Tau Activates Transposable Elements in Alzheimer's Disease

Abstract

Aging and neurodegenerative disease are characterized by genomic instability in neurons, including aberrant activation and mobilization of transposable elements (TEs). Integrating studies of human postmortem brain tissue and Drosophila melanogaster models, we investigate TE activation in association with Tau pathology in Alzheimer's disease (AD). Leveraging RNA sequencing from 636 human brains, we discover differential expression for several retrotransposons in association with neurofibrillary tangle burden and highlight evidence for global TE transcriptional activation among the long interspersed nuclear element 1 and endogenous retrovirus clades. In addition, we detect Tau-associated, active chromatin signatures at multiple HERV-Fc1 genomic loci. To determine whether Tau is sufficient to induce TE activation, we profile retrotransposons in Drosophila expressing human wild-type or mutant Tau throughout the brain. We discover heterogeneous response profiles, including both age- and genotype-dependent activation of TE expression by Tau. Our results implicate TE activation and associated genomic instability in Tau-mediated AD mechanisms.

Keywords: Alu; Alzheimer's disease; ERV; LINE1; MAPT; RNA sequencing; chromatin; genomic instability; neurodegeneration; retrotransposon.

Figure 1.. Tau Pathologic Burden Is Associated with Increased TE Expression in Human Brains

Boxplots display regression t-statistics aggregated based on TE clade annotations. The dotted lines indicate the significance threshold, denoting those TEs (red) with the most extreme associations listed in Table 1. The mean t-statistic was significantly inflated for L1 (p = 7.1 × 10⁻⁸), ERV1 (p = 6.9 × 10⁻¹⁴), ERV2 (p = 1.9 × 10⁻⁹), and ERV3 (p = 8.2 × 10⁻⁸), consistent with a global impact of Tau pathologic burden on TE expression. See also Table S2.

Figure 2.. Tau Activates Expression of Selected TEs in the Drosophila Brain

(A) In 20-day-old animals, the copia, het-a, and gypsy retrotransposons were activated following neuronal expression of wild-type and/or mutant human Tau. Expression of 12 TEs was profiled by qPCR in fly heads from the following genotypes: (1) ELAV-GAL4/+; (2) ELAV-GAL4/+; UAS-Tau^WT/+; and (3) ELAV-GAL4/+;UAS-Tau^R406W/+. One-way ANOVA model F-test was significant (p < 0.05) for copia, het-a, and gypsy. Analyses of 1- and 10-day-old animals are shown in Figure S2. (B) Expression of the copia retrotransposon is enhanced by age and mutant Tau. Two-way ANOVA testing was significant (p < 0.0001) for both age and genotype. All results (A and B) were normalized to RpL32 expression, and fold-change relative to 1-day-old ELAV-GAL4/+ control flies is shown (mean ± SEM). Subsetted t tests were performed for post hoc comparisons. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 3.. Hypothetical Model and Remaining Questions

Our results along with other published evidence inform a causal model for Tau-mediated TE activation in AD, along with key knowledge gaps for further investigation (1–3). Tau is sufficient to induce TE transcriptional activation. Analyses of HERV-Fc1 suggest that chromatin changes may in part be responsible, but other mechanisms may also contribute. (1) Besides Tau, it is likely that additional brain pathologies promote TE activation. (2) It remains to be determined whether TE transcriptional activation in AD leads to mobilization, potentially contributing to DNA damage and genomic instability. In the absence of transposition, TE expression may provoke an innate immune response. (3) Although DNA damage and neuroinflammation are strongly implicated in AD neurodegeneration, additional studies will be required to assess whether TEs contribute.