Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 131 (Pt 3), 651-64

Cyclooxygenase-2 Inhibition Improves Amyloid-Beta-Mediated Suppression of Memory and Synaptic Plasticity

Affiliations

Cyclooxygenase-2 Inhibition Improves Amyloid-Beta-Mediated Suppression of Memory and Synaptic Plasticity

Linda A Kotilinek et al. Brain.

Abstract

Non-steroidal anti-inflammatory agents (NSAIDs) are associated with a marked reduction in the risk of developing Alzheimer's disease, a form of dementia characterized by the accumulation of amyloid plaques containing the amyloid-beta protein (Abeta). Studies of the effects of NSAIDs upon the inflammatory response surrounding amyloid plaques and upon the generation of Abeta from the amyloid precursor protein (APP) have led to two proposed mechanisms by which NSAIDs may protect against Alzheimer's disease: one, the selective lowering of Abeta42 by a subset of NSAIDs; and two, the reduction of inflammation. Although Alzheimer's disease is a disorder of brain and synaptic function, the effects of NSAIDs on Abeta-mediated suppression of synaptic plasticity and memory function have never been reported. We therefore investigated how three different NSAIDs, chosen for their distinct effects on Abeta42 production and the inhibition of the cyclooxygenase (COX) isoenzymes, COX-1 and COX-2, affect memory function and synaptic plasticity. By focusing upon brain and synapse function, we made novel observations about the effects of NSAIDs on Abeta-mediated neural processes. Here we report that the selective inhibition of COX-2, but not COX-1, acutely prevented the suppression of hippocampal long-term plasticity (LTP) by Abeta. The non-selective NSAIDs, ibuprofen and naproxen, and a selective COX-2 inhibitor, MF-tricyclic, each restored memory function in Tg2576 mice over-expressing APP, and also blocked Abeta-mediated inhibition of LTP. There was no advantage of ibuprofen, a selective Abeta42-lowering agent (SALA), over the non-SALAs, naproxen and MF-tricyclic. The beneficial effects on memory did not depend upon lowered levels of Abeta42 or the inflammatory cytokines, tumour necrosis factor alpha (TNF-alpha) and interleukin 1beta (IL-1beta). Intriguingly, improved memory function was inversely related to prostaglandin E2 (PGE2) levels. Conversely, exogenous PGE2 prevented the restorative effects of COX-2 inhibitors on LTP. The data indicate that the inhibition of COX-2 blocks Abeta-mediated suppression of LTP and memory function, and that this block occurs independently of reductions in Abeta42 or decreases in inflammation. The results lead us to propose a third possible mechanism by which NSAIDs may protect against Alzheimer's disease, involving the blockade of a COX-2-mediated PGE2 response at synapses.

Figures

Fig. 1
Fig. 1
Possible targets of NSAIDS in the prevention of Alzheimer’s disease and proposed pathophysiology of memory loss in the Tg2576 mouse model of Alzheimer’s disease. (a) The activities of enzymes and factors (blue and red lettering) are affected by NSAIDs in the direction indicated by the arrow in parentheses or by the delta sign signifying a qualitative alteration in activity. The cyclooxygenases COX-1 and COX-2 are major targets of NSAIDs because their inhibition can be achieved at doses between 1 and 10 μM. The activation or suppression of other targets, including peroxisome proliferators-activated receptors (PPARs) and nuclear factor-kB (NF-kB), generally requires higher concentrations (>50 μM) of NSAIDs. Non-selective NSAIDs inhibit both COX-1 and COX-2. The coxibs selectively inhibit COX-2. A subset of NSAIDs, including ibuprofen but not naproxen or the coxibs, lowers production of Aβ42, the highly amyloidogenic 42-residue form of Aβ, by modulating γ-secretase activity. (b) In Alzheimer’s disease COX-2 is stimulated by Aβ, glutamate and inflammatory cytokines, which in turn are modulated by COX-2.
Fig. 2
Fig. 2
Selective COX-2, but not COX-1, inhibitors prevent the inhibition of LTP induction by synthetic, soluble Aβ42. (a) Control LTP induced by a single brief high-frequency stimulation (HFS) (closed circles) and LTP induction in the presence of synthetic, soluble Aβ42 (500 nM), applied 40 min prior to HFS (open circles). LTP in the presence of synthetic, soluble Aβ42 was significantly reduced from control. (b) The COX-2 inhibitor MF tricyclic (3 μM), applied 60 min prior to HFS, does not inhibit LTP induction (closed circles) but prevents the Aβ-mediated inhibition of LTP induction (open circles). (c) The COX-2 inhibitor NS-398 (20 μM), applied 60 min prior to HFS, does not inhibit LTP induction (closed circles) but prevents the Aβ-mediated inhibition of LTP induction (open circles). (d) The COX-1 inhibitor piroxicam (10 μM), applied 60 min prior to HFS, does not inhibit LTP induction (closed circles) and also does not prevent the Aβ-mediated inhibition of LTP induction (open circles). The traces a, b and c are field EPSPs at the times indicated by a, b and c on the graphs, with the top set of traces corresponding to the closed circles and the lower set of traces to the open circles.
Fig. 3
Fig. 3
Reversal of memory deficits and restoration of spatial memory by administration of the conventional NSAIDs, ibuprofen and naproxen, and the selective COX-2 inhibitor, MF tricyclic, in Tg2576 mice. (a) Experimental design. Wedges and rectangles above the timeline qualitatively represent amyloid plaques (stippled), Aβ*56 (dark grey) and memory ability (black) inTg2576 mice as a function of age. Mice<2 months were too young to test. Bars underneath the timeline indicate the duration of treatment. Arrows indicate the initiation of behavioural tests. To avoid possible retest effects related to the short inter-test interval, Tests 1 and 2 were performed on different groups of mice. (b) Acquisition of memory in NSAID-treated and control Tg2576 mice and non-transgenic littermates. Data are the mean target quadrant occupancy during probe trials as a function of number of training trials. (c) The mean probe score (MPS) in NSAID-treated and control Tg2576 mice and non-transgenic littermates. MPS is the mean target quadrant occupancy in probes performed between the eighth and 17th trials. Random swimming in all three probe trials would yield an MPS of 25. Ages at initiation of testing are indicated. Data are the mean±SEM of the MPS for each mouse (*P<0.05; **P<0.01, compared to age-matched Tg2576 mice on control diet; #P< during the final three training trials were equivalent for 11.6-month and 12.5-monthTg2576 and non-transgenic littermates (data not shown).
Fig. 4
Fig. 4
Preservation of spatial memory by chronic administration of ibuprofen inTg2576 mice. (a) Experimental design. (b) Spatial working memory in ibuprofen-treated and control Tg2576 mice, assessed using the T-maze. Data are the mean±SEM of the percentage of correct responses averaged over two days for each mouse (F = 5.6, P = 0.02, for treatment effect by multiple-measures ANOVA, **P<0.01 by t-test). Numbers of mice are in parentheses. (c) Spatial reference memory in ibuprofen-treated and control Tg2576 mice, assessed using the Morris water-maze. Data are the mean±SEM of the MPS for each mouse (F = 5.13, *P = 0.03, for treatment effect by ANOVA).
Fig. 5
Fig. 5
TNF-α and IL-1β levels inTg2576 mice. (a) TNF-α is significantly elevated in 13-monthTg2576 mice. Data are the mean ± SEM of hippocampal TNF-α measured by ELISA (#P<0.01; *P<0.001). TNF-α is not significantly reduced in 13-monthTg2576 mice treated with NSAIDs (data not shown). (b) Age-related increase in IL-1β inTg2576 mice. Data are the mean±SEM of hippocampal IL-1β measured by ELISA (*P<0.001). (c) IL-1β is suppressed in 10.5-month Tg2576 mice given ibuprofen prophylactically (#P<0.01). (d) IL-1β remains significantly elevated in13-monthTg2576 mice treated with NSAIDs (**P<0.0001). Control (C), Ibuprofen (IB), Naproxen (Nap), MF tricyclic (MF). (e) Memory-intact 13-monthTg2576 mice treated with NSAIDs showed significantly higher levels of IL-1β than memory-impaired 10.5-month Tg2576 mice on control diet. (*P<0.001; **P<0.0001). A similar relationship was observed for TNF-α (data not shown). Numbers of mice denoted inside bars.
Fig. 6
Fig. 6
Aβ levels inTg2576 mice. (a) No significant effects of treatment on Aβ levels in mice treated with NSAIDs from 11.5 to 13 months of age. Forebrain Aβ40 and Aβ42 levels were measured by ELISA in SDS and formic acid (FA) fractions. Numbers of mice denoted inside bars. (b) No significant effects of treatment on Aβ42 levels in 2-month-old mice treated with NSAIDs for 30 days. Brains were extracted in SDS and Aβ42 levels were measured by ELISA. (c) Aβ*56 levels are not modulated by ibuprofen treatment. Tg2576 mice (+) and non-transgenic littermates () were treated with ibuprofen (Ib) or control (C) diets from 11.5 to 13 months of age. Forebrain Aβ*56 was measured in membrane-enriched extracts by Western blotting (WB; 100 μg protein/lane) using 6E10 monoclonal antibodies (1:10,000). CTFβ and monomeric Aβ (1-mer) levels also showed no significant alterations.
Fig. 7
Fig. 7
COX-2 mRNA and PGE2 levels inTg2576 mice. (a) COX-2 expression is not elevated inTg2576 mice. COX-2 mRNA was measured by in situ hybridization in 4-, 8- and 12-monthTg2576 mice (n = 10) and non-transgenic littermates (n = 9). Data are the mean±SEM of COX-2 mRNA in the CA3 region of the hippocampus (HIP) and frontal cortex (CTX). (b) PGE2 levels are significantly lower in 13-monthTg2576 mice. No significant effect of NSAID treatment on PGE2 levels was observed. Data are mean±SEM of forebrain (minus hippocampus) PGE2 measured by EIA (*P<0.01).
Fig. 8
Fig. 8
Relationship between PGE-2, TNF-α, IL-1β, Aβ levels and spatial reference memory inTg2576 mice. (a) 13-month transgene positive Tg2576 mice (Tg+) on NSAID or control diets were divided into three evenly sized bins based upon performance in the water maze (low, medium and high MPS categories). There was a significant effect of MPS category in both control and NSAID-treated mice (P<0.001 and P<0.0001, respectively. a:b, b:c, a:c within control and NSAID treatment groups represent significant Fisher’s PLSD post hoc comparisons. *P<0.01 by unpaired t-test.) (b) There was an inverse relationship between water maze performance (MPS) and cortical PGE-2 levels in NSAID-treated mice (P<0.05). *P<0.01 by Fisher’s PLSD. No relationship between MPS and PGE-2 levels was observed in mice receiving Control diet. (c–f) No relationship was observed between MPS and hippocampal TNF-α, hippocampal IL-1β, cortical total Aβ40 or cortical total Aβ42 respectively in control or NSAID-treated mice. Data are means ± SEM. Numbers of mice denoted inside bars.
Fig. 9
Fig. 9
Exogenous PGE2 blocks the ability of a selective COX-2 inhibitor to prevent the inhibition of LTP induction by synthetic, soluble Aβ42. (a) Control LTP induced by a single brief high-frequency stimulation (HFS) (closed circles) and LTP induction in the presence of PGE2 (5μM), applied 40 min prior to HFS (open circles). PGE2 at this concentration has no effect on LTP. (b) The COX-2 inhibitor NS398 (20 μM), applied 60 min prior to HFS prevents the Aβ-mediated inhibition of LTP induction (closed circles), but is unable to do so in the presence of 5 μM PGE2 (open circles).

Similar articles

See all similar articles

Cited by 80 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms

Grant support

Feedback