G-CSF and cognitive dysfunction in elderly diabetic mice with cerebral small vessel disease: Preventive intervention effects and underlying mechanisms.

Aims: Although cognitive dysfunction is a common neurological complication in elderly patients with diabetes, the mechanisms underlying this relationship remain unclear, and effective preventive interventions have yet to be developed. Thus, this study investigated the preventive effects and mechanisms of action associated with granulocyte colony-stimulating factor (G-CSF) on cognitive dysfunction in elderly diabetic mice with cerebral small vessel disease.
Methods: This study included 40 male db/db diabetic and wild-type (WT) mice that were categorized into the following four groups at the age of 3 weeks: db/db group (DG), db/db+G-CSF group (DGG), WT group (WG), and WT+G-CSF group (WGG). The mice were fed normal diets for 4 months and then given G-CSF (75 μg/kg) via intraperitoneal injections for 1 month. At 7.5 months of age, the cognitive abilities of the mice were assessed with the Y-maze test and the Social Choice Test; body weight, blood pressure (BP), and blood glucose measurements were obtained throughout the study. Brain imaging and blood oxygen level-dependent (BOLD) contrast imaging analyses were performed with a small animal magnetic resonance imaging (MRI) system, autophagosome levels were detected with a transmission electron microscope (TEM), hippocampal neurons were assessed with hematoxylin and eosin (HE) staining, and protein expressions and distributions were evaluated using immunohistochemistry and Western blot analyses.
Results: (i) The body weight and blood glucose levels of the DG and DGG mice were significantly higher than those of the WG and WGG mice; (ii) social choice and spatial memory capabilities were significantly reduced in DG mice but were recovered by G-CSF in DGG mice; (iii) the MRI scans revealed multiple lacunar lesions and apparent hippocampal atrophy in the brains of DG mice, but G-CSF reduced the number of lacunar lesions and ameliorated hippocampal atrophy; (iv) the MRI-BOLD scans showed a downward trend in whole-brain activity and reductions in the connectivities of the hippocampus and amygdala with subcortical structures in DG mice, but G-CSF clearly improved the altered brain activity as well as the connectivity of the hippocampus in DGG mice; (v) HE staining revealed fewer neurons in the hippocampus in DG mice; (vi) TEM analyses revealed significantly fewer autophagosomes in the hippocampi of DG mice, but G-CSF did not increase these numbers; (vii) there were significant reductions in mechanistic target of rapamycin (mTOR) and LC3-phosphatidylethanolamine conjugate (LC3)-II/I levels in the hippocampi of DG mice, whereas p62 was upregulated, and G-CSF significantly enhanced the levels of Beclin1, mTOR, and LC-II/I in DGG mice; and (viii) G-CSF significantly reversed increases in nuclear factor κB (NF-κB) protein levels in DG but not in WG mice.
Conclusions: In this study, aged diabetic mice were prone to cognitive dysfunction and cerebral small vessel disease. However, administration of G-CSF significantly improved cognitive function in elderly db/db diabetic mice, and this change was likely related to the regulation of autophagy and NF-κB signaling pathways.
(© 2017 John Wiley & Sons Ltd.)