Edlund, Anna

Background: Altered cerebral glucose metabolism, especially prominent in APOE ɛ4 carriers, occurs years prior to symptoms in Alzheimer’s disease (AD). We recently found an association between a higher ratio of plasma apolipoprotein E4 (apoE4) over apoE3, and cerebral glucose hypometabolism in cognitively healthy APOE ɛ3/ɛ4 subjects. Plasma apoE does not cross the blood-brain barrier, hence we speculate that apoE is linked to peripheral glucose metabolism which is known to affect glucose metabolism in the brain. Objective: Explore potential associations between levels of plasma insulin and glucose with previously acquired plasma apoE, cerebral metabolic rate of glucose (CMRgl), gray matter volume, and neuropsychological test scores. Methods: Plasma insulin and glucose levels were determined by ELISA and a glucose oxidase assay whereas apoE levels were earlier quantified by mass-spectrometry in 128 cognitively healthy APOE ɛ3/ɛ4 subjects. Twenty-five study subjects had previously undergone FDG-PET and structural MRI. Results: Lower plasma apoE3 associated with higher plasma glucose but not insulin in male subjects and subjects with a body mass index above 25. Negative correlations were found between plasma glucose and CMRgl in the left prefrontal and bilateral occipital regions. These associations may have functional implications since glucose levels in turn were negatively associated with neuropsychological test scores. Conclusion: Plasma apoE3 but not apoE4 may be involved in insulin-independent processes governing plasma glucose levels. Higher plasma glucose, which negatively affects brain glucose metabolism, was associated with lower plasma apoE levels in APOE ɛ3/ɛ4 subjects. High plasma glucose and low apoE levels may be a hazardous combination leading to an increased risk of AD.

AbstractBackground APOEε4 is the strongest genetic risk factor for AD[1]. Although plasma apoE does not cross the blood‐brain‐barrier[2], altered plasma apoE levels have been linked to increased risk of AD[3] and to unfavorable CSF Aβ42 and tau levels[4]. Translation of results from studies of apoE in rodents to the human situation is difficult due species‐specific difference for instance in lipid metabolism[5]. Here we have studied the effects of a humanized APOEε4 liver[6] on synaptic integrity, insulin signaling and gliosis in a mouse model that reproduces the human plasma profile and that retains its endogenous apoE expression in the brain.MethodTissue samples from the cortex, hippocampus, cerebellum and thalamus were obtained from FRGN mice with humanized APOEε2/ε3 (n=4) versus APOEε4/ε4 (n=8) livers, and from APOEε3/ε3 (n=3) and APOEε4/ε4 (n=3) TR mice. Using established protocols[7], nuclei‐ and synaptosome‐enriched, and synaptosome‐depleted fractions were generated and analyzed for contents of markers relating to synapse integrity, gliosis and insulin signaling using western blotting. Levels of apoE4 were assessed in plasma from FRGN mice with humanized APOEε4/ε4 livers.ResultCortical and hippocampal levels of endogenous apoE, cortical levels of tubulin‐βIII, AMPAR, NMDAR, bassoon, mTOR, pAkt and a‐synuclein as well as hippocampal levels of synaptophysin, EAAT2, GFAP, NMDAR, pmTORS2448 and pIRS616 were reduced in FRGN mice with humanized APOEε4/ε4 versus APOEε2/ε3 livers (p≤0.05). Several of these findings were recapitulated in brain tissues from APOEε4/ε4 versus APOEε3/ε3 TR mice (p≤0.05). In both mouse models we observed a shift of a‐synuclein content from the synaptosomal (reduction) to the extra‐synaptosomal (increase) compartment (p<0.05). Altered expression levels of synaptic markers were also found in the thalamus and to a lesser extent in the cerebellum. Plasma levels of apoE appeared lower in the humanized APOEε4/ε4 liver mice and correlated positively to cortical levels of endogenous apoE but negatively to various markers of synaptic integrity and insulin signaling mainly in the hippocampus.ConclusionWe here provide the first evidence of a profound and negative effect of a liver APOEε4 phenotype in a human‐like setting, on key markers related to brain synaptic integrity, insulin signaling and glial cells involved in neuroinflammation.