Exogenous ketone bodies (KBs), acetoacetate (AA), and β-hydroxybutyrate (BHB) act as alternative energy substrates in neural cells under starvation. The present study examined the endogenous ketogenic capacity of astroglia under hypoxia with/without glucose and the possible roles of KBs in neuronal energy metabolism. Cultured neurons and astroglia were prepared from Sprague-Dawley rats. Palmitic acid (PAL) and l-carnitine (LC) were added to the assay medium. The 4- to 24-hr production of AA and BHB was measured using the cyclic thio-NADH method. 14C-labeled acid-soluble products (KBs) and 14CO2 produced from [1-14C]PAL were also measured. l-[U-14C]lactic acid ([14C]LAC), [1-14C]pyruvic acid ([14C]PYR), or β-[1-14C]hydroxybutyric acid ([14C]BHB) was used to compare the oxidative metabolism of the glycolysis end products with that of the KBs. Some cells were placed in a hypoxic chamber (1% O2). PAL and LC induced a higher production of KBs in astroglia than in neurons, while the CO2 production from PAL was less than 5% of the KB production in both astroglia and neurons. KB production in astroglia was augmented by the AMP-activated protein kinase activators, AICAR and metformin, as well as hypoxia with/without glucose. Neuronal KB production increased under hypoxia in the absence of PAL and LC. In neurons, [14C]LAC and [14C]PYR oxidation decreased after 24 hr of hypoxia, while [14C]BHB oxidation was preserved. Astroglia responds to ischemia in vitro by enhancing KB production, and astroglia-produced KBs derived from fatty acid might serve as a neuronal energy substrate for the tricarboxylic acid cycle instead of lactate, as pyruvate dehydrogenase is susceptible to ischemia.

BackgroundToll-like receptor 4 (TLR4) plays a pivotal role in the pathophysiology of stroke-induced inflammation. Both astroglia and microglia express TLR4, and endogenous ligands produced in the ischemic brain induce inflammatory responses. Reactive oxygen species (ROS), nitric oxide (NO), and inflammatory cytokines produced by TLR4 activation play harmful roles in neuronal damage after stroke. Although astroglia exhibit pro-inflammatory responses upon TLR4 stimulation by lipopolysaccharide (LPS), they may also play cytoprotective roles via the activation of the pentose phosphate pathway (PPP), reducing oxidative stress by glutathione peroxidase. We investigated the mechanisms by which astroglia reduce oxidative stress via the activation of PPP, using TLR4 stimulation and hypoxia in concert with microglia.MethodsIn vitro experiments were performed using cells prepared from Sprague–Dawley rats. Coexisting microglia in the astroglial culture were chemically eliminated using l-leucine methyl ester (LME). Cells were exposed to LPS (0.01 μg/mL) or hypoxia (1 % O2) for 12–15 h. PPP activity was measured using [1-14C]glucose and [6-14C]glucose. ROS and NO production were measured using 2′,7′-dichlorodihydrofluorescein diacetate and diaminofluorescein-FM diacetate, respectively. The involvement of nuclear factor-erythroid-2-related factor 2 (Nrf2), a cardinal transcriptional factor under stress conditions that regulates glucose 6-phosphate dehydrogenase, the rate-limiting enzyme of PPP, was evaluated using immunohistochemistry.ResultsCultured astroglia exposed to LPS elicited 20 % increases in PPP flux, and these actions of astroglia appeared to involve Nrf2. However, the chemical depletion of coexisting microglia eliminated both increases in PPP and astroglial nuclear translocation of Nrf2. LPS induced ROS and NO production in the astroglial culture containing microglia but not in the microglia-depleted astroglial culture. LPS enhanced astroglial ROS production after glutathione depletion. U0126, an upstream inhibitor of mitogen-activated protein kinase, eliminated LPS-induced NO production, whereas ROS production was unaffected. U0126 also eliminated LPS-induced PPP activation in astroglial–microglial culture, indicating that microglia-derived NO mediated astroglial PPP activation. Hypoxia induced astroglial PPP activation independent of the microglia–NO pathway. Elimination of ROS and NO production by sulforaphane, a natural Nrf2 activator, confirmed the astroglial protective mechanism.ConclusionsAstroglia in concert with microglia may play a cytoprotective role for countering oxidative stress in stroke.Electronic supplementary materialThe online version of this article (doi:10.1186/s12974-016-0564-0) contains supplementary material, which is available to authorized users.

D-Serine is known to be essential for the activation of the N-methyl-D-aspartate (NMDA) receptor in the excitation of glutamatergic neurons, which have critical roles in long-term potentiation and memory formation. D-Serine is also thought to be involved in NMDA receptor-mediated neurotoxicity. The deletion of serine racemase (SRR), which synthesizes D-Serine from L-Serine, was recently reported to improve ischemic damage in mouse middle cerebral artery occlusion model. However, the cell type in which this phenomenon originates and the regulatory mechanism for D-/L-Serine remain elusive. The D-/L-Serine content in ischemic brain increased until 20 hours after recanalization and then leveled off gradually. The results of in vitro experiments using cultured cells suggested that D-Serine is derived from neurons, while L-Serine seems to be released from astroglia. Immunohistochemistry studies of brain tissue after cerebral ischemia showed that SRR is expressed in neurons, and 3-phosphoglycerate dehydrogenase (3-PGDH), which synthesizes L-Serine from 3-phosphoglycerate, is located in astrocytes, supporting the results of the in vitro experiments. A western blot analysis showed that neither SRR nor 3-PGDH was upregulated after cerebral ischemia. Therefore, the increase in D-/L-Serine was not related to an increase in SRR or 3-PGDH, but to an increase in the substrates of SRR and 3-PGDH.

No abstract

Purpose: Although exogenous ketone bodies (KBs), acetoacetate (AA) and β-hydroxybutyrate (BHB) can serve as alternative energy substrates in neural cells under starvation, the exact roles and the regulation of ketogenesis in the brain remain uncertain. The present study examined the ketogenic capacity of cultured rat astroglia under hypoxia and possible roles and the regulation of KBs in neuronal energy metabolism. Methods: Primary neurons and secondary astroglia were prepared from SD rats. Palmitic acid (PA) and L-carnitine (LC) were added to the nutrient medium. AA and BHB produced and released in the medium during 24h were measured using the cyclic thio-NADH method. 14 C-labeled acid-soluble products (i.e., KBs) and 14 CO 2 produced from [1- 14 C]PA were measured to assess the role of PA and KBs as energy substrates in the TCA cycle. [U- 14 C]lactate or [1- 14 C]BHB was used to compare the oxidative metabolism of the end-products of glycolysis with those of the β-oxidation of fatty acid. Some cells were placed in a hypoxic chamber (1%O 2 ) for 12h-24h to evaluate the effects of hypoxia and re-oxygenation on KB metabolism. Results: PA (100 μM) and LC (1 mM) induced higher KB production (mean ± SD in pmol/μg /24h, n=6) in astroglia (AA: 40.8 ± 1.3, BHB: 9.6 ± 1.4) than in neurons (AA: 0.3 ± 1.5, BHB: 2.2 ± 2.7), while CO 2 production from PA was less than 5% of the KB production in astroglia. KB production was augmented by AICAR (500 μM), a cell-permeable AMPK activator (AA: 66.8 ± 8.7, BHB: 12.7 ± 0.9) as well as hypoxia (AA: 59.4 ± 5.1, BHB: 30.0 ± 14.4) in astroglia. [1- 14 C]BHB oxidation (mean ± SD in pmol/μg/h, n=4) in neurons (2.9 ± 0.6) was about 1.5 times as high as that in astroglia (1.8 ± 0.1). [U- 14 C]lactate oxidation in astroglia (14.0 ± 4.6) was not affected by addition of BHB (13.3 ± 4.7) in astroglia, while that in neurons (21.5 ± 2.8) was reduced by 15% (18.1 ± 1.2, P <0.05). Conclusions: Astroglia responded to hypoxia by enhancing KB production in the presence of PA and LC, and KBs produced by astroglia might serve as a neuronal energy substrate for the TCA cycle in place of lactate, since pyruvate dehydrogenase is susceptible to ischemia. The activation of astroglial ketogenesis may reduce ischemic cell damage after stroke.