The Impact of Dnmt3a or Tet2 CH on Alzheimer's Disease in a Murine Model

Clonal Hematopoiesis of Indeterminate Potential (CHIP) represents a clonal expansion of mutant hematopoietic stem cells as a consequence of aging and environmental stress and is most commonly driven by mutations in DNMT3A or TET2. CHIP has been associated with a number of age-associated diseases; but controversy exists as to whether CHIP predisposes to or protects from Alzheimer's Disease (AD). Indeed, mutations in DNMT3A and TET2 have been reported to drive increased myeloid cell inflammation which would seem to increase AD; however, the effect of these mutations on microglia, a resident macrophage-like cell in the brain that plays an important role in neurodegenerative diseases including AD, remains unknown. Here, we investigated the impact of Dnmt3a and Tet2 loss-of-function CHIP in a murine model of AD. Our findings suggest opposing and mutation-specific effects by Dnmt3a versus Tet2 on AD pathogenesis.

We performed total body irradiation of 5xFAD transgenic mice predisposed to familial AD and performed intravenous transplantation of either Dnmt3a Knock-out (KO), Tet2 KO or wildtype (WT) whole bone marrow. To mimic systemic chronic inflammation seen in aging, we challenged all groups weekly with LPS. We conducted behavioral assays to assess differences in recognition and cognitive memory as well as anxiety levels between the experimental and control groups. Six months following transplantation, mouse brains were harvested for cytokine analysis, measurement of beta-amyloid plaques, and characterization of myeloid and microglial populations in the brain.

Mice harboring Dnmt3a -mutant bone marrow showed increased anxiety, impaired recognition, and cognitive phenotypes consistent with AD. Unexpectedly, cytokine analysis revealed that both Dnmt3a and Tet2 experimental groups showed an overall decrease in pro-inflammatory cytokines such as IFNg, IL-1 and G-CSF in the brain, suggesting lower neuroinflammatory levels compared to the WT control group. Furthermore, we found decreased CD11a expression by qRT-PCR in the brain, suggesting a decrease in infiltrating peripheral immune cells into the brain. However, quantification of immunological cells by flow cytometry revealed that 5xFAD mice transplanted with Dnmt3a -mutant marrow showed an increase in the number of CD11a- CD11b+ TMEM119+ microglial cells, but no change in CD11a+ CD11b+ TMEM119- myeloid cells. Phenotypically, the microglia from the Dnmt3a -mutant marrow mice showed lower expression of activation markers CD45, CD68 and CD11c. Interestingly, while no difference was observed in infiltrating myeloid cells, we also observed a decrease in infiltrating CD11a+ CD3+ CD4+ T cells in the brain of Dnmt3a^-/- transplanted mice. These data suggest that the decreased T cell infiltration and subsequent decrease in cytokine signaling in the 5xFAD mice with Dnmt3a^-/- bone marrow may contribute to decreased activation of microglia, impairing amyloid uptake and clearance.

In contrast, 5xFAD mice transplanted with Tet2 mutant bone marrow showed improved recognition and cognitive memory levels compared to control, suggesting a protective effect against AD. Mice transplanted with Tet2-mutant marrow demonstrated an increase in CD11a+ cells, suggesting an increase in infiltrating peripheral immune cells into the brain. There was no significant difference in the overall frequency of microglial cells in mice transplanted with Tet2-mutant cells compared to WT. and MHC II expression was lower for both infiltrating myeloid and microglial cells in Tet2-mutant transplanted mice. Altogether, these data show that an increased myeloid cell population in the brain of Tet2 KO transplanted 5xFAD mice, despite having lower MHC II expression, may contribute to protecting against AD in our mouse model.

Overall, our data suggest that CHIP-associated Dnmt3a and Tet2 mutations may confer different effects on AD pathogenesis, with DNMT3A mutations accelerating progression of AD while TET2 mutations reduce it. Further studies are ongoing to conduct immunohistochemical analysis on brain samples to measure microglia and amyloid plaques. Our work is an important first step to understanding the underlying mechanism by which CHIP mutations affect the inflammatory state and progression of AD.

No relevant conflicts of interest to declare.