Age-related toxicity of amyloid-beta associated with increased pERK and pCREB in primary hippocampal neurons: reversal by blueberry extract

Abstract

Further clarification is needed to address the paradox that memory formation, aging and neurodegeneration all involve calcium influx, oxyradical production (ROS) and activation of certain signaling pathways. In aged rats and in APP/PS-1 mice, cognitive and hippocampal Ca(2+) dysregulation was reversed by food supplementation with a high antioxidant blueberry extract. Here, we studied whether neurons were an important target of blueberry extract and whether the mechanism involved altered ROS signaling through MAP kinase and cyclic-AMP response element binding protein (CREB), pathways known to be activated in response to amyloid-beta (Aβ). Primary hippocampal neurons were isolated and cultured from embryonic, middle-age or old-age (24 months) rats. Blueberry extract was found to be equally neuroprotective against Aβ neurotoxicity at all ages. Increases in Aβ toxicity with age were associated with age-related increases in immunoreactivity of neurons to pERK and an age-independent increase in pCREB. Treatment with blueberry extract strongly inhibited these increases in parallel with neuroprotection. Simultaneous labeling for ROS and for glutathione with dichlorofluorescein and monochlorobimane showed a mechanism of action of blueberry extract to involve transient ROS generation with an increase in the redox buffer glutathione. We conclude that the increased age-related susceptibility of old-age neurons to Aβ toxicity may be due to higher levels of activation of pERK and pCREB pathways that can be protected by blueberry extract through inhibition of both these pathways through an ROS stress response. These results suggest that the beneficial effects of blueberry extract may involve transient stress signaling and ROS protection that may translate into improved cognition in aging rats and APP/PS1 mice given blueberry extract.

Fig. 1

Neuron culture model of aging with virtually equal yields and regeneration of adult neurons from old brain (24 mo.) as younger ages. Cultures from rat hippocampus of A) embryonic day 18, B) middle-age (9–11 mo.) and C) old (22–24 mo.) imaged live at 8 days in culture.

Fig. 2

Concurrent treatment with blueberry extract rescues age-related toxicity of Aβ in adult neurons at 8 days in culture. Compared to cultures treated with blueberry extract alone (0.125 mg/mL, open squares) for 24 hr., treatment with fibrillar Aβ (10 µM, solid squares) for 24 hr. causes an age and treatment-dependent increase in dead neurons. Co-treatment of cultures with both Aβ and blueberry extract (solid circles) greatly attenuates the killing caused by treatment with Aβ alone (ANOVA Aβ vs. Blueberry extract + Aβ: p < 10⁻⁴). N= 12 fields, 3 from each of 4 cultures at each condition.

Fig. 3

Age-related increases of neuronal PKCα and PKCγ to Aβ, but failure of blueberry extract to reverse changes that result from treatment of adult neurons with Aβ. (A) Compared to untreated neurons (open circles) or neurons treated with blueberry extract (open squares), treatment with Aβ increased pPKCγ at all ages. Concurrent treatment with Aβ + blueberry extract (solid circles) partially reversed the Aβ-associated increase in embryonic neurons, but failed to affect the increase in either middle-age or old neurons. (B) Treatment with Aβ reduces the pPKCα signal in embryonic neurons, but raises the signal in adult neurons. Concurrent treatment with Aβ + blueberry extract (solid circles) reversed the Aβ-associated decrease in embryonic neurons, partially reversed the increase in middle-age neurons, but failed to reverse the increase seen in old neurons. Digital analysis of immunofluorescence of 30–90 individual neurons from 4 cultures of each age and treatment.

Fig. 4

Immunostain for pCREB (green), pERK1/2 (red) and nuclear DNA (blue) in neurons from old rat brain. (A) Strong cytoplasmic immunoreactivity for pERK (red) and punctate nuclear pCREB (green) in untreated old neurons. (B) Treatment with Aβ for 24 hr. increases the red cytoplasmic pERK as well as green pCREB together with red pERK in the nucleus (green+red+blue=white puncta). (C) Concurrent treatment with Aβ reduces the cytoplasmic pERK and nuclear pCREB closer to untreated conditions.

Fig. 5

Digital analysis shows neuron immunoreactivity for pCREB is elevated with treatment with Aβ (filled squares), but largely restored to baseline (open symbols) by treatment with Aβ + blueberry extract (closed circles). (A) Whole cell analysis. (B) Nuclear density of pCREB for middle-age (open bars) and old neurons (closed bars).

Fig. 6

Digital analysis shows neuron immunoreactivity for pERK is elevated with treatment with Aβ (filled squares), but partially reversed toward baseline (open symbols) by treatment with Aβ + blueberry extract (closed circles). (A) Whole cell analysis. (B) Separate immunostains for panERK and pERK for middle-age (open bars) and old neurons (closed bars) shows little change in the ratio of the staining, suggesting that the changes seen in figure 6A for pERK are not due to more total ERK per cell.

Fig. 7

Treatment with blueberry extract raises neuronal glutathione without raising ROS, while treatment with Aβ raises ROS, mostly without changing glutathione in live neurons. (A) Untreated old neurons with moderate blue stain with monocholorobimane (MCB) for glutathione and low green stain with DCF for ROS (colors in A, B and C inverted). (B) Blueberry extract for 30 min. raises blue glutathione signal with little change in green ROS. (C) Treatment with Aβ for 30 min. greatly elevates ROS levels (green) with only a few neurons resistant to the elevated ROS with associated higher levels of glutathione (arrows). Digital analysis of image at 10 min. intervals for neurons treated with (D) blueberry extract rapidly increases ROS levels in all three ages of neurons. Treatment with Aβ (E) causes a rapid but transient increase in ROS in embryonic and middle-age neurons, but a prolonged increase in old neurons. In all three ages, the addition of blueberry extract with Aβ (F) reduced ROS production over the 30 min. period.

Fig. 8

(A)Tracking individual old neurons for coincident changes in DCF density (ROS) and MCB (glutathione) after 30 min. reveals four populations of neurons for each of three treatments. Note that concurrent treatment with Aβ and blueberry extract (x) greatly increases the population of neurons with high glutathione and low ROS (quadrant 4). In contrast to treatment with Aβ alone (solid circles) with high proportions of ROS have both low and high glutathione levels, or untreated neurons (open circles) with low levels of both ROS and glutathione. (B) Time course of population in quadrant 2 with low glutathione and high ROS shows a sustained increase with time for treatment with Aβ alone (open squares) that is reversed by concurrent treatment with Aβ and blueberry extract. (C) Time course of population in quadrant 4 with high glutathione and low ROS shows a transient increase with time for treatment with Aβ alone (open squares) that grows into a much larger population by concurrent treatment with Aβ and blueberry extract.