Behavioral comparison of 4 and 6 month-old Ts65Dn mice: age-related impairments in working and reference memory.

Ts65Dn mice are partially trisomic for a segment of murine chromosome 16 similar to the gene segment on human chromosome 21 affected in Down's syndrome (DS). These animals display cognitive deficits, neurochemical imbalances, and cholinergic degeneration resembling alterations in DS and early onset Alzheimer's disease. The loss of basal forebrain cholinergic phenotype in Ts65Dn mice begins at approximately 6 months of age and may be due to an improperly functioning neurotrophic system. We compared 4 and 6 month-old Ts65Dn mice in a water-escape radial-arm maze task to investigate working and reference memory before and after the reported onset of cholinergic decline. Both 4 and 6 month-old Ts65Dn mice exhibited impaired performance compared to age-matched controls. However, the younger Ts65Dn mice displayed the capability to learn all working and reference memory measures, while the older Ts65Dn mice did not. Ts65Dn mice failed to maintain performance as working memory load increased, and the ability to handle an increasing working memory load also diminished with age. Collectively, these data suggest that major alterations in cognitive function occur in Ts65Dn mice between the ages of 4 and 6 months.

Age-related deficits as working memory load increases: relationships with growth factors.

Young and aged female rats were tested on a water radial-arm maze designed to measure performance as working memory load increased, followed by brain-derived neurotrophic factor (BDNF), nerve growth factor (NGF), and neurotrophin 3 (NT3) protein assessments in hippocampus and frontal cortex. Aged rats showed deficiencies in both working and reference memory. There were also profound age-related working memory load effects. Aged rats made more errors as working memory load increased and showed learning only during early trials when memory load was low, while young rats exhibited learning over all trials. Neurotrophin assessment showed that frontal cortex NGF and BDNF levels were positively, and hippocampal NT3 negatively, correlated with number of errors made during specific trials in aged animals. Comparison to untested rats showed that testing increased NT3, but not BDNF or NGF, protein levels in both age groups. Findings suggest that young rats learn to handle a higher working memory load as testing progresses, while aged rats do not, and that frontal cortex and hippocampal neurotrophin levels may relate to working memory proficiency in aged female rats.

Estrogen restores cognition and cholinergic phenotype in an animal model of Down syndrome.

Estrogen maintains normal function of basal forebrain (BF) cholinergic neurons and estrogen replacement therapy (ERT) has therefore been proposed as a therapy for Alzheimer's disease (AD). We provide evidence to support this hypothesis in an animal model of Down syndrome (DS), a chromosome 16 segmental trisomy (Ts65Dn) mouse. These mice develop cholinergic degeneration similar to young adults with DS and AD patients. ERT has not been tested in women with DS, even though they are more likely than normosomic women to develop early menopause and AD. Female Ts65Dn and normosomic mice (11-15 months) received a subcutaneous estrogen pellet or a sham operation. After 60 days, estrogen treatment improved learning of a T-maze task and normalized behavior in the Ts65Dn mice in reversal learning of the task, a measure of cognitive flexibility. Stereological evaluation of choline acetyltransferase (ChAT) immunopositive BF neurons showed that estrogen increased cell size and total number of cholinergic neurons in the medial septum of Ts65Dn mice. In addition, estrogen increased NGF protein levels in the BF of trisomic mice. These findings support the emerging hypothesis that estrogen may play a protective role during neurodegeneration and cognitive decline, particularly in cholinergic BF neuronal systems underlying cognition. The findings also indicate that estrogen may act, at least partially, via endogenous growth factors. Collectively, the data suggest that ERT may be a viable therapeutic approach for women with DS coupled with dementia.

Spatial memory testing decreases hippocampal amyloid precursor protein in young, but not aged, female rats.

Using young and aged rats, we investigated relationships between amyloid precursor protein (APP) and working or reference memory, as well as assessed whether cognitive testing altered APP levels. In young rats, higher APP levels were related to more working memory errors as a linear function. Aged rats exhibited a curvilinear relationship between APP and working memory, with moderate APP levels associated with better relative performance. A comparison of rats that received cognitive testing with those that did not showed that testing decreased APP levels in young, but not aged, rats. Collectively, the data suggest that young and aged rats exhibit different relationships between APP and working memory, and that aged rats do not maintain the capacity to decrease APP in response to cognitive testing.

Spatial and nonspatial Morris maze learning: impaired behavioral flexibility in mice with ectopias located in the prefrontal cortex.

About half of BXSB/MpJ-Yaa (BXSB) mice have neocortical ectopias (misplaced clusters of neurons located in layer I of cortex). Previous behavioral studies have suggested that ectopic mice have superior spatial, but equivalent nonspatial, reference memory learning. However, since spatial and nonspatial learning were not assessed in the same apparatus and with the same testing procedure, it is unclear if this conclusion is accurate. We have created a new nonspatial Morris maze for mice that differs from the spatial task only in the type of cues that must be utilized to efficiently locate the platform (intra-maze black/white patterns vs. extra-maze room cues) and does not differ in the level of task complexity or the presence of objects within the maze. Ectopic mice were very good in utilizing extra-maze cues when learning the spatial version and in utilizing intra-maze cues when learning the nonspatial version of the Morris maze, while non-ectopics were not, suggesting that ectopics have superior spatial and nonspatial reference memory. Ectopias in BXSB mice are usually located in prefrontal and/or motor cortex. The prefrontal cortex is involved in behavioral flexibility (e.g. being able to easily switch from using spatial to nonspatial cues). Only ectopic mice with ectopias specifically located in the prefrontal region of cortex demonstrated difficulty switching from using extra-maze to intra-maze cues and vice versa. Thus, the presence of one or more ectopias in the prefrontal region of cortex disrupted one of the normal functions of the prefrontal cortex.

Differential learning strategies in spatial and nonspatial versions of the Morris water maze in the C57BL/6J inbred mouse strain.

We recently developed a new nonspatial version of the Morris water maze that requires the use of four visually distinct intra-maze patterns to efficiently locate a hidden platform. The nonspatial version was designed to match the spatial version on complexity of cue usage, and differs only on spatiality of cues, thereby allowing more meaningful comparisons between the two versions. Following a previous experiment that demonstrated nonspatial learning with the BXSB inbred mouse strain, C57 inbred mice were tested in this study. They received spatial and nonspatial training in a counter-balanced order so that Test Order and information transfer could be assessed. Subjects that received spatial training first had superior performance in both the spatial and the nonspatial tasks when compared to mice that received nonspatial training first. The mice that received spatial training first used extra-maze cues as a spatial strategy. However, during nonspatial testing they did not use the intra-maze cues to locate the platform; instead, the mice used an egocentric strategy of circling through the platform annulus. Subjects that received spatial testing first were superior on the nonspatial task to those subjects that received nonspatial training first. Moreover, subjects that received nonspatial testing first were unable to learn the spatial version. Overall, C57 mice can learn both the spatial and nonspatial versions of the Morris maze presented here; however, the nonspatial version is more difficult and is solved using an egocentric strategy.