Abstract
Long-term potentiation (LTP) is an activity-dependent long-lasting increase in the efficacy of synaptic transmission. LTP is vital for learning and memory and must be regulated to keep synaptic strength within a dynamic range. LTP is proposed to be regulated in part by metaplasticity, i.e., the activity-dependent changes in neuronal state that orchestrate the magnitude of future synaptic plasticity. Previously, a type of metaplasticity has been described in the hippocampus called heterodendritic metaplasticity in which strong high-frequency priming stimulation in stratum oriens inhibits subsequent LTP in the stratum radiatum of area CA1. This form of metaplasticity is dependent on the release of Ca2+ from intracellular stores during priming and independent of postsynaptic depolarization, N-methyl-D-aspartate receptors (NMDARs) activity, and postsynaptic action potential firing. Moreover, the non-selective gap junction blocker carbenoxolone, an astrocyte-specific peptide inhibitor of connexin-43 channels, plus an adenosine A2B receptor antagonist, all block the priming effect on LTP inhibition. Evidence so far suggests that astrocytes are activated in area CA1 of the hippocampus following priming stimulation in SO. All these experimental data indicate the involvement of non-neuronal cells, possibly astrocytes, in mediating the heterodendritic metaplasticity effect.
One objective of this thesis addressed what signaling molecule may be feeding back to neurons to inhibit future LTP. Our electrophysiology experiments demonstrated that blocking tumor necrosis factor α (TNFα) (but not interleukin-1β) and bath-applying antagonists of signaling molecules known to be activated by TNFα such as p38 mitogen-activated protein kinase (p38 MAPK), c-Jun N-terminal kinase (JNK), and extracellular signal-regulated kinase (ERK) blocked the heterodendritic metaplasticity effect. In addition, our western blot data complemented the electrophysiology findings and p38 MAPK, ERK, and JNK were significantly phosphorylated in the primed group when compared to the non-primed group. Surprisingly, the metaplastic inhibition of LTP was also protein-synthesis dependent, mediated via the mTOR pathway. However, the SO LTP was unaffected by the protein synthesis inhibitors, suggesting possible differential regulation of synaptic plasticity in apical and basal dendrites.
The other objective of this thesis was to examine whether heterodendritic metaplasticity is dysregulated in APP/PS1 mouse model of the AD before (young mice) and after (aged mice) amyloid plaque formation. Since basal inhibition of LTP is reported in aged APP/PS1 mice, we predicted aberrant engagement of heterodendritic metaplasticity in aged mice but not in young mice. Surprisingly, LTP was already impaired in young mice, and this was perhaps due to the presence of soluble Aβ oligomers at this age in the APP/PS1 mice. TNFα antibody blocked the basal inhibition of LTP in aged mice, but not the basal impairment of LTP in young mice. In contrast, a block of IL-1β function did not affect LTP in aged APP/PS1 mice. This suggests TNFα-dependent, basal aberrant engagement of LTP in aged mice. However, the basal impairment of LTP in young mice was TNFα-independent. Consistent with the electrophysiological results, ELISA results demonstrated a significant elevation in the concentration of TNFα in aged Tg mice when compared to age-matched wt littermates, however, no such difference was observed in young mice.
The SO LTP was impaired in aged but not in young APP/PS1 mice, perhaps indicating the age-dependent progressive loss of synaptic plasticity, at least in this part of the hippocampus. Interestingly, the TNFα antibody did not affect the impaired SO LTP in aged APP/PS1 mice indicating the involvement of TNFα-independent mechanisms mediating the impairment of SO LTP. Overall, our findings support a vital role of TNFα, possibly released from astrocytes, in mediating heterodendritic metaplasticity in rat and mouse hippocampus, and blocking TNFα might be of therapeutic value in AD.