Stimulation-dependent intraspinal microtubules and synaptic failure in Alzheimer's disease: a review.

There are many microtubules in axons and dendritic shafts, but it has been thought that there were fewer microtubules in spines. Recently, there have been four reports that observed the intraspinal microtubules. Because microtubules originate from the centrosome, these four reports strongly suggest a stimulation-dependent connection between the nucleus and the stimulated postsynaptic membrane by microtubules. In contrast, several pieces of evidence suggest that spine elongation may be caused by the polymerization of intraspinal microtubules. This structural mechanism for spine elongation suggests, conversely, that the synapse loss or spine loss observed in Alzheimer's disease may be caused by the depolymerization of intraspinal microtubules. Based on this evidence, it is suggested that the impairment of intraspinal microtubules may cause spinal structural change and block the translocation of plasticity-related molecules between the stimulated postsynaptic membranes and the nucleus, resulting in the cognitive deficits of Alzheimer's disease.

Amyloid beta: a putative intra-spinal microtubule-depolymerizer to induce synapse-loss or dentritic spine shortening in Alzheimer's disease.

A loss or shortening of dendritic spines has been described in patients with neurodegenerative disorders such as Alzheimer's disease, but the underlying mechanisms are poorly understood. Recently, there have been four reports of capture of the plus-ends of microtubules in the dendritic spines. One report, based on acute hippocampal slices that were fixed by a microtubule preserving process after LTP-inducing stimulation, showed that microtubules of the dendritic shaft ramified into spines in a manner that was specific to the stimulated postsynaptic membranes. This resulted in enlarged protrusion of the dendritic spines. Other reports using living cultured neurons, showed that growing microtubule plus-ends enter spines and modulate spine morphology. Since microtubules originate from the centrosome, these four reports strongly suggest a stimulation-dependent connection between the nucleus and the stimulated postsynaptic membrane by microtubules. Several pieces of evidence suggest that spine elongation may be caused by microtubule polymerization. Firstly, the entry of plus-ends of microtubules into spines accompanies spine enlargement. Further, microtubule-associated protein-1B is over-expressed in Fragile X syndrome, in which spines are much elongated. Chronic stress causes neurite outgrowth and spine elongation. Polymerization of microtubules caused neurite outgrowth and microtubules-depolymerizing agents neurite retraction, both consistent with the proposition that spine elongation is caused by microtubule polymerization. This structural mechanism for spine elongation suggests, conversely, that synapse loss or spine shortening observed in Alzheimer's disease may be caused by depolymerization of intraspinal microtubules. The fact that a new drug, dimebon, shows promising results against memory disturbance in Alzheimer's patients and can also cause neurite outgrowth in cultured neurons may also support this idea. Amyloid activates GSK-3beta and it causes the abnormal hyperphosphorylation of tau and depolymerization of axonal microtubules, resulting in the impairment of axonal transport. Normal tau is mainly present in the axon, but hyperphosphorylated tau newly distributes to the dendrites and sequesters normal tau, MAP1A/MAP1B and MAP2, and may cause disruption of intraspinal microtubules by losing the microtubule-preserving effect of MAPs. Nevertheless, it may be strongly suspected that amyloid beta may be a putative intra-spinal microtubule-depolymerizer to induce spine shortening, synaptic loss and finally the memory disturbance in Alzheimer's disease.

Redistribution of microtubules in dendrites of hippocampal CA1 neurons after tetanic stimulation during long-term potentiation.

It is now well accepted that the trafficking of AMPA receptors to the postsynaptic plasma membrane plays an essential role in long-term potentiation at the hippocampal Schaffer collateral synapses on CA1 pyramidal cells, but the motor mechanism of trafficking is unknown. We suspected that this trafficking of AMPA receptors during long-term potentiation may be carried out along microtubules by their motors. To ascertain this hypothesis, we light- and electron-microscopically studied the distribution of microtubules in dendrites of CA1 neurons of non-stimulated and stimulated rat hippocampal slices by using very strong tetanic stimulation for inducing long-term potentiation. As a result, we observed the following changes: 1. In immunofluorescence for microtubules and IP3 receptor using ultrathin-cryosections, linear signals of microtubules in main dendritic shafts were changed into fragmented. 2. Many spotty signals of microtubules emerged at the peripheral area of dendrites. Electron-microscopically, there was redistribution of microtubules in dendritic spines and dendritic shafts, and the thickening of post-synaptic density. 3. Many microtubules concentrated to thickened postsynaptic density in spines and new ones emerged, going to spines from dendritic shafts. These results strongly suggest that new tracks of microtubules from cell bodies to the stimulated postsynaptic membranes were produced after tetanic stimulation during long-term potentiation. This newly produced microtubules between stimulated postsynaptic membranes and the cell body must be the most promising candidate of the track for the trafficking of AMPA receptors to the stimulated postsynaptic plasma membrane.

Microtubules to form memory.

Microtunbule-depolymerizing agents cause amnesia. Some signal translocations to the stimulated postsynaptic membrane are essential for inducing LTP in CA1 neurons like AMPA receptors, CaMKII and mRNA. On the other hand, LTP requires protein synthesis and gene expression. This indicates that signals generated at the synapse might be transmitted to the nucleus. Recently, we have reported that LTP-producing stimulation makes new microtubule track between cell body and the stimulated postsynaptic membrane in CA1 neurons. This newly produced microtubule track only to the stimulated postsynaptic membrane might be the route of these bi-directional transportation of signals during LTP formation. This lead us the hypothesis of the "endless memory amplifying circuit" that means gene expression-promoting molecules are translocated from postsynaptic membrane to the cell body and enter into nucleus and activate transcription factors, and gene products, which will probably promote plasticity, may be re-translocated only to the stimulated postsynaptic membrane along microtubules.

Microinjected F-actin into dividing newt eggs moves toward the next cleavage furrow together with Ca2+ stores with inositol 1,4,5-trisphosphate receptor in a microtubule- and microtubule motor-dependent manner.

It has been reported that F-actin is transported to the presumptive cleavage furrow along the cortex during anaphase-cytokinesis, an event termed cortical actin flow in animal cultured cells. The motor source has remained unknown. We reported that Ca2+ stores with IP3 receptor (IP3R) was re-distributed from the polar cortex during metaphase to the presumptive cleavage furrow just before the onset of furrowing, and that Ca2+ stores with IP3R microinjected into dividing newt eggs moved toward the presumptive cleavage furrow during anaphase-cytokinesis in a microtubule-dependent manner, and that Ca2+ store-enriched microsome fractions induced the cleavage furrow as the putative cleavage stimulus. Because the distribution of F-actin and Ca2+ stores with IP3R during metaphase to cytokinesis is similar, we considered that this cortical actin flow may be powered by transportation of Ca2+ stores with IP3R. Purified F-actin labeled with phalloidin-rhodamine was microinjected into the dividing newt eggs and the eggs observed under a confocal microscope. We found that the microinjected F-actin moved linearly toward the next cleavage furrow and that this movement was blocked by nocodazole, microtubule-depolarizing agent and AMP-PNP, a blocking agent of microtubule motors. Co-microinjected rhodamine-labeled F-actin and sacro/endoplasmic reticulum Ca2+-ATPase (SERCA)-GFP-labeled Ca2+ stores with IP3R co-moved and co-accumulated to the next cleavage furrow. These results strongly suggest that Ca2+ stores with IP3R, which is transferred by microtubule-based motility as cleavage stimulus, act as an F-actin translocator.