Abstract

This laboratory investigates the signal transduction in synaptic transmission and plasticity by biochemical, behavioral, and electrophysiological approaches using a genetically modified mice model. A strain of mutant mice was established by deleting the gene coding for a neural-specific protein, neurogranin (Ng). This protein is expressed at high levels in cerebral cortex, hippocampus, and amygdala, and has been implicated in the modulation of synaptic plasticity. Ng is a small molecular weight protein, which binds calmodulin (CaM) at low level of calcium under basal physiological conditions. The binding of Ng and CaM is weakened upon neuronal stimulation that results in an increase in calcium. When dissociated from CaM, Ng is phosphorylated by PKC and/or oxidized by nitric oxide and other oxidants. The phosphorylated and oxidized Ngs are poor binding partners of CaM. These multifaceted regulations of Ng provide a fine tune mechanism to set the levels of free calcium and calcium/CaM depending on the strength of stimulation to the neurons and, thus, gate the output response. Neuronal stimulation that leads to the enhancement of synaptic plasticity requires the activation of calcium- and calcium/CaM-dependent signaling pathways, which are hypothesized to be modulated by Ng through its interaction with CaM. Previous in vitro studies have provided ample evidences to support this hypothesis;however, the interaction of Ng and CaM in vivo has yet to be proven. We have employed acute hippocampal slices to investigate the interaction of these two proteins. Fluorescence immunohistochemical staining of hippocampal slices bathed in calcium-containing artificial cerebral spinal fluid (ACSF) revealed that Ng and CaM were co-localized in the soma and dendrites of principle neurons. In the CA1 region the concentrations of CaM and Ng in the soma were greater than those in the dendrites. Surprisingly, majority of the somatic CaM was sequestered in the nucleus. In contrast, Ng was abundantly present in the soma as well as in the dendrites. Changing the bathing fluid from the calcium-containing to the calcium-free ACSF resulted in a suppression of synaptic transmission with a concomitant redistribution of CaM and Ng from soma to dendrites. Confocal calcium-imaging showed that a reduction of merely 15 and 40 nM of intracellular calcium were sufficient to cause half-maximum translocation of Ng and CaM, respectively, from soma to dendrites. Switching the bathing fluid back to the calcium-containing ACSF restored the synaptic transmission and the original compartmentalization of these two proteins. The hippocampal CA1 pyramidal neurons were the most responsive to this calcium-sensitive translocation as compared to their neighboring CA2 and CA3 neurons. These studies illustrated the unique sensitivity of the hippocampal CA1 neurons in the mobilization of CaM and Ng between soma and dendrites. These findings also provide a positive proof that Ng and CaM respond to changing intracellular calcium in a coordinated manner. CaM and Ng are abundant neuronal proteins in the forebrain and deletion of Ng gene in mice causes deficits in learning the hippocampus- and amygdala-dependent behavioral tasks, and the high-frequency stimulation (HFS)-induced long-term potentiation (LTP). The concentration of Ng in the hippocampus was approximately twice that of CaM and the latter protein was largely concentrated in the nucleus and much less in the dendrites. The asymmetrical distribution of CaM in hippocampal neurons was in contrast to that of Ng, which was abundantly present in both soma and dendrites. Thus, in the distal dendrites binding of CaM at a lower concentration by a relatively higher concentration of Ng renders very little CaM available for the activation of CaM-dependent enzymes. We explored the possibility that the HFS could trigger the mobilization of CaM from soma to dendrites for the maintenance of LTP. Tetanic stimulation (a single train of 1 s, 100 Hz) of the Schaffer-collateral fiber of the hippocampal slices caused an increase of CaM in the dendrites. The HFS-mediated mobilization of CaM was detected surrounding the stimulating electrode within the stratum radiatum but not near a second electrode that didnt deliver HFS. Both CaM and Ng was found to associate with the dendritic spines nearby the stimulating electrode. For Ng KO mice, the same HFS didnt generate LTP and there was no mobilization of CaM from soma to dendrites. Redistribution of CaM from soma to dendrites was also observed in the phorbol ester-induced synaptic facilitation in these tissues. It seems that synaptic stimulation induced by the activation of PKC could also trigger the exit of CaM from soma (nucleus) to dendrites. These findings suggest that association of CaM and Ng at the stimulated dendritic spines may enhance the synaptic efficacy by increasing the calcium transients as predicted by the mass-action mechanism. The association of CaM and Ng at the stimulated synapses may also serve as synaptic tags for the stimulated dendritic branches. Deletion of Ng in mice caused deficits in cognitive functions and the HFS-induced LTP in hippocampal slices. Further characterization revealed that these animals also exhibited other behavioral abnormalities, including hyperactivity, inattentiveness, and impulsivity. These behavioral phenotypes were likely resulting from disruption of the Ng-regulated signaling. One of the most prominent roles of Ng is the enhancement of the NMDA-receptor mediated calcium-transients. Thus, we predict that stimulation of down stream signaling components post calcium influx or increase of the presynaptic transmitter release may rescue the deficits of Ng KO mice. In neurons, stimulation of PKCs is known to enhance the transmitter release and facilitate the synaptic responses. Treatment of hippocampal slices from Ng KO mice with the PKC-activating phorbol ester, phorbol-12, 13-dibutylate (PDBu), caused a persistent synaptic facilitation. The PDBu-mediated effects were most prominent among those tissue slices from the dorsal hippocampus, which is an area thought to associate with cognitive functions. In conjunction with these in vitro studies, we have tested the treatment of Ng KO mice with Ritalin, a psychostimulant drug known to increase the extracellular neurotransmitters. Four groups of animals kept in an enriched environment, including control and drug-treated wild type and Ng KO mice, were injected with Ritalin (10 mg/kg/day, i.p.) for three weeks and, afterward, subjected to behavioral tests. The drug-treated Ng KO mice exhibited improvement in their cognitive functions as evidenced by a reduction of the latency time to locate the hidden platform in the water maze and an increase in the freezing time after fear-conditioning. Ritalin also appeared to reduce the hyperactivity of the Ng KO in the open field and an increase in the immobility time in the forced-swim chamber. The drug treatment, however, only had a marginal effect on the performances of the wild type mice. Measurement of the HFS (one train of 100 Hz for 1s)-induced LTP in the hippocampal CA1 region in vitro showed a positive effect of the drug on the Ng KO. These results indicate that Ritalin, a drug commonly used for the treatment of attention-deficit hyperactivity disorder (ADHD), can exert beneficial effects on the Ng KO as it does for the human patients. These studies also suggest that Ng KO mice could serve as an animal model for ADHD and they will be useful for the development of new treatment strategy for this illness.

  Funding Agency

Agency
National Institute of Health (NIH)
Institute
Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD)
Type
Investigator-Initiated Intramural Research Projects (ZIA)
Project #
1ZIAHD000187-30
Application #
7968467
Study Section
Project Start
Project End
Budget Start
Budget End
Support Year
30
Fiscal Year
2009
Total Cost
$717,252
Indirect Cost

  Institution

Name
Eunice Kennedy Shriver National Institute of Child Health & Human Development
Department
Type
DUNS #
City
State
Country
Zip Code