Elsevier

Neurobiology of Disease

Volume 13, Issue 3, August 2003, Pages 246-253
Neurobiology of Disease

Regular article
Dendritic spine loss in the hippocampus of young PDAPP and Tg2576 mice and its prevention by the ApoE2 genotype

https://doi.org/10.1016/S0969-9961(03)00079-2Get rights and content

Abstract

Postmortem AD brains exhibit dendritic spine loss in the hippocampus. To determine whether this pathology may be associated with amyloid burden, the present study used the Golgi stain technique to assess age- and genotype-dependent changes in dendritic spine density in CA1 hippocampus of two transgenic mouse lines that produce high levels of Aβ. Tg2576 and PDAPP mice, as well as a group of Tg2576 mice crossed with human apoE2-expressing transgenic mice, were compared to respective transgene-negative controls. Since the time course of amyloid plaque deposition in the PDAPP and Tg2576 mice is well characterized, we examined changes in spine density at ages that corresponded to different levels of amyloid plaque load. The data show age- and genotype-dependent reductions in spine density in both Tg2576 and PDAPP mice, albeit at somewhat different time courses. The spine loss occurred prior to plaque deposition and was ameliorated by the overexpression of human apoE2. These results suggest that a soluble Aβ species may affect hippocampal synapses and thereby contribute to functional deficits evident in these animals.

Introduction

The characteristic neuropathology of Alzheimer’s disease (AD) includes vascular and parenchymal amyloid deposits, neurofibrillary tangles (NFTs), synaptic loss, and neuronal death. Many, but not all, of these pathologies and corresponding behavioral disturbances of AD have been replicated in animal models of the disease. Mice that overexpress mutant forms of human amyloid-precursor protein (APP) develop amyloid plaques, dystrophic neurites, and some deficits in memory and learning in an age-dependent manner Hsiao et al 1996, Irizarry et al 1997, Chapman et al 2001, Arendash et al 2001. On the other hand, the loss of neurons or synapses demonstrated in postmortem AD brains (Davidsson and Blennow, 1998) is not consistently replicated in all transgenic models. Some groups have reported decreases in the number of cresyl-violet-stained neurons and diminished levels of microtubule-associated protein-2 (MAP-2) and synaptophysin in APP-transgenic mice Calhoun et al 1998, Hsia et al 1999, Mucke et al 2000, while others have not shown any difference from age-matched controls in each of these measures Irizarry et al 1997, Stephenson et al 1999, Takeuchi et al 2000. The inconsistencies related to synaptophysin immunolabeling may be attributed to the confounding presence of dystrophic terminals (Takeuchi et al., 2000).

We hypothesized that a morphological correlate of synapse number, such as dendritic spine density, may be better suited for assessing synaptic viability. As postsynaptic dendritic contacts, an increase or decrease in spine density appears to correlate with the growth or loss of afferent synaptic terminals Parnavelas et al 1974, Gould et al 1990, Cheng et al 1997. Specifically, dendritic spines have been shown to be the predominant site of excitatory, but not inhibitory, synapses on CA1 pyramidal cells in the rat hippocampus (Megias et al., 2001). Additionally, spine-mediated plasticity in the hippocampus is thought to underlie learning and memory since memory deficits in rats and senile dementia in humans correlate with diminished spine density in the CA1 area de Ruiter and Uylings 1987, Avila-Costa et al 1999. Finally, Golgi-stained sections of human AD hippocampus and cortex reveal a significant decrease in dendritic spine density when compared to age-matched controls Probst et al 1983, Ferrer et al 1990, Baloyannis et al 1992, Einstein et al 1994.

Based on these data, the present study analyzed age-dependent changes in dendritic spine density in the CA1 region of the hippocampus in two strains of mice described in the literature as Tg2576 and PDAPP mice. The Tg2576 and PDAPP mice express the human APP transgene with the so-called Swedish double mutation (APPSw, K670N/M671L driven by the prion protein promoter) near the β-secretase cleavage site or the V717F mutation near the γ-secretase site (driven by the PDGF-β promoter), respectively. Both strains have been well characterized with respect to the time courses of Aβ production and amyloid plaque deposition Hsiao et al 1996, Games et al 1995. The CA1 region was chosen for its known vulnerability in AD; it is generally affected earlier and more severely than other brain regions during the course of the disease Van Hoesen and Hyman 1990, Arendt et al 1998, Arendt 2001. Learning and memory impairments in Tg2576 and PDAPP mice in the absence of sensory or motor deficits also suggest that the hippocampus is functionally affected in APP overexpressing mice Dodart et al 2000, Hsiao et al 1996. In order to determine correlations between progressive amyloid deposits and spine density, four age groups for each strain were selected on the basis of the time course of plaque deposition in each strain (Janus and Westaway, 2001). The youngest ages examined (2 and 5 months) are not associated with amyloid plaque deposition in either strain, the middle-aged groups (8 months for PDAPP, 11 months for Tg2576) exhibit moderate Aβ deposition, and the oldest groups (11 months for PDAPP, 20 months for Tg2576) show abundant levels of diffuse plaques and moderate density of dense core plaques Hsiao et al 1996, Irizarry et al 1997, Kawarabayashi et al 2001, Larson et al 1999. Although methodological constraints precluded us from measuring Aβ plaque load and spine density in the same animals, routine analysis of Aβ load in satellite groups of mice confirm the above-cited time courses of plaque deposition. In addition to the PDAPP and Tg2576 mice, we assessed CA1 dendritic spine density in 8-month-old double-transgenic mice generated by crossing Tg2576 animals with mice expressing the human apoE2 transgene. Since the apoE2 allele affects Aβ metabolism and deposition (Fagan et al., 2002), we reasoned that effects of over expression of mutant APP on a human apoE2 genetic background would allow us to better understand the role of Aβ on dendritic spine density changes in the Tg2576 mice.

Consistent with the loss of the synaptic marker, synapophysin (Hsia et al., 1999), the data reported here demonstrate early loss in hippocampal dendritic spines that precedes the development of amyloid plaques in both the Tg2576 and the PDAPP mouse. In addition, the apoE2 genotype conferred protection in the Tg2576 mice with respect to spine density loss in the hippocampus. Taken together, these data indicate that the mutant APP genotype-dependent loss in hippocampal spine density likely results from effects of a soluble Aβ species.

Section snippets

Animals

Male transgenic mice heterozygous for the APPSw mutations (K670N/M671L driven by the prion protein promoter) in the human APP gene (Tg2576 line) and their transgene-negative controls were studied at approximately 2, 5, 11, and 20 months of age. Heterozygous, male PDAPP mice (expressing the V717F APP mutation driven by the PDGF-β promoter) and their transgene-negative littermates were assessed at 2, 5, 8, and 11 months of age. Each age group consisted of eight transgene-positive (Tg+) and eight

PDAPP mice

In the CA1 region of the hippocampus, PDAPP mice exhibited dendritic spine loss due to both age and APP overexpression (Fig. 2, Fig. 3, Table 1). A two-way ANOVA revealed significant effects of age (F3,13 = 125, p < 0.001) and genotype (F1,31 = 38.7, p < 0.001), as well as an interaction between age and genotype (F = 7.36, p < 0.001). Going from 2 to 5 months of age, Tg− and Tg+ PDAPP mice exhibited 18% and 10% reduction in dendritic spines, respectively, without further significant loss in

Discussion

The present study demonstrates a selective decrease in dendritic spine density in the hippocampal CA1 neurons in two distinct mouse lines, the PDAPP and Tg2576 mice, expressing mutant human APP. We observed two factors contributing to dendritic spine loss in both lines of mice: age and the overexpression of the mutant APP. To our knowledge, this is the first report of reduced spine density in CA1 hippocampal neurons in APP transgenic mice relative to controls at an early age. The evidence of

Acknowledgements

The authors are grateful to Kathleen Hoenow and Karen Chen from Elan Pharmaceuticals for providing PDAPP mice, Diane Stephenson for her help with the optical scope, and Jerrie-Lynn Morrison and Erika Walker for technical assistance with perfusion and sectioning of tissue.

References (38)

  • C. Janus et al.

    Transgenic mouse models of Alzheimer’s disease

    Physiol. Behav.

    (2001)
  • J. Larson et al.

    Alterations in synaptic transmission and long-term potentiation in hippocampal slices from young and aged PDAPP mice

    Brain Res.

    (1999)
  • M. Megias et al.

    Total number and distribution of inhibitory and excitatory synapses on hippocampal CA1 pyramidal cells

    Neuroscience

    (2001)
  • D. Moechars et al.

    Early phenotypic changes in transgenic mice that overexpress different mutants of amyloid precursor protein in brain

    J. Biol. Chem.

    (1999)
  • A. Probst et al.

    Neuritic plaques in senile dementia of Alzheimer typea Golgi analysis in the hippocampal region

    Brain Res.

    (1983)
  • A. Takeuchi et al.

    Age-related amyloid beta deposition in transgenic mice overexpressing both Alzheimer mutant presenilin 1 and amyloid beta precursor protein Swedish mutant is not associated with global neuronal loss

    Am. J. Pathol.

    (2000)
  • H.B.M. Uylings et al.

    Neuronal changes in normal human aging and Alzheimer’s disease

    Brain Cognit.

    (2002)
  • S.J. Baloyannis et al.

    The acoustic cortex in Alzheimer’s disease

    Acta Otolaryngol. Suppl.

    (1992)
  • B.H. Bowman et al.

    Discovery of a brain promoter from the human transferring gene and its utilization for development of transgenic mice that express human apolipoprotein E alleles

    Proc. Natl. Acad. Sci. USA

    (1995)
  • Cited by (205)

    • Pin1 and Alzheimer's disease

      2023, Translational Research
    View all citing articles on Scopus
    View full text