Hypermetabolic State in the 7-Month-Old Triple Transgenic Mouse Model of Alzheimer'S Disease and the Effect of Lipoic Acid: A 13 C-NMR Study

Abstract

Alzheimer's disease (AD) is characterized by age-dependent biochemical, metabolic, and physiologic changes. These age-dependent changes ultimately converge to impair cognitive functions. This study was carried out to examine the metabolic changes by probing glucose and tricarboxylic acid cycle metabolism in a 7-month-old triple transgenic mouse model of AD (3×Tg-AD). The effect of lipoic acid, an insulin-mimetic agent, was also investigated to examine its ability in modulating age-dependent metabolic changes. Seven-month-old 3×Tg-AD mice were given intravenous infusion of [1-¹³C]glucose followed by an ex vivo ¹³C nuclear magnetic resonance to determine the concentrations of ¹³C-labeled isotopomers of glutamate, glutamine, aspartate, gamma aminobutyric acid, and N-acetylaspartate. An intravenous infusion of [1-¹³C]glucose+[1,2-¹³C]acetate was given for different periods of time to distinguish neuronal and astrocytic metabolism. Enrichments of glutamate, glutamine, and aspartate were calculated after quantifying the total (¹²C+¹³C) concentrations by high-performance liquid chromatography. A hypermetabolic state was clearly evident in 7-month-old 3×Tg-AD mice in contrast to the hypometabolic state reported earlier in 13-month-old mice. Hypermetabolism was evidenced by prominent increase of ¹³C labeling and enrichment in the 3×Tg-AD mice. Lipoic acid feeding to the hypermetabolic 3×Tg-AD mice brought the metabolic parameters to the levels of nonTg mice.

Keywords

Alzheimer's disease glutamate isotopomers hypermetabolism lipoic acid

INTRODUCTION

Alzheimer's disease (AD) is a complicated neurodegenerative disease with several gaps in our understanding about its etiology and a marked absence (to date) of drug therapies targeting the underlying mechanisms of neurodegeneration. However, hallmarks of AD include the widespread presence of β-amyloid (Aβ) plaques and neurofibrillary tangles; their detection in postmortem tissue validates a definite diagnosis.¹ In addition, there is a clear association of AD with biochemical,² metabolic,³ and physiologic changes.⁴ These three factors also show age-dependent changes that ultimately converge to impair cognitive abilities. Research from several preclinical and clinical studies has established the presence of multiple metabolic changes in brain and disturbances in glucose metabolism.³ These disturbances are not limited to brain glucose uptake but also extend to glycolytic metabolism, tricarboxylic acid (TCA) cycle, and the subsequent formation of neurotransmitters and neurochemicals like Glu, gamma aminobutyric acid (GABA), Gln, N-acetylaspartate (NAA), and several others.^5,6 Ultimately, the impaired metabolic pathways may become incapable of supporting the highly intricate supply of energy and metabolites required for neuronal function, leading to a synaptic failure that impinges on cognitive abilities.

Several mouse models with one or multiple transgenes have been designed to further our understanding about AD and test potential therapeutics.⁷ Each mouse model highlights one or multiple facets of AD, i.e., metabolic abnormalities, cognitive impairment, Aβ plaques, hyperphosphorylated tau and/or tangles; thus, choosing the correct mouse model according to the hypothesis being addressed is critical. In that regard, the triple transgenic mouse model of AD (3×Tg-AD) shows agedependent appearance of pathology type (plaques and tangles), pathology location (restricted to the hippocampus, amygdala, and cerebral cortex), and synaptic impairments. In addition, it shows a sequence of development of one hallmark pathology (plaques) leading to the development of other signature lesions (tau pathology)—a feature missing in all the previous mouse models of AD.⁸ In an earlier study, we characterized the brain glucose utilization of 13-month-old 3×Tg-AD mice and the insulinmimetic effect of lipoic acid. The important finding of that study was the occurrence of a glucose hypometabolic state and an insulin-mimetic effect by lipoic acid leading to partial restoration of glycolytic metabolism.⁹ It is worth noting that 13-month-old 3×Tg-AD mice show extracellular Aβ deposits in cortex and human tau immunoreactivity.8 Because a hypometabolic state was predominant in the 13-month-old 3×Tg-AD mice, it was critical to examine its preceding metabolic status that leads to hypometabolism (at an earlier time point with less pathology). Thus, we selected the 7-month-old 3×Tg-AD mice that are characterized by ‘diffuse Aβ plaques in the neocortex but no human tau immunoreactivity’ as the appropriate age for further investigation of brain metabolism.

The current study was conducted to examine metabolic differences in the brain of 7-month-old 3×Tg-AD and nonTg mice and assess the therapeutic effects of insulin mimetic, lipoic acid.^>10-12 In addition to [1-¹³C]glucose infusion, the metabolic state was also probed with [1-¹³C]glucose+[1,2-¹³C]acetate coinfusion to assess neuronal and astrocytic metabolism. Glucose is metabolized by both neurons and astrocytes; however, its major oxidative metabolism occurs in neurons, whereas, acetate is exclusively metabolized by astrocytes; thus, neuronal and astrocytic metabolism can be studied simultaneously by a co-infusion of ¹³C-labeled glucose and acetate.¹³

MATERIALS AND METHODS

Materials

Deuterium oxide (99.9%) and [1,2-¹³C]acetate (99%) were obtained from Cambridge Isotope Laboratories (Andover, MA, USA). [1-¹³C]glucose (99%) was purchased from Sigma-Aldrich (St Louis, MO, USA) and all other chemicals were the purest grade available from Sigma-Aldrich. Rodent tail vein catheter and restraining apparatus were obtained from Braintree Scientific (Braintree, MO, USA). The constant infusion of [1-¹³C]glucose and [1,2-¹³C]acetate was carried out by using a pump from Bio-Rad Laboratories (Hercules, CA, USA).

Mice Colonies and Lipoic Acid Feeding

Colonies of 3×Tg-AD and nonTg mouse strain (C57BL6/129S; Gift from Dr Frank Laferla, University of California, Irvine) were bred and maintained at the University of Southern California (Los Angeles, CA) following National Institutes of Health guidelines on use of laboratory animals and an approved protocol by the University of Southern California Institutional Animal Care and Use Committee. The triple transgenic mouse model of AD (3×Tg-AD) was first developed by Oddo et al. ⁸ Mice were housed on 12- hour light/dark cycles and provided ad libitum access to food and water. Triple transgenic mouse model of AD and nonTg male mice were either fed with water containing 0.23% R-sodium lipoic acid (gift from GeroNova Research) or normal water for 4 weeks. Thus, at the time of glucose infusion, mice were approximately 7 months of age. Over a period of 4 weeks, the amount of water consumed by mice drinking lipoic acid containing water or regular water was similar with no statistically significant differences. The sodium salt of lipoic acid used in this study is quite stable in water and we changed the water weekly. No protection from light was required.

Intravenous Glucose and Acetate Infusion

The mouse to be infused was first restrained using a rotating tail vein injection restrainer. Anesthesia was not used during the entire procedure; this allowed us to measure metabolism in awake non-anesthetized mice. After restraining the mice, we tested the basal blood glucose levels as described.⁹ The puncture made for testing the basal blood glucose levels was also used for inserting a vein catheter in the mouse tail (Braintree Scientific). The catheter was inserted following the manufacturer's instructions and was ensured to be in the tail vein by pushing some saline through the catheter. Any bulge at the bottom of the tail, resistance, or back flow of saline was considered as an improper insertion of catheter and performed again at a more proximal point in the same vein or the next vein was used. The glucose and acetate infusion protocol was carried out as described earlier.¹⁴,¹⁵ Briefly, 1-¹³C]glucose (0.3 M)+[1,2-¹³C]acetate (0.6 M) solution was prepared. A bolus injection to increase the blood glucose levels to normoglycemic range was followed by exponentially decreasing the amount of glucose for 5 minutes. Finally, infusion at a constant rate was performed for different durations as specified (20, 60, and 150 minutes). Mice were kept in a quiet and warm environment to avoid too much stress during the glucose and acetate infusion. The labeling pattern for [1-¹³C]glucose and [1,2-¹³C]glucose has been well described earlier.¹⁴,¹⁶

Tissue Collection and Extraction

The infusion was stopped after the specified time and the catheter removed, followed by testing for final blood glucose levels as described below. The mouse brain was then immediately snap frozen using liquid nitrogen. It was ensured that the total time taken for snap freezing of the brain (from the end of infusion) was less than 1 minute for all mice to minimize the postmortem metabolic changes. Subsequently, weighing and perchloric acid extraction was carried out as described previously.^9,17 Briefly, frozen brain was powdered using a mortar and pestle (while keeping the brain cold by constantly adding liquid nitrogen) followed by addition of perchloric acid and subsequent centrifugations. A final centrifugation was carried out to remove precipitation and the final brain extract supernatant was stored in −80 °C freezer until used for nuclear magnetic resonance or high-performance liquid chromatography analysis.

Nuclear Magnetic Resonance

The stored brain extracts were thawed and mixed in appropriate proportion with D₂O, 1.5 μL 1, 4-dioxane (chemical shift reference and internal standard) and a preservative (sodium azide). ¹³C nuclear magnetic resonance analysis was carried out on a Varian VNMRS 600 MHz instrument at 150.86 MHz. ¹³C spectra were acquired with proton decoupling and nuclear Overhauser enhancement with the following parameters: pulse angle of 45°, acquisition time of 1 second and a relaxation delay of 5 seconds, 251 p.p.m. spectral width with 32,768 spectral points. A total of 7,312 scans were acquired at 25 °C. The chemical shift reference peak of 1, 4-dioxane was set to exactly 67.4 p.p.m. This was followed by peak identification using chemical shift values from previous literature.^17,18 The peak areas were normalized by using the peak area of 1, 4-dioxane as the internal standard. MestRenova software from Mestrelab Research (Escondido, CA, USA) was used to integrate relevant peaks after normalization of peak areas. The quantification of each peak was carried out as follows: ¹³C spectra of Glu, Gln, Asp, NAA, and GABA at natural abundance of ¹³C were acquired in a single solution at different concentrations to construct a standard curve of peak area versus ¹³C concentration for each carbon of the compound. This standard curve was used to convert the observed peak area of each ¹³C metabolite (after normalization using the peak area of 1,4-dioxane as internal standard) to ¹³C concentration. From this ¹³C concentration, the contribution of natural abundance ¹³C, determined from the concentration of total [¹²C+¹³C]metabolite measured by highperformance liquid chromatography and the natural abundance of 13C (1.1%) was subtracted to obtain the ¹³C concentration resulting from the infusion of ¹³C-labeled substrates.

13C-Labeling Patterns and Interpretation

Labeling of brain metabolites from [1-¹³C]glucose and [1,2-¹³C]acetate is well described in several earlier publications by Dr Sonnewald.19 Briefly, after the formation of [3-¹³C]pyruvate through glycolysis, it can be either decarboxylated to [2-¹³C]acetyl CoA, or transaminated to [3-¹³C]alanine, or reduced to [3-¹³C]lactate or caboxylated to oxaloacetate (in astrocytes). However, once the ¹³C label originating from [1-¹³C]glucose enters the TCA cycle, it undergoes several steps to form [4-¹³C]α-ketoglutarate that can be transaminated to form [4-¹³C]glu. In GABAergic neurons, [4-¹³C]glu can be further decarboxylated to [2-¹³C]GABA. Astrocytes remove [4-¹³C]glu from the synaptic cleft (to prevent excitotoxicity), followed by its conversion to [4-¹³C]gln (via astrocyte-specific glutamine synthetase) or to [4-¹³C]α- ketoglutarate, which can enter TCA cycle and form [2-¹³C]-/[3-¹³C] oxaloacetate and can be further transaminated to [2-¹³C]-/[3-¹³C]aspartate. If the ¹³C label is not released in the first turn of TCA cycle, it would form [2-¹³C]-/[3-¹³C]glutamate and glutamine and [4-¹³C]-/[3-¹³C]GABA can be formed after several steps if oxaloacetate labeled from the first turn of the cycle condenses with unlabeled acetyl CoA. With regard to the exclusive metabolism of [1,2-¹³C]acetate in astrocytes, it is first converted to [1,2-¹³C] acetyl CoA in astrocytes. Once [1,2-¹³C]acetyl CoA enters the astrocytic TCA cycle, it can form, after several steps, [4,5-¹³C]α-ketoglutarate, which is the precursor of [4,5-¹³C]glutamate and [4,5-¹³C]glutamine. After transfer of [4,5-¹³C]glutamine to the neurons, it is converted to [4,5-¹³C]glutamate (by phosphate activated glutaminase) in glutamatergic neurons. It can also be converted to [1,2-¹³C]GABA in GABAergic neurons.

The ¹³C-labeling pattern can be influenced by pyruvate carboxylation; however, it occurs only in astrocytes, and an earlier study found that, in the brain of normal anesthetized rat given intravenous infusion of ¹³C glucose with the same protocol as used by us, it comprised only 19% to 26% of the total glutamine synthesis and thus, the majority of the ¹³C labeling was determined by pyruvate dehydrogenase.²⁰ Similarly, in our mice, the ¹³C labeling of Glu and Gln is determined mainly by the pyruvate dehydrogenase pathway, which labels C4 in the 1st turn and C3 and C2 equally in the second turn.

Metabolic Ratios

Relevant metabolic ratios have been calculated according to several publications by Dr Sonnewald.^21–23 We calculated the glycolytic activity,²² TCA cycle activity,²¹ ¹³C glucose, and acetate cycling ratio in terms of glutamate and glutamine,²³ and transfer ratio from astrocytes to neurons.²²

Regional Analysis of Brain Glucose Uptake

The positron emission tomography (PET) images from an earlier published study¹² of the 7-month-old nonTg and 3×Tg-AD with the same lipoic acid treatment were reanalyzed to study brain region-specific differences, as only the whole-brain glucose uptake was measured in the earlier publication. The imaging intensity indicated the level of brain glucose uptake. It did not express in any radioactivity unit, such as Ci or Bq, nor reflect in quantitative calibration of glucose metabolism. A structural MRI Rapid Acquisition with Refocusing Echoes T2 image of a normal 5-month-old C57BL6 mouse brain (dimensions 128 × 128 × 14; voxel size 0.1 × 0.1 × 1mm³) was used as a template for [¹⁸F]-fluorodeoxyglucose (FDG)-PET analysis. [18F]-fluorodeoxyglucose images (dimensions 128 × 128 × 63; voxel size 0.845 × 0.845 × 1.212mm³) were first preprocessed using AMIDE (Andreas Loening, http://.amide.sourceforge.net) and Mango (Research Imaging Institute, University of Texas Health Science Center at San Antonio, Texas) by converting them to nifti files followed by rotation to match MRI template's orientation, and re-sliced in Z to MRI voxel size. Then they were registered to the MRI template using Mango pointmatching co-registration tool with mutual information cost function for transformation, 12 degrees of full search, and tri-linear cost function for interpolation. All images were morphed into similar shape and size and all voxels were aligned to similar slice position as in the MRI template. In the end, the mean value of FDG image intensity for each animal was extracted on bilateral hippocampus, motor somatosensory cortex, as well as the whole-brain regions, which were first drawn on the high resolution MRI image. When ratio was used (region/whole brain), the individual difference of whole-brain uptake was accounted for. In the end, we actually compared regional distribution pattern among groups. In other words, the regional value was compared when the whole brain is ‘1’ for each animal and thus, the regional glucose uptake analysis was not affected by differences in the whole-brain glucose uptake. All FDG images were done in the same batch and the image groups were masked to the image analyst.

High-Performance Liquid Chromatography

High-performance liquid chromatography analysis was carried out to measure the total (¹²C+¹³C) Glu, Gln, Asp, and GABA concentrations in the brain extracts as described previously.⁹,²⁴ Briefly, precolumn derivatization with o-phthalaldehyde and 2-mercaptoethanol and separation on a reverse-phase column with fluorometric detection was carried out with the following modification in the program for gradient elution with methanol and 0.05M phosphate buffer (pH 5.29); the percentage of methanol was increased from 25% to 40% in 29.5 minutes, then to 84% in 22.3 minutes to resolve GABA from adjacent peaks. The peak areas of the pure chemicals (standards) were used to quantify the metabolites. The percentage ¹³C enrichment of [4-¹³C]Glu for example, was calculated from the concentration of [4-¹³C]Glu (after correction for natural abundance ¹³C) and the total Glu (¹²C+¹³C) concentration in each mouse.

Data Analysis

Student's two-tailed t-test was used for statistical analysis of paired data. The level of statistical significance and the values of n are indicated in the respective figures. *P≤0.05, **P≤0.01.

RESULTS

Comparison of 13C Metabolite Concentrations among Different Mouse Groups after [1-13C]Glucose Infusion

To compare the results with our earlier studies performed in 13-month-old mice, we carried out [1-¹³C]glucose infusion in 7-month-old nonTg and 3×Tg-AD mice. Well-resolved peaks of ¹³C-labeled isotopomers of alanine, lactate, Glu, Gln, Asp, GABA, NAA, MI, and glucose (C1α and β) were observed as shown previously.⁹ The concentrations of the different isotopomers of Glu, Gln, Asp, GABA, and NAA in 7-month-old nonTg and 3×Tg-AD mice plus/minus lipoic acid feeding are shown in the Table 1 (expressed in nmol/mg brain tissue ± s.e.m.). In all the four groups of mice, the highest labeling was of [4-¹³C]Glu followed by [3-¹³C] Glu and [2-¹³C]Glu (almost identical), with the lowest labeling of [1-¹³C]Glu. Overall, labeling pattern was in good agreement with the known ¹³C-labeling pattern of brain metabolites after [1-¹³C] glucose infusion.²⁵ Slightly decreasing labeling was found in the 7- month-old nonTg mice compared with previous results in 13- month-old nonTg mice.⁹ Surprisingly, a prominent increase of labeling was observed in the 7-month-old 3×Tg-AD mice in comparison with the age-matched nonTg mice.

Table 1.

Concentrations of the different isotopomers of ¹³C Glu, Gln, Asp, GABA, and NAA after [1-¹³C]glucose infusion

	nonTg (A)	nonTg+LA (B)	3×Tg-AD (C)	3×Tg-AD+LA (D)	A versus B	A versus C	C versus D

					P value
Metabolite
[4-¹³C]Glu	0.94 ± 0.08	1.25 ± 0.09	1.59 ± 0.26	1.05 ±0.14	0.010**	0.010**	0.046*
[3-¹³C]Glu	0.61 ±0.06	0.75 ±0.09	1.00 ± 0.20	0.58 ± 0.06	0.197	0.041*	0.034*
[2-¹³C]Glu	0.58 ± 0.03	0.66 ±0.03	0.81 ±0.18	0.46 ± 0.06	0.040*	0.158	0.050
[1-¹³C]Glu	0.18 ±0.02	0.22 ±0.02	0.37 ± 0.08	0.18 ±0.04	0.160	0.020*	0.027*
[4-¹³C]Gln	0.34 ± 0.06	0.43 ± 0.06	0.54 ± 0.07	0.33 ± 0.04	0.205	0.019*	0.008**
[3-¹³C]Gln	0.25 ± 0.03	0.36 ±0.06	0.44 ± 0.05	0.24 ±0.04	0.060	0.004**	0.004**
[2-¹³C]Gln	0.22 ±0.02	0.30 ±0.05	0.36 ± 0.06	0.21 ±0.02	0.144	0.028*	0.012*
[1-¹³C]Gln	0.08 ± 0.02	0.11 ±0.01	0.16 ±0.03	0.08 ±0.01	0.029*	0.014*	0.012*
[4-¹³C]Asp	0.08 ± 0.02	0.07 ±0.02	0.14 ±0.05	0.08 ± 0.02	0.716	0.139	0.141
[3-¹³C]Asp	0.24 ±0.01	0.23 ±0.01	0.26 ±0.01	0.26 ±0.01	0.410	0.103	0.498
[2-¹³C]Asp	0.14±0.01	0.16±0.02	0.22 ± 0.03	0.13 ±0.02	0.078	0.015*	0.015*
[1-¹³C]Asp	0.09 ±0.01	0.10±0.02	0.14 ±0.03	0.07 ±0.01	0.864	0.078	0.022*
[4-¹³C]GABA	0.14 ±0.02	0.17 ±0.02	0.21 ±0.11	0.11 ±0.02	0.310	0.034*	0.009**
[3-¹³C]GABA	0.14 ±0.02	0.14±0.02	0.20 ± 0.02	0.13 ±0.02	0.872	0.070	0.019*
[2-¹³C]GABA	0.20 ± 0.05	0.30 ±0.02	0.34 ± 0.05	0.22 ±0.02	0.124	0.075	0.033*
[1-¹³C]GABA	0.13 ±0.02	0.14±0.02	0.19 ±0.02	0.11 ±0.02	0.645	0.048*	0.017*
[3-¹³C]NAA	0.07 ±0.01	0.08 ±0.01	0.10 ±0.02	0.07 ±0.02	0.109	0.121	0.197
[2-¹³C]NAA	0.06 ±0.01	0.06 ±0.01	0.07 ± 0.02	0.05 ± 0.00	0.189	0.295	0.116

Concentrations of the different isotopomers of ¹³C Glu, Gln, Asp, GABA, and NAA in 7-month-old nonTg and 3×Tg-AD mice plus/minus lipoic acid, after 60 minutes of [1-¹³C]glucose infusion. Results in the column 2–5 are presented as average nmol/mg brain tissue ±s.e.m.; results in the columns 6–8 are the P values obtained from a two-tailed student t-test after comparing between the groups as indicated.

P≤0.05,

P≤0.01 (indicated in parenthesis); total n=22 and n≥5 per group. The results for Glu, Gln, and Asp are corrected for natural abundance (the results for NAA and GABA are not corrected for natural abundance).

¹³C isotopomers of Glu showed a marked ‘increase’ in the 3×Tg-AD mice compared with nonTg mice; (values in parenthesis represents the % increase in 3×Tg-AD compared with nonTg mice): [4-¹³C]Glu (∼69%), [3-¹³C]Glu (∼64%), [2-¹³C]Glu (∼40%), [1-¹³C]Glu (∼100%). Similarly, ¹³C isotopomers of Gln also showed the trend of marked increase in the 3×Tg-AD mice compared with nonTg mice: [4-¹³C]Gln (∼58%), [3-¹³C]Gln (∼77%), [2-¹³C]Gln (∼61%), [1-¹³C]Gln (∼100%). Overall, more than 60% increase in the levels of ¹³C Glu and Gln labeling was observed in the 3×Tg-AD mice compared with the nonTg mice. Similar trends were also observed in the labeling for the different isotopomers of Asp, NAA, and GABA (Table 1). Lipoic acid feeding to the nonTg mice had almost no effect, but it normalized the metabolite levels in the 3×Tg-AD mice by bringing it down to the levels of nonTg mice. (Table 1).

Fractional 13C (%) Enrichment of Metabolites after Glucose Infusion—Comparison among Different Mouse Groups

The enrichment for each of the ¹³C-labeled metabolites was calculated to assess the flux of ¹³C label. The calculated percentage of ¹³C-enrichment levels for the different isotopomers of Glu, Gln, and Asp are represented in Figure 1 as mean ± s.e.m. The trend of increase in ¹³C-labeled metabolites levels was also replicated in the percent enrichment levels (values in parenthesis represents the percentage of increase in the 3×Tg-AD mice compared with nonTg mice); [4-¹³C]Glu (∼92%) (Figure 1 Aii), [3-¹³C]Glu (∼84%) (Figure 1 Aiii), and [2-¹³C]Glu (∼60%) (Figure 1 Aiv). The ¹³C isotopomers of Gln also showed a similar trend of marked increase in the 3×Tg-AD mice compared with nonTg mice: [4-¹³C]Gln (∼94%) (Figure 1 Bii), [3-¹³C]Gln (∼114%) (Figure 1 Biii), [2-¹³C]Gln (∼90%) (Figure 1 Biv). Similarly, the ¹³C isotopomers of Asp also showed a similar trend of marked increase in the 3×Tg-AD mice compared with nonTg mice: [4-¹³C]Asp (∼101%) (Figure 1 Cii), [3-¹³C]Asp (∼8%) (Figure 1 Ciii), [2-¹³C]Asp (∼69%) (Figure 1 Civ). As seen in the metabolite isotopomer concentration, the lipoic acid-fed 3×Tg-AD mice showed a prominent decrease and the levels in these mice were close to the nonTg mice (Figure 1). However, no effect of lipoic acid was seen in the nonTg mice.

Figure 1

Total metabolite levels and percent ¹³C enrichment of the different metabolite isotopomers after [1-¹³C]glucose infusion and regional brain glucose uptake. Total metabolite levels (¹²C+¹³C) of Glu (Ai), Gln (Bi), and Asp (Ci) in 7-month-old nonTg and 3×Tg-AD mice±lipoic acid feeding shown as mean nmol/mg brain tissue ±s.e.m. Enrichment percentage for glutamate isotopomers (Aii-iv), glutamine isotopomers (Bii-iv), and aspartate isotopomers (Cii-iv) shown as mean percentage ±s.e.m.; P values obtained from a two-tailed student t-test comparing the specific groups is indicated under those groups. *P≤0.05, **P≤0.01; total n=22 and n≥ 5 per group. [¹⁸F]-fluorodeoxyglucose- positron emission tomography images after [¹⁸F]-FDG injection were co-registered to a high resolution magnetic resonance imaging to obtain region-specific glucose uptake as described in the materials and methods section. (Di) Representation of how the regions of interest were drawn to calculate the glucose uptake for hippocampus and motor and somatosensory cortex. The mean intensity values of glucose uptake calculated in regions of (Dii) hippocampus and (Diii) motor and somatosensory cortex. As shown, no statistically significant differences were seen among the different groups; total n=24, n≥5 per group.

The total [¹²C+¹³C] concentrations of Glu, Gln and Asp after glucose infusion were measured by high-performance liquid chromatography and represented as mean nmol/mg of brain tissue ± s.e.m. for each group (Figure 1 Ai-Ci). Similar to the results in 13-month-old mice,⁹ there was no statistically significant difference in total [¹²C+¹³C] Glu, Gln, and Asp concentrations among different groups of mice at 7 months of age (except the effect of lipoic acid in increasing total Gln of the 3×Tg-AD mice).

Regional Brain Glucose Uptake in 7-Month-Old nonTg and 3×Tg-AD Mice Plus/Minus Lipoic Acid Feeding

Raw PET data from a previous brain glucose uptake study in the same age group of mice with the same lipoic acid treatment¹² was re-sliced and the [¹⁸F]FDG-PET images were reanalyzed to calculate the region-specific brain glucose uptake in the hippocampus and motor and somatosensory cortex (Figure 1 Di); no regional differences were observed among the different groups (Figure 1 Dii and iii), although glucose uptake in the whole brain was slightly decreased in the 7-month-old 3×Tg-AD mice as found in the earlier study.¹²

Comparison of Glial and Neuronal Metabolism—13C Metabolite Concentration after [1-13C]Glucose+[1,2-13C]Acetate Infusion

These studies were carried out by infusing [1-¹³C]glucose+[1,2-¹³C] acetate for different periods of time (5 minutes, 20 minutes, 60 minutes, and 150 minutes) to assess time-dependent changes in ¹³C flux. Co-infusion of [1-¹³C]glucose+[1,2-¹³C]acetate leads to a typical pattern of labeling (Figure 2) and nuclear magnetic resonance trace (Figure 3). A hypermetabolic state in both neurons and astrocytes of the 3×Tg-AD mice was evidenced by the increase in absolute levels of the different ¹³C metabolite isotopomers at 60 minutes (data not shown) and 150 minutes (Table 2) (data expressed as nmol/mg brain tissue ± s.e.m.).

Figure 2.

Typical labeling pattern after [1-¹³C]glucose+[1,2-¹³C]acetate infusion. Labeling pattern after co-infusion of [1-¹³C]glucose+[1,2-¹³C] acetate as described in the Materials and Methods section (adapted from Melo et al⁴⁰).

Figure 3

A representative ¹³C nuclear magnetic resonance spectrum of brain extract. A proton-decoupled nuclear Overhauser effect (NOE)-enhanced representative ¹³C spectrum (150.86 MHz) of typical perchloric acid brain extract after [1-¹³C]glucose and [1,2-¹³C]acetate infusion for 150 minutes showing several metabolite isotopomers; parts of the spectra have been zoomed to clearly show all peaks (A) and spectra focusing on glutamate/glutamine isotopomers between 27 and 35 p.p.m. and changes in the spectra as a function of time after [1-¹³C]glucose+[1,2-¹³C]acetate infusion for 5 minutes (bolus), 20 minutes, 60 minutes, and 150 minutes (B). The chemical shift and internal standard, 1,4-dioxane is at 67.4 p.p.m. and isopropyl alcohol with three equivalent methyl carbons (solvent in the pH indicator) at 24.6 p.p.m.

Table 2.

Concentrations of the different isotopomers of ¹³C Glu, Gln, Asp, GABA, MI, and NAA after [1-¹³C]glucose+[1,2-¹³C] acetate infusion

	nonTg (A)	nonTg+LA (B)	3×Tg-AD (C)	3×Tg-AD+LA (D)	A versus B	A versus C	C versus D

					P value
Metabolite
[4-¹³C]Glu	1.24 ± 0.06	1.24 ± 0.12	1.18 ± 0.12	1.52 ±0.12	0.991	0.681	0.096
[3-¹³C]Glu	0.97 ± 0.12	0.90 ± 0.08	1.65 ± 0.17	1.10 ±0.08	0.686	0.010**	0.026*
[2-¹³C]Glu	0.87 ± 0.08	0.87 ± 0.10	1.69 ± 0.23	1.02 ±0.07	0.998	0.009**	0.036*
[4,5-¹³C]Glu	0.33 ± 0.04	0.36 ± 0.05	0.30 ± 0.03	0.39 ±0.07	0.571	0.712	0.282
[1,2-¹³C]Glu	0.25 ± 0.04	0.17 ± 0.02	0.41 ± 0.04	0.27 ±0.02	0.969	0.029*	0.016*
[2,3-¹³C]Glu	0.32 ± 0.06	0.37 ± 0.05	0.55 ± 0.05	0.38 ±0.05	0.543	0.019*	0.037*
[2,3-¹³C]Gluoo	0.70 ± 0.12	0.73 ± 0.05	1.08 ± 0.10	0.74 ±0.05	0.830	0.046*	0.037*
[4-¹³C]Gln	0.35 ± 0.03	0.31 ± 0.02	0.48 ± 0.12	0.39 ±0.02	0.391	0.292	0.490
[3-¹³C]Gln	0.38 ± 0.03	0.34 ± 0.03	0.74 ± 0.06	0.47 ±0.03	0.401	0.001**	0.008**
[2-¹³C]Gln	0.34 ± 0.03	0.32 ± 0.04	0.46 ± 0.10	0.37 ±0.09	0.655	0.250	0.487
[4,5-¹³C]Gln	0.37 ± 0.09	0.49 ± 0.04	0.45 ± 0.08	0.38 ±0.08	0.298	0.540	0.593
[1,2-¹³C]Gln	0.14 ± 0.03	0.13 ± 0.01	0.22 ± 0.01	0.15 ±0.02	0.775	0.039*	0.029*
[2,3-¹³C]Gln	0.14 ± 0.02	0.14 ± 0.02	0.18 ± 0.03	0.13 ±0.02	0.906	0.375	0.301
[2,3-¹³C]Glnoo	0.34 ± 0.05	0.33 ± 0.03	0.53 ± 0.05	0.38 ±0.04	0.787	0.034*	0.065
[4-¹³C]Asp	0.29 ± 0.02	0.26 ± 0.03	0.52 ± 0.08	0.35 ±0.04	0.516	0.015*	0.102
[3-¹³C]Asp	0.34 ± 0.03	0.31 ± 0.03	0.64 ± 0.06	0.37 ±0.02	0.528	0.003**	0.006**
[2-¹³C]Asp	0.20 ± 0.02	0.18 ± 0.03	0.30 ± 0.07	0.22 ±0.06	0.657	0.177	0.366
[1-¹³C]Asp	0.22 ± 0.06	0.18 ± 0.02	0.50 ± 0.12	0.38 ±0.12	0.716	0.065	0.485
[3,4-¹³C]Asp	0.06 ± 0.02	0.06 ± 0.01	0.11 ± 0.02	0.08 ±0.02	0.804	0.116	0.323
[2,3-¹³C]Asp	0.07 ± 0.01	0.06 ± 0.01	0.12 ± 0.02	0.06 ±0.02	0.232	0.030*	0.035*
[4-¹³C]GABA	0.30 ± 0.08	0.31 ± 0.02	0.47 ± 0.06	0.32 ±0.02	0.793	0.058	0.057
[3-¹³C]GABA	0.26 ± 0.03	0.24 ± 0.05	0.41 ± 0.04	0.23 ±0.07	0.789	0.004**	0.074(*)
[2-¹³C]GABA	0.38 ± 0.03	0.32 ± 0.02	0.52 ± 0.06	0.33 ±0.10	0.090	0.023*	0.156
[1-¹³C]GABA	0.15 ± 0.02	0.10 ± 0.02	0.22 ± 0.02	0.14 ±0.03	0.078	0.060	0.120
[1,2-¹³C]GABA	0.10 ± 0.02	0.10 ± 0.02	0.13 ± 0.02	0.09 ±0.02	0.709	0.304	0.174
[2,3-¹³C]GABA	0.14 ± 0.03	0.14 ± 0.01	0.19 ± 0.03	0.14 ±0.02	0.916	0.172	0.182
[3,4-¹³C]GABA	0.05 ± 0.02	0.06 ± 0.01	0.07 ± 0.02	0.05 ±0.01	0.359	0.201	0.130
[4,6-¹³C]MI	0.10 ± 0.01	0.09 ± 0.01	0.14 ± 0.02	0.10 ±0.02	0.108	0.100	0.190
[2-¹³C]MI	0.04 ± 0.01	0.03 ± 0.01	0.10 ± 0.02	0.05 ±0.01	0.387	0.023*	0.043*
[1,3-¹³C]MI	0.11 ± 0.01	0.09 ± 0.02	0.14 ± 0.02	0.09 ±0.02	0.203	0.146	0.112
[5-¹³C]MI	0.06 ± 0.00	0.06 ± 0.01	0.10 ± 0.01	0.07 ±0.01	0.085	0.001**	0.009**
[6-¹³C]NAA	0.11 ± 0.02	0.11 ± 0.02	0.11 ± 0.01	0.10 ±0.02	0.902	0.992	0.543
[3-¹³C]NAA	0.25 ± 0.06	0.14 ± 0.01	0.26 ± 0.05	0.21 ±0.03	0.132	0.917	0.413
[2-¹³C]NAA	0.10 ± 0.01	0.08 ± 0.01	0.10 ± 0.01	0.06 ±0.02	0.148	0.686	0.114

Concentrations of the different isotopomers of ¹³C Glu, Gln, Asp, GABA, MI, and NAA in 7-month-old nonTg and 3×Tg-AD mice ±lipoic acid, after 150 minutes of [1-¹³C]glucose+[1,2-¹³C] acetate infusion. Results in the columns 2 to 5 are presented as average nmol/mg brain tissue ±s.e.m.; results in the columns 6 to 8 are the P values obtained from a two-tailed student t-test after comparing between the groups as indicated.

P≤ 0.05,

P≤0.01 (indicated in parenthesis); total n=17 and n≥4 per group. The results for Glu, Gln, and Asp are corrected for natural abundance (the results for GABA, MI, and NAA are not corrected for natural abundance). ∞[2,3 and 3,4-¹³C]Glu/Gln.

The calculated ¹³C metabolite isotopomers levels at 150 minutes infusion from 7-month-old nonTg and 3×Tg-AD mice are shown in the Table 2. Metabolites labeled in the second and third turns of TCA cycle showed a clear trend of increase in the 3×Tg-AD mice compared with the nonTg mice (Table 2). The glutamate isotopomers labeled in the second and third turns showed an average increase of >60% in comparison with agematched nonTg mice. Similarly, the glutamine isotopomers labeled in the second and third turns showed an average increase of greater than 40% in comparison with the age-matched nonTg mice. Similar trends of increase in the 3×Tg-AD for ¹³C-labeled metabolites for Asp, GABA, and MI were observed. Mice fed lipoic acid showed similar trend as seen with just [1-¹³C]glucose infusion, i.e., there were almost no changes in the nonTg mice fed lipoic acid but the 3×Tg-AD mice fed lipoic acid showed a decrease in the metabolite isotopomers concentration and almost brought the levels of metabolite isotopomer concentration to those of nonTg mice (Table 2).

The 3×Tg-AD mice show no increase of the major neuronal metabolite from the first turn of TCA cycle from [1-¹³C]glucose i.e., [4-¹³C]Glu; similarly, there was no increase in the major astrocytic metabolite from the first turn TCA cycle of [1,2-¹³C]acetate i.e., [4,5-¹³C]Glu. No differences were seen in the NAA isotopomers after 150 minutes infusion of [1-¹³C]glucose+[1,2-¹³C]acetate. A clear trend of increase in the ¹³C labeling of 3×Tg-AD mice was not seen at 5 minutes and 20 minutes (data not shown). This is not unexpected because, at 5 minutes, mainly the metabolites labeled from first cycle of TCA cycle are detected. Although several of them showed an increase in the 3×Tg-AD compared with nonTg, a uniform trend was not seen.

Comparison of Glial and Neuronal Metabolism—Fractional 13C (%) Enrichment of Metabolites after [1-13C]Glucose+[1,2-13C]Acetate Infusion

The fractional enrichment of the different metabolite isotopomers for glutamate and glutamine was calculated at the different time points (Figure 4). Interestingly, the metabolites labeled in the subsequent turns of TCA cycle (after the first cycle) at 60 minutes and 150 minutes show a general trend of increasing enrichment in the 7-month-old 3×Tg-AD compared with age-matched nonTg mice (Figures 4 Aii, iii, v, vi and 4 Bii, iii, v, vi). For several of the metabolite isotopomers, the enrichment was increased by 100% and thus clearly showing a state of hypermetabolism in the 7-month-old 3×Tg-AD mice. Lipoic acid feeding to the nonTg mice did not result in statistically significant differences but lipoic acid feeding to the 3×Tg-AD mice showed similar decrease of enrichment as seen in the metabolite isotopomer concentrations (Figures 4 Aii, iii, v, vi and 4 Bii, iii, v, vi).

Figure 4

¹³C enrichment percentage of the different metabolite isotopomers after [1-¹³C]glucose and [1,2-¹³C]acetate infusion for different time periods. The different graphs show the labeled glutamate/glutamine isotopomers after the specified period of [1-¹³C]glucose+[1,2-¹³C] acetate infusion for either 5 minutes, 20 minutes, 60 minutes, or 150 minutes. (A) Left shows glutamate isotopomers generated primarily from glucose metabolism (single labeled) (Figs. Ai-iii) and acetate metabolism (double labeled) (Figs. Aiv-vi). (B) Right shows glutamine isotopomers generated primarily from glucose metabolism (single labeled) (Figs. Bi-iii) and acetate metabolism (double labeled) (Figs. Biv-vi). *P≤0.05, **P≤ 0.01; total n =62 and n≥15 per time point. Note that [2,3 and 3,4-¹³C]glu/gln doublets at 60 and 180 minutes also contain those derived from [1-¹³C]glucose.

However, statistically significant differences were not observed for enrichment of glutamate isotopomers from the first turn of TCA cycle i.e., [4-¹³C]Glu and [4,5-¹³C]Glu from [1-¹³C]glucose and [1,2-¹³C]acetate, respectively at 150 minutes (Figures 4 Ai and 4 Aiv). Albeit, not statistically significant, trends of increasing [4-¹³C] Glu and [4,5-¹³C]Glu enrichment were also observed in the 3×Tg- AD mice after 5 minutes and 20 minutes (magnitude varied across metabolites). Similar trends were also seen in the enrichment of glutamine isotopomers labeled from first turn of TCA cycle, i.e., [4-¹³C]Gln and [4,5-¹³C]Gln from [1-¹³C]glucose and [1,2-¹³C] acetate, respectively (Figures 4Bi and 4Biv). Lipoic acid treatment had no statistically significant effect on these metabolites in both nonTg and 3×Tg mice (Figures 4 Ai, iv and 4Bi, iv).

Metabolic Ratios in 7-Month-Old nonTg and 3×Tg-AD Mice Plus/Minus Lipoic Acid Feeding

As shown in Figure 5Ai, ii, TCA cycle activity and percentage of glycolytic activity were increased by twofold in the 3×Tg-AD mice as compared with the age-matched nonTg mice. Lipoic acid had almost no effect in the nonTg mice but substantially decreased the TCA cycle activity and percentage of glycolytic activity in the 3×Tg-AD mice, bringing it close to the levels of nonTg mice (Figures 5Ai, ii). Looking at the ¹³C glucose cycling ratio in terms of glutamate, it was increased by ∼ 100% in the 3×Tg-AD mice but brought back to the levels of nonTg mice when 3×Tg-AD mice were fed lipoic acid (Figure 5Bi). ¹³C acetate cycling ratio in terms of glutamate also showed a similar increase in the 3×Tg-AD mice (Figure 5Ci).

Figure 5

Metabolic ratios calculated after [1-¹³C]glucose+[1,2-¹³C] acetate infusion for 150 minutes. Relevant metabolic ratios, calculated as described in the Materials and Methods section, after [1-¹³C] glucose+[1,2-¹³C]acetate infusion for 150 minutes are shown in the graphs (A-D). Percentage of glycolytic activity based on the levels of [3-¹³C]alanine (Ai), TCA cycle activity based on glutamate formation from [1-¹³C]glucose (Aii), ¹³C glucose cycling ratio for glutamate and glutamine respectively (Bi, ii), ¹³C acetate cycling ratio for glutamate and glutamine, respectively (Ci, ii), transfer of glutamine from astrocytes to glutamatergic neurons where [4,5-¹³C]glutamine is converted to [4,5-¹³C]glutamate (Di), and transfer of glutamine to GABAergic neurons where [4,5-¹³C]glutamine is converted to [1,2-¹³C]GABA (Dii).

Notably, glutamine transfer from astrocytes to glutamatergic neurons was decreased in the 3×Tg-AD mice (Figure 5Di). It should be noted that the transfer ratio between astrocytes to glutamatergic or GABAergic neurons represents the substrate transfer (glutamine) from astrocytes to the specific neurons.

DISCUSSION

This study was aimed at understanding the age-related metabolic changes in the 7-month-old mouse model of AD (3×Tg-AD) and the effect of lipoic acid. The study used ¹³C ex vivo nuclear magnetic resonance after supplying a single (glucose) or multiple (glucose and acetate) brain substrates to dissect neuronal and astrocytic metabolism. The flux of ¹³C from glucose and acetate through the TCA cycle, synthesis of metabolic and neurotransmitter pools of Glu and GABA, their precursor Gln and of several other metabolites, have been calculated. The 7-month-old 3×Tg-AD mice show a clear hypermetabolic state evidenced by an increase (>50%) in the concentration of several [¹³C] metabolites. Treatment with lipoic acid brought the levels of these ¹³C metabolites close to those observed in the nonTg mice, thus hinting at its ability to regulate glucose uptake and metabolism.

[1-13C]Glucose Metabolism

The concentrations of ¹³C-labeled Glu, Gln, Asp, GABA, and NAA isotopomers in Table 1 show a clear increase of ¹³C flux in the 3×Tg-AD mice by ∼ 50% in key metabolites. This signals greater incorporation of ¹³C label from metabolism of glucose and thus faster turnover of TCA cycle-related metabolites. A simple explanation of these results would be an increase in brain glucose uptake; however, as reported in an earlier study, whole-brain glucose uptake in the 7-month-old 3×Tg-AD mice was slightly decreased.¹² Whole-brain FDG-PET can mask regional differences owing to the spill-over effects (strong near facial glands). Accordingly, previous data obtained from a different study but involving animals of same age and same lipoic acid treatment were reanalyzed to calculate regional glucose uptake in the hippocampus and motor and somatosensory cortex (Figure 1Di). No regional glucose uptake differences were found among the four groups. Hence, the hypermetabolic state in the 3×Tg-AD mice cannot be explained by increased FDG-PET determined glucose uptake in these regions; alternatively, it might be more appropriate to use autoradiography methods for assessing regional brain differences.

[1-13C]Glucose+[1,2-13C]Acetate Metabolism

The question of whether the hypermetabolism observed with [1-¹³C]glucose infusion (single substrate) can be ascribed to mainly hypermetabolic neurons or a similar hypermetabolic state also exists in astrocytes was addressed with co-infusion of [1-¹³C] glucose and [1,2-¹³C]acetate over four different time periods (with 150 minutes being the last time period); it must be noted that by 150 minutes of [1-¹³C]glucose+[1,2-¹³C]acetate infusion, the fractional amounts of ¹³C multiplets in glutamate and glutamine reach approximately steady-state levels.¹⁴ The concentration of [13C] metabolites at 150 minutes (Table 2) and the enrichment of different metabolites (Figure 4) revealed a general increase in the 3×Tg-AD mice, especially at 60 minutes and 150 minutes. There were almost no differences among the four groups in metabolite and enrichment levels of [4-¹³C]glu/gln, [4,5-¹³C]glu/gln at 150 minutes after infusion. These are the metabolites that are derived mainly from the first turn of the TCA cycle. It is possible that after the 150-min infusion, inter-group differences in the rate of formation of the TCA cycle-related metabolites from its first turn are masked. However, the metabolites labeled in the second and subsequent turns of TCA i.e., [3/2-¹³C]glu/gln and [1,2-¹³C]glu/gln show an almost generalized increase of enrichment at 60 minutes and 150 minutes (Figure 4).

Metabolic Ratios

Hypermetabolism was also evidenced by a 100% increase in (a) glycolytic activity (measured from alanine levels) and (b) TCA cycle activity²¹ (Figure 5); these increases were based on comparisons with the nonTg mice. ¹³C glucose or acetate cycling ratio in terms of glutamate or glutamine gives information about how long does the ¹³C label derived from [1-¹³C]glucose and [1,2-¹³C]acetate stays in the TCA cycle before being incorporated into glutamate or glutamine; e.g., ¹³C glucose cycling ratio in terms of glutamate means how long does the ¹³C label derived from [1-¹³C]glucose stays in the TCA cycle before being incorporated into glutamate.²³ Approximately 100% increase of cycling ratio from glucose and acetate in terms of glutamate were also observed in the 3×Tg-AD mice (Figures 5Bi and 5Ci). The results in this study suggest that the ¹³C label turns over faster in the TCA cycle of 7-month-old 3×Tg-AD mice.

Astrocytes and neurons interact actively, with astrocytes supporting neuronal metabolic requirements and clearing excess glutamate from extracellular space to prevent neuronal excitotoxicity.²³ A large fraction of the glutamate and GABA present in the neurons is synthesized from glutamine transferred from astrocytes. Using the data of glutamine labeled specifically in the astrocytes (from [1,2-¹³C]acetate) and comparing it with the similarly labeled glutamate, the transfer ratio between astrocytes and neurons may be calculated. Interestingly, the transfer of glutamine from astrocytes to neurons for glutamate was decreased in the 3×Tg-AD mice (Figure 5). This indicates that the hypermetabolic state in the 7-month-old 3×Tg-AD mice does not have a strong astrocytic support system to fulfill the neuronal metabolic demands. It is possible that the reduction in supply of metabolites from astrocytes would render the neurons energy deficient and thus force the neurons into a hypermetabolic state to meet their energy demands; that might result in greater flux of glutamate. Alternatively, it is also possible that the uptake of glutamine by neurons could be impaired.

Effect of Lipoic Acid

The data shown in this study indicates that lipoic acid treatment brought the high metabolite levels observed in the 3×Tg-AD mice close to those of nonTg mice. Exogenous lipoic acid is expected to equilibrate among the different intracellular and extracellular compartments but cannot substitute for covalently bound lipoic acid (as the cofactor of mitochondrial complexes such as pyruvate- and α-ketoglutarate dehydrogenases).²⁶ It is likely that thiol/disulfide exchange reactions facilitated by lipoic acid are involved in activation or stimulation of cysteine-rich members of insulin signaling, such as the insulin receptor itself and insulin receptor substrate. This, in turn, regulates metabolism of glucose and the subsequent TCA cycle-related metabolites. Studies using adipocytes showed insulin-mimetic effects of lipoic acid, demonstrated by the activation of insulin receptor and increased glucose uptake.^10,11 Lipoic acid fed to 7-month-old 3×Tg-AD mice did not improve the reduced LTP but was effective in restoring the reduced LTP in 13-month-old 3×Tg-AD mice.¹² We found opposite effects of lipoic acid on the 3×Tg-AD mice at the ages of 7 and 13 months with respect to its ability to modulate metabolism. Lipoic acid countered the hypometabolic state of the 13-month old 3×Tg-AD mice and increased glycolytic metabolism, evidenced by increased ¹³C labeling of several TCA cycle-related metabolites like glutamate, glutamine, aspartate, NAA, and GABA after [1-¹³C] glucose infusion. The percent enrichments for the different isotopomers of glutamate and glutamine were also increased substantially by lipoic acid feeding after they were found to be reduced by ∼ 50% in the 13-month-old 3×Tg-AD mice.⁹ However, lipoic acid feeding to 7-month old 3×Tg-AD mice reversed the hypermetabolic state by reducing the metabolite levels, evidenced by the data presented in this manuscript. The functional effects of lipoic acid in the 7- (this study) and 13-month-old⁹ 3×Tg-AD mice seem to be related or determined by the cellular redox environment. Changes in the cellular redox environment would change the thiol-disulfide exchange abilities of lipoic acid, as it could be modified from its oxidized form i.e., lipoic acid, to reduced form i.e., dihydrolipoic acid or vice versa.

Association of Hypermetabolism, β-Amyloid Plaques, and Hyperactivity Related to Seizures in Alzehimer's Disease

Of key importance in understanding the hypermetabolic state reported in the present study is the stage of pathology in the 3×Tg-AD mice at 7 months of age i.e., diffusion of Aβ plaques in neocortex is present but immunoreactivity to human tau is absent.⁸ Thus, at 7-months of age, the 3×Tg-AD mice are expected to be comparable with several amyloid mouse models of AD (e.g., Tg2576 and APP/PS1) in terms of the presence of plaque pathology and the absence of tau pathology. The 7-month-old Tg2576 mice show an increased cerebral glucose uptake²⁷ and the APP/PS1 mouse model revealed increased glucose uptake only close to plaques but not in amyloid-free cerebral tissues²⁸ and increased basal metabolic rate of cells surrounding amyloid plaques.²⁹ The 5XFAD mouse model of AD also showed an increased brain glucose uptake.³⁰ Studies carried out in the APP23×PS45 mouse model of AD found that the synchronized neuronal firing could well increase the risk for seizure-like activity.³¹ This is correlated with increased epileptic seizures in AD patients.³² In addition, the risk of epileptic activity is 87-fold greater in AD patients with early-onset dementia (that is typically characterized by Aβ pathology) and the relationship between AD and seizures is even tighter in early-onset familial AD^33,34 (the 3×Tg-AD mouse model perhaps resembles early-onset familial AD better than late onset AD, as it has genetic mutations from birth). These results raise an important question of whether the extent of subclinical epileptic activity in AD has been underestimated.

The fact that total metabolite levels (¹²C+¹³C—Glu, Gln, Asp) did not change (Figure 1) but was accompanied by major changes in the flux of ¹³C from [1-¹³C]glucose and [1,2-¹³C]acetate to metabolites strengthens the possibility of a shift in metabolic pathways. More importantly, whether the sudden increase of labeled glutamate after [1-¹³C]glucose and [1,2-¹³C]acetate infusion leads to excitotoxic damage remains to be determined. Excitotoxicity occurs when glutamate receptors are overstimulated by excess glutamate in the extracellular synaptic fluid where the concentration is ∼ 1/2,000th of the intracellular concentration,³⁵ so any elevation of extracellular glutamate in 3×Tg mice during infusion of ¹³C-labeled substrate will not be detectable in whole-brain glutamate measured at end-point. Our results clearly show that the turnover (flux) of ¹³C was faster in 3×Tg mice, suggesting increased neuronal firing and faster release of the excitatory neurotransmitter glutamate into the synaptic fluid and fits well with the finding in other mouse models at similar stage of pathology as described above. Taken together, these results suggest an intriguing possibility that the increased flux of glutamate after [1-¹³C]glucose and [1,2-¹³C]acetate infusion can lead to excitotoxicity. With these stipulations, it may be speculated whether this excitotoxic damage (that might be initiated at 7-months or earlier) may be partially responsible for the hypometabolic state seen in the 13-month-old 3×Tg-AD mice⁹ (the values in the previous study⁹ were reported in mmol/L; for conversion of values from mmol/L to nmol/mg brain tissue, divide the values by 0.8 based on the fact that water constitutes 80% of the brain weight). One of the few drugs approved to treat AD is memantine, a drug that acts against glutamate-mediated hyperactivity by partially blocking the NMDAR and in turn hypothesized to control excitotoxicity.³⁶ It would be interesting to assess the effect of memantine and lipoic acid given as a two-hit combination therapy wherein the former targets the glutamatemediated excitotoxicity and the latter addresses the regulation of glucose metabolism. Examination of (a) behavioral or electrographic seizures, (b) possible changes in extracellular glutamate during infusion of ¹³C-labeled substrates, and (c) histologic examination of end-point brain for possible excitotoxic damage remain challenging issues for future investigation. There are also reports showing physiologic changes in the 3×Tg-AD mice evidenced by increased food consumption, changes in weight, increased oxygen consumption, carbon dioxide production, defective gutbrain signaling, changes in core body temperature; thus, showing physiologic hyperactivity.^37–39 In addition to the association of hypermetabolism with plaques, a metabolic study in the tau transgenic mouse model of AD showed hypermetabolism in cerebral cortex.¹⁹

Concluding Remarks

The major finding of this study is the hypermetabolic state in the 7-month-old 3×Tg-AD mice as compared with age-matched nonTg mice and these results are in contrast to the hypometabolism observed in 13-month-old 3×Tg-AD mice. Importantly, the specific role of metabolism in the coordinated picture of hyperactivity (electrophysiologic and physiologic), previously reported in multiple mouse models of AD was highlighted. In view of the studies referenced in this section, an important question emerges, i.e., does cerebral glucose metabolism in very early stage of AD patients show an increase that has not been examined closely enough? Although, whole-brain glucose uptake measured by PET-computed tomography imaging may show a decrease associated with mild cognitive impairment (herein, a slight but noticeable and measureable decline is seen in cognitive abilities and having mild cognitive impairment increases the risk of AD or other forms of dementia) and/or AD, this does not necessarily mean a decreased cerebral glucose metabolism. Finally, the effects of lipoic acid in stabilizing glucose metabolism emphasize its utility as an insulin-mimetic agent with multidimensional effects that need to be assessed in a major clinical study.

Footnotes

The authors declare no conflict of interest.

ACKNOWLEDGMENTS

We thank Dr David Carlson (GeroNova Research) for providing the lipoic acid used in this study and Dr Brian D. Ross for his encouragement in initiating these studies.

References

Dubois

Feldman

Jacova

DeKosky

Barberger-Gateau

Cummings

Research criteria for the diagnosis of Alzheimer's disease: revising the NINCDS-ADRDA criteria. Lancet Neurol 2007; 6:734–746.

Gottfries

Adolfsson

Aquilonius

Carlsson

Eckernas

Nordberg

Biochemical changes in Dementia disorders of Alzheimer type (AD/SDAT). Neurobiol Aging 1983; 4:261–271.

Mosconi

Brain glucose metabolism in the early and specific diagnosis of Alzheimer's disease. FDG-PET studies in MCI and AD. Eur J Nucl Med Mol Imaging 2005; 32:486–510.

Farkas

Luiten

Cerebral microvascular pathology in aging and Alzheimer's disease. Prog Neurobiol 2001; 64:575–611.

Lin

Shic

Enriquez

Ross

. Reduced glutamate neurotransmission in patients with Alzheimer's disease-an in vivo ¹³C magnetic resonance spectroscopy study. MAGMA 2003; 16:29–42.

Moats

Ernst

Shonk

Ross

Abnormal cerebral metabolite concentrations in patients with probable Alzheimer disease. Magn Reson Med 1994; 32:110–115.

Götz

Ittner

. Animal models of Alzheimer's disease and frontotemporal dementia. Nat Rev Neurosci 2008; 9:532–544.

Oddo

Caccamo

Shepherd

Murphy

Golde

Kayed

Tripletransgenic model of Alzheimer's disease with plaques and tangles: intracellular A-beta and synaptic dysfunction. Neuron 2003; 39:409–421.

Sancheti

Kanamori

Patil

Brinton

Ross

Cadenas

Reversal of metabolic deficits by lipoic acid in a triple transgenic mouse model of Alzheimer's disease: a ¹³C NMR study. J Cereb Blood Flow Metab 2014; 34:288–296.

10.

Cho

Moini

Shon

Chung

Packer

Alpha-lipoic acid decreases thiol reactivity of the insulin receptor and protein tyrosine phosphatase 1B in 3T3-L1 adipocytes. Biochem Pharmacol 2003; 66:849–858.

11.

Yaworsky

Somwar

Ramlal

Tritschler

Klip

Engagement of the insulinsensitive pathway in the stimulation of glucose transport by α-lipoic acid in 3T3-L1 adipocytes. Diabetologia 2000; 43:294–303.

12.

Sancheti

Akopian

Yin

Brinton

Walsh

Cadenas

Age-dependent modulation of synaptic plasticity and insulin mimetic effect of lipoic acid on a mouse model of Alzheimer's disease. PLoS One 2013; 8:e69830.

13.

Taylor

McLean

Morris

Bachelard

Approaches to studies on neuronal/ glial relationships by ¹³C-MRS analysis. Dev Neurosci 1996; 18:434–442.

14.

Marin-Valencia

Good

Jeffrey

Malloy

Pascual

. High-resolution detection of ¹³C multiplets from the conscious mouse brain by ex vivo NMR spectroscopy. J Neurosci Methods 2012; 203:50–55.

15.

Fitzpatrick

Hetherington

Behar

Shulman

. The flux from glucose to glutamate in the rat brain in vivo as determined by 1H-observed, ¹³C-edited NMR spectroscopy. J Cereb Blood Flow Metab 1990; 10:170–179.

16.

Sonnewald

Therrien

Butterworth

. Portacaval anastomosis results in altered neuron-astrocytic metabolic trafficking of amino acids: evidence from 13C-NMR studies. J Neurochem 1996; 67:1711–1717.

17.

Preece

Cerdán

Metabolic precursors and compartmentation of cerebral GABA in vigabatrin-treated rats. J Neurochem 1996; 67:1718–1725.

18.

Blüml

Hwang

J-H

Moreno

Ross

. Novel peak assignments of in vivo ¹³C MRS in Human Brain at 1.5 T. J Magn Reson 2000; 143:292–298.

19.

Nilsen

Rae

Ittner

Götz

Sonnewald

Glutamate metabolism is impaired in transgenic mice with tau hyperphosphorylation. j Cereb Blood Flow Metab 2013; 33:684–691.

20.

Sibson

Mason

Shen

Cline

Herskovits

Wall

In vivo ¹³C NMR measurement of neurotransmitter glutamate cycling, anaplerosis and TCA cycle flux in rat brain during [2-¹³C]glucose infusion. J Neurochem 2001; 76: 975–989.

21.

Walls

Eyjolfsson

Smeland

Nilsen

Schousboe

Knockout of GAD65 has major impact on synaptic GABA synthesized from astrocyte-derived glutamine. J Cereb Blood Flow Metab 2010; 31:494–503.

22.

Morken

Brekke

Håberg

Widerøe

Brubakk

Sonnewald

Neuron–Astrocyte Interactions, Pyruvate Carboxylation and the Pentose Phosphate Pathway in the Neonatal Rat Brain. Neurochem Res 2014; 39:556–569.

23.

Kondziella

Brenner

Eyjolfsson

Markinhuhta

Carlsson

Sonnewald

Glial-neuronal interactions are impaired in the schizophrenia model of repeated MK801 exposure. Neuropsychopharmacology 2005; 31:1880–1887.

24.

Kanamori

Ross

. The first in vivo observation of ¹³C-¹⁵N coupling in mammalian brain. J Magn Reson 2001; 153:193–202.

25.

Sibson

Dhankhar

Mason

Behar

Rothman

Shulman

. In vivo ¹³C NMR measurements of cerebral glutamine synthesis as evidence for glutamate—glutamine cycling. Proc Natl Acad Sci USA 1997; 94:2699–2704.

26.

Packer

Cadenas

Lipoic acid: energy metabolism and redox regulation of transcription and cell signaling. J Clin Biochem Nutr 2011; 48:26–32.

27.

Luo

Rustay

Ebert

Hradil

Cole

Llano

Characterization of 7- and 19-month-old Tg2576 mice using multimodal in vivo imaging: limitations as a translatable model of Alzheimer's disease. Neurobiol Aging 2012; 33:933–944.

28.

Poisnel

Hérard

El Tannir El Tayara

Bourrin

Volk

Kober

Increased regional cerebral glucose uptake in an APP/PS1 model of Alzheimer's disease. Neurobiol Aging 2012; 33:1995–2005.

29.

Puzzo

Privitera

Leznik

Fà

Staniszewski

Palmeri

Picomolar amyloid-β positively modulates synaptic plasticity and memory in hippocampus. J Neurosci 2008; 28:14537–14545.

30.

Rojas

Herance

Gispert

Abad

Torrent

Jiménez

In vivo evaluation of amyloid deposition and brain glucose metabolism of 5XFAD mice using positron emission tomography. Neurobiol Aging 2013; 34:1790–1798.

31.

Busche

Eichhoff

Adelsberger

Abramowski

Wiederhold

Haass

Clusters of hyperactive neurons near amyloid plaques in a mouse model of Alzheimer's Disease. Science 2008; 321:1686–1689.

32.

Amatniek

Hauser

DelCastillo-Castaneda

Jacobs

Marder

Bell

Incidence and predictors of seizures in patients with Alzheimer's Disease. Epilepsia 2006; 47:867–872.

33.

Edwards-Lee

Ringman

Chung

Werner

Morgan

Hyslop

An African American family with early-onset Alzheimer disease and an APP (T714I) mutation. Neurology 2005; 64:377–379.

34.

Marcon

Giaccone

Cupidi

Balestrieri

Beltrami

Finato

Neuropathological and clinical phenotype of an italian alzheimer family with M239V mutation of presenilin 2 Gene. J Neuropathol Exp Neurol 2004; 63:199–209.

35.

Danbolt

Glutamate uptake. Prog Neurobiol 2001; 65:1–105.

36.

Lipton

Paradigm shift in neuroprotection by NMDA receptor blockade: memantine and beyond. Nat Rev Drug Discov 2006; 5:160–170.

37.

Knight

Verkhratsky

Luckman

Allan

Lawrence

Hypermetabolism in a triple-transgenic mouse model of Alzheimer's disease. Neurobiol Aging 2012; 33: 187–193.

38.

Adebakin

Bradley

Gümüsgöz

Waters

Lawrence

Impaired satiation and increased feeding behaviour in the triple-transgenic Alzheimer's disease mouse model. PLoS One 2012; 7:e45179.

39.

Knight

Brown

Gümüsgöz

Smith

Waters

Allan

Age-related changes in core body temperature and activity in triple-transgenic Alzheimer's disease (3×TgAD) mice. Dis Model Mech 2013; 6:160–170.

40.

Melø

Sonnewald

AinaBastholm

Nehlig

Astrocytes may play a role in the etiology of absence epilepsy: A comparison between immature GAERS not yet expressing seizures and adults. Neurobiol Aging 2007; 28:227–235.