Age- and Brain Region-Specific Changes of Glucose Metabolic Disorder, Learning, and Memory Dysfunction in Early Alzheimer's Disease Assessed in APP/PS1 Transgenic Mice Using 18F-FDG-PET

Abstract

Alzheimer's disease (AD) is a leading cause of dementia worldwide, associated with cognitive deficits and brain glucose metabolic alteration. However, the associations of glucose metabolic changes with cognitive dysfunction are less detailed. Here, we examined the brains of APP/presenilin 1 (PS1) transgenic (Tg) mice aged 2, 3.5, 5 and 8 months using ¹⁸F-labed fluorodeoxyglucose (¹⁸F-FDG) microPET to assess age- and brain region-specific changes of glucose metabolism. FDG uptake was calculated as a relative standardized uptake value (SUVr). Morris water maze (MWM) was used to evaluate learning and memory dysfunction. We showed a glucose utilization increase in multiple brain regions of Tg mice at 2 and 3.5 months but not at 5 and 8 months. Comparisons of SUVrs within brains showed higher glucose utilization than controls in the entorhinal cortex, hippocampus, and frontal cortex of Tg mice at 2 and 3.5 months but in the thalamus and striatum at 3.5, 5 and 8 months. By comparing SUVrs in the entorhinal cortex and hippocampus, Tg mice were distinguished from controls at 2 and 3.5 months. In MWM, Tg mice aged 2 months shared a similar performance to the controls (prodromal-AD). By contrast, Tg mice failed training tests at 3.5 months but failed all MWM tests at 5 and 8 months, suggestive of partial or complete cognitive deficits (symptomatic-AD). Correlation analyses showed that hippocampal SUVrs were significantly correlated with MWM parameters in the symptomatic-AD stage. These data suggest that glucose metabolic disorder occurs before onset of AD signs in APP/PS1 mice with the entorhinal cortex and hippocampus affected first, and that regional FDG uptake increase can be an early biomarker for AD. Furthermore, hippocampal FDG uptake is a possible indicator for progression of Alzheimer's cognition after cognitive decline, at least in animals.

Keywords: 18F-FDG PET; APP/PS1 mice; Alzheimer’s disease; cognitive dysfunction; glucose metabolism; hippocampus.

Figure 1

Seven regions of interest (ROI). The schematic boundaries of ROIs were drawn on the coronal (a–e), axial (f), and sagittal (g) T2-weighted images of a mouse brain according to the mouse brain atlas by Paxinos and Watson [12].

Figure 2

Morris water maze (MWM) results. (A) Comparisons of escape latencies in the cued test between Tg and wide-type (WT) mice aged 2, 3.5, 5 and 8 mo; (B) Escape latencies in 5 days in the training test and (C) typical training traces (the fifth day) for mice of each age. (* p < 0.05; ** p < 0.01; vs. aged-matched WT mice); (D) Comparisons of the percentage of time spent in the target quadrant, percentage of path travelled in the target quadrant, mean speed, as well as number of target crossings in the probe test between Tg and WT mice aged 2, 3.5, 5 and 8 mo; and (E) typical probe traces. (* p < 0.05; ** p < 0.01; vs. aged-matched WT mice). mo: month-old.

Figure 3

Results of PET/MRI fusion. Typical coronal ¹⁸F-FDG-PET images from WT (8 mo) and Tg mice (2, 3.5, 5 and 8 mo), projected on standard T2-weighed MR images from a WT mouse. mo: month-old.

Figure 4

Quantitative analysis of PET data. ROI-based SUVr (relative standardized uptake value) group comparisons for Tg and WT mice aged 2, 3.5, 5 and 8 mo. Graphs highlight age-specific changes in SUVr values for the entorhinal cortex (A); hippocampus (B); frontal cortex (C); corpus callosum (D); striatum (E); and thalamus (F). (* p < 0.05, ** p < 0.01, vs. Tg mice of other age groups; # p < 0.05, ## p < 0.01 vs. Tg mice of the same age group). mo: month-old.

Figure 5

Correlations of hippocampal glucose utilization with MWM performance in mice at ages 3.5, 5 and 8 mo. Hippocampal SUVr values of individual animals were correlated with the escape latency (A) and percentage of path travelled in the target quadrant (B). A linear regression model was used for the correlation analysis. The regression curve line, correlation coefficients (r), and significance values are shown in the plots. Statistical significance was set at p < 0.05. mo: month-old.

Figure 6

Immunohistochemical results. (A) Representative images of Aβ and plaques in the entorhinal cortex and hippocampus (dental gyrus) of WT and Tg mice aged 2, 3.5, 5 and 8 mo, visualized by staining Aβ_1–16 peptides. Arrows indicate the presence of Aβ plaques. Scale bar: 50 μm; (B) Percentage of Aβ plaque area in each microscopic view (at 200×) in the entorhinal cortex and hippocampus in Tg and WT mice aged 2, 3.5, 5 and 8 mo. (* p < 0.05, ** p < 0.01, vs. Tg mice of other age groups). mo: month-old.