Insulin-induced recurrent hypoglycemia exacerbates diabetic brain mitochondrial dysfunction and oxidative imbalance

Neurobiology of Disease 49 (2013) 1–12

Contents lists available at SciVerse ScienceDirect

Neurobiology of Disease journal homepage: www.elsevier.com/locate/ynbdi

Insulin-induced recurrent hypoglycemia exacerbates diabetic brain mitochondrial dysfunction and oxidative imbalance Susana Cardoso a, b, Renato X. Santos a, b, Sónia C. Correia a, b, Cristina Carvalho a, b, Maria S. Santos a, b, Inês Baldeiras a, c, d, Catarina R. Oliveira a, e, Paula I. Moreira a, f,⁎ a

Center for Neuroscience and Cell Biology, University of Coimbra, Portugal Department of Life Sciences, Faculty of Sciences and Technology, University of Coimbra, Portugal Laboratory of Neurochemistry, Coimbra University Hospital, Portugal d Neurology Department, Faculty of Medicine, University of Coimbra, Portugal e Institute of Biochemistry, Faculty of Medicine, University of Coimbra, Portugal f Institute of Physiology, Faculty of Medicine, University of Coimbra, 3000-354 Coimbra, Portugal b c

a r t i c l e

i n f o

Article history: Received 3 February 2012 Revised 21 July 2012 Accepted 16 August 2012 Available online 24 August 2012 Keywords: Cortex Hippocampus Long-term hyperglycemia Mitochondria Neurodegeneration Oxidative stress Recurrent hypoglycemia

a b s t r a c t Intensive insulin therapy can prevent or slow the progression of long-term diabetes complications but, at the same time, it increases the risk for episodes of severe hypoglycemia. In our study, we used a protocol intended to mimic the levels of blood glucose that occur in type 1 diabetic patients under an intensive insulin therapy. Streptozotocin (STZ)-induced diabetic rats were treated subcutaneously with twice-daily insulin injections for 2 weeks to induce hypoglycemic episodes. Brain cortical and hippocampal mitochondria were isolated and mitochondrial bioenergetics (respiratory chain and phosphorylation system) and oxidative status parameters (malondialdehyde (MDA) levels, mitochondrial aconitase activity and enzymatic and non-enzymatic antioxidant defenses) were analyzed. The protein levels of synaptophysin, a marker of synaptic integrity, and caspase 9 activity were also evaluated in cortical and hippocampal homogenates. Brain cortical mitochondria isolated from hyper- and recurrent hypoglycemic animals presented higher levels of MDA and α-tocopherol together with an increased glutathione disulfide reductase activity, lower manganese superoxide dismutase (MnSOD) activity and glutathione-to-glutathione disulfide (GSH/GSSG) ratio. No significant alterations were found in cortical mitochondrial respiratory chain and oxidative phosphorylation system. Hippocampal mitochondria from both experimental groups presented an impaired oxidative phosphorylation system characterized by a decreased mitochondrial energization potential and ATP levels and higher repolarization lag phase. In addition, higher MDA levels and decreased GSH/GSSG, α-tocopherol levels, and aconitase, glutathione peroxidase and MnSOD activities were observed in both groups of animals. Hippocampal mitochondria from recurrent hypoglycemic animals also showed an impairment of the respiratory chain characterized by a lower state 3 of respiration, respiratory control ratio and ADP/O index, and a higher state 4 of respiration. Additionally, a non-statistically significant decrease in synaptophysin protein levels was observed in cortical homogenates from recurrent hypoglycemic rats as well as in hippocampal homogenates from hyperglycemic and recurrent hypoglycemic rats. An increase in caspase 9 activity was also observed in hippocampal homogenates from hyperglycemic and recurrent hypoglycemic animals. Our results show that mitochondrial dysfunction induced by long-term hyperglycemic effects is exacerbated by recurrent hypoglycemia, which may compromise the function and integrity of brain cells. © 2012 Elsevier Inc. All rights reserved.

Introduction Diabetes mellitus (DM) is a metabolic disorder of carbohydrate metabolism resulting from inadequate insulin release, which

⁎ Corresponding author at: Center for Neuroscience and Cell Biology, University of Coimbra and Institute of Physiology, Faculty of Medicine, University of Coimbra, 3000-354 Coimbra, Portugal. Fax: +351 239480034. E-mail addresses: [email protected], [email protected] (P.I. Moreira). Available online on ScienceDirect (www.sciencedirect.com). 0969-9961/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nbd.2012.08.008

characterizes type 1 diabetes (T1DM), or insulin insensitivity that occurs in type 2 diabetes (T2DM), both of which result in hyperglycemia if not controlled. Although T1DM accounts for only 5–10% of all diabetes cases, it represents a significant public health concern. T1DM begins early in life and leads to long-term complications in several body systems including cardiovascular, renal, and nervous systems (Modi, 2007). Intensive insulin therapy, the standard treatment for individuals with T1DM, aims to provide a tight glycemic control. However, insulin treatment is unable to fully compensate for the tightly regulated insulin secretion of a normally functioning pancreas

2

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

and has the common side-effect of producing periodic hypoglycemic episodes (Leese et al., 2003; McNay, 2005). Although still controversial, growing evidence suggests an association between T1DM and cognitive performance impairments that is reflected in a mild to moderate slowing of mental speed and a diminished mental flexibility (Brands et al., 2005; Jacobson et al., 2011). By depriving the brain of glucose, severe hypoglycemia can cause brain damage leading to long-term impairments in learning and memory (Auer, 2004). And so, in diabetic patients, there have been concerns about the effects of recurrent hypoglycemia and chronic hyperglycemia on cognitive function (Jacobson et al., 2011). Mitochondria are essential organelles for neuronal function because the limited glycolytic capacity of these cells makes them highly dependent on aerobic oxidative phosphorylation for their energetic needs (Moreira et al., 2010). However, it has been established that reactive oxygen species (ROS) production is also inherent to mitochondrial oxidative metabolism (Adam-Vizi and Starkov, 2010; Sullivan et al., 2004). It has been described that DM leads to an oversupply of electrons in the mitochondrial electron transport chain that results in membrane hyperpolarization and ROS formation, mitochondrial energy metabolism dysfunction and oxidative stress being recognized as the main players in diabetes-related complications (Maiese et al., 2007; Moreira et al., 2009a). A recent study demonstrated that recurrent hypoglycemia exacerbated cerebral ischemic damage in type 1 diabetic rats, the increased mitochondrial ROS production being suggested as the possible cause of the ischemic damage exacerbation (Dave et al., 2011). A previous study from our laboratory showed that an acute episode of hypoglycemia potentiated lipid peroxidation and the imbalance of the antioxidant defenses occurring in brain mitochondria isolated from streptozotocin (STZ)-induced diabetic rats (Cardoso et al., 2010). As far as we know, no studies exist regarding the effects of insulin-induced recurrent hypoglycemia in brain mitochondrial function and oxidative status. Therefore, this work aimed to evaluate the effects of long-term hyperglycemia and recurrent hypoglycemia induced by insulin in brain cortical and hippocampal mitochondrial bioenergetics and oxidative status. Several parameters, namely, respiratory chain parameters [states 3 and 4 of respiration, respiratory control ratio (RCR), and ADP/O index], phosphorylation system [transmembrane potential (ΔΨm), ADP-induced depolarization, repolarization lag phase, ATP levels], malondialdehyde (MDA) levels, mitochondrial aconitase activity, and non-enzymatic [glutathione-to-glutathione disulfide (GSH/GSSG) ratio and α-tocopherol levels] and enzymatic antioxidant [glutathione peroxidase (GPx), glutathione disulfide reductase (GR) and manganese superoxide dismutase (MnSOD)] defenses, were evaluated. The levels of synaptophysin, a marker of synaptic integrity, and caspase 9 activity were also analyzed in brain cortical and hippocampal homogenates. Materials and methods

overnight and randomly divided into two groups. One group received an intraperitoneal (i.p.) injection of STZ (50 mg/kg body weight) freshly dissolved in 100 mM citrate, pH 4.5. The volume administered was always 0.5 ml/200 g body weight. The control group received an i.p. injection with an equal volume of citrate (vehicle). In the following 24 h, animals were orally fed with glycosylated serum in order to avoid hypoglycemia resulting from the massive destruction of β-cells and release of intracellular insulin associated with STZ treatment (Moreira et al., 2005). Three days after STZ administration, the tail vein blood glucose levels were measured in all animals and those presenting levels above 250 mg/dl were considered diabetic. After 3 months of the induction of diabetes, the STZ-induced diabetic rats were randomly divided into two groups and one group was subjected to recurrent hypoglycemia achieved by twice-daily subcutaneous injections of insulin (dose adjusted to blood glucose levels) for 2 weeks. Animal handling and sacrifice followed the procedures approved by the Federation of European Laboratory Animal Science Associations (FELASA). Determination of blood glucose and glycated hemoglobin levels Blood glucose concentration was determined from the tail vein using a commercial glucometer (Glucometer-Elite, Bayer, Portugal). Hemoglobin A1C (HbA1c) levels were determined using Systems SYNCHRON CX 4 (Beckman). This system utilizes two cartridges, Hb and A1c to determine A1c concentration as a percentage of the total Hb. The hemoglobin is measured by a colorimetric method and the A1c concentration by a turbidimetric immunoinhibition method. Preparation of mitochondrial fraction Brain cortical and hippocampal mitochondria were isolated from rats according to Moreira et al. (2001, 2002). In brief, the rat was decapitated, and the cortex and hippocampus were rapidly removed, washed, minced, and homogenized at 4 °C in 10 ml of isolation medium (225 mM mannitol, 75 mM sucrose, 5 mM Hepes, 1 mM EGTA, 1 mg/ml BSA, pH 7.4) containing 5 mg of bacterial protease type VIII (Subtilisin). Single brain homogenates were brought to 20 ml and then centrifuged at 2500 rpm (Sorvall RC-5B Refrigerated Superspeed Centrifuge) for 5 min. The pellet, including the fluffy synaptosomal layer, was resuspended in 10 ml of the isolation medium containing 0.02% digitonin and centrifuged at 10,000 rpm for 10 min. The brown mitochondrial pellet without the synaptosomal layer was then resuspended again in 10 ml of the medium and centrifuged at 10,000 rpm for 5 min. The pellet was resuspended in 10 ml of a washing medium (225 mM mannitol, 75 mM sucrose, 5 mM Hepes, pH 7.4) and centrifuged at 10,000 rpm for 5 min. The final mitochondrial pellet was resuspended in the washing medium and mitochondrial protein was determined by the biuret method calibrated with BSA (Gornall et al., 1949).

Chemicals Mitochondrial respiration measurements Streptozotocin was obtained from Sigma (St. Louis, MO, USA). Insulin (Humulin NPH) was obtained from Eli Lilly and Company (USA). Anti-α-tubulin antibody was obtained from Cell Signaling (Danvers, MA, USA). Ac-LEHD-pNA was obtained from Calbiochem, Merck KGaA (Darmstadt, Germany). All the other chemicals were of the highest grade of purity commercially available. Animals treatment Male Wistar rats (2-month-old) were housed in our animal colony (Laboratory Research Center, Faculty of Medicine, University of Coimbra) and maintained under controlled light (12 h day/night cycle) and humidity with free access to water and powdered rodent chow (except in the fasting period). Rats were deprived of food

Oxygen consumption of mitochondria was registered polarographically with a Clark oxygen electrode (Estabrook, 1967) connected to a suitable recorder in a thermostated water-jacketed closed chamber with magnetic stirring. The reactions were carried out at 30 °C in 1 ml of standard respiratory medium (100 mM sucrose, 100 mM KCl, 2 mM KH2PO4, 5 mM Hepes and 10 μM EGTA; pH 7.4) with 0.5 mg of protein. State 3 of respiration (consumption of oxygen in the presence of substrate and ADP) was initiated with ADP (75 nmol/mg protein). States 3 and 4 (consumption of oxygen after ADP phosphorylation) of respiration, respiratory control ratio (RCR= state 3/state 4), and ADP/O index (a marker of the mitochondrial ability to couple oxygen consumption to ADP phosphorylation during state 3 of respiration) were determined according to Chance and Williams (1956).

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

Mitochondrial membrane potential measurements The transmembrane potential (ΔΨm) was monitored by evaluating the transmembrane distribution of the lipophilic cation TPP + (tetraphenylphosphonium) with a TPP +-selective electrode prepared according to Kamo et al. (1979) using an Ag/AgCl-saturated electrode (Tacussel, model MI 402) as reference. TPP + uptake has been measured from the decreased TPP + concentration in the medium sensed by the electrode. The potential difference between the selective electrode and the reference electrode was measured with an electrometer and recorded continuously in a Linear 1200 recorder. The voltage response of the TPP + electrode to log[TPP +] was linear with a slope of 59 ± 1, which is in a good agreement with the Nernst equation. Reactions were carried out in a chamber with magnetic stirring in 1 ml of the standard medium (100 mM sucrose, 100 mM KCl, 2 mM KH2PO4, 5 mM Hepes and 10 μM EGTA; pH 7.4) containing 3 μM TPP +. This TPP + concentration was chosen in order to achieve high sensitivity in measurements and to avoid possible toxic effects on mitochondria (Jensen and Gunter, 1984). The ΔΨm was estimated by the equation: ΔΨm (mV) = 59 log(v/V) − 59 log(10 ΔE/59 − 1), as indicated by Kamo et al. (1979) and Muratsugu et al. (1977). v, V, and ΔE stand for mitochondrial volume, volume of the incubation medium and deflection of the electrode potential from the baseline, respectively. This equation was derived assuming that TPP + distribution between the mitochondria and the medium follows the Nernst equation, and that the law of mass conservation is applicable. A matrix volume of 1.1 μl/mg protein was assumed. No correction was made for the “passive” binding contribution of TPP + to the mitochondrial membranes, because the purpose of the experiments was to show relative changes in potentials rather than absolute values. As a consequence, we can anticipate a slight overestimation on ΔΨm values. However, the overestimation is only significant at ΔΨm values below 90 mV, therefore, far from our measurements. Mitochondria (0.5 mg/ml) were energized with 5 mM succinate (substrate of complex II) in the presence of 2 μM rotenone in order to activate the mitochondrial electron transport chain. After a steady-state distribution of TPP + had been reached (ca. 1 min of recording), ΔΨm fluctuations were recorded. Determination of adenine nucleotide levels At the end of each ΔΨm measurement, 250 μl of each sample was promptly centrifuged at 14,000 rpm (Eppendorf centrifuge 5415C) for 2 min with 250 μl of 0.3 M perchloric acid (HClO4). The supernatants were neutralized with 10 M KOH in 5 M Tris and again centrifuged at 14,000 rpm for 2 min. The resulting supernatants were assayed for adenine nucleotide by separation in a reversephase high performance liquid chromatography (HPLC). The HPLC apparatus was a Beckman-System Gold, consisting of a 126 Binary Pump Model and a 166 Variable UV detector controlled by a computer. The detection wavelength was 254 nm, and the column was a Lichrospher 100 RP-18 (5 μm) from Merck. An isocratic elution with 100 mM phosphate buffer (KH2PO4; pH 6.5) and 1.2% methanol was performed with a flow rate of 1 ml/min. The required time for each analysis was 5 min. Adenine nucleotides were identified by their chromatographic behavior (retention time, absorption spectra and correlation with standards).

3

calculated from a standard curve prepared using the thiobarbituric acid–MDA complex and was expressed as mmol/mg protein. Measurement of aconitase activity Aconitase activity was determined according to Krebs and Holzach (1952). Briefly, mitochondrial fractions (200 μg) were diluted in a 0.6 ml buffer containing 50 mM Tris–HCl and 0.6 mM MnCl2 (pH 7.4), and sonicated for 10 s. Aconitase activity was immediately measured spectrophotometrically by monitoring at 240 nm the cis-aconitase after addition of 20 mM isocitrate at 25 °C. The activity of aconitase was calculated using a molar coefficient of 3.6 mM−1 cm−1 and expressed as U/mg protein/min. One unit was defined as the amount of enzyme necessary to produce 1 μM cis-aconitate per minute. Measurement of glutathione (GSH) and glutathione disulfide (GSSG) levels GSH and GSSG levels were determined with fluorescence detection after reaction of the supernatant containing H3PO4/NaH2PO4–EDTA or H3PO4/NaOH, respectively, of the deproteinized homogenate solution with ophthalaldehyde (OPT), pH 8.0, according to Hissin and Hilf (1976). In brief, freshly isolated mitochondria (0.5 mg) resuspended in 1.5 ml phosphate buffer (100 mM NaH2PO4, 5 mM EDTA, pH 8.0) and 500 μl H3PO4 4.5% were rapidly centrifuged at 50,000 rpm (Beckman, TL-100 Ultracentrifuge) for 30 min. For GSH determination, 100 μl of supernatant was added to 1.8 ml phosphate buffer and 100 μl OPT. After thorough mixing and incubation at room temperature for 15 min, the solution was transferred to a quartz cuvette and the fluorescence was measured at 420 nm and 350 nm emission and excitation wavelengths, respectively. For GSSG determination, 250 μl of the supernatant was added to 100 μl of N-ethylmaleimide and incubated at room temperature for 30 min. After the incubation 140 μl of the mixture was added to a 1.76 ml NaOH (100 mM) buffer and 100 μl OPT. After mixing and incubation at room temperature for 15 min, the solution was transferred to a quartz cuvette and the fluorescence was measured at 420 nm and 350 nm emission and excitation wavelengths, respectively. The GSH and GSSG levels were determined from comparisons with a linear GSH or GSSG standard curve, respectively. Measurement of α-tocopherol (vitamin E) content Extraction and separation of reduced α-tocopherol (vitamin E) from mitochondria were performed by following a previously described method by Vatassery and Younoszai (1978). Briefly, 1.5 ml sodium dodecyl sulfate (10 mM) was added to 0.5 mg of freshly isolated brain mitochondria, followed by the addition of 2 ml ethanol. Then 2 ml hexane and 50 μl of 3 M KCl were added, and the mixture was vortexed for about 3 min. The extract was centrifuged at 2000 rpm (Sorvall RT6000 Refrigerated Centrifuge) and 1 ml of the upper phase, containing n-hexane (n-hexane layer), was recovered and evaporated to dryness under a stream of N2 and kept at −80 °C. The extract was dissolved in n-hexane, and the α-tocopherol content was analyzed by reverse-phase HPLC. A Spherisorb S10w column (4.6 × 200 nm) was eluted with n-hexane modified with 0.9% methanol, at a flow rate of 1.5 ml/min. Detection was performed by an UV detector at 287 nm. The content of mitochondrial vitamin E was calculated as mmol/mg protein. Measurement of glutathione disulfide reductase (GR) activity

Measurement of malondialdehyde (MDA) levels MDA levels were determined by HPLC (Wong et al., 1987). Liquid chromatography was performed using a Gilson HPLC apparatus with a reverse phase column (RP18 Spherisorb, S5 OD2). The samples were eluted from the column at a flow rate of 1 ml/min and detection was performed at 532 nm. The MDA content of the samples was

0.1 mg of each sample was incubated for 1 min with 0.2 mM phosphate buffer (containing 0.2 M K2HPO4 and 2 mM EDTA, pH 7.0) and 2 mM NADPH. The activity of the enzyme was measured at 340 nm and initiated with the addition of 20 mM GSSG, at 30 °C, with continuous magnetic stirring, for 4 min, against blanks prepared in the absence of GSSG, using a Jasco V560 UV/VIS spectrophotometer

4

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

(Carlberg and Mannervik, 1985). GR activity was determined using the molar extinction coefficient 6220 M −1 cm −1 and expressed as nmol/min/mg protein.

imaging system (Bio-Rad) were used and densities from each band were obtained with Quantity One Software (Bio-Rad). Statistical analysis

Measurement of glutathione peroxidase (GPx) activity GPx activity was determined spectrophotometrically at 340 nm by following the method of Flohe and Gunzler (1984). Briefly, the activity of GPx was measured upon a 5 min incubation, in the dark, of 0.1 mg of each sample with 0.5 mM phosphate buffer (0.25 M KH2PO4, 0.25 M K2HPO4 and 0.5 mM EDTA, pH 7.0), 0.5 mM EDTA, 1 mM GSH and 2.4 U/ml glutathione reductase. The quantification occurred after the addition of 0.2 mM NADPH and 1.2 mM tert-butyl hydroperoxide, at 30 °C with continuous magnetic stirring, for 5 min, in a Jasco V560 UV/VIS spectrophotometer. The measurements were made against blanks prepared in the absence of NADPH. GPx activity was determined using the molar extinction coefficient 6220 M −1 cm −1 and expressed as nmol/min/mg protein. Measurement of manganese superoxide dismutase (MnSOD) activity MnSOD activity was determined spectrophotometrically, at 550 nm (Flohe and Otting, 1984). After the incubation of 0.1 mg of protein with 0.07 mM of phosphate buffer (50 mM K2HPO4 and 100 μM EDTA, pH 7.8), 0.025 mM hypoxanthine, 0.025 Triton X-100, 0.1 mM nitrobluetetrazolium (NBT) and 1.33 mM KCN, the reaction was started with the addition of 0.025 U/ml xanthine oxidase, and the reaction was allowed to continue for 3 min at 25 °C, with continuous magnetic stirring. The measurements were performed in a Jasco V560 UV/VIS spectrophotometer, against a blank, prepared in the absence of hypoxanthine. The activity of MnSOD was calculated using a standard curve, prepared with different concentrations of SOD. Measurement of caspase 9 activation Caspase 9 activation was measured using a colorimetric method. Brain cortical and hippocampal tissues were homogenized in cold RIPA buffer (50 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 1% Triton-x100, 1% deoxycorticosterone, 0.1% SDS) and frozen and defrozen three times. The lysates were centrifuged for 10 min at 14,000 rpm. Protein concentrations were measured by using the bicinchoninic acid (BCA) protein assay kit (Pierce, Rockford, IL). Samples (50 μg of protein) were incubated at 37 °C for 2 h in 25 mM Hepes, pH 7.5 containing 0.1% CHAPS, 10% sucrose, 2 mM DTT, and 40 μM Ac-LEHD-pNA. Caspase-9-like activity was determined by measuring the substrate cleavage at 405 nm in a microplate reader (SpectraMax Plus 384, Molecular Devices). Western blotting Brain cortical and hippocampal tissues were homogenized in RIPA buffer, protease and phosphatase inhibitors (commercial protease and phosphatase inhibitor cocktails from Roche), 0.1 mol/l phenylmethylsulfonyl fluoride, and 0.2 mol/l dithiothreitol, frozen three times in liquid nitrogen, and centrifuged at 14,000 rpm for 10 min, and samples (50 μg of protein per lane) were run on 10% SDS-polyacrylamide gels. After electrophoresis, proteins were transferred to PVDF membranes, and blocked membranes (1 h in 5% BSA and 0.1% Tween in TBS for 1 h at room temperature) were incubated overnight at 4 °C with a mouse anti-synaptophysin antibody (1:20,000, Sigma) and a rabbit antibody against α-tubulin (1:1000, Cell Signaling). The proteins were detected separately with the secondary antibodies, anti-mouse (1:10,000) for synaptophysin detection and anti-rabbit (1:1000) for α-tubulin (used as a loading control). The ECF detection system (GE Healthcare) and Versa Doc

Data were analyzed and results are presented as mean ± SEM of the indicated number of experiments. Statistical significance between groups was defined using the non-parametric Mann–Whitney U-test. A p-value b 0.05 was considered significant. Results Characterization of the experimental animals As expected, long-term STZ-induced diabetic rats presented significantly higher values of glycemia and glycated hemoglobin (HbA1C) and a significant decrease in body weight when compared with control rats (Table 1), confirming their diabetic state. Insulininduced recurrent hypoglycemic rats presented a significant decrease of body weight when compared with control rats. Glycemia in recurrent hypoglycemic rats reflects the levels of blood glucose under a hypoglycemic episode (Table 1). HbA1C values in the recurrent hypoglycemic rats were similar to those of STZ rats, which were significantly higher compared to the control group (Table 1). Long-term hyperglycemia affects the respiratory chain and phosphorylation system of hippocampal mitochondria: insulin-induced recurrent hypoglycemia potentiates respiratory chain impairment induced by hyperglycemia ΔΨm is essential for oxidative phosphorylation to occur, which results in the conversion of ADP to ATP via ATP synthase. Mitochondrial respiratory chain pumps H + out of the mitochondrial matrix across the inner mitochondrial membrane. The H + gradient originates an electrochemical potential (Δp) resulting in a pH (ΔpH) and a voltage gradient (ΔΨm) across the inner membrane. Our results show that neither insulin-induced recurrent hypoglycemia nor long-term STZ-induced diabetes induced any significant alteration in the respiratory chain and phosphorylation system of brain cortical mitochondria (Fig. 1 and Table 2). However, in hippocampal mitochondria obtained from long-term hyperglycemic rats a significant decrease in state 3 of respiration (Fig. 2), ΔΨm and ATP levels and an increase in the repolarization lag phase (Table 3) were observed. Recurrent hypoglycemia potentiated STZ-induced respiratory chain impairments by significantly increasing state 4 of respiration and decreasing RCR and ADP/O index (Fig. 2). Insulin-induced recurrent hypoglycemia potentiates oxidative stress and damage induced by long-term hyperglycemia To quantify the extent of lipid peroxidation, a marker of oxidative damage, the levels of MDA were measured. Fig. 3A shows that brain cortical mitochondria from long-term and recurrent hypoglycemic rats

Table 1 Characterization of the experimental animal models.

Body weight (g) Brain weight (g) Glucose (mg/dl) HbA1C (%)

Control

STZ

Hypoglycemia

440 ± 6.0 2.05 ± 0.03 127 ± 5.8 3.6 ± 0.03

296 ± 8.6⁎⁎⁎ 2.10 ± 0.07 453 ± 22.2⁎⁎⁎ 8.5 ± 0.40⁎⁎

330 ± 10.1⁎⁎⁎ 2.17 ± 0.10 45 ± 5.8⁎⁎⁎,+++ 7.1 ± 0.27⁎⁎

Data are the means ± SEM of 6 animals from each condition studied. Statistical significance: ⁎⁎⁎p b 0.001; ⁎⁎p b 0.01 when compared with control rats; +++p b 0.001 when compared with STZ-treated rats. HbA1C — glycated hemoglobin.

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

5

Fig. 1. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia in brain cortical mitochondria respiratory chain parameters: states 3 (A) and 4 (B) of respiration, RCR (C) and ADP/O index (D). Data are the mean ± SEM of 5–6 animals from each condition studied.

presented higher levels of MDA compared with control rats. However, in hippocampal mitochondria, only insulin-induced recurrent hypoglycemia promoted a significant increase in MDA levels (Fig. 3B). Mitochondrial aconitase activity is a sensitive redox sensor of reactive oxygen and nitrogen species in cells. As shown in Fig. 4A, brain cortical mitochondria isolated from recurrent hypoglycemic rats presented a significantly lower aconitase activity when compared with control and long-term STZ-induced diabetic mitochondria. Concerning hippocampal mitochondria, both long-term hyperglycemia and insulin-induced recurrent hypoglycemia induced a significant decrease in aconitase activity, this decrease being more pronounced in mitochondria from recurrent hypoglycemic rats (Fig. 4B). Long-term hyperglycemia and insulin-induced recurrent hypoglycemia impact the antioxidant defense system of brain cortical and hippocampal mitochondria Mitochondria possess a multi-leveled ROS defense network that includes non-enzymatic antioxidants like glutathione and α-tocopherol

Table 2 Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia in the cortical mitochondrial oxidative phosphorylation system (ΔΨm, ADP-induced depolarization, repolarization lag phase and ATP levels).

ΔΨm (mV) ADP-induced depolarization (mV) Repolarization lag phase (min) ATP levels (nmol/mg protein)

Control

STZ

Hypoglycemia

210.2 ± 1.7 23.4 ± 2.6 1.23 ± 0.1 197.1 ± 12.6

216.7 ± 1.7 23.3 ± 2.3 1.27 ± 0.0 205.6 ± 12.0

217.7 ± 0.5 24.4 ± 2.1 1.18 ± 0.1 205.4 ± 19.9

The oxidative phosphorylation parameters were evaluated in freshly isolated cortical mitochondrial fractions (0.5 mg) in 1 ml of the reaction medium supplemented with 3 μM of TPP+ and energized with 5 mM succinate in the presence of 2 μM rotenone. Data are the mean ± SEM of 5–6 animals from each condition studied.

and antioxidant enzymes such as GPx, GR and MnSOD, acting as free radical scavengers and, consequently, protecting cells against oxidative damage. As shown in Fig. 5A, long-term hyperglycemic and recurrent hypoglycemic brain cortical mitochondria presented a significant decrease in GSH/GSSG ratio when compared to control mitochondria. In hippocampal mitochondria only recurrent hypoglycemia promoted a significant decrease in GSH/GSSG ratio (Fig. 5B). Interestingly, α-tocopherol levels in brain cortical mitochondria were significantly increased in both recurrent hypoglycemic and long-term hyperglycemic rats; this increase being more pronounced in mitochondria from recurrent hypoglycemic rats (Fig. 6A). In contrast, a significant decrease in α-tocopherol levels was observed in hippocampal mitochondria obtained from recurrent hypoglycemic rats (Fig. 6B). Concerning antioxidant enzymatic defenses, long-term STZ-induced diabetes significantly increased GR activity in both brain mitochondrial preparations (Figs. 7A and B), an effect that was potentiated by insulin-induced recurrent hypoglycemia in cortical mitochondria (Fig. 7A). Although GPx activity remained statistically unchanged in cortical mitochondria (Fig. 8A), in hippocampal mitochondria a significant decrease in the activity of this enzyme in both long-term diabetic and recurrent hypoglycemic groups was observed (Fig. 8B). Additionally, long-term STZ-induced diabetes caused a significant decrease in MnSOD activity in both brain cortical and hippocampal mitochondria (Figs. 9A and B), an effect that was potentiated by recurrent hypoglycemia in hippocampal mitochondria (Fig. 9B). Long-term hyperglycemia and insulin-induced recurrent hypoglycemia alter synaptophysin protein levels and caspase 9 activity Diabetic conditions are associated with alterations in brain function and cognitive performance. Because the structure and function of neuronal synapses are closely related to cognitive performance, we further analyzed the protein levels of synaptophysin, which

6

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

Fig. 2. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia in brain hippocampal mitochondria respiratory chain parameters: states 3 (A) and 4 (B) of respiration, RCR (C) and ADP/O index (D). Data are the mean± SEM of 5–6 animals from each condition studied. *p b 0.05 when compared with control mitochondria.

plays an important role in the number of synapses and synaptic plasticity (Ding et al., 2002). We observed a decrease, although not statistically significant, in synaptophysin levels in brain cortical homogenates from recurrent hypoglycemic animals (Fig. 10A) while a non-statistically significant decrease in the levels of synaptophysin was observed in hippocampal homogenates from both hyperglycemic and recurrent hypoglycemic rats (Fig. 10B). Mitochondrial abnormalities are intimately associated with oxidative stress and apoptosis (Russell et al., 2002). Under this perspective, caspase 9 activation was also evaluated in brain homogenates. An increase in caspase 9 activation was observed in hippocampal homogenates of both hyperglycemic and recurrent hypoglycemic animals, although only statistically significant in hyperglycemic rats (Fig. 11B). No statistically significant alterations in caspase 9 activation were observed in brain cortical homogenates from the three groups of experimental animals (Fig. 11A).

Table 3 Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia in the hippocampal mitochondrial oxidative phosphorylation system (ΔΨm, ADP-induced depolarization, repolarization lag phase and ATP levels).

ΔΨm (mV) ADP-induced depolarization (mV) Repolarization lag phase (min) ATP levels (nmol/mg protein)

Control

STZ

Hypoglycemia

213.6 ± 4.3 21.2 ± 2.5 1.01 ± 0.11 207.3 ± 10.73

201.7 ± 2.7⁎

199.4 ± 2.0⁎ 20.4 ± 1.2 1.31 ± 0.05⁎ 166.2 ± 12.49⁎

19.4 ± 1.4 1.49 ± 0.09⁎⁎ 166.2 ± 12.54⁎

The oxidative phosphorylation parameters were evaluated in freshly isolated hippocampal mitochondrial fractions (0.5 mg) in 1 ml of the reaction medium supplemented with 3 μM of TPP+ and energized with 5 mM succinate in the presence of 2 μM rotenone. Data are the mean±SEM of 6–8 animals from each condition studied. Statistical significance: ⁎⁎pb 0.01; ⁎pb 0.05 when compared with control rats.

Fig. 3. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on MDA levels in brain cortical (A) and hippocampal (B) mitochondria. Data are the mean ± SEM of 5–6 animals from each condition studied. Statistical significance: **p b 0.01; *p b 0.05 when compared with control mitochondria.

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

Fig. 4. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on aconitase activity of brain cortical (A) and hippocampal (B) mitochondria. Data are the mean± SEM of 5–6 animals from each condition studied. *p b 0.05 when compared with control mitochondria; +p b 0.05 when compared with STZ mitochondria.

7

in agreement with characteristics described in T1DM patients (Ramakrishna and Jailkhani, 2007). Because mitochondria are the main power supply of cells, they play a central role in cell life and death particularly in tissues having high metabolic rates such as the brain. The respiratory chain is the focal point for energy metabolism in the brain since it links substrate and oxygen consumption with ATP production, with neuronal cells being highly dependent on aerobic oxidative phosphorylation for their energetic needs (Moreira and Oliveira, 2011). We observed that recurrent hypoglycemia potentiated the impairment of the respiratory chain induced by long-term hyperglycemia (Fig. 2). Recently, Dave et al. (2011) evaluated the effects of recurrent hypoglycemia in hippocampal mitochondria of diabetic rats exposed to cerebral ischemia. An alteration in the ratio of mitochondrial respiratory chain complex I subunits was observed, which may underlie the increase in free radical levels under those experimental conditions suggesting that hippocampal mitochondria are extremely sensitive to severe glucose oscillations. The mitochondrial electron transport chain generates superoxide (O2•−) as an unavoidable by-product and primary ROS at mitochondrial complexes I and III (St-Pierre et al., 2002). Mitochondrial ROS dysregulation due to altered production or decomposition of ROS is linked to diabetes, obesity, neurodegenerative disorders and aging (Moreira et al., 2009b). When the concentration of ROS and reactive nitrogen species crosses a certain threshold it overwhelms the antioxidant defenses resulting into oxidative stress and damage. An important marker of oxidative stress is the loss of the mitochondrial aconitase activity. Prolonged exposure of mitochondria to oxidants results in disassembly of the [4Fe–4S]2+ cluster and concomitant release of Fe2+, carbonylation, and inactivation of the enzyme, potentially establishing the link between oxidative stress and mitochondrial

Discussion Without the intricate regulation of glucose metabolism normally mediated by insulin, T1DM patients develop high blood glucose levels, with hyperglycemia being considered the primary pathogenic factor for the development of diabetic complications, which include memory and cognitive impairments. Hypoglycemia remains the major factor limiting the application of intensive insulin therapy designed to prevent complications in patients with T1DM (McCrimmon, 2012). Our study shows that long-term hyperglycemia and insulininduced recurrent hypoglycemic episodes exert different effects in mitochondria from different brain areas, namely the cortex and hippocampus, with hippocampal mitochondria being the most severely affected by both metabolic insults. We also show that recurrent hypoglycemia potentiates the effects of long-term hyperglycemia exacerbating brain hippocampal mitochondria impairment and oxidative stress. A previous study from our laboratory showed that an acute episode of hypoglycemia and short-term hyperglycemia differently affected brain cortical and hippocampal mitochondria but, interestingly, in that study brain cortical mitochondria were the most affected (Cardoso et al., 2010), which suggests that different brain regions behave differently according to the severity and/or duration of hyperglycemic and insulin-induced hypoglycemic insults. As discussed earlier, we used a protocol to induce recurrent hypoglycemic episodes in long-term hyperglycemic rats in order to mimic the fluctuation of blood glucose that often occur in T1DM patients under intensive insulin therapy (Herzog et al., 2008; Jacob et al., 1999). The long-term STZ-induced diabetes was confirmed by the significant increase in glucose and HbA1c levels and decrease in body weight when compared with control rats (Table 1), which is

Fig. 5. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on the glutathione content of brain cortical (A) and hippocampal (B) mitochondria. Data are the mean ± SEM of 5–6 animals from each condition studied. Statistical significance: *p b 0.05 when compared with control mitochondria.

8

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

(Fig. 5), while only hippocampal recurrent hypoglycemic mitochondria presented a significant decrease in this antioxidant defense (Fig. 5B). GSH protects against oxidative stress by several ways; one of them is the ability to regenerate α-tocopherol to its active form. GSH can reduce the tocopherol radical of vitamin E directly, or indirectly, via reduction of semidehydroascorbate to ascorbate (Valko et al., 2007). This may explain why brain cortical mitochondria exposed to long-term hyperglycemia and recurrent hypoglycemia presented a lower GSH/GSSG ratio (Fig. 5A), since GSH is probably being used to regenerate α-tocopherol (Fig. 6A), an antioxidant that protects against lipid peroxidation (Fig. 3A). However, in hippocampal mitochondria from long-term hyperglycemic animals no statistically significant differences were observed in MDA (Fig. 3B) and α-tocopherol (Fig. 6B) levels, and GSH/GSSG ratio (Fig. 5B) suggesting that hippocampal mitochondria under hyperglycemic conditions are able to maintain an equilibrium between oxidant and antioxidant processes. On the other hand, hippocampal mitochondria from recurrent hypoglycemic animals presented higher MDA levels (Fig. 3B) and lower α-tocopherol levels (Fig. 6B) and GSH/GSSG ratio (Fig. 5B) implying that hippocampal mitochondria under an additional metabolic insult, recurrent hypoglycemia, lose the capacity to counteract oxidative stress and damage. The enzymatic antioxidant defenses include, among others, MnSOD, GPx, and GR. Under normal physiological conditions, a balance exists between their activities allowing cells to maintain a redox balance. MnSOD is an enzyme that dismutates O2• − into H2O2, which can be converted to water through the action of catalase or GPx in the presence of GSH, which in turn is regenerated by GR. In brain cortical mitochondria, both metabolic insults lead to an increase

Fig. 6. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on α-tocopherol (vitamin E) levels of brain cortical (A) and hippocampal (B) mitochondria. Data are the mean±SEM of 5–6 animals from each condition studied. Statistical significance: ***pb 0.001; **pb 0.01; *pb 0.05 when compared with control mitochondria.

aconitase inactivation (Addabbo et al., 2009; Yan et al., 1997). In the present study, brain cortical mitochondria obtained from recurrent hypoglycemic animals showed a significant decrease in mitochondrial aconitase activity (Fig. 4A), whereas in hippocampal mitochondria a significant decrease was observed in both long-term hyperglycemic and recurrent hypoglycemic animals (Fig. 4B). These results show that brain cortical and hippocampal mitochondria present a distinct behavior when exposed to an oxidative environment induced by metabolic insults. An increase in lipid peroxidation was also observed (Figs. 3A and B). The extension of oxidative stress and damage during hyperglycemia has been shown by marked increases in lipid peroxidation in total brain tissue (Karaagac et al., 2011), brain cortical and hippocampal tissues (Bhutada et al., 2011) and brain cortical mitochondria (Kamboj and Sandhir, 2011) from diabetic animals. We have previously shown that brain cortical mitochondria of short-term hyperglycemic and acute-hypoglycemic animals were more susceptible to lipid peroxidation (Cardoso et al., 2010). α-Tocopherol is the major lipid-soluble chain-breaking antioxidant and, similar to Karaagac et al. (2011), we observed higher levels of α-tocopherol in cortical mitochondria of long-term hyperglycemic and recurrent hypoglycemic rats (Fig. 6A). In opposite, α-tocopherol levels were found to be decreased in hippocampal mitochondria of insulin-induced recurrent hypoglycemic rats (Fig. 6B). Another major cellular non-enzymatic antioxidant and redox buffer is GSH. A previous study showed that acute hypoglycemia and 3 months of diabetes duration led to a decrease in GSH content and an increase in lipid peroxidation in several brain areas (Singh et al., 2004). In our study, both brain cortical long-term hyperglycemic and recurrent hypoglycemic mitochondria presented a decreased GSH/GSSG ratio

Fig. 7. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on glutathione disulfide reductase (GR) activity of brain cortical (A) and hippocampal (B) mitochondria. Data are the mean ± SEM of 5–6 animals from each condition studied. Statistical significance: **p b 0.01; *p b 0.05 when compared with control mitochondria; +p b 0.05 when compared with STZ mitochondria.

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

9

viewed as a mechanism for replenishment of GSH (Srivastava and Shivanandappa, 2005). We believe that the differences observed between brain cortical and hippocampal mitochondria depend on distinct characteristics of each brain region. Indeed, different brain regions have dissimilar patterns of oxidative stress markers and levels/activities of antioxidant defenses (Srivastava and Shivanandappa, 2005). It has been previously shown that brain cortex from control adult rats presented higher levels of GSH (Hussain et al., 1995) and MDA (Rongzhu et al., 2009) compared to the hippocampus. In agreement, our results show that hippocampal mitochondria from control rats presented a lower GSH/ GSSG ratio (Fig. 5B) and MDA levels (Fig. 3B) when compared with control cortical mitochondria (Figs. 3A and 5A). However, hippocampal mitochondria presented higher α-tocopherol levels (Fig. 6B) and aconitase activity (Fig. 4B) and lower GR activity (Fig. 7B) than control cortical mitochondria (Figs. 4A, 6A and 7A). Therefore, the notion that antioxidant systems are not evenly distributed across brain regions (Baek et al., 1999) may explain the differential sensitivity of distinct brain regions in response to metabolic insults. Indeed, regional variations in other cellular and tissue characteristics may also contribute to metabolic impairment. Such environmental variables include the location of mitochondria within a neuron, neurotransmitter phenotype, high tonic firing rates, and differences in tissue perfusion pressure (Dubinsky, 2009). Although not evaluated in this study, blood pressure fluctuations can also be responsible for the mitochondrial abnormalities induced by hyperglycemia and recurrent insulin-induced hypoglycemia. Indeed, diabetes and hypertension are considered potent risk factors for cerebrovascular disease. Kario et al. (2005) reported that brain

Fig. 8. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on glutathione peroxidase (GPx) activity of brain cortical (A) and hippocampal (B) mitochondria. Data are the mean ± SEM of 5–6 animals from each condition studied. Statistical significance: *p b 0.05 when compared with control mitochondria.

in GR activity (Fig. 7A) and a decrease in MnSOD activity (Fig. 9A), whereas GPx activity remained statistically unchanged (Fig. 8A). The decreased MnSOD activity has been already reported in diabetic dorsal root ganglions (Zherebitskaya et al., 2009) and in brain cortical mitochondria of diabetic rats (Kamboj and Sandhir, 2011), which may cause an accumulation of O2• − potentiating the formation of other radicals and disturbing the cellular redox status. In the mitochondrial matrix, MnSOD activity should be in balance with the activity of GPx, a H2O2 detoxifying enzyme, and the matrix GSH redox cycle must be in coordination with MnSOD-mediated scavenging of O2• − in order to prevent excessive H2O2 accumulation (Jezek and Hlavata, 2005). In contrast to Kamboj and Sandhir (2011), we did not observe significant alterations in H2O2 levels in long-term hyperglycemic and recurrent hypoglycemic brain cortical mitochondria (data not shown), although a decrease in GSH/GSSG ratio (Fig. 5A) was observed. We can speculate that the lower GSH/GSSG ratio together with the reduced GR and GPx activities may be a coordinated attempt to overcome the decreased MnSOD activity in order to maintain unchanged H2O2 levels (Perez-Matute et al., 2009). Similarly, in hippocampal mitochondria we also observed unchanged levels of H2O2 (data not shown), higher GR (Fig. 7B) and lower MnSOD (Fig. 9B) activities in both long-term hyperglycemic and recurrent hypoglycemic animals. However, GPx activity in this mitochondrial population was found to be decreased (Fig. 8B). Previous data suggested that the decreased GPx activity found in some brain regions of diabetic rats could be due to the increased endogenous production of O2•− (Kapoor et al., 2009). Consequently, the high levels of unscavenged free radicals would result in oxidative deterioration of polyunsaturated lipids leading to MDA formation whereas the increased activity of GR could be

Fig. 9. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on manganese superoxide dismutase (MnSOD) activity of brain cortical (A) and hippocampal mitochondria (B). Data are the mean ± SEM of 5–6 animals from each condition studied. Statistical significance: **p b 0.01; *p b 0.05 when compared with control mitochondria; +p b 0.05 when compared with STZ mitochondria.

10

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

Fig. 11. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on caspase-9 activity in brain cortical (A) and hippocampal (B) homogenates. Data are the mean (% of control)±SEM of 3–5 animals from each condition studied. Statistical significance: *pb 0.05 when compared with control mitochondria.

Fig. 10. Effect of long-term hyperglycemia and insulin-induced recurrent hypoglycemia on synaptophysin protein levels in brain cortical (A) and hippocampal (B) homogenates. Data are the mean (% of control)±SEM of 3–5 animals from each condition studied.

damage in diabetic subjects was exacerbated by hypertension. A recent study performed in animal models of diabetes showed that diabetes or hypertension led to synaptic loss and together caused neuronal cell loss (Devisser, 2011). A recent paper also showed that experimental hypertension induced by deoxycorticosterone acetate (DOCA) salt resulted in vascular dementia-like features (Sharma and Singh, 2012). There is also evidence of alterations in the permeability of blood–brain barrier (BBB) in diabetic hypertensive rats (Oztaş and Küçük, 1995). Similarly, Awad (2006) found that hypertensive rats presented an increased permeability of BBB, a decrease in brain levels of GSH and SOD and an increase in lipid peroxides. Dysfunctional mitochondria and mitochondria dependent-apoptosis contribute to neuronal loss in most neurodegenerative diseases. It was previously shown that high glucose levels induced apoptosis in neuronal (Sharifi et al., 2009) and endothelial cells (Correia et al., 2012) by a mechanism involving oxidative stress and mitochondrial abnormalities. Moreover, hypoglycemic insults were also associated to apoptotic cell

death (Ouyang et al., 2000; Turner et al., 2004). We also observed an increase in caspase 9 activity in the hippocampus of hyperglycemic and recurrent hypoglycemic rats (Figs. 11A and B). In line with our findings, a recent in vitro study showed that glucose fluctuations adversely affected mitochondrial activity and enhanced apoptosis gene expression (Russo et al., 2012). Evidence corroborates that diabetic conditions affect brain function and cognitive performance. Abnormalities in proteins involved in neurotransmission and functional synaptic plasticity, such as synaptophysin, have been reported in diverse neurodegenerative disorders as markers of dysregulated neural connectivity (Al-Noori et al., 2008; Valtorta et al., 2004). The levels of synaptophysin, a protein associated to structural remodeling of synapses, were decreased in brain cortical homogenates from hypoglycemic rats and in hippocampal homogenates from hyperglycemic and recurrent hypoglycemic rats (Figs. 10A and B). This observation is in agreement with previous studies reporting that chronic hyperglycemia triggered synaptic degeneration in the hippocampus of STZ-induced diabetic animals by promoting a decrease in synaptic proteins density, such as syntaxin, SNAP25 and synaptophysin (Duarte et al., 2009; Grillo et al., 2005; Malone et al., 2006). A decrease in synaptophysin protein levels was also observed in the hypothalamic regions of recurrent hypoglycemic rats (Al-Noori et al., 2008). It was also reported that anoxic–hypoglycemic episodes rapidly modified the hippocampal structural and functional components of synaptic networks, at both the pre- and postsynaptic levels (Jourdain et al., 2002). In opposite, Singh et al. (2002) found an enhanced expression of synaptophysin in the cerebellum, brain stem and diencephalon regions under an acute hypoglycemic event. As previously

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

discussed, results disparities found in different brain regions can be due to the severity and/or duration of the metabolic insults. We must also emphasize that our analyses were performed in brain tissue homogenates composed by several cell types such as neurons, endothelial cells and astrocytes, which may camouflage the alterations occurring at the neuronal level. Several years after insulin discovery and with all the advances in its therapy, hypoglycemia remains as the major barrier avoiding insulin from achieving its full therapeutic promise. Our data suggest that blood glucose fluctuations may compromise brain mitochondrial function and neuronal integrity, these effects being more pronounced in the hippocampus. The hippocampus is an important structure for the integration of learning and memory in the mammalian CNS and it is particularly sensitive to changes in glucose homeostasis (Wrighten et al., 2009), with evidence suggesting that recurrent severe hypoglycemia may occasionally cause sub-clinical cerebral injury or permanent cognitive impairment (Warren and Frier, 2005; Wright et al., 2009). Our results support the view that brain mitochondrial dysfunction and oxidative stress underlie brain injury that result from long-term hyperglycemic and recurrent hypoglycemic episodes. Although the ideal solution would be the maintenance of tight blood glucose control, perhaps therapies focusing on reducing mitochondrial dysfunction and oxidative stress would be of interest for type 1 diabetic subjects at loose glycemic control. Acknowledgments Susana Cardoso has a PhD fellowship from the Portuguese Foundation for Science and Technology (SFRH/BD/43968/2008). We thank Dr. Paula Agostinho for kindly providing the synaptophysin antibody. References Adam-Vizi, V., Starkov, A.A., 2010. Calcium and mitochondrial reactive oxygen species generation: how to read the facts. J. Alzheimers Dis. 20 (Suppl. 2), S413–S426. Addabbo, F., et al., 2009. The Krebs cycle and mitochondrial mass are early victims of endothelial dysfunction: proteomic approach. Am. J. Pathol. 174, 34–43. Al-Noori, S., et al., 2008. Recurrent hypoglycemia alters hypothalamic expression of the regulatory proteins FosB and synaptophysin. Am. J. Physiol. Regul. Integr. Comp. Physiol. 295, R1446–R1454. Auer, R.N., 2004. Hypoglycemic brain damage. Metab. Brain Dis. 19, 169–175. Awad, A.S., 2006. Role of AT1 receptors in permeability of the blood–brain barrier in diabetic hypertensive rats. Vascul. Pharmacol. 45, 141–147. Baek, B.S., et al., 1999. Regional difference of ROS generation, lipid peroxidation, and antioxidant enzyme activity in rat brain and their dietary modulation. Arch. Pharm. Res. 22, 361–366. Bhutada, P., et al., 2011. Protection of cholinergic and antioxidant system contributes to the effect of berberine ameliorating memory dysfunction in rat model of streptozotocin-induced diabetes. Behav. Brain Res. 220, 30–41. Brands, A.M., et al., 2005. The effects of type 1 diabetes on cognitive performance: a meta-analysis. Diabetes Care 28, 726–735. Cardoso, S., et al., 2010. Cortical and hippocampal mitochondria bioenergetics and oxidative status during hyperglycemia and/or insulin-induced hypoglycemia. Biochim. Biophys. Acta 1802, 942–951. Carlberg, I., Mannervik, B., 1985. Glutathione reductase. Methods Enzymol. 113, 484–490. Chance, B., Williams, G.R., 1956. The respiratory chain and oxidative phosphorylation. Adv. Enzymol. Relat. Subj. Biochem. 17, 65–134. Correia, S.C., et al., 2012. Cyanide preconditioning protects brain endothelial and NT2 neuron-like cells against glucotoxicity: role of mitochondrial reactive oxygen species and HIF-1α. Neurobiol. Dis. 45, 206–218. Dave, K.R., et al., 2011. Recurrent hypoglycemia exacerbates cerebral ischemic damage in streptozotocin-induced diabetic rats. Stroke 42, 1404–1411. Devisser, A., 2011. Differential impact of diabetes and hypertension in the brain: adverse effects in grey matter. Neurobiol. Dis. 44, 161–173. Ding, Y., et al., 2002. Functional improvement after motor training is correlated with synaptic plasticity in rat thalamus. Neurol. Res. 24, 829–836. Duarte, J.M., et al., 2009. Caffeine consumption attenuates neurochemical modifications in the hippocampus of streptozotocin-induced diabetic rats. J. Neurochem. 111, 368–379. Dubinsky, J.M., 2009. Heterogeneity of nervous system mitochondria: location, location, location! Exp. Neurol. 218, 293–307. Estabrook, R.E., 1967. Mitochondrial respiratory control and the polarographic measurement of ADP/O ratios. Methods Enzymol. 10, 41–47. Flohe, L., Gunzler, W.A., 1984. Assays of glutathione peroxidase. Methods Enzymol. 105, 114–121.

11

Flohe, L., Otting, F., 1984. Superoxide dismutase assays. Methods Enzymol. 105, 93–104. Gornall, A.G., et al., 1949. Determination of serum proteins by means of the biuret reaction. J. Biol. Chem. 177, 751–766. Grillo, C.A., et al., 2005. Immunocytochemical analysis of synaptic proteins provides new insights into diabetes-mediated plasticity in the rat hippocampus. Neuroscience 136, 477–486. Herzog, R.I., et al., 2008. Effect of acute and recurrent hypoglycemia on changes in brain glycogen concentration. Endocrinology 149, 1499–1504. Hissin, P.J., Hilf, R., 1976. A fluorometric method for determination of oxidized and reduced glutathione in tissues. Anal. Biochem. 4, 214–226. Hussain, S., et al., 1995. Age-related changes in antioxidant enzymes, superoxide dismutase, catalase, glutathione peroxidase and glutathione in different regions of mouse brain. Int. J. Dev. Neurosci. 13, 811–817. Jacob, R.J., et al., 1999. Effects of recurrent hypoglycemia on brainstem function in diabetic BB rats: protective adaptation during acute hypoglycemia. Diabetes 48, 141–145. Jacobson, A.M., et al., 2011. Biomedical risk factors for decreased cognitive functioning in type 1 diabetes: an 18 year follow-up of the Diabetes Control and Complications Trial (DCCT) cohort. Diabetologia 54, 245–255. Jensen, B.D., Gunter, T.R., 1984. The use of tertaphenylphosphonium (TPP+) to measure membrane potentials in mitochondria: membrane binding and respiratory effects. Biophys. J. 45, 92. Jezek, P., Hlavata, L., 2005. Mitochondria in homeostasis of reactive oxygen species in cell, tissues, and organism. Int. J. Biochem. Cell Biol. 37, 2478–2503. Jourdain, P., et al., 2002. Remodeling of hippocampal synaptic networks by a brief anoxia–hypoglycemia. J. Neurosci. 22, 3108–3116. Kamboj, S.S., Sandhir, R., 2011. Protective effect of N-acetylcysteine supplementation on mitochondrial oxidative stress and mitochondrial enzymes in cerebral cortex of streptozotocin-treated diabetic rats. Mitochondrion 11, 214–222. Kamo, N., et al., 1979. Membrane potential of mitochondria measured with an electrode sensitive to tetraphenyl phosphonium and relationship between proton electrochemical potential and phosphorylation potential in steady state. J. Membr. Biol. 49, 105–121. Kapoor, R., et al., 2009. Bacopa monnieri modulates antioxidant responses in brain and kidney of diabetic rats. Environ. Toxicol. Pharmacol. 27, 62–69. Karaagac, N., et al., 2011. Changes in prooxidant–antioxidant balance in tissues of rats following long-term hyperglycemic status. Endocr. Res. 36, 124–133. Kario, K., et al., 2005. Diabetic brain damage in hypertension: role of renin–angiotensin system. Hypertension 45, 887–893. Krebs, H.A., Holzach, O., 1952. The conversion of citrate into cis-aconitate and isocitrate in the presence of aconitase. Biochem. J. 52, 527–528. Leese, G.P., et al., 2003. Frequency of severe hypoglycemia requiring emergency treatment in type 1 and type 2 diabetes: a population-based study of health service resource use. Diabetes Care 26, 1176–1180. Maiese, K., et al., 2007. Oxidative stress biology and cell injury during type 1 and type 2 diabetes mellitus. Curr. Neurovasc. Res. 4, 63–71. Malone, J.I., et al., 2006. Hyperglycemic brain injury in the rat. Brain Res. 1076, 9–15. McCrimmon, R.J., 2012. Update in the CNS response to hypoglycemia. J. Clin. Endocrinol. Metab. 97, 1–8. McNay, E.C., 2005. The impact of recurrent hypoglycemia on cognitive function in aging. Neurobiol. Aging 26 (Suppl. 1), 76–79. Modi, P., 2007. Diabetes beyond insulin: review of new drugs for treatment of diabetes mellitus. Curr. Drug Discov. Technol. 4, 39–47. Moreira, P.I., Oliveira, C.R., 2011. Mitochondria as potential targets in antidiabetic therapy. Handb. Exp. Pharmacol. 331–356. Moreira, P.I., et al., 2001. Amyloid beta-peptide promotes permeability transition pore in brain mitochondria. Biosci. Rep. 21, 789–800. Moreira, P.I., et al., 2002. Effect of amyloid beta-peptide on permeability transition pore: a comparative study. J. Neurosci. Res. 69, 257–267. Moreira, P.I., et al., 2005. Insulin protects against amyloid beta-peptide toxicity in brain mitochondria of diabetic rats. Neurobiol. Dis. 18, 628–637. Moreira, P.I., et al., 2009a. Mitochondria as a therapeutic target in Alzheimer's disease and diabetes. CNS Neurol. Disord. Drug Targets 8, 492–511. Moreira, P.I., et al., 2009b. An integrative view of the role of oxidative stress, mitochondria and insulin in Alzheimer's disease. J. Alzheimers Dis. 16, 741–761. Moreira, P.I., et al., 2010. Mitochondrial dysfunction is a trigger of Alzheimer's disease pathophysiology. Biochim. Biophys. Acta 1802, 2–10. Muratsugu, M., et al., 1977. Selective electrode for dibenzyl dimethyl ammonium cation as indicator of the membrane potential in biological systems. Biochim. Biophys. Acta 464, 613–619. Ouyang, Y.B., et al., 2000. Is neuronal injury caused by hypoglycemic coma of the necrotic or apoptotic type? Neurochem. Res. 25, 661–667. Oztaş, B., Küçük, M., 1995. Influence of acute arterial hypertension on blood–brain barrier permeability in streptozocin-induced diabetic rats. Neurosci. Lett. 188, 53–56. Perez-Matute, P., et al., 2009. Reactive species and diabetes: counteracting oxidative stress to improve health. Curr. Opin. Pharmacol. 9, 771–779. Ramakrishna, V., Jailkhani, R., 2007. Evaluation of oxidative stress in insulin dependent diabetes mellitus (IDDM) patients. Diagn. Pathol. 2, 22. Rongzhu, L., et al., 2009. Effects of acrylonitrile on antioxidant status of different brain regions in rats. Neurochem. Int. 55, 552–557. Russell, J.W., et al., 2002. High glucose-induced oxidative stress and mitochondrial dysfunction in neurons. FASEB J. 16, 1738–1748. Russo, V.C., et al., 2012. Effects of fluctuating glucose levels on neuronal cells in vitro. Neurochem. Res. 37, 1768–1782. Sharifi, A.M., et al., 2009. Involvement of caspase-8, -9, and ‐3 in high glucose-induced apoptosis in PC12 cells. Neurosci. Lett. 459, 47–51.

12

S. Cardoso et al. / Neurobiology of Disease 49 (2013) 1–12

Sharma, B., Singh, N., 2012. Experimental hypertension induced vascular dementia: Pharmacology, biochemical and behavioral recuperation by angiotensin receptor blocker and acetylcholinesterase inhibitor. Pharmacol. Biochem. Behav. 102, 101–108. Singh, P., et al., 2002. Expression of neuronal plasticity markers in hypoglycemia induced brain injury. Mol. Cell. Biochem. 247, 69–74. Singh, P., et al., 2004. Impact of hypoglycemia and diabetes on CNS: correlation of mitochondrial oxidative stress with DNA damage. Mol. Cell. Biochem. 260, 153–159. Srivastava, A., Shivanandappa, T., 2005. Hexachlorocyclohexane differentially alters the antioxidant status of the brain regions in rat. Toxicology 214, 123–130. St-Pierre, J., et al., 2002. Topology of superoxide production from different sites in the mitochondrial electron transport chain. J. Biol. Chem. 277, 44784–44790. Sullivan, P.G., et al., 2004. Mitochondrial uncoupling as a therapeutic target following neuronal injury. J. Bioenerg. Biomembr. 36, 353–356. Turner, C.P., et al., 2004. A1 adenosine receptors mediate hypoglycemia-induced neuronal injury. J. Mol. Endocrinol. 32, 129–144. Valko, M., et al., 2007. Free radicals and antioxidants in normal physiological functions and human disease. Int. J. Biochem. Cell Biol. 39, 44–84.

Valtorta, F., et al., 2004. Synaptophysin: leading actor or walk-on role in synaptic vesicle exocytosis? Bioessays 26, 445–453. Vatassery, G.T., Younoszai, R., 1978. Alpha tocopherol levels in various regions of the central nervous systems of the rat and guinea pig. Lipids 13, 828–831. Warren, R.E., Frier, B.M., 2005. Hypoglycaemia and cognitive function. Diabetes Obes. Metab. 7, 493–503. Wong, S.H., et al., 1987. Lipoperoxides in plasma as measured by liquidchromatographic separation of malondialdehyde–thiobarbituric acid adduct. Clin. Chem. 33, 214–220. Wright, R.J., et al., 2009. Effects of acute insulin-induced hypoglycemia on spatial abilities in adults with type 1 diabetes. Diabetes Care 32, 1503–1506. Wrighten, S.A., et al., 2009. A look inside the diabetic brain: contributors to diabetesinduced brain aging. Biochim. Biophys. Acta 1792, 444–453. Yan, L.J., et al., 1997. Oxidative damage during aging targets mitochondrial aconitase. Proc. Natl. Acad. Sci. U. S. A. 94, 11168–11172. Zherebitskaya, E., et al., 2009. Development of selective axonopathy in adult sensory neurons isolated from diabetic rats: role of glucose-induced oxidative stress. Diabetes 58, 1356–1364.