Pharmacologically increasing O-GlcNAcylation increases complexity of astrocytes in the dentate gyrus of TgF344-AD rats

Melissa L Garcia^1,2Adam R Denton³Nateka L Jackson²Michael David Scofield^2*Lori L McMahon^1,2*

• ¹The University of Alabama at Birmingham, Birmingham, United States
• ²Medical University of South Carolina, Charleston, United States
• ³Tusculum University, Greeneville, United States

Background: Preclinical studies report that increasing the post-translational modification, O-GlcNAcylation can reduce amyloidogenic processing of amyloid precursor protein (APP) and compete with serine phosphorylation on tau, decreasing hyperphosphorylated tau accumulation. Protein O-GlcNAcylation can have anti-inflammatory effects, suggesting the possibility that increasing O-GlcNAcylation may decrease reactive gliosis and other pathological changes in AD. Methods: The selective OGA inhibitor thiamet-G (TMG; 10 mg/kg, subcutaneously (s.c.)) was administered three times per week for 3 months starting at 6 months of age, a time point when Aβ pathology is evident in the hippocampus. Western blot analysis was used to measure protein levels of GFAP, Iba-1, and Aβ. Immunohistochemistry and confocal imaging were used to assess Aβ plaques, astrocyte and microglia complexity, and degeneration of tyrosine hydroxylase-positive (TH+) axons. Results: In TgF344-AD rats, we found significantly increased astrocyte complexity, increased numbers of microglia, loss of noradrenergic axons (NA), and significant Aβ plaques compared to WT, confirming previous work by us and others. Notably, pharmacologically increasing O-GlcNAcylation further increased astrocyte complexity in TgF344-AD rats, specifically those located in close proximity to Aβ plaques, while microglia morphology and Aβ staining were unaffected. O-GlcNAcylation was not able to lessen the loss of TH+ axons in TgF344-AD rats, although fewer dystrophic axons were observed, suggesting a possible beneficial effect. Discussion: Our findings demonstrate that increasing O-GlcNAcylation in TgF344-AD rats using a cyclical treatment protocol at a time when Aβ pathology is already significant does not provide broad beneficial effects on Aβ accumulation, microglial reactivity, or noradrenergic axon loss, although there appears to be fewer dystrophic axons. Importantly, increasing O-GlcNAcylation in TgF344-AD rats has dual beneficial effects on astrocyte reactivity. Astrocytes in close proximity to Aβ plaques are more demonstrate longer processes and more branches compared to those in saline-treated TgF344-AD rats at the same distance, enabling them to surround plaques and protect nearby neurons. Astrocytes located at more distal locations from plaques are less reactive than those at the same distance in saline-treated TgF344-AD rats. Our findings offer new insights into the possible mechanisms that might contribute to the beneficial therapeutic effects of increasing O-GlcNAcylation during progressive AD pathology