Hippocampal neural circuit connectivity alterations in an Alzheimer's disease mouse model revealed by monosynaptic rabies virus tracing - PubMed (original) (raw)

Hippocampal neural circuit connectivity alterations in an Alzheimer's disease mouse model revealed by monosynaptic rabies virus tracing

Qiao Ye et al. Neurobiol Dis. 2022.

Abstract

Alzheimer's disease (AD) is a progressive neurodegenerative disorder with growing major health impacts, particularly in countries with aging populations. The examination of neural circuit mechanisms in AD mouse models is a recent focus for identifying new AD treatment strategies. We hypothesize that age-progressive changes of both long-range and local hippocampal neural circuit connectivity occur in AD. Recent advancements in viral-genetic technologies provide new opportunities for semi-quantitative mapping of cell-type-specific neural circuit connections in AD mouse models. We applied a recently developed monosynaptic rabies tracing method to hippocampal neural circuit mapping studies in AD model mice to determine how local and global circuit connectivity to hippocampal CA1 excitatory neurons may be altered in the single APP knock-in (APP-KI) AD mouse model. To determine age-related AD progression, we measured circuit connectivity in age-matched littermate control and AD model mice at two different ages (3-4 vs. 10-11 months old). We quantitatively mapped the connectivity strengths of neural circuit inputs to hippocampal CA1 excitatory neurons from brain regions including hippocampal subregions, medial septum, subiculum and entorhinal cortex, comparing different age groups and genotypes. We focused on hippocampal CA1 because of its clear relationship with learning and memory and that the hippocampal formation shows clear neuropathological changes in human AD. Our results reveal alterations in circuit connectivity of hippocampal CA1 in AD model mice. Overall, we find weaker extrinsic CA1 input connectivity strengths in AD model mice compared with control mice, including sex differences of reduced subiculum to CA1 inputs in aged female AD mice compared with aged male AD mice. Unexpectedly, we find a connectivity pattern shift with an increased proportion of inputs from the CA3 region to CA1 excitatory neurons when comparing young and old AD model mice, as well as old wild-type mice and old AD model mice. These unexpected shifts in CA3-CA1 input proportions in this AD mouse model suggest the possibility that compensatory circuit increases may occur in response to connectivity losses in other parts of the hippocampal circuits. We expect that this work provides new insights into the neural circuit mechanisms of AD pathogenesis.

Keywords: Alzheimer's disease; CA1; Monosynaptic; Neural circuit; Rabies tracing; Retrograde; hippocampus.

Copyright © 2022. Published by Elsevier Inc.

PubMed Disclaimer

Conflict of interest statement

Declaration of Competing Interest

The authors declare no competing interests.

Figures

Fig. 1.

Fig. 1.

Age-progressive Aβ pathology in APP-KI mice. A) Representative images of amyloid plaques in age-matched young and old APP-KI and WT mice. The upper four panels show half coronal brain sections in red channel autofluorescence without immunostaining. The bottom four panels depict enlarged views of 6E10 amyloid antibody immunostained hippocampal CA1 slices. White arrows point to the amyloid plaques in brain slices prepared from APP-KI young and APP-KI old mice. No plaques are detected in WT young and old mice. More amyloid deposits are detected in APP-KI old mice relative to APP-KI young mice. DAPI is labeled blue, amyloid plaque is labeled red. Scale bars are in figure panels. B–D) Quantification of amyloid plaques stained by 6E10 antibody. No amyloid plaques are found in either WT young or old mice, but in APP-KI mice there is a significant age-dependent increase in both plaque density and intensity (B, density, APP-KI young: 654.3 ± 94.9, APP-KI old: 1339 ± 81.3, Wilcoxon rank-sum test, p = 2.0 × 10−4; D, intensity, APP-KI young: 1057 ± 27.4, APP-KI old: 1254 ± 18.9, linear mixed effects model, p = 0.045573). C, no significance is detected in plaque size measurement. Two brain sections from each of 5 brains from each group of mice were used for quantification. AU: arbitrary unit. * p < 0.05, ** p < 0.01, *** p < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2.

Fig. 2.

Cre-dependent monosynaptic rabies tracing reveals overall reduced circuit connectivity in APP-KI mice. A) Schematic of cell-type-specific retrograde monosynaptic rabies tracing. To specifically label excitatory CA1 neurons, AAV helper virus (AAV-DIO-TC66T-GFP-oG and AAV-CaMKII-EGFP-Cre), labeled green, was injected into the dorsal hippocampal CA1 pyramidal layer, followed 17 days later at the same site by an injection of rabies virus (EnvA-SADG-DsRed), labeled red. The neurons labeled both green and red represent the starter neurons. The neurons labeled only red represent the presynaptic inputs to the starter neurons. B) Major input regions to CA1 pyramidal layer excitatory neurons, including MS-DB, CA1, CA2, CA3, SUB, and EC. Red colour labels represent presynaptic input neurons. C) Overall connectivity revealed by monosynaptic rabies tracing. Left to right, wild type and APP-KI mice at young and old age. The connectivity strength is determined by the overall number of labeled neurons in the whole brain divided by the total number of starter neurons. APP-KI young mice (n = 10) have significantly weaker connectivity strength compared to WT young mice (n = 7) (WT young: 22.10 ± 1.97, APP-KI young: 14.41 ± 0.72, Wilcoxon rank-sum test, p = 2.0 × 10−3). APP-KI old mice (n = 8) have significantly weaker connectivity strength relative to WT old mice (n = 10) (WT old: 21.01 ± 2.14, APP-KI old: 14.36 ± 1.92, Wilcoxon rank-sum test, p = 3.40 × 10−2). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3.

Fig. 3.

Monosynaptic rabies tracing maps specific regional inputs to excitatory CA1 cells along the anterior-posterior (AP) axis in young WT and APP-KI mice. A, C) Representative fluorescent coronal section images from WT young (n = 7) and APP-KI young (n = 10), respectively. Rabies virus-infected neurons are labeled by DsRed, and AAV-infected neurons are in green. All slices are counterstained by DAPI in blue. For both A and C, (1) shows the ipsilateral (ipsi) hippocampal formation including the CA1 injection site. (2), rabies virus mapped presynaptic inputs in the contralateral (contra) hippocampus. (3), enlarged image of CA1 starter neurons expressing both EGFP and DsRed fluorescent proteins from AAV and rabies virus. For both A and C (4–10), results of rabies virus-mediated retrograde monosynaptic tracing from CA1. The input regions include hippocampal CA2 (4), CA3 (5), medial septum and diagonal band (MS-DB) (6), subiculum (SUB) (7 and 8), and entorhinal cortex (EC) (9 and 10). (3), (4), (5), (8), and (10) are enlarged views of the white boxed areas shown in (1), (7) and (9), respectively. Scale bars are labeled for each panel. B, D) The connectivity strength index (CSI) distribution along AP positions across the whole brain. The CSI is defined as the number of input neurons normalized by the number of starter neurons; the AP position is given relative to bregma values. The red arrow at AP = −1.94 mm shows the position of the injection site. Representative input regions are used for the AP plot, including the ipsilateral hippocampal CA1 oriens layer (Or), ipsilateral CA2 and ipsilateral CA3 pyramidal layers (Py), as well as MS-DB, SUB, and EC. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4.

Fig. 4.

Monosynaptic rabies tracing maps specific regional inputs to excitatory CA1 cells along the anterior-posterior (AP) axis in old WT and APP-KI mice. A, C) Representative fluorescent coronal section images from WT old (n = 10) and APP-KI old (n = 8), respectively. Rabies virus-infected neurons are labeled by DsRed, and AAV-infected neurons are in green. All slices are counterstained by DAPI in blue. For both A and C, (1) shows the ipsilateral hippocampal formation including the CA1 injection site. (2), rabies virus mapped presynaptic inputs in the contralateral hippocampus. (3), enlarged image of CA1 starter neurons expressing both EGFP and DsRed fluorescent proteins from AAV and rabies virus. For both A and C (4–10), results of rabies virus-mediated retrograde monosynaptic tracing from CA1. The input regions include hippocampal CA2 (4), CA3 (5), medial septum and diagonal band (MS-DB) (6), subiculum (SUB) (7 and 8), and entorhinal cortex (EC) (9 and 10). (3), (4), (5), (8), and (10) are enlarged views of the white boxed areas shown in (1), (7), and (9), respectively. Scale bars are labeled for each panel. B, D) The connectivity strength index (CSI) distribution along AP positions across the whole brain. CSI is defined as the number of input neurons normalized by the number of starter neurons; AP position is given relative to bregma values. The red arrow at AP = −1.94 mm shows the position of the injection site. Representative input regions were used for the AP plot, including the ipsilateral hippocampal CA1 oriens layer (Or), ipsilateral CA2 and ipsilateral CA3 pyramidal layers (Py), as well as MS-DB, SUB, and EC. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5.

Fig. 5.

Connectivity strength index (CSI) quantification of presynaptic input regions. Quantitative analysis of CA1 CSI values across WT young and old mice and APP-KI young and old mice. The CSI is the ratio of the CA1 input neuron number in a subregion to the total starter neuron number in a brain. The input regions include hippocampal CA1, CA2, and CA3 oriens layer (or) and pyramidal layer (Py), from both the ipsilateral and contralateral sides of the CA1 injection site, as well as the MS-DB, SUB, and EC. A) WT young mice (n = 7) have significantly higher CSI values relative to APP-KI young mice (n = 10) in the CA1_Or_ipsi, CA2_Py_ipsi, CA2_Or_ipsi, and MS-DB regions (Wilcoxon rank-sum test, p = 4.11 × 10−4, p = 4.11 × 10−4, p = 1.36 × 10−2, p = 1.03 × 10−4, respectively). B) WT old mice (n = 10) have significantly higher CSI values relative to APP-KI old mice (n = 8) in the CA1_Py_contra, CA2_Py_contra, and MS-DB regions (Wilcoxon rank-sum test, p = 2.64 × 10−2, p = 1.17 × 10−2, p = 1.55 × 10−2, respectively). C) WT young mice have significantly higher CSI values relative to WT old mice in the CA1_Py_contra region (Wilcoxon rank-sum test, p = 2.34 × 10−2). D) APP-KI old mice have significantly higher CSI values relative to APP-KI young mice in the CA1_Or_ipsi region (Wilcoxon rank-sum test, p = 1.17 × 10−2). APP-KI young mice have significantly higher CSI values relative to APP-KI old mice in the CA1_Py_contra, CA2_Py_contra, and CA3_Py_contra regions (Wilcoxon rank-sum test, p = 4.6 × 10−5, p = 1.55 × 10−2, p = 4.34 × 10−2, respectively). * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. All data are represented with mean ± SEM. See also Supplementary Tables S1 and S2.

Fig. 6.

Fig. 6.

Proportion of inputs (PI) quantification of presynaptic input regions. Quantitative analysis of PI values across four groups of mice. The PI is the ratio of the input neuron number in a subregion to the total input neuron number in a brain. The input regions include hippocampal CA1, CA2, and CA3 oriens layer (Or) and pyramidal layer (Py), from both the ipsilateral and contralateral sides of the injection site, as well as the MS-DB, SUB, and EC. A) APP-KI young mice (n = 10) have significantly lower PI values relative to WT young mice (n = 7) in the CA1_Or_ipsi, CA2_Py_ipsi, CA2_Or_ipsi and MS-DB regions (Wilcoxon rank-sum test, p = 4.63 × 10−3, p = 4.11 × 10−4, p = 1.85 × 10−2, p = 2.50 × 10−2, respectively). APP-KI young mice have significantly higher PI values relative to WT young mice in the CA3_Py_contra region (Wilcoxon rank-sum test, p = 1.95 × 10−3). B) APP-KI old mice (n = 8) have significantly lower PI values relative to WT old mice (n = 8) in the CA1_Py_contra, CA2_Py_contra, MS-DB and SUB regions (Wilcoxon rank-sum test, p = 3.39 × 10−2, p = 2.66 × 10−2, p = 1.17 × 10−2, p = 1.55 × 10−2, respectively). APP-KI old mice have significantly higher PI values relative to WT old mice in the CA3_Py_ipsi region (Wilcoxon rank-sum test, p = 1.37 × 10−3). C) WT young mice have significantly higher PI values relative to WT old mice in the CA1_Py_contra region (Wilcoxon rank-sum test, p = 1.36 × 10−2). D) APP-KI old mice have significantly higher PI values relative to APP-KI young mice in the CA1_Or_ipsi, CA2_Py_ipsi regions, and CA3_Py_ipsi (Wilcoxon rank-sum test, p = 2.06 × 10−3, p = 3.06 × 10−3, p = 3.2 × 10−4, respectively). APP-KI young mice have significantly higher PI values relative to APP-KI old mice in the CA1_Py_contra, CA2_Py_contra, CA3_Py_contra, SUB and MnR/PMnR regions (Wilcoxon rank-sum test, p = 4.6 × 10−5, p = 6.22 × 10−3, p = 3.06 × 10−3, p = 3.06 × 10−3, p = 8.48 × 10−3, respectively). All data are represented with mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. See also Supplementary Tables S1 and S2.

References

    1. Andrade-Moraes CH, Oliveira-Pinto AV, Castro-Fonseca E, da Silva CG, Guimaraes DM, Szczupak D, et al. , 2013. Cell number changes in Alzheimer’s disease relate to dementia, not to plaques and tangles. Brain 136 (Pt 12), 3738–3752. 10.1093/brain/awt273. -DOI -PMC -PubMed
    1. Busche MA, Eichhoff G, Adelsberger H, Abramowski D, Wiederhold KH, Haass C, et al. , 2008. Clusters of hyperactive neurons near amyloid plaques in a mouse model of Alzheimer’s disease. Science 321 (5896), 1686–1689. 10.1126/science.1162844. -DOI -PubMed
    1. Busche MA, Chen X, Henning HA, Reichwald J, Staufenbiel M, Sakmann B, Konnerth A, 2012. Critical role of soluble amyloid-beta for early hippocampal hyperactivity in a mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. U. S. A. 109 (22), 8740–8745. 10.1073/pnas.1206171109. -DOI -PMC -PubMed
    1. Busche MA, Kekus M, Adelsberger H, Noda T, Forstl H, Nelken I, Konnerth A, 2015. Rescue of long-range circuit dysfunction in Alzheimer’s disease models. Nat. Neurosci. 18 (11), 1623–1630. 10.1038/nn.4137. -DOI -PubMed
    1. Calvo-Rodriguez M, Bacskai BJ, 2021. Mitochondria and calcium in Alzheimer’s disease: from cell signaling to neuronal cell death. Trends Neurosci. 44 (2), 136–151. 10.1016/j.tins.2020.10.004. -DOI -PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources