Cerebral glucose transporters expression and spatial learning in the K-ATP Kir6.2(-/-) knockout mice.

K-ATP channels formed of the Sur and Kir subunits are widely distributed in the brain. Sur1-Kir6.2 is the most common combination of K-ATP channel subunits in the brain and Kir6.2 plays an important role in glucose metabolism through pancreatic insulin secretion or hypothalamic glucose sensing. K-ATP channels have also been reported to play a role in memory processing. Therefore, the aim of the present experiment is to assess the gene and protein expression of GLUT1, GLUT3 and GLUT4 in various brain regions of Kir6.2(-/-) K-ATP knockout mice and to test their working memory performance. GLUT4 was measured using two antibodies, one recognizing an intracellular epitope and the other, an extracellular epitope. Relative to their corresponding wild type, semi-quantitative immunohistochemistry showed that GLUT4 protein expression as measured by a GLUT4 antibody recognizing an extracellular epitope was increased in the Kir6.2(-/-) K-ATP mice. However, there was only a small increase in GLUT4 labeling using the GLUT4 antibody recognizing the intracellular epitope. These results suggest a compensatory higher GLUT4 inclusion at the cellular neuronal membrane in the cerebral cortex, hippocampus and cerebellum of the Kir6.2(-/-) K-ATP knockout mice. However, there was no change in GLUT4 gene expression assessed by TaqMan PCR except for a decrease in the cerebellum of these mice. Working memory performance of the Kir6.2(-/-) K-ATP mice was disrupted at age of 12 weeks but not at 5 weeks. The mild glucose intolerance that is observed in the Kir6.2 knockout mice is unlikely to have created the memory deficits observed. Rather, in light of the effects of K-ATP channel modulators on memory, the memory deficits in the Kir6.2(-/-) K-ATP mice are more likely due to the absence of the Kir6.2 and possible disruption of the GLUT4 activity in the brain.

Glucose transporter plasticity during memory processing.

Various types of learning, including operant conditioning, induce an increase in cellular activation concomitant with an increase in local cerebral glucose utilization (LCGU). This increase is mediated by increased cerebral blood flow or changes in brain capillary density and diameter. Because glucose transporters are ultimately responsible for glucose uptake, we examined their plastic expression in response to cellular activation. In vitro and in vivo studies have demonstrated that cerebral glucose transporter 1 (GLUT1) expression consistently parallels changes in LCGU. The present study is the first to investigate the effect of memory processing on glucose transporters expression. Changes in GLUT expression produced by training in an operant conditioning task were measured in the brain of CD1 mice. Using semi-quantitative immunohistochemistry, Western blot and real time RT-PCR the cerebral GLUT1 and GLUT3 expression was quantified immediately, 220 min and 24 h following training. Relative to sham-trained and naive controls, operant conditioning training induced an immediate increase in GLUT1 immunoreactivity level in the hippocampus CA1 pyramidal cells as well as in the sensorimotor cortex. At longer post-learning delays, GLUT1 immunoreactivity decreased in the sensorimotor cortex and putamen. Parallel to the changes in protein levels, hippocampus GLUT1 mRNA level also increased immediately following learning. No effect of learning was found on hippocampal GLUT3 protein or mRNA expression. Measures of changes in glucose transporters expression present a link between cellular activation and glucose metabolism. The learning-induced localized increases in GLUT1 protein as well as mRNA levels observed in the present study confirm the previous findings that GLUT1 expression is plastic and respond to changes in cellular metabolic demands.

Immunohistochemical localization and quantification of glucose transporters in the mouse brain.

A family of seven facilitative glucose transporters (Glut1-5, 7 and 8) mediates the cellular uptake of glucose. In the brain, Glut2, Glut5 and Glut8 are found at relatively low levels whereas Glut1, Glut3 and Glut4 were reported in abundance in several brain regions. Using immunofluorescence, this study investigated, compared and quantified the localization of the brain major glucose transporters, Glut1, Glut3 and Glut4, in the different cerebral areas of CD1 mice. Most of the staining of Glut1, Glut3 and Glut4 in the mouse brain coincides with observations made in rats. The results confirm the cortical neuropil distribution of Glut3, the prominence of this transporter in the mossy fiber field of the hippocampus and the Glut3 and Glut4 immunostaining of the hippocampal pyramidal cell layer. The present study also reports novel localizations of the transporters such as the presence of Glut3 in neuronal perikarya, Glut4-labeled neurons in the CA3 of the hippocampus and the subiculum. In the cerebellum, Glut3 shows subcellular localization to the base of the Purkinje cell bodies near the axon hillock. Furthermore, an important population of Golgi cells was found to be strongly immunostained for Glut4 in the granular cell layer of the cerebellum. The quantification results suggest that the relative abundance of Glut1 in the frontal and motor cortices coincides well with the high-energy demands of these brain regions. However, the Glut4-selective abundance in cerebral motor areas supports its suggested role in providing the energy needed for the control of the motor activity. The reported neuropil distribution of Glut3 seems to uphold its suggested role in synaptic energy provision and neurotransmitter synthesis. We conclude that the cellular and regional distributions of the glucose transporters in the rodent brain seem to be relevant to their corresponding functions.