Augmentation of Normal and Glutamate-Impaired Neuronal Respiratory Capacity by Exogenous Alternative Biofuels

Abstract

Mitochondrial respiratory capacity is critical for responding to changes in neuronal energy demand. One approach toward neuroprotection is administration of alternative energy substrates (“biofuels”) to overcome brain injury-induced inhibition of glucose-based aerobic energy metabolism. This study tested the hypothesis that exogenous pyruvate, lactate, β-hydroxybutyrate, and acetyl-L-carnitine each increase neuronal respiratory capacity in vitro either in the absence of, or following transient excitotoxic glutamate receptor stimulation. Compared to the presence of 5 mM glucose alone, the addition of pyruvate, lactate, or β-hydroxybutyrate (1.0 – 10.0 mM) to either day in vitro (DIV) 14 or 7 rat cortical neurons resulted in significant, dose-dependent stimulation of respiratory capacity, measured by cell respirometry as the maximal O₂ consumption rate in the presence of the respiratory uncoupler FCCP. A thirty minute exposure to 100 μM glutamate impaired respiratory capacity for DIV 14 but not DIV 7 neurons. Glutamate reduced the respiratory capacity for DIV 14 neurons with glucose alone by 25% and also reduced respiratory capacity with glucose plus pyruvate, lactate or β-hydroxybutyrate. However, respiratory capacity in glutamate-exposed neurons following pyruvate or β-hydroxybutyrate addition was still at least as high as that obtained with glucose alone in the absence of glutamate exposure. These results support the interpretation that previously observed neuroprotection by exogenous pyruvate, lactate, or β-hydroxybutyrate is at least partially mediated by their preservation of neuronal respiratory capacity.

Introduction

Mitochondrial bioenergetic dysfunction is a primary cause of impaired cerebral energy metabolism which precedes and contributes to neuronal death after brain ischemia, trauma, and excitotoxicity [1]. One approach to neuroprotection for brain injury is the administration of exogenous alternative potential energy substrates (“biofuels”) for aerobic energy metabolism that may overcome inhibition of glucose-based energy production [1]. These alternative fuels include pyruvate, lactate, acetyl-L-carnitine, and β-hydroxybutyrate.

Extracellular pyruvate crosses the plasma membrane and mitochondrial inner membrane by respective monocarboxylate transporters. Once inside the mitochondria, pyruvate is oxidized to form acetyl-CoA, which then enters the tricarboxylic acid cycle, driving aerobic energy metabolism. Exogenous pyruvate may be neuroprotective by compensating for inhibition of glycolysis and it significantly stimulates uncoupled respiration by normal, organotypic hippocampal slices [2], isolated synaptic nerve endings (synaptosomes) [3], and cerebellar granule neurons exposed to excitotoxic glutamate [4].

Both exogenous and endogenous lactate can also be neuroprotective [5]. Extracellular lactate also enters neurons via the plasmalemmal monocarboxylic acid transporter. Endogenous lactate is produced primarily by astrocytes, and is transported into neurons via this pathway [6]. Lactate is reduced in the cytoplasm by lactate dehydrogenase to form pyruvate. Pyruvate then enters mitochondria and is used for aerobic energy metabolism.

Acetyl-L-carnitine (ALCAR) enters neurons through at least two organic cation transporters [7]. ALCAR then enters the mitochondrial matrix in exchange for free carnitine via the acyl-carnitine/carnitine transporter. Once inside, the carnitine acyltransferase catalyzes the reaction between ALCAR and coenzyme A to produce free carnitine and acetyl-coenzyme A, the latter of which is also the product of the pyruvate dehydrogenase reaction that enters the tricarboxylic acid cycle. ALCAR is metabolized by the brain for both energy and neurotransmitter metabolism [8]. and may exert neuroprotection by compensating for inhibition of glycolysis or the pyruvate dehydrogenase complex [9][10]. β-hydroxybutyrate (BHB) is a ketone body used by the brain as an energy substrate during early development and under conditions of reduced glucose availability [11][12]. Like ALCAR, BHB carbon atoms enter the TCA cycle in the form of acetyl-CoA. Thus, BHB may also exert neuroprotection by compensating for inhibition of glycolysis or pyruvate dehydrogenation [13][14].

The mechanism by which each of these biofuels provides neuroprotection may be due to their direct metabolism by bioenergetically impaired neurons; however, evidence also exists for alternative mechanisms of action. The relative contributions of different mechanisms may also depend on neurodevelopmental age, since the activities of enzymes and transporters that use these metabolites as substrates vary considerably from birth to maturity [15]. No direct comparisons of the degree to which these alternative biofuels contribute to neuronal aerobic energy metabolism have been reported, either for normal, healthy neurons or for injured neurons exhibiting impaired glucose-based respiration. This study tested the hypothesis that the respiratory capacity of primary cultures of rat cortical neurons is dose-dependently increased by pyruvate, lactate, ALCAR, or BHB. The additional objective of this study was to determine which alternative biofuel is most effective at preserving respiratory capacity following exposure of neurons to glutamate excitotoxicity. Finally, we also determined if the in vitro age of cortical neurons affects the degree to which respiration is influenced by these oxidizable substrates.

Materials

Neurobasal media, B27 supplement, Penicillin-Streptomycin, GlutaMAX and fetal bovine serum were purchased from Invitrogen (Carlsbad, CA). V7 XF24 cell culture microplates were obtained from Seahorse Bioscience (North Billerica, MA). All other reagents required for either XF24 microplate-based respirometry or cell culture were purchased from Sigma-Aldrich (St. Louis, MO) including the following alternative fuels: pyruvate (P2256), β-hydroxybutyrate (54965), acetyl-L-carnitine (A6706) and lactic acid (L1750). Alternative biofuels were prepared in aCSF on the day of the experiment (pyruvate) or diluted from concentrated stocks which were stored at −20°C.

Methods

Cell Culture

Rat cortical neurons were cultured from cerebral cortices of Sprague Dawley embryos at embryonic day 18 by aseptic dissection and trypsin dissociation as previously described [16][17]. Cells were plated at 8 × 10⁴ cells/well (0.32 cm²) on poly-D-lysine-coated 24-well V7 plates (Seahorse Bioscience). Neurons were initially plated in Neurobasal medium supplemented with 2% B27, 1% Glutamax, 10% fetal bovine serum and antibiotics and then, after 2 h, changed to and maintained in the same medium lacking fetal bovine serum. Cytosine β-D-arabinofuranoside (5 μM) was added on day in vitro (DIV) 4 to minimize glial contamination and media was partially replenished on DIV6. Neurons were maintained in vitro at 37°C in a 95% air / 5% CO₂ humidified incubator until they were used for experiments on either DIV 14 or DIV 7. Astrocytic contamination of DIV 14 and DIV 7 neuronal cultures was consistently less than 5% as determined by GFAP immunocytochemistry to identify astrocytes and NeuN counterstaining to identify neurons.

XF24 Microplate-Based Respirometry

Neuronal O₂ consumption rate (OCR) was measured using the Seahorse XF24 Extracellular Flux Analyzer (Seahorse Bioscience) in an artificial cerebrospinal fluid (aCSF) medium consisting of 120 mM NaCl, 3.5 mM KCl, 1.3 mM CaCl₂, 0.4 mM KH₂PO₄, 1 mM MgCl₂, 5 mM HEPES, and 5.0 mM glucose, pH 7.4 at 37°C, supplemented with 0.4% defatted bovine serum albumin. This form of cell respirometry utilizes a 24 well plate format and quantifies the OCR at different times and following additions of metabolites, drugs, or their vehicles. Prior to each experiment, neurons in each plate well were washed twice with 500 μl of aCSF. Three wells per plate did not contain neurons, thus serving as “blank” wells, to control for temperature-sensitive fluctuations in O₂ fluorophore emission. Following washing, each well was filled with 675 μl aCSF and plates were placed in a CO₂-free incubator (37°C) for ~45 minutes before each set of measurements to further purge CO₂ and to allow temperature and pH equilibration. Plates were then loaded into the XF24 respirometer and further equilibrated for 15 min by three 3-min mix, 2-min wait cycles prior to the first measurement. XF24 assays consisted of 3-min mix, 3-min wait, and 2-min measurement cycles and were performed at 37°C. Drugs of interest prepared in aCSF assay medium (75 μl) were preloaded into reagent delivery chambers (A–D) at 10X, 11X, 12X, and 13X the final working concentrations, respectively, to account for changes in total fluid volumes, and injected sequentially at pre-designated intervals. HEPES (5 mM) was included in aCSF to ensure pH stability over the two hour time course of measurements.

Statistical Analysis

Comparison between all maximal OCR values obtained for DIV 14 and 7 neurons in the presence of glucose without added fuels used a student’s t test where n = 18 – 19 per group. The effects of exposure to alternative biofuels and to glutamate were analyzed using a Kruskal-Wallace one-way analysis of variance followed by Student Newman Keul’s post-hoc test. The values used for each OCR for each independent experiment were the means obtained from n = 3 respirometer plate wells. Statistical analyses were performed on data obtained from 3 – 6 independent experiments using different primary neuronal cultures. Results are expressed as means ± standard error. A p-value < 0.05 was considered significant.

Results

Oxygen consumption by primary cultures of rat cortical neurons was measured using the Seahorse XF24 Extracellular Flux Analyzer. A representative example of O₂ consumption rate (OCR) data obtained from one preparation of DIV 14 neurons is shown in Figure 1. Each value represents the mean ± standard error for n = 3 wells. The baseline OCR was stable during the first 20 min (Fig. 1A). Other experiments where no additions were made indicate that the initial OCR is unperturbed for at least 100 min (not shown). Basal cellular respiration is generally limited by the rate at which ATP is hydrolyzed (energy demand), which in turn determines the rate of trans-mitochondrial membrane proton cycling into the matrix through the ATP synthetase and out of the matrix through components of the electron transport chain. The maximal rate at which O₂ is consumed can be elicited by the addition of a proton ionophore, e.g., FCCP, which collapses proton electrochemical gradients and therefore allows the respiratory chain to operate maximally without limitation by a positive outside membrane potential. Thus, FCCP uncouples respiration from ATP synthesis and allows determination of the respiratory capacity of the electron transport chain, which, in the presence of FCCP, is only limited by substrate supply. As seen in Fig. 1A, the addition of FCCP resulted in a substantial, approximately 100% increase in OCR. The concentration of FCCP was titrated for the different fuels and for DIV 14 and 7 neurons to obtain maximal OCR. Based on these titrations, the optimal FCCP concentration of 3 μM was used in all experiments. Subsequent addition of the respiratory Complex III inhibitor antimycin A resulted in a >90% inhibition of O₂ consumption, confirming that the OCRs are due almost exclusively to mitochondrial respiration.

Figure 1.

Representative oxygen consumption rate (OCR) measurements obtained with primary cultures of DIV 14 rat cortical neurons. At 45 min prior to OCR measurements, culture medium was replaced with aCSF containing 5.0 mM glucose as the only exogenous metabolic fuel. Maximal OCR was obtained in the presence of 3.0 μM FCCP and minimum OCR was obtained in the presence of 1.0 μM antimycin A. Glutamate, when added, was at a final concentration of 100 μM (along with 10 μM glycine) and was followed 30 min later by the addition of MK801 (10 μM) plus CNQX (10 μM). (A). OCR values obtained in the absence of exposure to glutamate. The first and second additions of FCCP were 1 μM and 2 μM, respectively. (B). OCR values obtained before and after 30 min excitotoxic glutamate receptor stimulation. The addition of FCCP was 3 μM. Each OCR value, expressed as pmoles O₂ per min, represents the mean ± standard error of measurements obtained from three identically treated wells.

Supraphysiologic concentrations of pyruvate increase uncoupled O₂ consumption in both primary cultures of neurons and organotypic brain slice cultures [2][4], indicating that substrate supply is rate limiting for maximum respiration in these cells. Figure 1A shows that when pyruvate was added to DIV14 neurons in the absence of FCCP, the baseline OCR was unaffected, as expected based on respiration being limited by metabolic demand. However, subsequent addition of FCCP resulted in a respiratory capacity that was approximately twice that of the respiratory capacity measured in the absence of pyruvate.

Figure 1B provides an example of a similar test of pyruvate on respiratory capacity but following transient exposure to an excitotoxic concentration of glutamate. Addition of 100 μM glutamate (along with the N-methyl D-aspartate (NMDA) receptor co-agonist glycine, 10 μM) resulted in an immediate, approximately 100% stimulation of OCR, as expected, due to the increased ATP turnover associated with glutamate-stimulated ion fluxes. Thirty minutes after glutamate addition, the glutamate receptor antagonists MK801 (an NMDA receptor inhibitor) and CNQX (an AMPA/kainite receptor inhibitor) were added. Subsequent addition of the uncoupler FCCP to the neurons metabolizing glucose alone resulted in respiratory stimulation but to an OCR level that was approximately 60% of that obtained in the absence of glutamate (compare Figure 1B, open squares to Figure 1A, open squares). Pyruvate (10.0 mM), when added together with FCCP, increased O₂ consumption to an OCR level that was similar to that obtained in control, glucose metabolizing neurons in the absence of excitotoxic glutamate exposure (compare Figure 1B, filled squares to Figure 1A, open squares). However, this respiratory capacity measured in the presence of FCCP plus pyruvate after excitotoxic glutamate exposure was still reduced compared to the respiratory capacity measured in the presence of pyruvate without prior glutamate treatment (compare Figure 1B, filled squares, to Figure 1A, filled squares).

Figure 2 provides a summary of OCR data obtained from n = 3 – 6 independent experiments such as that described in Figure 1. These experiments were performed independently with DIV 14 and 7 cortical neurons and used pyruvate concentrations ranging from 1.0 to 10.0 mM. Figure 2A describes the relationship between the dose of pyruvate and the respiratory capacity for DIV 14 neurons in the presence of 5.0 mM glucose and in the absence of exposure to glutamate. Values are expressed as a percentage of the OCR obtained with FCCP in the absence of pyruvate. Significant stimulation (20%) of respiratory capacity was observed at 1.0 mM pyruvate. Greater stimulation was observed at 2.5 to 10.0 mM pyruvate and a maximum stimulation of approximately 60% was evident at 10.0 mM pyruvate. Addition of 10.0 mM sodium chloride to control for the effect of 10.0 sodium pyruvate on osmolarity failed to stimulate OCR (not shown).

Figure 2.

Concentration-dependent effects of pyruvate on the respiratory capacity of DIV 14 and 7 rat cortical neurons in the presence of glucose. Experiments were performed as described in Figure 1. (A) Effects of pyruvate on maximal OCR obtained with DIV 14 neurons. (B) Effects of pyruvate on maximal OCR obtained with DIV 7 neurons. (C) Effects of transient excitotoxic glutamate receptor stimulation and pyruvate on uncoupled OCR obtained with DIV 14 neurons. (D) Effects of transient excitotoxic glutamate receptor stimulation and pyruvate on uncoupled OCR obtained with DIV 7 neurons. Values for OCR obtained after FCCP addition are expressed as a percentage of the raw rates of oxygen consumption obtained just prior to the addition of glutamate. Values are expressed as means ± s.e. for n = 3–6 different neuronal preparations. ^* p < 0.05 compared to the absence of pyruvate (A and B). ^* p < 0.05 compared to the absence of glutamate (C and D). ^# p < 0.05 compared to the absence of pyruvate following exposure to glutamate (C and D).

The same experiments described in Figures 2A and C were performed with cortical neurons at DIV 7 (Figures 2B and D). Neurons at this stage of in vitro development are often considered to be comparable in some phenotypic characteristics to neurons present within the brains of immature rats. For instance, the expression of several NMDA receptor subunits in DIV 7 cortical neurons is substantially lower than those expressed in DIV 12 neurons [18]. The OCR obtained with FCCP in the presence of glucose alone was significantly lower for DIV7 than for DIV14 neurons (548 ± 73 vs. 814 ± 59 pmoles O₂/min; n = 18–19 per age group; p < 0.05). Figure 2B describes the effects of different pyruvate concentrations on respiratory capacity measured following FCCP addition using DIV 7 neurons metabolizing 5.0 mM glucose. The degree to which pyruvate increased respiratory capacity was similar for DIV 7 and 14 neurons. Significant (40%) stimulation was observed at 1.0 mM pyruvate and 60% stimulation was observed at 10.0 mM pyruvate.

The effects of pyruvate supplementation on the uncoupled rate of O₂ consumption by DIV 14 and 7 neurons following exposure to an excitotoxic concentration of glutamate are summarized in Figures 2C and D. A 30 min exposure to 100 μM glutamate, followed by the addition of the glutamate receptor antagonists MK801 and CNQX, resulted in an average 25% reduction in maximal OCR by DIV 14 neurons. Pyruvate at all concentrations elevated the respiratory capacity measured after excitotoxic glutamate receptor stimulation, with a maximum stimulation of respiratory capacity of 60% at 10.0 mM pyruvate. Despite this stimulation, the neuronal respiratory capacity measured after excitotoxic glutamate in the presence of pyruvate (Figure 2C, 6^th bar) was approximately 35% lower than that obtained in the absence of excitotoxic glutamate exposure (Figure 2C, 3^rd bar). Nevertheless, the respiratory capacity was still at least as great as that measured in the absence of pyruvate or glutamate treatment (Figure 2C, first bar). In contrast to the glutamate-induced inhibition of respiratory capacity observed in DIV 14 neurons, no inhibition by glutamate occurred in DIV 7 neurons. Therefore, the degree to which 10.0 mM pyruvate stimulated respiratory capacity for DIV7 neurons was the same in the absence and presence of glutamate.

Lactate is another energy metabolite that is purported to be neuroprotective through its ability to sustain neuronal energy metabolism [5]. The effects of exogenous lactate on neuronal respiratory capacity are described in Figure 3. For DIV14 neurons in the absence of exposure to glutamate (Figure 3A), significant, 30% stimulation of respiratory capacity was achieved at 2.5 mM lactate and a maximum stimulation of 45% at 10.0 mM lactate. As for Figure 2, stimulation of respiratory capacity by exogenous biofuel was expressed as a percentage of the uncoupled OCR of neurons metabolizing 5.0 mM glucose in the absence of additional substrate. Lactate stimulation of respiratory capacity was less pronounced in DIV 7 neurons, with a significant, 30% stimulation observed only at 10.0 mM lactate. Figures 3C and D provide results from the measurements obtained after exposure to excitotoxic glutamate receptor stimulation. In these experiments using DIV 14 neurons, glutamate exposure decreased respiratory capacity by an average of 30% (Figure 3C). Lactate at 1.0 – 10.0 mM did not stimulate the respiratory capacity measured in the presence or absence of prior glutamate exposure (Figure 3C), indicating a variable response to lactate among different sets of experiments. As seen with the experiments described in Figure 2D, glutamate did not inhibit respiration by DIV 7 neurons (Figure 3D). Lactate did not significantly stimulate the respiratory capacity of DIV 7 neurons measured in the absence of glutamate exposure in this set of experiments (10.0 mM, white bar), but did significantly stimulate uncoupled respiration after exposure to glutamate (10.0 mM, black bar).

Figure 3.

Concentration-dependent effects of lactate on the respiratory capacity of DIV 14 and 7 rat cortical neurons in the presence of glucose. Experiments were performed as described in Figure 1. (A) Effects of lactate on uncoupled OCR obtained with DIV 14 neurons. (B) Effects of lactate on uncoupled OCR obtained with DIV 7 neurons. (C) Effects of transient excitotoxic glutamate receptor stimulation and lactate on uncoupled OCR obtained with DIV 14 neurons. (D) Effects of transient excitotoxic glutamate receptor stimulation and lactate on uncoupled OCR obtained with DIV 7 neurons. Values for OCR obtained after FCCP addition are normalized as in Figure 2. Values are expressed as means ± s.e. for n = 3–6 different neuronal preparations. ^* p < 0.05 compared to the absence of lactate (A and B). ^* p < 0.05 compared to the absence of glutamate (C and D). ^# p < 0.05 compared to the absence of lactate following exposure to glutamate (D).

Acetyl-L-carnitine (ALCAR) is another natural endogenous metabolite that when present at pharmacologic concentrations exhibits neuroprotective properties, often ascribed to its oxidative metabolism [19][20]. When added to DIV 14 or 7 cortical neuronal cultures, ALCAR at 1.0 – 10.0 mM exhibited no significant stimulation of respiratory capacity either in the absence or presence of exposure to glutamate (Figures 4A–D). Surprisingly, 10.0 mM ALCAR instead significantly reduced the respiratory capacity measured in the experiments described in Figure 4D.

Figure 4.

Concentration-dependent effects of acetyl-L-carnitine (ALCAR) on the respiratory capacity of DIV 14 and 7 rat cortical neurons in the presence of glucose. Experiments were performed as described in Figure 1. (A) Effects of ALCAR on uncoupled OCR obtained with DIV 14 neurons. (B) Effects of ALCAR on uncoupled OCR obtained with DIV 7 neurons. (C) Effects of transient excitotoxic glutamate receptor stimulation and ALCAR on uncoupled OCR obtained with DIV 14 neurons. (D) Effects of transient excitotoxic glutamate receptor stimulation and ALCAR on uncoupled OCR obtained with DIV 7 neurons. Values for OCR obtained after FCCP addition are normalized as in Figure 2. Values are expressed as means ± s.e. for n = 3–6 different neuronal preparations. ^* p < 0.05 compared to the absence of glutamate (C and D). ^# p < 0.05 compared to the absence of ALCAR following exposure to glutamate (D).

β-hydroxybutyrate (BHB) is an endogenously produced ketone body that also exhibits neuroprotection when administered at pharmacologic doses. Significant stimulation of DIV 14 neuronal respiratory capacity was observed at all BHB concentrations, with 60% stimulation at 10.0 mM (Figure 5A). Very similar results were obtained with DIV 7 neurons (Figure 5B). Following exposure of DIV 14 neurons to glutamate, the respiratory capacity with glucose alone was significantly inhibited (30%) (Figure 5C). The respiratory capacity was also inhibited by glutamate in the presence of 10.0 mM BHB (25%). The maximal OCRs obtained after glutamate exposure with BHB present at 1.0 – 10.0 mM were not significantly different from the respiratory capacity seen with glucose alone and no glutamate exposure. BHB (10.0 mM) increased neuronal respiratory capacity after excitotoxic glutamate exposure compared to that observed with glutamate and glucose alone. As with the other experiments, exposure of DIV 7 neurons to glutamate did not reduce respiratory capacity with or without BHB.

Figure 5.

Concentration-dependent effects of β-hydroxybutyrate (BHB) on the respiratory capacity of DIV 14 and 7 rat cortical neurons in the presence of glucose. Experiments were performed as described in Figure 1. (A) Effects of BHB on maximal OCR obtained with DIV 14 neurons. (B) Effects of BHB on uncoupled OCR obtained with DIV 7 neurons. (C) Effects of transient excitotoxic glutamate receptor stimulation and BHB on uncoupled OCR obtained with DIV 14 neurons. (D) Effects of transient excitotoxic glutamate receptor stimulation and BHB on uncoupled OCR obtained with DIV 7 neurons. Values for OCR obtained after FCCP addition are normalized as in Figure 2. Values are expressed as means ± s.e. for n = 3–6 different neuronal preparations. ^* p < 0.05 compared to the absence of BHB (A and B). ^* p < 0.05 compared to the absence of glutamate (C and D). ^# p < 0.05 compared to the absence of BHB following exposure to glutamate (C and D).

Discussion

This study represents the first direct comparison of the extent to which different alternative biofuels elevate the respiratory capacity of primary cultures of neurons, measured in the presence of the uncoupler FCCP. Administration of each of these potential fuels to cultured neurons or to animals provides neuroprotection in acute CNS injury models, including glutamate excitotoxicity [21][19][22][23], cerebral ischemia/reperfusion [24][25][26][27], and traumatic brain and spinal cord injury [28][29][30]. The typical explanation for neuroprotection by these metabolic intermediates is that they are actively metabolized to support aerobic energy metabolism and therefore to also limit tissue acidosis caused by anaerobic lactate production. This mechanism of action is based primarily on indirect measures of their metabolism, e.g. the effects of these agents on tissue or cellular lactate and ATP levels. Some studies have directly measured the incorporation of labeled carbons in these compounds to downstream metabolites but have not actually quantified the extent to which their utilization actually supports cerebral or neuronal energy metabolism [8]. In contrast, our study directly measured the extent to which neuronal respiratory capacity is stimulated by different alternative biofuels in the presence of a physiologically relevant glucose concentration. Although ADP-stimulated respiration cannot be directly measured in intact cells, the relative extent to which the uncoupled respiration is stimulated by these fuels likely approximates the relative extent to which they can support aerobic ATP formation.

In our experiments, both pyruvate and BHB stimulated uncoupled respiration by up to 60%. Lactate stimulation of respiratory capacity was approximately 45% and ALCAR had no effect on neuronal respiration under control conditions. We also assessed the ability of these biofuels to support neuronal respiration after transient excitotoxic glutamate receptor activation to model their effects on metabolically injured neurons. Depending on the molecular site(s) at which aerobic glucose metabolism is inhibited, different alternative fuels may compensate for impaired glucose utilization and therefore maintain ATP production at a level sufficient for normal neuronal electrical activity and cell viability. When DIV 14 cortical neurons were exposed to 100 μM glutamate for 30 min, maximal respiration was inhibited by approximately 25%. In the presence of 10.0 mM pyruvate, glutamate decreased respiratory capacity by approximately 30%. Nevertheless, the maximal OCR observed after glutamate exposure in the presence of both glucose plus 1.0 – 10.0 mM pyruvate met or exceeded that observed with glucose alone in the absence of pyruvate, thus establishing the ability of pyruvate to compensate for glutamate-impaired aerobic glucose metabolism. Similar results were obtained with BHB. Neither lactate nor ALCAR significantly preserved respiratory capacity after exposure of DIV 14 neurons to excitotoxic glutamate. Therefore under the conditions used in these experiments, pyruvate and BHB are the most effective at stimulating neuronal aerobic energy metabolism, both in the absence and presence of excitotoxicity-impaired respiration.

In addition to testing the effects of alternative biofuels on O₂ consumption by mature, DIV 14 cortical neurons, we performed identical tests with immature DIV 7 neurons using aCSF containing 5.0 mM glucose. Measurements of the maximal OCR obtained after addition of the uncoupler FCCP in the presence of glucose alone revealed a 47% greater maximal rate with DIV 14 compared to DIV 7 neurons. This new observation is consistent with increases in mitochondrial mass and mitochondrial protein levels during neuronal maturation both in vitro and in vivo [31]. In the absence of glutamate exposure, the presence of pyruvate, lactate, and BHB each stimulated maximal respiration by 60%, 30%, and 50%, respectively. No significant stimulation was observed with ALCAR.

Exposure of DIV 7 neurons to 100 μM glutamate for 30 min had no effect on respiratory capacity. Hippocampal neurons at DIV 5–7 are resistant to NMDA excitotoxicity and express 15% of the NMDA receptor binding sites present at DIV 21 [32]. Moreover, the level of post-synaptic density 95 protein (PSD95), which is necessary for mature NMDA receptor activity, is 70% lower in DIV 7 compared to DIV 14 cortical neurons [33]. Taken together, these observations suggest that the lack of respiratory inhibition by glutamate in DIV 7 neurons is due to the relatively low influx of Ca²⁺ and Na⁺ through ionotropic glutamate receptors. In light of the fact that respiration by DIV 7 neurons was unaffected by glutamate, it is not surprising that the extent to which pyruvate, lactate and BHB stimulated maximal OCR after exposure to glutamate was comparable to the degree of stimulation observed in the absence of glutamate. As with DIV 14 neurons and DIV 7 neurons not exposed to glutamate, ALCAR did not stimulate maximal respiration. In fact, in DIV 7 neurons 10.0 mM ALCAR actually significantly inhibited uncoupled O₂ consumption by a mechanism that remains to be defined.

There are several aspects of these in vitro O₂ consumption measurements that limit their extrapolation to potential metabolic effects of alternative biofuels on the normal or injured brain. One major limitation is that pure cultures of cortical neurons were used, rather than mixed cultures or pure astrocyte cultures. It is distinctly possible that one or more of these alternative energy sources would be utilized to a greater or lesser extent by astrocytes compared to neurons. For instance, measurements of incorporation of ¹³C from ¹³C-labeled acetyl-L-carnitine in the immature brain (21 day old) suggest that ALCAR is utilized for energy metabolism and that a substantial metabolic flux of carbon from the acetyl moiety occurs in astrocytes [8]. Based on evidence that a substantial fraction of glucose is metabolized to lactate in astrocytes and that this lactate is shuttled to neurons where it is metabolized aerobically, any increase in aerobic astrocyte metabolism by alternative biofuels could further promote anaerobic glucose metabolism, thereby providing additional lactate to neurons for their aerobic metabolism. Experiments are in progress to determine the relative influence of the four alternative biofuels on respiration by primary cultures of cortical astrocytes. In addition to the potential difference in fuel consumption by different brain cells, the difference in ambient conditions present in cell culture compared to those in the intact brain are likely to influence cellular energy metabolism. For example, cells are normally cultured under 95% air (20% O₂) and 5% CO₂ whereas cells in the normal brain are exposed to much lower levels of O₂ closer to around 2–5% [34]. These relatively low levels of physiological O₂ are accompanied by expression of hypoxia inducible factor 1 (HIF1), which transcriptionally activates the expression of many genes that includes several coding for glycolytic enzymes and glucose transporters [35];[36];[35];[36];[37]. The expression of different isoforms of monocarboxylic acid transporters (MCT), which transport pyruvate, lactate, and ketone bodies, can also be influenced by HIF1 [38] and by ambient levels of different fuels, including ketone bodies [39]. In contrast to pyruvate, lactate, and ketones, ALCAR is transported into cells by members of the organic cation transporter novel family (OCTN2 and 3). These transporters are abundant in astrocytes but are also present in neurons in vivo [40]; however, little is known regarding the regulation of their expression in cultured neurons. The fact that fasting increases the expression of these transporters suggests that their expression could be suppressed by the high glucose and very low fatty acid and ketone body concentrations present in culture media [41]. Considering these and many other factors that could influence the utilization of potential fuels by different brain cells, future studies should be conducted under a variety of cell culture conditions to characterize the spectrum of metabolic responses to both physiological and pathological variables.

ABBREVIATIONS

aCSF

artificial cerebrospinal fluid

AMPA

2-amino-3-(3-hydroxy-5-methyl-isoxazol-4-yl)propanoic acid

ALCAR

acetyl-L-carnitine

ATP

adenosine triphosphate

BHB

β-hydroxybutyrate

CNQX

6-cyano-7-nitroquinoxaline-2,3-dione

DIV

days in vitro

FCCP

carbonyl cyanide-p-trifluoromethoxyphenylhydrazone

MK801

(+)-5-methyl-10,11- dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine maleate

NMDA

N-methyl D-aspartate

OCR

oxygen consumption rate