Brain Poster Session: Cerebral Metabolic Regulation

10. Inter-individual variations of CBF and oxygen consumption in relation to hemoglobin concentration: a PET study

M. Ibaraki, Y. Shinohara, K. Nakamura, S. Miura and T. Kinoshita

Akita Research institute of Brain and Blood Vessels, Akita, Japan

Introduction: Control mechanism of CBF and its relation to oxygen metabolism are not fully understood. Hemoglobin (Hb) concentration is one of important factors that directly affect oxygen delivery to brain tissue. Variations of Hb concentration, even in normal range, may induce biological response in cerebral circulation to keep oxygen consumption constant. The aim of the study was to investigate the relationship between Hb and CBF and oxygen metabolism measured by ¹⁵O-PET in humans.

Methods: Healthy volunteers (n = 17) and patients with unilateral major arterial steno-occlusive disease who had no obvious infarct lesion in T2-weighted MR image (n = 44) were studied retrospectively. CBF, CBV, CMRO₂, and OEF were measured by the sequential administration of H₂¹⁵O, ¹⁵O₂, and C¹⁵O (autoradiographic 3-step method) with 3D-dedicated PET scanner, SET-3000GCT/M. ¹ Region-of-interests with elliptical shape (16-mm × 48-mm) were defined on cerebral cortical region on a slice at the centrum semiovale level. Values of CBF, CBV, CMRO₂, and OEF obtained from both hemispheres of the volunteers and non-affected hemisphere of the patients were pooled and examined by multiple linear regression analysis. Independent variables included Hb concentration (g/dL), age (yr), sex (male or female), and subject type (healthy or stroke). The optimal models were determined by backward selection method based on Akaike information criterion.

Results: Average values (mean±standard deviation) for Hb concentration, CBF, CBV, CMRO₂, and OEF were 12.9±1.6 (g/dL), 40±6 (mL/100 mL per min), 3.2±0.5 (mL/100 mL), 3.1±0.5 (mL/100 mL per min), and 45±6 (%), respectively. CBF and OEF were inversely related to Hb concentration (Figure 1). In contrast, the analyses for CBV and CMRO₂ showed no statistically-significant relation to Hb. The optimal model for CBF was described as CBF = 73.3−(0.191 × Age)−(1.65 × Hb). In the OEF analysis, adding CBF term as an independent variable improved goodness of fit described as OEF (%) = 80.5−(1.75 × Hb)−(0.317 × CBF). CMRO₂ calculated from the optimal models for CBF and OEF (= arterial oxygen content × CBF × OEF) were relatively insensitive to changes in Hb concentration (Figure 1).

Figure 1

Relation between Hb and CBF, OEF, and CMRO2. Solid lines represent predictions by the optimal models for Age = 64 yr (group average).

Conclusion: In contrast to the inverse relations of CBF and OEF to Hb concentration, CMRO₂ tends to keep constant value over the normal range in Hb concentration. The present result may support an existence of CBF control mechanism that keeps oxygen consumption constant against the variation of Hb concentration.

265. Monitoring oxygen partial pressure in cerebral microvasculature with high spatial resolution using phosphorescence lifetime imaging

M. Yaseen¹, V. Srinivasan¹, S. Sakadžiæ¹, S. Vinogradov², C. Ayata^3,4 and D. Boas¹

¹Photon Migration Imaging Laboratory, MGH/MIT/HMS Athinuola A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital/Harvard Medical School, Charlestown, Massachusetts; ²Department of Biochemistry and Biophysics, University of Pennsylvania, Philadelphia, Pennsylvania; ³Stroke and Neurovascular Regulation Laboratory, Department of Radiology, Massachusetts General Hospital/Harvard Medical School; ⁴Stroke Service and Neuroscience Intensive Care Unit, Department of Neurology, Massachusetts General Hospital/Harvard Medical School, Charlestown, Massachusetts, USA

Objective: Identifying changes in metabolic activity in brain tissue is important for gaining insight into functional activation, head trauma, and various neuropathologies such as stroke and Alzheimer's disease. Determining the distribution of oxygen in the cerebral vasculature is essential to the evaluation of metabolic activity. Currently, the most widely accepted techniques for measuring oxygen levels in vivo suffer from at least one of several possible limitations, including poor spatial resolution, the risk of physiological disruption resultant from an invasive measurement, or uncertainty associated with the fact that the technique does not provide a direct measure of O₂.

We have developed and tested a confocal optical imaging system to determine oxygen partial pressure (pO₂) in microvessels with high spatial resolution. The system quantifies pO₂ by measuring phosphorescence quenching of exogenous, O₂-sensitive dyes confined to blood plasma, such as Oxyphor R2. The phosphorescence lifetimes of these compounds vary with O₂ concentration as described by the Stern-Volmer equation. The technique, therefore, provides a minimally invasive, direct measurement of dissolved O₂ concentration in the cortical microvasculature.

Methods: The system's ability to perform lifetime imaging was evaluated by imaging heterogeneous phosphorescent phantoms. Theoretical calculations were also performed to determine the optimal imaging parameters to yield pO₂.

Experiments were performed to quantify pO₂ in cortical vessels of Sprague Dawley rats under a variety of conditions such as hypoxia, hyperoxia, and functional stimulation. The vessels were exposed through a sealed cranial window. After intravenously administering R2, the dye was excited using a low power, continuous wave 532 nm diode laser. An electro-optical modulator was used to control and gate the intensity of the excitation beam. Phosphorescence signal was collected through time-correlated single photon counting at a 50 MHz sampling rate using an avalanche photodiode. We developed custom software to control data acquisition and image processing. pO₂ was determined at select points of interest by binning the photon counts, fitting the phosphorescence decay profiles to an exponential decay, and subsequently using a calibration curve to relate lifetime measurement to pO₂.

Results: Figure 1 displays an image of merging cortical veins of a rat under normoxic conditions, collected by integrating the phosphorescence decay profiles at each pixel. The non-uniform brightness both between and within the vessels illustrates the heterogeneity of vascular pO₂ over short distances. By varying the fraction of inspired oxygen, changes in phosphorescence lifetime within the vessels are observed.

Figure 1

Merging veins in rat cortex.

Conclusions: Our system demonstrates potential for characterizing spatial and temporal variation of cerebral pO₂ under several different physiological conditions. When coupled with simultaneous blood flow measurements, our system will allow for high-resolution quantification of cerebral metabolic rate of oxygen (CMRO₂), providing a better understanding of metabolic dynamics during functional stimulation and under various neuropathologies.

324. Two-photon measurement and modeling of NADH emission changes in cortex during hypoxia, cortical spreading depression, and functional activation

S. Sakadžić¹, V.J. Srinivasan¹, M.A. Yaseen¹, A. Devor^1,2, C. Ayata^3,4 and D.A. Boas¹

¹Department of Radiology, MGH/MIT/HMS Athinuola A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, Massachusetts; ²Department of Neurosciences and Radiology, University of California at San Diego, San Diego, California; ³Stroke and Neurovascular Regulation Laboratory, Department of Radiology; ⁴Stroke Service and Neuroscience Intensive Care Unit, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, Massachusetts, USA

Objectives: Quantification of NADH changes in astrocytes and neurons during functional brain activation and pathological conditions provides critical insight in mechanisms of neurovascular coupling and brain metabolism.^1–3 Two-photon (2P) microscopy is becoming a tool of choice for in vivo measurement of NADH changes on the cellular level that is driving current progress in our understanding of astrocytic and neuronal metabolism.^1,2,4 However, the NADH emission spectrum is strongly attenuated by hemoglobin and in in vivo measurements the changes in NADH emission are affected by hemodynamic responses. We used modeling of excitation and emission light propagation based on microscopic 3D maps of rat cerebral vasculature to determine the effect of hemodynamic responses on detected NADH emission in 2P microscopy at various depths and conditions. The modeling procedure was then applied to evaluate the true NADH changes in our measurements during pathological changes and functional activation at different depths.

Methods: Two-photon imaging of the rat's cortex was performed through a sealed cranial window. Blood plasma and astrocytes were labeled with fluorescein-conjugated dextran (FITC) and Sulforhodamine101 (SR101), respectively. Simultaneous imaging of NADH fluorescence emission, astrocytes, and blood vessels was performed with three separate detectors in a commercial 2P microscope (Ultima, Prairie Technologies Inc.) using 740 nm excitation wavelength. Imaging was performed during hypoxia, cortical spreading depression, and functional activation of the forepaw region. Based on the measured 3D map of the micro-vasculature we modeled the effect of blood volume and oxygen saturation changes on both excitation and emission of NADH fluorescence using ray tracing and Monte Carlo simulations.

Results: Based on the fluorescence excitation and emission propagation modeling we established a range of depths and distances from the cortical vessels of different diameter where the detection of the NADH emission is not significantly affected by the hemodynamic responses. We determined individual contributions of blood volume and oxygen saturation changes on detected NADH emission signal, and found that the blood volume changes are playing a dominant role. Changes in the detected signal of other dyes (SR101 and FITC) at all depths can also be used as valuable indicators of the hemodynamic response's influence on the NADH signal. Finally, light propagation modeling was applied to validate the measurements during pathological conditions and functional activation at different depths.

Conclusions: We developed a detailed excitation and emission light propagation modeling for 2P imaging of the NADH changes. The modeling was used to determine the hemodynamic response influence on measured NADH emission and it was applied to validate the NADH measurements during pathological conditions and functional activation. This study will help to improve the accuracy of in vivo 2P measurement of cortical metabolism. Analogous modeling can be used to assist 2P imaging of other fluorophores important for neuronal activation and brain metabolism such as calcium dyes and flavins.

350. Uncoupling of glucose uptake in neurons and astrocytes in somatosensory cortex in vivo

J. Chuquet, P. Quilichini and G. Buzsáki

Rutgers University, Newark, New Jersey, USA

Glucose is the primary energetic substrate of the brain and measurements of its metabolism are the basis of major functional cerebral imaging methods. Contrary to the dogmatic view that neurons are fueled solely by glucose in proportion to their energetic needs, recent in vitro and ex vivo analyses suggest that glucose preferentially feeds astrocytes. However, the cellular fate of glucose in the intact brain has not yet been directly observed. We have used a real-time method for measuring glucose uptake in astrocytes and neurons in vivo by imaging the trafficking of the non-hydrolysable glucose analog 6-NBDG using two-photon microscopy. During resting conditions we found that astrocytes and neurons both absorbed 6-NBDG at the same rate in the barrel cortex of the rat. However during intense neuronal activity triggered by whisker stimulation, astrocytes rapidly accelerated their uptake whereas neuronal uptake remained unchanged. Following the stimulation period, astrocytes returned to their pre-activation rates of uptake paralleling the neuronal rate of uptake. These observations suggest that glucose is metabolized primarily in astrocytes, supporting the hypothesis that astrocytes play a central role in the energetic fueling of neurons.

363. Neurochemical and pharmcological studies in rat model of diet restriction-induced anorexia

T. Malik and D. Haleem

Neurochemistry and Biochemical Neuropharmacology Research Laboratory Biochemistry, Karachi University, Karachi, Pakistan

Anorexia nervosa (AN) patients exhibit extreme dieting, body weight loss and hyperactivity. The 5-hydroxytryptamine (5-HT; serotonin) system involved in the regulation of appetite and mood is the major neurotransmitter system of interest in research on AN. Pharmacological studies show that manipulations that tend to increase brain serotonin functions are anorexiogenic. The hypothesis of suppression of appetite through excessive release of 5-HT to receptors is not supported by data on subjects with clinical symptoms of AN as cerebrospinal (CSF) levels of 5-hydroxyindoleacetic acid (5-HIAA), a major metabolite of 5-HT, are reduced in AN patients and returned to normal in recovered patients. Loss of appetite in AN may simply follow self imposed dieting and diet restriction (DR). The hypothalamus is believed to be the site of the brain transducing satiety signals of serotonin. Studies on animal models show that excessive DR decreases 5-HT metabolism and synthesis in the brain and hypothalamus. The present lecture explains mechanism involved in DR-induced decreases of brain 5-HT. Possible role of regional 5-HT change in the elicitation of DR-induced deficits of behavior is discussed. Research on animal models of AN may help in developing strategies for the treatment of AN patients.

Figure

Diet restriction-induced anorexia.

371. Assessing the quantification of CMR_O2 from blood oxygen measurements

A. Vazquez, M. Fukuda and S.-G. Kim

Radiology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

Objective: The metabolic rate of oxygen consumption (CMR_O2) is a direct index of brain function under resting and neurally active conditions. Although it was initially thought that CMR_O2 does not change much with neural activity, studies have shown measurable increases with increased function.^1–4 CMR_O2 is difficult to measure and most methods rely on its calculation from arterio-venous differences in blood oxygenation. Several assumptions are typically made to simplify this calculation:

• (1)

  the arterial blood is fully saturated,

• (2)

  the transport of oxygen from blood to tissue is rapid enough such that the average tissue oxygen tension is in equilibrium with the capillary oxygen tension, and

• (3)

  the changes in CMR_O2 in mitochondria are rapidly equilibrated with the average tissue oxygen tension.

In this work, these assumptions are investigated.

Methods: For this purpose, the oxygen tension (P_O2) in a pre-penetrating arteriole and an emerging venule were measured using two oxygen probes (4 μm tip diameter each) over the somato-sensory cortex of isoflurane anesthetized rats (n = 6). Optical imaging (620 nm light) was performed to map the somato-sensory area responding to electrical forepaw stimulation and target the placement of the probes. The tissue oxygen tension and regional cerebral blood flow (CBF) were also measured using an oxygen probe (30 μm tip diameter positioned 300 μm in depth) and a LDF probe (450 μm tip diameter), respectively. These measurements were recorded during forepaw stimulation (20 secs duration, 1.6 mA, 1.0 ms pulses at 3 Hz, every 80 secs) under two experimental conditions: control and suppressed-CBF (achieved using a vasodilatory agent). Local field potential activity was measured using a metal electrode embedded in the tissue P_O2 sensor to monitor electrical activity and ensure that there are no significant differences in the activity between experimental conditions.

Results: Under control conditions, stimulation induced a measurable, average increase in arteriolar P_O2 from 77.9 to 88.4 mm Hg, or from 87.6 to 90.9% in oxygen saturation (calculated using the Hill equation). As expected, tissue and venous P_O2 increases were also observed with evoked stimulation. Under CBF-suppressed conditions, significant decreases in the tissue and venous P_O2 responses were observed with forepaw stimulation. The time-to-20%-peak and time-to-80%-peak of the average tissue P_O2 response were 1.4 and 6.7 secs, respectively. In addition, the lag between the venous and tissue P_O2 responses was measured to range between 1.3 and 2.1 secs based on their time-to-20%-peak and time-to-80%-peak.

Conclusions: Collectively, these findings show that a simple arterio-venous difference model may suffer from inaccuracies that depend on the volume encompassing the CMR_O2 calculation and also on the temporal scale. For example, the CMR_O2 baseline for these data was calculated to be 7.6 mL/min and stimulation increased CMR_O2 by 17.5%; if the arterial P_O2 is assumed to be 100 mm Hg, the baseline CMR_O2 would be 10.7 mL/min with a 13.6% increase with stimulation. In addition, the CMR_O2 values over transition regions are generally overestimated.

431. Effect of valproate on rat cerebral glutamine metabolism: a carbon 13 cellular metabolomic approach applicable to drug development

M. El Hage, B. Ferrier, G. Baverel and G. Martin

Metabolys Inc., Lyon Cedex, France

Objectives: It is well established that glutamine metabolism plays a central role in brain functions; indeed, glutamine is both synthesized in astrocytes to detoxify glutamate and ammonia and metabolized in neurons for the provision of the neuromediators glutamate and GABA. The objectives of the present study were:

• (i)

  to provide a panoramic view of cerebral glutamine metabolism in brain slices in vitro and

• (ii)

  to evaluate this model and the cellular metabolomic approach as tools for drug research and development.

Methods: For this, slices prepared from rat brain hemispheres were incubated under appropriate conditions with 5 mmol/L L-[3-¹³C]glutamine in the absence and the presence of 1 mmol/L valproate, a widely used antiepileptic drug. At the end of incubation, substrate removal and product formation were measured by both enzymatic and carbon 13 NMR spectroscopy methods. When combined with an original mathematical model of cerebral glutamine metabolism, the latter methods allowed to calculate fluxes through the enzymatic steps involved.

Results: In the absence of valproate, 3-¹³C-glutamine was used as substrate at high rates by glutaminase but the glutamine removal measured enzymatically was much smaller than the 3-¹³C-glutamine removal indicating that glutamine synthesis also occurred in astrocytes. Concomitant removal of the glutamate present at zero-time and 3-¹³C-glutamate accumulation were also observed. The lactate found at zero-time was also used as substrate and only a small ¹³C-lactate synthesis occurred. Small amounts of unlabeled alanine but not of pyruvate accumulated. GABA accumulation was found by both enzymatic and NMR measurements; GABA was found labeled mainly on its C3 (direct synthesis) and, to a very small extent, on its C2 (indirect synthesis). Accumulation of aspartate, labeled mainly on its C2 and C3 and, to a lesser extent, on its C1 and C4 was also observed. Labeling of glutamate and glutamine on carbons other than their C3 reveals that resynthesis of both amino acids occurred. Approximately half of the C3 of the 3-¹³C-glutamine removed was released as CO₂.

In the presence of valproate, 3-¹³C-glutamine removal was slightly inhibited, but net glutamine synthesis, glutamate accumulation and labeling, lactate removal and labeling as well as alanine accumulation remained unchanged. Aspartate accumulation and labeling on its C2 and C3 (but not on its C1 and C4) and GABA labeling on its C3 were decreased by valproate. The addition of valproate also diminished the production of ¹³CO₂.

Flux calculations indicate that valproate slightly inhibited flux through glutaminase and caused an accumulation of GABA by the direct pathway. It also inhibited fluxes through succinate semialdehyde dehydrogenase plus alpha-ketoglutarate dehydrogenase, succinate dehydrogenase, fumarase, malic enzyme, aspartate aminotransferase and pyruvate carboxylase. By contrast, valproate did not alter fluxes through alanine aminotransferase, lactate dehydrogenase, pyruvate dehydrogenase, citrate synthase, aconitase, isocitrate dehydrogenase, glutamate dehydrogenase, glutamine synthetase and glycogen phosphorylase.

Conclusion: It is concluded that the model used in vitro and the cellular metabolomic approach employed are suitable for studying the beneficial and adverse interactions of drugs with brain energy and intermediary metabolism.

437. Metabolic role of aquaglyceroporin 9 in astrocytes

C. Guérin¹, J.-F. Brunet¹, N. Mastour¹, L. Regli², L. Pellerin³ and J. Badaut^1,4,5

¹Neurosurgery, Lausanne University Hospital, Lausanne, Switzerland; ²Neurosurgery, University Medical Center Utrecht, Utrecht, Netherlands Antilles; ³Physiology, Lausanne University, Lausanne; ⁴Clinical and Fundamental Neurosciences, University of Geneva, Geneva, Switzerland; ⁵Pediatrics, Loma Linda University, Loma Linda, California, USA

Aim: Aquaglyceroporin-9 (AQP9) is a member of the Aquaporin channel family involved in water flux through plasma membranes and exhibits the distinctive feature of also being permeable to glycerol and monocarboxylates. AQP9 is detected in astrocytes and catecholaminergic neurons. ¹ However, the presence of AQP9 in the brain is now debated after a recent publication claiming that AQP9 is not expressed in the brain. ² Based on our results, ³ we have evidence of the presence of AQP9 in the brain and we further hypothesize that AQP9 plays a functional role in brain energy metabolism.

Methods: The presence of AQP9 in brain of OF1 mice was studied by RT-PCR and immunohistochemistry. To address the role of AQP9 in brain, we used commercial siRNA against AQP9 to knockdown its expression in 2 cultures of astrocytes from two distinct sources (from differentiated stem cells ⁴ and primary astrocyte cultures). After assessment of the decrease of AQP9, glycerol uptake was measured using [H³]-glycerol. Then, modifications of the astrocytic energy metabolism was evaluated by measurement of glucose consumption, lactate release ⁵ and evaluation of the mitochondrial activity by MTT staining.

Results: AQP9 is expressed in astrocytes of OF1 mouse brain (mRNA and protein levels). We also showed that AQP9 mRNA and protein are present in cultured astrocytes. Four days after AQP9 siRNA application, the level of expression is significantly decreased by 76% compared to control. Astrocytes with AQP9 knockdown exhibit a 23% decrease of glycerol uptake, showing that AQP9 is a glycerol channel in cultured astrocytes. In parallel, astrocytes with AQP9 knockdown have a 155% increase of their glucose consumption without modifications of lactate release. Moreover, considering the observed glucose consumption increase and the absence of proliferation induction, the significant MTT activity increase (113%) suggests an increase of oxidative metabolism in astrocytes with AQP9 knockdown.

Discussion: The involvement of AQP9 in astrocyte energy metabolism adds a new function for this channel in the brain. The determination of the role of AQP9 in astrocytes provides a new perspective on the controversial expression of AQP9 in brain. We also suggest that AQP9 may have a complementary role to monocarboxylate transporters in the regulation of brain energy metabolism.

Financial support: SNF-3100AO-108001 and 31003A-122166, Novartis foundation.

443. Green tea (Camellia sinensis) potentiates extrapyrmidal symptoms of haloperidol: a study in rat model

T. Malik and D. Haleem

Neurochemistry and Biochemical Neuropharmacology Research Laboratory, Biochemistry, Karachi University, Karachi, Pakistan

Objectives: Schizophrenia, a psychiatric illness, has been treated with typical antipsychotic drug haloperidol (HAL). Although the treatment is associated with a high rate success but Extrapyramidal Symptoms (EPS) associated with the treatment is a serious limitation of the therapy. Some studies have suggested that oxidative stress induced during the metabolism of HAL is involved in the elicitation of EPS. The components of green tea (Camellia sinensis) have been demonstrated to have therapeutic efficacy in reducing oxidative stress. It has been reported that flavoniods of green tea have free radical scavenging properties thus antioxidant in nature. Therefore, it can be speculated that green tea may prevent the EPS induced by HAL.

Methods: In the cohort design of study we examined the efficacy of Green Tea Extract (GTE) (1 g/L) on HAL (0.1 g/kg)—induced EPS. Treatment was done for 5 weeks. Groups were tested for motor coordination on Rota-rod each week. Tardive Vacuous Chewing Movements (t VCMs) were also monitored by an observer blind to the treatment. The animals were scarified after 35th hour of drug withdrawal. The brain samples from dorsal and ventral striatum were taken and analyzed by HPLC.

Results: We found that HAL—induced impairment of motor coordination were greater in GTE than water treated animals. The elicitation of t VCMs were also significantly greater in GTE treated animals. The metabolism of dopamine was greater in the Nuclease Accumbens (NA) in dorsal striatum of GTE plus HAL model than water plus HAL treated animals. The results suggest that increases of DA and metabolism in the dorsal striatum may be involved in the elicitation of greater EPS induced by HAL in GTE treated groups, conversely the levels of DA was decreased in the rest of the brain regions of GTE treated animals. It is suggested that an increase in DA metabolism observed in the Ventral Tegmental Area (VTM) of ventral striatum of GTE than water treated animals may potentiate schizophrenic symptoms.

Conclusion: Our results recommend that GTE intake may provoke more vulnerability to HAL—induced side effects and more likely to relapsed schizophrenic symptoms in patients treated with haloperidol. We suggest that patients on HAL therapy should avoid green tea. The mechanism by which green tea may potentiate HAL—induced EPS is discussed.

Figure

Dopamine and metabolism in the nuclease accumbens.

462. Brain glutamatergic activity and EAC cell development: effect of caffeine

M.K. Poddar and A. Mandal

Department of Biochemistry, University of Calcutta, Kolkata, India

Brain is thought to function normally under equilibrium condition in the interaction of inhibitory and excitatory neurotransmitters, γ-amino butyric acid (GABA) and glutamate respectively. The excitation and inhibition homeostatic balance is known to alter by exogenous and endogenous factors that may alter due to stress or some pathological conditions including neoplastic transformation. ¹ Such changes in neuronal function bring about alteration at the biochemical and physiological levels including endocrine and immune systems. It is well known that centrally active drugs interacts with different types of CNS activities, whose actions having clinical values in animals including human depending on their (drugs) dosages, toxicity, duration of exposure as well as their efficacy. Caffeine (plant alkaloid, 1,3,7-trimethyl xanthine), present in coffee, is one of the CNS stimulant having significant place in modern medicine due to its worldwide consumption and high bio potency. It (caffeine) also acts as chemopreventive, antioxidant and anticancer agent. ² It is biochemically as well as pharmacologically acts as an adenosinergic antagonist. ² Recent-past observation from our laboratory has shown that caffeine attenuates the EAC cell progression through reversal of EAC cell-induced dysfunction of gonadal hormones, suppression of specific and nonspecific immune response, diminution of antioxidant activity and induction of central GABA-ergic activity. ² Since glutamatergic and GABA-ergic systems interact each other ³ and adenosine, the inhibitory neurotransmitter through its receptors plays a modulatory role on glutamatergic activity, ⁴ the present authors are interested to study the effect of long-term consumption of caffeine on brain glutamatergic activity during the development of Ehrlich ascites tumor cells in adult albino female mice of Swiss strain (25 to 30 gm). In the present study animals were treated either with caffeine (20 mg/kg/day, p.o.) for 22 to 27 consecutive days or inoculated with EAC cells (5 × 10⁶ cells/mL, i.p.) and allowed to develop for 10 to 15 days or animals bearing EAC cells were treated with caffeine for 22 to 27 consecutive days. Control animals were treated with corresponding vehicle(s) under similar conditions. Development of EAC cells in mice impaired the whole brain GABA/glutamate-glutamine cycle regulation along with a significant reduction in central glutamatergic activity (by measuring the steady state levels of glutamine, glutamate and GABA; enzyme activities of glutamic acid decarboxylase, GABA-transaminase, glutaminase and glutamine synthetase along with the specific binding of glutamate receptor (using ¹⁴C-glutamate) including the kinetic parameters, maximum binding (Bmax) and binding affinity (1/Kd)). Although caffeine consumption (20 mg/kg/day, p.o.) for 22 to 27 consecutive days did not change the central glutamatergic activity, its consumption during EAC cell development retarded EAC cell growth and also restored/attenuated the EAC cell-induced inhibition of brain glutamatergic activity. These results therefore, suggest that long-term consumption caffeine may reduce the development of EAC cells through the attenuation/restoration of EAC cell-induced inhibition of central GABA-glutamate regulation and its activity in mice.

Supported by University of Calcutta, Kolkata & University Grants Commission, New Delhi, India.

608. Gamma oscillations in the hippocampus are associated with high oxygen consumption

O. Kann¹, C. Huchzermeyer¹, R. Kovács¹, S. Wirtz² and M. Schuelke²

¹Neurophysiology; ²Neuropediatrics, Charité-Universitätsmedizin Berlin, Berlin, Germany

Background and aims: Fast neuronal network oscillations in the gamma range (30 to 80 Hz) have been implicated in complex brain functions such as sensory processing and memory formation, and they might be associated with high energy demands.

Methods: We applied local field potential recordings, oxygen sensor microelectrodes, fluorescence imaging techniques (NAD(P)H, FAD) and pharmacology.

Purpose: We explored the fundamental relationships between gamma oscillations and mitochondrial functions (oxygen consumption, mitochondrial redox state) in hippocampal slice preparations.

Results: Activation of cholinergic receptors in hippocampal slices by acetylcholine evokes robust and persistent gamma oscillations and is associated with high oxygen consumption. Both findings are more prominent in subfield CA3 as compared to CA1 and the dentate gyrus. Moreover, gamma oscillations are associated with significant changes in mitochondrial redox state and are highly sensitive to decreases in interstitial partial oxygen pressure (pO₂).

Conclusions: Our data suggest that gamma oscillations are highly energy consuming and require a strong functional performance of neuronal mitochondria. This might explain the exceptional vulnerability of complex brain functions under pathological conditions.

922. Plasma concentration of pentraxin3 (PTX3) in acute cerebral infarction

T. Tajima, M. Yamazato, W. Hara, A. Saitou, J. Akiyama, A. Kubota, S. Narukawa, M. Kojima, S. Izaki, N. Yoshida, S. Ohji, T. Iguchi, T. Mitui, M. Tkahama, M. Ohnuki and K. Nomura

Neuroogy, Saitama Medical Center, Saitama Medical University, Kawagoe, Saitama, Japan

Objectives: The utility of high sensitivity CRP (hs-CRP) is reported as a blood vessel inflammation marker of arteriosclerosis. However, the uniqueness to the blood vessel inflammation is low. Pentraxin3(PTX3) is an acute inflammation protein that belongs to Petraxin family as well as CRP, and it has exhibted on endothelial cell, smooth muscle cell, and macrophage. It has the possibility to become a specific biomarker for the blood vessel inflammation. There are some reports that plasma PTX3 concentration increased in the acute coronary syndrome. ¹ However, there is no report in acute cerebral infarction (ACI).

Methods: We measured the concentration of plasma PTX3 and serum hs-CRP, 45 patients with ACI containing five patients treated with t-PA, admitted acute stroke within 48 h (36 men and 9 women; mean age, 69.1±7.8 years; 15 cardiac embolism (CE), 16 atherothrombotic infarction (ATI), 10 lacnar infarction (LI), 4 TIA and 56 age-matched healthy control (HC)).

Results:

• (1)

  Plasma PTX3 concentration within 24 h was no significant difference in four types of ACI (CE 4.31±1.89, ATI 3.57±3.86, LI 2.02±0.24, and TIA 2.29±0.47 ng/mL).

• (2)

  Plasma PTX3 concentration of CE in 48 h was significantly higher than that of ATI, LI, TIA, and HC (P<0.01, respectively) (CE 12.52±8.68, ATI 4.62±3.38, LI 2.58±1.59, TIA 2.07±0.76, and HC 1.98±1.14 ng/mL, Figure 1).

• (3)

  Serum hs-CRP concentration within 48 h was no significant difference in four types of ACI (n = 48, P = 0.13).

• (4)

  There is no correlation between PTX3 and hs-CRP within 48 h (n = 24, P = 0.45).

• (5)

  There is a significant correlation between PTX3 in 48 h and modified Rankin scale (mRS) in the discharge (n = 40, P<0.01, Figure 2).

Conclusions:

• (1)

  Plasma PTX3 concentration increased after cardiac embolism, and it was suggested to be a useful indicator of ACI.

• (2)

  Plasma PTX3 within 48 h after stroke was suggested to be a prognostic indicator to estimate of functional grade in the discharge.

982. Alteration of brain glycogen turnover in the awake rat

F.D. Morgenthaler¹, B. Lanz¹, J.-M. Petit², H. Frenkel¹, P.J. Magistretti² and R. Gruetter^1,3

¹Center for Biomedical Imaging; ²Laboratory of Neuroenergetics and Cellular Dynamics, EPFL, Lausanne; ³Department of Radiology, University of Geneva, Geneva, Switzerland

Objectives: Sleep deprivation elicit complex responses in brain glycogen (Glyc) synthesis and breakdown.^1,2 Moreover, brief sensory stimulation alters Glyc turnover. ³ The increased Glyc turnover during activation complicates interpretation of standard approaches that assess concentration ([ ]) and metabolism to evaluate roles or utilization of Glyc under various conditions. ³

The aim of this study was to determine brain Glyc concentration ([ ]) and turnover time (tau) in euglycemic conscious rats maintained awake during 5 h using stable isotopes.

Methods: The metabolism of [1-¹³C]-labeled glucose (Glc) into Glyc was measured in rats that had previously received a 10% w/v [1-¹³C]-labeled Glc solution ad libitum as their sole source of exogenous carbon for either 5 (n = 6+7), 24 (n = 5) or 48 h (n = 5). Six animals were continuously aroused by acoustic, tactile and olfactory stimuli for 5 h, and otherwise treated identically, resulting in elevated plasma corticosterone levels (97±21 ng/mL, n = 6) compared to the control—undisturbed—rats (66±52 ng/mL, n = 7, (mean)±s.d.). To minimize postmortem degradation, rats were sacrificed using a focused microwave fixation device, brains dissected and assayed for tissue glycogen and Glc concentrations using biochemical measurements. ^{4 13}C enrichment of N-acetyl-aspartate (NAA) and Glc (as well as digested glycogen) was evaluated in vitro by high field ¹H Nuclear Magnetic Resonance (NMR) spectroscopy (600 MHz). ⁵

Results: Brain [Glyc] of the aroused rats (3.9±0.6 μmol/g, n = 6) was not significantly different from that of the control group (4.0±0.4 μmol/g, n = 7; t-test, P>0.5). To account for potential variations in plasma Glc Isotopic Enrichment (IE), Glyc IE was normalized by NAA IE, which has a turnover time on the same order of magnitude. A simple mathematical model was developed to derive brain Glyc and NAA turnover times as 5.3±3.2 h and 15.6±6.5 h (fitted value±s.d.) respectively, in the control rats. Finally, assuming that tau_NAA was not altered by 5 h of awakening, a faster tau_Glyc (2.9±1.2 h) was estimated in the aroused rats. Moreover, the model established in the present study shows that a calculated value of tau_Glyc in case of a shorter tau_NAA would anyway be smaller than the value found with an assumption of unchanged tau_NAA. Subsequently, the difference in tau_Glyc in case of a shorter tau_NAA between control and aroused rats would be even bigger than reported.

Conclusion: In conclusion, 5 h of prolonged wakefulness mainly activates glycogen metabolism, but has minimal effect on brain [Glyc].

989. The brain releases IL-6 in recovery from exercise concomitant with increased IL-6 mRNA expression

P. Rasmussen¹, J.-C. Vedel¹, M.V. Pedersen¹, E. Hart², N.H. Secher¹, H. Adser¹, H. Pilegaard¹ and B.K. Pedersen¹

¹University of Copenhagen, Copenhagen, Denmark; ²Brunel University, Uxbridge, UK

Interleukin-6 influences metabolism including inducing lipolysis and fat oxidation. Plasma IL-6 increases during exercise as well as in the recovery maybe mediating some of the health benefits of exercise. Both active muscles and the brain contribute to plasma IL-6, but whether the brain also increases IL-6 release in the recovery from exercise is unknown. To address this question, nine healthy male subjects completed a 4 h bout of ergometer rowing, while arterio-jugular venous difference (a–v diff) of IL-6 was measured. The IL-6 a–v diff was −2.2±1.9 pg mL⁻¹ (P<0.05) after 4 h of exercise and importantly, the IL-6 a–v diff over the brain was still −2.1±2.1 pg mL⁻¹ (P<0.05) after 1 h of recovery. To examine whether the exercise-induced IL-6 release may be associated with increased cerebral IL-6 expression, the IL-6 mRNA content was determined in mice hippocampus, cerebellum and cortex immediately after a single treadmill exercise and at 2, 6 and 24 h hours of recovery and compared with mice not run acutely. IL-6 mRNA was expressed in all three brain parts examined with higher (P<0.05) content in cortex than in hippocampus and cerebellum. Exercise induced 2 to 3 fold increase (P<0.05) in the IL-6 mRNA content in hippocampus only. In conclusion, the results indicate that hippocampus is a likely source of exercise-induced brain-derived IL-6. As IL-6 release in the human brain correlated to cerebral lactate release, we speculate that increased cerebral IL-6 expression and release may be elicited by changes in cerebral metabolism.

1018. Oxygen extraction fraction is unchanged in healthy aging

J. Aanerud¹, P. Borghammer¹ and A. Gjedde²

¹PET Centre, Aarhus University Hospitals; ²CFIN, Aarhus University, Aarhus, Denmark

Introduction: We tested the hypothesis that the oxygen extraction fraction increases with age. Kety (1956) reported an increased arteriovenous oxygen difference with age in a meta-analysis of 16 studies, in which mean age ranged from 5 to 93 years. In our case PET measurements of cerebral blood flow (CBF) and oxygen consumption (CMRO2) were used to estimate oxygen extraction fraction (OEF), which corresponds to arteriovenous oxygen difference.

Materials and methods: Fifty-nine healthy subjects (39 men) aged 21 to 73 years were included. Parametric maps of CBF, OEF and CMRO2 were calculated from the PET image volumes, and non-linearly co-registered to common space, via individual anatomical MR images. Linear voxelwise regression of CBF, CMRO2, and OEF on age were performed using fMRIstat. OEF-maps were created by dividing CMRO2- maps with CBF-maps and hemoglobin concentration.

Results: The regression showed no significant changes in OEF with increasing age. Throughout the brain t-values remained close to zero, i.e. 95.7% of the intra-cerebral voxels exhibited t-values ranging from −2 (decreased OEF) to 2 (increased OEF).

Discussion: Based on this dataset the hypothesis of increased OEF with aging cannot be upheld. On the contrary, the result indicates that blood flow is tightly regulated across the age span investigated.

1053. Traumatic brain injury-induced expression and phosphorylation of pyruvate dehydrogenase: a mechanism of dysregulated glucose metabolism

G. Xing, M. Ren, W. Watson, J.T. O'Neill and A. Verma

Departments of Psychiatry, Pediatrics and Neurology, Uniformed Services University, Bethesda, Maryland, USA

Background: Dysregulated brain glucose metabolism and lactate acidosis are metabolic characteristics in people with traumatic brain injury (TBI). The molecular mechanism is poorly understood. Pyruvate dehydrogenase (PDH), the rate-limiting enzyme coupling cytosolic glycolysis to mitochondrial citric acid cycle, plays a critical role in maintaining homeostasis of brain glucose metabolism. PDH activity is maintained by the expression of its E1α1 subunit (PDHE1α1) and inhibited by the phosphorylation of PDHE1α1 (p-PDHE1α1). We hypothesized that PDHE1α1 expression and phosphorylation was altered in rat brain following controlled cortical impact (CCI)-induced TBI.

Objectives: To determine PDHE1α1 mRNA and protein expression and phosphorylated PDHE1α1 (p-PDHE1α1) protein in rat brain homogenates at various time post- traumatic brain injuries induced by controlled cortical impact (CCI).

Methods: CCI was induced in young adult male Sprague-Daley rats (170 to 200 g) with a penetration width of 6 mm, depth of 2.5 mm, and velocity of 4 m/s. Brains were analyzed at 4 h, 1 d, 3 d and 7 d post TBI with immunohistochemistry, RT-PCR and western blot.

Results: Compared to naïve controls (= 100%), PDHE1α1 protein decreased significantly in ipsilateral CCI (62%, P<0.05; 75%, P<0.05; 57%, P<0.05; and 39%, P<0.01) and contralateral CCI (77%, 78%, 78% and 36% P<0.01) at 4 h, 24 h, 3- and 7-day post-CCI, respectively. p-PDHE1α1 protein level decreased significantly in ipsilateral CCI (31%, P<0.01; 102%, P>0.05; 64%, P<0.05; and 14%, P<0.01) and contralateral CCI (35%, 74%, P<0.05; 60%, P<0.05; 20% P<0.01) at 4 h, 24 h, 3- and 7-day post-CCI, respectively. Similar reduction in PDHE1α1 and p-PDHE1α1 protein was found in craniotomy (Sham CCI) group.

Conclusions: Our data showed that TBI-induced reduction in PDHE1α1 expression and phosphorylation could alter PDH activity and thus affect brain glucose metabolism.

1073. Characterization of lactate dehydrogenase isozymes in japanese quail brain (Coturnix Coturnix Japonica)

R.P. Singh¹, K.V.H. Sastry¹, N.K. Pandey¹, N. Shit¹, A. Ahuja¹, D. Radha¹, J. Mohan¹, K.B. Singh² and R.P. Moudgal¹

¹Division of Physiology and Reproduction, Central Avian Research Institute, Izatnagar, Bareilly; ²Depertment of Animal Science, M.J. P. Rohilkhand University Bareilly, Bareilly, India

Lactate dehydrogenase (LDH, EC 1.1.1.27) is a tetramer with two different types of polypeptide chains (subunits) viz. H and M. It exists in different molecular forms with different electrophoretic mobility. The distribution of different LDH isozymes is mostly tissue specific making them ideal molecular markers for the study of tissue energy metabolism. The LDH present in heart muscle consists of H-Type isozyme whereas in skeletal muscle it is of M-type isozyme. Presence of more than one LDH isozyme has been reported in brain. This distribution has been correlated with local oxygen tensions, pyruvate inhibition and lactate accumulation. Lactate dehydrogenase converts Pyruvate into lactate may use as a substrate in brain. The old hypothesis of brain energy metabolism was that oxidation of glucose is responsible for providing energy to the brain. However, recent research has suggested that lactate may be a major source in cerebral energy metabolism. In the functioning brain, lactate is the major end product of anaerobic glycolysis formed by one of the LDH isozyme and that lactate subsequently serves as a substrate for generation of significant amount of ATP to properly serve the metabolic functions of brain. Lactate dehydrogenase isozyme pattern of adult Japanese quail brain was investigated to examine brain energy metabolism by polyacrylamide-gel electrophoresis (PAGE) followed by densitometry. Twenty adult quails were sacrificed and separate brain tissue homogenate (10%) were prepared in 0.02 M Tris-Cl buffer (pH 7.4). The tissue homogenate was subjected to centrifuge. Supernatant was collected and used for enzyme activity and non-denaturing PAGE. The chicken brain tissue subjected to the same treatment was used as a reference for electrophoresis. The enzyme activity in the supernatant was 28.62±0.8400 μmol min-1 mg-1protein. The zymogram of quail brain isozymes revealed four different bands of LDH-1, LDH-2, LDH-3 and LDH-4 with a marked difference in their enzyme activity respectively as compare to chicken. The zymogram of chicken brain isozymes also showed four bands of equal enzyme activity at identical electrophoretic mobility to the quail brain. Among all four isozymes present in the quail brain the highest band intensity was observed in LDH-4. Densitometry again confirmed that LDH-4 was in highest (75.6%) proportion in quail brain. These results indicated that specifically higher amount of LDH-4 isozyme was only observed in quails was unique. The results are in support of the hypothesis that adult quail brain favors anaerobic tissue metabolism though it may function in both aerobic and anaerobic conditions.

1082. Chromium histidinate increases brain GLUT-1 and GLUT-3 levels impaired by insulin resistance

J. Komorowski¹, M. Tuzcu², N. Sahin² and K. Sahin²

¹Technical Services & Scientific Affairs, Nutrition 21, Purchase, New York, USA; ²Firat University, Elazig, Turkey

Objectives: Chromium is an essential trace mineral and a cofactor for insulin function. Chromium, in the form of chromium picolinate, has been shown to lower elevated blood glucose levels in people with type 2 diabetes. ¹ Chromium's beneficial effects on blood glucose levels may be due to its ability to increase insulin dependent membrane-associated GLUT-4 levels. ² Hou et al ³ found that chronic hyperglycemia downregulated GLUT-1 and GLUT-3 expression in the brain. The current study was conducted to evaluate the effect of chromium, in the form of chromium histidinate (CrHis), on brain GLUT-1 and GLUT-3 levels in overweight, insulin resistant rats.

Methods: Insulin resistance and hyperglycemia were induced in male Wistar rats by feeding a high fat diet (HFD, 40% of calories as fat). A healthy control group received a standard diet (12% of calories as fat). Rats (10/group) were then divided into groups: healthy control (Control), HFD, and HFD+CrHis. Chromium (Cr) treated rats were fed CrHis (110 mcg/kg body wt/d). After 12 weeks, Cr levels and GLUT-1 and GLUT-3 expression (Western blot analysis) were measured from brain tissue homogenates. Body weight, serum Cr, and blood glucose levels were also measured.

Results: The study results are shown in Table 1. HFD significantly decreased brain Cr (−26%), GLUT-1 (−38%) and GLUT-3 (−11.2%) levels compared to Control. HFD also significantly increased body weight (+14%), increased blood glucose levels (+30%), and decreased serum Cr levels (−20%). Addition of CrHis to the HFD significantly increased brain and serum Cr levels, increased brain GLUT-1 and GLUT-3 levels, and decreased body weight and blood glucose levels, compared to HFD alone.

Table 1

Comparison of efficacy variables after treatment (mean±s.e.m.)

	Control	HFD	HFD+CrHis
Body weight (g)	288.7±3.0a	329.2±1.3b	315.1±1.9c
Blood glucose (mg/dL)	101.5±1.6a	132.3±2.3b	116.4±2.1c
Serum chromium (ng/g)	16.9±0.4a	13.6±0.3b	22.0±0.6c
Brain chromium (ng/g)	15.6±0.3a	11.2±0.2b	17.7±0.2c
GLUT-1 (% of control)	100.0±1.2a	62±1.5b	89.5±0.3c
GLUT-3 (% of control)	100.0±0.6a	88.8±3.8b	106.3±1.5a

Different letters represent statistical significance (P<0.01) between groups.

Conclusions: This is the first study to demonstrate that chromium administration can enhance brain GLUT-1 and GLUT-3 expression. This study also shows that a HFD decreases brain chromium levels, and brain GLUT-1 and GLUT-3 levels. The ability of chromium histidinate to enhance brain GLUT-1 and GLUT-3 levels may be directly related to: (1) actions of the chromium in the brain, (2) the indirect effects of chromium on systemic blood glucose levels, or (3) both. Additional research is warranted to further evaluate additional cerebral metabolic effects of chromium histidinate.