Altered 13 C Glucose Metabolism in the Cortico—Striato—Thalamo—Cortical Loop in the MK-801 Rat Model of Schizophrenia

Abstract

Using a modified MK-801 (dizocilpine) N-methyl-d-aspartic acid (NMDA) receptor hypofunction model for schizophrenia, we analyzed glycolysis, as well as glutamatergic, GABAergic, and monoaminergic neurotransmitter synthesis and degradation. Rats received an injection of MK-801 daily for 6 days and on day 6, they also received an injection of [1-¹³C]glucose. Extracts of frontal cortex (FCX), parietal and temporal cortex (PTCX), thalamus, striatum, nucleus accumbens (NAc), and hippocampus were analyzed using ¹³C nuclear magnetic resonance spectroscopy, high-performance liquid chromatography, and gas chromatography—mass spectrometry. A pronounced reduction in glycolysis was found only in PTCX, in which ¹³C labeling of glucose, lactate, and alanine was decreased. ¹³C enrichment in lactate, however, was reduced in all areas investigated. The largest reductions in glutamate labeling were detected in FCX and PTCX, whereas in hippocampus, striatum, and Nac, ¹³C labeling of glutamate was only slightly but significantly reduced. The thalamus was the only region with unaffected glutamate labeling. γ-Aminobutyric acid (GABA) labeling was reduced in all areas, but most significantly in FCX. Glutamine and aspartate labeling was unchanged. Mitochondrial metabolites were also affected. Fumarate labeling was reduced in FCX and thalamus, whereas malate labeling was reduced in FCX, PTCX, striatum, and NAc. Dopamine turnover was decreased in FCX and thalamus, whereas that of serotonin was unchanged in all regions. In conclusion, neurotransmitter metabolism in the cortico—striato—thalamo—cortical loop is severely impaired in the MK-801 (dizocilpine) NMDA receptor hypofunction animal model for schizophrenia.

Keywords

dopamine GABA glutamate lactate

Introduction

The N-methyl-d-aspartic acid (NMDA) receptor hypofunction hypothesis, originally proposed by Olney and Farber (1995), suggests dysfunction of glutamatergic neurotransmission in schizophrenia. Indeed, NMDA receptor antagonists such as phencyclidine and ketamine exacerbate positive, negative, and cognitive symptoms in patients with schizophrenia (Lahti et al, 1995; Rujescu et al, 2006). Moreover, repeated MK-801 injections in rats serve as an animal model of schizophrenia (Kondziella et al, 2007), and induce negative, positive, and cognitive symptoms in rodents. This is in contrast to hyperdopaminergic agents such as amphetamines, which only induce positive symptoms (Rung et al, 2005). Consequently, attempts have been made to integrate the NMDA receptor hypofunction hypothesis with the well-known dopamine (DA) hypothesis (for review see Kondziella et al, 2007). Glutamatergic neurons in the frontal cortex (FCX) and the hippocampus modulate subcortical dopaminergic systems and vice versa, and Kegeles et al (2000) showed that disruption of glutamatergic neurotransmission with ketamine increased amphetamine-induced striatal DA release in healthy volunteers. For the first time, direct support was found for the interaction of glutamate and DA. Others have shown that blocking of NMDA receptors in the FCX increases DA release in nucleus accumbens (NAc; Del Arco and Mora, 2008). Taken together, this implies that schizophrenia might involve dysregulation of subcortical DA due to failure of control and/or loss of feedback by the FCX. In addition, disturbed interaction with other neurotransmitter systems, notably γ-aminobutyric acid (GABA), noradrenalin, and serotonin (5-HT), may have a role in schizophrenia (Carlsson et al, 2004).

Much attention has been given to the potential neuroprotective role of glutamate receptor antagonists, especially to those acting on the NMDA subtype. However, in addition to their neuroprotective potential, these compounds also have neurotoxic properties, which have been discussed earlier by our group with respect to the schizophrenia model of repeated injections of MK-801 (Eyjolfsson et al, 2006).

This paper examines whether glycolytic and mitochondrial dysfunction in the cortico—striato—thalamo—cortical loop has a role in schizophrenia (for a review see Kondziella et al, 2007). The main aim was to study how NMDA receptor hypofunction influenced glycolysis- and mitochondria-based amino-acid synthesis and degradation. In an earlier study, [1-¹³C]glucose metabolism was assessed with ¹³C nuclear magnetic resonance spectroscopy (NMRS) in rats subjected to daily injections of 0.5 mg/kg body weight MK-801 for 6 days (Kondziella et al, 2006). This study showed disturbed metabolism in the frontal lobe of the rat brain. However, the study was confined to examination of the frontal and temporal lobes only, and monoamine analysis was suboptimal as the animals had been killed by decapitation. In this study, we used microwave fixation and Rodent Brain Matrix (Asi-Instruments, Warren, MI, USA; RBM 4000C) to enable assessment of [1-¹³C]glucose metabolism, DA and 5-HT levels, and turnover. FCX, parietal and temporal cortex (PTCX), thalamus, striatum, NAc, and hippocampus were examined using ¹³C and ¹H NMRS to analyze glycolysis and amino-acid synthesis and degradation; high-performance liquid chromatography (HPLC) to measure levels of amino acids; and gas chromatography—mass spectrometry (GC—MS) to measure ¹³C incorporation from [1-¹³C]glucose into amino acids and tricarboxylic acid (TCA) cycle intermediates. Amino acid and monoamine neurotransmission were indirectly assessed based on metabolite levels and turnover.

Materials and methods

A total of 18 male Sprague—Dawley rats (Taconic, Ry, Denmark) with an average weight of 270 g were included in this experiment. Nine animals were used in the intervention group and nine as controls. [1-¹³C]Glucose (99%, ¹³C enriched) and D₂O (99.9%) were purchased from Cambridge Isotopes Laboratories (Woburn, MA, USA), ethylene glycol was purchased from Merck (Darmstadt, Germany), and MK-801 (dizocilpine; [5R, 10S]-[+]-5-methyl-10, 11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine) from Sigma-Aldrich (St Louis, MO, USA). All other chemicals were of the purest grade available from local commercial sources.

All animal procedures were approved by the Norwegian Animal Research Authority. Before experiments, animals were housed in individual cages with food and water available ad libitum. Animals were kept in a light/dark cycle of 12 hours, humidity 60%, and at a temperature of 22°C. In total, 10 days were allowed for acclimatization. After this, animals were given intraperitoneal injections of saline (nine animals) or MK-801 (nine animals; 0.5 mg/kg body weight) daily for 6 days. MK-801 induced hyperlocomotion, ataxia, abducted hind limbs, flat body posture, and stereotyped behavior, such as head waving, which were characterized by considerable inter- and intrasubject variability. The final dose of MK-801 or saline was given 30 minutes before intraperitoneal injection of [1-¹³C]glucose (543 mg/kg; 0.3 M solution). At 15 minutes after the last injection, the brains of the animals were subjected to microwave fixation at 4 kW for 2.2 seconds (Model GA5013, Gerling Applied Engineering, Modesto, CA, USA). Brains were dissected using Rodent Brain Matrix (Asi-Instruments, RBM 4000C) for coronal sections in adult rats. The following brain areas were obtained: FCX, PTCX, striatum, thalamus, Nac, and hippocampus. Dissections were performed on a Harris cutting mat and tissue samples were stored at −75°C. The frozen samples were extracted using 0.7% perchloric acid, and ultrasound was applied using a Vibra Cell sonicator (Model VCX 750, Sonics & Materials, Newtown, CT, USA). Homogenized tissue samples were centrifuged at 3,000 g at 4°C for 5 minutes. The pH of the supernatants was adjusted to 6.5 to 7.5 and the samples were lyophilized before analysis with NMRS, GC—MS, and HPLC.

Gas Chromatography—Mass Spectrometry

Lyophilized brain extracts were redissolved in HCl (10 mmol/L), adjusted to pH<2 using 6 M HCl, and dried under atmospheric air. The amino acids were extracted into an organic phase of ethanol and benzene and dried again under atmospheric air before derivatization with N-methyl-N-(tert-butyldimethylsilyl)trifluoracetamide (MTBSTFA) in the presence of 1% tert-butyldimethylchlorosilane (t-BDMS-Cl; both from Regis Technologies, Morton Grove, IL, USA). Samples were analyzed using a GC (6890N, Agilent, Santa Clara, CA, USA) linked to a MS (5975B, Agilent) with an electron ionization source. The percentwise distribution of mass isotopomers for the derivatized amino and keto acids was determined. M represents the percentage of the specified metabolite without ¹³C labeling and M+1 represents the percentage of the metabolite with ¹³C labeling in one carbon atom in the molecule. Glutamate M+1 in excess of 1.1% natural abundance in FCX and PTCX was calculated from NMRS data by summarizing the percent enrichment for all mono-labeled glutamate isotopomers.

High-Performance Liquid Chromatography

To determine the total amounts of amino acids, samples were analyzed using HPLC (1100 HPLC system, Agilent) with fluorescence detection after derivatization with o-phthaldialdehyde. The components were separated using a ZORBAX SB-C18 column (4.6 × 250 mm, 5 μm, Agilent) with a gradient of two eluents that were used to obtain optimal separation. One eluent consisted of phosphate buffer (50 mmol/L, pH=5.9) and tetrahydrofuran (2.5%), and the other consisted of methanol (98.75%) and tetrahydrofuran (1.25%). Relevant peaks in the HPLC chromatograms were identified and integrated. Amino acids were quantified by comparison with an external standard curve derived from standard solutions of amino acids. A standard solution was also run after every fourth sample as a control.

To determine the total amounts of monoamines (noradrenaline (NA), DA, 5-HT) and acid metabolites 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), and 5-hydroxyindoleacetic acid (5-HIAA), samples were analyzed by HPLC (1200 HPLC system, Agilent) with an electrochemical detector (Coulochem III, ESA, Chelmsford, MA, USA). Components were separated using an Eclips XDB-C18 column (4.6 × 150 mm, 5 μm, Agilent) with an aqueous mobile phase (0.90 mL/min) containing 90 mmol/L NaH₂PO₄, 50 mmol/L citric acid, 0.1 mmol/L EDTA, 0.5 mmol/L octanesulfonic acid, and 7% methanol solution. Relevant peaks were identified and integrated. Monoamine transmitter substances and acid metabolites were quantified by comparison with a standard curve. A standard solution was run after every fifth sample as a control. Net turnover of DA (DOPAC/DA and HVA/DA) and 5-HT (5-HIAA/5-HT) was calculated by dividing the amounts of the respective compounds.

¹³C and ¹H Nuclear Magnetic Resonance Spectroscopy

Lyophilized samples were dissolved in 200 μl of 99% D₂O containing 0.05% ethylene glycol as an internal standard, and pH was readjusted to values between 6.5 and 7.5. Samples were transferred into 5 mm Shigemi NMR microtubes (Shigemi, Allison Park, PA, USA). Proton-decoupled ¹³C NMR spectra were acquired using a BRUKER DRX-500 spectrometer (BRUKER Analytic GmbH, Rheinstetten, Germany). Spectra were recorded at 25°C with the following parameters: a 30° pulse angle and 30 kHz spectral width with 64K data points. The number of scans was typically 30,000. The acquisition time was 1.08 seconds and the relaxation delay was 0.5 seconds.

¹H NMR spectra were acquired using the same spectrometer, with the following parameters: a 90° pulse angle and a spectral width with 32K data points, number of scans was 128. The acquisition time was 1.36 seconds and relaxation delay was 10 seconds. Water suppression was achieved by applying a low-power pre-saturation pulse at the water frequency.

Relevant peaks in the spectra were identified and integrated using XWINNMR software (Bruker BioSpin), and quantified using ethylene glycol as an internal standard. The ¹³C labeling of metabolites was quantified from the integrals of the peak areas. Correction for natural abundance, nuclear Overhauser, and relaxation effects were applied to all relevant resonances. N-acetylaspartate (NAA) was quantified from the naturally abundant [2-¹³C]NAA (Choi and Gruetter, 2004). Percentage of ¹³C enrichment in glucose at the C-1 position was calculated from the ¹H spectra. The percentage difference between the ¹H on the ¹³C and the sum of the ¹H on ¹³C plus ¹²C in the C-1 position was used.

Labeling Patterns from [1-¹³C]Glucose

Interpretation of NMRS and GC—MS results is based on the knowledge of pathways for glucose metabolism, the differential distribution of enzymes between cell types, and the ¹³C labeling patterns (for labeling schemes see Eyjolfsson et al, 2006 and Kondziella et al, 2006) derived from different metabolic fates of ¹³C-labeled glucose. Glutamine synthetase and pyruvate carboxylase are located in astrocytes (Cesar and Hamprecht, 1995), whereas glutamic acid decarboxylase is found mainly in GABAergic neurons. Through glycolysis, [1-¹³C]glucose can be converted to [3-¹³C]pyruvate, which can further be converted to [3-¹³C]alanine or [3-¹³C]lactate. [3-¹³C]Pyruvate can also enter the TCA cycle as [2-¹³C]acetyl-CoA, which can condense with oxaloacetate, and after several steps result in [4-¹³C]α-ketoglutarate. This TCA cycle intermediate can be converted to [4-¹³C]glutamate, [4-¹³C]glutamine, or [2-¹³C]GABA, or stay in the cycle and give rise to [2-¹³C] or [3-¹³C]aspartate and in the next turn to [2-¹³C] or [3-¹³C]glutamate and glutamine and to [4-¹³C] or [3-¹³C]GABA. If pyruvate carboxylase is active, [2-¹³C]glutamate and glutamine and [4-¹³C]GABA can be formed. For more details on labeling patterns see Eyjolfsson et al (2006) and Kondziella et al (2006). Using GC—MS analysis, it is only possible to detect the number of ¹³C-labeled carbon atoms in different metabolites, but not their specific positions, as is possible with ¹³C NMRS.

When the term ‘labeling’ is used in the following sections, it refers to the level of M+1 for the respective metabolite, except when a positional isotopomer, for example, [2-¹³C]GABA, is mentioned.

Statistics

All data from HPLC, NMRS, and GC—MS analyses were tested for statistically significant differences between the control group and the MK-801 group. As the data from each group were independent, the two-tailed Student's t-test was used and P<0.05 was considered significant. Trimmed values were used, that is, two rats were excluded in each group, the ones with the highest and the lowest values (Keselman et al, 2002). Data are represented as mean±s.e.m.

Results

Each injection of MK-801 led to a distinct set of behavioral alterations such as head waving, hyperlocomotion, ataxia, flat body posture, and difficulties with balance and coordination of the hind limbs.

¹³C Labeling of Metabolites

As can be seen in Table 1, metabolism of glycolysis-related metabolites was affected. Lactate ¹³C enrichment was reduced in all brain regions. Glucose enrichment was calculated for FCX and PTCX from the NMR spectra. ¹³C enrichment of glucose in PTCX was reduced by ∼45% compared with control animals. Furthermore, ¹³C enrichment in alanine was reduced in PTCX only.

Table 1

Percent metabolite with ¹³C labeling in one carbon atom measured by GC—MS for lactate and alanine, and percent enrichment with [1-¹³C]glucose from ¹H NMRS for glucose in brain extracts from controls and animals treated with MK-801 for 6 days

	Glucose ^a		Lactate		Alanine
	Control	MK-801	Control	MK-801	Control	MK-801
FCX	22.3±1.2	18.0±2.4	10.0±0.7	7.1±0.6^b	11.4±0.4	10.3±0.5
PTCX	35.1±5.7	20.1±1.7^b	15.9±1.5	8.0±0.6^b	20.8±2.9	12.1±1.2^b
Hippocampus	ND		10.2±0.8	7.7±0.5^b	12.7±0.7	12.7±1.5
Thalamus	ND		9.3±0.7	6.8±0.7^b	13.3±0.7	11.0±0.7
Striatum	ND		9.4±0.8	7.0±0.4^b	16.2±1.2	15.5±1.2
NAc	ND		5.0±0.6	3.0±0.4^b	14.1±1.0	14.0±1.6

FCX, frontal cortex; GC—MS, gas chromatography—mass spectrometry; i.p., intraperitoneal; NAc, nucleus accumbens; ND, not determined; NMRS, nuclear magnetic resonance spectroscopy; PTCX, parietal and temporal cortex.

Data for glucose were from ¹H NMRS analysis and thus only for FCX and PTCX.

Significantly different from control group, P<0.05.

Data are presented as mean±s.e.m. and were analyzed with the two-tailed Student's t-test.

Metabolism of TCA cycle-related metabolites (Table 2) was affected to varying degrees in the brain regions analyzed. Labeling of fumarate was reduced in FCX and NAc, but was not detectable in PTCX. ¹³C labeling of malate was reduced in FCX, PTCX, striatum, and NAc in the MK-801 group.

Table 2

Percent metabolite with ¹³C labeling in one carbon atom (M+1) in TCA cycle-related metabolites measured by GC—MS in brain extracts from controls (n=9) and animals treated with MK-801 (n=9) for 6 days

	Fumarate		Malate
	Control	MK-801	Control	MK-801
FCX	9.2±0.8	6.3±0.7^a	10.3±0.8	7.7±0.6^a
PTCX	ND		11.5±0.7	8.5±0.5^a
Hippocampus	8.0±0.8	6.0±0.7	9.4±0.7	7.5±0.7
Thalamus	7.8±0.6	5.5±0.5^a	8.9±0.7	6.9±0.6
Striatum	9.1±0.8	7.1±0.7	9.9±0.8	7.6±0.6^a
NAc	8.3±0.7	6.5±0.7	9.2±0.7	6.6±0.6^a

FCX, frontal cortex; GC—MS, gas chromatography—mass spectrometry; i.p., intraperitoneal; NAc, nucleus accumbens; ND, not determined; PTCX, parietal and temporal cortex; TCA, tricarboxylic acid.

Significantly different from control group, P<0.05.

Fumarate was not detectable in PTCX (ND).

Data are presented as mean±s.e.m. and were analyzed with the two-tailed Student's t-test.

In Figures 1, results for glutamate, glutamine, GABA, and aspartate are presented as percentage of label (M+1). The percentage of metabolites with M+2 was negligible. In FCX, ¹³C labeling was reduced by ∼50% in glutamate and by ∼40% in GABA (Figure 1A). Metabolites in PTCX showed similar changes, with a reduction of ∼50% in the labeling in glutamate and ∼20% in GABA (Figure 1B). In addition, the percentage of [4-¹³C]glutamate and [2-¹³C]GABA of the total amount of these metabolites was calculated from NMRS and HPLC data for FCX and PTCX (Figures 2A and 2B). A reduction of ∼40% was found in [4-¹³C]glutamate. Unfortunately, ¹³C NMRS of hippocampus, thalamus, striatum, and NAc did not give sufficient signal for quantification of [4-¹³C]glutamate and [2-¹³C]GABA. All results from these regions were, thus, obtained by GC—MS. It can be seen in Figures 1C-1F that a similar reduction in ¹³C labeling of glutamate (M+1) occurred in hippocampus, striatum, and NAc, but not in thalamus. Labeling in GABA was reduced in hippocampus, striatum, thalamus, and NAc. Labeling of glutamine and aspartate remained unchanged in all brain regions analyzed.

Figure 1

Percent enrichment of ¹³C in M+1 isotopomers of metabolites derived from [1-¹³C]glucose in (A) frontal cortex (FCX), (B) parietal and temporal cortex (PTCX), (C) hippocampus, (D) thalamus, (E) striatum, and (F) nucleus accumbens (NAc). PTCX control n=8, MK-801 n=8, all other areas control n=9, and MK-801 n=9. ^# M+1 for glutamate was calculated from NMRS data by summarizing all mono-labeled glutamate. M+1 for glutamine, GABA, and aspartate was obtained using GC—MS. All animals were injected with [1-¹³C]glucose 15 minutes before microwave fixation. Gray bars represent controls and white bars represent MK-801-treated animals. Data are presented as mean±s.e.m and were analyzed with the two-tailed Student's t-test. ∗Significantly different from control group, P<0.05.

Figure 2

Percent enrichment of ¹³C-labeled metabolites derived from [1-¹³C]glucose in (A) frontal cortex (FCX; control n=8, MK-801 n=7) and (B) parietal and temporal cortex (PTCX; control n=8, MK-801 n=7). This figure illustrates percent enrichment of [4-¹³C]glutamate and [2-¹³C]GABA calculated from data obtained with ¹³C NMRS. [4-¹³C]Glutamate and [2-¹³C]GABA are labeled substances derived from the first turn in the TCA cycle. All animals were injected with [1-¹³C]glucose 15 minutes before microwave fixation. Gray bars represent controls and white bars represent MK-801-treated animals. Data are presented as mean±s.e.m. and were analyzed with the two-tailed Student's t-test. ∗Significantly different from control group, P<0.05.

The ratio of ¹³C label from pyruvate carboxylation over pyruvate dehydrogenation (¹³C label in singlets (C2-C3)/C4) was calculated for glutamate, glutamine, and GABA. Only glutamate labeling in FCX was affected by MK-801 injection (controls 0.08±0.2 versus MK-801 0.12±0.05).

Amounts of Metabolites

The concentration of lactate was reduced in FCX in animals treated with MK-801 compared with controls (controls 1.14±0.24 μmol/g versus MK-801 0.72±0.28 μmol/g; P<0.01). Moreover, a reduction in the amount of NAA was found in this area (controls 45.5±10.2 nmol/g versus MK-801 32.3±6.3 nmol/g; P<0.01). However, no differences in the total amounts of glutamate, glutamine, GABA, aspartate, or alanine were detected in any of the examined brain regions (results not shown).

The amounts of DA, its metabolite DOPAC, and HVA, and the 5-HT metabolite 5-HIAA are shown in Figure 3. DOPAC and DA were increased in the MK-801 group in FCX only. Dopamine concentration was increased by ∼40% (474.2±178.8 μmol/g versus 680.6±117.9 μmol/g; P<0.01), whereas DOPAC was increased by ∼100% (69.5±14.6 μmol/g versus 137.0±35.9 μmol/g; P<0.0001) in FCX of animals treated with MK-801 compared with controls. An increase in HVA by ∼50% (33.5±10.7 μmol/g versus 50.3±9.4 μmol/g; P<0.02) compared with controls was found in the PTCX. In contrast, HVA was decreased by ∼21% (594.6±83.2 μmol/g versus 472.1±80.6 μmol/g; P<0.01) in the striatum. The 5-HIAA was also decreased in PTCX (140.2±3.1 versus 161.9±5.2 μmol/g; P<0.011). No differences were observed in noradrenalin and 5-HT, and results are not presented. Ratios between some of these metabolites were also calculated: DOPAC/DA and 5-HIAA/5-HT were unchanged in all examined areas, whereas HVA/DA was decreased in FCX and thalamus of MK-801-treated animals (Table 3).

Table 3

Turnover rates for dopamine (DA) and serotonin (5-HT) in brain extracts of controls (n=9) and animals treated with MK-801 (n=9) for 6 days

	DOPAC/DA		HVA/DA		5-HIAA/5-HT
	Control	MK-801	Control	MK-801	Control	MK-801
FCX	0.16±0.10	0.21±0.20	0.28±0.03	0.19±0.02^a	ND
PTCX	0.2±0.0	0.2±0.0	0.16±0.04	0.18±0.0	0.53±0.06	0.55±0.04
Hippocampus	ND		ND		0.76±0.04	0.72±0.02
Thalamus	0.11±0.02	0.14±0.04	0.1±0.01	0.09±0.01^a	ND
Striatum	0.70±0.00	0.80±0.00	0.10±0.01	0.10±0.01	ND
NAc	0.40±0.08	0.51±0.06	0.17±0.03	0.15±0.01	1.10±0.02	1.10±0.02

5-HIAA, 5-hydroxyindoleacetic acid; DOPAC, 3,4-dihydroxyphenylacetic acid; FCX, frontal cortex; HPLC, high-performance liquid chromatography; HVA, homovanillic acid; i.p., intraperitoneal; NAc, nucleus accumbens; ND, not determined; PTCX, parietal and temporal cortex.

Significantly different from control group, P<0.05.

Animals were given i.p. injections of saline (nine animals) or MK-801 (nine animals; 0.5 mg/kg body weight) daily for 6 days. The final dose of MK-801 or saline was given 30 minutes before i.p. injection of [1-¹³C]glucose (543 mg/kg, 0.3 M solution). For details see ‘Materials and methods’ section. Turnover rates were calculated from HPLC results, for DA as DOPAC/DA and/or HVA/DA, and for serotonin as 5-HIAA/5-HT. For details see ‘Materials and methods’ section.

Data are presented as mean±s.e.m. and were analyzed with the two-tailed Student's t-test.

Discussion

It has been shown that MK-801 reaches maximal concentrations in plasma and brain within 10 to 30 minutes after injection, with an elimination half-life of 1.9 and 2.05 hours, respectively (Vezzani et al, 1989). The distribution of MK-801 in cortex, hippocampus, striatum, brainstem, and cerebellum 30 and 180 minutes after injection showed no preferential concentration or retention (Vezzani et al, 1989). MK-801 binds to NMDA receptors in the areas investigated (Wong et al, 1988). However, it has been demonstrated that binding is unevenly distributed among brain regions, with binding to hippocampus membranes greater than cortex, which is greater than striatum, which is still greater than medulla-pons (Wong et al, 1988).

Figure 3

Amounts of monoamines (mmol/g) in the analyzed brain regions of controls (n=9) and animals treated with MK-801 (n=9) for 6 days. (A) Dopamine, (B) 3,4-dihydroxyphenylacetic acid (DOPAC), (C) homovanillic acid (HVA), (D) 5-hydroxyindoleacetic acid (5-HIAA). For details see ‘Material and methods’ section. Amounts of metabolites were measured with HPLC. FCX, frontal cortex; PTCX, parietal and temporal cortex; NAc, nucleus accumbens. Gray bars represent controls and white bars represent MK-801-treated animals. Data are presented as mean±s.e.m. and were analyzed with the two-tailed Student's t-test. ∗Significantly different from control group, P<0.05.

This study shows that 45 minutes after the sixth daily dose of MK-801, both glycolysis and mitochondrial function were slightly affected throughout the cortico—striato—thalamo—cortical loop (Figure 4). Apart from lactate and NAA, both of which were decreased in the FCX, no changes were detected in metabolite levels. This is in contrast to an earlier study that analyzed metabolite levels 15 minutes after a series of six daily injections (as carried out in this study) in rats and showed increased levels of glutamate (Kondziella et al, 2006). Thus, it appears that this increase in glutamate is transient, as it was not observed 45 minutes after MK-801 injection. It should be noted that first-episode schizophrenia patients have been reported to have an increased level of glutamate, glutamine, or both (glx; when peaks in ¹H NMR spectra could not be separated), whereas chronic patients show a decrease in these metabolites (Bustillo et al, 2009; Lutkenhoff et al, 2010; Theberge et al, 2003). Even though effects of MK-801 were very uniform throughout the cortico—striato—thalamo—cortical loop, there were important differences that can shed light on the malfunctioning of the cortico—striato—thalamo—cortical loop in schizophrenia.

Figure 4

Schematic illustration of the metabolic changes induced by MK-801 in central areas of the cortico—striato—thalamo—cortical loop. The short black arrows indicate the direction of change in rats injected with MK-801, showing either an increase or a decrease. The thin black arrows illustrate the main direction of the loop. The thick gray arrows illustrate the sensory input to cortex. The extent of sensory information from thalamus to the cortical areas should normally be reduced compared with the incoming sensory input to the thalamus, because of the thalamic filter. However, the results in this study indicate that MK-801, through a series of actions, might disturb this filter function, which may mean that sensory information is not sufficiently filtered in the thalamus and continues to cortical areas without any substantial decrease in information content. This may in turn lead to sensory overload of the cortex and psychosis or confusion. 5-HIA, 5-hydroxyindoleacetic acid; DA, dopamine; DOPAC, 3,4-dihydroxyphenylacetic acid; Glu, glutamate; HVA, homovanillic acid; NAA, N-acetylaspartate; TCA, tricarboxylic acid cycle.

Cytosolic Glucose Metabolism (Glycolysis)

Reduced glycolysis, as evidenced by reduced FDG consumption (Kurumaji and McCulloch, 1989), could be the cause for the reduction in ¹³C enrichment in glucose, lactate, and alanine seen in this study in the PTCX. The only area in which a decrease in lactate amount was observed was FCX. Thus, the cortex showed the most severe impairment of glycolysis. All other areas exhibited only slightly reduced glycolysis, as only ¹³C labeling of lactate was decreased. An effect of MK-801 on glycolysis was also reported by Schroeder et al (1994) in some of these areas. Holmes et al (2006) found a significant decrease in cerebrospinal fluid lactate levels in drug-naive patients with first-episode schizophrenia. However, other findings are in disagreement on this point. On the basis of reports in the literature, it seems that lactate levels fluctuate with disease duration. After one MK-801 injection, both Loubinoux et al (1994) and Brenner et al (2005) found increased amounts of lactate in some of the brain regions investigated. Beasley et al (2009) found decreased levels of alanine and lactate in white matter of patients with established schizophrenia and interpreted this as an alteration in astrocyte—neuron metabolic coupling. The reduced amount of lactate in FCX in this study can be interpreted in the same way. Using both single and repeated injection models of MK-801 in rats, we have demonstrated that the metabolic disturbance depends on exposure time to MK-801, which is considered to translate to the number of psychotic episodes in schizophrenia (Brenner et al, 2005; Kondziella et al, 2006). The present findings are in line with studies in humans (Beasley et al, 2009; Holmes et al, 2006) and further strengthen this hypothesis as ¹³C enrichment in lactate was reduced in all brain regions studied.

Mitochondrial Metabolism of Glucose Metabolites

N-acetylaspartate, synthesized in mitochondria, has been suggested to reflect neuronal function and integrity (Baslow and Guilfoyle, 2006). A reduced amount of NAA was found in FCX. This is in agreement with several studies that have reported reduced levels of NAA in the FCX and temporal lobe in patients with schizophrenia (Bertolino et al, 1996; Steen et al, 2005; Tanaka et al, 2006). This reduction is not necessarily related to cell loss, but can be an indicator of mitochondrial dysfunction. Labeling of the TCA cycle intermediates, malate and fumarate, was also reduced in the FCX of the MK-801-treated animals, indicating impairments of mitochondrial enzymes, as also reported in humans (Martins-de-Souza et al, 2009) and in the MK-801 rat model using proteomics (Schmitt et al, 2004).

Using ¹³C NMRS and/or MS it is possible to detect mitochondrial dysfunction at the metabolite level. Two of these metabolites are the most predominant neurotransmitters in the brain (glutamate and GABA) that through α-ketoglutarate are coupled to mitochondrial function.

Glutamate and γ-Aminobutyric Acid Turnover, and the Cortico—Striato—Thalamo—Cortical Loop

Significant reductions in glutamatergic metabolism (∼50%) were found in FCX and PTXC of MK-801-treated animals. Decreases in the other areas were much smaller, and no difference was found in thalamus. In FCX, the ¹³C label in glutamate derived from astrocytes, via pyruvate carboxylation, was less reduced than the neuronal glutamate labeling. Glutamate can be converted to GABA by glutamic acid decarboxylase in GABAergic neurons. Both glutamate and GABA de novo synthesis are dependent on glutamine delivery from astrocytes (Waagepetersen et al, 2003). There was a tendency toward decreased labeling in glutamine in all areas except PTXC. This could indicate decreased glutamatergic neurotransmission, as ∼40% of [4-¹³C]glutamine is derived from [4-¹³C]glutamate labeled in the neuronal compartment (Hassel et al, 1995). Statistical significance was, however, not reached due to the relatively large standard deviation.

Hippocampus has the greatest binding affinity for MK-801 (Wong et al, 1988). Glutamate-containing subicular hippocampal axons innervate cortical pyramidal neurons and interneurons, in which AMPA (α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate) and NMDA receptors are strongly involved in synaptic transmission (Kiss et al, 2010). In this study, glutamate and GABA labeling were decreased in the hippocampus, possibly leading to decreased input into FCX. GABA labeling was most dramatically decreased in FCX and this could well be the reason for the psychosis-like behavior seen in the rats. GABAergic networks of interneurons and cortico—cortico connections are thought to be responsible for generation of β- and γ-band activity that has been shown to be abnormal in patients with schizophrenia (Uhlhaas and Singer, 2010). Changes in cortical GABA concentration and GABA receptors have been reported in patients with schizophrenia (Duncan et al, 2010; Yoon et al, 2010). The disruption in β- and γ-band activity, together with the decreased glutamate labeling, could affect interaction between cortex and striatum and might lead to decreased excitation of striatal neurons (Figure 4). In striatum and NAc, GABA labeling was more affected than that of glutamate. These areas communicate with the thalamus through GABAergic fibers. It can be postulated that the decrease in GABA turnover in striatum and NAc might have a disinhibiting effect on glutamatergic neurons in the thalamus that project to the cerebral cortex (Figure 4). It has been suggested that hypofunction of the corticostriatal glutamate pathway disrupts the postulated thalamic filter, in which inhibition of glutamatergic neurons in the thalamus normally filters sensory input to protect the cortex from sensory overload (Carlsson et al, 2004). Reduced function of the thalamic filter is suggested to lead to confusion or psychosis. However, at this stage, the model we propose is only tentative.

Monoamine Turnover

Laruelle et al (2003) showed that dopaminergic innervations of the dorsolateral FCX appear to be decreased in schizophrenia. In this study, dopaminergic turnover was decreased in the thalamus. In spite of decreased dopamine transmission, there is little direct evidence for low dopamine levels in the FCX of humans with schizophrenia (Howes and Kapur, 2009). In fact, an increase was observed in the levels of dopamine and its degradation product DOPAC in the FCX in this study. This is in agreement with results from Loscher et al (1991), as both DA and DOPAC were increased. The most likely explanation for this finding is increased release of dopamine and not reduced turnover. This can be reconciled with decreased dopaminergic activity because of the fact that there is an inverse U-shaped curve between dopaminergic activity in FCX and the DA concentration (Vijayraghavan et al, 2007). In PTCX, the HVA/DA ratio was significantly reduced, as also reported by Jentsch and Roth (1999). This was due to the significant increase in HVA level, corroborating results from Loscher et al (1991). In this study, we detected a decrease in HVA in the striatum. As there was a general decrease in amounts of metabolites, it is likely that DA turnover was decreased in this region. This finding is not in agreement with Loscher et al (1991), but might be due to differences in experimental setups reflecting metabolic changes caused by single versus repeated injections of MK-801. It must be kept in mind that all changes in dopaminergic transmission are necessarily secondary to NMDA receptor hypofunction in the schizophrenia model of repeated MK-801 exposure used in this study. The only observed indication of disturbed 5-HT metabolism was a slight increase in the 5-HIAA level in PTCX, which is in line with an earlier report by Loscher et al (1991).

Conclusion

We have shown that perturbations of NMDA receptor function in the model of repeated injections of MK-801 cause pronounced metabolic changes in the areas of the cortico—striato—thalamo—cortical loop, and include the dopaminergic, as well as the glutamatergic and GABAergic systems.

Footnotes

Acknowledgements

The technical assistance of Lars Evje is gratefully acknowledged.

The authors declare no conflict of interest.

References

Baslow

Guilfoyle

(2006) Functions of N-acetylaspartate and N-acetylaspartyl-glutamate in brain—Evidence of a role in maintenance of higher brain integrative activities of information processing and cognition. Adv Exp Med Biol 576:95–112.

Beasley

Dwork

Rosoklija

Mann

Mancevski

Jakovski

Davceva

Tait

Straus

Honer

(2009) Metabolic abnormalities in fronto-striatal-thalamic white matter tracts in schizophrenia. Schizophr Res 109:159–66.

Bertolino

Nawroz

Mattay

Barnett

Duyn

Moonen

Frank

Tedeschi

Weinberger

(1996) Regionally specific pattern of neurochemical pathology in schizophrenia as assessed by multislice proton magnetic resonance spectroscopic imaging. Am J Psychiatry 153:1554–63.

Brenner

Kondziella

Haberg

Sonnewald

(2005) Impaired glutamine metabolism in NMDA receptor hypofunction induced by MK801. J Neurochem 94:1594–603.

Bustillo

Rowland

Mullins

Jung

Chen

Qualls

Hammond

Brooks

Lauriello

(2009) 1H-MRS at 4 Tesla in minimally treated early schizophrenia. Mol Psychiatry 15:629–36.

Carlsson

Nilsson

(2004) Schizophrenia: from dopamine to glutamate and back. Curr Med Chem 11:267–77.

Cesar

Hamprecht

(1995) Immunocytochemical examination of neural rat and mouse primary cultures using monoclonal antibodies raised against pyruvate carboxylase. J Neurochem 64:2312–8.

Choi

Gruetter

(2004) Dynamic or inert metabolism? Turnover of N-acetyl aspartate and glutathione from D-[1–13C]glucose in the rat brain in vivo . J Neurochem 91:778–87.

Del Arco

Mora

(2008) Prefrontal cortex-nucleus accumbens interaction: in vivo modulation by dopamine and glutamate in the prefrontal cortex. Pharmacol Biochem Behav 90:226–35.

10.

Duncan

Webster

Rothmond

Bahn

Elashoff

Shannon Weickert

(2010) Prefrontal GABA(A) receptor alpha-subunit expression in normal postnatal human development and schizophrenia. J Psychiatr Res 44:673–81.

11.

Eyjolfsson

Brenner

Kondziella

Sonnewald

(2006) Repeated injection of MK801: an animal model of schizophrenia? Neurochem Int 48:541–6.

12.

Hassel

Sonnewald

Fonnum

(1995) Glial-neuronal interactions as studied by cerebral metabolism of [2–13C]acetate and [1–13C]glucose: an ex vivo 13C NMR spectroscopic study. J Neurochem 64:2773–82.

13.

Holmes

Tsang

Huang

Leweke

Koethe

Gerth

Nolden

Gross

Schreiber

Nicholson

Bahn

(2006) Metabolic profiling of CSF: evidence that early intervention may impact on disease progression and outcome in schizophrenia. PLoS Med 3:e327.

14.

Howes

Kapur

(2009) The dopamine hypothesis of schizophrenia: version III—the final common pathway. Schizophr Bull 35:549–62.

15.

Jentsch

Roth

(1999) The neuropsychopharmacology of phencyclidine: from NMDA receptor hypofunction to the dopamine hypothesis of schizophrenia. Neuropsychopharmacol 20:201–25.

16.

Kegeles

Abi-Dargham

Zea-Ponce

Rodenhiser-Hill

Mann

Van Heertum

Cooper

Carlsson

Laruelle

(2000) Modulation of amphetamine-induced striatal dopamine release by ketamine in humans: implications for schizophrenia. Biol Psychiatry 48:627–40.

17.

Keselman

Cribbie

Holland

(2002) Controlling the rate of Type I error over a large set of statistical tests. Br J Math Stat Psychol 55:27–39.

18.

Kiss

Hoffmann

Hajos

(2010) Delta oscillation and short-term plasticity in the rat medial prefrontal cortex: modelling NMDA hypofunction of schizophrenia. Int J Neuropsychopharmacol 25:1–14.

19.

Kondziella

Brenner

Eyjolfsson

Markinhuhta

Carlsson

Sonnewald

(2006) Glial-neuronal interactions are impaired in the schizophrenia model of repeated MK801 exposure. Neuropsychopharmacol 31:1880–7.

20.

Kondziella

Brenner

Eyjolfsson

Sonnewald

(2007) How do glial-neuronal interactions fit into current neurotransmitter hypotheses of schizophrenia? Neurochem Int 50:291–301.

21.

Kurumaji

McCulloch

(1989) Effects of MK-801 upon local cerebral glucose utilisation in conscious rats and in rats anaesthetised with halothane. J Cereb Blood Flow Metab 9:786–94.

22.

Lahti

Koffel

LaPorte

Tamminga

(1995) Subanesthetic doses of ketamine stimulate psychosis in schizophrenia. Neuropsychopharmacol 13:9–19.

23.

Laruelle

Kegeles

Abi-Dargham

(2003) Glutamate, dopamine, and schizophrenia: from pathophysiology to treatment. Ann N Y Acad Sci 1003:138–58.

24.

Loscher

Annies

Honack

(1991) The N-methyl-D-aspartate receptor antagonist Mk-801 induces increases in dopamine and serotonin metabolism in several brain-regions of rats. Neurosci Lett 128:191–4.

25.

Loubinoux

Meric

Borredon

Correze

Gillet

Beloeil

Tiffon

Mispelter

Lhoste

Jacques

(1994) Cerebral metabolic changes induced by MK-801: a 1D (phosphorus and proton) and 2D (proton) in vivo NMR spectroscopy study. Brain Res 643:115–24.

26.

Lutkenhoff

Van Erp

Thomas

Therman

Manninen

Huttunen

Kaprio

Lonnqvist

O'Neill

Cannon

(2010) Proton MRS in twin pairs discordant for schizophrenia. Mol Psychiatry 15:308–18.

27.

Martins-De-Souza

Gattaz

Schmitt

Novello

Marangoni

Turck

Dias-Neto

(2009) Proteome analysis of schizophrenia patients Wernicke's area reveals an energy metabolism dysregulation. BMC Psychiatry 9:17.

28.

Olney

Farber

(1995) NMDA antagonists as neurotherapeutic drugs, psychotogens, neurotoxins, and research tools for studying schizophrenia. Neuropsychopharmacol 13:335–45.

29.

Rujescu

Bender

Keck

Hartmann

Ohl

Raeder

Giegling

Genius

McCarley

Moller

Grunze

(2006) A pharmacological model for psychosis based on N-methyl-D-aspartate receptor hypofunction: molecular, cellular, functional and Behavioral abnormalities. Biol Psychiatry 59:721–9.

30.

Rung

Carlsson

Markinhuhta

Carlsson

(2005) (+)-MK-801 induced social withdrawal in rats; a model for negative symptoms of schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry 29:827–32.

31.

Schmitt

Wilczek

Blennow

Maras

Jatzko

Petroianu

Braus

Gattaz

(2004) Altered thalamic membrane phospholipids in schizophrenia: a postmortem study. Biol Psychiatry 56:41–5.

32.

Schroeder

Buchsbaum

Siegel

Geider

Haier

Lohr

Potkin

(1994) Patterns of cortical activity in schizophrenia. Psychol Med 24:947–55.

33.

Steen

Hamer

Lieberman

(2005) Measurement of brain metabolites by H-1 magnetic resonance spectroscopy in patients with review and meta-analysis. Neuropsychopharmacol 30:1949–62.

34.

Tanaka

Obata

Sassa

Yoshitome

Asai

Ikehira

Suhara

Okubo

Nishikawa

(2006) Quantitative magnetic resonance spectroscopy of schizophrenia: relationship between decreased N-acetylaspartate and frontal lobe dysfunction. Psychiatry Clin Neurosci 60:365–72.

35.

Theberge

Al-Semaan

Williamson

Menon

Neufeld

Rajakumar

Schaefer

Densmore

Drost

(2003) Glutamate and glutamine in the anterior cingulate and thalamus of medicated patients with chronic schizophrenia and healthy comparison subjects measured with 4.0-T proton MRS. Am J Psychiatry 160:2231–3.

36.

Uhlhaas

Singer

(2010) Abnormal neural oscillations and synchrony in schizophrenia. Nat Rev Neurosci 11:100–13.

37.

Vezzani

Serafini

Stasi

Caccia

Conti

Tridico

Samanin

(1989) Kinetics of MK-801 and its effect on quinolinic acid-induced seizures and neurotoxicity in rats. J Pharmacol Exp Ther 249:278–83.

38.

Vijayraghavan

Wang

Birnbaum

Williams

Arnsten

(2007) Inverted-U dopamine D1 receptor actions on prefrontal neurons engaged in working memory. Nat Neurosci 10:376–84.

39.

Wong

Knight

Woodruff

(1988) [3H]MK-801 labels a site on the N-methyl-D-aspartate receptor channel complex in rat brain membranes. J Neurochem 50:274–81.

40.

Waagepetersen

Sonnewald

Schousboe

(2003) Compartmentation of glutamine, glutamate, and GABA metabolism in neurons and astrocytes: functional implications. Neuroscientist 9:398–403.

41.

Yoon

Maddock

Rokem

Silver

Minzenberg

Ragland

Carter

(2010) GABA concentration is reduced in visual cortex in schizophrenia and correlates with orientation-specific surround suppression. J Neurosci 30:3777–81.

Altered 13 C Glucose Metabolism in the Cortico—Striato—Thalamo—Cortical Loop in the MK-801 Rat Model of Schizophrenia

Abstract

Keywords

Introduction

Materials and methods

Gas Chromatography—Mass Spectrometry

High-Performance Liquid Chromatography

13C and 1H Nuclear Magnetic Resonance Spectroscopy

Labeling Patterns from [1-13C]Glucose

Statistics

Results

13C Labeling of Metabolites

Amounts of Metabolites

Discussion

Cytosolic Glucose Metabolism (Glycolysis)

Mitochondrial Metabolism of Glucose Metabolites

Glutamate and γ-Aminobutyric Acid Turnover, and the Cortico—Striato—Thalamo—Cortical Loop

Monoamine Turnover

Conclusion

Footnotes

Acknowledgements

References

¹³C and ¹H Nuclear Magnetic Resonance Spectroscopy

Labeling Patterns from [1-¹³C]Glucose

¹³C Labeling of Metabolites