Species: Mouse
Genes: App
Modification: App: Knock-In
Disease Relevance: Alzheimer's Disease
Strain Name: B6(SJL)-App^tm1.1Aduci/J
Genetic Background: mixed C57BL/6J and C57BL/6NJ
Availability: Available from The Jackson Laboratory (JAX Stock No. 030898).

Summary

This knock-in model, designated “hAβ-KI,” carries a humanized Aβ sequence within the murine App gene. Like the App knock-in (humanized Aβ) model, hAβ-KI mice do not carry any Alzheimer’s disease-linked mutations in APP. hAβ-KI mice exhibit age-dependent impairments in synaptic plasticity, behavioral abnormalities, and changes in gene expression that overlap those seen in sporadic AD, but they do not develop amyloid plaques. In this model, the Aβ-encoding exon of App is flanked by loxP sites, allowing Cre-mediated conditional ablation of APP.

The description below refers to hAβ-KI mice that are homozygous for the humanized Aβ sequence.

Expression of App is controlled by the cognate locus, and hAβ-KI mice produce APP at mouse physiological levels.

Age-dependent increases in levels of insoluble Aβ40 and Aβ42 were seen in the brains of hAβ-KI mice, accompanied by decreases in the levels of the soluble peptides. The Aβ42/Aβ40 ratio, measured in the hippocampus, did not change with age in the knock-in mice. Seeding-competent Aβ aggregates were detected in hippocampal extracts from 18- to 21-month hAβ-KI mice, using the protein misfolding cyclic amplification assay.

Elevated levels of CTF-β have been found in the brains of other rodent models with humanized Aβ sequences —a rat model and the App knock-in (humanized Aβ) mouse mentioned above—compared with wild-type conspecifics. It is not yet known whether CTF-β is similarly elevated in hAβ-KI mice.

Neuropathology | Gliosis/Inflammation | Neuron loss | Synapse loss | LTP | Behavior | Transcriptomics | Conditional ablation of App | Modification Details |Related Strains

Neuropathology

Amyloid plaques were not found in the brains of hAβ-KI mice, observed through 22 months of age.

No plaques were seen in the brains of 18-month mice homozygous for the humanized Aβ sequence in App and for the M146V mutation in Psen1 (these mice were generated through crossing hAβ-KI mice with PSEN1(M146V) Knock-In mice, then interbreeding the offspring). This finding is consistent with observations of another double knock-in mouse with a humanized Aβ sequence and an AD-linked Psen1 mutation (App knock-in (humanized Aβ) (Leuven); Psen1 knock-in (M139T)), which also failed to generate plaques.

Mouse brains contained granules positive for the Periodic Acid Schiff (PAS) stain for polysaccharides. PAS-positive granules were found in both hAβ-KI mice and wild-type mice but appeared earlier and were more abundant in the former. The PAS-positive granules in mice are believed to be equivalent to the corpora amylacea of the human brain, which accumulate with age and in neurodegenerative diseases and may function in transferring waste products from the brain parenchyma to the CSF (Augé et al., 2017). Granules also stained using OC antibodies, which are believed to detect protofibrils of amyloid-forming proteins, in a conformation-dependent, sequence-independent manner (Kayed et al., 2007). Surprisingly, however, the OC-positive profiles in hAβ-KI mice do not appear to contain Aβ (Kim Green, personal communication), and the apparent staining with OC may reflect the propensity of these structures to bind IgMs that contaminate many antibody preparations (Augé et al., 2017). Granule formation was accelerated in mice that carried the Psen1 M146V mutation in addition to the humanized Aβ sequence, compared with hAβ-KI animals.

Gliosis/Inflammation

Neither microgliosis nor astrogliosis was observed through 22 months of age. However, astrocyte processes may associate with PAS-positive granules.

Levels of the pro-inflammatory cytokines IL-1β and TNFα increased with age in hAβ -KI mice, while levels of the anti-inflammatory cytokines IL-2, IL- 4, and IL-10 decreased.

Neuron loss

Neuron loss has not yet been documented in hAβ-KI mice. Neuron numbers in hippocampal CA1 were similar in aged (22-month-old) hAβ-KI and wild-type mice. However, the volume of the hippocampus was decreased in the knock-in mice—possibly due to loss or shrinkage of neurons in other hippocampal subfields or shrinkage of the neuropil.

Synapse loss

There is immunohistochemical evidence of synapse loss in hAβ-KI mice: Fewer synaptophysin-immunoreactive puncta were found in the hippocampi of 18-month hAβ-KI mice, compared with wild-type mice; however, the number of puncta immunoreactive for the postsynaptic marker PSD95 did not differ between the genotypes.

LTP

Long-term potentiation at Schaffer collateral-CA1 synapses, induced by theta-burst stimulation, was normal in hippocampal slices obtained from 2-month-old hAβ-KI mice, but was impaired in slices from 18-month animals.

Behavior

hAβ-KI mice did not perform as well as wild-type mice in two memory tests: the genotypes differed in the contextual fear conditioning test by 10 months of age and in the novel object recognition task by 14 months.

Transcriptomics

Bulk RNA-Seq identified five genes that were differentially expressed in 2-month-old hAβ-KI mice, compared with wild-type mice. This number increased to 14 differentially expressed genes (DEG) in 22-month mice. The DEG in the older animals included genes involved in metabolism (Dhcr7, Sdhd, Hspe1, and Wdfy1), neuroplasticity and neurotransmission (Atp2b1, Gabra2, Lppr4, Tppp3, Diras2, and Nsmf), and transcriptional and splicing-related factors (Chd4, Psip1, and R3hdm4).

Weighted gene co-expression network analysis (WGCNA) identified three modules downregulated in hAβ-KI at 22 months, but not 2 months. Functional annotation showed enrichment in gene-ontology (GO) terms related to glutamate-ion channel binding and signaling and histone acetylation, neurotransmitter transport and metabolic processes, and mitochondrial energetics. Using additional bioinformatics tools, these modules were found to be downregulated in temporal and frontal cortices of human AD subjects.

Conditional ablation of App

The floxed App Aβ-encoding exon in hAβ-KI mice permits conditional deletion of the modified App allele. Cre-mediated excision of exon 14 led to reductions in App mRNA and APP protein (deletion of this exon causes a frameshift that may result in mRNA instability).

Cre-mediated ablation of App in adult hAβ-KI mice led to recovery of cognitive function assessed using the object location memory test, accompanied by reduction in levels of insoluble, but not soluble, levels of Aβ40 and Aβ42. In a second set of experiments, ablation of App reduced the number of OC-immunoreactive puncta in hAβ-KI mice.

Modification Details

Homologous recombination in mouse embryonic stem cells was used to introduce the following mutations into the endogenous App gene: G676R (G5R), F681Y (F10Y), R684H (R13H), numbered according to the 770 amino-acid isoform of human APP (position within the Aβ sequence); loxP sites flanking exon 14.

Related Strains

Mice with the hAβ-KI allele are also available on a congenic C57BL/6J (Jax Stock No. 031050) or a congenic C57BL/6NJ (Jax Stock No. 032013) background.

Phenotype Characterization

COMMENTS / QUESTIONS

1. • Mathias Jucker
     Hertie Institute for Clinical Brain Research, University of Tübingen, and DZNE Tübingen
   • Posted: 04 May 2021
   • Paper: Generation of a humanized Aβ expressing mouse demonstrating aspects of Alzheimer's disease-like pathology.

   Happy birthday, astroglial inclusions!

   Thirty years ago, the first APP transgenic mouse model was published claiming that it develops Aβ deposits in the brain (Wirak et al., 1991). Then it turned out that these structures were not Aβ deposits, but rather Periodic acid Schiff-positive (PAS) astrocytic inclusions that occur with aging in some, but not all, aged mouse brains, and which are heavily dependent on the genetic background of the mice. The PAS inclusions are rich in polyglucosan, and they tend to bind nonspecifically to many different antibodies. The paper was corrected, the authors stating that their data no longer support the claim that the Aβ transgene causes the formation of these inclusions in the brain (see comment after Jucker et al., 1992). 

   Now, 30 years later, this study by Baglietto-Vargas and colleagues finds that the expression of human-sequence Aβ in knock-in mice promotes the appearance of these still-enigmatic lesions. The authors do not cite the 30-year-old paper, and thus I am not sure whether they are aware of the previous work.

   I remain skeptical—as I was 30 years ago—that there is a meaningful link between Aβ and these interesting, age-related, glial inclusions. Nevertheless, hopefully the current publication will stimulate the field to work again on the nature and functional consequences of these glial lesions in the aging brain (see also Jucker et al., 1994; Jucker et al., 1994; Mitsuno et al., 1999). 

   References:

   Wirak DO, Bayney R, Ramabhadran TV, Fracasso RP, Hart JT, Hauer PE, Hsiau P, Pekar SK, Scangos GA, Trapp BD. Deposits of amyloid beta protein in the central nervous system of transgenic mice. Science. 1991 Jul 19;253(5017):323-5. PubMed.

   Jucker M, Walker LC, Martin LJ, Kitt CA, Kleinman HK, Ingram DK, Price DL. Age-associated inclusions in normal and transgenic mouse brain. Science. 1992 Mar 13;255(5050):1443-5. PubMed.

   Jucker M, Walker LC, Schwarb P, Hengemihle J, Kuo H, Snow AD, Bamert F, Ingram DK. Age-related deposition of glia-associated fibrillar material in brains of C57BL/6 mice. Neuroscience. 1994 Jun;60(4):875-89. PubMed.

   Jucker M, Walker LC, Kuo H, Tian M, Ingram DK. Age-related fibrillar deposits in brains of C57BL/6 mice. A review of localization, staining characteristics, and strain specificity. Mol Neurobiol. 1994 Aug-Dec;9(1-3):125-33. PubMed.

   Mitsuno S, Takahashi M, Gondo T, Hoshii Y, Hanai N, Ishihara T, Yamada M. Immunohistochemical, conventional and immunoelectron microscopical characteristics of periodic acid-Schiff-positive granules in the mouse brain. Acta Neuropathol. 1999 Jul;98(1):31-8. PubMed.

   View all comments by Mathias Jucker
2. • Sigrid Veasey
     University of Pennsylvania
   • Posted: 06 Aug 2024

   I share Dr. Jucker's skepticism for what is reported as being detected by the Aβ (OC+) antibody used.  We have found several different antibodies to label similar clusters adjacent to the foot processes of GFAP-labeled astrocytes in older mice. We have immunolabeled hippocampal mouse brain sections with reported OC+ antibody and GFAP in amyloid precursor protein knock out (APP KO) mice, as well as age-matched wild-type mice, and we find OC+ bodies (clusters) at astrocyte extensions equally in APPKO and wild-type mice. 

   Thus, the labeling does not represent Aβ or any other APP product but, rather, represents corpora amylacea, which are polyglucosan bodies found in older brains. IgM antibodies can be present in various polyclonal antibodies, and IgM will (non-specifically) attach to the polyglucosan or other components of the corpora amylacea to produce false-positive labeling. 

   Because corpora amylacea are age-dependent, they may be responsive to cellular stress and thus could be downregulated in the humanized APP knock-in mice after cre-lox reduction in APP, if brain health is improved by lessening APP or its products. This would be quite interesting to assess. Auge et al., 2018, describe this predilection of corpora amylacea for IgM and the composition of these in humans.

   References:

   Augé E, Duran J, Guinovart JJ, Pelegrí C, Vilaplana J. Exploring the elusive composition of corpora amylacea of human brain. Sci Rep. 2018 Sep 10;8(1):13525. PubMed.

   View all comments by Sigrid Veasey

Make a comment or submit a question

To make a comment you must login or register.

References

Research Models Citations

1. App knock-in (humanized Aβ)
2. App knock-in (humanized Aβ) (Leuven)
3. PSEN1(M146V) Knock-In
4. App knock-in (humanized Aβ) (Leuven); Psen1 knock-in (M139T)

Paper Citations

1. Augé E, Cabezón I, Pelegrí C, Vilaplana J. New perspectives on corpora amylacea in the human brain. Sci Rep. 2017 Feb 3;7:41807. PubMed.
2. Kayed R, Head E, Sarsoza F, Saing T, Cotman CW, Necula M, Margol L, Wu J, Breydo L, Thompson JL, Rasool S, Gurlo T, Butler P, Glabe CG. Fibril specific, conformation dependent antibodies recognize a generic epitope common to amyloid fibrils and fibrillar oligomers that is absent in prefibrillar oligomers. Mol Neurodegener. 2007;2:18. PubMed.

External Citations

1. Jax Stock No. 031050
2. Jax Stock No. 032013
3. JAX Stock No. 030898

Further Reading

No Available Further Reading