Early and Persistent O-GlcNAc Protein Modification in the Streptozotocin Model of Alzheimer’s Disease

Abstract

O-GlcNAc transferase (OGT), an enzyme highly expressed in brain tissue, catalyzes the addition of N-acetyl-glucosamine (GlcNAc) to hydroxyl residues of serine and threonine of proteins. Brain protein O-GlcNAcylation is diminished in Alzheimer’s disease (AD), and OGT targets include proteins of the insulin-signaling pathway (e.g., insulin receptor susbtrate-1, IRS-1). We hypothesized that ICV streptozotocin (STZ) also affects O-GlcNAc protein modification. We investigated hippocampal metabolic changes in Wistar rats, particularly OGT levels and insulin resistance, as well as related astroglial activities, immediately after ICV STZ administration (first week) and later on (fourth week). We found an early (at one week) and persistent (at fourth week) decrease in OGT in the ICV STZ model of AD, characterized by a spatial cognitive deficit. Consistent with this observation, we observed a decrease in protein O-GlnNAc modification at both times. Increased phosphorylation at serine-307 of IRS-1, which is related to insulin resistance, was observed on the fourth week. The decrease in OGT and consequent protein O-GlnNAc modifications appear to precede the decrease in glucose uptake and increment of the glyoxalase system observed in the hippocampus. Changes in glial fibrillary acidic protein and S100B in the hippocampus, as well as the alterations in cerebrospinal fluid S100B, confirm the astrogliosis. Moreover, decreases in glutamine synthetase and glutathione content suggest astroglial dysfunction, which are likely implicated in the neurodegenerative cascade triggered in this model. Together, these data contribute to the understanding of neurochemical changes in the ICV STZ model of sporadic AD, and may explain the decreases in protein O-GlcNAc levels and insulin resistance observed in AD.

Keywords

Astrocyte GFAP hippocampus insulin-resistance O-GlcNAc transferase streptozotocin S100B

INTRODUCTION

Alzheimer’s disease (AD) progresses to serious and disabling cognitive impairment, causing enormous social and economic damage [1, 2]. From a histopathological point of view, AD is characterized by extracellular deposits of amyloid-β (Aβ) peptides (senile plaques), derived from neuronmembrane amyloid-β protein precursor (AβPP),and the derangement of neuronal microtubules(neurofibrillary tangles) due to hyperphosphorylation of the tau protein. Such histopathological alterations are accompanied and/or preceded by changes in glucose metabolism and astroglial dysfunction [3]. Late-onset or sporadic AD is found in more than 98% of cases and causes remain elusive.

Both AβPP and tau protein are enzymatically modified by N-acetyl-glucosamine (GlcNAc) in addition to hydroxyl residues of serine and threonine (O-GlcNAc). The addition of O-GlcNAc is catalyzed by uridine diphosphate-N-acetyl-D-glucosamine:polypeptidyl transferase or O-GlcNAc transferase (OGT), which has a trimeric structure comprising two subunits of 110 kDa and one of 78 kDa [4]. This enzyme is ubiquitous, but is highly expressed in the pancreas and brain tissue [5]. The post-translational modification by OGT is dependent on glucose flux through the hexosamine biosynthetic pathway [6].

Therefore, considering the altered glucose metabolism and insulin-resistance observed in AD, it would be interesting to evaluate possible changes in O-GlcNAc protein modification in an animal model of AD. Protein O-GlcNAcylation is reportedly diminished in AD brain tissue [7] and this may be due to reduced glucose flux [6]. Moreover, molecular data of insulin resistance indicate an alteration in O-GlcNAc protein modification, as OGT is able to modify the activities of insulin receptor substrates (IRS) and other downstream targets (e.g., Akt) [8]. Consistent with this hypothesis, there is much evidence to suggest a potential involvement of O-GlcNAc protein modification in AD and other neurodegenerative diseases (see [9] for a review).

Intracerebroventricular (ICV) administration of streptozotocin (STZ) has been used as a valuable model for AD, suitable for understanding neurochemical alterations and the development of therapeutic strategies (see [10] for a review). STZ is an analog of GlcNAc and is able to inhibit theenzyme, O-GlnNAcase (OGA), which removes the GlnNAc residue from proteins [11, 12]. This compound may stimulate OGT activity; however, in diabetic rats (induced by intraperitoneal STZadministration), a decrease in OGT activity and an increase in OGA have been reported [13]. A decrease in protein O-GlcNAcylation was reported in theSTZ model of AD in the cerebrum of rats, but not in the cerebellum [14]. More recently, a study usingthis model of AD reported no changes in the gene expression (mRNA levels) of OGT in mouse hippocampus [15].

Considering the astrocyte dysfunction that precedes and/or accompanies AD, we have investigated astroglial parameters in the STZ model of sporadic AD in rats [16 –18]. We found cognitive impairment and a decreased glucose uptake in these animals, in association with signs of astrocyte dysfunction, such as decreased cerebrospinal fluid (CSF) S100B levels at 4–6 weeks after ICV STZ. We hypothesized that ICV STZ affects O-GlcNAc protein modification in these rats, and that this, together with astroglial dysfunction, contributes to the cognitive impairment observed. In the present study, our aim was to investigate hippocampal metabolic changes in the STZ-induced model of AD, particularly OGT protein levels and insulin resistance, as well as related astroglial activities, immediately after STZ administration (first week) and later on, after administration (fourth week).

MATERIALS AND METHODS

Chemicals

Streptozotocin, methylglyoxal, standard reduced glutathione (GSH), o-phthaldialdehyde (OPA),o-phenylenediamine (OPD), meta-phosphoric acid, cytochalasin B, anti-S100B (SH-B1) and anti-O-linked-N-acetylglucosamine transferase (OGT) were purchased from Sigma (Saint Louis, MO, USA). 2-Deoxy-D-Glucose-3H(G) was purchased from American Radiolabeled Chemicals, Inc. (Saint Louis, MO, USA). Polyclonal anti-S100B, anti-GFAP and anti-rabbit peroxidase-linked antibodies were purchased from DAKO (São Paulo, Brazil) and GE, respectively (Little Chalfont, United Kingdom). Primary antibody anti-phospho-IRS1, anti-panIRS1, anti-O-linked N-Acetylglucosamine (RL2), and anti-Actin were purchased from EMD Millipore (Darmstadt, Germany) and anti-glutamine synthetase (GS) was from Santa Cruz Biotechnology (Texas, USA). All other chemicals were purchased from local commercial suppliers.

Animals

A total of a forty male Wistar rats (90–120 days old) were obtained from our breeding colony (at the Department of Biochemistry, Federal University of Rio Grande do Sul), and were maintained under controlled light and environmental conditions (12 h light/12 h dark cycle at a constant temperature of 22±1°C) with free access to food and water. All animal experiments were carried out in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80–23) revised 1996, and following the regulations of the local animal house authorities.

Rats were divided into two groups: Sham and STZ. After the first week (1 w) and behavioral tasks in the fourth week (4 w), rats were anaesthetized for CSF puncture and hippocampi were dissected. Transverse slices of hippocampi were prepared for the glucose uptake assay or were immediately frozen and stored at –80°C, for the subsequent evaluation of S100B and glial fibrillary acidic protein (GFAP) contents, Glyoxalase 1 (GLO1) activity, GSH content and OGT protein and GlcNAc-protein levels, phospho-IRS1/pan-IRS1, and glutamine synthetase content.

Surgical procedure

Streptozotocin was intracerebroventricularly infused, based on previous studies [16, 18]. Briefly, on the day of the surgery, animals were anesthetized with ketamine/xylazine (80 and 10 mg/kg, respectively, IP) and placed in a stereotaxic apparatus. A midline sagittal incision was made in the scalp. The lateral ventricles were accessed using the following coordinates: 0.9 mm posterior to bregma; 1.5 mm lateral to sagittal suture; 3.6 mm beneath the surface of the brain. Rats received a single bilateral infusion of 5μL STZ (3 mg/kg) or vehicle (Hank’s balanced salt solution –HBSS –containing in mM: 137 NaCl; 0.63 Na2HPO4; 4.17 NaHCO3; 5.36 KCl; 0.44 KH2PO4; 1.26 CaCl2; 0.41 MgSO4; 0.49 MgCl2 and 10 glucose, in pH 7.4) using a Hamilton syringe. After the surgical procedure, rats were placed on a heating pad to maintain body temperature at 37.5±0.5°C and were kept there until recovery from anesthesia. The animals were submitted to behavioral tasks at 3 weeks after STZ infusion and biochemical analysis at 1 and 4 weeks after STZ injection. A schematic design of the experimental procedure is shown in Fig. 1, indicating times of surgery (STZ or vehicle infusion), cognitive behavior and biochemical analysis.

Fig.1

Schematic representation of the experimental protocol. Adult Wistar rats were submitted to ICV STZ administration, assumed as time 0. Neurochemical analyses (in the hippocampus tissue and cerebrospinal fluid) were performed at one week or 4 weeks afterwards. Cognitive analysis by Morris water maze (MWM) was performed only during the last week.

Morris water maze test

Spatial learning and memory of animals were tested in the Morris water maze. The test began three weeks from STZ injection [18, 19]. The apparatus consisted of a circular pool (180 cm diameter, 60 cm high) filled with water (depth 30 cm; 25±1°C), placed in a room with consistently-located spatial cues. An escape platform (10 cm diameter) was placed in the middle of one of the quadrants, 2 cm below the water surface, equidistant from the sidewall and the middle of the pool. The platform provided the only escape from the water and was located in the same quadrant every trial. Four different starting positions were equally spaced around the perimeter of the pool. The rats were placed for four trials (once from each starting position) per session for 5 days, where each trial had a ceiling time of 60 s. The inter-trial interval was 10 min. After each trial, the rats were dried, and returned to their cages at the end of the session. At 24 h after the last training session, the rats were submitted to a test session. Before this session, the submerged platform was removed. The retention test consisted of placing the animals in the water for 60 s. The time to reach the platform (latency in seconds) was measured. The number of crossings over the original position of the platform and time spent in the target quadrant, compared to the opposite quadrant, were measured.

Western blotting

Hippocampi were quickly removed, homogenized in RIPA lysis buffer (150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, 1 mM Na3VO4, 10 mM NaF, 1μg.ml-1 leupeptin, 1 mM PMSF, 50 mM Tris–HCl, pH 7.5), and centrifuged at 10,000 rpm for 10 min. The supernatants were collected and protein concentration was measured using the Lowry method [20]. To obtain representative western blots, equal amounts of lysates of each hippocampus were loaded per lane on 8% and 10% polyacrylamide gels and transferred to nitrocellulose membranes (Hybond ECL, GE Healthcare Bio-Sciences Corp., NJ, USA). Membranes were blocked for 2 h in 5% Tris-buffered saline with Tween 20^® (TTBS) and bovine serum albumin (BSA). Western immunoblot analysis was performed using anti-IRS1 (1 : 1000, EMD Millipore) and anti-phospho-IRS1 (p-IRS1) Ser307 (1 : 1000, EMD Millipore), anti-OGT (1 : 1000, Sigma), anti-GS (1 : 1000, Santa Cruz Biotechnology), anti-O-Linked GlcNAc (1 : 1000, EMD Millipore), and anti-Actin (1 : 2000, EMD Millipore) antibodies. Secondary antibodies used were horseradish peroxidase (HRP)-conjugated anti-mouse IgG (1 : 10,000, Dako) and anti-rabbit IgG (1 : 10,000, Millipore). Chemiluminescent bands were detected using ImageQuant LAS4000 GE Healthcare, and densitometric analyses were performed using Image-J software. Results are expressed as percentages of the control.

Quantification of S100B and GFAP

The S100B contents in the CSF and hippocampal slice were measured by enzyme-linked immunosorbent assay (ELISA), as described previously [21]. Briefly, 50μL of sample plus 50μL of Tris buffer were incubated for 2 h on a microtiter plate that was previously coated with monoclonal anti-S100B. Polyclonal anti-S100 was incubated for 30 min and then peroxidase-conjugated anti-rabbit antibody was added for a further 30 min. The color reaction with OPD was measured at 492 nm. The standard S100B curve ranged from 0.02 to 10 ng.mL-1. ELISA for GFAP in the hippocampal slice was carried out by coating the microtiter plate with 100μL samples overnight at 4°C. Incubation with a rabbit polyclonal anti-GFAP for 2 h was followed by incubation with a secondary antibody conjugated with peroxidase for 1 h, at room temperature; the standard GFAP curve ranged from 0.1 to 7.5 ng.mL-1 [22].

Glucose uptake assay

Glucose uptake was measured in hippocampal slices, as previously standardized [23]. Briefly, slices were incubated for 30 min at 37°C in a Hank’s balanced salt solution (HBSS). The assay was started by the addition of 0.1μCi/mL 2-Deoxy-D-Glucose-3H(G). The incubation was stopped after 30 min by the removal of the medium and rinsing the slices twice with ice-cold HBSS. The slices were then lysed in a solution containing 0.5 M NaOH. Radioactivity was measured in a scintillation counter. Non-specific uptake was determined using 10μM cytochalasin B. Final glucose uptake was obtained by subtracting the non-specific uptake from the total uptake to obtain the specific uptake.

Glyoxalase 1 activity assay

Slices were lysed and homogenized in phosphate buffered saline (PBS), pH 7.4. Subsequently, slices were centrifuged at 13,000 rpm for 15 min at 4°C and the supernatant was used for enzymatic activity and protein content measurements. GLO1 activity was assayed according to [24]. The reaction mixture contained 50 mM sodium-phosphate buffer pH 7.2, 2 mM methylglyoxal and 1 mM GSH (pre-incubated for 30 min at room temperature). Protein from the sample was added to the buffer. The formation of S-(d)-lactoylglutathione was linear and monitored at 240 nm for 15 min at 30°C. A unit of GLO1 activity is defined as the amount of enzyme that catalyzes the formation of 1μmol of S-(d)-lactoylglutathione per minute. Specific activity was calculated in milliunits per milligram of protein (mU.mg-1 protein).

Reduced glutathione (GSH) content assay

The glutathione content was determined as described [25]. The hippocampal slices were homogenized in a 100 mM sodium phosphate buffer, pH 8.0, containing 5 mM EDTA and the protein was precipitated with 1.7% metaphosphoric acid. The supernatant was assayed with OPA (at a concentration of 1 mg.mL-1 methanol) at room temperature for 15 min. Fluorescence was measured using excitation and emission wavelengths of 350 and 420 nm, respectively. A calibration curve was performed with standard GSH solutions at concentrations ranging from 0 to 500μM. GSH concentrations were calculated as nmol.mg-1 protein. Results are expressed as nmol.mg-1 protein.

Statistical analysis

Data are reported as the mean±standard error, the Morris water maze task was analyzed statistically by two-way analysis of variance (ANOVA), followed by Bonferroni’s test, when the F-test was significant. An unpaired Student’s t-test was used for comparing the two samples. Differences were considered to be significant when p < 0.05 (indicated by one asterisk or two, when p < 0.01). All analyses were performed using Prism 6.0 (GraphPad).

RESULTS

STZ-induced dementia causes cognitive and ponderal deficit

One week after ICV-STZ treatment, rats exhibited a decrease in body weight (Fig. 2A, p = 0.0001). This decrease became discrete at 4 weeks, although it remained significant (Fig. 2B, p = 0.0001). However, no changes in glycemia or signs of undernutrition were observed in these animals (data not shown). Cognitive performance was evaluated in the Morris water maze. There was a decline in the average time to find the platform during training (escape latency), from day 2 onwards, only in the sham group (Fig. 2C) (F (4, 36) = 8.511; p = 0.0001). In the test group, escape latency was clearly different between groups (Fig. 2D, p = 0.0001). In addition, STZ rats spent less time in the target quadrant, as compared to the sham group (Fig. 2E, F (1, 18) = 14.23; p = 0.0014). The number of crossings over the platform location was significantly lower in the STZ group, compared to the sham group (Fig. 2F,p = 0.0002).

Fig.2

Body weight changes and cognitive performance of rats submitted to ICV STZ models of AD. All animals were weighed at time 0, before vehicle (sham) or STZ administration, and later at 1 or 4 weeks, before euthanasia. Differences in body weights in grams (Δ) at 1 and 4 weeks are represented in A and B, respectively. C shows the latency scape in MWM during the 5 consecutive days of training; D shows the latency scape on the test day. Times in the target and opposite quadrant (Q) of the water tank are shown in E, and number of crossings over the target are in shown in F. Values are means±standard error of 10 rats in each group. ^*Significant differences from the sham group by Student’s t test or two-way ANOVA (C and E), p < 0.05.

Hippocampal OGT is reduced in the STZ-model of AD

The immunocontent of OGT in the hippocampus of rats after ICV administration of STZ was reduced within first week after administration (Fig. 3A, p = 0.0104) and remained reduced in the fourth week (Fig. 3B, p = 0.0425). We then investigated IRS-1, a target of OGT, by measuring IRS-1 phosphorylation at the Ser 307 site, which is related to insulin resistance. We found an increase in phosphorylation at Ser 307, at 4 weeks (p = 0.0047) (Fig. 3C, D). However, we did not observe differences in the amount of total IRS-1 among the groups and times (Fig. 3E, p = 0.9861 and Fig. 3F, p = 0.9214, respectively). The phospho-IRS-1/IRS-1 ratio (Fig. 3G, H) demonstrates a clear increase in IRS-1 phosphorylation at the fourth week (p = 0.0296).

Fig.3

Hippocampal OGT content and IRS-1 phosphorylation changes at 1 and 4 weeks after ICV STZ administration. A, B) Immunocontents of hippocampal OGT, as determined by western blotting, normalized by actin, and assuming Sham samples as 100%, at 1 or 4 weeks after ICV STZ administration, respectively. C, D) Hippocampal phosphorylation of IRS-1 at the serine 307 site, at 1 and 4 weeks, respectively. Phosphorylation was determined by western blotting, which was normalized by actin with sham samples assumed as 100%. Total IRS-1 contents at 1 and 4 weeks are shown in E and F. IRS-1 phosphorylation at ser 307/IRS-1 at 1 and 4 weeks are shown in G and H, respectively. Values are means±standard error of 9 rats in each group. ^*Significant differences from the sham group by Student’s t test, p < 0.05.

Glucose flux is altered in the STZ-model of AD

Insulin resistance in the brain tissue is accompanied by changes in glucose transport and metabolism. In association with the decrease in OGT protein level, we observed a decrease in protein O-GlcNAc modification in hippocampal tissue at one week (p = 0.0485) and four weeks (p = 0.0052) after ICV STZ administration (Fig. 4A, B). Note that quantification refers to the value of all detected bands and it is possible to observe changes that are specific for each band. We found a decrease in hippocampal glucose uptake at 4 weeks after STZ infusion (p < 0.0001), that was not observed earlier on at one week (p = 0.3430) (Fig. 4C, D). Moreover, we also found an increase in hippocampal glyoxalase 1 (GLO 1) activity at 4 weeks after STZ treatment (p = 0.0213), which was not seen earlier on (p = 0.7143) (Fig. 4E, F). GLO 1 is a key enzyme in the glyoxalase system, and is responsible for converting methylglyoxal (a subproduct of the glycolytic pathway) to lactate. GLO 2 activity was not altered at1 or 4 weeks (data not shown).

Fig.4

Protein O-GlcNAcylation, glucose uptake and glyoxalase 1 activities in ICV STZ-treated rats. Protein O-GlcNAc levels were determined by immunoblotting at 1 and 4 weeks (A and B, respectively) after ICV STZ administration. Quantification refers to the value of all detected bands in the lane. Representative immunoblots are shown below. Glucose uptake and glyoxalase-1 activity were determined in hippocampus tissues at 1 or 4 weeks after ICV STZ treatment by deoxy-3H-glucose incorporation and lactoylgluthatione formation, respectively. Changes in glucose uptake measured in nmol of deoxy-3H-glucose/mg protein/min, at 1 and 4 weeks, are shown in C and D, respectively. E and F show the activities of glyoxalase–1 (GLO-1), measured in milliunits per milligram of protein, at 1 and 4 weeks. Values are means±standard error of 9 rats in each group. ^*Significant differences from the sham group by Student’s t test, p < 0.05.

Changes in astrocyte markers in the STZ-model of AD

Glucose transport and metabolism to lactate (including from methyglyoxal) mainly occur in astrocytes. Two specific astroglial markers, GFAP and S100B, were evaluated. We observed an increase in hippocampal GFAP at one (p = 0.0053) and four (p = 0.0153) weeks after ICV STZ administration (Fig. 5A and B, respectively). On the other hand, a decrease was observed in hippocampal S100B at one week after STZ treatment (p = 0.0335), but an increase was seen at four weeks (p = 0.0234) (Fig. 5C and D, respectively). Moreover, S100B is secreted and extracellular changes also indicate astrocyte activation and/or dysfunction. In fact, we found an increase in cerebrospinal (CSF) S100B at one week (p = 0.0010), but a decrease in this protein in CSF, later on (p = 0.0321) (Fig. 5E, F).

Fig.5

Changes in astrocyte markers at 1 and 4 weeks after ICV STZ administration. Hippocampal Glial fibrillary acidic protein (GFAP) and S100B, and cerebrospinal fluid (CSF) content of S100B were measured by ELISA. A, B) GFAP contents at 1 and 4 weeks after STZ administration, respectively. C, D) S100B content. E, F) CSF S100B concentrations. Values are means±standard error of 8-9 rats in each group. ^*Significant differences from the sham group by Student’s t test, p < 0.05.

Changes in other biomarkers suggest astrocyte dysfunction

Two markers of astroglial activity were also investigated: glutamine synthetase (GS) and glutathione (GSH) content. GS is a specific astrocyte enzyme responsible for the conversion of glutamate to glutamine, which is transported to neurons as a source for neurotransmitter synthesis (glutamate or GABA). We observed a decrease in GS content at one (p = 0.0364) and four (p = 0.0022) weeks after ICV STZ administration (Fig. 6A and B, respectively). Although GSH content is not a specific marker of astrocytes, its brain synthesis and recycling is fully dependent on astrocyte activity. We observed a decrease in GSH in the STZ-model of AD within one week (p = 0.0226), which was maintained at four weeks (p = 0.0075) (Fig. 6C, D).

Fig.6

Glutamine synthetase immunocontent and reduced content of glutathione in ICV STZ models of AD. Glutamine synthetase (GS) content was determined by western blotting; reduced glutathione (GSH) content was measured by fluorescence in the presence ofo-phthaldialdehyde (and expressed as nmol of GSH/mg protein). Changes in GS content, at 1 and 4 weeks, are shown in A and B, respectively. C and D show changes in GSH contents. Values are means±standard error of 9 rats in each group. ^*Significant differences from the sham group by Student’s t test, p < 0.05.

DISCUSSION

AD is the most common type of dementia. AD patients exhibit progressive changes in memory and cognition and cortical brain tissues present extensive neuronal loss and gliosis, in addition to two classic lesions; the accumulation of Aβ peptide and hyperphosphorylation of the tau protein, forming neurofibrillary tangles. ICV STZ administration has been used as an appropriate model for sporadic AD, as it shares several biochemical alterations with the disease, including insulin resistance, Aβ deposition and tau phosphorylation [10, 26]. Moreover, ICV STZ administration has been used to accelerate and exacerbate signals in a transgenic model of AD.

The ICV STZ mechanism of damage to brain cells is unknown, but in the pancreas, it depends on the glucose transporter type 2 (GluT2). In the brain tissue, this transporter appears to be restricted to some hypothalamic neurons and parasynaptic astrocytes [27]. Neuronal damage may explain body weight changes in these animals; however, feeding behavior was not evaluated in these animals.

There is an evident association between protein hypo-O-GlcNAcylation and AD, but the mechanistic connections remain unclear. Several pathways have been proposed to explain the low levels of O-GlcNAc in the brain tissue of AD patients and models (see Fig. 7). About 2–3% of incoming glucose is physiologically shifted to UDP-GlcNAc synthesis, the OGT substrate for O-GlcNAc protein modification [28]. Therefore, the first possible mechanism may be that a reduced glucose flux led to reduced levels of UDP-GlcNAc and consequently to reduced levels of O-GlcNAcylated proteins. In fact, inhibition of UDP-GlcNAc synthesis led to decreased protein O-GlcNAcylation [6, 14].

Fig.7

Schematic representation of pathways related to protein O-GlcNAcylation. Glucose 6-phosphate is involved in three pathways: glycolysis itself (on left), pentose phosphate pathway (PPP), as a source of NADPH (right) and in the hexosamine pathway (down), as a source of UDP-GlcNAc. This compound leads to O-GlcNAc modification of proteins by OGT catalysis. Asterisks indicate three possible mechanisms involved in the hypo-GlcNAcylation observed in AD: ^*1 (due to glucose hypometabolism); ^*2 (due to nitrosylation of OGT, mediated by NO) and ^*3 (due to decreased OGT, as proposed in this study). GFAT, glutamine-fructose-6-phosphate-transaminase; GS, glutamine synthetase; Glu, glutamate; Gln, glutamine; Pr, protein.

Another proposed mechanism involves NO as a mediator. Aβ peptide exposure of neuroblastoma cells led to NO synthesis and then nitrosylation of OGT, which, in turn, reduces OGT activity [29]. Moreover, it is important to mention that a reduced glucose flux or increased NO synthesis, could be accompanied by decreased levels of GSH due to reduced recycling (dependent on glucose for NADPH synthesis) or consumption (as a consequence of NO quenching), respectively. Consistent with this hypothesis, decreased levels of GSH are found in AD and in the STZ-induced model of dementia [17, 30].

Herein, we propose a third possible mechanism to explain decreased O-GlcNAc in ICV STZ-treated rats, namely due to a decrease in hippocampal OGT content. This decrease may explain the reduced levels of protein O-GlcNAcylation in the model. Accordingly, a decrease in hepatic OGT in STZ-induced diabetic rats has been reported [31]. On the other hand, specific increments in proteinO-GlcNAcylation and mislocation of OGT have been found in the diabetic heart of rats [13]. Lower levels of protein O-GlcNAc modification have been reported in the STZ model of AD in the cerebrum of rats, but not in the cerebellum [14]. In another study in mice, using this model of sporadic AD, mRNA levels of OGT and OGA, the enzyme that removes GlcNAc from proteins, were not altered [15]. It is important to mention that OGA has been less than OGT studied in AD [32].

There is much evidence to suggest that insulin resistance in the brain might be the primary event in AD [33], and brain insulin resistance is relatively well characterized in the ICV STZ model of AD [34]. Phosphorylation at Ser 307 of IRS-1 is the basis of the insulin resistance observed in inflammatory and metabolic disorders. Consistent with this, we found an increment of IRS-1 at Ser 307 in the ICV STZ model of AD. Notably, the OGT decrease precedes IRS-1 phosphorylation. However, the mechanism involved in Ser 307 phosphorylation is unclear at moment, and the connection between OGT and insulin resistance is a matter of debate [9]. Ser/Thr residues of IRS-1 are O-GlcNAc-modified, which may affect the phosphorylation status of the protein [35]. It is not known whether Ser 307 of IRS-1, specifically, is a direct target of OGT, but it is possible to conceive that reduced levels of OGT may lead to a decrease in the O-GlcNAc-modification of IRS-1, which could favor phosphorylation at determined site(s), in turn, favoring Ser-307 phosphorylation, recognized as a site related to insulin resistance. However, this issue deserves further experimental investigation. It is noteworthy that the RL2 antibody predominantly recognized bands of lower than 65 kDa in hippocampal tissue, in contrast to observations in the whole brain tissue samples reported by Liu’s work [36], possibly due to methodological differences.

We found a reduced hippocampal glucose uptake in ICV STZ-treated rats at four weeks, in association with the insulin resistance and spatial cognitive deficit observed in these animals. This is in agreement with reduced levels of the glucose transporter subtypes 1 and 3 reported in this model at 3 weeks after STZ [14]. However, we observed a decrease in protein O-GlcNAc modification at one and four weeks in STZ-treated rats. Accordingly, glucose metabolism impairment and reduced protein O-GlcNAcylation modified are observed in AD [6]. Glucose hypometabolism in AD is interpreted as a reduction in neuronal glucose consumption, but it is not possible to ignore concomitant changes in glucose flow through astrocytes, as these cells also use glucose for synthesis of substrates for neuronal metabolism such as glutamine, glutathione and cholesterol, in addition to energetic requirements. Astrocytes exhibit a higher expression and activity of glyoxalase enzymes, responsible for the detoxification of methylglyoxal (MG), a glycolytic subproduct that is increased during a higher glucose flux [37]. We also observed an increase in GLO-1 in ICV STZ-treated animals at four weeks, suggesting that this glucose uptake impairment does not uniformly affect all brain cells and that some cells, possibly astrocytes, are submitted to an increased glucose flux.

Although it seems probable that glucose hypometabolism can lead to decreased protein O-GlcNAcylation, it is possible that a cross-talk may occur, in which glucose transporters and/or glycolytic enzymes in the brain tissue are affected by O-GlcNAcyl changes, causing the decrease in glucose metabolism. In support of this hypothesis, O-GlcNAc modification affects GAPDH in the glycolytic pathway [38] and glucose transporter 4 [39]. However, in peripheral tissues, O-GlcNAc has been proposed to act as a nutrient sensor when glucose is offered to limit glucose metabolism.

The involvement of astrocytes is confirmed once more in the ICV STZ model of AD by changes in GFAP and S100B. Notably, in the brain tissue of AD patients, in addition to Aβ deposition and neurofibrillary tangles, gliosis is always present next to plaques and degenerative neurons. Transgenic animal models of AD that overexpress AβPP show signs of gliosis, but also sign of atrophy in astrocytes far from amyloid deposition [40]. We, herein, describe early and persistent sign of gliosis, based on GFAP immunocontent. Notice that at this time (1 or 4 weeks), no signs of Aβ deposition were observed in this model (data not shown). In fact, Aβ deposition is a long-term finding in the ICV STZ model of AD [41]. Therefore, gliosis precedes Aβ deposition in this model and this is in agreement with the current view that astroglial alterations appear before classic neuronal alterations [3]. S100B is another marker of tissue astrogliosis, but in contrast to GFAP, we observed an increase only at 4 weeks and a decrease at 1 week. These data reinforce the idea of the heterogeneity of glial cells and/or of the glial response. Other studies have reported different changes (in time, intensity or directions) in these proteins under conditions of injury (e.g., [42]). Regardless of the different profiles of changes at one week, both proteins reinforce the idea of hippocampal astrogliosis in the ICV STZ modelof AD.

In contrast to GFAP, which is a cytoskeletal protein, S100B is soluble and a small fraction is secreted by astrocytes. In vitro studies suggest extracellular trophic or apoptotic effects of S100B, depending on its concentration, on neighboring cells [43]. More recently, this protein has been suggested to act as an alarmin (e.g., [44]). Nevertheless, extracellular changes in the levels of this protein (in the CSF, serum, or extracellular medium of in vitro preparations) have been interpreted as signs of astroglial activation, both in vivo and in vitro [45]. An increment of CSF S100B was reported in patients with mild cognitive deficit, but not in the dementia phase of AD [46]. However, a decrease in the CSF content of S100B has been described in the ICV STZ model of AD [16] and other dementia models, such as chronic cerebral hypoperfusion [47] and ICV okadaic acid administration [23]. Herein, we confirmed a later decrease in CSF S100B, at 4 weeks after STZ administration; however, for the first time, we observed an earlier increase at one week post-administration, likely related to the acute injury induced by STZ, as described in other acute brain injury conditions (e.g., [48, 49]).

Interestingly, CSF S100B levels also drop during ketogenic conditions [50, 51] and brain protein O-GlcNAc levels are lower during fasting [52]. Therefore, when brain glucose utilization is reduced, both brain O-GlcNAc levels and CSF S100B decline; however, the relationship between these events, if it exists, is far from clear. We recently reported that insulin modulates S100B secretion in acute hippocampal slices and it is possible that the decrease in the ICV STZ model of AD is related to insulin resistance [53].

Changes in specific astroglial markers do not necessarily reflect any dysfunction in these cells. As such, we investigated two brain markers, which are directly dependent on astrocyte –glutamine synthetase and the content of reduced glutathione. The ICV STZ model of AD displayed alterations in both of these markers, at one and four weeks after STZ administration.

In the brain, astrocytes predominantly express GS and are, therefore, able to synthetize glutamine, the amine source for glutamine-fructose-6-phosphate-transaminase (GFAT), a key regulator enzyme in the hexosamine pathway of UDP-GlcNAc synthesis (Fig. 7). A decrease in GS activity has been proposed in AD genesis [54]; corroborating this, GS activity is decreased in the STZ model of dementia [17]. In addition, a recent meta-analysis of genomic data suggests that GS may represent the link between AD and diabetes mellitus [55]. Our data indicate that this decrease in GS activity is an early alteration in ICV STZ models of AD.

The hippocampal GSH content is also reduced during the first week after ICV STZ administration. Deficient GSH synthesis and recycling in astrocytes, as well as glutathione trafficking between astrocytes and neurons, make neuronal cells particularly susceptible to damage and death [56, 57]. We have shown a decrease in GSH in the ICV STZ model of dementia [18] and present data to indicate the time point at which this occurs.

It is not known whether GS or antioxidant enzymes dependent on GSH are affected by changes in O-GlcNAc levels, but this possibility cannot be ruled out and a cross-talk does seem to exist. Ammonia exposure induces reversible protein O-GlcNAcylation in astrocyte cultures in a GS-dependent manner [38]. Moreover, glutathione depletion by diethyl maleate appears to increase OGT expression and O-GlcNAc protein modification in the rat skeletal muscle [58]. These findings indicate that glutamine synthesis affects GFAT, the regulator enzyme of the hexosamine pathway and that the oxidative environment promotes protein O-GlcNAcmodification.

STZ is an analog compound of GlcNac and therefore could, per se, affect protein O-GlcNAc modification [11]. In fact, STZ acts as an inhibitor of OGA, promoting augmented O-GlcNAc protein modification [12, 59]. However, in the diabetic models induced by STZ, a general decrease in O-GlcNAcylation is observed as a long-term consequence of this compound. Similarly, STZ releases NO, and OGT nitrosylation could reduce its activity, but there is no evidence that NO is long-term mediator of STZ.

In summary, we identified an early (at one week) and persistent (at four weeks) decrease in OGT protein levels in the ICV STZ model of sporadic AD, as characterized by the spatial cognitive deficit in the Morris water maze. We found increased phosphorylation at Ser 307 of IRS-1, a molecular sign of resistance in the insulin signaling. IRS-1 is a target of OGT and increased phosphorylation at this site, which can explain the brain insulin resistance in these animals. However, further experiments will be necessary to identify specific changes in O-GlcNAc levels in targets of OGT. The decrease in OGT and consequent protein O-GlnNAc modifications appear to precede the decrease in glucose uptake and increment of glyoxalase system observed in the hippocampus. The astroglial reactivity in this model was confirmed by hippocampal changes in GFAP and S100B, as well as alterations in CSF S100B. Moreover, decreases in hippocampal glutamine synthetase and GSH content suggest astroglial dysfunction, which is possibly involved in the cognitive decline and neurodegenerative cascade triggered in this model. Together, these data contribute to understanding the biochemical changes and astrocyte dysfunction in the ICV STZ model of AD, and may help to explain cognitive and metabolic alterations (particularly the decrease in protein O-GlcNAc levels and insulin resistance) observed in AD.

Footnotes

ACKNOWLEDGMENTS

This study was supported by the National Council for Scientific and Technological Development (CNPq, Brazil), Ministry of Education (MEC/CAPES, Brazil), State Foundation for Scientific Research of Rio Grande do Sul (FAPERGS), and National Institute of Science and Technology for Excitotoxicity and Neuroprotection (MCT/INCTEN).

Authors’ disclosures available online ().

References

Medeiros

, Chabrier

, LaFerla

(2013) Elucidating the triggers, progression, and effects of Alzheimer’s disease. J Alzheimers Dis 33(Suppl 1), S195–210.

Zetterberg

, Mattsson

(2014) Understanding the cause of sporadic Alzheimer’s disease. Expert Rev Neurother 14, 621–630.

Leclerc

, Abulrob

(2013) Perspectives in molecular imaging using staging biomarkers and immunotherapies in Alzheimer’s disease. ScientificWorldJournal 2013, 589308.

Haltiwanger

, Blomberg

, Hart

(1992) Glycosylation ofnuclear and cytoplasmic proteins. Purification andcharacterization of a uridinediphospho-N-acetylglucosamine:polypeptidebeta-N-acetylglucosaminyltransferase. J Biol Chem 267, 9005–9013.

Kreppel

, Blomberg

, Hart

(1997) Dynamic glycosylation of nuclear and cytosolic proteins. Cloning and characterization of a unique O-GlcNAc transferase with multiple tetratricopeptide repeats. J Biol Chem 272, 9308–9315.

Liu

, Shi

, Tanimukai

, Gu

, Grundke-Iqbal

, Iqbal

, Gong

(2009) Reduced O-GlcNAcylation links lower brain glucose metabolism and tau pathology in Alzheimer’s disease. Brain 132, 1820–1832.

Liu

, Iqbal

, Grundke-Iqbal

, Hart

, Gong

(2004) O-GlcNAcylation regulates phosphorylation of tau: A mechanism involved in Alzheimer’s disease. Proc Natl Acad Sci U S A 101, 10804–10809.

Hart

, Slawson

, Ramirez-Correa

, Lagerlof

(2011) Cross talk between O-GlcNAcylation and phosphorylation: Roles in signaling, transcription, and chronic disease. Annu Rev Biochem 80, 825–858.

Yuzwa

, Vocadlo

(2014) O-GlcNAc and neurodegeneration: Biochemical mechanisms and potential roles in Alzheimer’s disease and beyond. Chem Soc Rev 43, 6839–6858.

10.

Salkovic-Petrisic

, Knezovic

, Hoyer

, Riederer

(2013) Whathave we learned from the streptozotocin-induced animal model ofsporadic Alzheimer’s disease, about the therapeutic strategies inAlzheimer’s research. J Neural Transm (Vienna) 120, 233–252.

11.

Roos

, Xie

, Su

, Clark

, Yang

, Chin

, Paterson

, Kudlow

(1998) Streptozotocin, an analog ofN-acetylglucosamine, blocks the removal of O-GlcNAc fromintracellular proteins. Proc Assoc Am Physicians 110, 422–432.

12.

Akimoto

, Kreppel

, Hirano

, Hart

(2000) Increased O-GlcNAc transferase in pancreas of rats with streptozotocin-induced diabetes. Diabetologia 43, 1239–1247.

13.

Ramirez-Correa

, Ma

, Slawson

, Zeidan

, Lugo-Fagundo

, Xu

, Shen

, Gao

, Caceres

, Chakir

, DeVine

, Cole

, Marchionni

, Paolocci

, Hart

, Murphy

(2015) Removal ofabnormal myofilamentO-GlcNAcylation restores Ca2+sensitivity in diabetic cardiac muscle. Diabetes 64, 3573–3587.

14.

Deng

, Li

, Liu

, Iqbal

, Grundke-Iqbal

, Gong

(2009) Dysregulation of insulin signaling, glucose transporters, O-GlcNAcylation, and phosphorylation of tau and neurofilaments in the brain: Imlication for Alzheimer’s disease. Am J Pathol 175, 2089–2098.

15.

Chen

, Tian

, Liang

, Sun

, Dai

, Lee

, LaFerla

, Grundke-Iqbal

, Iqbal

, Liu

, Gong

(2012) Brain gene expression of a sporadic (icv-STZ Mouse) and a familial mouse model (3xTg-AD mouse) of Alzheimer’s disease. PLoS One 7, e51432.

16.

Rodrigues

, Biasibetti

, Swarowsky

, Leite

, Quincozes-Santos

, Quilfeldt

, Achaval

, Goncalves

(2009) Hippocampal alterations in rats submitted to streptozotocin-induced dementia model are prevented by aminoguanidine. J Alzheimers Dis 17, 193–202.

17.

Tramontina

, Wartchow

, Rodrigues

, Biasibetti

, Quincozes-Santos

, Bobermin

, Tramontina

, Goncalves

(2011) The neuroprotective effect of two statins: Simvastatin and pravastatin on a streptozotocin-induced model of Alzheimer’s disease in rats. J Neural Transm (Vienna) 118, 1641–1649.

18.

Biasibetti

, Tramontina

, Costa

, Dutra

, Quincozes-Santos

, Nardin

, Bernardi

, Wartchow

, Lunardi

, Goncalves

(2013) Green tea (-)epigallocatechin-3-gallate reverses oxidative stress and reduces acetylcholinesterase activity in a streptozotocin-induced model of dementia. Behav Brain Res 236, 186–193.

19.

Morris

(1984) Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Methods 11, 47–60.

20.

Peterson

(1977) A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal Biochem 83, 346–356.

21.

Leite

, Galland

, Brolese

, Guerra

, Bortolotto

, Freitas

, Almeida

, Gottfried

, Goncalves

(2008) A simple, sensitive and widely applicable ELISA for S100B: Methodological features of the measurement of this glial protein. J Neurosci Methods 169, 93–99.

22.

Tramontina

, Leite

, Cereser

, de Souza

, Tramontina

, Nardin

, Andreazza

, Gottfried

, Kapczinski

, Goncalves

(2007) Immunoassay for glial fibrillary acidic protein: Antigen recognition is affected by itshosphorylation state. J Neurosci Methods 162, 282–286.

23.

Broetto

, Hansen

, Brolese

, Batassini

, Lirio

, Galland

, Dos Santos

, Dutra

, Goncalves

(2016) Intracerebroventricular administration of okadaic acid induces hippocampal glucose uptake dysfunction and tau phosphorylation. Brain Res Bull 124, 136–143.

24.

Hansen

, Pandolfo

, Galland

, Torres

, Dutra

, Batassini

, Guerra

, Leite

, Goncalves

(2016) Methylglyoxal can mediate behavioral and neurochemical alterations in rat brain. Physiol Behav 164, 93–101.

25.

Browne

, Armstrong

(1998) Reduced glutathione and glutathione disulfide. Methods Mol Biol 108, 347–352.

26.

Chen

, Liang

, Blanchard

, Dai

, Sun

, Lee

, Grundke-Iqbal

, Iqbal

, Liu

, Gong

(2013) A non-transgenic mouse model (icv-STZ mouse) of Alzheimer’s disease: Similarities to and differences from the transgenic model (3xTg-AD mouse). Mol Neurobiol 47, 711–725.

27.

Nardin

, Tramontina

, Sesterheim

, Rodrigues

, Biasibetti

, Gonçalves

(2014) Cognitive impairment induced by streptozotocin:an exerimental link between diabetes and Alzheimer’s disease. In Streptozotocin: Uses, mechanism of action and side effects. Gauthier

, ed. New Developments in Medical Research, United States.

28.

Marshall

, Bacote

, Traxinger

(1991) Discovery of a metabolic pathway mediating glucose-induced desensitization of the glucose transport system. Role of hexosamine biosynthesis in the induction of insulin resistance. J Biol Chem 266, 4706–4712.

29.

Ryu

, Lee

, Do

(2016) Abeta-affected pathogenic induction of S-nitrosylation of OGT and identification of Cys-NO linkage triplet. Biochim Biophys Acta 1864, 609–621.

30.

Ishrat

, Parveen

, Hoda

, Khan

, Yousuf

, Ansari

, Saleem

, Islam

(2009) Effects of Pycnogenol and vitamin E on cognitive deficits and oxidative damage induced by intracerebroventricular streptozotocin in rats. Behav Pharmacol 20, 567–575.

31.

Majumdar

, Wright

, Markowitz

, Martinez-Hernandez

, Raghow

, Solomon

(2004) Insulin stimulates and diabetes inhibits O-linked N-acetylglucosamine transferase and O-glycosylation of Sp1. Diabetes 53, 3184–3192.

32.

Forster

, Welleford

, Triplett

, Sultana

, Schmitz

, Butterfield

(2014) Increased O-GlcNAc levels correlate with decreased O-GlcNAcase levels in Alzheimer disease brain. Biochim Biophys Acta 1842, 1333–1339.

33.

Bedse

, Di Domenico

, Serviddio

, Cassano

(2015) Aberrant insulin signaling in Alzheimer’s disease: Current knowledge. Front Neurosci 9, 204.

34.

Salkovic-Petrisic

, Hoyer

(2007) Central insulin resistance as a trigger for sporadic Alzheimer-like pathology: An experimental approach. J Neural Transm Suppl 217–233.

35.

Klein

, Berkaw

, Buse

, Ball

(2009) O-linked N-acetylglucosamine modification of insulin receptor substrate-1 occurs in close proximity to multiple SH2 domain binding motifs. Mol Cell Proteomics 8, 2733–2745.

36.

Liu

, Li

, Yu

, Shi

, Liang

, Run

, Li

, Dai

, Grundke-Iqbal

, Iqbal

, Liu

, Gong

(2012) Developmental regulation of protein O-GlcNAcylation, O-GlcNAc transferase, and O-GlcNAcase in mammalian brain. PLoS One 7, e43724.

37.

Allaman

, Belanger

, Magistretti

(2015) Methylglyoxal, the dark side of glycolysis. Front Neurosci 9, 23.

38.

Karababa

, Gorg

, Schliess

, Haussinger

(2014) O-GlcNAcylation as a novel ammonia-induced posttranslational protein modification in cultured rat astrocytes. Metab Brain Dis 29, 975–982.

39.

Buse

, Robinson

, Marshall

, Hresko

, Mueckler

(2002) Enhanced O-GlcNAc protein modification is associated with insulin resistance in GLUT1-overexpressing muscles. Am J Physiol Endocrinol Metab 283, E241–250.

40.

Rodriguez

, Witton

, Olabarria

, Noristani

, Verkhratsky

(2010) Increase in the density of resting microglia precedes neuritic plaque formation and microglial activation in a transgenic model of Alzheimer’s disease. Cell Death Dis 1, e1.

41.

Knezovic

, Osmanovic-Barilar

, Curlin

, Hof

, Simic

, Riederer

, Salkovic-Petrisic

(2015) Staging of cognitive deficits and neuropathological and ultrastructural changes in streptozotocin-induced rat model of Alzheimer’s disease. J Neural Transm (Vienna) 122, 577–592.

42.

Vicente

, Boer

, Leite

, Silva

, Tramontina

, Porciuncula

, Dalmaz

, Goncalves

(2004) Cerebrospinal fluid S100B increases reversibly in neonates of methyl mercury-intoxicated pregnant rats. Neurotoxicology 25, 771–777.

43.

Donato

, Sorci

, Riuzzi

, Arcuri

, Bianchi

, Brozzi

, Tubaro

, Giambanco

(2009) S100B’s double life: Intracellular regulator and extracellular signal. Biochim Biophys Acta 1793, 1008–1022.

44.

, Wang

, Won

, Basu

, Kapfhamer

, Swanson

(2016) Microglial activation induced by the alarmin S100B is regulated by poly(ADP-ribose) polymerase-1. Glia 64, 1869–1878.

45.

Goncalves

, Leite

, Nardin

(2008) Biological and methodological features of the measurement of S100B, a putative marker of brain injury. Clin Biochem 41, 755–763.

46.

Peskind

, Griffin

, Akama

, Raskind

, Van Eldik

(2001) Cerebrospinal fluid S100B is elevated in the earlier stages of Alzheimer’s disease. Neurochem Int 39, 409–413.

47.

Vicente

, Degerone

, Bohn

, Scornavaca

, Pimentel

, Leite

, Swarowsky

, Rodrigues

, Nardin

, de Almeida

, Gottfried

, Souza

, Netto

, Goncalves

(2009) Astroglial and cognitive effects of chronic cerebral hypoperfusion in the rat. Brain Res 1251, 204–212.

48.

Kleindienst

, Hesse

, Bullock

, Buchfelder

(2007) The neurotrophic protein S100B: Value as a marker of brain damage and possible therapeutic implications. Prog Brain Res 161, 317–325.

49.

Batassini

, Broetto

, Tortorelli

, Borsoi

, Zanotto

, Galland

, Souza

, Leite

, Goncalves

(2015) Striatal injury with 6-OHDA transiently increases cerebrospinal GFAP and S100B. Neural Plast 2015, 387028.

50.

Ziegler

, Oliveira

, Pires

, Ribeiro

, Leite

, Mendez

, Goncalves

, Tramontina

, Portela

, Wofchuk

, Perry

, Goncalves

(2004) Ketogenic diet fed rats have low levels of S100B in cerebrospinal fluid. Neurosci Res 50, 375–379.

51.

Vizuete

, de Souza

, Guerra

, Batassini

, Dutra

, Bernardi

, Costa

, Goncalves

(2013) Brain changes in BDNF and S100B induced by ketogenic diets in Wistar rats. Life Sci 92, 923–928.

52.

, Lu

, Wang

, Gong

(2006) Concurrent alterations of O-GlcNAcylation and phosphorylation of tau in mouse brains during fasting. Eur J Neurosci 23, 2078–2086.

53.

Wartchow

, Tramontina

, de Souza

, Biasibetti

, Bobermin

, Goncalves

(2016) Insulin stimulates S100B secretion and these proteins antagonistically modulate brain glucose metabolism. Neurochem Res 41, 1420–1429.

54.

Butterfield

(2002) Amyloid beta-peptide (1-42)-induced oxidative stress and neurotoxicity: Implications for neurodegeneration in Alzheimer’s disease brain. A review. Free Radic Res 36, 1307–1313.

55.

Mirza

, Ali

, Ashraf

, Kamal

, Abuzenadah

, Choudhary

, Damanhouri

, Sheikh

(2014) Proteomics approaches to understand linkage between Alzheimer’s disease and type 2 diabetes mellitus. CNS Neurol Disord Drug Targets 13, 213–225.

56.

Dringen

(2000) Metabolism and functions of glutathione in brain. Prog Neurobiol 62, 649–671.

57.

Fernandez-Fernandez

, Almeida

, Bolanos

(2012) Antioxidant and bioenergetic coupling between neurons and astrocytes. Biochem J 443, 3–11.

58.

Peternelj

, Marsh

, Strobel

, Matsumoto

, Briskey

, Dalbo

, Tucker

, Coombes

(2015) Glutathione depletion and acute exercise increase O-GlcNAc protein modification in rat skeletal muscle. Mol Cell Biochem 400, 265–275.

59.

Toleman

, Paterson

, Shin

, Kudlow

(2006) Streptozotocin inhibits O-GlcNAcase via the production of a transition state analog. Biochem Biophys Res Commun 340, 526–534.