ApoE APP PS-2 ApoE4
+/- developmental defects, -/- lethal; Forebrain:decreased hippocampal neurogenesis (104, 105) Memory and LTP deficits (106, 107) Hippocampal gliosis (108) Modest phenotype (109)
No learning phenotype, accelerated plaque disposition when combined with APP V717F(66, 110)
Tg2576 mice exhibit many behavioral and pathological features of AD including elevated production of Ap peptides, age-dependent accumulation of amyloid fibrils, and plaque formation with subsequent age-dependent hippocampal learning and memory deficits (45, 46, 70, 72-75). This mouse model demonstrates a correlation between hippocampal dysfunction at both the cellular and behavioral levels versus its increased burden of extracellular Ap42(43), as I will describe briefly next.
At 6 months of age and older, mice exhibit a deficit in spatial memory (Morris water maze). In addition to impairment in spatial memory at 14-16 months of age, mice also exhibit a working memory deficit as measured using a forced-alternation paradigm (75, 76). Tg2576 mice also exhibit deficits in contextual fear conditioning (see references 74 and 76 and Figure 5). Several
FIGURE 5 Fear conditioning in mouse models of AD. PS-1, APP, doubly transgenic, and control mice were subjected to a standard fear conditioning paradigm in which the animals learn to associate neutral stimuli with an aversive one. The mice were placed in a novel context (fear-conditioning box) and exposed to two pairings of a white-noise cue and mild foot shock. Fear learning was assessed 24 hours later by measuring freezing behavior in response to re-presentation of the context or of the auditory cue within a completely different context. At 5 months of age, there were no apparent differences in the freezing behavior of the different mouse genotypes during the two-pairing training phase of fear conditioning (top). In the contextual test for fear learning, the APP and doubly transgenic animals exhibited decreased freezing behavior compared to both the control littermates and PS-1 transgenic group (lower right). One-way ANOVA and Tukey post hoc analysis detected a significant difference in freezing behavior at the 1- through 4-minute time epochs compared to control littermates and at the 1-, 2-, 3-, and 5-minute epochs compared to the PS-1 transgenic group [min 1: F(3,60) = 5.60; min 2: F = 7.68; min 3: F = 8.51; min 4: F = 4.26; min 5: F = 4.35; p < .05 all groups]. Analysis of overall freezing behavior indicates that APP and doubly transgenic animals freeze significantly less than control and PS-1 transgenic animals [Tukey's multiple comparison test: F(5,16) = 27.97; control versus APP, control versus doubly, APP versus PS-1, doubly versus PS-1 all p < .001]. These data indicate that APP and doubly transgenic animals have a deficit in contextual fear learning. These same animals did not exhibit a deficit in cued fear conditioning (lower left). One-way ANOVA and Tukey post hoc analysis determined that all animals displayed similar and significant freezing in the cued test for associative learning, indicating that the impairment in contextual fear learning exhibited by the APP and doubly transgenic animal groups is not the result of an inability to freeze or to detect the aversive foot-shock stimulus (p < .001, all groups). Therefore, 5-month-old APP and doubly transgenic mice appear to have a selective hippocampus-dependent impairment in associative learning following two pairings of conditioned and unconditioned stimuli for fear conditioning. Adapted from Dineley et al. (74; see also reference 76).
AD-like pathologies are not present in these mice. For example, there is no neuronal loss in CA1 or other brain areas (37). Thus, these mice do not model AD-associated neuronal loss. Nonetheless, the age-dependent amyloidogenesis and working memory deficits are powerful correlates of AD and make this animal model an attractive launching point for investigations into the cellular signaling processes underlying neuronal dysfunction induced by an increased burden of extracellular A-beta.
It has also been reported that accompanying the behavioral deficits in working memory, Tg2576 mice exhibit disruptions of LTP in both the Schaffer collateral and perforant pathways of the hippocampus. Synaptic transmission and paired-pulse facilitation appear normal (indicating that Ca2+-dependent synaptic vesicle release is normal), and there is no decrease in the number of CA1 neurons or synaptic density in the dentate gyrus (75). Thus, in Tg2576 mice, selective impairments in synaptic plasticity correlate with deficits in cognitive abilities.
Our take-home message from all of this is that the Tg2576 mouse strain exhibits learning deficits and alterations in synaptic plasticity with no neuronal cell loss. In that respect, it, along with all other available mouse strains modeling AD, does not recapitulate one of the major hallmarks of late-stage AD—cell death. However, these mouse lines accurately model amyloidosis and likely model early-stage AD. The lack of cell death in all these lines has an important implication as well. Because there is no appreciable cell loss nor pronounced alterations in neuronal morphology, the impairment that leads to the memory phenotype is likely in the normal cellular signaling cascades involved in learning and synaptic plasticity.
What is the site of derangement in neuronal signaling that underlies the Tg2576 mouse phenotype? Ongoing work by Kelly Dineley in my lab, done in collaboration with a number of investigators including Karen Hsiao and Hui Zheng, suggests the hypothesis that the increased burden of extracellular Ap peptide in Tg2576 mice leads to derangement of hippocampal ERK MAPK activity (77). These derangements potentially result in subsequent learning and memory deficits in a fashion reminiscent of the various human mental retardation syndromes that we discussed in the last chapter. However, this is just one specific finding among a multitude of identified derangements in mouse lines modeling AD. I note the finding because, in my mind, it presents an interesting example of a potential AD-associated derangement in one of the principal memory-associated signal trans-duction pathway that we have been discussing throughout the book.
Crossing PS1 transgenic mice carrying the A246E FAD mutation with Tg2576 mice causes acceleration of CNS amyloid accumulation and plaque formation (see Figure 4 and Table 2). The doubly trans-genic animals also have exacerbated associative learning deficits relative to the Tg2576 transgenic mice. These data indicate that, as might be expected, there is an interaction of the APP and PS-1 gene products in vivo. The point in raising this observation is twofold. First, over time their will likely be improvement in modeling AD in mice through these kinds of mix-and-match genetic experiments. Second, this result illustrates a use of genetically engineered mouse models of AD that is independent of their utility as a model for human disease per se. That is, the emergence of these various transgenic mouse lines will allow the testing of various specific predictions of current working models for how the various gene products identified as relevant to amyloid beta peptide production interact in the living animal.
Overall, the studies published so far indicate cause for optimism that transgenic mouse models for AD will be of great utility both in studying the biochemistry and physiology of the disease and in assessing potential new treatment avenues for AD (see Box 3). Important work in the near future will allow further evaluation and optimization of mouse models for investigating AD. In addition, as we obtain further basic information concerning human AD (for example, identification of additional genetic factors predisposing us to AD), further enhancement of progress in mouse models is likely.
These practical considerations should not completely overshadow the additional important implications of transgenic mouse experiments, however. The transgenic mouse studies to date can be looked at as experiments testing predictions of the amyloid beta hypothesis of AD. The measure experiments using human AD tissues identified an association of Ap with
AD. Transgenic mouse experiments are the mimic experiments. Thus far, studies of transgenic mice overexpressing Ap peptide certainly indicate that Ap is sufficient to cause many of the pathological hallmarks and cognitive features of early AD. These are important experimental results in their own right.
Was this article helpful?
It seems like you hear it all the time from nearly every one you know I'm SO stressed out!? Pressures abound in this world today. Those pressures cause stress and anxiety, and often we are ill-equipped to deal with those stressors that trigger anxiety and other feelings that can make us sick. Literally, sick.