Cristina Lemos

Depression as a Glial-Based Synaptic Dysfunction

Recent studies combining pharmacological, behavioral, electrophysiological and molecular approaches indicate that depression results from maladaptive neuroplastic processes occurring in defined frontolimbic circuits responsible for emotional processing such as the prefrontal cortex, hippocampus, amygdala and ventral striatum. However, the exact mechanisms controlling synaptic plasticity that are disrupted to trigger depressive conditions have not been elucidated. Since glial cells (astrocytes and microglia) tightly and dynamically interact with synapses, engaging a bi-directional communication critical for the processing of synaptic information, we now revisit the role of glial cells in the etiology of depression focusing on a dysfunction of the “quad-partite” synapse. This interest is supported by the observations that depressive-like conditions are associated with a decreased density and hypofunction of astrocytes and with an increased microglia “activation” in frontolimbic regions, which is expected to contribute for the synaptic dysfunction present in depression. Furthermore, the traditional culprits of depression (glucocorticoids, biogenic amines, brain-derived neurotrophic factor, BDNF) affect glia functioning, whereas antidepressant treatments (serotonin-selective reuptake inhibitors, SSRIs, electroshocks, deep brain stimulation) recover glia functioning. In this context of a quad-partite synapse, systems modulating glia-synapse bidirectional communication—such as the purinergic neuromodulation system operated by adenosine 5′-triphosphate (ATP) and adenosine—emerge as promising candidates to “re-normalize” synaptic function by combining direct synaptic effects with an ability to also control astrocyte and microglia function. This proposed triple action of purines to control aberrant synaptic function illustrates the rationale to consider the interference with glia dysfunction as a mechanism of action driving the design of future pharmacological tools to manage depression.

Early synaptic deficits in the APP/PS1 mouse model of Alzheimer's disease involve neuronal adenosine A2A receptors.

Synaptic plasticity in the autoassociative network of recurrent connections among hippocampal CA3 pyramidal cells is thought to enable the storage of episodic memory. Impaired episodic memory is an early manifestation of cognitive deficits in Alzheimers disease (AD). In the APP/PS1 mouse model of AD amyloidosis, we show that associative long-term synaptic potentiation (LTP) is abolished in CA3 pyramidal cells at an early stage. This is caused by activation of upregulated neuronal adenosine A2A receptors (A2AR) rather than by dysregulation of NMDAR signalling or altered dendritic spine morphology. Neutralization of A2AR by acute pharmacological inhibition, or downregulation driven by shRNA interference in a single postsynaptic neuron restore associative CA3 LTP. Accordingly, treatment with A2AR antagonists reverts one-trial memory deficits. These results provide mechanistic support to encourage testing the therapeutic efficacy of A2AR antagonists in early AD patients.

Ketone bodies effectively compete with glucose for neuronal acetyl-CoA generation in rat hippocampal slices

Ketone bodies can be used for cerebral energy generation in situ, when their availability is increased as during fasting or ingestion of a ketogenic diet. However, it is not known how effectively ketone bodies compete with glucose, lactate, and pyruvate for energy generation in the brain parenchyma. Hence, the contributions of exogenous 5.0 mM [1‐13C]glucose and 1.0 mM [2‐13C]lactate + 0.1 mM pyruvate (combined [2‐13C]lactate + [2‐13C]pyruvate) to acetyl‐CoA production were measured both without and with 5.0 mM [U‐13C]3‐hydroxybutyrate in superfused rat hippocampal slices by 13C NMR non‐steady‐state isotopomer analysis of tissue glutamate and GABA. Without [U‐13C]3‐hydroxybutyrate, glucose, combined lactate + pyruvate, and unlabeled endogenous sources contributed (mean ± SEM) 70 ± 7%, 10 ± 2%, and 20 ± 8% of acetyl‐CoA, respectively. With [U‐13C]3‐hydroxybutyrate, glucose contributions significantly fell from 70 ± 7% to 21 ± 3% (p < 0.0001), combined lactate + pyruvate and endogenous contributions were unchanged, and [U‐13C]3‐hydroxybutyrate became the major acetyl‐CoA contributor (68 ± 3%) – about three‐times higher than glucose. A direct analysis of the GABA carbon 2 multiplet revealed that [U‐13C]3‐hydroxybutyrate contributed approximately the same acetyl‐CoA fraction as glucose, indicating that it was less avidly oxidized by GABAergic than glutamatergic neurons. The appearance of superfusate lactate derived from glycolysis of [1‐13C]glucose did not decrease significantly in the presence of 3‐hydroxybutyrate, hence total glycolytic flux (Krebs cycle inflow + exogenous lactate formation) was attenuated by 3‐hydroxybutyrate. This indicates that, under these conditions, 3‐hydroxybutyrate inhibited glycolytic flux upstream of pyruvate kinase. Copyright

Adenosine A2b receptors control A1 receptor‐mediated inhibition of synaptic transmission in the mouse hippocampus

Adenosine is a neuromodulator mostly acting through A1 (inhibitory) and A2A (excitatory) receptors in the brain. A2B receptors (A2BR) are Gs/q‐protein‐coupled receptors with low expression in the brain. As A2BR function is largely unknown, we have now explored their role in the mouse hippocampus. We performed electrophysiological extracellular recordings in mouse hippocampal slices, and immunological analysis of nerve terminals and glutamate release in hippocampal slices and synaptosomes. Additionally, A2BR‐knockout (A2BR‐KO) and C57/BL6 mice were submitted to a behavioural test battery (open field, elevated plus‐maze, Y‐maze). The A2BR agonist BAY60‐6583 (300 nm) decreased the paired‐pulse stimulation ratio, an effect prevented by the A2BR antagonist MRS 1754 (200 nM) and abrogated in A2BR‐KO mice. Accordingly, A2BR immunoreactivity was present in 73 ± 5% of glutamatergic nerve terminals, i.e. those immunopositive for vesicular glutamate transporters. Furthermore, BAY 60‐6583 attenuated the A1R control of synaptic transmission, both the A1R inhibition caused by 2‐chloroadenosine (0.1–1 μm) and the disinhibition caused by the A1R antagonist DPCPX (100 nm), both effects prevented by MRS 1754 and abrogated in A2BR‐KO mice. BAY 60‐6583 decreased glutamate release in slices and also attenuated the A1R inhibition (CPA 100 nm). A2BR‐KO mice displayed a modified exploratory behaviour with an increased time in the central areas of the open field, elevated plus‐maze and the Y‐maze and no alteration of locomotion, anxiety or working memory. We conclude that A2BR are present in hippocampal glutamatergic terminals where they counteract the predominant A1R‐mediated inhibition of synaptic transmission, impacting on exploratory behaviour.

High sucrose consumption induces memory impairment in rats associated with electrophysiological modifications but not with metabolic changes in the hippocampus

High sugar consumption is a risk factor for metabolic disturbances leading to memory impairment. Thus, rats subject to high sucrose intake (HSu) develop a metabolic syndrome and display memory deficits. We now investigated if these HSu-induced memory deficits were associated with metabolic and electrophysiological alterations in the hippocampus. Male Wistar rats were submitted for 9 weeks to a sucrose-rich diet (35% sucrose solution) and subsequently to a battery of behavioral tests; after sacrifice, their hippocampi were collected for ex vivo high-resolution magic angle spinning (HRMAS) metabolic characterization and electrophysiological extracellular recordings in slices. HSu rats displayed a decreased memory performance (object displacement and novel object recognition tasks) and helpless behavior (forced swimming test), without altered locomotion (open field). HRMAS analysis indicated a similar hippocampal metabolic profile of HSu and control rats. HSu rats also displayed no change of synaptic transmission and plasticity (long-term potentiation) in hippocampal Schaffer fibers-CA1 pyramid synapses, but had decreased amplitude of long-term depression in the temporoammonic (TA) pathway. Furthermore, HSu rats had an increased density of inhibitory adenosine A1 receptors (A1R), that translated into a greater potency of A1R in Schaffer fiber synapses, but not in the TA pathway, whereas the endogenous activation of A1R in HSu rats was preserved in the TA pathway but abolished in Schaffer fiber synapses. These results suggest that HSu triggers a hippocampal-dependent memory impairment that is not associated with altered hippocampal metabolism but is probably related to modified synaptic plasticity in hippocampal TA synapses.

Hierarchical glucocorticoid-endocannabinoid interplay regulates the activation of the nucleus accumbens by insulin.

Here we asked if insulin activation of the nucleus accumbens in vitro is reflected by an increase in (3)H-deoxyglucose ([(3)H]DG) uptake, thus subserving a new model to study molecular mechanisms of central insulin actions. Additionally, we investigated the dependence of this insulin effect on endocannabinoids and corticosteroids, two major culprits in insulin resistance. We found that in acute accumbal slices, insulin (3 and 300nM but not at 0.3nM) produced an increase in [(3)H]DG uptake. The synthetic cannabinoid agonist, WIN55212-2 (500nM) and the glucocorticoid dexamethasone (10μM), impaired insulin (300nM) action on [(3)H]DG uptake. The glucocorticoid receptor (GcR) antagonist, mifepristone (10μM) prevented dexamethasone from inhibiting insulins action. Strikingly, this anti-insulin action of dexamethasone was also blocked by two CB1 cannabinoid receptor (CB1R) antagonists, O-2050 (500nM) and SR141716A (500nM), as well as by tetrahydrolipstatin (10μM), an inhibitor of diacylglycerol lipases-the enzymes responsible for the synthesis of the endocannabinoid, 2-arachidonoyl-glycerol (2-AG). On the other hand, the blockade of the post-synaptic 2-AG metabolizing enzymes, α,β-serine hydrolase domain 6/12 by WWL70 (1μM) also prevented the action of insulin, probably via increasing endogenous 2-AG tone. Additionally, an anti-insulin receptor (InsR) antibody immunoprecipitated CB1Rs from accumbal homogenates, indicating a physical complexing of CB1Rs with InsRs that supports their functional interaction. Altogether, insulin stimulates glucose uptake in the nucleus accumbens. Accumbal GcR activation triggers the synthesis of 2-AG that in turn binds to the known CB1R-InsR heteromer, thus impeding insulin signaling.

Decreased synaptic plasticity in the medial prefrontal cortex underlies short-term memory deficits in 6-OHDA-lesioned rats

Parkinsons disease (PD) is characterized by motor dysfunction associated with dopaminergic degeneration in the dorsolateral striatum (DLS). However, motor symptoms in PD are often preceded by short-term memory deficits, which have been argued to involve deregulation of medial prefrontal cortex (mPFC). We now used a 6-hydroxydopamine (6-OHDA) rat PD model to explore if alterations of synaptic plasticity in DLS and mPFC underlie short-term memory impairments in PD prodrome. The bilateral injection of 6-OHDA (20μg/hemisphere) in the DLS caused a marked loss of dopaminergic neurons in the substantia nigra (>80%) and decreased monoamine levels in the striatum and PFC, accompanied by motor deficits evaluated after 21 days in the open field and accelerated rotarod. A lower dose of 6-OHDA (10μg/hemisphere) only induced a partial degeneration (about 60%) of dopaminergic neurons in the substantia nigra with no gross motor impairments, thus mimicking an early premotor stage of PD. Notably, 6-OHDA (10μg)-lesioned rats displayed decreased monoamine levels in the PFC as well as short-term memory deficits evaluated in the novel object discrimination and in the modified Y-maze tasks; this was accompanied by a selective decrease in the amplitude of long-term potentiation in the mPFC, but not in DLS, without changes of synaptic transmission in either brain regions. These results indicate that the short-term memory dysfunction predating the motor alterations in the 6-OHDA model of PD is associated with selective changes of information processing in PFC circuits, typified by persistent changes of synaptic plasticity.

Lack of presynaptic interaction between glucocorticoid and CB1 cannabinoid receptors in GABA- and glutamatergic terminals in the frontal cortex of laboratory rodents

Corticosteroid and endocannabinoid actions converge on prefrontocortical circuits associated with neuropsychiatric illnesses. Corticosteroids can also modulate forebrain synapses by using endocannabinoid effector systems. Here, we determined whether corticosteroids can modulate transmitter release directly in the frontal cortex and, in doing so, whether they affect presynaptic CB1 cannabinoid receptor- (CB1R) mediated neuromodulation. By Western blotting of purified subcellular fractions of the rat frontal cortex, we found glucocorticoid receptors (GcRs) and CB1Rs enriched in isolated frontocortical nerve terminals (synaptosomes). CB1Rs were predominantly presynaptically located while GcRs showed preference for the post-synaptic fraction. Additional confocal microscopy analysis of cortical and hippocampal regions revealed vesicular GABA transporter-positive and vesicular glutamate transporter 1-positive nerve terminals endowed with CB1R immunoreactivity, apposing GcR-positive post-synaptic compartments. In functional transmitter release assay, corticosteroids, corticosterone (0.1-10 microM) and dexamethasone (0.1-10 microM) did not significantly affect the evoked release of [(3)H]GABA and [(14)C]glutamate in superfused synaptosomes, isolated from both rats and mice. In contrast, the synthetic cannabinoid, WIN55212-2 (1 microM) diminished the release of both [(3)H]GABA and [(14)C]glutamate, evoked with various depolarization paradigms. This effect of WIN55212-2 was abolished by the CB1R neutral antagonist, O-2050 (1 microM), and was absent in the CB1R KO mice. CB2R-selective agonists did not affect the release of either neurotransmitter. The lack of robust presynaptic neuromodulation by corticosteroids was unchanged upon either CB1R activation or genetic inactivation. Altogether, corticosteroids are unlikely to exert direct non-genomic presynaptic neuromodulation in the frontal cortex, but they may do so indirectly, via the stimulation of trans-synaptic endocannabinoid signaling.

Impaired hippocampal glucoregulation in the cannabinoid CB1 receptor knockout mice as revealed by an optimized in vitro experimental approach

Several techniques exist to study the rate of glucose uptake and metabolism in the brain but most of them are not sufficiently robust to permit extensive pharmacological analysis. Here we optimized an in vitro measurement of the simultaneous accumulation of the metabolizable and non-metabolizable (3)H and (14)C d-glucose analogues; permitting convenient large-scale studies on glucose uptake and metabolism in brain slices. Next, we performed an extensive pharmacological characterization on the putative glucoregulator role of the endocannabinoid system in the hippocampal slices of the rat, and the wild-type and the CB(1) cannabinoid receptor (CB(1)R) knockout mice. We observed that (3)H-3-O-methylglucose is a poor substrate to measure glucose uptake in the hippocampus. (3)H-2-deoxyglucose is a better substrate but its uptake is still lower than that of (14)C-U-d-glucose, from which the slices constantly metabolize and dissipate (14)C atoms. Thus, uptake and the metabolism values are not to be used as standalones but as differences between a control and a treatment. The CB(1)R knockout mice exhibited ∼10% less glucose uptake and glucose carbon atom dissipation in comparison with the wild-type mice. This may represent congenital defects as acute treatments of the rat and mouse slices with cannabinoid agonists, antagonists and inhibitors of endocannabinoid uptake/metabolism failed to induce robust changes in either the uptake or the metabolism of glucose. In summary, we report here an optimized technique ideal to complement other metabolic approaches of high spatiotemporal resolution. This technique allowed us concluding that CB(1)Rs are at least indirectly involved in hippocampal glucoregulation.

Blockade of adenosine A 2A receptors recovers early deficits of memory and plasticity in the triple transgenic mouse model of Alzheimer's disease

Alzheimers disease (AD) begins with a deficit of synaptic function and adenosine A2A receptors (A2AR) are mostly located in synapses controlling synaptic plasticity. The over-activation of adenosine A2A receptors (A2AR) causes memory deficits and the blockade of A2AR prevents memory damage in AD models. We now enquired if this prophylactic role of A2AR might be extended to a therapeutic potential. We used the triple transgenic model of AD (3xTg-AD) and defined that the onset of memory dysfunction occurred at 4 months of age in the absence of locomotor or emotional alterations. At the onset of memory deficits, 3xTg mice displayed a decreased density of markers of excitatory synapses (10.6 ± 3.8% decrease of vGluT1) without neuronal or glial overt damage and an increase of synaptic A2AR in the hippocampus (130 ± 22%). After the onset of memory deficits in 3xTg-AD mice, a three weeks treatment with the selective A2AR antagonist normalized the up-regulation of hippocampal A2AR and restored hippocampal-dependent reference memory, as well as the decrease of hippocampal synaptic plasticity (60.0 ± 3.7% decrease of long-term potentiation amplitude) and the decrease of global (syntaxin-I) and glutamatergic synaptic markers (vGluT1). These findings show a therapeutic-like ability of A2AR antagonists to recover synaptic and memory dysfunction in early AD.