Characterizing region-specific glucose metabolic profile of the rodent brain using Seahorse XFe96 analyzer

Abstract

The brain is highly complex with diverse structural characteristics in accordance with specific functions. Accordingly, differences in regional function, cellular compositions, and active metabolic pathways may link to differences in glucose metabolism at different brain regions. In the current study, we optimized an acute biopsy punching method and characterized region-specific glucose metabolism of rat and mouse brain by a Seahorse XFe96 analyzer. We demonstrated that 0.5 mm diameter tissue punches from 180-µm thick brain sections allow metabolic measurements of anatomically defined brain structures using Seahorse XFe96 analyzer. Our result indicated that the cerebellum displays a more quiescent phenotype of glucose metabolism than cerebral cortex, basal ganglia, and hippocampus. In addition, the cerebellum has higher AMPK activation than other brain regions evidenced by the expression of pAMPK, upstream pLKB1, and downstream pACC. Furthermore, rodent brain has relatively low mitochondrial oxidative phosphorylation efficiency with up to 30% of respiration linked to proton leak. In summary, our study discovered region-specific glucose metabolic profile and relative high proton leak coupled respiration in the brain. Our study warrants future research on spatial mapping of the brain glucose metabolism in physiological and pathological conditions and exploring the mechanisms and significance of mitochondrial uncoupling in the brain.

Keywords

Brain glucose hippocampus metabolism respiration

Introduction

The mammalian brain is the most expensive organ in terms of energy expenditure in the whole body with fine regulatory mechanisms to ensure adequate energy substrates supply in register with neuronal activity. The brain uses glucose as the primary fuel for energy production predominately through mitochondrial oxidative phosphorylation.¹ Nonetheless, glucose metabolism is not of homogeneity in the whole brain. The brain is highly complex with diverse structural characteristics in accordance with specific functions. Accordingly, differences in regional cellular compositions, axonal and dendritic density, neurotransmitter distribution, and active metabolic pathways, may link to difference in neuronal function and glucose metabolism for different brain regions.^2,3 A recent study using imaging mass spectrometry demonstrated that some of the glucose metabolism enzymes and ATP level vary dramatically cross the brain.⁴ Disruption of glucose metabolism forms the pathophysiological basis for many brain disorders.^5–7 Therefore, the brain spatial metabolic signatures are of high relevance in our understanding of the normal brain physiology and neuropathology of neurological diseases.

Extracellular flux assay by the Seahorse XFe24 and XFe96 analyzers has provided a technique for simultaneous measurement of respiration and glycolysis in 24 and 96-well plates.⁸ The dynamic measure of oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) enables characterization of metabolic parameters in cultured cells and isolated mitochondria. However, the cultured cells do not recapitulate the complex neuron-glia interaction and extracellular matrix in the brain. The purified mitochondria from brain tissue provide neither intracellular or extracellular environment nor spatial metabolic measurement due to the relatively large amount of tissue required for mitochondrial isolation. Recently, methods have been developed to assess respiration of acute tissue biopsy from anatomically defined rodent brain regions by the Seahorse XFe24 and XFe96 analyzers.^9–12In the current study, we modified the acute biopsy punching procedure and characterized region-specific glucose metabolic profile of rat and mouse brain by the Seahorse XFe96 analyzer.

Materials and methods

Animals

Sprague Dawley male and female rats (3-month-old) were purchased from the Charles River Laboratories (Hollister, CA). C57BL/6J female mice (3-month-old) were purchased from the Jackson Laboratory (Bar Harbor, ME). All animals were kept in independent ventilating cages with access to laboratory chow and water ad libitum under a fixed 12:12 light-dark cycle. All animal procedures were reviewed and approved by the University of North Texas Health Science Center Institutional Animal Care and Use Committee and followed EU Directive 2010/63/EU Guide for the Care and Use of Laboratory Animals.

Preparation of acute brain slices

Animals were anesthetized using isoflurane inhalation and brains were immediately removed and placed in ice-cold oxygenated artificial cerebrospinal fluid (aCSF; in mmol/L: 120 NaCl, 3.5 KCl,1.3 CaCl₂, 1 MgCl₂, 0.4 KH₂PO₄, 5 HEPES, 10 D-glucose, pH 7.4). The aCSF was oxygenated by 95% O₂ with 5% CO₂ at least one-hour prior to brain dissection. Agarose (4%) was used to make gel block and brains were super-glued to the block and then fixed to the tissue holder of a vibratome (VT1000S, Leica, Nussloch, Germany) filled with oxygenated ice-cold aCSF. Coronal and sagittal sections (180 μm thickness) were obtained from cerebrum and cerebellum, respectively. The sections were transferred to a chamber containing aCSF with continuous oxygenation at room temperature.

Tissue punching and Seahorse XFe96 analysis

Brian sections were transferred to a 60 mm cell culture dish containing oxygenated aCSF at room temperature. Reusable biopsy punch (0.5 mm or 0.75 mm diameter, World Precision Instrument, FL, USA) was used to obtain tissue punches from different regions of the cerebral and cerebellar sections under a Zeiss surgical microscope (Carl Zeiss, Germany). The punch was gently pressed overtop the area of interest in the brain section and the obtained tissue punch was immediately injected into a poly-L-lysine (Sigma-Aldrich, USA) coated XFe96 Cell Culture Microplate (Agilent Technologies, CA) based on pre-defined layout. Each well contained 180 μl aCSF with 1.0 mM pyruvate solution as assay media. After loading the 96-well plate, punches were manually manipulated to position in the bottom center of each well. For each Seahorse assay, brain punches derived from the same animal were seeded in a 96-well plate with n = 12 to 16 for each group. The XFe96 Cell Culture Microplate was then transferred to a non-CO2 incubator at 37 °C for 30 min. During this period, 10X concentration of each drug for assay (prepared in aCSF) was loaded in each injection port (A, B, C, D) of the Seahorse XFe96 Assay sensor cartridge that had been hydrated overnight and exchanged for XF Calibrant solution 3 hours prior to assay initiation. The drug-filled sensor cartridge was inserted into the Seahorse XFe96 Analyzer (Agilent Technologies, CA) for calibration. Once the calibration finished, the calibration plate was replaced by the tissue-containing microplate to initiate the assay.

Assay protocol contained 4-cycle measurement of OCR/ECAR baseline. Four to twelve cycles measurement was made after injection of oligomycin (Sigma Aldrich, USA), gboxin, or genipin (Cayman Chemical, MI, USA). Four cycles measurement was made after injection of carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP) supplied with pyruvate (Sigma Aldrich, USA). Four cycles measure were finally made after injection of rotenone (EMD Millipore, Germany)/antimycin A (Sigma Aldrich, USA). Each cycle consists of 2-min mix, 1-min wait and 3-min measure period. All the concentrations indicated are final concentration in the well after injection.

Immunohistochemistry and confocal microscopy

Young adult female rats were sacrificed by anesthesia and cardiac perfusion with ice-cold normal saline. The brains were harvested and fixed in 4% paraformaldehyde and processed for paraffin embedding. Coronal and sagittal brain sections (7-µm thick) were obtained from the cerebrum and cerebellum, respectively, for immunostaining. Sections were blocked with 0.2% Triton-X in Superblock Blocking Buffer™ (Thermo Scientific, USA) for 20 min at room temperature after de-paraffinization and antigen retrieval. Sections were then washed and incubated with primary antibodies (NeuN: Mouse mAb, 1:150 dilutions, catalog #MAB377, Miillipore; GFAP, Rabbit mAb, 1:250 dilutions, catalog #12389, Cell Signaling Technology) at 4 °C overnight followed by 60-min incubation of Alexa Fluor conjugated secondary antibodies (Thermo Fisher Scientific, USA). The sections were mounted with ProLong™ Gold antifade reagent with DAPI (Invitrogen, USA). Z-stack microscopic images were obtained by Zeiss LSM 510 META confocal microscope (Carl Zeiss, USA). Images were processed for maximum intensity Z-projection by ImageJ software (NIH, USA).

Western blots

Young adult mice were euthanized and perfused with ice-cold normal saline. The cerebral cortex, hippocampus, and cerebellum were dissected and protein extractions were obtained for Western blots carried out as previously detailed,¹³ using following primary antibodies: β3-Tubulin (D71G9) (Rabbit mAb, 1:1000 dilutions, catalog #5568, Cell Signaling Technology); Aldehyde dehydrogenase 1 family, member L1 (ALDH1L1, Rabbit Ab, 1:1000 dilutions, catalog #ab87117, Abcam); GFAP (D1F4Q) (Rabbit mAb, 1:1000 dilutions, catalog #12389, Cell Signaling Technology); GAPDH (Mouse mAb,1:5000 dilutions, catalog #sc-3223, Santa Cruz Technology); Phospho-AMPKα (Thr172) (Rabbit mAb, 1:1000 dilutions, catalog #4188, Cell Signaling Technology); AMPKα (23A3) (Rabbit mAb, 1:1000 dilutions, catalog #2603, Cell Signaling Technology); Phospho-Acetyl-CoA Carboxylase (Ser79) (Rabbit Ab, 1:1000 dilutions, catalog #3661, Cell Signaling Technology); Acetyl-CoA Carboxylase (C83B10) (Rabbit Ab, 1:1000 dilutions, catalog #3676, Cell Signaling Technology); Phospho-LKB1 (Ser428) (C67A3) (Rabbit Ab, 1:1000 dilutions, catalog #3482, Cell Signaling Technology); LKB1 (D60C5F10) (Rabbit Ab, 1:1000 dilutions, catalog #13031, Cell Signaling Technology). The membrane was then incubated by HRP-conjugated secondary antibodies (The Jackson ImmunoReseach Laboratories). Chemiluminescence was detected with Biospectrum 500 UVP imaging system. Protein band density was quantified by ImageJ (NIH, USA) and normalized to GAPDH.

Data collection and statistical analysis

The animal data reporting of the current study has followed the ARRIVE 2.0 guidelines.¹⁴ Raw data of OCR and ECAR curve for each well were complied with Agilent’s Wave 2.6.0 software. Wells with detached or broken tissue punch observed after assay were excluded from analysis. The basal respiration was measured as OCR in the absence of mitochondria inhibitors. The residual OCR values and OCR reduction after oligomycin or gboxin injection was used as proton leak-linked respiration and ATP-production coupled respiration, respectively. The OCR measurement after FCCP/pyruvate injection without oligomycin treatment was used as maximal respiration. The difference between maximal OCR and basal OCR was considered as spare respiration capacity. OCR measurement after rotenone/antimycin A (Rot/AA) injection represented non-mitochondria respiration (Figure 1(a)). ECAR was used for measure of glycolysis rate. The baseline ECAR was obtained at the starting assay conditions, while the stressed ECAR was obtained in the presence of stressor mix (oligomycin + FCCP). The metabolic phenotype showed the baseline OCR/ECAR of punches, as well as the maximal OCR (obtained after FCCP injection) and ECAR (after oligomycin/FCCP injection). In each figure (panel), data were derived from the brain punches from the same animal and seeded in the same plate. Each experiment was repeated at least 3 times under the same condition from different animals.

Figure 1.

Metabolic analysis of acute tissue punches (0.5 mm diameter) from rodent brain sections (180 µm thick). (a) Schematic diagram depicts the step of brain harvesting, brain sectioning, tissue punching and seeding in 96-well XFe plate for Seahorse extracellular flux analysis (Created with BioRender.com). (b, c, d, e, f) Metabolic analysis of cerebral cortex (CX), basal ganglia (BG), and cerebellum (CE) punches from a 3-month-old female rat using Seahorse XF96 analyzer. (b) Continuous tracings of oxygen consumption rate (OCR) of CX, BG, and CE punches before and after inhibition of mitochondrial complex V (oligomycin, oligo), addition of mitochondrial oxidative phosphorylation uncoupler FCCP, and inhibition of mitochondrial complexes I/III (rotenone/antimycin A) (Rot/AA) in the same 96-well plate. Non-oligomycin groups (no oligo) were included to obtain the maximal OCR. (c) Scatter dot plots depict basal, maximal respiration, and non-mitochondria respiration (Non-MOC) of each brain region. (d) Continuous tracing of extracellular acidification rate (ECAR) of CX, BG, and CE punches before and after each treatment in the same 96-well plate. (e) Scatter dot plot depicts baseline and stressed ECAR of each brain region. (f) Metabolic profile of CX, BG, and CE from the same assay. Final concentrations: oligomycin 5 μM, FCCP 1 μM/pyruvate 0.75 mM, rotenone 1 μM/antimycin A 1 μM. Results of OCR/ECAR curve and scatter dot plots are shown as mean ± SD for normally distributed data, and median with interquartile range for non-normally distributed data. Metabolic profiles are shown as mean ± SEM. n = 4–17, *p < 0.05, **p < 0.01; ***p < 0.001; ****p < 0.0001 in ordinary one-way ANOVA followed by Tukey’s multiple comparison test or Kruskal-Wallis test followed by Dunn’s multiple comparison test.

Data adhering to normal distribution were depicted as mean ± SD and non-normally distributed data were presented as median and interquartile range (being the 25th and 75th percentile) unless otherwise stated. The Shapiro-Wilk test was used to test for normality. Two-tailed Student t-tests or Mann-Whitney U test were used to identify any significant difference when comparing two groups. One-way ANOVA followed by Tukey’s multiple comparison test or Kruskal-Wallis test followed by Dunn’s multiple comparison test were used for multiple groups comparison with one independent variable. The α level was set at 0.05 for all analyses, and p-value <0.05 was considered statistically significance. Statistical analyses were performed using Graph Pad Prism V.7.

Results

Region-specific glucose metabolic profile of the young and old rat brain

To determine the glucose metabolism in the presence of native intracellular and extracellular environment in well-defined anatomical brain structure, we have optimized a method that enables dynamic metabolic function profiling by the Seahorse XFe96 analyzer using acute rat brain tissue biopsy punches modified from previously reports.^11,12 We obtained tissue punches from different brain regions at frontoparietal cortex, basal ganglia, hippocampal CA1, dentate gyrus, and cerebellar cortex for metabolic analysis using a Seahorse XFe96 analyzer (Figure 1(a)).

We compared different size cerebral cortical and cerebellar punches of 0.5 and 0.75 mm in diameter to obtain reliable OCR measurements. We found that the OCR readings of the 0.5 mm diameter cortical and cerebellar punches were within the 20–200 pmol/min range recommended by the manufacturer. In addition, the cerebellum demonstrated a lower basal OCR as compared with the cerebral cortex (Supplemental Figure 1).

We determined the glucose metabolism profile of cerebral cortical, basal ganglia, and cerebellar punches obtained from 3-month-old rats using Seahorse XF Cell Mito Stress Test. Inhibition of mitochondrial complex V by oligomycin has been found to attenuate the maximal OCR response to mitochondrial uncoupler FCCP.¹⁵ Thus, non-oligomycin groups were included to obtain the maximal OCR rate. Consistently, we observed that cerebellar punches exhibited significantly lower basal OCR than cerebral cortical and basal ganglia punches. Data analysis further revealed that the cerebellum was significantly lower than cerebral cortex and basal ganglia in maximal respiration and non-mitochondria respiration (Figure 1(b) and (c)). Similarly, significantly lower basal and stressed ECAR were observed in the cerebellum as compared to basal ganglia and/or cerebral cortex (Figure 1(d) and (e)). Assessment of glucose metabolic phenotype indicated more energetic phenotypes for all groups in response to stress and that the cerebellum displays a more quiescent phenotype than cerebral cortex and basal ganglia evidenced by the lower glycolytic and aerobic activity at the cerebellum than the cerebral cortex and basal ganglia (Figure 1(f)).

We obtained tissue punches from hippocampal dentate gyrus, CA1, and cerebellum of 3-month-old rats for extracellular flux analysis in the same 96-well plate. Our results demonstrated that dentate gyrus exhibited basal OCR values similar to the cerebellum while the basal OCR of the CA1 was within the range of cerebral cortex and basal ganglia (Figure 2(a)). Data analysis further revealed that dentate gyrus and cerebellum were significantly lower than hippocampal CA1 in basal respiration and non-mitochondria respiration. No significant difference was observed in maximal respiration among CA1, dentate gyrus and cerebellum (Figure 2(b)). Dentate gyrus displayed similar basal ECAR as hippocampal CA1, while CA1 was significantly higher than the cerebellum and dentate gyrus in stressed ECAR (Figure 2(c) and (d)). Assessment of glucose metabolic phenotype indicated that the cerebellum had a quiescent phenotype as compared with hippocampal CA1 and dentate gyrus (Figure 2(e)).

Figure 2.

Metabolic analysis of in hippocampal CA1, dentate gyrus (DG), and cerebellum (CE) punches from 3-month-old female rat using a Seahorse XFe96 analyzer. (a) Continuous tracings of OCR of CA1, DG, and CE punches before and after inhibition of mitochondrial complexes and addition of uncoupler in the same 96-well plate. Non-oligomycin groups (no oligo) were included to obtain the maximal OCR. (b) Scatter dot plots depict basal, maximal respiration, and non-mitochondria respiration (Non-MOC) of each brain region. (c) Continuous tracing of ECAR of CA1, DG, and CE punches before and after each treatment in the same 96-well plate. (d) Scatter dot plot depicts baseline and stressed ECAR of each brain region. (e) Metabolic profile of CA1, DG, and CE from same assay. Final concentrations: oligomycin 5 μM, FCCP1 μM/pyruvate 0.75 mM, rotenone (Rot) 1 μM/antimycin A (AA) 1 μM. Results of OCR/ECAR curve and scatter dot plots are shown as mean ± SD for normally distributed data, and median with interquartile range for non-normally distributed data. Metabolic profiles are shown as mean ± SEM. n = 4–19, *p < 0.05, **p < 0.01; ***p < 0.001; ****p < 0.0001 in ordinary one-way ANOVA followed by Tukey’s multiple comparison test or Kruskal-Wallis test followed by Dunn’s multiple comparison test.

We further determined metabolic phenotype in different brain regions of middle-aged rats. Tissue punches were obtained from cerebral cortex, basal ganglia, and cerebellum of 12-month-old rats, and glucose metabolism was assessed by Seahorse XF Cell Mito Stress test in the same 96-well plate. Consistently, the cerebellum displayed lower basal respiration and non-mitochondrial respiration than cerebral cortex and basal ganglia (Supplemental Figure 2a, b). The cerebellum also displayed lower baseline and stressed ECAR as compared with cerebral cortex and basal ganglia (Supplemental Figure 2c, d).

Region-specific glucose metabolic profile of the mouse brain

We conducted extracellular flux analysis in different brain regions from 3-month-old mice to explore the region-specific metabolic phenotype in young mouse brain. Consistently, the cerebellum displayed lower basal OCR and ECAR values compared to cerebral cortex and basal ganglia. The cerebellum was significantly lower than cerebral cortex and basal ganglia in basal respiration, maximal respiration, and non-mitochondrial respiration (Figure 3(a) and (b)). A significantly lower baseline and stressed ECAR was observed in the cerebellum as compared with cerebral cortex and basal ganglia (Figure 3(c) and (d)). Assessment of glucose metabolic phenotype indicated that the cerebellum displayed a more quiescent phenotype with lower glycolytic and aerobic activity than cerebral cortex and basal ganglia (Figure 3(e)).

Figure 3.

Metabolic analysis of cerebral cortex (CX), basal ganglia (BG), and cerebellum (CE) punches from a 3-month-old female mouse using a Seahorse XFe96 extracellular flux analyzer. (a) Continuous tracings of OCR of CX, BG, and CE punches before and after inhibition of mitochondrial complexes and treatment of uncoupler in the same 96-well plate. Non-oligomycin groups (no oligo) were included to obtain the maximal OCR (not shown). (b) Scatter dot plots depict basal, maximal respiration, and non-mitochondria respiration (Non-MOC) of each brain region. (c) Continuous tracing of ECAR of CX, BG, and CE punches before and after each treatment in the same 96-well plate. (d) Scatter dot plot depict baseline and stressed ECAR of each brain region. (e) Metabolic profile of CX, BG, and CE from same assay. Final concentrations: oligomycin 5 μM FCCP 1 μM/pyruvate 0.75 mM, rotenone (Rot) 1 μM/antimycin A (AA) 1 μM. Results of OCR/ECAR curve and scatter dot plots are shown as mean ± SD for normally distributed data, and median with interquartile range for non-normally distributed data. Metabolic profiles are shown as mean ± SEM. n = 6–20, *p < 0.05, **p < 0.01; ***p < 0.001; ****p < 0.0001 in ordinary one-way ANOVA followed by Tukey’s multiple comparison test or Kruskal-Wallis test followed by Dunn’s multiple comparison test.

In separate analysis, dentate gyrus exhibited similar basal respiration, as the cerebellum. Data analysis suggested that the cerebellum was significantly lower than hippocampal CA1 in basal, maximal respiration, and non-mitochondrial respiration, as well as significantly lower than dentate gyrus in maximal respiration and non-mitochondrial respiration (Supplemental Figure 3). Glucose metabolic phenotype assessment implicated a more glycolytic and aerobic activity for hippocampal CA1 as compared to dentate gyrus and cerebellum that both maintained a more quiescent phenotype (Supplemental Figure 3e). Overall, the region-specific metabolic phenotype in cerebral cortex, basal ganglia, hippocampal CA1, dentate gyrus, and cerebellum displayed similar region-specific profile in rat and mouse.

Oxidative phosphorylation efficiency of mouse brain

We conducted titration experiments to select optimal concentrations of oligomycin on OCR. Oligomycin treatment induced a dose-dependent OCR reduction which reach plateau at 5 to 10 μM (Figure 4(a) and (b)). We further determined the ATP production and proton leak-linked respiration in the mouse brain using different concentrations of complex V inhibitors. Oligomycin and gboxin decreased OCR to similar levels in the cerebral cortex and cerebellum at 5 µM concentration. Increase of concentration of oligomycin and gboxin to 10 µM did not provide further inhibition of OCR (Figure 4(c) to (f)). Data analysis suggested that cerebral cortex has significant higher ATP production, proton leak-linked, and non-mitochondrial respiration than the cerebellum. Interestingly, ATP production-linked respiration is about 50% of the basal respiration in both cerebral cortex and cerebellum. Proton leak-linked respiration comprised up to 30% of basal respiration in the cerebral cortex and cerebellum (Figure 4(d) and (f)).

Figure 4.

Optimization of mitochondrial complex V inhibition and measurement of non-mitochondria respiration, ATP production-linked respiration, and proton leak-linked respiration in the cerebral cortex and cerebellum. (a) Continuous tracings of OCR of CX punches from a 3-month-old female mouse before and after inhibition of complex V by different concentration of oligomycin (0, 1, 5, 10, 20, 50 μM). OCR was measured for 12 cycles of 85 minutes after addition of oligomycin. (b) Bar graph showing dose-dependent effect of oligomycin on OCR. Results are mean ± SD. n = 11–14. *p < 0.05, **p < 0.01 in ordinary one-way ANOVA followed by Tukey’s multiple comparison test. (c) Continuous tracings of OCR of CX and CE punches from a 3-month-old mouse before and after inhibition of complex V by oligomycin (5 or 10 µM) (oligo 5/10), addition of FCCP/pyruvate, and rotenone/antimycin A (Rot/AA). (d) Stacked bar graph showing non-mitochondria respiration, ATP production-linked respiration, and proton leak-linked respiration of CX and CE punches. (e) Continuous tracings of OCR of CX and CE punches from 3-month-old female mouse before and after inhibition of complex V by gboxin (5 or 10 µM) (gboxin 5/10), addition of FCCP/pyruvate, and rotenone/antimycin A. (f) Stacked bar graph showing non-mitochondria respiration, ATP production-linked respiration, and proton leak-linked respiration of CX and CE punches. Results are mean ± SD. n = 8-12. *p < 0.05, **p < 0.01; ^***p < 0.001 in either un-paired Student t-test for data or Mann-Whitney U test.

We determined the effect of uncoupling proteins (UCPs) inhibitor, genipin, on mitochondrial proton leak-linked respiration. Acute treatment of genipin at 50 and 100 μM significantly decreased proton leak coupled OCR with minimal effect on ATP production coupled OCR (Figure 5(a) to (c)). In addition, chronic treatment of genipin at 100 and 150 μM in aCSF supplied with pyruvate for 60 min displayed a trend to attenuate the basal OCR and significant reduction of proton leak coupled OCR without impact on ATP production coupled respiration (Figure 5(d) to (f)).

Figure 5.

Inhibition of uncoupling proteins decreased proton leak coupled OCR in cerebral cortex punches. (a) Continuous tracings of OCR of the cerebral cortex punches from a 3-month-old female mouse before and after treatment of UCPs inhibitor (genipin 50, 100 μM), oligomycin 5 μM, FCCP 1 μM/pyruvate 0.75 mM, and rotenone (Rot) 1 μM/antimycin A (AA) 1 μM. (b, c) Bar graphs showing the effect of UCPs inhibition by genipin on ATP production coupled (b) and proton leak coupled (c) respiration. (d, e, f) Bar graphs show the effect of chronic treatment (60 minutes) of genipin on basal OCR (d), ATP production-linked OCR (e), and proton leak-linked OCR in the cerebral cortex punches from a 3-month-old female mouse. Results are mean ± SD. n = 12-15. ^*p < 0.05, ^**p < 0.01; ^***p < 0.001 in ordinary one-way ANOVA followed by Tukey’s multiple comparison test.

Region-specific metabolic signaling in the mouse brain

We determined the cellular components in the brain punches of different brain regions. Immunohistochemistry and confocal microscopy of the corresponding punch area demonstrated distinct brain structures with different cellular composites (Figure 6(a) to (c)). We further identified the neuronal and astrocytic markers and major metabolic signaling expression in different brain regions of 3-month-old mice. Representative Western blots and quantitative analysis indicated a significantly higher neuronal marker of β3-tubulin in cerebral cortex than hippocampus and cerebellum (Figure 7(a) and (d)). The expression of ALDH1L1, an astrocyte marker, was significantly higher in cerebral cortex than hippocampus (Figure 7(a) and (e)), while GFAP expression was higher in hippocampus and cerebellum than in cerebral cortex (Figure 7(a) and (f)). Activation of AMPK signaling was significantly higher in the cerebellum than in cerebral cortex and hippocampus evidenced by higher expression of phospho-AMPKα (Figure 7(b) and (g)), phospho-ACC (Figure 7(b) and (h)), and phospho-LKB1 (Figure 7(c) and (i)). These data indicated that cerebral cortex, hippocampus, and cerebellum displayed a diverse cellular composition and difference in activation of metabolic signaling.

Figure 6.

Immunohistochemistry of GFAP, NeuN, and DAPI depict different cellular components of different brain regions. (a) Coronal section of the cerebrum depicts the location of tissue punches at cerebral cortex and basal ganglia and representative confocal images show neuron and astrocyte components of cerebral cortical and basal ganglia punches. (b) Coronal section of hippocampus depicts the location of tissue punches at hippocampal CA1 and dentate gyrus and representative confocal images show neuron and astrocyte components of CA1 and dentate gyrus punches. (c) Sagittal section of the cerebellum depicts the location of the tissue punches at cerebellar cortex and representative confocal images show neuron and astrocyte components of cerebellar cortex punches. Brain sections were stained by GFAP (green), NeuN (red), and nucleus counterstained by DAPI (blue).

Figure 7.

Cellular composition and metabolic signaling in different mouse brain regions. (a) Western blots of β3Tubulin, GFAP, ALDH1L1, GAPDH at the cerebral cortex (CX), hippocampus (HP), and cerebellum (CE) of 3-month-old mouse. (b) Western blots of AMPKα, pAMPKα, ACC, pACC, GAPDH at CX, HP, CE of 3-month-old mouse. (c) Western blots of LKB1, pLAKB1, GAPDH at CX, HP, CE of 3-month-old mouse. (d-i). Quantitative analysis of β3-Tubulin, ALDH1, GFAP, pAMPKα, pACC, and pLKB1 expression at CX, HP, and CE. Results are mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 in repeated measure for one-way ANOVA followed by Tukey’s multiple comparison.

Discussion

There are three important findings derived from the present study. First, we have optimized a method that enables metabolic function assessment of anatomically defined brain structures by the Seahorse XFe96 analyzer in adult rat and mouse. Second, the rodent brain has region-specific glucose metabolic profile that the cerebellum displays a more quiescent phenotype than cerebral cortex, basal ganglia, and hippocampus. Third, the rodent brain has relatively low mitochondrial oxidative phosphorylation efficiency with high proton leak-linked respiration.

The Seahorse XFe analyzers measure real-time oxygen consumption rate and extracellular acidification rate in a multi-well plate to provide a systemic view of metabolic function in term of basal respiration, ATP production, proton leak, maximal respiration, spare respiratory capacity, and non-mitochondrial respiration in cultured cells and ex-vivo tissue samples. By simultaneously assessing mitochondrial respiration and glycolysis under basal and stressed conditions, it also provides insight into the metabolic phenotype of cells and tissues. The Seahorse XFe analysis of acute brain punches with same thickness, diameter, and weight enables high-resolution spatial mapping of brain glucose metabolism in a relative intact brain microenvironment. In 24-well plate-based assay, 1 mm diameter tissue punches from 250 µm thick rat and mouse brain sections provide reproducible respiratory measurement using Seahorse XFe24 analyzer.^10,12 In 96-well plate based assay, 0.5 and 0.75 mm diameter tissue punches from 220 μm rat brain sections seems are optimal for metabolic flux analysis.¹¹ The height of micro-chamber of XFe96 cell culture microplates is 200 μm. We obtained tissue punches from 180 µm thick brain sections to avoid tissue damage by the compression of the microsensors during the assay. The small brain tissue punches and the using of the Seahorse XFe96 analyzer allow metabolic measurements of anatomically defined brain structures with sufficient sample size to compare metabolic phenotype of different brain regions in the same 96-well plate assay. Our study demonstrated that this method could be used to define glucose metabolic profile of different brain regions at high-resolution in mouse and rat.

Imaging mass spectrometry of mouse brain sections has demonstrated significant regional differences in glucose metabolism enzymes.⁴ The human cerebellum seems are unique in term of energy metabolism with lower metabolic rate for glucose than the cerebrum.¹⁶ In addition, the human cerebellum displays dramatically lower aerobic glycolysis than the cerebral cortex.^17,18 In the current study, we consistently observed lower glucose metabolism in the cerebellum than in other brain regions of young and old rats as well as young adult mice. The low glucose metabolic phenotype in the cerebellum was evidenced by the low rate of basal respiration, maximal respiration, non-mitochondrial respiration, as well as glycolysis. The cerebellum displays a more quiescent glucose metabolic phenotype at baseline as compared with cerebral cortex, basal ganglia, and hippocampal CA1 region. The glucose metabolism at cerebral cortex, basal ganglia, and hippocampal CA1 region shift dramatically toward a higher energetic phenotype under stressed condition. The cerebellum have different energy budget distribution from the neocortex with majority of energy on the maintenance of resting potentials while action potentials account for only small part.¹⁹ Our study indicated that the cerebellum has lower spare respiratory capacity than the cerebrum and that the quiescent state of glucose metabolism in the cerebellum has less transition toward active state upon stress.

The brain consists of heterogeneous cells and each cell type has distinct metabolic phenotype. The region-specific glucose metabolism might be due to the difference in cellular compositions of different brain regions. Oxidative phosphorylation is the main mechanism to power neuronal activity in the brain.²⁰ The cerebellum, although represents only ∼10% of total brain mass, contains ∼80% of the total brain neurons that are mostly organized in the densest granule layer.²¹ Interestingly, the high neuron to glia ratio in the cerebellum is not associated with high glucose metabolism. The granule cells are the most numerous cell type in the cerebellum and dominate the energy use of the cerebellar cortex.¹⁹ Our cerebellar punches contained molecular layer, Purkinje cell layer, and granular layer with granule cells as the predominant cell type. We speculated that the low metabolism rate at the cerebellum was mainly due to the granule cell. Indeed, a similar glucose metabolic profile was observed in the dentate gyrus as in the cerebellum in which granule cell is the principal cell type. The low glucose metabolism of the cerebellar granule cell might be attributed to their small cell body and less dendrites.²² However, the low glucose metabolism of the cerebellar cortex with a large amount of neurons indicates that the cerebellum might have unique mechanism underlying its higher energy efficient than the cerebrum.

The brain is widely recognized as a highly oxidative organ with disproportionately high fraction of oxygen consumption.¹ The coupling of ATP synthesis and substrate oxidation in mitochondria is not complete as proton can return to mitochondrial matrix without producing ATP. The classic oxygen electrode experiments have been used since 1950s to determine mitochondrial bioenergetics function.²³ The state 4 respiration evaluates the oxygen consumption rate of the mitochondrial respiratory chain predominately due to a proton leak. The ratio of the state 3 and state 4, termed as respiratory control ratio, indicates the level of coupling between oxidation and phosphorylation process in the mitochondria.²³ In the permeabilized adult mouse brain tissue, the respiratory control ratio, calculated as the ratio of the state 3 respiration after adding ADP and the state 4 after adding oligomycin, was around 7 in the cerebral cortex, indicating high mitochondrial oxidative phosphorylation efficiency of the brain.²⁴ Nonetheless, the measurement of the state 3_ADP and state 4_oligo respiration may include non-mitochondrial respiration in the permeabilized brain tissue, thus, may not necessarily reflect the mitochondrial oxidative phosphorylation efficiency. In isolated mitochondria, the mitochondrial phosphate/oxygen ratio of rat brain has lower mitochondrial oxidative phosphorylation efficiency than that of heart and liver.²⁵ In the present study, we consistently observed high levels of OCR even after complex V inhibition by oligomycin in different brain regions. We suspected that the less OCR inhibition by oligomycin was potentially due to the low concentration at 5 μM or short measurement period. We conducted a titration experiment to determine the dose dependent effect of oligomycin on OCR with 12-cycle measurement for 85 minutes. Consistently, we observed ∼50% OCR reduction upon oligomycin treatment at the concentration of 5, 10, 20, and 50 µM. Without oligomycin, no spontaneous OCR reduction was observed within 12-cycles measurement. However, the brain punches were less response to FCCP after 12-cycles measurement. In a separate experiment, oligomycin at 10 μM did not provide further OCR reduction after up to 8-cycle measurements. We used another ATP synthase inhibitor, gboxin, to assess the ATP production and proton leak-linked respiration and similar results were observed. Our results provide further evidence that brain glucose metabolism might be less efficient with substantial uncoupling.

Mitochondrial proton leak is mediated mainly through adenine nucleotide translocase and UCPs.²⁶ Inhibition of UCPs has been found to reduce OCR in cell cultures.^27,28 We determined the effect of UCPs inhibition on OCR in cerebral cortex punches. UCPs inhibition by genipin significantly decreased proton leak coupled respiration with minimal impact on ATP production coupled respiration, suggesting that the observed proton leak-linked respiration is partially mediated through UCPs in the brain.

The significance of unexpected mitochondrial proton leak in the brain is unclear. Expression of UCP-2 has been found in the brain to provide neuroprotective effect against ischemic insult.²⁹ Similarly, expression of human UCP-2 in the mitochondria of adult fly neurons increases proton leak-dependent state 4 respiration, decrease ROS production, and extend life span.³⁰ Thus, the mitochondrial proton leak-linked respiration might play important roles in the oxidative defense and survival of the brain cells. UCP1 is well known to be responsible for thermogenesis in brown adipose tissue and body temperature maintenance. There is increasing evidence that brain temperature is significantly higher than arterial blood and that increases in local brain temperature occur in response to various stressful and emotionally arousing environment stimuli although the source of the heat production is not clear.^31–35 The global distribution of UCP2/4/5 in the brain and their abilities to decrease mitochondrial membrane potential in neurons imply that neuronal UCPs activation may lead to heat generation.^36,37 We speculated that the high proton leak may be coupled to heat production and high temperature in the brain.

We expected that the differences in regional glucose metabolism identified by the Seahorse analysis might be associated with different activation of metabolic pathways. AMPK is an evolutionarily conserved energy sensor and regulator for energy metabolism.³⁸ In mammalian adult brain, AMPKs are mainly expressed and constitutively active in neurons, with AMPKα2 as the predominant catalytic subunit.³⁹ AMPK is phosphorylated, hence, activated by two main mammalian upstream kinases: LKB1 and CaMKKβ.⁴⁰ LKB1 activates AMPKα2 but not AMPKα1.^41–43 We observed different activation of AMPK signaling in different brain regions. The lowest glucose metabolism in the cerebellum is associated with the highest AMPK activation than other brain regions evidenced by the expression of pAMPK, upstream pLKB1, and downstream pACC. AMPK is activated by events that either compromise cellular ATP production or increase ATP consumption.^38,44 The high AMPK activation, together with the low rate of basal respiration, maximal respiration, and non-mitochondrial respiration in the cerebellum, indicated that the cerebellum has unique glucose metabolic phenotype as compared with other brain regions.

In summary, the present study determined the region-specific glucose metabolic profile of rodent brain using acute biopsy punches and Seahorse XFe96 analyzer. The metabolic flux analysis indicated that the cerebellum has a more quiescent phenotype of glucose metabolism as compared with the cerebrum. In addition, glucose metabolism might be less efficient in the brain than we expected, with relatively large component of proton leak-linked respiration. Our study warrants future research on spatial mapping of the brain glucose metabolism in physiological and pathological conditions and exploring the potential mechanisms and significance of mitochondrial uncoupling in the brain.

Supplemental Material

sj-pdf-1-jcb-10.1177_0271678X221077341 - Supplemental material for Characterizing region-specific glucose metabolic profile of the rodent brain using Seahorse XFe96 analyzer

Supplemental material, sj-pdf-1-jcb-10.1177_0271678X221077341 for Characterizing region-specific glucose metabolic profile of the rodent brain using Seahorse XFe96 analyzer by Linshu Wang, Kiran Chaudhari, Ali Winters, Yuanhong Sun, Ran Liu and Shao-Hua Yang in Journal of Cerebral Blood Flow & Metabolism

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partly supported by National Institutes of Health grants 1R01NS109583-01 (SY).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Authors’ contributions

LW, AW, KC, YS, RL, and SY conceived and designed the experiments; LW, AW, KC performed the experiments. LW and KC analyzed the data. LW, RL, and YS wrote and edited the manuscript; all the authors reviewed the manuscript before publication.

Supplemental material

Supplemental material for this article is available online.

References

Dienel

GA.

Brain glucose metabolism: Integration of energetics with function. Physiol Rev 2019; 99: 949–1045.

van Aalst

Ceccarini

Sunaert

, et al. In vivo synaptic density relates to glucose metabolism at rest in healthy subjects, but is strongly modulated by regional differences. J Cereb Blood Flow Metab 2021; 41: 1978–1987.

Shokri-Kojori

Tomasi

Alipanahi

, et al. Correspondence between cerebral glucose metabolism and BOLD reveals relative power and cost in human brain. Nat Commun 2019; 10: 690.

Kleinridders

Ferris

Reyzer

, et al. Regional differences in brain glucose metabolism determined by imaging mass spectrometry. Mol Metab 2018; 12: 113–121.

Mergenthaler

Lindauer

Dienel

, et al. Sugar for the brain: the role of glucose in physiological and pathological brain function. Trends Neurosci 2013; 36: 587–597.

Chen

Wei

Chan

, et al. D-Glucose uptake and clearance in the tauopathy Alzheimer's disease mouse brain detected by on-resonance variable delay multiple pulse MRI. J Cereb Blood Flow Metab 2021; 41: 1013–1025.

Grist

Miller

Zaccagna

, et al. Hyperpolarized (13)C MRI: a novel approach for probing cerebral metabolism in health and neurological disease. J Cereb Blood Flow Metab 2020; 40: 1137–1147.

Rogers

Brand

Petrosyan

, et al. High throughput microplate respiratory measurements using minimal quantities of isolated mitochondria. PLoS One 2011; 6: e21746.

Schuh

Clerc

Hwang

, et al. Adaptation of microplate-based respirometry for hippocampal slices and analysis of respiratory capacity. J Neurosci Res 2011; 89: 1979–1988.

10.

Fried

Moffat

Seifert

, et al. Functional mitochondrial analysis in acute brain sections from adult rats reveals mitochondrial dysfunction in a rat model of migraine. Am J Physiol Cell Physiol 2014; 307: C1017–30.

11.

Underwood

Redell

Zhao

, et al. A method for assessing tissue respiration in anatomically defined brain regions. Sci Rep 2020; 10: 13179.

12.

Yin

Characterizing brain metabolic function ex vivo with acute mouse slice punches. STAR Protoc 2021; 2: 100559.

13.

Chaudhari

Wang

Kruse

, et al. Early loss of cerebellar Purkinje cells in human and a transgenic mouse model of Alzheimer's disease. Neurol Res 2021; 43: 570–581.

14.

Percie Du Sert

Hurst

Ahluwalia

, et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. J Cereb Blood Flow Metab 2020; 40: 1769–1777.

15.

Ruas

Siqueira-Santos

Amigo

, et al. Underestimation of the maximal capacity of the mitochondrial electron transport system in oligomycin-treated cells. PLoS One 2016; 11: e0150967.

16.

Bentourkia

Bol

Ivanoiu

, et al. Comparison of regional cerebral blood flow and glucose metabolism in the normal brain: effect of aging. J Neurol Sci 2000; 181: 19–28.

17.

Vaishnavi

Vlassenko

Rundle

, et al. Regional aerobic glycolysis in the human brain. Proc Natl Acad Sci U S A 2010; 107: 17757–17762.

18.

Goyal

Hawrylycz

Miller

, et al. Aerobic glycolysis in the human brain is associated with development and neotenous gene expression. Cell Metab 2014; 19: 49–57.

19.

Howarth

Gleeson

Attwell

Updated energy budgets for neural computation in the neocortex and cerebellum. J Cereb Blood Flow Metab 2012; 32: 1222–1232.

20.

Hall

Klein-Flugge

Howarth

, et al. Oxidative phosphorylation, not glycolysis, powers presynaptic and postsynaptic mechanisms underlying brain information processing. J Neurosci 2012; 32: 8940–8951.

21.

Azevedo

Carvalho

Grinberg

, et al. Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain. J Comp Neurol 2009; 513: 532–541.

22.

Badura

De Zeeuw

CI.

Cerebellar granule cells: dense, rich and evolving representations. Curr Biol 2017; 27: R415–R418.

23.

Brand

Nicholls

DG.

Assessing mitochondrial dysfunction in cells. Biochem J 2011; 435: 297–312.

24.

Arias-Reyes

Losantos-Ramos

Gonzales

, et al. NADH-linked mitochondrial respiration in the developing mouse brain is sex-, age- and tissue-dependent. Respir Physiol Neurobiol 2019; 266: 156–162.

25.

Cocco

Pacelli

Sgobbo

, et al. Control of OXPHOS efficiency by complex I in brain mitochondria. Neurobiol Aging 2009; 30: 622–629.

26.

Divakaruni

Brand

MD.

The regulation and physiology of mitochondrial proton leak. Physiology (Bethesda) 2011; 26: 192–205.

27.

Mailloux

Adjeitey

Harper

ME.

Genipin-induced inhibition of uncoupling protein-2 sensitizes drug-resistant cancer cells to cytotoxic agents. PLoS One 2010; 5: e13289.

28.

Jeong

Yun

, et al. Genipin induces mitochondrial dysfunction and apoptosis via downregulation of Stat3/mcl-1 pathway in gastric cancer. BMC Cancer 2019; 19: 739.

29.

Rubattu

Stanzione

Bianchi

, et al. Reduced brain UCP2 expression mediated by microRNA-503 contributes to increased stroke susceptibility in the high-salt fed stroke-prone spontaneously hypertensive rat. Cell Death Dis 2017; 8: e2891.

30.

Fridell

Sanchez-Blanco

Silvia

, et al. Targeted expression of the human uncoupling protein 2 (hUCP2) to adult neurons extends life span in the fly. Cell Metab 2005; 1: 145–152.

31.

Kiyatkin

Wise

RA.

Striatal hyperthermia associated with arousal: intracranial thermorecordings in behaving rats. Brain Res 2001; 918: 141–152.

32.

Kiyatkin

Wise

RA.

Brain and body hyperthermia associated with heroin self-administration in rats. J Neurosci 2002; 22: 1072–1080.

33.

Kiyatkin

Brown

Wise

RA.

Brain temperature fluctuation: a reflection of functional neural activation. Eur J Neurosci 2002; 16: 164–168.

34.

Brown

Wise

Kiyatkin

EA.

Brain hyperthermia is induced by methamphetamine and exacerbated by social interaction. J Neurosci 2003; 23: 3924–3929.

35.

Kiyatkin

EA.

Brain temperature and its role in physiology and pathophysiology: lessons from 20 years of thermorecording. Temperature (Austin) 2019; 6: 271–333.

36.

Andrews

Diano

Horvath

TL.

Mitochondrial uncoupling proteins in the CNS: in support of function and survival. Nat Rev Neurosci 2005; 6: 829–840.

37.

Rajagopal

Brown

Gelda

, et al. Transient heat release during induced mitochondrial proton uncoupling. Commun Biol 2019; 2: 279.

38.

Hardie

DG.

AMP-activated protein kinase: an energy sensor that regulates all aspects of cell function. Genes Dev 2011; 25: 1895–1908.

39.

Culmsee

Monnig

Kemp

, et al. AMP-activated protein kinase is highly expressed in neurons in the developing rat brain and promotes neuronal survival following glucose deprivation. J Mol Neurosci 2001; 17: 45–58.

40.

Steinberg

Kemp

BE.

AMPK in health and disease. Physiol Rev 2009; 89: 1025–1078.

41.

Imai

Inukai

Ikegami

, et al. LKB1, an upstream AMPK kinase, regulates glucose and lipid metabolism in cultured liver and muscle cells. Biochem Biophys Res Commun 2006; 351: 595–601.

42.

Sakamoto

Zarrinpashneh

Budas

, et al. Deficiency of LKB1 in heart prevents ischemia-mediated activation of AMPKalpha2 but not AMPKalpha1. Am J Physiol Endocrinol Metab 2006; 290: E780–8.

43.

Alessi

Sakamoto

Bayascas

JR.

LKB1-dependent signaling pathways. Annu Rev Biochem 2006; 75: 137–163.

44.

Hardie

Carling

Gamblin

SJ.

AMP-activated protein kinase: also regulated by ADP?

Trends Biochem Sci 2011; 36: 470–477.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.42 MB