PKMζ is necessary and sufficient for synaptic clustering of PSD-95.

The persistent activity of protein kinase Mzeta (PKMζ), a brain-specific, constitutively active protein kinase C isoform, maintains synaptic long-term potentiation (LTP). Structural remodeling of the postsynaptic density is believed to contribute to the expression of LTP. We therefore examined the role of PKMζ in reconfiguring PSD-95, the major postsynaptic scaffolding protein at excitatory synapses. In primary cultures of hippocampal neurons, PKMζ activity was critical for increasing the size of PSD-95 clusters during chemical LTP (cLTP). Increasing PKMζ activity by overexpressing the kinase in hippocampal neurons was sufficient to increase PSD-95 cluster size, spine size, and postsynaptic AMPAR subunit GluA2. Overexpression of an inactive mutant of PKMζ did not increase PSD-95 clustering, and applications of the ζ-pseudosubstrate inhibitor ZIP reversed the PKMζ-mediated increases in PSD-95 clustering, indicating that the activity of PKMζ is necessary to induce and maintain the increased size of PSD-95 clusters. Thus the persistent activity of PKMζ is both necessary and sufficient for maintaining increases of PSD-95 clusters, providing a unified mechanism for long-term functional and structural modifications of synapses.

Protein kinase M zeta regulation of Na/K ATPase: a persistent neuroprotective mechanism of ischemic preconditioning in hippocampal slice cultures.

In ischemic preconditioning, a sublethal ischemic insult protects neurons from subsequent ischemia. In organotypic hippocampal slice cultures a sublethal 5-minute hypoxia-hypoglycemia treatment prevented neuronal loss after a 10-minute experimental ischemic (EI) treatment of hypoxia-hypoglycemia. Whereas preconditioning protected against EI given 24 h later, it did not protect when EI was given 2 h later, suggesting a slow development of neuroprotection. This model identified two regulators of ischemic preconditioning: the atypical protein kinase M zeta (PKMzeta), and the Na/K ATPase. Two hours following preconditioning, when there was no neuroprotection, Na/K ATPase activity was unchanged. In contrast, Na/K ATPase activity significantly increased 24 h after the preconditioning treatment. Elevated Na/K ATPase activity was accompanied by increased surface expression of the alpha1 and alpha2 isoforms of the Na/K ATPase. Similarly, active PKMzeta levels were increased at 24 h, but not 2 h, after preconditioning. PKMzeta overexpression by sindbis virus vectors also increased Na/K ATPase activity. To examine PKMzeta regulation of Na/K ATPase, occlusion experiments were performed using marinobufagenin to inhibit alpha1, dihydroouabain to inhibit alpha2/3 and a zeta-pseudosubstrate peptide to inhibit PKMzeta. These experiments showed that PKMzeta regulated both the activity and surface expression of the alpha1 isoform of the Na/K ATPase. Marinobufagenin, dihydroouabain, and zeta-pseudosubstrate peptide were used to determine if PKMzeta or the alpha1 and alpha2 Na/K ATPase isoforms protected neurons. All three compounds blocked neuroprotection following ischemic preconditioning. PKMzeta levels were elevated 3 days after ischemic preconditioning. These data indicate key roles of PKMzeta and Na/K ATPase in ischemic preconditioning.

Actin polymerization regulates the synthesis of PKMzeta in LTP.

Changes in the actin-based synaptic cytoskeleton are important for long-term potentiation (LTP), but the mechanism linking actin filament formation and the persistence of enhanced synaptic transmission has not been described. Actin filaments form rapidly during LTP induction without new protein synthesis, but their effects on synaptic potentiation occur during protein synthesis-dependent late-LTP. Previous studies have shown that late-LTP is mediated by the synthesis of the persistently active, atypical PKC isoform, protein kinase Mzeta (PKMzeta). Here we show that preventing actin polymerization during LTP with latrunculin B blocks de novo protein synthesis of PKMzeta. In contrast, the agent has minimal effects on the potentiation of AMPA receptors mediated by postsynaptically perfused PKMzeta or on LTP expression. Thus actin filaments formed by tetanization enhance the efficiency of the synthesis of PKMzeta that maintains LTP.

The G protein-coupled 5-HT1A receptor causes suppression of caspase-3 through MAPK and protein kinase Calpha.

The 5-HT(1A) agonist 8-hydroxy-2 (di-n-propylamino) tetralin (8-OH-DPAT) causes inhibition of caspase-3 and apoptosis via the extracellular signal-regulated kinases (ERK1/2) in hippocampal HN2-5 cells. Two 5-HT(1A) agonists, Repinotan hydrochloride (BAY x 3702) and 8-OH-DPAT, block caspase-3 activation and apoptosis caused by anoxia/reoxygenation and H(2)O(2) treatment. This is reversed upon transient expression of dominant negative Ras (N17Ras) and Raf-1 (Raf301), confirming the involvement of Ras and Raf-1 in this 5-HT(1A)-R-->ERK1/2-->caspase-3 pathway. A selective inhibitor of phospholipase Cbeta (PLCbeta) (U73122) but not a general protein kinase C (PKC) inhibitor (GFX) reversed the 5-HT(1A)-R-mediated ERK1/2 stimulation. However, both GFX and the PKCalpha and PKCbeta(1) inhibitor Gö6976 reversed the ERK1/2-mediated inhibition of caspase-3. ERK-dependent activation of only PKCalpha was observed in immunoprecipitates obtained from 5-HT(1A) agonist-treated HN2-5 cells. Finally, transient expression of kinase-negative PKCalpha eliminated the 8-OH-DPAT-evoked block on the H(2)O(2)-triggered caspase-3 stimulation, establishing PKCalpha as a link between ERK and caspase-3 (5-HT(1A)-R-->PLC-->ERK1/2-->PKCalpha-->caspase-3). Our results elucidate a novel yet general, neuroprotective pathway through which G protein-coupled receptors could cause inhibition of effector caspases, such as caspase-3.