Effects of concentrated ambient ultrafine particulate matter on hallmarks of Alzheimer's disease in the 3xTgAD mouse model

Abstract

Background: Exposure to air pollution has been identified as a possible environmental contributor to Alzheimer's Disease (AD) risk. As the number of people with AD worldwide continues to rise, it becomes vital to understand the nature of this potential gene-environment interaction. This study assessed the effects of short-term exposures to concentrated ambient ultrafine particulates (UFP, <100 nm) on measurements of amyloid-β, tau, and microglial morphology.

Methods: Two cohorts of aged (12.5-14 months) 3xTgAD and NTg mice were exposed to concentrated ambient UFP or filtered air for 2 weeks (4-h/day, 4 days/week). Bronchoalveolar lavage fluid and brain tissue were collected twenty-four hours following the last exposure to evaluate lung inflammation, tau pathology, amyloid-β pathology, and glial cell morphology.

Results: No exposure- or genotype-related changes were found with any of the measures of lung inflammation or in the hippocampal staining density of astrocyte marker glial fibrillary acidic protein. The microglia marker, ionized calcium binding adaptor molecule 1, and amyloid-β marker, 6E10, exhibited significant genotype by exposure interactions such that levels were lower in the UFP-exposed as compared to filtered air-exposed 3xTgAD mice. When microglia morphology was assessed by Sholl analysis, microglia from both NTg mouse groups were ramified. The 3xTgAD air-exposed mice had the most ameboid microglia, while the 3xTgAD UFP-exposed mice had microglia that were comparatively more ramified. The 3xTgAD air-exposed mice had more plaques per region of interest as measured by Congo red staining as well as more plaque-associated microglia than the 3xTgAD UFP-exposed mice. The number of non-plaque-associated microglia was not affected by genotype or exposure. Levels of soluble and insoluble human amyloid-β₄₂ protein were measured in both 3xTgAD groups and no exposure effect was found. In contrast, UFP-exposure led to significant elevations in phosphorylated tau in 3xTgAD mice as compared to those that were exposed to air, as measured by pT205 staining.

Conclusions: Exposure to environmentally relevant levels of ultrafine particulates led to changes in tau phosphorylation and microglial morphology in the absence of overt lung inflammation. Such changes highlight the need to develop greater mechanistic understanding of the link between air pollution exposure and Alzheimer's disease.

Keywords: Alzheimer’s disease; Amyloid-β; Inhalation exposure; Microglia; Tau; Ultrafine particles.

Figure 1-

Daily particle number concentration over the duration of the 8 days of exposure. Mean ± SD. Closed circles=Cohort A, open triangles=Cohort B.

Figure 2-

Hippocampal GFAP staining in mice exposed to filtered air or UFP. (A) Quantitation of GFAP percent area stained. Representative images of GFAP staining in (B) a NTg mouse that was exposed to filtered air, (C) a NTg mouse that was exposed to UFP, (D) a 3xTgAD mouse exposed to filtered air, and (E) a 3xTgAD mouse exposed to UFP. Mean ± SEM, n=4–5.

Figure 3-

Hippocampal Iba1 staining in mice exposed to filtered air or UFP. (A) Quantitation of Iba1 percent area stained in hippocampus. (B) Quantitation of Iba1 percent area stained with analysis restricted to the subiculum. Representative images of Iba1 staining in (C) a NTg mouse that was exposed to filtered air, (D) a NTg mouse that was exposed to UFP, (E) a 3xTgAD mouse exposed to filtered air, and (F) a 3xTgAD mouse exposed to UFP. (G) Magnification of the subiculum that is shown in panel 3E. Note the presence of prominent Iba1+ ameboid microglia. Mean ± SEM, n=4–5, *p<0.05.

Figure 4-

Sholl analysis in the subiculum of 3xTgAD and NTg mice. (A) 60x image in the subiculum of a filtered air-exposed 3xTgAD mouse showing Aβ plaques (red), microglia (green), and cell nuclei (blue). Inset illustrates the concentric ring structure used for Sholl analysis. (B) Sholl analysis in the subiculum (error expressed in AUC graph): NTg air=open triangles, NTg UFP=closed triangles, 3xTgAD air=open circles, 3xTgAD UFP=closed circles. (C) Area under the curve for the subiculum Sholl graph shown in (B). Mean ± SEM, n=4–5, *p<0.05.

Figure 5-

Sholl analysis in the cortex of 3xTgAD and NTg mice. (A) Sholl analysis in the cortex (error expressed in AUC graph): NTg air=open triangles, NTg UFP=closed triangles, 3xTgAD air=open circles, 3xTgAD UFP=closed circles. (B) Area under the curve for the cortex Sholl graph shown in (A). Mean ± SEM, n=4–5.

Figure 6-

Quantitation of microglia and plaques in confocal images of subiculum. (A) Average number of microglia in each image. (B) Average number of plaques in each image. (C) Average number of microglia that were touching, surrounding, or otherwise associated with plaques. (D) Percentage of microglia in each image that were associated with plaques. (E) Average number of microglia in each image that were not associated with plaques. (F) Average number of microglia per plaque. Mean ± SEM, n=4–5, *p<0.05.

Figure 7-

Measures of Aβ in 3xTgAD or NTg mice exposed to filtered air or UFP. (A) Soluble and (B) insoluble Aβ protein levels from hippocampal homogenates. Mean ± SEM, n=6–7. (C) Quantitation of 6E10 percent area stained. (D) Quantitation of 6E10 percent area stained with analysis restricted to the subiculum. Representative images of 6E10 staining in (E) a NTg mouse that was exposed to filtered air, (F) a NTg mouse that was exposed to UFP, (G) a 3xTgAD mouse exposed to filtered air, and (H) a 3xTgAD mouse exposed to UFP. Mean ± SEM, n=4–5, *p<0.05.

Figure 8-

Hippocampal pT205 staining in mice exposed to filtered air or UFP. (A) Quantitation of pT205 percent area stained. Representative images of pT205 staining in (B) a NTg mouse that was exposed to filtered air, (C) a NTg mouse that was exposed to UFP, (D) a 3xTgAD mouse exposed to filtered air, and (E) a 3xTgAD mouse exposed to UFP. Mean ± SEM, n=4–5, *p<0.05.