Introduction: Microglia, central nervous system resident immune cells, play a pivotal role in the neuroinflammatory response associated with Alzheimer's disease (AD). Their dysregulation and morphological changes contribute to AD pathogenesis, activating in response to beta amyloid plaques (Doens and Fernández, 2014). However, the timing of microglial activation in relation to early Alzheimer's pathology remains unclear. This study aims to elucidate whether microglial morphological changes occur in the early stages of amyloid-beta (Aβ) accumulation, using a mouse model to selectively induce amyloid precursor protein expression (thus Aβ-production).
Methods: We bred transgenic mice with human APOE3/APOE4 alleles, using a Tet-off system and dietary doxycycline to modulate APP expression and control Aβ production. The study assessed the impact of APOE genotypes, entorhinal cortex and hippocampus regions, and Aβ production on Alzheimer's pathology. Microglia visualization was through IBA1 immunohistochemistry, Aβ quantification via the MSD 6E10 Aβ panel (protein-normalized), and amyloid aggregation detected with MethoxyX04 staining. We assessed microglial morphology in 150 cells from 28 animals using ImageJ's Moti-Q plugin, following Hansen et al., 2022. Statistical analysis involved principal component analysis (PCA) for morphological data dimensionality reduction and a targeted approach focusing on neuroinflammation-related metrics, as per Olabiyi et al., 2023.
Results: Without doxycycline for 18 weeks, APPSwe/Ind mice exhibited increased Aβ levels (38, 40, 42 isoforms; p = 0.012) without amyloid aggregation. Neither Aβ levels nor microglial morphology were affected by APOE genotype. Unbiased PCA revealed nine principal components, explaining 80% of the variance in microglial morphology, with the primary component explaining 46%, largely reflecting cell size and branching complexity. However, multi-way ANOVA found no significant morphological differences between control mice and those accumulating Aβ.
Targeted analysis on key microglial activation markers—Iba1 intensity, tree length, branching, and ramification — did however reveal significant inflammation-linked changes in Aβ-producing mice: heightened intensity (100.52 ± 5.69 vs. 68.48 ± 2.90; p < 0.0001), fewer branches (98.27 ± 9.23 vs. 162.72 ± 13.67; p = 0.0038), lower ramification index (7.05 ± 0.26 vs. 10.37 ± 0.36; p < 0.00001), and reduced tree lengths (615.08 ± 47.89 µm vs. 935.32 ± 65.32 µm; p = 0.00282). However, cell volume remained unchanged (4304.56 ± 312.92 µm³ vs. 4699.72 ± 290.89 µm³; p = 0.424), despite its previous association with microglial activation (Olabiyi et al., 2023). Post-hoc analysis revealed marked intensity and ramification differences in Aβ mice in the assessed regions (intensity: cortex p=0.00086, hippocampus p=0.00218; ramification: cortex p=0.00005, hippocampus p=0.007) and significant cross-regional variations (intensity: cortex vs. control hippocampus p<0.00001; ramification: hippocampus vs. control cortex p=0.0081, cortex vs. control hippocampus p=0.00006). These results indicate early, albeit incomplete, microglial activation in early Alzheimer’s disease pathology, prior to beta amyloid plaque formation.
Conclusion: Our results demonstrated elevated neuroinflammation and Aβ levels in early AD mouse model, irrespective of APOE genotype. Initial PCA and multi-way ANOVA showed no significant microglial morphology changes due to Aβ. Yet, targeted analysis of microglial activation markers—like increased intensity, reduced branches, shorter tree length, and lower ramification index—indicates early microglial activation signs during initial Aβ accumulation, before plaque formation.