Ketogenic diet versus ketoacidosis: what determines the influence of ketone bodies on neurons?

Glucose is the main energy substrate for neurons, however, at certain conditions, e.g. in starvation, these cells could also use ketone bodies. This approach is used in clinical conditions as the ketogenic diet. The ketogenic diet is actually a biochemical model of fasting. It includes replacing carbohydrates by fats in daily meal. Synthesis of ketone bodies ß-hydroxubutirate, acetoacetate and acetone begins once glycogen stores have depleted in the liver. The ketogenic diet can be used to treat clinical conditions, primarily epilepsy. The mechanism of neuroprotective action of ketogenic diet is not very clear. It is shown that ketone bodies influence neurons at three different levels, namely, metabolic, signaling and epigenetic levels. Ketone bodies are not always neuroprotective. Sometimes they can be toxic for the brain. Ketoacidosis which is a very dangerous complication of diabetes mellitus or alcoholism can be taken as an example. The exact mechanism of how neuroprotective properties of ketone bodies reverse to neurotoxic is yet to be established.

Metabolic regulation of synaptic activity.

Brain tissue is bioenergetically expensive. In humans, it composes approximately 2% of body weight and accounts for approximately 20% of calorie consumption. The brain consumes energy mostly for ion and neurotransmitter transport, a process that occurs primarily in synapses. Therefore, synapses are expensive for any living creature who has brain. In many brain diseases, synapses are damaged earlier than neurons start dying. Synapses may be considered as vulnerable sites on a neuron. Ischemic stroke, an acute disturbance of blood flow in the brain, is an example of a metabolic disease that affects synapses. The associated excessive glutamate release, called excitotoxicity, is involved in neuronal death in brain ischemia. Another example of a metabolic disease is hypoglycemia, a complication of diabetes mellitus, which leads to neuronal death and brain dysfunction. However, synapse function can be corrected with "bioenergetic medicine". In this review, a ketogenic diet is discussed as a curative option. In support of a ketogenic diet, whereby carbohydrates are replaced for fats in daily meals, epileptic seizures can be terminated. In this review, we discuss possible metabolic sensors in synapses. These may include molecules that perceive changes in composition of extracellular space, for instance, ketone body and lactate receptors, or molecules reacting to changes in cytosol, for instance, KATP channels or AMP kinase. Inhibition of endocytosis is believed to be a universal synaptic mechanism of adaptation to metabolic changes.

Biogenetic and morphofunctional heterogeneity of mitochondria: the case of synaptic mitochondria.

The mitochondria of different cells are different in their morphological and biochemical properties. These organelles generate free radicals during activity, leading inevitably to mitochondrial DNA damage. It is not clear how this problem is addressed in long-lived cells, such as neurons. We propose the hypothesis that mitochondria within the same cell also differ in lifespan and ability to divide. According to our suggestion, cells have a pool of 'stem' mitochondria with low metabolic activity and a pool of 'differentiated' mitochondria with significantly shorter lifespans and high metabolic activity. We consider synaptic mitochondria as a possible example of 'differentiated' mitochondria. They are significantly smaller than mitochondria from the cell body, and they are different in key enzyme activity levels, proteome, and lipidome. Synaptic mitochondria are more sensitive to different damaging factors. It has been established that neurons have a sorting mechanism that sends mitochondria with high membrane potential to presynaptic endings. This review describes the properties of synaptic mitochondria and their role in the regulation of synaptic transmission.

ß-Hydroxybutyrate supports synaptic vesicle cycling but reduces endocytosis and exocytosis in rat brain synaptosomes.

The ketogenic diet is used as a prophylactic treatment for different types of brain diseases, such as epilepsy or Alzheimer's disease. In such a diet, carbohydrates are replaced by fats in everyday food, resulting in an elevation of blood-borne ketone bodies levels. Despite clinical applications of this treatment, the molecular mechanisms by which the ketogenic diet exerts its beneficial effects are still uncertain. In this study, we investigated the effect of replacing glucose by the ketone body ß-hydroxybutyrate as the main energy substrate on synaptic vesicle recycling in rat brain synaptosomes. First, we observed that exposing presynaptic terminals to nonglycolytic energy substrates instead of glucose did not alter the plasma membrane potential. Next, we found that synaptosomes were able to maintain the synaptic vesicle cycle monitored with the fluorescent dye acridine orange when glucose was replaced by ß-hydroxybutyrate. However, in presence of ß-hydroxybutyrate, synaptic vesicle recycling was modified with reduced endocytosis. Replacing glucose by pyruvate also led to a reduced endocytosis. Addition of ß-hydroxybutyrate to glucose-containing incubation medium was without effect. Reduced endocytosis in presence of ß-hydroxybutyrate as sole energy substrate was confirmed using the fluorescent dye FM2-10. Also we found that replacement of glucose by ketone bodies leads to inhibition of exocytosis, monitored by FM2-10. However this reduction was smaller than the effect on endocytosis under the same conditions. Using both acridine orange in synaptosomes and the genetically encoded sensor synaptopHluorin in cortical neurons, we observed that replacing glucose by ß-hydroxybutyrate did not modify the pH gradient of synaptic vesicles. In conclusion, the nonglycolytic energy substrates ß-hydroxybutyrate and pyruvate are able to support synaptic vesicle recycling. However, they both reduce endocytosis. Reduction of both endocytosis and exocytosis together with misbalance between endocytosis and exocytosis could be involved in the anticonvulsant activity of the ketogenic diet.

Influence of Glucose Deprivation on Membrane Potentials of Plasma Membranes, Mitochondria and Synaptic Vesicles in Rat Brain Synaptosomes.

Hypoglycemia can cause neuronal cell death similar to that of glutamate-induced cell death. In the present paper, we investigated the effect of glucose removal from incubation medium on changes of mitochondrial and plasma membrane potentials in rat brain synaptosomes using the fluorescent dyes DiSC3(5) and JC-1. We also monitored pH gradients in synaptic vesicles and their recycling by the fluorescent dye acridine orange. Glucose deprivation was found to cause an inhibition of K(+)-induced Ca(2+)-dependent exocytosis and a shift of mitochondrial and plasma membrane potentials to more positive values. The sensitivity of these parameters to the energy deficit caused by the removal of glucose showed the following order: mitochondrial membrane potential > plasma membrane potential > pH gradient in synaptic vesicles. The latter was almost unaffected by deprivation compared with the control. The pH-dependent dye acridine orange was used to investigate synaptic vesicle recycling. However, the compound's fluorescence was shown to be enhanced also by the mixture of mitochondrial toxins rotenone (10 µM) and oligomycin (5 µg/mL). This means that acridine orange can presumably be partially distributed in the intermembrane space of mitochondria. Glucose removal from the incubation medium resulted in a 3.7-fold raise of acridine orange response to rotenone + oligomycin suggesting a dramatic increase in the mitochondrial pH gradient. Our results suggest that the biophysical characteristics of neuronal presynaptic endings do not favor excessive non-controlled neurotransmitter release in case of hypoglycemia. The inhibition of exocytosis and the increase of the mitochondrial pH gradient, while preserving the vesicular pH gradient, are proposed as compensatory mechanisms.

Role of iron, zinc and reduced glutathione in oxidative stress induction by low pH in rat brain synaptosomes.

Brain ischemia leads to a decrease in pHo. We have shown previously in synaptosomes that the extracellular acidification induces depolarization of mitochondria followed by synthesis of superoxide anions and oxidative stress. Here, we investigated the effects of lowered pHo on oxidative stress and membrane potentials in synaptosomes treated by the iron chelator deferoxamine and zinc chelator TPEN. We demonstrated that chelating of metals has no impact on superoxide anion synthesis and intrasynaptosomal mitochondria depolarization. Meanwhile, deferoxamine was able to inhibit oxidative stress induced by low pHo and hydrogen peroxide application. Compared to deferoxamine, TPEN was less effective but it decreased the DCF fluorescence induced by pHo 6.0 which had no effects in other oxidative stress models. We found that the chelators were able to inhibit slightly plasma membrane depolarization. Synaptosomes preincubation at low pHo caused no effects on the reduced glutathione level. Depletion of glutathione by CDNB produced no additional increase in the DCF fluorescence induced by pHo 7.0. Our results suggest that free iron is crucial for the development of oxidative stress elicited by acidification in synaptosomes. Chelating of this metal seems to be a promising strategy for protecting the neuronal presynaptic terminals against oxidative stress developed at stroke.

Influence of intra- and extracellular acidification on free radical formation and mitochondria membrane potential in rat brain synaptosomes.

Brain ischemia is accompanied by lowering of intra- and extracellular pH. Stroke often leads to irreversible damage of synaptic transmission by unknown mechanism. We investigated an influence of lowering of pH(i) and pH(o) on free radical formation in synaptosomes. Three models of acidosis were used: (1) pH(o) 6.0 corresponding to pH(i) decrease down to 6.04; (2) pH(o) 7.0 corresponding to the lowering of pH(i) down to 6.92: (3) 1 mM amiloride corresponding to pH(i) decrease down to 6.65. We have shown that both types of extracellular acidification, but not intracellular acidification, increase 2',7'-dichlorodihydrofluorescein diacetate fluorescence that reflects free radical formation. These three treatments induce the rise of the dihydroethidium fluorescence that reports synthesis of superoxide anion. However, the impact of amiloride on superoxide anion synthesis was less than that induced by moderate extracellular acidification. Superoxide anion synthesis at pH(o) 7.0 was almost completely eliminated by mitochondrial uncoupler carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone. Furthermore, using fluorescent dyes JC-1 and rhodamine-123, we confirmed that pH(o) lowering, but not intracellular acidification, led to depolarization of intrasynaptosomal mitochondria. We have shown that pH(o) but not pH(i) lowering led to oxidative stress in neuronal presynaptic endings that might underlie the long-term irreversible changing in synaptic transmission.

Glutamate-induced free radical formation in rat brain synaptosomes is not dependent on intrasynaptosomal mitochondria membrane potential.

Glutamate induces reactive oxygen species formation (ROS) in neurons. Free radicals can potentially be synthesized by NADPH oxidase or mitochondria. The primary source of ROS origin has yet to be identified. In addition, pro-oxidant action of glutamate receptors on neuronal presynaptic terminals is still not characterized. We investigated the influence of glutamate and agonists of its ionotropic receptors on ROS formation detected by fluorescent dye DCFDA in rat brain synaptosomes. Glutamate in concentration 10 and 100µM led to an increase of probe fluorescence pointing to free radical accumulation. This effect was mimicked by 100µM of NMDA or 100µM of kainate. Glutamate-induced ROS formation was sensitive to NMDA inhibitors MK-801 (10µM), NO synthase (NOS) inhibitor l-NAME (100µM) and NADPH oxidase inhibitors DPI (30µM) and not affected by mitochondrial uncoupler CCCP (10µM) and mitochondrial toxins rotenone (10µM)+oligomycin (5µg/ml). We also showed that 100µM of glutamate leads to a decrease of intrasynaptosomal mitochondrial potential monitored by fluorescent dye Rhodamine-123. Hence, the depolarization of intrasynaptosomal mitochondria is not a primary cause of glutamate-induced ROS formation in neuronal presynaptic terminals. Activation of NMDA receptors might be responsible for a certain part of glutamate pro-oxidant action. Most likely, sources of glutamate-induced ROS formation in neuronal presynaptic terminals are NADPH oxidase and NOS activation.

Presynaptic glycine receptors influence plasma membrane potential and glutamate release.

Glycine is a classical inhibitory neurotransmitter however presynaptic glycine receptors have rather depolarizing action. Reasons for latter phenomenon are unknown. In the present paper we have investigated how glycine influences cytosolic chloride level monitored by fluorescent dye SPQ, membrane potential monitored by fluorescent dye DiSC3(5) and [(14)C]-glutamate release in synaptosomes. We estimated that cytosolic chloride concentration in synaptosomes was about 52 +/- 1 mM. Glycine (1 mM) induced chloride efflux and caused slow plasma membrane depolarization. Chloride efflux was almost completely blocked by 100 microM strychnine whilst glycine-induced depolarization was only partially. We also showed that 1 mM glycine induced [(14)C]-glutamate release via a strychnine-insensitive pathway. Hence we have concluded that glycine was able to induce two independent effects in synaptosomes: (1) Chloride efflux with following depolarization. This efflux was sensitive to strychnine and thereby is probably conducted through glycine-gated ion channels. (2) Glutamate release seems to be mediated by glycine transporters.

Are synapses targets of nanoparticles?

The last few years have been marked by real breakthroughs in the field of nanotechnology. Application of nanoparticles was proposed for diagnosis and treatment of different central nervous system diseases. Exposure to nanoparticles in vivo increases the risk of onset of neurodegenerative diseases and nanoparticles are apparently able to kill neurons in vitro. We suggested that presynaptic terminals of neurons are another target for nanoparticles, beyond the already established microglial cells. Ferritin was chosen as a prototypic nanoparticle model. We found that even a high concentration of ferritin, 800 microg/ml, was not able to induce spontaneous release of [(14)C]glutamate. In contrast, [(14)C]glutamate uptake was inhibited by ferritin in a dose-dependent fashion. As a next step, the influence of ferritin on the formation of reactive oxygen species was monitored using the fluorescent dye DCFH-DA (2',7'-dichlorofluorescein diacetate). It was shown that ferritin leads to a dose-dependent formation of free radicals. We found that the ferritin-mediated changes in glutamatergic neurotransmission at presynaptic endings can result in neuronal damage and finally neurodegeneration.

Ferritin, a protein containing iron nanoparticles, induces reactive oxygen species formation and inhibits glutamate uptake in rat brain synaptosomes.

Nanoparticles are currently used in medicine as agents for targeted drug delivery and imaging. However it has been demonstrated that nanoparticles induce neurodegeneration in vivo and kill neurons in vitro. The cellular and molecular bases of this phenomenon are still unclear. We have used the protein ferritin as a nanoparticle model. Ferritin contains iron particles (Fe(3+)) with size 7 nm and a protein shell. We investigated how ferritin influences uptake and release of [(14)C]glutamate and free radical formation as monitored by fluorescent dye DCFDA in rat brain synaptosomes. We found that even a high concentration of ferritin (800 microg/ml) did not induce spontaneous [(14)C]glutamate release. In contrast the same concentration of this protein inhibited [(14)C]glutamate uptake two fold. Furthermore ferritin induced intrasynaptosomal ROS (reactive oxygen species) formation in a dose-dependent manner. This process was insensitive to 30 microM DPI, an inhibitor of NADPH oxidase and to 10 microM CCCP, a mitochondrial uncoupler. These results indicate that iron-based nanoparticles can cause ROS and decreased glutamate uptake, potentially leading to neurodegeneration.

Influence of integrin-blocking peptide on gadolinium- and hypertonic shrinking-induced neurotransmitter release in rat brain synaptosomes.

Polyvalent cations and hypertonic shrinking of presynaptic endings lead to calcium-independent exocytosis in various synapses. In the present study we have investigated the contribution of integrins to this phenomenon. It was found that hypertonic shrinking, polyvalent cations ruthenium red and gadolinium results in dose-dependent calcium-independent neurotransmitter release in rat brain synaptosomes. The exocytotic mechanism of neurotransmitter release induced by 300 microM gadolinium was additionally verified by the fluorescent dye FM2-10. We found that 200 microM of RGDS peptide, an inhibitor of integrins, decreased polyvalent gadolinium-induced [3H]D: -aspartate release by 26%. This compound had no effect upon hypertonicity-induced release. The peptide RGES, a negative control for RGDS; genistein, an inhibitor of tyrosine kinases; and citrate, an inhibitor of lanthanides-induced aggregation were ineffective in both cases. Therefore, we have shown that integrins did not influence hypertonicity-evoked [3H]D: -aspartate release, but partially mediated that evoked by gadolinium ions.

Hypertonic shrinking but not hypotonic swelling increases sodium concentration in rat brain synaptosomes.

Neurotransmitter release is dependent on both calcium and sodium influx. Hypotonic swelling and hypertonic shrinking of neurons evokes calcium-independent exocytosis of neurotransmitters into the synaptic cleft. To date, there are not too much data available on relationship between extracellular osmolarity and sodium concentration in presynaptic endings. In the present study we investigated the effects of hypotonic swelling and hypertonic shrinking on sodium levels, as measured using fluorescent dyes SBFI-AM and Sodium Green in rat brain synaptosomes. Reduction of incubation medium osmolarity from 310 to 230 mOsm did not raise the intrasynaptosomal sodium concentration. An increase of osmolarity from 310 to 810 mOsm is accompanied by a dose-dependent elevation of sodium concentration from 8.1+/-0.5 to 46.5+/-2.8mM, respectively. This effect was insensitive to several channel inhibitors such as: tetrodotoxin, an inhibitor of voltage-gated sodium channels, bumetanide, an inhibitor of Na(+)/K(+)/2Cl(-) cotransport, gadolinium, an inhibitor of nonselective mechanosensitive channels, ruthenium red, an inhibitor of transient receptor potential channel and amiloride, an inhibitor of epithelial sodium channel/degenerin. Additionally, using the fluorescent dye BCECF-AM, we have shown that hypertonic shrinking caused a dose-dependent acidification of intrasynaptosomal cytosol, which suggests that the Na(+)/H(+) exchanger is not involved in the effect of increased osmolarity on cytosolic sodium levels. The increase in intrasynaptosomal sodium concentrations following increases in osmolarity is probably due to sodium influx through another sodium channels.

Influence of cholesterol depletion in plasma membrane of rat brain synaptosomes on calcium-dependent and calcium-independent exocytosis.

It is well established that calcium-dependent neurotransmitter release and exocytosis can be regulated by altering the cholesterol content of the plasma membrane. We have compared the influence of cholesterol depletion of synaptosomal plasma membrane by 15 mM methyl-beta-cyclodextrin (MCD) treatment on calcium-dependent release of D-[(3)H]aspartate induced by the calcium ionophore A23187 and on calcium-independent release induced by hypertonic shrinking or polyvalent cations. We found that decrease of cholesterol concentration by 9.3% inhibited calcium-dependent release of d-[(3)H]aspartate induced by calcium ionophore A23187 by four times while release induced by 300 microM Gd(3+), 150 mM and 500 mM sucrose remained unchanged. Further we have investigated the influence of MCD on exocytosis monitored by the fluorescent dye, acridine orange. Cholesterol depletion inhibited calcium-dependent exocytosis induced by calcium ionophore A23187 but had virtually no influence on calcium-independent exocytosis induced by hypertonic shrinking or Gd(3+). In summary, we found that the cholesterol content in synaptosomal plasma membrane is important for calcium-dependent exocytosis.

Role of calcium in exocytosis induced by hypotonic swelling.

We studied the influence of Ca(2+) on exocytosis induced by hypotonic shock using the fluorescent dyes acridine orange FM1-43 and FM2-10. It was shown using acridine orange that lowering osmolarity to 230 mOsm/L induces exocytosis both in calcium-containing and calcium-free media. By contrast, we were able to demonstrate calcium dependence of exocytosis using styryl dyes. Lowering osmolarity leads to an increase of neurotransmitter release in a calcium-independent manner. Thus, our data suggest that hypotonic swelling induces calcium-independent exocytosis. Calcium influx mediated by stretch channels is able to switch mode of exocytosis from "kiss and run" to full fusion.

Calcium regulates the mode of exocytosis induced by hypotonic shock in isolated neuronal presynaptic endings.

A decrease in the osmolarity of incubation medium is accompanied by calcium influx in neuronal presynaptic endings. We studied the influence of Ca2+ on exocytosis induced by hypotonic shock using the hydrophilic fluorescent dye acridine orange and the hydrophobic fluorescent dye FM2-10. It was shown using acridine orange that lowering of osmolarity to 230 mOsm/l induces exocytosis both in calcium-containing and calcium-free medium. By contrast, we were able to demonstrate calcium-dependence of exocytosis using styryl dye FM2-10. Lowering of osmolarity leads to increase of [3H]D-aspartate and [3H]GABA release in calcium-free medium. Addition of calcium inhibits hypotonic-induced neurotransmitter release. Decreasing of NaCl concentration to 92 mM in isotonic medium is able to induce d-aspartate and GABA release. Thus, our data suggest that hypotonic swelling induces calcium-independent exocytosis possibly by a "kiss and run" mechanism. Calcium influx mediated by stretch channels is able to provoke full fusion between plasma membrane and synaptic vesicles. [3H]D-aspartate and [3H]GABA released by hypotonic shock is determined by sodium lowering rather than by osmolarity decreasing itself.

Influence of hypotonic shock on glutamate and GABA uptake in rat brain synaptosomes.

Hyponatremia leads to hyperexcitability of neurons, seizures, and coma. It is well established that uptake of neurotransmitters is a sodium-dependent process. Therefore, we suggest that inhibition of neurotransmitter uptake can lead to the clinical manifestations of hyponatremia. Decreasing of sodium concentration down to 92 mM in incubation medium, which corresponds to lowering the osmolarity down to 230 mOsm/l, leads to a 45% decrease in glutamate uptake and a 46% decrease in gamma-aminobutyric acid (GABA) uptake. However, this effect was mediated by the nonspecific lowering of osmolarity rather than by decreasing sodium concentration. Hypotonic shock was able to reduce glutamate uptake in the presence of protein kinase inhibitors staurosporine and genistein, the phosphatase inhibitor okadaic acid, the phosphatidylinositol 3-kinase inhibitor wortmannin, and cytoskeleton modulators colchicine and cytochalasin B. Therefore, we suggest that intracellular signaling is not mediating the effect of osmolarity reduction on neurotransmitter uptake.