TREM2 deficiency exacerbates tau pathology through dysregulated kinase signaling in a mouse model of tauopathy

Abstract

Background: Genetic variants of the Triggering Receptor Expressed on Myeloid Cells-2 (TREM2) confer increased risk of developing late-onset Alzheimer's Disease (LOAD) and other neurodegenerative disorders. Recent studies provided insight into the multifaceted roles of TREM2 in regulating extracellular β-amyloid (Aβ) pathology, myeloid cell accumulation, and inflammation observed in AD, yet little is known regarding the role of TREM2 in regulating intracellular microtubule associated protein tau (MAPT; tau) pathology in neurodegenerative diseases and in AD, in particular.

Results: Here we report that TREM2 deficiency leads to accelerated and exacerbated hyperphosphorylation and aggregation of tau in a humanized mouse model of tauopathy. TREM2 deficiency also results, indirectly, in dramatic widespread dysregulation of neuronal stress kinase pathways.

Conclusions: Our results suggest that deficiency of microglial TREM2 leads to heightened tau pathology coupled with widespread increases in activated neuronal stress kinases. These findings offer new insight into the complex, multiple roles of TREM2 in regulating Aβ and tau pathologies.

Keywords: Alzheimers disease; Immunity; Inflammation; TREM2; Tauopathy.

Fig. 1

Increased soluble and insoluble tau phosphorylation in TREM2 deficient mice. a-d Microdissected cortices and hippocampi from hTau (Trem2 ^+/+) and hTau;Trem2 ^−/− mice were analyzed using western blot with antibodies against AT8, AT180, PHF-1 phospho-epitopes and Tau5 (total tau). b Quantification of cortex western blot data revealed highly significant increases in the ratio of phosphorylated AT8, AT180, and PHF-1 to total tau but importantly, no increases were detected in total tau (Tau5). d Quantification of hippocampus western blot data reveals significant hyperphosphorylation at the AT8 epitope. e Sarkosyl extractions were performed on cortical tissue lysates from hTau (Trem2 ^+/+) and hTau;Trem2 ^−/− mice and protein levels analyzed using western blot. f Quantification of western blot optical densities revealed significant increases in AT180 and highly significant increases in PHF-1 and Tau5 (total tau). All experiments used n = 4–6 (equal males and females) mice per group unless otherwise noted. At least two independent experiments were performed for each analysis. Error bars represent SEM. *, P < 0.05, **, P < 0.01, ***, P < 0.001

Fig. 2

Immunohistochemistry reveals increased tau hyperphosphorylation in TREM2 deficient hTau mice. a Immunohistochemistry was performed on 6-month hTau (Trem2 ^+/+) and hTau;Trem2 ^−/− mice using antibodies against AT8 and AT180 tau phospho-epitopes and revealed dramatically increased AT8 and AT180 staining in the cortex. b Quantitation of total p-Tau⁺ neurons revealed highly significant increases in total AT8⁺ and AT180⁺. c, d Laminar specific p-tau⁺ counts reveal significant increases in specific layers II-III in AT8⁺ and AT180⁺ neurons in hTau;Trem2 ^−/− mice. Additionally, robust increases in layers IV-VI were detected in hTau;Trem2 ^−/− mice. e Immunofluorescent co-labeling was performed with AT8 and macrophage/microglia specific Iba1 revealing reactive microglia in direct association with neurons heavily laden with phosphorylated tau in hTau;Trem2 ^−/− mice compared to hTau controls. All experiments used n = 4–6 (equal males and females) mice per group unless otherwise noted. At least two independent experiments were performed for each analysis. Bars, a-30 μm, e-100 μm

Fig. 3

TREM2 deficiency leads to altered microglial activation. a IHC staining was performed on 30 μm thick sagittal sections from 6-month hTau (Trem2 ^+/+) and hTau;Trem2 ^−/− mice (n = 6 per group; equal males and females) revealing altered morphological activation. b qRT-PCR was performed on hemi-brains from 6- and 12- month hTau (Trem2 ^+/+) and B6 control mice (n = 3–4 per group), revealing significantly upregulated TREM2 transcripts over time. c and d Individual microglia were selected randomly, isolated, thresholded, and skeletonized for analysis (20–30 per genotype) and compared between hTau and hTau;Trem2 ^−/− mice within cortex and CA3 region of the hippocampus. (e,l) IHC staining was performed on hTau (Trem2 ^+/+) and hTau;Trem2 ^−/− mice using antibodies against the pan-macrophage marker F4/80 (n = 6 per group). f, g Iba1 transcripts were analyzed from 6-month hTau (Trem2 ^+/+ ) and hTau;Trem2 ^−/− mice using qRT-PCR (n = 4–6 mice per group) alongside IHC staining with Iba1 (n = 6 per group) to perform a cell count which revealed no differences in the levels of Iba1 produced, or in the total number of cells/area h-k Detailed quantification of microglial morphology revealed decreased soma, cell area, branches, and branch points in TREM2 deficient mice compared to wild type hTau. At least two independent experiments were performed for each analysis. Error bars represent SEM., (b,f,l) Student’s t; (h-k) 2-way ANOVA with Sidak multiple comparisons correction) * P < 0.05. **, P < 0.01, ***, P < 0.001

Fig. 4

Increased activation of stress signaling molecules in TREM2 deficient hTau mice. a-d Western blot analysis was performed on microdissected cortices (a) and hippocampi (b) of hTau (Trem2 ^+/+) and hTau;Trem2 ^−/− mice using antibodies directed against phosphorylated and total JNK, ERK1/2, GSK3β, and P38 to determine relative levels of MAPK activation. Quantification of western blot data revealed highly significant increases in total levels of JNK, as well as the ratio of pJNK/total JNK, and also very highly significant increases in the ratio pGSK3β(Ser9)/GSK3β in cortex (c), while robust increases in total levels of GSK3β were accompanied by significant increases in the ratios pERK1/2/total ERK1/2, pGSK3β(Ser9)/Total GSK3β, pGSK3β(Y216)/total GSK3β, pP38/total P38, and pJNK/total JNK in hippocampi of 6-month mice (d). All experiments used n = 6 (equal males and females) mice per group unless otherwise noted. At least two independent experiments were performed for each analysis. Error bars represent SEM. ANOVA with Tukey correction *, P < 0.05, **, P < 0.01, ***, P < 0.001