Direct neuronal glucose uptake heralds activity-dependent increases in cerebral metabolism

Abstract

Metabolically, the brain is a highly active organ that relies almost exclusively on glucose as its energy source. According to the astrocyte-to-neuron lactate shuttle hypothesis, glucose is taken up by astrocytes and converted to lactate, which is then oxidized by neurons. Here we show, using two-photon imaging of a near-infrared 2-deoxyglucose analogue (2DG-IR), that glucose is taken up preferentially by neurons in awake behaving mice. Anaesthesia suppressed neuronal 2DG-IR uptake and sensory stimulation was associated with a sharp increase in neuronal, but not astrocytic, 2DG-IR uptake. Moreover, hexokinase, which catalyses the first enzymatic steps in glycolysis, was highly enriched in neurons compared with astrocytes, in mouse as well as in human cortex. These observations suggest that brain activity and neuronal glucose metabolism are directly linked, and identify the neuron as the principal locus of glucose uptake as visualized by functional brain imaging.

Figure 1. Preferential neuronal uptake of glucose and glucose analogues in vitro

(A) Immunocytochemistry of NeuN (grey), GFAP (red) and DAPI (blue) in neuronal cultures from wild type mice (left) and astrocyte cultures from GLT1-eGFP mice (right). Insert: GFAP (left) and NeuN (right) staining signals alone showed no staining signals. Scale bar, 50 µm. (B) ¹⁴C-glucose uptake and (C) ¹⁴C-2DG and 2DG-IR uptake in neurons and astrocytes plotted as a function of time. Slopes for initial linear phase ¹⁴C-glucose neurons, 1.15 ± 0.06; ¹⁴C-glucose astrocytes, 0.20 ± 0.01; ¹⁴C-2DG neurons, 1.39 ± 0.03; ¹⁴C-2DG astrocytes, 0.23 ±0.01 (nmol/mg protein/min); 2DG-IR neurons, 245.00 ± 30.52; 2DG-IR astrocytes, 28.17 ± 1.41 (in arbitrary units, au/min). (D) Substrate competition of ¹⁴C-2DG and 2DG-IR uptake plotted as a function of increasing concentration of D-glucose (Ki in mM: ¹⁴C-2DG neurons, 0.41 ± 0.01, 2DG-IR neurons, 0.87 ± 0.04, ¹⁴C-2DG astrocytes, 0.71 ± 0.0 (E) Effect of increasing concentrations of the nonspecific glucose transporter inhibitor, cytochalasin B, on ¹⁴C-glucose uptake (Ki in µM: ¹⁴C-glucose neurons, 2.58 ± 0.11, ¹⁴C-glucose astrocytes, 4.39 ± 0.03). (F) The effect of cytochalasin B effect on ¹⁴C-2DG and 2DG-IR uptake (Ki in µM: ¹⁴C-2DG neurons, 2.59 ± 0.04, 2DG-IR neurons, 3.02 ± 0.23; ¹⁴C-2DG astrocytes, 4.64 ± 0.06; 2DG-IR astrocytes, 1.66 ± 0.06). The cultures were pretreated with cytochalasin B for 10 min prior to addition of ¹⁴C-glucose, ¹⁴C-2DG, or 2DG-IR. All values are expressed as mean ± s.e.m.

Figure 2. In vivo imaging of cellular uptake of a near-infrared glucose analogue, 2DG-IR

(A) Diagram illustrating the entry route of 2DG-IR via the peri-arterial space. 2DG-IR can be delivered either to the CSF via injection in cisterna magna or via micro-injection in the peri-arterial space surrounding cortical penetrating arteries. The CSF influx pathway is similar to vascular delivery of glucose, since glucose must, after crossing the blood brain barrier (BBB), pass the peri-vascular space prior to gaining access to neurons and astrocytes. (B) Representative high magnification in vivo images at serial depths of 2DG-IR in peri-arterial space surrounding penetrating arteriole at 5 min after injection (N=8). (C) 2DG-IR is transported along the peri-vascular space around both arteries and capillaries followed by diffusion into the tissue 5 min after injection (white arrows) (N=8). (D) Upper panels 2DG-IR uptake in a CamKII-EGFP neuronal reporter mouse showing that the highest 2DG-IR signal co-localizes with CamKII-EGFP⁺ neuronal cells bodies (white arrows) (N=3). Lower panels In contrast, 2DG-IR uptake in GLT1-EGFP⁺ astrocytes (green arrows) is either lower or comparable to surrounding neuropil (N=8). Insert Higher magnification of the capillary wall showing that the high 2DG-IR signal is localized in the peri-capillary space and surrounded by EGFP⁺ vascular endfeet of astrocytes. Scale bars, 10 µm in C–D.

Figure 3. Quantitative comparison of neuronal and astrocytic uptake of 2DG-IR

(A) Diagram illustrating entry of 2DG-IR into the brain via peri-arterial routes following CSF injection in GFAP-eGFP reporter mice. After 30 min circulation of 2DG-IR, the animals are perfusion-fixed with 4% paraformaldehyde and 100 µm thick vibratome sections prepared. Before imaging, the slices are washed repeatedly to remove the glucose analogue not trapped in the cytosol and the nuclei were stained with yellow-Hoechst. (B) Neurons were identified in the freshly prepared vibratome sections by their large round nuclei and absence of cytosolic eGFP (10.2 ± 0.7 µm diameter). Astrocytes were recognized by their smaller nuclei and eGFP signal (7.1 ± 0.1 µm diameter, P=0.01, t-test. N=3 mice). Scale bars, 10 µm. (C) Representative images of 2DG-IR uptake by neurons and astrocytes in cortex, striatum, CA1 and CA3 in awake GLT1-eGFP mice. Nuclei were stained with yellow-Hoechst (white). Scale bars, 20 µm. (D) Quantification of the ratio of 2DG-IR uptake in neurons vs astrocytes. P<0.001 in cortex and striatum, p=0.002 in CA1, P=0.006 in CA3, paired t test, neurons compared to astrocytes (N=9, 6, 8, 7 for cortex, striatum, CA1 and CA3, respectively). Bar graphs represent mean ± s.e.m.

Figure 4. Anesthesia preferentially reduces neuronal 2DG-IR uptake

(A) Left: Representative ¹⁴C-2DG autoradiographs of awake and anesthetized mice. Right: Quantifications of glucose consumption based on ¹⁴C-2DG autoradiography in cortex of awake and ketamine/xylazine anesthetized mice. P=0.001, t-test (N=5). (B) Left: Imaging of 2DG-IR in cortex of awake and mice anesthetized with ketamine, 100 mg/kg and xylazine, 10 mg/kg, i.p (KX). Scale bars, 20 µm. Right: Anesthesia reduced the ratio of neuronal vs. astrocytic 2DG-IR uptake in cortex. P=0.002, Mann-Whitney test, N=9 awake, N=6, anesthetized. Bar graphs represent mean ± s.e.m.

Figure 5. Functional activation increases neuronal uptake, but not astrocytic uptake of 2DG-IR

(A) Upper panel: Diagram of experimental setup. Whisker stimulation (air puffs, 3Hz, 30 min) induced an increase in uptake of 2DG-IR and ¹⁴C-2DG in contralateral somatosensory barrel cortex (S1BF). Lower panel: Local field potential (LFP) recording in contralateral cortex 1 min before, 1, 15 and 29 min into the stimulation and 1 min after whisker stimulation. (B) Representative images of 2DG-IR fluorescence and (C) ¹⁴C-2DG autoradiography of whisker stimulated mice. White dotted lines outline hippocampus. Contralateral cortex was 19 ± 4% higher than ipsilateral cortex. P<0.01, paired t test (N=5). (D) Quantification of 2DG-IR fluorescence in contralateral and ipsilateral barrel cortex in mice with whisker stimulation. P=0.003, paired t-test (N=7). Representative images of 2DG-IR uptake in neurons and astrocytes in (E) ipsilateral and (F) contralateral cortex of whisker stimulated mice. Scale bars, 20 µm. Right panel: Close-up images of 2DGIR in neurons and astrocytes in ipsi- and contralateral cortex with and without eGFP and yellow-Hoechst signals. (G) Quantification of 2DG-IR fluorescence in neurons relative to astrocytes in the respective regions. P=0.006, t test (N=7 contralateral and N=9 ipsilateral). Bar graphs represent mean ± s.e.m.

Figure 6. Neurons express high levels of hexokinases

(A) Left: Diagram of FACS sorting strategy for qPCR of HK1-3. Right: qPCR analysis of hexokinase (HK) 1–3 expression in FACS sorted cortical neurons (tdTomato⁺ vs tdTomato⁻ cell population harvested from Camk2a-CreERT2/CAG-tdTomato reporter mice) and astrocytes (eGFP⁺ vs eGFP⁻ cell populations harvested from GLT1-eGFP reporter mice) expressed as log2 of the ratio of positive vs negative population. P<0.05 compared to negative populations for HK1-3 in neurons and HK3 in astrocytes (N=3 mice in each group). (B–C) Immunohistochemistry of HK1 (red) in neurons (NeuN⁺, grey arrow) and astrocytes (green arrow, eGFP⁺) with DAPI (blue) in cortex of a GLT1-eGFP reporter mouse. Scale bar, 50 µm. (D) Orthogonal projections of a 0.1 µm step size Z-stack. Green arrow indicates an astrocyte endfoot. (E) Quantification of HK1 immunoflourescence shown in percentage difference in HK1 immunolabeling intensity relative to cell-free parenchyma in the cytosol of NeuN⁺ neurons (grey) and GLT1-eGFP⁺ astrocytes (green) in cortex. P<0.01, t-test (40 neurons and 40 astrocytes from N=4 mice). (F–H) Immunohistochemistry of HK1 and NeuN in GLT1-eGFP mouse and (I,J,K) percentage difference in immunolabeling intensity relative to cell-free parenchyma in striatum, CA1 and CA3, respectively. Scale bars, 50 µm in F–H, P<0.001 for striatum, P<0.05 for CA1 and CA3, t test (40 neurons and 40 astrocytes in N=4 mice). Bar graphs represent mean ± s.e.m.

Figure 7. Neurons express high levels of hexokinase 2 and 3

(A) Immunohistochemistry of HK2 (red) in neurons (NeuN⁺, grey arrow) and astrocytes (green arrow, eGFP⁺) with DAPI (blue) of a GLT1-eGFP reporter mouse from the surface of cortex through to the corpus callosum. Scale bar, 50 µm. (B) HK2 (red) and NeuN (white) immunohistochemistry in GLT1-eGFP mouse cortex. Scale bar, 50 µm. (C) Percentage difference in HK2 immunolabeling intensity relative to cell-free parenchyma in the cytosol of NeuN⁺ neurons (grey) and GLT1-eGFP⁺ astrocytes (green) in cortex (P<0.001), striatum (P<0.05), CA1 (P<0.05) and CA3 (P<0.01), t test (30–50 neurons and 30–50 astrocytes in N=4 mice). (D) HK3 (red) and NeuN (white) immunohistochemistry in GLT1-eGFP mouse cortex. Scale bar, 50 µm. (E) Percentage difference in HK3 immunolabeling intensity relative to cell-free parenchyma in the cytosol of NeuN⁺ neurons (grey) and GLT1-eGFP⁺ astrocytes (green) in cortex, striatum, CA1 and CA3. P<0.01, t test (35–50 neurons and 35–50 astrocytes in N=4 mice). Bar graphs represent mean ± s.e.m.

Figure 8. Correlation of hexokinase with glucose consumption and its expression in human brain

(A) Hexokinase 1 (HK1) immunohistochemistry of mouse brain. (B) Glucose consumption calculated from ¹⁴C-2DG signal in cortex, striatum, thalamus, and corpus callosum plotted against HK1 immunoreactivity in the same area. Correlation coefficient R²=0.85 (N=5 for HK1 quantification, N=5 for autoradiography). (C) The ratio of neuronal vs astrocytic 2DG-IR uptake plotted against HK1 immunoreactivity in cortex, striatum, hippocampal CA1 and CA3 regions. Correlation coefficient R²=0.78 (N=5 for HK1 quantification, N=6–9 for 2DG-IR uptake). (D) Orthogonal projections of a 0.1 µm step size Z-stack and image of HK1 (red) and NeuN (grey) immunohistochemistry of adult human cortex. (E) Immunohistochemistry of HK1 (red) and GFAP (green) in adult human cortex. Scale bars in D–E, 10 µm.