Intrahippocampal Inoculation of Aβ1-42 Peptide in Rat as a Model of Alzheimer's Disease Identified MicroRNA-146a-5p as Blood Marker with Anti-Inflammatory Function in Astrocyte Cells

Abstract

Circulating microRNAs (miRNAs) have aroused a lot of interest as reliable blood diagnostic biomarkers of Alzheimer's disease (AD). Here, we investigated the panel of expressed blood miRNAs in response to aggregated Aβ_1-42 peptides infused in the hippocampus of adult rats to mimic events of the early onset of non-familial AD disorder. Aβ_1-42 peptides in the hippocampus led to cognitive impairments associated with an astrogliosis and downregulation of circulating miRNA-146a-5p, -29a-3p, -29c-3p, -125b-5p, and-191-5p. We established the kinetics of expression of selected miRNAs and found differences with those detected in the APP_swe/PS1_dE9 transgenic mouse model. Of note, miRNA-146a-5p was exclusively dysregulated in the Aβ-induced AD model. The treatment of primary astrocytes with Aβ_1-42 peptides led to miRNA-146a-5p upregulation though the activation of the NF-κB signaling pathway, which in turn downregulated IRAK-1 but not TRAF-6 expression. As a consequence, no induction of IL-1β, IL-6, or TNF-α was detected. Astrocytes treated with a miRNA-146-5p inhibitor rescued IRAK-1 and changed TRAF-6 steady-state levels that correlated with the induction of IL-6, IL-1β, and CXCL1 production, indicating that miRNA-146a-5p operates anti-inflammatory functions through a NF-κB pathway negative feedback loop. Overall, we report a panel of circulating miRNAs that correlated with Aβ_1-42 peptides' presence in the hippocampus and provide mechanistic insights into miRNA-146a-5p biological function in the development of the early stage of sporadic AD.

Keywords: Alzheimer’s disease; Aβ1–42 peptide; biomarkers; diagnosis; miRNA-146a-5p; microRNAs.

Figure 1

Effects of FAβ_1–42 infusion on learning and memory capacity evaluated by the Morris Water Maze (MWM) test. The distance and the escape latency performed by rats (n = 10) injected with FAβ_1–42 at a concentration of 1 µg/µL was evaluated at day 14 post-injection. (A) Total distance traveled and (B) Total escape latency used by rats during the 5 days of training. Control group (n = 10) of animals was injected with PBS. Data are represented as mean ± SEM. Statistical comparisons were performed using Kruskal–Wallis. * p ≤ 0.05.

Figure 2

Quantification of GFAP⁺ astrocytes in the hippocampal areas. Brain sections from rats infused with FAβ_1–42 (1 µg/µL) or PBS solutions were analyzed 14 days post-injection. (A1,A2) Representative immunohistochemical staining of GFAP expression (in green) in the DG area. The slides were counterstained with DAPI (blue). Scale bar, 500 µm. The box corresponds to GFAP staining with a higher magnification (B) Histograms showing the number of GFAP + cells per mm² quantified in the total hippocampus and (C) in each area of the hippocampus (CA1/CA2, CA3 and DG). The quantification of GFAP + cells was analyzed in 4–5 sections per animal (n = 5 for each group). Statistical comparisons between both groups were analyzed using a Student’s t-test. *** p ≤ 0.001.

Figure 3

Relative expression of circulating miRNA-146a-5p, miRNA-29a-3p, miRNA-9a-5p, miRNA-29c-3p in serum samples of rats infused with 1 µg/µL of FAβ_1–42 peptides solution. Sera from rats infused with FAβ_1–42 peptides (treated group) or PBS (control group) solutions were collected 21 days post-infusion. MiRNA were extracted and quantified by qRT-PCR. Data are represented as mean ± SEM of 7–8 samples evaluated in triplicate. Statistical comparisons between FAβ_1–42 and PBS group were performed using the Mann–Whitney test * p ≤ 0.05.

Figure 4

Relative expression of circulating miRNA-146a-5p, miRNA-29a-3p, miRNA-9a-5p, miRNA-29c-3p, miRNA-125b, miRNA-191-5p in serum of rats infused with 2.5 µg/µL of FAβ_1–42 peptide solution. Serum samples were collected 21 days post-infusion. MiRNAs were extracted and quantified by qRT-PCR. Data are represented as mean ± SEM of 7–8 samples evaluated in triplicate. Statistical comparisons between FAβ_1–42 and PBS group were performed using the Mann–Whitney test. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001.

Figure 5

Kinetics of circulating microRNAs detected in the serum of rats infused with FAβ _1–42 peptide solution used at 2.5 µg/µL final concentration. The quantification of miRNAs in serum of rats was investigated at 0, 7, 14, and 21 days post-infusion with FAβ_1–42 peptides or with PBS control solutions. Relative expression profiles of (A) miR-146a-5p, (B) miR-29a-3p, (C) miR-29c-3p, and (D) miR-9a-5p quantified by qRT-PCR performed in triplicates and expressed as mean ± SEM (n = 8 rats per group and for each time point). Statistical comparisons between groups at each time were made using the Mann–Whitney test. * p ≤ 0.05, ** p ≤ 0.01.

Figure 6

Relative expression of circulating miRNA-125b-5p, miRNA-29a, miRNA-29c, and miRNA-191-5p in serum of APP_swe/PS1_dE9 transgenic animal model of AD. (A) Serum samples from transgenic mice at 15 months of age were extracted and processed for miRNA detection by qRT-PCR. Data are represented as mean ± SEM of 5 sera samples performed in triplicate. (B) Table illustrating the statistically significant differences in terms of relative expression of miRNAs between the FAβ_1–42-infused animal model (day 21) and the APPswe/PS1dE9 transgenic animal model of AD. Statistical comparisons between the transgenic and control group were performed using the Mann–Whitney test. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001.

Figure 7

Relative expression of circulating miRNA-146a-5p in CSF of rats infused with FAβ_1–42 peptides solution. CSF samples from rats infused with FAβ (2.5 µg/µL) or PBS solutions were collected at 14 days post-injection. The amount of miR-146-5p was then quantified by qRT-PCR. Data are represented as mean ± SEM of n = 7 samples performed in triplicate. Statistical comparisons between the FAβ-infused and control group were performed using the Mann–Whitney test. ** p ≤ 0.01.

Figure 8

MiRNA-146a-5p is up-regulated in primary astrocytes treated with FAβ_1–42 peptides, and its expression is dependent on the NF-κB pathway. (A) Relative expression of miRNA-146a-5p in primary astrocytes was 0.5, 1, and 2 µM of FAβ_1–42 peptides for 3 days in tissue culture. (B) Relative expression of miRNA-146a-5p in primary astrocytes as function of treatment with 2 µM of FAβ_1–42 solution in the presence of 5 µM of BMS-345541, IκB kinase pharmacology inhibitor. As positive control, cells were treated with LPS (100 ng/mL). (C) Relative expression of IRAK-1/2 and TRAF-6 in astrocytes treated with 2 µM of FAβ_1–42 peptides. Data are represented as the mean ± SEM performed in triplicate. Statistical comparisons between FAβ_1–42-treated cells and control, DMSO-treated cells were performed with a Student’s t-test. * p ≤ 0.05, ** p ≤ 0.01.

Figure 9

Expression of inflammatory markers in astrocyte cells treated with FAβ_1–42 peptides solution. (A) Three days post-stimulation with FAβ solution (2 µM), the expression of IL-6, IL-1β, and TNF-α were quantified by RT-qPCR. (B) CXCL1 was quantified by ELISA. (C) Astrocytes were also stimulated with LPS at 100 ng/mL for 3 days as positive control. Data are represented as mean ± SEM of experiments made in triplicate. Statistical comparisons between groups were made using a Student’s t-test. * p ≤ 0.05.

Figure 10

Transfection of miRNA-146a-5p inhibitor in FAβ_1–42-treated astrocyte cells rescues expression of IRAK-1 and TRAF-6 but not IRAK-2. (A) qRT-PCR evaluation of performance of anti-miR-146a oligonucleotides (AMO-146a) to down-regulate expression of miRNA-146a-5p in astrocytes. (B) qRT-PCR evaluation of impact of AMO-146a transfection on expression of transcriptional targets of miRNA-146a-5p in FAβ_1–42-astrocyte-treated cells. Cells were first transfected with AMO-146a or AMO-CTL (100 nM), then treated with FAβ_1–42 peptides solutions (2 µM) before quantification of the relative expression of IRAK1/2 and TRAF-6 transcriptional targets of miRNA-146a-5p. Non-treated astrocytes (NT, no FAβ_1–42 treatment) were used as additional control. Data are represented as mean ± SEM of one representative experiment performed twice in triplicate. Statistical comparisons between groups were made using a Student’s t-test. * p ≤ 0.05, ** p ≤ 0.01.

Figure 11

Transfection of miRNA-146-5p inhibitor in FAβ_1–42-treated astrocytes rescues as well promote the expression of pro-inflammatory molecules. Astrocytes were first transfected with AMO-146a or AMO-CTL (100 nM), then treated with FAβ_1–42 peptides (2 µM) before quantification relative of the expression of (A) IL-6, (B) IL-1β, (C) TNF-α, and (D) CXCL1 by qRT-PCR. Data are represented as mean ± SEM of one representative experiment performed twice in triplicate. Statistical comparisons between groups were made using a Student’s t-test. * p ≤ 0.05, ** p ≤ 0.01.