A Mechanism of Action of Metformin in the Brain: Prevention of Methylglyoxal-Induced Glutamatergic Impairment in Acute Hippocampal Slices.

Metformin, a biguanide compound (N-1,1-dimethylbiguanide), is widely prescribed for diabetes mellitus type 2 (T2D) treatment. It also presents a plethora of properties, such as anti-oxidant, anti-inflammatory, anti-apoptosis, anti-tumorigenic, and anti-AGE formation activity. However, the precise mechanism of action of metformin in the central nervous system (CNS) needs to be clarified. Herein, we investigated the neuroprotective role of metformin in acute hippocampal slices exposed to methylglyoxal (MG), a highly reactive dicarbonyl compound and a key molecule in T2D developmental pathophysiology. Metformin protected acute hippocampal slices from MG-induced glutamatergic neurotoxicity and neuroinflammation by reducing IL-1ß synthesis and secretion and RAGE protein expression. The drug also improved astrocyte function, particularly with regard to the glutamatergic system, increasing glutamate uptake. Moreover, we observed a direct effect of metformin on glutamate transporters, where the compound prevented glycation, by facilitating enzymatic phosphorylation close to Lys residues, suggesting a new neuroprotective role of metformin via PKC ζ in preventing dysfunction in glutamatergic system induced by MG. Proposed neuroprotection role of metformin in acute hippocampal slices against impairment in glutamatergic system induced in a model of methylglyoxal glycotoxicity. Metformin reversed methylglyoxal (MG)-induced neuroinflammation by reducing pro-inflammatory IL-1ß synthesis and secretion and RAGE protein expression. Metformin did not alter the effect of MG on S100B secretion (1). Both MG and metformin also influenced astrocyte function in hippocampal slices. Metformin did not reverse the elevation in GLO1 activity induced by glycotoxicity; however, it abrogated the high GSH level and the expression of the co-factor of GLO1 (2). Both treatments per se changed bioenergetic metabolism and increased glucose uptake, extracellular lactate content, and pyruvate kinase (PK) activity. The increment in glucose uptake and lactate levels ceased during the co-incubation of MG with metformin. Metformin reversed the elevation of hexokinase activity by MG (3). We suggest a new role of metformin in the glutamate system, whereby it protects the hippocampus against the derangements of the glutamatergic system induced by MG, possibly by phosphorylation via PKC ζ (4). The neuroprotective action of metformin may be mediated by the phosphorylation of specific amino acid residues (Lysine) of the glutamate transporters (GLAST and GLT-1), since metformin activated the PKC ζ signaling and promoted cascades of phosphorylation in p38 MAPK and Akt proteins. The transporter protein phosphorylation prevented the Lys-glycation and the impairment of glutamate uptake induced by MG (5).

Lipopolysaccharide impairs neurodevelopment and induces changes in astroglial reactivity, antioxidant defenses and bioenergetics in the cerebral cortex of neonatal rats.

Neonates have an immature immune system, which increases their vulnerability to infectious agents and inflammatory insults. The administration of the immunostimulatory agent lipopolysaccharide (LPS) has been shown to induce the expression of pro-inflammatory cytokines and cause behavior alterations in rodents at different ages. However, the effects of LPS administration during the neonatal period and its consequences during immune system maturation remain to be elucidated. We showed here that a single intraperitoneal administration of LPS in rats on postnatal day (PND) 7 caused early and variable alterations in TNF-α, S100B and GFAP levels in the cerebral cortex, CSF and serum of the animals, indicating long-term induction of neuroinflammation and astroglial reactivity. However, on PND 21, only GFAP levels were increased by LPS. Additionally, LPS induced oxidative stress and altered energy metabolism enzymes in the cerebral cortex on PND 21, and caused neurodevelopment impairment over time. These data suggest that neuroinflammation induction during the neonatal period induces glial reactivity, oxidative stress and bioenergetic disruption that may lead to neurodevelopment impairment and cognitive deficit in adult life.

Astroglial Alterations in the Hippocampus of Rats Submitted to a Single Trans-Cranial Direct Current Stimulation Trial.

Evidence indicates that transcranial direct current stimulation (tDCS) provides therapeutic benefits in different situations, such as epilepsy, depression, inflammatory and neuropathic pain. Despite the increasing use of tDCS, its cellular and molecular basis remains unknown. Astrocytes display a close functional and structural relationship with neurons and have been identified as mediators of neuroprotection in tDCS. Considering the importance of hippocampal glutamatergic neurotransmission in nociceptive pathways, we decided to investigate short-term changes in the hippocampal astrocytes of rats subjected to tDCS, evaluating specific cellular markers (GFAP and S100B), as well as markers of astroglial activity; glutamate uptake, glutamine synthesis by glutamine synthetase (GS) and glutathione content. Data clearly show that a single session of tDCS increases the pain threshold elicited by mechanical and thermal stimuli, as evaluated by von Frey and hot plate tests, respectively. These changes involve inflammatory and astroglial neurochemical changes in the hippocampus, based on specific changes in cell markers, such as S100B and GS. Alterations in S100B were also observed in the cerebrospinal fluid of tDCS animals and, most importantly, specific functional changes (increased glutamate uptake and increased GS activity) were detected in hippocampal astrocytes. These findings contribute to a better understanding of tDCS as a therapeutic strategy for nervous disorders and reinforce the importance of astrocytes as therapeutic targets.

Homocysteine May Decrease Glucose Uptake and Alter the Akt/GSK3ß/GLUT1 Signaling Pathway in Hippocampal Slices: Neuroprotective Effects of Rivastigmine and Ibuprofen.

Homocysteine (Hcy) is a risk factor for neurodegenerative diseases, such as Alzheimer's Disease, and is related to cellular and tissue damage. In the present study, we verified the effect of Hcy on neurochemical parameters (redox homeostasis, neuronal excitability, glucose, and lactate levels) and the Serine/Threonine kinase B (Akt), Glucose synthase kinase-3ß (GSK3ß) and Glucose transporter 1 (GLUT1) signaling pathway in hippocampal slices, as well as the neuroprotective effects of ibuprofen and rivastigmine alone or in combination in such effects. Male Wistar rats (90 days old) were euthanized and the brains were dissected. The hippocampus slices were pre-treated for 30 min [saline medium or Hcy (30 µM)], then the other treatments were added to the medium for another 30 min [ibuprofen, rivastigmine, or ibuprofen + rivastigmine]. The dichlorofluorescein formed, nitrite and Na+, K+-ATPase activity was increased by Hcy at 30 µM. Ibuprofen reduced dichlorofluorescein formation and attenuated the effect of Hcy. The reduced glutathione content was reduced by Hcy. Treatments with ibuprofen and Hcy + ibuprofen increased reduced glutathione. Hcy at 30 µM caused a decrease in hippocampal glucose uptake and GLUT1 expression, and an increase in Glial Fibrillary Acidic Protein-protein expression. Phosphorylated GSK3ß and Akt levels were reduced by Hcy (30 µM) and co-treatment with Hcy + rivastigmine + ibuprofen reversed these effects. Hcy toxicity on glucose metabolism can promote neurological damage. The combination of treatment with rivastigmine + ibuprofen attenuated such effects, probably by regulating the Akt/GSK3ß/GLUT1 signaling pathway. Reversal of Hcy cellular damage by these compounds may be a potential neuroprotective strategy for brain damage.

Arundic acid (ONO-2506) downregulates neuroinflammation and astrocyte dysfunction after status epilepticus in young rats induced by Li-pilocarpine.

Astrocytes, the most abundant glial cells, have several metabolic functions, including ionic, neurotransmitter and energetic homeostasis for neuronal activity. Reactive astrocytes and their dysfunction have been associated with several brain disorders, including the epileptogenic process. Glial Fibrillary Acidic Protein (GFAP) and S100 calcium-binding protein B (S100B) are astrocyte biomarkers associated with brain injury. We hypothesize that arundic acid (ONO-2506), which is known as an inhibitor of S100B synthesis and secretion, protects the hippocampal tissue from neuroinflammation and astrocyte dysfunction after status epileptics (SE) induction by Li-pilocarpine in young rats. Herein, we investigated the effects of arundic acid treatment, at time points of 6 or 24 h after the induction of SE by Li-pilocarpine, in young rats. In SE animals, arundic acid was able to prevent the damage induced by Li-pilocarpine in the hippocampus, decreasing neuroinflammatory signaling (reducing IL-1ß, COX2, TLR4 and RAGE contents), astrogliosis (decreasing GFAP and S100B) and astrocytic dysfunction (recovering levels of GSH, glutamine synthetase and connexin-43). Furthermore, arundic acid improved glucose metabolism and reduced the glutamate excitotoxicity found in epilepsy. Our data reinforce the role of astrocytes in epileptogenesis development and the neuroprotective role of arundic acid, which modulates astrocyte function and neuroinflammation in SE animals.

Early effects of LPS-induced neuroinflammation on the rat hippocampal glycolytic pathway.

Neuroinflammation is a common feature during the development of neurological disorders and neurodegenerative diseases, where glial cells, such as microglia and astrocytes, play key roles in the activation and maintenance of inflammatory responses in the central nervous system. Neuroinflammation is now known to involve a neurometabolic shift, in addition to an increase in energy consumption. We used two approaches (in vivo and ex vivo) to evaluate the effects of lipopolysaccharide (LPS)-induced neuroinflammation on neurometabolic reprogramming, and on the modulation of the glycolytic pathway during the neuroinflammatory response. For this, we investigated inflammatory cytokines and receptors in the rat hippocampus, as well as markers of glial reactivity. Mitochondrial respirometry and the glycolytic pathway were evaluated by multiple parameters, including enzymatic activity, gene expression and regulation by protein kinases. Metabolic (e.g., metformin, 3PO, oxamic acid, fluorocitrate) and inflammatory (e.g., minocycline, MCC950, arundic acid) inhibitors were used in ex vivo hippocampal slices. The induction of early inflammatory changes by LPS (both in vivo and ex vivo) enhanced glycolytic parameters, such as glucose uptake, PFK1 activity and lactate release. This increased glucose consumption was independent of the energy expenditure for glutamate uptake, which was in fact diverted for the maintenance of the immune response. Accordingly, inhibitors of the glycolytic pathway and Krebs cycle reverted neuroinflammation (reducing IL-1ß and S100B) and the changes in glycolytic parameters induced by LPS in acute hippocampal slices. Moreover, the inhibition of S100B, a protein predominantly synthesized and secreted by astrocytes, inhibition of microglia activation and abrogation of NLRP3 inflammasome assembly confirmed the role of neuroinflammation in the upregulation of glycolysis in the hippocampus. Our data indicate a neurometabolic glycolytic shift, induced by inflammatory activation, as well as a central and integrative role of astrocytes, and suggest that interference in the control of neurometabolism may be a promising strategy for downregulating neuroinflammation and consequently for diminishing negative neurological outcomes.

The Methylglyoxal/RAGE/NOX-2 Pathway is Persistently Activated in the Hippocampus of Rats with STZ-Induced Sporadic Alzheimer's Disease.

Alzheimer's disease (AD) is the leading cause of dementia in humans, with a high social and economic cost. AD is predominantly a sporadic disease, and the intracerebroventricular (ICV) administration of streptozotocin (STZ) has been widely used as an AD-like model of dementia. While the etiology of AD remains unknown, changes such as glucose metabolism and activation of receptors for advanced glycation end products (RAGE) seem to underlie its pathogenesis. We hypothesized that methylglyoxal, an endogenous toxin derived from the glycolytic pathway, could be the precursor of advanced glycated end products that activates RAGE and that, consequently, may activate membrane NADPH oxidase (NOX), contributing to the inflammatory status of the model and the disease. We administered ICV-STZ to Wistar rats and evaluated several neurochemical parameters in the hippocampus, particularly glyoxalase 1 (GLO-1) activity, which serves as an index of high levels of methylglyoxal, and the contents of RAGE and NOX-2, the most abundant brain NOX isoform. At the times evaluated (4 and 24 weeks after STZ), we observed cognitive deficit, increased beta-amyloid content, and increased tau phosphorylation. A persistent increase in GLO-1 activity was found, as well as increases in RAGE and NOX-2 contents, suggesting astroglial and microglial commitment. The increase in NOX-2 may reflect elevated microglial activity (confirmed by IBA-1 marker), which may contribute to the synaptic dysfunction and pruning described in the literature, both in this model and AD patients. Furthermore, reinforcing this possibility, we found a reduction in cholinergic communication in the hippocampus (as shown by decreased choline acetyltransferase), a reduction in BDNF, and an increase in TGF-ß, the combination of which may result in synaptic deterioration.

Arundic acid (ONO-2526) inhibits stimulated-S100B secretion in inflammatory conditions.

Astrocytes respond to injury by modifying the expression profile of several proteins, including the S100 calcium-binding protein B (S100B), assumed to be a marker as well as a mediator of brain injury. AA is an inhibitor of S100B synthesis and plays a protective role in different models of brain injury, as decreases in S100B expression cause decreases in extracellular S100B. However, S100B mRNA expression, S100B protein content and S100B secretion do not always occur in association; as such, we herein investigated the effect of AA on S100B secretion, using different approaches with three stimulating conditions for S100B secretion, namely, low potassium medium, TNF-α (in hippocampal slices) and LPS exposure (in astrocyte cultures). Our data indicate that AA directly affects S100B secretion, indicating that the extracellular levels of this astroglial protein may be mediating the action of this compound. More importantly, AA had no effect on basal S100B secretion, but inhibited stimulated S100B secretion (stimulated either by the proinflammatory molecules, LPS or TNF-α, or by low potassium medium). Data from hippocampal slices that were directly exposed to AA, or from animals that received the acid by intracerebroventricular infusion, contribute to understanding its neuroprotective effect.

Anti-inflammatory effect of rosmarinic acid isolated from Blechnum brasiliense in adult zebrafish brain.

Neuroinflammation has been associated to neurodegenerative disease development, with evidence suggesting that high levels of proinflammatory cytokines promote neuronal dysfunction and death. Therefore, it is necessary to study new compounds that may be used as adjuvant treatments of neurodegenerative diseases by attenuating the inflammatory response in the central nervous system (CNS). The aim of this study was to utilize the lipopolysaccharide (LPS) induction model of neuroinflammation to evaluate the modulation of inflammation by rosmarinic acid (RA) isolated from Blechnum brasiliense in adult zebrafish. First, we investigated the toxicity and antioxidant properties of fractionated B. brasiliense extract (ethyl acetate fraction- EAF) and the isolated RA in zebrafish embryos. Next, we developed a model of neuroinflammation induction by intraperitoneal (i.p.) injection of LPS to observe the RA modulation of proinflammatory cytokines. The median lethal concentration (LC50) calculated was 185.2 ± 1.24 µg/mL for the ethyl acetate fraction (EAF) and 296.0 ± 1.27 µM for RA. The EAF showed free radical inhibition ranging from 23.09% to 63.44% at concentrations of 10-250 µg/mL. The RA presented a concentration-dependent response ranging from 18.24% to 47.63% at 10-250 µM. Furthermore, the RA reduced LPS induction of TNF-α and IL-1ß levels, with the greatest effect observed 6 h after LPS administration. Thus, the data suggested an anti-inflammatory effect of RA isolated from B. brasiliense and reinforced the utility of the new model of neuroinflammation to test the possible neuroprotective effects of novel drugs or compounds.

GABAA Modulation of S100B Secretion in Acute Hippocampal Slices and Astrocyte Cultures.

Astrocytes are the major glial cells in brain tissue and are involved, among many functions, ionic and metabolic homeostasis maintenance of synapses. These cells express receptors and transporters for neurotransmitters, including GABA. GABA signaling is reportedly able to affect astroglial response to injury, as evaluated by specific astrocyte markers such as glial fibrillary acid protein and the calcium-binding protein, S100B. Herein, we investigated the modulatory effects of the GABAA receptor on astrocyte S100B secretion in acute hippocampal slices and astrocyte cultures, using the agonist, muscimol, and the antagonists pentylenetetrazol (PTZ) and bicuculline. These effects were analyzed in the presence of tetrodotoxin (TTX), fluorocitrate (FLC), cobalt and barium. PTZ positively modify S100B secretion in hippocampal slices and astrocyte cultures; in contrast, bicuculline inhibited S100B secretion only in hippocampal slices. Muscimol, per se, did not change S100B secretion, but prevented the effects of PTZ and bicuculline. Moreover, PTZ-induced S100B secretion was prevented by TTX, FLC, cobalt and barium indicating a complex GABAA communication between astrocytes and neurons. The effects of two putative agonists of GABAA, ß-hydroxybutyrate and methylglyoxal, on S100B secretion were also evaluated. In view of the neurotrophic role of extracellular S100B under conditions of injury, our data reinforce the idea that GABAA receptors act directly on astrocytes, and indirectly on neurons, to modulate astroglial response.

Effects of dexamethasone on the Li-pilocarpine model of epilepsy: protection against hippocampal inflammation and astrogliosis.

BACKGROUND: Temporal lobe epilepsy (TLE) is the most common form of partial epilepsy and is accompanied, in one third of cases, by resistance to antiepileptic drugs (AED). Most AED target neuronal activity modulated by ionic channels, and the steroid sensitivity of these channels has supported the use of corticosteroids as adjunctives to AED. Assuming the importance of astrocytes in neuronal activity, we investigated inflammatory and astroglial markers in the hippocampus, a key structure affected in TLE and in the Li-pilocarpine model of epilepsy. METHODS: Initially, hippocampal slices were obtained from sham rats and rats subjected to the Li-pilocarpine model of epilepsy, at 1, 14, and 56 days after status epilepticus (SE), which correspond to the acute, silent, and chronic phases. Dexamethasone was added to the incubation medium to evaluate the secretion of S100B, an astrocyte-derived protein widely used as a marker of brain injury. In the second set of experiments, we evaluated the in vivo effect of dexamethasone, administrated at 2 days after SE, on hippocampal inflammatory (COX-1/2, PGE2, and cytokines) and astroglial parameters: GFAP, S100B, glutamine synthetase (GS) and water (AQP-4), and K+ (Kir 4.1) channels. RESULTS: Basal S100B secretion and S100B secretion in high-K+ medium did not differ at 1, 14, and 56 days for the hippocampal slices from epileptic rats, in contrast to sham animal slices, where high-K+ medium decreased S100B secretion. Dexamethasone addition to the incubation medium per se induced a decrease in S100B secretion in sham and epileptic rats (1 and 56 days after SE induction). Following in vivo dexamethasone administration, inflammatory improvements were observed, astrogliosis was prevented (based on GFAP and S100B content), and astroglial dysfunction was partially abrogated (based on Kir 4.1 protein and GSH content). The GS decrease was not prevented by dexamethasone, and AQP-4 was not altered in this epileptic model. CONCLUSIONS: Changes in astroglial parameters emphasize the importance of these cells for understanding alterations and mechanisms of epileptic disorders in this model. In vivo dexamethasone administration prevented most of the parameters analyzed, reinforcing the importance of anti-inflammatory steroid therapy in the Li-pilocarpine model and possibly in other epileptic conditions in which neuroinflammation is present.

Phase-Dependent Astroglial Alterations in Li-Pilocarpine-Induced Status Epilepticus in Young Rats.

Epilepsy prevalence is high in infancy and in the elderly population. Lithium-pilocarpine is widely used to induce experimental animal models of epilepsy, leading to similar neurochemical and morphological alterations to those observed in temporal lobe epilepsy. As astrocytes have been implicated in epileptic disorders, we hypothesized that specific astroglial changes accompany and contribute to epileptogenesis. Herein, we evaluated time-dependent astroglial alterations in the hippocampus of young (27-day-old) rats at 1, 14 and 56 days after Li-pilocarpine-induced status epilepticus (SE), corresponding to different phases in this model of epilepsy. We determined specific markers of astroglial activation: GFAP, S100B, glutamine synthetase (GS), glutathione (GSH) content, aquaporin-4 (AQP-4) and potassium channel Kir 4.1; as well as epileptic behavioral, inflammatory and neurodegenerative changes. Phase-dependent signs of hippocampal astrogliosis were observed, as demonstrated by increments in GFAP, S100B and GS. Astrocyte dysfunction in the hippocampus was characterized, based on the decrease in GSH content, AQP-4 and Kir 4.1 channels. Degenerating neurons were identified by Fluoro-Jade C staining. We found a clear, early (at SE1) and persistent (at SE56) increase in cerebrospinal fluid (CSF) S100B levels. Additionally, serum S100B was found to decrease soon after SE induction, implicating a rapid-onset increase in the CSF/serum S100B ratio. However, serum S100B increased at SE14, possibly reflecting astroglial activation and/or long-term increase in cerebrovascular permeability. Moreover, we suggest that peripheral S100B levels may represent a useful marker for SE in young rats and for follow up during the chronic phases of this model of epilepsy. Together, results reinforce and extend the idea of astroglial involvement in epileptic disorders.