Search

Understanding the molecular and cellular changes that underlie memory, the engram, requires the identification, isolation and manipulation of the neurons involved. This presents a major difficulty for complex forms of memory, for example hippocampus-dependent declarative memory, where the participating neurons are likely to be sparse, anatomically distributed and unique to each individual brain and learning event. In this paper, I discuss several new approaches to this problem. In vivo calcium imaging techniques provide a means of assessing the activity patterns of large numbers of neurons over long periods of time with precise anatomical identification. This provides important insight into how the brain represents complex information and how this is altered with learning. The development of techniques for the genetic modification of neural ensembles based on their natural, sensory-evoked, activity along with optogenetics allows direct tests of the coding function of these ensembles. These approaches provide a new methodological framework in which to examine the mechanisms of complex forms of learning at the level of the neurons involved in a specific memory.

Mice lacking NMDA receptors in the dentate gyrus and CA1 subfields of the hippocampus form spatial memories just as well as wild-type mice, but they disregard them when confounded by ambiguous local cues. Hippocampal NMDA receptors may influence spatial memory more subtly than previously thought.

The stabilization of long-term memories requires de novo protein synthesis. How can proteins, synthesized in the soma, act on specific synapses that participate in a given memory? We studied the dynamics of newly synthesized AMPA-type glutamate receptors (AMPARs) induced with learning using transgenic mice expressing the GluR1 subunit fused to green fluorescent protein (GFP-GluR1) under control of the c-fos promoter. We found learning-associated recruitment of newly synthesized GFP-GluR1 selectively to mushroom-type spines in adult hippocampal CA1 neurons 24 hours after fear conditioning. Our results are consistent with a "synaptic tagging" model to allow activated synapses to subsequently capture newly synthesized receptor and also demonstrate a critical functional distinction in the mushroom spines with learning.

Quantitative analysis of synaptic proteomes from specific brain regions is important for our understanding of the molecular basis of neuroplasticity and brain disorders. In the present study we have optimized comparative synaptic proteome analysis to quantitate proteins of the synaptic membrane fraction isolated from the hippocampus of wild type mice and 3'UTR-calcium/calmodulin-dependent kinase II alpha mutant mice. Synaptic proteins were solubilized in 0.85% RapiGest and digested with trypsin without prior dilution of the detergent, and the peptides from two groups of wild type mice and two groups of CaMKIIalpha 3'UTR mutants were tagged with iTRAQ reagents 114, 115, 116, and 117, respectively. The experiment was repeated once with independent biological replicates. Peptides were fractionated with tandem liquid chromatography and collected off-line onto MALDI metal plates. The first iTRAQ experiment was analyzed on an ABI 4700 proteomics analyzer, and the second experiment was analyzed on an ABI 4800 proteomics analyzer. Using the criteria that the proteins should be matched with at least three peptides with the highest CI% of a peptide at least 95%, 623 and 259 proteins were quantified by a 4800 proteomics analyzer and a 4700 proteomics analyzer, respectively, from which 249 proteins overlapped in the two experiments. There was a 3 fold decrease of calcium/calmodulin-dependent kinase II alpha in the synaptic membrane fraction of the 3'UTR mutant mice. No other major changes were observed, suggesting that the synapse protein constituents of the mutant mice were not substantially altered. A first draft of a synaptic protein interaction network has been constructed using commercial available software, and the synaptic proteins were organized into 10 (interconnecting) functional groups belonging to the pre- and postsynaptic compartments, e.g., receptors and ion channels, scaffolding proteins, cytoskeletal proteins, signaling proteins, adhesion molecules, and proteins of synaptic vesicles and those involved in membrane recycling.

Local protein translation in dendrites could be a means for delivering synaptic proteins to their sites of action, perhaps in a spatially regulated fashion that could contribute to plasticity. To directly test the functional role of dendritic translation of calcium/calmodulin-dependent protein kinase IIalpha (CaMKIIalpha) in vivo, we mutated the endogenous gene to disrupt the dendritic localization signal in the mRNA. In this mutant mouse, the protein-coding region of CaMKIIalpha is intact, but mRNA is restricted to the soma. Removal of dendritic mRNA produced a dramatic reduction of CaMKIIalpha in postsynaptic densities (PSDs), a reduction in late-phase long-term potentiation (LTP), and impairments in spatial memory, associative fear conditioning, and object recognition memory. These results demonstrate that local translation is important for synaptic delivery of the kinase and that local translation contributes to synaptic and behavioral plasticity.

To examine the physiological role of the Fyn tyrosine kinase in neurons, we generated transgenic mice that expressed a fyn cDNA under the control of the calcium/calmodulin-dependent protein kinase IIalpha promoter. With this promoter, we detected only low expression of Fyn in the neonatal brain. In contrast, there was strong expression of the fyn-transgene in neurons of the adult forebrain. To determine whether the impairment of long-term potentiation (LTP) observed in adult fyn-deficient mice was caused directly by the lack of Fyn in adult hippocampal neurons or indirectly by an impairment in neuronal development, we generated fyn-rescue mice by introducing the wild-type fyn-transgene into mice carrying a targeted deletion in the endogenous fyn gene. In fyn-rescue mice, Schaffer collateral LTP was restored, even though the morphological abnormalities characteristic of fyn-deficient mice were still present. These results suggest that Fyn contributes, at least in part, to the molecular mechanisms of LTP induction.

To relate different forms of synaptic plasticity to the formation and maintenance of place cells in the hippocampus, we have recorded place cells in freely behaving, transgenic mice that express a mutated Ca2+-independent form of CaM Kinase II. These mice have normal long-term potentiation (LTP) at 100 Hz, but they lack LTP in response to stimulation at 5-10 Hz and are impaired on spatial memory tasks. In these transgenic mice, the place cells in the CA1 region have three important differences from those of wild types: they are less common, less precise, and less stable. These findings suggest that LTP in the 5-10 Hz range may be important for the maintenance of place-field stability and that this stability may be essential for the storage of spatial memory.

Using the phage P1-derived Cre/loxP recombination system, we have developed a method to create mice in which the deletion (knockout) of virtually any gene of interest is restricted to a subregion or a specific cell type in the brain such as the pyramidal cells of the hippocampal CA1 region. The Cre/loxP recombination-based gene deletion appears to require a certain level of Cre protein expression. The brain subregional restricted gene knockout should allow a more precise analysis of the impact of a gene mutation on animal behaviors.

To investigate the function of the autophosphorylated form of CaMKII in synaptic plasticity, we generated transgenic mice that express a kinase that is Ca2+ independent as a result of a point mutation of Thr-286 to aspartate, which mimics autophosphorylation. Mice expressing the mutant form of the kinase show an increased level of Ca(2+)-independent CaMKII activity similar to that seen following LTP. The mice nevertheless exhibit normal LTP in response to stimulation at 100 Hz. However, at lower frequencies, in the range of 1-10 Hz, there is a systematic shift in the size and direction of the resulting synaptic change in the transgenic animals that favors LTD. The regulation of this frequency-response function by Ca(2+)-independent CaMKII activity seems to account for two previously unexplained synaptic phenomena, the relative loss of LTD in adult animals compared with juveniles and the enhanced capability for depression of facilitated synapses.

We assessed hippocampal-dependent memory in mice with a Ca(2+)-independent form of CaMKII generated by the introduction of an aspartate at amino acid 286. The CaMKII-Asp-286 mice show normal LTP at high frequency stimulation, but in the 5-10 Hz range, they show a shift in the frequency-response curve favoring LTD. This range of frequencies is similar to the theta rhythm, which is associated with exploration in rodents. Using the Barnes maze to assess spatial memory, we found the transgenic mice could not learn to navigate to a specific location using spatial cues. In contrast, one line of transgenic mice performed normally in contextual fear conditioning, a task that is also hippocampal dependent. This dissociation between spatial and contextual memory suggests that even though both require the hippocampus, they may be mediated by different synaptic mechanisms.