaga

; Hotter

Infusion of IL-10-expressing cells protects against renal ischemia through induction of lipocalin-2. Kidney Int. 81:969–982; 2012.

15.

Kohr

M. J.

; Sun

; Aponte

; Wang

; Gucek

; Murphy

; Steenbergen

Simultaneous measurement of protein oxidation and S-nitrosylation during preconditioning and ischemia/reperfusion injury with resin-assisted capture. Circ. Res. 108:418–426; 2011.

16.

Kosieradzki

; Rowinski

Ischemia/reperfusion injury in kidney transplantation: Mechanisms and prevention. Transplant. Proc. 40:3279–3288; 2008.

17.

Kowaltowski

A. J.

; Cosso

R. G.

; Campos

C. B.

; Fiskum

Effect of Bcl-2 overexpression on mitochondrial structure and function. J. Biol. Chem. 277:42802–42807; 2002.

18.

Kowaltowski

A. J.

; de Souza-Pinto

N. C.

; Castilho

R. F.

; Vercesi

A. E.

Mitochondria and reactive oxygen species. Free Radic. Biol. Med. 47:333–343; 2009.

19.

Lenaz

The mitochondrial production of reactive oxygen species: Mechanisms and implications in human pathology. IUBMB Life 52:159–164; 2001.

20.

Lenaz

; Genova

M. L.

Structure and organization of mito-chondrial respiratory complexes: A new understanding of an old subject. Antioxid. Redox Signal. 12:961–1008; 2010.

21.

Liao

X. H.

; Chen

G. T.

; Li

; Zhang

; Liu

; Sun

; Guo

Augmenter of liver regeneration attenuates tubular cell apoptosis in acute kidney injury in rats: The possible mechanisms. Ren. Fail. 34:590–599; 2012.

22.

Lindoso

R. S.

; Araujo

D. S.

; Adao-Novaes

; Mariante

R. M.

; Verdoorn

K. S.

; Fragel-Madeira

; Caruso-Neves

; Linden

; Vieyra

; Einicker-Lamas

Paracrine interaction between bone marrow-derived stem cells and renal epithelial cells. Cell. Physiol. Biochem. 28:267–278; 2011.

23.

Liu

; Fiskum

; Schubert

Generation of reactive oxygen species by the mitochondrial electron transport chain. J. Neurochem. 80:780–787; 2002.

24.

Lizasoain

; Moro

M. A.

; Knowles

R. G.

; Darley-Usmar

; Moncada

Nitric oxide and peroxynitrite exert distinct effects on mitochondrial respiration which are differentially blocked by glutathione or glucose. Biochem. J. 314:877–880; 1996.

25.

Mannick

J. B.

; Schonhoff

; Papeta

; Ghafourifar

; Szibor

; Fang

; Gaston

S-nitrosylation of mito-chondrial caspases. J. Cell. Biol. 154:1111–1116; 2001.

26.

Matzinger

The danger model: A renewed sense of self. Science 296:301–305; 2002.

27.

Moncada

; Erusalimsky

J. D.

Does nitric oxide modulate mitochondrial energy generation and apoptosis?

Nat. Rev. Mol. Cell. Biol. 3:214–220; 2002.

28.

Morigi

; Introna

; Imberti

; Corna

; Abbate

; Rota

; Rottoli

; Benigni

; Perico

; Zoja

; Rambaldi

; Remuzzi

Human bone marrow mesenchymal stem cells accelerate recovery of acute renal injury and prolong survival in mice. Stem Cells 26:2075–2082; 2008.

29.

Murphy

; Kohr

; Sun

; Nguyen

; Steenbergen

S-nitrosylation: A radical way to protect the heart. J. Mol. Cell. Cardiol. 52:568–577; 2012.

30.

Nadtochiy

S. M.

; Burwell

L. S.

; Ingraham

C. A.

; Spencer

C. M.

; Friedman

A. E.

; Pinkert

C. A.

; Brookes

P. S.

In vivo cardioprotection by S-nitroso-2-mercaptopropionyl glycine. J. Mol. Cell. Cardiol. 46:960–968; 2009.

31.

Niikura

; Dixit

; Scott

; Perkins

; Kitagawa

BUB1 mediation of caspase-independent mitotic death determines cell fate. J. Cell. Biol. 178:283–296; 2007.

32.

Nogueira

; Figueiredo-Freitas

; Casimiro-Lopes

; Magdesian

M. H.

; Assreuy

; Sorenson

M. M.

Myosin is reversibly inhibited by S-nitrosylation. Biochem. J. 424:221–231; 2009.

33.

Olivares

E. L.

; Ribeiro

V. P.

; Werneck de Castro

J. P.

; Ribeiro

K. C.

; Mattos

E. C.

; Goldenberg

R. C.

; Mill

J. G.

; Dohmann

H. F.

; dos Santos

R. R.

; de Carvalho

A. C.

; Masuda

M. O.

Bone marrow stromal cells improve cardiac performance in healed infarcted rat hearts. Am. J. Physiol. Heart Circ. Physiol. 287:H464-H470; 2004.

34.

Portilla

; Kaushal

G. P.

; Basnaldan

A. G.

; Shah

S. V.

The pathophysiology of acute renal failure. In: Runge

M. S.

; Patterson

Humana Press (eds. Principles of Molecular Medicine. New York, NY):643-649; 2011.

35.

Pradelli

L. A.

; Beneteau

; Ricci

J. E.

Mitochondrial control of caspase-dependent and -independent cell death. Cell. Mol. Life Sci. 67:1589–1597; 2010.

36.

Schonhoff

C. M.

; Benhar

Analysis of protein S-nitrosylation. Curr. Protoc. Protein Sci. Chapter 14: Unit14 16; 2011.

37.

Sun

; Morgan

; Shen

R. F.

; Steenbergen

; Murphy

Preconditioning results in S-nitrosylation of proteins involved in regulation of mitochondrial energetics and calcium transport. Circ. Res. 101:1155–1163; 2007.

38.

Susin

S. A.

; Zamzami

; Castedo

; Hirsch

; Marchetti

; Macho

; Daugas

; Geuskens

; Kroemer

Bcl-2 inhibits the mitochondrial release of an apoptogenic protease. J. Exp. Med. 184:1331–1341; 1996.

39.

Szabo

; Ischiropoulos

; Radi

Peroxynitrite: Biochemistry, pathophysiology and development of therapeutics. Nat. Rev. Drug Discov. 6:662–680; 2007.

40.

Tahara

E. B.

; Navarete

F. D.

; Kowaltowski

A. J.

Tissue-, substrate-, and site-specific characteristics of mitochondrial reactive oxygen species generation. Free Radic. Biol. Med. 46:1283–1297; 2009.

41.

Togel

; Cohen

; Zhang

; Yang

; Hu

; Westenfelder

Autologous and allogeneic marrow stromal cells are safe and effective for the treatment of acute kidney injury. Stem Cells Dev. 18:475–485; 2009.

42.

Togel

; Isaac

; Hu

; Weiss

; Westenfelder

Renal SDF-1 signals mobilization and homing of CXCR4-positive cells to the kidney after ischemic injury. Kidney Int. 67:1772–1784; 2005.

43.

Whittembury

; Proverbio

Two modes of Na extrusion in cells from guinea pig kidney cortex slices. Pflugers Arch. 316:1–25; 1970.

44.

Yang

C. C.

; Lin

L. C.

; Wu

M. S.

; Chien

C. T.

; Lai

M. K.

Repetitive hypoxic preconditioning attenuates renal isch-emia/reperfusion induced oxidative injury via upregulating HIF-1 alpha-dependent bcl-2 signaling. Transplantation 88:1251–1260; 2009.

45.

Yilmaz

; McLaughlin

; Paavonen

; Taskinen

; Monroy

; Aavik

; Vamvakopoulos

; Hayry

Clinical predictors of renal allograft: A comparative study of single-lesion histology versus a composite, quantitative scoring system. Transplantation 83:671–676; 2007.