Treadmill Exercise Reduces Neuroinflammation, Glial Cell Activation and Improves Synaptic Transmission in the Prefrontal Cortex in 3 × Tg-AD Mice

Abstract

Physical exercise improves memory and cognition in physiological aging and Alzheimer's disease (AD), but the mechanisms remain poorly understood. Here, we test the hypothesis that Aβ oligomer accumulation, neuroinflammation, and glial cell activation may lead to disruption of synaptic transmission in the prefrontal cortex of 3 × Tg-AD Mice, resulting in impairment of learning and memory. On the other hand, treadmill exercise could prevent the pathogenesis and exert neuroprotective effects. Here, we used immunohistochemistry, western blotting, enzyme-linked immunosorbent assay, and slice electrophysiology to analyze the levels of GSK3β, Aβ oligomers (Aβ dimers and trimers), pro-inflammatory cytokines (IL-1β, IL-6, and TNFα), the phosphorylation of CRMP2 at Thr514, and synaptic currents in pyramidal neurons in the prefrontal cortex. We show that 12-week treadmill exercise beginning in three-month-old mice led to the inhibition of GSK3β kinase activity, decreases in the levels of Aβ oligomers, pro-inflammatory cytokines (IL-1β, IL-6, and TNFα), and the phosphorylation of CRMP2 at Thr514, reduction of microglial and astrocyte activation, and improvement of excitatory and inhibitory synaptic transmission of pyramidal neurons in the prefrontal cortex of 3 × Tg-AD Mice. Thus, treadmill exercise reduces neuroinflammation, glial cell activation and improves synaptic transmission in the prefrontal cortex in 3 × Tg-AD mice, possibly related to the inhibition of GSK3β kinase activity.

Keywords: 3 × Tg-AD mice; Aβ oligomer; GSK3β; exercise; glial cell; inflammatory factor; synaptic plasticity.

Figure 1

Treadmill exercise alleviated the Aβ pathology of the prefrontal cortex in 3 × Tg-AD mice. (A) Timeline of treadmill exercise or non-exercise control, histological test, and electrophysiology. (B) Representative microphotographs of the prefrontal cortex sections from 6-month-old mice stained with an anti-Aβ specific antibody (6E10, n = 8/group). No Aβ plaques were found in the prefrontal cortex sections. (C) Representative western blots for Aβ dimers, Aβ trimers and GAPDH of prefrontal cortex homogenates were prepared from these four groups of mice. Red in (C) represents western Blot image overexposure. The Aβ protein content was low, so the marker protein image showed overexposure. (D) Summarized data showed that Aβ trimers (*** p < 0.001, n = 8) in the prefrontal cortex were significantly increased in the 3 × Tg-AD mice compared to the non-Tg control group, and this increase was blocked by treadmill exercise pretreatment (*** p < 0.001, n = 8). See Supplementary Figure S1 for uncropped blots.

Figure 2

Treadmill exercise inhibited the kinase activity of GSK3β in 3 × Tg-AD mice. (A,B) Representative western blots for p-GSK3β-Tyr216 (A), p-GSK3β-Ser9 (B), and GSK3β (A,B) of prefrontal cortex homogenates were prepared from these four groups of mice. (C) Summarized data showed that p-GSK3β-Tyr216/GSK3β in the prefrontal cortex were significantly increased in the 3 × Tg-AD mice compared to the non-Tg control group (*** p < 0.001, n = 8-8) and this increase was blocked by treadmill exercise pretreatment (*** p < 0.001, n = 8-8). (D) Summarized data showed that p-GSK3β-Ser9/GSK3β in the prefrontal cortex were significantly decreased in the 3 × Tg-AD mice compared to the non-Tg control group (*** p < 0.001, n = 8-8), and this decrease was blocked by treadmill exercise pretreatment (*** p < 0.001, n = 8-8). See Supplementary Figure S2 for uncropped blots.

Figure 3

Treadmill exercise reduced astrocyte and microglia activation in 3 × Tg-AD mice. (A) Representative western blots for GFAP and CD68 of the prefrontal cortex homogenates were prepared from these four groups of mice. (B,C) Summarized data showed that GFAP ((B), *** p < 0.001, n = 8-8 and CD68 ((C), ** p = 0.003, n = 8-8) of the prefrontal cortex were significantly increased in the 3 × Tg-AD mice compared to the non-Tg control group and these increases were blocked by treadmill exercise pretreatment (*** p < 0.001, n = 8-8). (D) Representative microphotographs of the prefrontal cortex sections from six-month-old mice. Sections were stained antibody against GFAP (n = 10/group). Blue arrows represent GFAP positive cells. (E) Quantitative analysis revealed that the number of GFAP-positive cells was significantly higher in 3 × Tg-AD mice compared to the non-Tg control group (*** p < 0.001, n = 10-10), and this increase was blocked by treadmill exercise pretreatment (** p = 0.001, n = 10-10). See Supplementary Figure S3 for uncropped blots.

Figure 4

Treadmill exercise decreased the pro-inflammatory cytokines and p-CRMP2 in 3 × Tg-AD mice. (A–C) ELISA analysis of IL-1β (A), IL-6 (B), and TNFα (C) concentrations in the prefrontal cortex. The concentrations of IL-1β ((A), *** p < 0.001; n = 8-8), IL-6 ((B), *** p < 0.001; n = 8-8), and TNFα ((C), *** p < 0.001; n = 8-8) in the prefrontal cortex were significantly increased in the 3 × Tg-AD control group compared to the non-Tg control group, and these increases were blocked by treadmill exercise pretreatment (*** p < 0.001, n = 8-8). (D) Representative western blots for p-CRMP2-Thr514 and CRMP2 in the prefrontal cortex homogenates were prepared from these four groups of mice. (E) Quantitative analysis revealed that the levels of p-CRMP2-Thr514/CRMP2 was significantly higher in 3 × Tg-AD mice compared to the non-Tg control group (*** p < 0.001, n = 8-8), and this increase was blocked by treadmill exercise pretreatment (*** p < 0.001, n = 8-8). See Supplementary Figure S4 for uncropped blots.

Figure 5

Treadmill exercise blocked the reduction of synaptic transmission of 3 × Tg-AD Mice. (A,D) Representative mEPSCs (A) and mIPSCs (D) were recorded from prefrontal cortex pyramidal neurons in slices prepared from the non-Tg control group, non-Tg exercise group, 3 × Tg-AD control group, 3 × Tg-AD exercise group mice. (B,C) The mean frequency (B) and amplitude (C) of mEPSCs in the prefrontal cortex in these four groups of mice. The mean frequency (* p = 0.014, n = 15-15 and amplitude (*** p < 0.001, n = 15-15) of mEPSCs were significantly decreased in the 3 × Tg-AD mice compared to the non-Tg control group, and these decreases were attenuated by treadmill exercise pretreatment (frequency, * p = 0.012, n = 15-15; amplitude, *** p < 0.001, n = 15-15). (E,F) The mean frequency (E) and amplitude (F) of mIPSCs in the prefrontal cortex in these four groups of mice. The mean frequency and amplitude of mIPSCs were significantly decreased in the 3 × Tg-AD mice compared to the non-Tg control group (*** p < 0.001, n = 15-15), and these decreases were attenuated by treadmill exercise pretreatment (*** p < 0.001, n = 15-15). Treadmill exercise increased the mean frequency (* p = 0.012, n = 15-15) and amplitude of mIPSC (** p = 0.007, n = 15-15) in non-Tg mice. Each data set was obtained from 6 mice.

Figure 6

GSK3β inhibitor (ARA-014418) increased synaptic transmissions of 3 × Tg-AD mice. (A) Representative mEPSCs and mIPSCs were recorded from prefrontal cortex pyramidal neurons in slices prepared from mice microinjected with vehicle or ARA-014418. (B,C) The mean frequency (B) and amplitude (C) of mEPSCs in the prefrontal cortex in these two groups of mice. The mean frequency (** p = 0.004, n = 15-15) and amplitude (*** p < 0.001, n = 15-15) of mEPSCs were significantly increased in the ARA-014418 group compared to the vehicle group. (D,E) The mean frequency (D) and amplitude (E) of mIPSCs in the prefrontal cortex in these two groups of mice. The mean frequency (*** p < 0.001, n = 15-15) and amplitude (** p = 0.001, n = 15-15) of mIPSCs were significantly increased in the ARA-014418 group compared to the vehicle group. Each data set was obtained from 6 mice.