Impairment in long-term memory formation and learning-dependent synaptic plasticity in mice lacking glycogen synthase in the brain

Abstract

Glycogen is the only carbohydrate reserve of the brain, but its overall contribution to brain functions remains unclear. Although it has traditionally been considered as an emergency energetic reservoir, increasing evidence points to a role of glycogen in the normal activity of the brain. To address this long-standing question, we generated a brain-specific Glycogen Synthase knockout (GYS1(Nestin-KO)) mouse and studied the functional consequences of the lack of glycogen in the brain under alert behaving conditions. These animals showed a significant deficiency in the acquisition of an associative learning task and in the concomitant activity-dependent changes in hippocampal synaptic strength. Long-term potentiation (LTP) evoked in the hippocampal CA3-CA1 synapse was also decreased in behaving GYS1(Nestin-KO) mice. These results unequivocally show a key role of brain glycogen in the proper acquisition of new motor and cognitive abilities and in the underlying changes in synaptic strength.

Figure 1

Generation of a brain-specific Glycogen Synthase knockout (GYS1 KO). (A) Schematic representation of the alleles. The excision of the selection cassette from the knockout allele generates a conditional allele that expresses glycogen synthase normally. The subsequent excision of several exons from the conditional allele by a Cre Recombinase generates again a knockout allele. (B) Genotypes used in the study. Homozygous conditional, non-Cre-expressing littermates were used as controls of GYS1^Nestin-KO. (C) Western blot of glycogen synthase in total homogenates of brain and skeletal muscle using an antibody that recognizes specifically the muscle (MGS) or the liver isoform (LGS). Mouse liver extract (L) was used as a positive control for LGS. Actin was used as a loading control. (D) Total glycogen synthase activity in the presence of its allosteric activator Glucose 6-phosphate (P=2.80 × 10⁻⁷). (E) Glycogen levels (P=0.007) in brain extracts of control and GYS1^Nestin-KO mice. Control (n=4), KO (n=4). *Statistically significant (P⩽0.05).

Figure 2

Changes in the expression of proteins involved in glycogen metabolism in the GYS1^Nestin-KO mouse brain. (A) MGS, brain and muscle isoforms of Glycogen Phosphorylase (BGPh and MGPh, respectively), glycogen debranching enzyme (GDE), AMP kinase (AMPK), pAMPK, GSK3 α/β, and pGSK3 β were analyzed in total brain homogenates. (B) Glycogenin levels were analyzed in untreated and amylase-treated total brain extracts. Actin was used as a loading control.

Figure 3

Electrophysiologic changes of hippocampal synapses in GYS1^Nestin-KO alert behaving mice. (A) Experimental design. Animals were chronically implanted with stimulating electrodes in the hippocampal Schaffer collaterals and with a recording electrode in the ipsilateral pyramidal CA1 area. An extra wire was attached to the bone as ground (DG, dentate gyrus; Sub., subiculum). (B) Input/output curves of field excitatory postsynaptic potentials (fEPSPs) evoked by paired pulses of increasing intensities in control (n=11) and knockout (KO) (n=11) mice. The best nonlinear adjustments to the collected data are illustrated. (C) Paired-pulse facilitation analyses (mean±s.e.m. slopes of the second fEPSP expressed as a percentage of the first interpulse interval). Representative fEPSP paired traces (40 ms of interpulse interval) are shown on the right. Control (n=11), KO (n=11). (D) Time course of long-term potentiation (LTP) evoked in the CA3-CA1 synapse after a high-frequency stimulation (HFS) session (mean±s.e.m. fEPSP slopes given as a percentage of values collected during baseline recordings (100%)). Control (n=7), KO (n=7). Representative examples of fEPSPs collected at the indicated times are plotted at the top. *Statistically significant (P<0.05) differences between the two groups.

Figure 4

Impaired performance of GYS1^Nestin-KO mice in an operant conditioning task. (A) Experimental set-up. Mice were trained in a Skinner box to press a lever to obtain a food pellet. (B) Two tasks of increasing difficulty were assayed. In task 1 (fixed-ratio (1:1)), mice received a food pellet each time they pressed the lever. In task 2 (Light/Dark), lever presses were rewarded only when a light bulb was switched on. Pressing the lever during the dark period punished the animal with an additional delay in the reappearance of the light period. (C) Lever presses in the first 7 days of training of task 1. Dotted line corresponds to criterion. (D) Mean days required to reach the criterion. (E) Percentage of mice reaching the criterion during the training. (F) Evolution of field excitatory postsynaptic potential (fEPSPs) evoked at the CA3-CA1 synapse during task 1 (mean±s.e.m. fEPSP slopes given as a percentage of values collected before training (100%)). (G) Performance of control and GYS1^Nestin-KO mice during task 2. Control (CT) (n=13), KO (n=13). *Statistically significant (P<0.01) differences between the two groups.