Soluble oligomeric amyloid-β (Aβ) is thought to induce synaptic dysfunction during early stages of Alzheimer's disease (AD). In this report, we show that soluble Aβ downregulates the levels of two synaptic proteins, PSD-95 and synaptophysin, and that this effect can be blocked by MK-801 (NMDAR antagonist) and ifenprodil (NR2B antagonist). Low (1 μM) and high (10 μM) doses of NMDA, respectively, prevented and potentiated the actions of Aβ. Blockade of NR2A or synaptic NMDAR eliminated the protective effect of 1 μM NMDA, while the effects of 10 μM NMDA were only abolished by ifenprodil. Caspase-8, acting upstream of caspase-3, was found to mediate the synaptotoxic actions of Aβ in an ifenprodil-reversible fashion. Thus, Aβ leads to a loss of synaptic proteins by suppression of NR2A function and activation of NR2B function and subsequent induction of caspase-8 and caspase-3 activities. The identified novel mechanism through which Aβ initiates synaptic dysfunction suggests that selective enhancement of NR2A activity and/or reduction of NR2B activity can halt the manifestation of a key early-stage event in AD.

The dorsal spinocerebellar tract (DSCT) of the thoracic cord (the thoracic DSCT) consists of uncrossed ascending axons originating from Clarke's column, marginal neurons of Clarke's column, and lamina V neurons, and crossed ascending axons originating from neurons in lamina VIII and the ventromedial part of lamina VII. The present study has examined, by using anterograde labeling with biotinylated dextran, whether the thoracic DSCT projects to the cerebellar nuclei. The tracer was injected into the thoracic cord, for two to four segments at levels between the T4 and T9 segments in the rat. The distribution of anterogradely labeled axons and terminals was bilateral but predominantly ipsilateral to the injections. Labeled axons entered the medial nucleus from its rostromedial aspect and terminated widely in medial and ventral parts of the middle subdivision. Furthermore, they terminated in the medial and ventral part of the caudomedial subdivision. Labeled terminals were seen in rostromedial parts of the anterior interpositus nucleus and in medial to caudal parts of the posterior interpositus nucleus. A small number of labeled terminals were consistently seen in the ventral part of the lateral nucleus and the dorsolateral hump region. The present study shows that the thoracic DSCT projects bilaterally, but predominantly ipsilaterally, to the medial and the anterior and posterior interpositus nuclei and suggests that it conveys input related to posture and movement of the trunk and respiratory movement of the thorax.

The different expressions of NR2 and synaptic-associated proteins have been studied by protein and mRNA level with immunoblotting, in situ hybridization, or immunogold analysis. But the relationship between NR2 subunits and PSD-95 family proteins is still controversial. In this study, we used immunofluorescent staining to assess NR2B and PSD-95 expressions and the relationship between them in CA1, CA3, and DG of rat hippocampus on different postnatal day. In CA1, NR2B expression decreased with age. It was high at birth, reached a plateau at P4, and declined gradually after P7. In CA3, NR2B expression was similar to that in CA1. But the stratum lucidum was devoid of staining. In DG, the NR2B expression retained a higher level. From P0 to P2, the PSD-95 expression in CA1 increased gently, and then declined slightly. After P7, the PSD-95 expression increased sharply till P28, and decreased again. In CA3 and DG, the PSD-95 expression is very similar except that low-level of PSD-95 was found in the CA3 stratum lucidum. The expression of NR2B did not correlate with that of PSD-95 in CA1 and the DG granular and molecular layer. Only in CA3 and DG polymorphic layer, there was a negative correlation. The results suggested in hippocampal subregions, CA3 and DG may be more plastic than CA1.The NR2B and PSD-95 expression have distinct regional and cell specific distribution. The different regional distribution pattern may relate to the different physiological functions during postnatal development.

Projections from the spinal cord to the vestibular nuclei were examined following injections of Phaseolus vulgaris-leucoagglutinin, cholera toxin subunit B, or biotinylated dextran at various levels of the spinal cord in the rat. Labeled terminals were abundant after injections of the tracers into the C2 and C3 segments containing the central cervical nucleus. Labeled terminals were seen in the descending vestibular nucleus and the parvocellular, magnocellular, and caudal parts of the medial vestibular nucleus throughout its rostrocaudal extent. Labeled terminals were most numerous in the lateral vestibular nucleus throughout its rostrocaudal extent. The projections from the central cervical nucleus to the vestibular nuclei were exclusively contralateral to the cells of origin because the axons of the central cervical nucleus neurons cross in the spinal cord. Following tracer injections in the cervical enlargement, many labeled terminals were seen in the magnocellular part of the medial vestibular nucleus, but a few were seen in the lateral and the descending vestibular nucleus. Injections into more caudal segments resulted in sporadic terminal labeling in the magnocellular part of the medial vestibular nucleus, the descending vestibular nucleus, and the caudal part of the lateral vestibular nucleus. The results indicate that primary neck afferent input relayed at the central cervical nucleus is mediated directly to the contralateral vestibular nuclei. It is suggested that this projection serves as an important linkage from the upper cervical segments to the lateral vestibulospinal tract in the tonic neck reflex.

No abstract