Species: Mouse
Genes: APP
Mutations: APP K670_M671delinsNL (Swedish), APP S679C
Modification: APP: Transgenic
Disease Relevance: Alzheimer's Disease, MCI due to AD
Strain Name: N/A
Genetic Background: C57BL/6N
Availability: Available through Carsten Korth.

Summary

This unique model provides scientists the opportunity to study the effects of Aβ dimers independent of amyloid plaques, Aβ monomers, or other Aβ oligomers. The TgDimer mouse carries a transgene encoding the 751-amino-acid isoform of human APP with the Swedish mutation and a serine-to-cysteine substitution at amino acid 679 (amino acid 8 within the Aβ sequence), driven by the neuron-specific Thy1 promoter (Müller-Schiffmann et al., 2016). The latter mutation links pairs of APP molecules via a disulfide bridge and results in the generation of stable Aβ dimers that are structurally similar to those formed by the wild-type peptides (Müller-Schiffmann et al., 2011). Aβ dimers, but not monomers or other oligomers, are detected in the brains of TgDimer mice.

TgDimer mice do not develop amyloid plaques, tau hyperphosphorylation, or neuroinflammation, at least through 24 months of age. However, they display deficits in synaptic plasticity, age-dependent decreases in acetylcholine in the hippocampus, learning impairment, and behavioral abnormalities. These phenotypes have been attributed to the presence of Aβ dimers, although possible effects of overexpression of mutant APP or its metabolites—including  dimerized APP or CTF-β (C99)—cannot be excluded.

TgDimer mice crossed with plaque-producing APP transgenic strains have been used to study the interaction of Aβ dimers with amyloid plaques in vivo.

The mice described here are homozygous for the transgene.

Transgene expression | Neurochemistry | Neuropathology | Long-term potentiation | Behavior | Other | Applications of the model

Transgene expression

The transgene—as well as the linearized vector—used to generate TgDimer mice is the same as that used to create the APP23 strain, except that the S679C mutation was added. Similar to APP23 mice, TgDimer mice show a sevenfold overexpression of APP relative to levels of the endogenous protein in wild-type mice. Levels of soluble Aβ40 and Aβ42, measured by ELISA in brain homogenates from mice 3 to 24 months of age, did not change over time in TgDimer mice. APP, CTF-α (C83), CTF-β (C99), dimerized APP, dimerized CTF-β, and Aβ dimers—but not Aβ monomers—were detected by western blot (Müller-Schiffmann et al., 2016).

Neurochemistry

Neurotransmitter levels have been measured in the brains of 7- and 12-month-old male mice (Abdel-Hafiz et al., 2018). Acetylcholine levels declined over time in the hippocampi of TgDimer mice but did not differ between the age groups in the neostriatum, ventral striatum, prefrontal cortex, amygdala, or entorhinal cortex. Compared with age-matched, wild-type controls, the rate of serotonin turnover was lower in TgDimer brains (hippocampus, ventral striatum, and amygdala), but did not change with aging. Dopamine levels in the hippocampus declined with age in both transgenic and wild-type mice.

Neuropathology

No amyloid plaques were detected by immunohistochemistry in the brains of TgDimer mice through 24 months of age, but intracellular Aβ immunoreactivity was seen in the hippocampus and cortex, beginning by 12 months. Consistent with the immunostaining results, no insoluble Aβ was detected biochemically through 24 months (Müller-Schiffmann et al., 2016).

Hyperphosphorylation of endogenous mouse tau was not seen by immunohistochemistry or western blot, using antibodies AT8 (recognizes tau phosphorylated at serine 202 and threonine 205), AT180 (recognizes paired helical filament tau) or an antibody directed against tau phosphorylated at serine 396 and serine 404 (Müller-Schiffmann et al., 2016).

Microgliosis and astrogliosis—assessed by Iba1- and GFAP- immunoreactivity, respectively—were not seen. Nor was neuron loss evident (Müller-Schiffmann et al., 2016).

Long-term potentiation

Long-term potentiation (LTP) at hippocampal Schaeffer collateral-CA1 synapses decayed more rapidly in brain slices from TgDimer mice than wild-type mice, with differences becoming apparent approximately 90 minutes after high-frequency stimulation (Müller-Schiffmann et al., 2016).

Behavior

TgDimer mice displayed a learning deficit in the Morris Water Maze, which appeared to worsen with age: Although their performance never matched that of wild-type controls, 7-month-old transgenic mice did show a decreased latency to find the escape platform across trial blocks. When retested at 12 months of age, control mice continued to show improved performance across trials, but TgDimer mice did not (Muller-Schiffmann et al., 2016). The genotypes did not differ in swim speed or latency to find a visible escape platform, suggesting that the impaired performance of TgDimer mice in the water maze is unlikely to be due to motor or sensory deficits.

Further testing of 7-month-old mice showed additional behavioral abnormalities (Abdel-Hafiz et al., 2018). TgDimer mice performed similarly to wild-type controls on the rotarod at constant speed but fell off sooner when the rod accelerated. These findings suggest that the transgenic animals had normal motor coordination and resistance to fatigue but had impairments in motor learning. Compared with wild-type mice, TgDimer mice showed less rearing in the radial arm maze and the open field and a preference for the proximal versus the distal open arm in the elevated plus maze, differences that might indicate deficits in non-selective attention and increased risk assessment and anxiety. In the forced swim test, used to assess depression, transgenic animals showed a shorter latency to float. Working memory in the reward-free version of the radial arm maze was comparable between transgenic mice and wild-type controls, indicating an intact reinforced free spatial working memory.

Together these findings suggest that the cognitive deficits in the TgDimer mice may be limited to reward-governed learning and attention-related processes.

Other

Blood corticosterone levels after restraint stress were similar in transgenic and wild-type mice, at 12 months of age, suggesting an intact hypothalamic-pituitary axis (Abdel-Hafiz et al., 2018).

Applications of the model

The absence of plaques in TgDimer mice implies that dimers alone are incapable of initiating (“seeding”) the formation of amyloid plaques. To investigate whether dimers can interact with plaques in seeding-competent brains, TgDimer mice were crossed with TgCRND8 mice. The latter animals carry a human APP transgene with the Swedish and Indiana mutations and develop amyloid plaques beginning at 3 months of age. The double transgenic TgCRND8;TgDimer mice are heterozygous for each transgene.

The brains of TgCRND8;TgDimer mice were found to contain fewer amyloid plaques than those of age-matched TgCRND8 animals (van Gerresheim et al., 2020). This difference was apparent by 90 days in the cortex, where the double-transgenic mice had approximately twofold fewer dense-core and diffuse plaques than did TgCRND8. By 150 days, differences between genotypes were also seen in the hippocampus, where TgCRND8;TgDimer mice had approximately three- to fourfold fewer dense-core and diffuse plaques than TgCRND8 mice. Plaque size did not differ between genotypes. Despite the difference in plaque burden, markers of microgliosis and astrogliosis—Iba1- and GFAP- immunoreactivity measured by western blot of whole-brain homogenates—were similar in TgCRND8;TgDimer and TgCRND8 mice.

Biochemical studies suggested that the mutant (Aβ-S8C) dimers were incorporated into amyloid plaques in the TgCRND8;TgDimer mice: Elevated levels of dimers, presumably formed by Aβ-S8C, were found in the formic acid fraction of the brains of these mice, compared with TgCRND8 brains (Müller-Schiffmann et al., 2016).

Phenotype Characterization

Q&A with Model Creator

Q&A with Carsten Korth and Laila Abdel-Hafiz.

What would you say are the unique advantages of this model?

It allows one to investigate the biological effects of Aβ dimers in the absence of other Aβ species. The exclusive presence of Aβ dimers in the absence of Aβ plaques and the neurochemical changes make it an animal for early AD. The limitation, of course, is that the Aβ dimer is artificial—i.e. stabilized by an artificial disulfide bridge even though this has been designed at a position where similarities to presumed Aβ dimer structures have been preserved (Müller-Schiffmann et al., 2011).

What do you think this model is best used for?

1. To investigate the isolated effects of Aβ dimers or interactions of Aβ dimers with other Aβ species or proteins when crossed with other transgenes
2. To investigate aspects of early AD

What caveats are associated with this model?

The Aβ dimer is artificial (Aβ-S8C), even though it has been designed at a position where similarities to presumed Aβ-dimer structures have been preserved (Müller-Schiffmann et al., 2011).

COMMENTS / QUESTIONS

No Available Comments

Make a comment or submit a question

To make a comment you must login or register.

References

Research Models Citations

1. APP23
2. TgCRND8

Paper Citations

1. Müller-Schiffmann A, Herring A, Abdel-Hafiz L, Chepkova AN, Schäble S, Wedel D, Horn AH, Sticht H, de Souza Silva MA, Gottmann K, Sergeeva OA, Huston JP, Keyvani K, Korth C. Amyloid-β dimers in the absence of plaque pathology impair learning and synaptic plasticity. Brain. 2016 Feb;139(Pt 2):509-25. Epub 2015 Dec 10 PubMed.
2. Müller-Schiffmann A, Andreyeva A, Horn AH, Gottmann K, Korth C, Sticht H. Molecular engineering of a secreted, highly homogeneous, and neurotoxic aβ dimer. ACS Chem Neurosci. 2011 May 18;2(5):242-8. Epub 2011 Mar 11 PubMed.
3. Abdel-Hafiz L, Müller-Schiffmann A, Korth C, Fazari B, Chao OY, Nikolaus S, Schäble S, Herring A, Keyvani K, Lamounier-Zepter V, Huston JP, de Souza Silva MA. Aβ dimers induce behavioral and neurochemical deficits of relevance to early Alzheimer's disease. Neurobiol Aging. 2018 Sep;69:1-9. Epub 2018 Apr 17 PubMed.
4. van Gerresheim EF, Herring A, Gremer L, Müller-Schiffmann A, Keyvani K, Korth C. The interaction of insoluble Amyloid-β with soluble Amyloid-β dimers decreases Amyloid-β plaque numbers. Neuropathol Appl Neurobiol. 2020 Dec 18; PubMed.

Other Citations

1. Carsten Korth

Further Reading

No Available Further Reading