AD is a disease of lost memories. Studies in learning and memory field suggested that persistent increase in synaptic strength is used to achieve long-term modification of neuronal network properties and store memories. It has been proposed that the mushroom spines between excitatory neurons are stable “memory spines” that make functionally stronger synapses and therefore responsible for memory storage [
6]. By taking this one step further, we and others previously proposed that loss of mushroom spines may underlie cognitive decline during the progression of the AD [
7‐
11]. Recent experimental evidence in several cellular and animal models of AD provided support to this hypothesis. Fraction of mushroom spines was significantly reduced in hippocampal slice cultures from APP
SDL transgenic mice that express human APP695 with Swedish (KM595/596NL), Dutch (E618Q), and London (V642I) mutations under control of platelet-derived growth factor β promoter [
21]. More subtle model of amyloid toxicity was generated recently. In this model human APP transgene with Dutch mutation (E693Q) was expressed under control of neuronal Thy1 promoter (DU mice) [
23]. In contrast to most APP transgenic mice, DU mice generate soluble Aβ oligomers without formation of amyloid plaques [
23]. Such phenotype mimics phenotype of patients with Arctic (E693G) and E693Δ mutations that show AD phenotype in the absence of fibrillar amyloid accumulation [
24,
25]. Anatomical analysis of spines in CA1 area of hippocampus in DU mice revealed significant reduction in post-synaptic density (PSD) of mushroom spine synapses at 12 months of age [
11]. In the present study we observed significant reduction in the fraction of mushroom spines in neuronal cultures exposed to Aβ42 oligomers (Fig.
2) and in CA1 area (stratum radiatum) of hippocampus of the mice injected with Aβ42 oligomers (Fig.
5). Although Price at al have not reported reduction in the fraction of mushroom spines in their analysis of DU mice [
11], there was a trend to reduced mushroom spine density in their data (Fig.
3 in Price et al., 2014). Most likely concentration of Aβ42 oligomers in our study was higher than in the DU mice, leading to more dramatic destabilization of mushroom spines in our experiments. Interestingly, loss of mushroom spines is not unique to amyloid toxicity models. In the recent study we observed significant loss of mushroom spines in hippocampal neurons from PS1-M146V-KI mouse model [
10]. Thus, loss of mushroom spines appear to be a common feature of AD models, in agreement with our hypothesis [
7,
9].
Important to note that there is a correlation between dendritic spines alterations in CA1 area of hippocampus and memory impairments in different mice models of AD. Loss of mushroom spines in CA1 hippocampal neurons in PS1-M146V-KI [
10] may underlie disrupted late-phase LTP and age-related alterations in hippocampal spatial memory observed in these mice [
26]. Twelve month old DU mice have decreased PSD length and trend to reduced mushroom spine density in CA1 area as well as diminished performance in the water maze indicating that these mice exhibit perturbed hippocampus-associated spatial learning and memory [
11]. It has been demonstrated that the Aβ-mediated impairment of memory in rats is associated with lower density of synapses and altered synaptic structure in both the dentate gyrus and CA1 fields [
27].