CoChat

Abnormal accumulation of extracellular vesicles in hippocampal dystrophic axons and regulation by the primary cilia in Alzheimer's disease

Open access

The DOI matches its Crossref record; 87 of the 87 references it cites were found in a registry or an index; 55 of the 81 sources its claims cite were checked for support.

Abstract

Dystrophic neurites (DNs) are abnormal axons and dendrites that are swollen or deformed in various neuropathological conditions. In Alzheimer's disease (AD), DNs play a crucial role in impairing neuronal communication and function, and they may also contribute to the accumulation and spread of amyloid beta (Aβ) in the brain of AD patients. However, it is still a challenge to understand the DNs of specific neurons that are vulnerable to Aβ in the pathogenesis of AD. To shed light on the development of radiating DNs, we examined enriched dystrophic hippocampal axons in a mouse model of AD using a three-dimensional rendering of projecting neurons. We employed the anterograde spread of adeno-associated virus (AAV)1 and conducted proteomic analysis of synaptic compartments obtained from hippocampo-septal regions. Our findings revealed that DNs were formed due to synaptic loss at the axon terminals caused by the accumulation of extracellular vesicle (EV). Abnormal EV-mediated transport and exocytosis were identified in association with primary cilia, indicating their involvement in the accumulation of EVs at presynaptic terminals. To further address the regulation of DNs by primary cilia, we conducted knockdown of the Ift88 gene in hippocampal neurons, which impaired EV-mediated secretion of Aβ and promoted accumulation of axonal spheroids. Using single-cell RNA sequencing, we identified the septal projecting hippocampal somatostatin neurons (SOM) as selectively vulnerable to Aβ with primary cilia dysfunction and vesicle accumulation. Our study suggests that DNs in AD are initiated by the ectopic accumulation of EVs at the neuronal axon terminals, which is affected by neuronal primary cilia.

Overview

CoChat Summary

## TL;DR

This mouse and cell study argues that Alzheimer’s-related Aβ burden drives dystrophic neurite/axon spheroids by trapping extracellular vesicles at presynaptic terminals, and that primary cilia dysfunction worsens this process. Single-cell RNA-seq further identified septal-projecting hippocampal somatostatin (SOM) neurons as selectively vulnerable.

## Key findings

  • In 6-month-old 5xFAD mice, whole-brain 3D imaging showed neurite spheroids in the septum and hippocampal regions, with the septum appearing “dense and clustered” compared with WT mice.
  • Synaptosome proteomics found 319 upregulated and 112 downregulated proteins in 5xFAD versus WT; GO analysis highlighted altered synaptic vesicle transport/endocytosis/exocytosis pathways.
  • EV-related markers accumulated at axon terminals near Aβ plaques: in 5xFAD hippocampal synaptosomes, western blots showed significant increases in CD81 and CD9 in whole-cell lysates and CD9 and CD63 in synaptosomes; CD63-EGFP-labeled EVs aggregated in the septum of 5xFAD mice.
  • Single-cell RNA-seq of 22,233 cells from hippocampi identified 5,468 neurons for subclustering; among neuronal types, Sst/SOM, Chgb, and Calb2-expressing neurons were reduced in 5xFAD, with differential expression showing 620 up / 234 down genes in SOM, 738 up / 186 down in Chgb neurons, and 816 up / 185 down in Calb2 neurons.
  • Disrupting primary cilia via Ift88 knockdown increased axonal dystrophy and altered vesicle handling: in cultured neurons, siIft88 reduced Ift88, increased CD81-containing EV release, and in Aβ-treated neurons proteomics found 127 downregulated and 527 upregulated proteins, with vesicle trafficking/endocytosis/exocytosis pathways altered.

## Methods

  • Mouse AD models and tracing: 5xFAD, Thy1-YFP, Ai6, and Ift88-flox/flox mice were used; AAV tracing and Cre/LoxP strategies were used to label projecting neurons and manipulate Ift88 in vivo and in microfluidic neuron cultures.
  • Multi-modal analyses: active tissue clearing and 3D light-sheet imaging, immunohistochemistry, western blotting, co-IP, LC-MS/MS proteomics, and single-cell RNA sequencing were applied to hippocampal/septal tissues and cultured neurons.
  • Single-cell analysis: 8 inDrops libraries were sequenced; after QC, 22,233 cells remained, and 5,468 neurons were reanalyzed to identify 9 neuronal subclusters and compare WT versus 5xFAD gene expression.

## Limitations

  • The paper states that it did not further elucidate how endo-lysosomal deficits affect EV accumulation, and it did not clarify the mechanisms by which primary cilia are involved.
  • It also says future studies are needed to fully characterize the vesicular landscapes in dystrophic synapses in AD.
  • Evident from the text: many experiments are in mouse models and cultured neurons, so direct human disease validation is not shown in the provided text.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1186/s40478-023-01637-3

Cites

87 of 87 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Axonal spheroids are formed in the hippocampo-septal tracts of 5xFAD; Thy1-YFP transgenic mice. A Whole-brain imaging revealed numerous axonal spheroids in the hippocampo-septal tracts of 5xFAD; Thy1-YFP transgenic mice. Scale bars = 3 mm for the dorsal view, 300 μm for the lateral view. B Sagittal sections and higher magnification images of the septum revealed that the transgenic mice had insufficient axonal fasciculation and a distinct morphology with a series of bulbous YFP-positive projections with enlarged ends. Scale bars = 400 μm for the left, and 200 μm for the right. C Schematic drawing for the active clarity method in conjunction with immunostaining and representative coronal sections showing aggregation of YFP-positive axons in the septum of 5xFAD; Thy1-YFP mice. MBP (red) was used to detect the axons in the sections. Scale bars = 200 μm. All schematic drawings were created with Biorender.com. D The presence and location of axonal spheroids in relation to amyloid plaques in the transgenic mice were studied using the ACT combined with immunostaining for Aβ (Yellow arrow: the septum, scale bar: 3 mm). Higher magnification images show that the axonal spheroids contained vesicular features (bright YFP signals in the spheroid) and were found at the end of MBP-stained axons around Aβ plaques. Whole brain imaging of Thy1-YFP WT mouse brain and higher magnification Thy1-YFP expression in the septum did not reveal any spheroids. Scale bars = 2 mm (whole brain), 10 μm (middle), 100 μm (rightmost). E A schematic drawing depicts the injection of AAV into the somatosensory cortex. A low magnification image under the drawing shows a tiled coronal image of AAV expression. The numbered magnified images are presented on the right panels with dendrite (1), callosal axons crossing the corpus callosum (2), corticothalamic axonal ends (3), and callosal axons reaching the contralateral cortex (4). AAV-hSyn-EGFP virus was injected into the somatosensory cortex of 6-month-old (1–4) and 8-month-old (1*–4*) 5xFAD mouse brains to determine the selectivity of neuronal compartments for spheroid development in the presence of Aβ plaques. The spheroid-like or bead-like big bulbous formations were not found in the somatic or dendritic compartments (1, 1*) or in a callosal bundle of nerve fibers in the corpus callosum (2, 2*). However, the spheroids near Aβ plaques were typically observed at the axon terminals of the ipsilateral corticothalamic route (3, 3*) and the callosal axons reaching the contralateral cortex (4, 4*; asterisk (*). AAV-hSyn-EGFP virus was injected into the somatosensory cortex of 8-month-old C57BL/6 mice (5) and the typical axon terminals of the ipsilateral corticothalamic route (5*) was presented (n = 4). Scale bars = 100 μm except for the leftmost image, scale bar = 500 μm and 4* = 20 μm
Fig. 1 Axonal spheroids are formed in the hippocampo-septal tracts of 5xFAD; Thy1-YFP transgenic mice. A Whole-brain imaging revealed numerous axonal spheroids in the hippocampo-septal tracts of 5xFAD; Thy1-YFP transgenic mice. Scale bars = 3 mm for the dorsal view, 300 μm for the lateral view. B Sagittal sections and higher magnification images of the septum revealed that the transgenic mice had insufficient axonal fasciculation and a distinct morphology with a series of bulbous YFP-positive projections with enlarged ends. Scale bars = 400 μm for the left, and 200 μm for the right. C Schematic drawing for the active clarity method in conjunction with immunostaining and representative coronal sections showing aggregation of YFP-positive axons in the septum of 5xFAD; Thy1-YFP mice. MBP (red) was used to detect the axons in the sections. Scale bars = 200 μm. All schematic drawings were created with Biorender.com. D The presence and location of axonal spheroids in relation to amyloid plaques in the transgenic mice were studied using the ACT combined with immunostaining for Aβ (Yellow arrow: the septum, scale bar: 3 mm). Higher magnification images show that the axonal spheroids contained vesicular features (bright YFP signals in the spheroid) and were found at the end of MBP-stained axons around Aβ plaques. Whole brain imaging of Thy1-YFP WT mouse brain and higher magnification Thy1-YFP expression in the septum did not reveal any spheroids. Scale bars = 2 mm (whole brain), 10 μm (middle), 100 μm (rightmost). E A schematic drawing depicts the injection of AAV into the somatosensory cortex. A low magnification image under the drawing shows a tiled coronal image of AAV expression. The numbered magnified images are presented on the right panels with dendrite (1), callosal axons crossing the corpus callosum (2), corticothalamic axonal ends (3), and callosal axons reaching the contralateral cortex (4). AAV-hSyn-EGFP virus was injected into the somatosensory cortex of 6-month-old (1–4) and 8-month-old (1*–4*) 5xFAD mouse brains to determine the selectivity of neuronal compartments for spheroid development in the presence of Aβ plaques. The spheroid-like or bead-like big bulbous formations were not found in the somatic or dendritic compartments (1, 1*) or in a callosal bundle of nerve fibers in the corpus callosum (2, 2*). However, the spheroids near Aβ plaques were typically observed at the axon terminals of the ipsilateral corticothalamic route (3, 3*) and the callosal axons reaching the contralateral cortex (4, 4*; asterisk (*). AAV-hSyn-EGFP virus was injected into the somatosensory cortex of 8-month-old C57BL/6 mice (5) and the typical axon terminals of the ipsilateral corticothalamic route (5*) was presented (n = 4). Scale bars = 100 μm except for the leftmost image, scale bar = 500 μm and 4* = 20 μmCC BY · PubMed Central
Fig. 2 The formation of axonal spheroids in close proximity to axon terminals contributes to synaptic dysfunction due to anomalies in synaptic proteins involved in synaptic vesicle exocytosis. A To study the relationship between synaptic loss and axonal spheroid development, AAV1-hSyn-Cre was injected into the vDG of 6-month-old Ai6 control and 5xFAD; Ai6 transgenic mice. A schematic drawing depicts the Cre-Ai6 system used to record anterograde transsynaptic delivery events. B Whole-brain imaging of Ai6 and 5xFAD; Ai6 mice exhibited damaged synapses and reduced viral spreading across vDG connections in 5xFAD; Ai6 mice (bottom). Ai6 mice showed brain-wide spreading of Cre-mediated Ai6 expression (top). White arrow indicates Ai6 expression in the cortex, marked by transsynaptic transfer of AAV1-Cre. Scale bars = 2 mm (left), 1 mm (right). C Confocal images of coronal sections including the septum, hippocampus, and dentate gyrus of Ai6 and 5xFAD; Ai6 mouse brain immunostained for Aβ and Ctip2. In the septum, Ai6 mice showed transsynaptic labeling of a few neurons (yellow arrows) in contrast to the aggregation of Ai6 signals in the neuritic spheroids adjacent to Aβ plaques. In the cortex, Ai6 mice showed transsynaptic labeling of cortical neurons (yellow arrows), however, 5xFAD; Ai6 mice showed Ai6-positive neurites but did not show Ai6 labeled cortical neurons. In the dentate gyrus, Ai6-positive dentate granule neurons and hilus neurons were widely distributed, and a few Ai6-positive hilus neurons were visible in 5xFAD; Ai6 mice. Aggregation of Ai6 in the molecular layers depicts dystrophic axons near Aβ plaques. Scale bars = 100 μm. D Synaptosomes in the hippocampo-septal tracts of WT and 5xFAD mice were analyzed by LC–MS/MS to investigate the biological processes underlying axonal spheroid formation and synaptic dysfunction. E A total of 1,099 proteins, with 743 overlapping, 92 WT control-unique, and 264 5xFAD-unique proteins, were identified in the proteomic analysis of synaptosomes (Venn diagram). A scatter plot shows 337 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 109 down-regulated proteins (log2 fold-change < -0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between 5xFAD (12-month-old, n = 4, 4 male) and WT (12-month-old, n = 4, 4 male) control synaptosomes and the weight value of this quantification. The position of the representative proteins selected for gene-set enrichment analysis is indicated in colors (blue = downregulated, red = upregulated). F Gene-set enrichment analysis based on GO Biological Process showed that both the downregulated and the upregulated proteins were linked to vesicle-related pathways, synaptic signaling pathways including chemical synaptic transmission and signal release from the synapse, and the neuron projection morphogenesis pathway. G PPI network analysis using the STRING database identified a network with 30 nodes and 50 edges from proteins in 10 upregulated pathways and a network with 20 nodes and 32 edges from proteins in 10 downregulated pathways, as shown in Fig. 2F
Fig. 2 The formation of axonal spheroids in close proximity to axon terminals contributes to synaptic dysfunction due to anomalies in synaptic proteins involved in synaptic vesicle exocytosis. A To study the relationship between synaptic loss and axonal spheroid development, AAV1-hSyn-Cre was injected into the vDG of 6-month-old Ai6 control and 5xFAD; Ai6 transgenic mice. A schematic drawing depicts the Cre-Ai6 system used to record anterograde transsynaptic delivery events. B Whole-brain imaging of Ai6 and 5xFAD; Ai6 mice exhibited damaged synapses and reduced viral spreading across vDG connections in 5xFAD; Ai6 mice (bottom). Ai6 mice showed brain-wide spreading of Cre-mediated Ai6 expression (top). White arrow indicates Ai6 expression in the cortex, marked by transsynaptic transfer of AAV1-Cre. Scale bars = 2 mm (left), 1 mm (right). C Confocal images of coronal sections including the septum, hippocampus, and dentate gyrus of Ai6 and 5xFAD; Ai6 mouse brain immunostained for Aβ and Ctip2. In the septum, Ai6 mice showed transsynaptic labeling of a few neurons (yellow arrows) in contrast to the aggregation of Ai6 signals in the neuritic spheroids adjacent to Aβ plaques. In the cortex, Ai6 mice showed transsynaptic labeling of cortical neurons (yellow arrows), however, 5xFAD; Ai6 mice showed Ai6-positive neurites but did not show Ai6 labeled cortical neurons. In the dentate gyrus, Ai6-positive dentate granule neurons and hilus neurons were widely distributed, and a few Ai6-positive hilus neurons were visible in 5xFAD; Ai6 mice. Aggregation of Ai6 in the molecular layers depicts dystrophic axons near Aβ plaques. Scale bars = 100 μm. D Synaptosomes in the hippocampo-septal tracts of WT and 5xFAD mice were analyzed by LC–MS/MS to investigate the biological processes underlying axonal spheroid formation and synaptic dysfunction. E A total of 1,099 proteins, with 743 overlapping, 92 WT control-unique, and 264 5xFAD-unique proteins, were identified in the proteomic analysis of synaptosomes (Venn diagram). A scatter plot shows 337 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 109 down-regulated proteins (log2 fold-change < -0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between 5xFAD (12-month-old, n = 4, 4 male) and WT (12-month-old, n = 4, 4 male) control synaptosomes and the weight value of this quantification. The position of the representative proteins selected for gene-set enrichment analysis is indicated in colors (blue = downregulated, red = upregulated). F Gene-set enrichment analysis based on GO Biological Process showed that both the downregulated and the upregulated proteins were linked to vesicle-related pathways, synaptic signaling pathways including chemical synaptic transmission and signal release from the synapse, and the neuron projection morphogenesis pathway. G PPI network analysis using the STRING database identified a network with 30 nodes and 50 edges from proteins in 10 upregulated pathways and a network with 20 nodes and 32 edges from proteins in 10 downregulated pathways, as shown in Fig. 2FCC BY · PubMed Central
Fig. 3 The enrichment and direct interaction of CD81, a tetraspanin protein of EVs, with APP in the synaptosomes of the 5xFAD mouse brain. A A schematic drawing depicts the localization of axonal spheroids and pSer8-Aβ (Aβ with phosphorylation at the Ser8 residue) in the periphery of an Amylo-Glo-stained plaque core. Plaque-associated Iba1-positive microglia create barriers in their processes. A representative three-dimensional image of axonal spheroids with Aβ plaques consisting of the core (Amylo-Glo), the diffuse area (pSer8-Aβ), and a microglial barrier (Iba1). Scale bar = 5 μm. B Confocal mages of the septum from 5xFAD; Thy1-YFP mice at two different ages (2 and 6 months) were immunostained for Iba1 and Aβ. Higher magnification images of the white box are presented in (c). Aggregation of YFP-positive axons and clustering of microglia are visible at 6-months of age. Scale bars = 100 μm. C Higher magnification images of the white box in (b) show that the axonal spheroids became larger as the amount of pSer8-Aβ increased. Scale bar = 5 μm for 2-month-old mice, 10 μm for 6-month-old mice. D Primary hippocampal neurons were harvested at 48 h after treatment with 1 μM Aβ1-42 synthetic oligomer for 6 h. Proteins isolated from the whole lysate of the three independent experiments (n = 3) were analyzed by LC–MS/MS. E A scatter plot shows 907 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 233 downregulated proteins (log2 fold-change < − 0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between Aβ-treated and control primary hippocampal neurons and the weight value of this quantification. PPI network analysis using the STRING database identified a network of CD81, APP, and 65 other proteins with 310 interactions from the upregulated proteins of the differentially expressed protein analysis. F Gene-set enrichment analysis based on GO Biological Process showed 4 downregulated pathways related to cilia assembly and 18 upregulated pathways related to vesicle transport, endocytosis, and exocytosis. G The proteomic analysis of the APP C-terminal fragment-immunoprecipitated synaptosomes profiled the proteins involved in the initial formation of axonal spheroids during APP processing in the synaptosomes. A total of 1,070 proteins were identified with, 983 overlapping, 71 WT-unique, and 16 5xFAD-unique proteins. A volcano plot shows 519 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 307 down-regulated proteins (log2 fold-change < − 0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between 5xFAD (6-months-old, n = 4, one female and three male) and WT (6-months-old, n = 4, two female and two male) synaptosome samples and the adjusted p-value. The positions of the representative proteins selected for further PPI analysis are annotated. H 29 PPIs were linked to the 38 proteins from synaptosomes that were immunoprecipitated with APP. Several proteins, such as MIF (red) and CD81 (green), interacted directly with APP (blue)
Fig. 3 The enrichment and direct interaction of CD81, a tetraspanin protein of EVs, with APP in the synaptosomes of the 5xFAD mouse brain. A A schematic drawing depicts the localization of axonal spheroids and pSer8-Aβ (Aβ with phosphorylation at the Ser8 residue) in the periphery of an Amylo-Glo-stained plaque core. Plaque-associated Iba1-positive microglia create barriers in their processes. A representative three-dimensional image of axonal spheroids with Aβ plaques consisting of the core (Amylo-Glo), the diffuse area (pSer8-Aβ), and a microglial barrier (Iba1). Scale bar = 5 μm. B Confocal mages of the septum from 5xFAD; Thy1-YFP mice at two different ages (2 and 6 months) were immunostained for Iba1 and Aβ. Higher magnification images of the white box are presented in (c). Aggregation of YFP-positive axons and clustering of microglia are visible at 6-months of age. Scale bars = 100 μm. C Higher magnification images of the white box in (b) show that the axonal spheroids became larger as the amount of pSer8-Aβ increased. Scale bar = 5 μm for 2-month-old mice, 10 μm for 6-month-old mice. D Primary hippocampal neurons were harvested at 48 h after treatment with 1 μM Aβ1-42 synthetic oligomer for 6 h. Proteins isolated from the whole lysate of the three independent experiments (n = 3) were analyzed by LC–MS/MS. E A scatter plot shows 907 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 233 downregulated proteins (log2 fold-change < − 0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between Aβ-treated and control primary hippocampal neurons and the weight value of this quantification. PPI network analysis using the STRING database identified a network of CD81, APP, and 65 other proteins with 310 interactions from the upregulated proteins of the differentially expressed protein analysis. F Gene-set enrichment analysis based on GO Biological Process showed 4 downregulated pathways related to cilia assembly and 18 upregulated pathways related to vesicle transport, endocytosis, and exocytosis. G The proteomic analysis of the APP C-terminal fragment-immunoprecipitated synaptosomes profiled the proteins involved in the initial formation of axonal spheroids during APP processing in the synaptosomes. A total of 1,070 proteins were identified with, 983 overlapping, 71 WT-unique, and 16 5xFAD-unique proteins. A volcano plot shows 519 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 307 down-regulated proteins (log2 fold-change < − 0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between 5xFAD (6-months-old, n = 4, one female and three male) and WT (6-months-old, n = 4, two female and two male) synaptosome samples and the adjusted p-value. The positions of the representative proteins selected for further PPI analysis are annotated. H 29 PPIs were linked to the 38 proteins from synaptosomes that were immunoprecipitated with APP. Several proteins, such as MIF (red) and CD81 (green), interacted directly with APP (blue)CC BY · PubMed Central
Fig. 4 The accumulation of EVs in the axonal spheroids by dysfunctional vesicle exocytosis. A Tetraspanin proteins (CD9, CD63, and CD81), synaptic markers (Synapsin-1, PSD95: Postsynaptic density protein 95, and Synaptophysin), and Aβ protein were found in SDS-soluble lysates from hippocampal tissues and synaptosomes from 5xFAD (12-months-old, n = 4, four male) and WT control mice (12-months-old, n = 4, four male). 15 μg proteins from each sample were loaded. The plots are presented as the mean ± S.E.M. of four independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001 as determined by the unpaired t-test. B Annexin-based biosensor called pSIVA, which detects exosome-like vesicles by binding to externalized phosphatidylserine was used to observe the accumulation of exosome-like vesicles at the axon terminals in the 5xFAD; Thy1-YFP mouse septum. pSIVA signals were detected in the DNs at 2-month-old mice but more localized in the Amylo-Glo-positive plaque core at 6-month-old mice. Scale bars = 10 μm. C Axonal spheroids surrounding the Aβ plaque of the 5xFAD; Thy1-YFP mice were fluorescently labeled for Aβ and CD63 to offer direct evidence of the accumulation of exosome-like vesicles at the axon terminals. Scale bars = 10 μm. D LC–MS/MS was utilized to identify CD63-immunoprecipitated proteins that were enriched in the synaptosomes collected from the hippocampo-septal tracts (hippocampo-septal synaptosome) of WT (6-months-old, n = 11, one female and ten male mice) and 5xFAD mice (6-months-old, n = 10, four female and six male mice) in order to explore the proteomic changes of EV states in the presence of Aβ plaques. The quality of the CD63-immunoprecipitated synaptosomal fractions was checked by western blot analysis of the pull-down, flow-through, and whole synaptosomes for CD63 and Aβ. E A total of 1,780 proteins were identified, including 1605 overlapping, 44 WT-unique, and 131 5xFAD-unique proteins. A scatter plot showed 211 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 65 down-regulated proteins (log2 fold-change < -0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between 5xFAD and WT control CD63-IPed synaptosome samples and the weight value of this quantification. The positions of the significantly increased and decreased proteins are indicated. F The PPI network analysis using the STRING database identified a network with 19 nodes and 53 edges from the proteins in 11 upregulated pathways and a network with 7 nodes and 2 edges from the proteins in 5 downregulated pathways selected for gene-set enrichment analysis. G Gene-set enrichment analysis based on GO Biological Process showed 5 downregulated pathways related to vesicle-mediated transport and exocytosis and 11 upregulated pathways related to vesicle transport, endocytosis, and exocytosis
Fig. 4 The accumulation of EVs in the axonal spheroids by dysfunctional vesicle exocytosis. A Tetraspanin proteins (CD9, CD63, and CD81), synaptic markers (Synapsin-1, PSD95: Postsynaptic density protein 95, and Synaptophysin), and Aβ protein were found in SDS-soluble lysates from hippocampal tissues and synaptosomes from 5xFAD (12-months-old, n = 4, four male) and WT control mice (12-months-old, n = 4, four male). 15 μg proteins from each sample were loaded. The plots are presented as the mean ± S.E.M. of four independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001 as determined by the unpaired t-test. B Annexin-based biosensor called pSIVA, which detects exosome-like vesicles by binding to externalized phosphatidylserine was used to observe the accumulation of exosome-like vesicles at the axon terminals in the 5xFAD; Thy1-YFP mouse septum. pSIVA signals were detected in the DNs at 2-month-old mice but more localized in the Amylo-Glo-positive plaque core at 6-month-old mice. Scale bars = 10 μm. C Axonal spheroids surrounding the Aβ plaque of the 5xFAD; Thy1-YFP mice were fluorescently labeled for Aβ and CD63 to offer direct evidence of the accumulation of exosome-like vesicles at the axon terminals. Scale bars = 10 μm. D LC–MS/MS was utilized to identify CD63-immunoprecipitated proteins that were enriched in the synaptosomes collected from the hippocampo-septal tracts (hippocampo-septal synaptosome) of WT (6-months-old, n = 11, one female and ten male mice) and 5xFAD mice (6-months-old, n = 10, four female and six male mice) in order to explore the proteomic changes of EV states in the presence of Aβ plaques. The quality of the CD63-immunoprecipitated synaptosomal fractions was checked by western blot analysis of the pull-down, flow-through, and whole synaptosomes for CD63 and Aβ. E A total of 1,780 proteins were identified, including 1605 overlapping, 44 WT-unique, and 131 5xFAD-unique proteins. A scatter plot showed 211 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 65 down-regulated proteins (log2 fold-change < -0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between 5xFAD and WT control CD63-IPed synaptosome samples and the weight value of this quantification. The positions of the significantly increased and decreased proteins are indicated. F The PPI network analysis using the STRING database identified a network with 19 nodes and 53 edges from the proteins in 11 upregulated pathways and a network with 7 nodes and 2 edges from the proteins in 5 downregulated pathways selected for gene-set enrichment analysis. G Gene-set enrichment analysis based on GO Biological Process showed 5 downregulated pathways related to vesicle-mediated transport and exocytosis and 11 upregulated pathways related to vesicle transport, endocytosis, and exocytosisCC BY · PubMed Central
Fig. 5 Primary cilia dysfunction contributed to axonal spheroid formation and Aβ deposition through vesicle accumulation. A Immunostaining of the 5xFAD mouse septum for nucleus (DAPI), Aβ, and MIF showed that MIF protein was located on DNs adjacent to Aβ plaques, suggesting that its failure of secretion and accumulation in DNs. Scale bars = 10 μm, green dashed lines: DN, white dashed lines: Aβ plaque, red dashed lines: MIF protein in DNs. B To examine the localization of MIF protein in the hippocampal neurons, MIF was immunostained using 5xFAD hippocampus. MIF protein was found at the basal end of ACIII-labeled primary cilia (yellow arrowhead) and co-localized with the ubiquitin-labeled proteasome system. Scale bar = 10 μm for ACIII and scale bar = 5 μm for ubiquitin staining. C Primary hippocampal neurons were harvested after siIft88 (40 nM) transfection for 48 h and oligomeric Aβ1-42 (1 μM) treatment for 6 h (siCon: scrambled short interfering RNA, siIft88: short interfering RNA for Ift88). Whole cell lysates (n = 3) were analyzed by LC–MS/MS. A total of 3546 proteins, with 3152 overlapping, 32 siCon-unique, and 362 siIft88-unique proteins, were identified. A scatter plot showed 527 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 126 down-regulated proteins (log2 fold-change < − 0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between siIft88-transfected and siCon- transfected primary neurons and the weight value of this quantification. The position of the representative proteins selected for the pathway enrichment analysis is indicated. D Gene-set enrichment analysis based on GO Biological Process revealed 3 downregulated pathways related to vesicle-mediated transport and cilia assembly, and 16 upregulated pathways related to vesicle transport, endocytosis, and exocytosis. E Protein complexes were isolated from the primary neuron lysate (top) or the EV proteins collected from the conditioned media (bottom) followed by IP with the anti-CD63. Western blot analysis showed that Aβ was not complexed with CD63 in the lysate but was preferentially bound to CD63 in the control EVs. F Using hippocampal neurons derived from 5xFAD; Ift88-flox/flox mice grown in compartmentalized neuron culture platforms with Cre/loxP-mediated Ift88 deletion, the area of axon terminals was measured (left plot, axonal spheroid), and the co-localization coefficients of CD63 and Aβ in the axon terminals were measured (right plot). Primary neurons were derived from 5xFAD mice to induce intraneuronal Aβ accumulation without extracellular treatment. Both axon terminal areas and co-localization of CD63 and intraneuronal Aβ were significantly increased in primary neurons with Cre-induced knock-down of Ift88 expression (n = 3). Data are presented as the mean ± S.E.M. ****p < 0.001 as determined by the unpaired t-test (Ift88KD: knockdown of the Ift88 gene). G 5xFAD; Ift88-flox/flox; Ai6 (mut) and 5xFAD; Ift88flox/wt; Ai6 (het) mice were used for stereotaxic injection of AAV-Cre. AAV vectors were injected into the ventral hippocampus, and confocal imaging was conducted using the septum to measure the number of axonal spheroids surrounding the Aβ deposition (n = 3). Higher magnification images were taken from the box area of the upper panel. Data are presented as the mean ± S.E.M. *p < 0.05 as determined by the unpaired t-test. Scale bars = 50 μm (upper), 5 μm (lower)
Fig. 5 Primary cilia dysfunction contributed to axonal spheroid formation and Aβ deposition through vesicle accumulation. A Immunostaining of the 5xFAD mouse septum for nucleus (DAPI), Aβ, and MIF showed that MIF protein was located on DNs adjacent to Aβ plaques, suggesting that its failure of secretion and accumulation in DNs. Scale bars = 10 μm, green dashed lines: DN, white dashed lines: Aβ plaque, red dashed lines: MIF protein in DNs. B To examine the localization of MIF protein in the hippocampal neurons, MIF was immunostained using 5xFAD hippocampus. MIF protein was found at the basal end of ACIII-labeled primary cilia (yellow arrowhead) and co-localized with the ubiquitin-labeled proteasome system. Scale bar = 10 μm for ACIII and scale bar = 5 μm for ubiquitin staining. C Primary hippocampal neurons were harvested after siIft88 (40 nM) transfection for 48 h and oligomeric Aβ1-42 (1 μM) treatment for 6 h (siCon: scrambled short interfering RNA, siIft88: short interfering RNA for Ift88). Whole cell lysates (n = 3) were analyzed by LC–MS/MS. A total of 3546 proteins, with 3152 overlapping, 32 siCon-unique, and 362 siIft88-unique proteins, were identified. A scatter plot showed 527 up-regulated (log2 fold-change > 0.25, adjusted p-value < 0.1) and 126 down-regulated proteins (log2 fold-change < − 0.25, adjusted p-value < 0.1) as a result of the differentially expressed protein analysis. Shown are the fold changes in protein abundance between siIft88-transfected and siCon- transfected primary neurons and the weight value of this quantification. The position of the representative proteins selected for the pathway enrichment analysis is indicated. D Gene-set enrichment analysis based on GO Biological Process revealed 3 downregulated pathways related to vesicle-mediated transport and cilia assembly, and 16 upregulated pathways related to vesicle transport, endocytosis, and exocytosis. E Protein complexes were isolated from the primary neuron lysate (top) or the EV proteins collected from the conditioned media (bottom) followed by IP with the anti-CD63. Western blot analysis showed that Aβ was not complexed with CD63 in the lysate but was preferentially bound to CD63 in the control EVs. F Using hippocampal neurons derived from 5xFAD; Ift88-flox/flox mice grown in compartmentalized neuron culture platforms with Cre/loxP-mediated Ift88 deletion, the area of axon terminals was measured (left plot, axonal spheroid), and the co-localization coefficients of CD63 and Aβ in the axon terminals were measured (right plot). Primary neurons were derived from 5xFAD mice to induce intraneuronal Aβ accumulation without extracellular treatment. Both axon terminal areas and co-localization of CD63 and intraneuronal Aβ were significantly increased in primary neurons with Cre-induced knock-down of Ift88 expression (n = 3). Data are presented as the mean ± S.E.M. ****p < 0.001 as determined by the unpaired t-test (Ift88KD: knockdown of the Ift88 gene). G 5xFAD; Ift88-flox/flox; Ai6 (mut) and 5xFAD; Ift88flox/wt; Ai6 (het) mice were used for stereotaxic injection of AAV-Cre. AAV vectors were injected into the ventral hippocampus, and confocal imaging was conducted using the septum to measure the number of axonal spheroids surrounding the Aβ deposition (n = 3). Higher magnification images were taken from the box area of the upper panel. Data are presented as the mean ± S.E.M. *p < 0.05 as determined by the unpaired t-test. Scale bars = 50 μm (upper), 5 μm (lower)CC BY · PubMed Central
Fig. 6 Single-cell transcriptome profiling of the hippocampal tissues identified the specific neuron type associated with vesicle accumulation in the septum. A A UMAP plot shows the subtypes of neurons produced from WT (3-month-old, n = 3, 3 male; 6-month-old, n = 1, female) and 5xFAD (3-month-old, n = 4, male; 6-month-old, n = 2, female) hippocampus tissues. Using independent component analysis using canonical markers, nine different neuronal subclusters, including 1 undefined (u) were found from 5468 neurons, characterized by strong expression of Meg3 and Atp1b1. Each cluster is color-coded according to cell type. B A dot plot illustrated the expression of selected marker genes in each neuron subtype (u: undefined). C A permutation plot to calculate the average proportion of each neuron subtype between WT and 5xFAD mouse hippocampal tissues demonstrated that the number of neurons with Sst, Chgb, and Calb2 markers was significantly decreased in 5xFAD mice (ASC: astrocyte, Endo: endothelial cells, EPC: ependymal cells, NEUT: neutrophils, OPC: oligodendrocyte precursor cells, OLG: oligodendrocytes). D Differential gene expression analysis revealed 620 up-regulated (log2 fold-change > 0.25, p-value < 0.01) and 234 down-regulated genes (log2 fold-change < − 0.25, p-value < 0.01) in the SOM, and 738 up-regulated (log2 fold-change > 0.25, p-value < 0.01) and 186 down-regulated genes (log2 fold-change < -0.25, p-value < 0.01) in the Chgb-expressing neuron. Shown are the fold changes of gene expression between 5xFAD and WT control mouse hippocampus in SOM and the fold changes of gene expression between 5xFAD and WT control mouse hippocampus in Chgb-expressing neurons. E Gene-set enrichment analysis in GO biological process signatures showed that SOM and Chgb-expressing neurons commonly turned down pathways related to secretion, cell secretion, and the regulation of cell secretion, as well as the non-motile ciliary assembly pathway. F Gene network connections and associated functions were visualized based on GO biological process signatures derived from differential gene expression analysis of SOM. Gene lists for gene ontology terms are derived from SOM, as shown in Fig. 6E. The genes related to secretion, non-motile cilium assembly, phagocytosis, and synaptic plasticity were enclosed by black boxes. The gene symbols were matched with the STRING database, and we identified a network with 39 nodes and 132 edges. G Immunofluorescence analysis of hippocampal sections from WT and 5xFAD; Thy1-YFP mice showed that the number of SOM in the CA1 area of the hippocampus (left), and the dentate gyrus hilus (middle) was lower in 5xFAD mice than in WT. Plots represent the number of SOMs in each area. Higher magnification images of SST expression showed a bright aggregation of SST in the perinucleus of the SOM in 5xFAD. The data are presented as the mean ± S.E.M. **p < 0.01; ****p < 0.001 as determined by the unpaired t-test. Scale bars = 5 μm. H Axonal spheroids were observed in the vicinity of septal Aβ plaques for SOM by immunofluorescence staining for SST (white dashed line) in the 5xFAD; Thy1-YFP mouse septum. I Immunofluorescence analysis of the septum area in 5xFAD; Thy1-YFP mice showed that the axonal spheroids of SOM are greatly induced in the vicinity of Aβ plaques (Amylo-Glo) as early as 2-months old when Thy1-YFP-positive projection neurons initiate their axonal dystrophy. Scale bars = 5 μm. J Images of SOM in WT and 5xFAD mice were immunostained for Hoechst, CD63, ACIII, and SST. CD63 aggregates formed in the brightly stained perinuclear SST and primary cilia were short and weakly labeled in the SOMs of 5xFAD mice. Scale bar = 2 μm
Fig. 6 Single-cell transcriptome profiling of the hippocampal tissues identified the specific neuron type associated with vesicle accumulation in the septum. A A UMAP plot shows the subtypes of neurons produced from WT (3-month-old, n = 3, 3 male; 6-month-old, n = 1, female) and 5xFAD (3-month-old, n = 4, male; 6-month-old, n = 2, female) hippocampus tissues. Using independent component analysis using canonical markers, nine different neuronal subclusters, including 1 undefined (u) were found from 5468 neurons, characterized by strong expression of Meg3 and Atp1b1. Each cluster is color-coded according to cell type. B A dot plot illustrated the expression of selected marker genes in each neuron subtype (u: undefined). C A permutation plot to calculate the average proportion of each neuron subtype between WT and 5xFAD mouse hippocampal tissues demonstrated that the number of neurons with Sst, Chgb, and Calb2 markers was significantly decreased in 5xFAD mice (ASC: astrocyte, Endo: endothelial cells, EPC: ependymal cells, NEUT: neutrophils, OPC: oligodendrocyte precursor cells, OLG: oligodendrocytes). D Differential gene expression analysis revealed 620 up-regulated (log2 fold-change > 0.25, p-value < 0.01) and 234 down-regulated genes (log2 fold-change < − 0.25, p-value < 0.01) in the SOM, and 738 up-regulated (log2 fold-change > 0.25, p-value < 0.01) and 186 down-regulated genes (log2 fold-change < -0.25, p-value < 0.01) in the Chgb-expressing neuron. Shown are the fold changes of gene expression between 5xFAD and WT control mouse hippocampus in SOM and the fold changes of gene expression between 5xFAD and WT control mouse hippocampus in Chgb-expressing neurons. E Gene-set enrichment analysis in GO biological process signatures showed that SOM and Chgb-expressing neurons commonly turned down pathways related to secretion, cell secretion, and the regulation of cell secretion, as well as the non-motile ciliary assembly pathway. F Gene network connections and associated functions were visualized based on GO biological process signatures derived from differential gene expression analysis of SOM. Gene lists for gene ontology terms are derived from SOM, as shown in Fig. 6E. The genes related to secretion, non-motile cilium assembly, phagocytosis, and synaptic plasticity were enclosed by black boxes. The gene symbols were matched with the STRING database, and we identified a network with 39 nodes and 132 edges. G Immunofluorescence analysis of hippocampal sections from WT and 5xFAD; Thy1-YFP mice showed that the number of SOM in the CA1 area of the hippocampus (left), and the dentate gyrus hilus (middle) was lower in 5xFAD mice than in WT. Plots represent the number of SOMs in each area. Higher magnification images of SST expression showed a bright aggregation of SST in the perinucleus of the SOM in 5xFAD. The data are presented as the mean ± S.E.M. **p < 0.01; ****p < 0.001 as determined by the unpaired t-test. Scale bars = 5 μm. H Axonal spheroids were observed in the vicinity of septal Aβ plaques for SOM by immunofluorescence staining for SST (white dashed line) in the 5xFAD; Thy1-YFP mouse septum. I Immunofluorescence analysis of the septum area in 5xFAD; Thy1-YFP mice showed that the axonal spheroids of SOM are greatly induced in the vicinity of Aβ plaques (Amylo-Glo) as early as 2-months old when Thy1-YFP-positive projection neurons initiate their axonal dystrophy. Scale bars = 5 μm. J Images of SOM in WT and 5xFAD mice were immunostained for Hoechst, CD63, ACIII, and SST. CD63 aggregates formed in the brightly stained perinuclear SST and primary cilia were short and weakly labeled in the SOMs of 5xFAD mice. Scale bar = 2 μmCC BY · PubMed Central

Verification

Claims the paper makes

40 Valid · 0 Not valid · 7 Not checked · 154 Own results and interpretations

99 more claims · 5 with their cited sources checked

Create a free account to see all 201 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 42
  • It may state two separate claims as one. 73
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 87 references

Research works printed but not linked: 1

  1. Takahashi RH, Milner TA, Li F, Nam EE, Edgar MA, Yamaguchi H, Beal MF, Xu HX, Greengard P, Gouras GK (2002) Intraneuronal Alzheimer A beta 42 accumulates in multivesicular bodies and is associated with synaptic pathology. Am J Pathol 161:1869–1879. https://doi.org/10.1016/S0002-9440(10)64463-X · DOI · Paper page

Cited by

See all 20

Metadata

Authors
Jang, J., Yeo, S., Baek, S., Jung, H. J., Lee, M. S., Choi, S. H., & Choe, Y.
Affiliations
Korea Brain Research Institute (KR)
Corresponding author
Choe, Youngshik
Venue
Acta Neuropathologica Communications
Published
2023-09-04
Volume
11
Issue
1
Pages
142-142
ISSN
2051-5960 (ISSN-L)
OpenAlex
W4386422726
Cited by
20
References
87
Citations by year
2026: 2 · 2025: 11 · 2024: 6 · 2023: 1
FWCI
3.71
Type
journal article
Language
en
Open access
Licence
Locations
  • Acta Neuropathologica Communications (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • KBRI (23-BR-03-05, 23-BR-02-01)
  • NRF (2020M3A9B6037812, 2022R1A5A202954611)
Topics
  • Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Alzheimer's disease research and treatments
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Extracellular vesicles in disease
  • Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Neuroscience and Neuropharmacology Research
Keywords
Alzheimer's disease, dystrophic neurites, extracellular vesicles, primary cilia, IFT88, amyloid-beta, axonal swellings, presynaptic terminals, synaptic loss, hippocampal neurons, somatostatin neurons, axonal transport, exocytosis
MeSH terms
Extracellular Vesicles, Dependovirus, Alzheimer Disease, Animals, Axons, Cilia, Hippocampus, Amyloid beta-Peptides, Proteomics, Mice

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-09-28, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-09-28 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Abnormal accumulation of extracellular vesicles in hippocampal dystrophic axons and regulation by the primary cilia in Alzheimer's disease. https://app.cochat.ai/papers/10.1186/s40478-023-01637-3
The paper
Jang, J., Yeo, S., Baek, S., Jung, H. J., Lee, M. S., Choi, S. H., & Choe, Y. (2023). Abnormal accumulation of extracellular vesicles in hippocampal dystrophic axons and regulation by the primary cilia in Alzheimer's disease. Acta Neuropathologica Communications. https://doi.org/10.1186/s40478-023-01637-3

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.