Next, we investigated whether the variability in the A?? plaque burden at each age of testing also reflects inverse association with context and tone fear memory. We found that variability in A?? small molecule plaque burden (expressed by the coefficient of variation (CV)) increased with age, from 17% at three months to 34%, and 27% at six and 12 months, respectively. Moreover, the variability in memory scores of CRND8 mice differed from the variability in memory of nTg littermates. While the variance in memory scores of nTg mice was low and comparable across age groups (73%, 76%, 60%, and 38%, 49%, 50% for three, six, and 12 month context and tone memory, respectively), the variability in memory scores of CRND8 mice was higher, reaching high variance (CV > 100%) at the age of 12 months (70%, 93%, 144%, and 53%, 75%, 113% for three, six, and 12 month context and tone memory, respectively).

While our analysis revealed no significant association between the A?? plaque burden and the context fear memory in CRND8 mice at any age of testing, surprisingly, we found positive associations between A?? plaque burden and tone fear memory at three (rS = 0.67, P < 0.05) and six months of age (rS = 0.80, P < 0.01), but not at 12 months of age (rS = 0.30, NS). Despite a much reduced sample size, these post-hoc analyses revealed that at the ages of three and six months, which are characterized by rapid A?? plaque formation, those CRND8 mice which showed more A?? plaques, also showed higher tone fear memory. At present, these preliminary results have to be interpreted with caution.

These findings should be replicated in future studies and the relationship between the soluble A?? and A?? sequestered in plaques, and also other processes, such as reactive gliosis and inflammatory responses should be systematically evaluated in order to elucidate further the relationship between amyloid-?? and cognition at the early stage of plaque formation. Discussion The present results extend previous studies showing that other APP mouse models exhibit impairment in fear conditioned memory, by demonstrating that this impairment is progressive and correlates well with overall A?? burden. Also, the demonstrated greater sensitivity of the foreground tone conditioning test in the identification of age dependent onset of the memory impairment in CRND8, suggests that this testing paradigm might be particularly suitable in studies evaluating potential AV-951 therapeutic agents related to memory improvement in APP mouse models.

APP transgenic mouse models have been reported to show memory deficits similar to those observed in AD [21,48-53]. However, comprehensive cognitive profiles, including multiple memory systems, have often been based on comparative analyses from several independent clearly studies using APP mouse models (see [21]).