In vivo molecular neuroimaging of glucose utilization and its association with fibrillar amyloid-beta load in aged APPPS1-21 mice
Faculty of Medicine and Health Sciences
Alzheimer's research & therapy
, 11 p.
University of Antwerp
Introduction: Radioligand imaging is a powerful in vivo method to assess the molecular basis of Alzheimer's Disease. We therefore aimed to visualize the pathological deposition of fibrillar amyloid-beta and neuronal dysfunction in aged double transgenic mice. Methods: Using non-invasive positron emission tomography (PET) we assessed brain glucose utilization with [F-18]FDG and fibrillar amyloidosis with [C-11]PiB and [F-18]AV45 in 12 month old APPPS1-21 (n = 10) mice and their age-matched wild-type controls (n = 15). PET scans were analyzed with statistical parametric mapping (SPM) to detect significant differences in tracer uptake between genotypes. After imaging, mice were sacrificed and ex vivo measures of amyloid-beta burden with immunohistochemistry as well as glucose utilization with [C-14]-2DG autoradiography were obtained as gold standards. Results: Voxel-wise SPM analysis revealed significantly decreased [F-18]FDG uptake in aged APPPS1-21 mice in comparison to WT with the thalamus (96.96 %, maxT = 3.35) and striatum (61.21 %, maxT = 3.29) demonstrating the most widespread reductions at the threshold of p < 0.01. [C-11]PiB binding was significantly increased in APPPS1-21 mice, most notably in the hippocampus (87.84 %, maxT = 7.15) and cortex (69.08 %, maxT = 7.95), as detected by SPM voxel-wise analysis at the threshold of p < 0.01. Using the same threshold [F-18]AV45 uptake was comparably lower with less significant differences. Compared to their respective ex vivo equivalents [F-18]FDG demonstrated significant positive correlation to [C-14]2-DG autoradiography (r = 0.67, p < 0.0001) while [C-11]PiB and [F-18]AV45 binding did not correlate to ex vivo immunohistochemistry for amyloid-beta (r = 0.25, p = 0.07 and r = 0.17, p = 0.26 respectively). Lastly no correlation was observed between regions of high amyloid burden and those with decreased glucose utilization (r = 0.001, p = 0.99). Conclusions: Our findings support that fibrillar amyloid-beta deposition and reduced glucose utilization can be visualized and quantified with in vivo mu PET imaging in aged APPPS1-21 mice. Therefore, the combined use of [F-18]FDG and amyloid mu PET imaging can shed light on the underlying relationship between fibrillar amyloid-beta pathology and neuronal dysfunction.